Synthesis of 4,4-difluoroglycosides using ring-closing metathesis

Article abstract of DOI:10.1039/b313731g

4-Deoxy-4,4-difluoro-glycosides have been synthesised for the first time via a direct sequence involving ring-closing metathesis and indium-mediated difluoroallylation with 1-bromo-1,1-difluoropropene in water. Two protecting group strategies were explore

Full text of DOI:10.1039/b313731g

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of 4,4-difluoroglycosides using ring-closing metathesis
Christophe Audouard,a John Fawcett,a Gerry A. Griffiths,a Jonathan M. Percy,*a
Stéphane Pintat†a and Clive A. Smithb
a Department of Chemistry, University of Leicester, University Road, Leicester, UK LE1 7RH.
E-mail: jmp29@le.ac.uk; Fax: ϩ44 116 252 3789; Tel: ϩ44 116 252 2140
b GlaxoSmithKline Pharmaceuticals, New Frontiers Science Park, Third Avenue, Harlow,
UK CM19 5AW
Received 3rd November 2003, Accepted 11th December 2003
First published as an Advance Article on the web 28th January 2004
4-Deoxy-4,4-difluoro-glycosides have been synthesised for the first time via a direct sequence involving ring-closing
metathesis and indium-mediated difluoroallylation with 1-bromo-1,1-difluoropropene in water. Two protecting group
strategies were explored, one to allow protection of the primary C-6 hydroxyl group throughout the sequence, while
the second was intended to allow deprotection after RCM and before dihydroxylation. The benzyl ether could be
used in the first role, and pivaloyl is effective in the second. Dihydroxylations were highly stereoselective and
controlled by the orientation of the glycosidic C–O bond.
myces fradiae. The proposed mechanism (Scheme 2) involves
deprotonation of the ketodeoxysugar 1 (drawn in conformer
1a) to form enolate 2; methylation occurs at C-3 with inversion
of hydroxyl group configuration. Liu proposed that 3 would
undergo dehydrofluorination: the fluoroenol product would not
be able to undergo the C-3 methylation and the enzyme would
be inhibited, but exposure of the enzyme to 3 failed to result in
either inhibition or fluoride ion release. The proposal is a sur-
Introduction
Fluorosugars are of considerable interest because they retain
much of the reactivity of natural saccharides while lacking the
ability to enter into critical hydrogen bonding interactions with
nucleic acids or proteins. In addition, suitably placed fluorine
atoms can modulate the chemistry of the glycosidic bond
strongly,1 an effect that has been exploited in the study of
carbohydrate processing enzymes.2 Most spectacularly, substi-
prising one given the unsuitable stereoelectronic relationship
tution at the 2-position has allowed crystallographic studies of
between the C–H3 bond and the carbonyl π* orbital; conformer
glycosyl enzyme intermediates to be performed,3 while 2,2-
1b appears a more likely precursor to the enolate.10
difluorosugars have been used to label active site nucleophiles
in inverting glycosidases.4 In the latter cases, the glycosidic bond
is so stable that enzymatic digestion of the glycosyl enzyme
intermediate can be achieved, yielding valuable sequence
information (Scheme 1).
Scheme 1 Reaction leading to the inhibition of α-galactosidase
(i) from Phanerochaete chrysosporium.
Scheme
2
Proposed C-3 methyltransferase mechanism. i, TylC3,
AdoMet (ref. 9).
Both electrophilic and nucleophilic fluorination methods
have been applied to the synthesis of 2-deoxy-2-fluoro-5 and
2-deoxy-2,2-difluoro-sugars.4,6 Glycal fluorination typifies the
former approach, while transformations of hydroxyl or ketonic
carbonyl groups with DAST account for the latter. There have
been well-documented complications arising from the well-
defined stereoelectronic relationships around the sugar ring;7
neighbouring group participation, group shifts and elimination
reactions have all attended fluorinations and difluorinations
with DAST.8 While difluorination at the 2- and 3-positions has
been described, we found no papers describing the synthesis
of 4-deoxy-4,4-difluorosugars. Liu has however, published a
DAST-based 10-step route to TDP 4,4-difluoro-2,4,6-trideoxy-
sugar 3, a proposed methyltransferase inhibitor, from a known
sugar epoxide.9
Also, C–F bonds are very strong and dehydrofluorination
normally requires quite forcing conditions, so it seems unlikely
that an enzyme set up to enolise a ketone would be able to
achieve this more demanding conversion. An alternative appli-
cation for 4-deoxy-4,4-difluorosugars could involve activated
derivatives like 3 in other roles. The effect of the two fluorine
atoms would be to stabilize the bond to the TDP leaving group
considerably (given that a single fluorine atom has a strong
effect upon the ability to ascend to oxacarbenium ion-like
transition states). Such compounds could be used to probe the
binding sites of glycosyltransferase enzymes.11
Building block syntheses of 4,4-difluorohexose analogues
had not yet been completed at the start of our study. Taguchi12
synthesised dihydropyrone 5 from 1,1-difluoro-2,4-dialkoxy-
1,3-butadiene 4 using a hetero Diels–Alder reaction (Scheme 3),
and advanced the intermediate dihydropyrone to racemic
2,4-dideoxy-4,4-difluorosugar 6, but did not describe the
4-deoxy-4,4-difluorosugars 7. Since then, Uneyama13 has
shortened the synthesis of an analogue of 4, and of pyrones 5,
though further elaboration has not been described. In this
The reaction catalysed by the C-3 methyltransferase is a key
step in the biosynthesis of the antibiotic Tylosin by Strepto-
† Present address: Evotech OAI, 151 Milton Park, Abingdon, Oxon,
UK OX14 4SD.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
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 4
528

View Article Online
Methallyl ether 14d was prepared in the same way but failed
to undergo RCM, even after one week in DCM at reflux with
Grubbs’ catalyst; this result has implications for the sequence
of events in these RCM reactions, which we will discuss later.
The most direct approach to the synthesis of unknown
difluorosugars 7 would involve either direct oxidation of 12d to
the lactone 16 using CrO3/3,5-dimethylpyrazole21 or RCM of
acryloyl ester 15 (Scheme 5). Exposure of 12d to the oxidation
conditions resulted in destruction of the starting material;
no identifiable fluorinated products were obtained following
solvent extraction of the black tar that formed the reaction
products. Acryloyl ester 15 was prepared, but failed to undergo
RCM to any appreciable extent in the presence of either of the
Grubbs’ or Nolan catalysts,22 either in DCM or DCE over 7
days. We decided also to explore the transacetalisation with
acrolein diethyl acetal described by Mulzer,23 followed by RCM
which would deliver a mixture of diastereoisomeric dihydro-
pyrans. Ideally, these would be separated before dihydroxyl-
ation to avoid potential problems with the separation of highly
polar sugars (Scheme 6).
Scheme 3 Taguchi route to racemic 2,4-dideoxy-4,4-difluorosugar 6.
manuscript, we wish to describe completed syntheses of
racemic 7. The route delivers novel sugars, and tests the scope
and limitations of RCM reactions which contain a difluorin-
ated allyl fragment.
Results and discussion
Recently, we described a useful combination of difluoroallyl-
ation and ring-closing metathesis reactions for the rapid syn-
thesis of difluorinated pyrans.14 Seyferth15 and Burton16 had
described efficient additions of 1-bromo-1,1-difluoropropene to
aldehydes, but the sub-millimole scale procedure reported by
Kirihara and co-workers17 seemed particularly convenient. We
were surprised to note that stirring was used, given the pro-
pensity of indium powder to agglomerate; indeed, the method
failed to work in our hands when stirring was employed. How-
ever, when suspensions of reactants were shaken vigorously or
sonicated, we were able to reproduce their result with benzalde-
hyde (though not to match the reported yield of 99%)
and achieve reaction with furfural, pentanal and benzyloxy-
acetaldehyde as an illustrative set of aldehydes in moderate to
good yield (Scheme 4). The reaction required a modest excess of
the propene (1.5 equivalents).
Scheme 5 Reagents and conditions: i, acryloyl chloride, Et3N, DCM,
0 ЊC to rt, 16 h; ii, 5–10% Grubbs’ or Nolan catalyst, DCM or DCE,
reflux 7 days; iii, 3,5-dimethylpyrazole, CrO3, DCM, Ϫ20 ЊC.
Scheme 4 Synthesis of pyrans from 1-bromo-1,1-difluoropropene: i, In,
DMF, RCHO 9; ii, NaOH, TBAHSO4, allyl bromide, water, rt; iii, 5%
Grubbs’ catalyst, DCM, rt; iv, OsO4, NMO, t-BuOH, acetone, water, rt.
Scheme 6 Synthesis of fully-functionalised pyrans. i, acrolein diethyl
acetal, PhMe, heat, 60 mmHg; ii, 5% Grubbs’ catalyst, DCM, reflux,
see text; iii, separate; iv, OsO4, NMO, t-BuOH, acetone, water, rt.
Allylations of 10a–10d were facile under phase transfer-cata-
lysed conditions,18 delivering the ethers 11a–11d in high purity
(shown by GC). The ethers were sufficiently pure for use in the
RCM reaction, directly following removal of the extraction sol-
vent. All four ethers delivered pyran products upon exposure to
Grubbs’ catalyst19 in DCM for 24 hours at room temperature,
with a particularly good yield being obtained for 9d (97%). A
rather poor yield (23%) of 9d was obtained when polymer-sup-
ported Grubbs’ catalyst20 was used for the reaction, even after
an extended reaction time (5 days) with a relatively high loading
of catalyst (10%).
Alcohol 10d was transacetalised with acrolein diethyl acetal
to afford 17d in acceptable (63%) yield after column chromato-
graphy. The inseparable mixture of diastereoisomers underwent
RCM smoothly to afford a 1 : 1 mixture of cis-18d and trans-
19d (Fig. 1) in dichloromethane at reflux with either Grubbs’
(3 days, 76%) or Nolan22 (4 hours, 75%) catalysts under
relatively concentrated conditions (0.17 M). The reactions were
followed by TLC, so the ca. 20-fold difference in reaction time
represents a genuine difference in rate. Dihydroxylation of
the mixture under conventional UpJohn conditions24 was very
slow, taking 10 days to reach completion with the addition of
additional osmium tetroxide after 5 days, but affording the
Dihydroxylation occurred slowly and without discrimination
between the alkene faces to afford 13d as a 1 : 1 mixture of
separable cis and trans isomers in moderate yield.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
529

View Article Online
hydroxyl groups is likely; the adjacent CF2 centre would lower
the pKa of the secondary hydroxyl group by up to 3 pKa units. A
migratory mechanism (Scheme 7) involving a tetrahedral inter-
mediate could lead to what appears to be the expulsion of the
poorer of the two leaving groups. In DMF and THF, weak
chelation involving the ester alkyl oxygen is overwhelmed by
interactions with the solvent, and the monodentate indium
alkoxide 23 can exchange through tetrahedral intermediate 24.
The interaction between the C-1 hydroxyl group and indium is
stronger than that with the less basic C-2 hydroxyl group, so
when 24 breaks down, it releases 25 and not 23. In water where
chelation is extremely important, 26 is stabilized and the
exchange reaction through monodentate 23 is not established.
Fig. 1 Three-dimensional representations of 18d and 19d showing the
protons giving rise to the key NOESY cross peak.
sugars in 65% yield. The sluggishness with which the sub-
strates react may be due to the fact that each alkene bears two
-I substitutents at each allylic position. Fortunately, very careful
flash column chromatography with a non-polar eluent (5%
diethylether in 40–60 petroleum ether) allowed the almost com-
plete separation of 18d and 19d though the RF’s were rather
close (0.25 and 0.28); this procedure was used to prepare 500
mg of each pure racemic diastereoisomer.
The cis and trans dihydropyrans 18d and 19d were identified
by NOESY experiments. The H-5 proton can be identified easily
from COSY experiments and by the clear coupling to fluorine
within a rather complex splitting pattern. A clear cross peak
between H-1 and H-5 in the NOESY could be observed for only
one of the pyran diastereoisomers, which was assigned as 18d.
The H-5 signal occurs at significantly different chemical shifts
in the two diastereoisomers (4.50 ppm in the trans, 4.10 ppm in
the cis). In the trans-diastereoisomer, the location of the ethoxy
group in a pseudo-axial environment allows an n–σ* interaction
between the pyran oxygen and the glycosidic bond which may
deshield the H5 proton inductively relative to the corresponding
environment in the cis diastereoisomer. The NMR spectra of
3
both diastereoisomers show large JH–F couplings consistent
Scheme 7 Different solvents lead to different regiochemical outcomes
with the presence of pseudo-diaxial H–F relationships (22.8 Hz
trans, 17.8 Hz cis) and, where resolved, much smaller pseudo-
in the indium-mediated reaction.
3
axial–pseudo-equatorial JH–F couplings (ca. 3 Hz). Pleasingly,
Transacetalisation proceeded smoothly on a 20 gram scale to
afford 17f as a 1 : 1 mixture of diastereoisomeric acetals, then
RCM occurred in reasonable yield, though more slowly than
for 17d, under high dilution conditions (0.005 M) with com-
mercial first generation Grubbs’ catalyst (added in two portions
over 5 days) to afford inseparable pyrans 18f and 19f (Scheme 8).
the H-5 protons in both diastereoisomers resolved into identifi-
able splitting patterns at 400 MHz (trans ddt, cis dddd), allow-
ing all the coupling constants to be extracted. Dihydroxylation
afforded a single sugar product in each case; the assignment of
sugar stereochemistry shown in Scheme 6 will be justified later
in the manuscript.
The C-6 hydroxyl group is protected in the products from
these sequences, which may be advantageous in subsequent
transformations, but we also wished to explore the effect of
deprotection of the C-6 hydroxyl group just after RCM in the
hope that the diastereoisomer separation would be more facile,
and the dihydroxylation more rapid. It was difficult to see how
to cleave the benzyl group while leaving the alkenyl group and
the acetal function intact, so we explored other potential
protecting groups. Attempted transacetalisation of acetoxy-
acetaldehyde adduct 10e with acrolein diethyl acetal resulted in
cleavage of the acetate and the formation of a complex reaction
mixture. Next, we investigated the bulky pivaloyl group to
attempt to suppress side reactions in the transacetalisation. The
aldehyde was prepared from Z-1,4-butenediol on a 10g scale.
Surprisingly, allylation in DMF afforded a mixture of products
with the same mass and very similar retention times in the GC
(10.78 and 10.86 minutes). These were identified as isomeric
products 10f (58%) and 22 (13%), which contained the pivaloxy
group at the more hindered position. However, addition in water
afforded 10f alone in excellent (91%) yield on a multigram scale.
The formation of 22 is surprising; given the bulk of the pro-
tecting group, it would not be expected to migrate into a more
crowded environment. The behaviour of indium() alkoxides
in solution is quite complex, leading to issues of kinetic and
thermodynamic control in crotylation reactions. Chelation is
also documented quite extensively, and strong solvent effects
have been observed. Solvents such as DMF and THF are
believed to coordinate strongly to indium() salts,25 while water
seems to favour the formation of chelated species wherever pos-
sible.26 Of course, a significant pKa difference between the two
Scheme 8 Reagents and conditions: i, acrolein diethyl acetal, PhMe,
heat, 60 mmHg; ii, 5% Grubbs’ catalyst, DCM, reflux, see text; iii,
separate; iv, DIBAl–H, Ϫ78 ЊC to 0 ЊC; v, OsO4, NMO, t-BuOH,
acetone, water, 0 ЊC to rt, 5d, 29, 62% and 30, 62%.
The need for more rigorous conditions (30-fold higher dilu-
tion, catalyst replenished, longer reaction time) compared to
those employed for 17d suggests that the presence of an ester
protecting group impedes the RCM reaction to a significant
extent. After removal of the pivaloyl group by reduction (80%
combined isolated yield, 1 : 1 mixture), flash column chromato-
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
530

View Article Online
graphic separation of 27 and 28 was straightforward. The 19F
NMR spectra of 27 and 28 raised the same issues encountered
for 18d and 19d. Dihydroxylation under UpJohn conditions24
occurred in acceptable (60% isolated) yields in each case. The
higher (twofold) rate of dihydroxylation is presumably due to a
lower degree of steric hindrance as the smaller hydroxyl group
replaces the benzyloxy. Dihydroxylations were spot-to-spot by
TLC: the relatively low yields are a result of performing triol
isolations from the highly aqueous reaction media.
Both reactions were highly stereospecific; indeed we were
only able to detect a single product in each case in the 19F NMR
1
spectra of crude reaction mixtures. In the H NMR spectra in
d6-DMSO, distinct signals were obtained for the three OH
protons, whereas most of the CH protons were overlapped. The
hydroxyl proton at C-6 appears as a triplet, allowing identifi-
cation of the H-6 AB system (COSY) and allowing partial
assignment of the rest of the spectrum (HMBC). The sense of
stereoselection predicted by literature precedent involves attack
trans to the exocyclic glycosidic C–O bond in each case,27 so
that cis 27 affords 29, and trans 28 leads to 30. The diastereo-
isomeric triols displayed quite different splitting patterns for the
axial fluorine atom in each case; in the first instance, we
assigned the relative configurations as shown in Scheme 6 on
the basis of the number of large 3JH–F splittings (Fig. 2).
Fig. 3 ORTEP diagram for 29 showing thermal ellipsoids at 50%.
at the dihydropyran stage. Pyrans 18d and 19d behaved
identically, leading to the formation of 20 and 21 respectively;
product stereochemistries for these sugars, and for pyrans
cis-13d and trans-13d were assigned in the same way. Pyrans 18d
and 27, and 19d and 28 have almost identical 19F NMR spectra,
so the NOESY experiment is validated by the crystal structure
of 29.
Elemental analysis was obtained for 29 and for the triacetate
31 of 30 (Scheme 9). The triacetate shows an extremely well
dispersed 1H NMR spectrum which could be assigned straight-
forwardly using 2D methods (COSY/HMBC).
Scheme 9 Peracetylation and glycoside hydrolysis. i, Ac2O, pyridine, 17
h, rt; ii, HCl, 80 ЊC, 3 days.
Glycoside hydrolysis is
a pre-requisite for any useful
chemistry of our target molecules, so we explored hydrolysis
reactions under simple aqueous conditions. Hydrolysis was
complete after 3 days in hot (80 ЊC) dilute aqueous hydrochloric
acid, but the product 32 proved surprisingly difficult to isolate
and characterise. The sugar was extracted successfully with
ethyl acetate; the aqueous phase was checked carefully for the
presence of product, by TLC, concentrating spots of the reac-
tion solution repeatedly and using the phosphomolybdic acid
stain to visualise the plates. We were unable to detect any
pyranose in the aqueous phase, but the removal of the ethyl
acetate returned an extremely disappointing (<20%) yield
of product. We were more successful when an ethyl acetate
solution was allowed to evaporate slowly at room temperature
and pressure but could not obtain a dry sample of material
for NMR characterisation. We believe that 32 is surprisingly
volatile. Dramatically increased volatility has been described as
a consequence of fluorination of sugars. DiMagno synthesised
a volatile 2,2,3,3,4,4-hexafluoro-2,3,4-trideoxy ‘sugar’ from
hexafluoroglutaric anhydride and 2-lithiofuran,28 but we were
surprised that 32 proved so elusive, given that only a single
hydroxyl group has been replaced. We therefore carried out the
hydrolysis in DCl–D2O to allow direct NMR characterisation
of the sugar products. Fortunately, minimal overlap occurred
between the ethanol released and the sugar (we could resolve
the overlap completely by homodecoupling of the methyl group
of the ethanol).
Fig. 2 Partial 19F NMR spectra (axial fluorine only shown for each)
for 29 and 30. Both partial spectra show the large 2JF–F coupling (249.5
Hz and 242.8 Hz respectively).
Isomer 30 should show two relatively large H–F couplings
given that there are two large H–F dihedral angles (to H-3 and
H-5) and should therefore display the axial fluorine as a doublet
of triplets, whereas 29 should show only one large coupling (to
H-5). The axial fluorine in this latter case is a more finely split
doublet of doublets. We were able to crystallise 29‡ and obtain
the crystal structure (Fig. 3) to verify the NMR assignment.
This result also confirms the sense of stereoselection in the
dihydroxylation and the assignment of relative stereochemistry
1
Both anomers can be seen clearly in the H and 19F NMR
‡ Crystallographic data for 29: C8H14F2O5, crystal size 0.29 × 0.18 ×
0.16 mm, M = 228.19, triclinic, a = 6.9777(8), b = 8.3569(10),
c = 9.3808(11) Å, α = 82.847(2), β = 71.230(2), γ = 77.901(2) deg,
U = 505.45(10) Å3, T = 150(2) K, space group P-1, Z = 2, µ(Mo–Kα) =
0.145 mmϪ1, 3650 reflections measured, 1756 unique (Rint = 0.0127)
which were used in all calculations. Final R indices [F 2>2σ(F 2)] R1 =
0.0334, wR2 = 0.0892; R indices (all data) R1 = 0.0359, wR2 = 0.0907.
CCDC reference numbers 196719. See http://www.rsc.org/suppdata/ob/
b3/b313731g/ for crystallographic data in.cif or other electronic format.
spectra. The anomers were distinguished on the basis of
the appearance of the H-1 or anomeric proton. In the major
β-anomer (the β : α ratio is 6 : 1), the hemiacetal proton appears
as a dd with one large (8.0 Hz) coupling consistent with a
diaxial coupling to H-2, and a small (0.6 Hz) coupling which
was not visible in the 1H{19F} spectrum, indicating that one of
the small splittings arises from a long range (5JH–F) coupling.29
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
531

View Article Online
In the minor anomer, H-1 appears as a triplet with a 3.2 Hz
It is very difficult to compare reaction rates accurately from
the literature data, but it appears that the electronic properties
of the alkene that takes part in the intramolecular metallo-
cycloaddition reaction exert a modest effect on the overall rate.
Furthermore, there are no consistent and major differences
between the reactions of the ethers, acetals and esters, which
have very different conformational properties. The key intra-
molecular metallocycloaddition reaction must involve a con-
formation in which the reacting termini can encounter each
other; for the esters, this means s-cis arrangement 42, whereas
s-trans 43 would always be expected to be more stable and more
highly populated because of the the n–π* interaction shown
(Fig. 5).38 This preference has a dramatic effect on free radical
cyclisation reactions of esters to lactones under tin hydride
conditions.
splitting; again simplification to a doublet occurred in the
5
1H{19F} spectrum so the JH–F coupling is considerably larger
in this case. The size of the J value is consistent with an
equatorial-axial splitting to H-2. Unfortunately, neither of the
long range couplings could be identified readily in the 19F NMR
spectrum at the highest field strength available to us locally
(376.5 MHz). Sugar 32 is related most closely to allose and
gulose, for which the normal β : α anomer ratios are 5.5 : 1 and
5.1 : 1 respectively30 so the introduction of a more electrophilic
centre at C-4 has exerted only a modest effect on the com-
position of the anomeric mixture. We would retain protecting
groups on one or more of the other hydroxyl groups during any
subsequent processing, to attenuate this volatility.
We were pleased that simple ethers 11a–d underwent RCM
smoothly, and also that acetals 17d and 17f could be ring
closed, albeit slowly. The failure of acrylate ester 15 to undergo
RCM is interesting. There appear to be different fluorine substi-
tutent effects at the three oxidation levels: we believe these
effects are electronic rather than conformational. At the ether
level, there is no difference in reactivity between the fluorinated
and non-fluorinated systems, indicating that the fluorine atoms
exert no conformational effect. The fluorinated and non-
fluorinated acetals are different; 17d and 17f both ring-close
more slowly than comparable examples 35–39 from the
literature (Fig. 4).23,31–37
Fig. 5 Conformations and conformationally-dependent reactions of
acetals and esters.
In the favoured s-trans conformer, the precursor radical is
removed rapidly by reduction, which competes very effectively
with the conformational equilibrium. Atom transfer methods,39
in which the free radical precursor is constantly regenerated,
provide an effective solution, as does a change of oxidation state
to the acetal level. The n–σ* interactions in the acetals are much
weaker, so these more flexible precursor radicals interconvert
rapidly between conformers such as 44 and 45. Stork used this
property effectively in his eponymous lactonisation sequence
(the oxidation level is raised after radical cyclisation)40 and
Itoh41 used a similar indirect approach for the synthesis of
difluorolactones from bromodifluoroacetic acid esters, proving
that the conformational barriers in the acetal systems are
significantly lower.
For acetals 17d and 17f, which are significantly less reactive
than their non-fluorinated counterparts, we believe that 46 is
the key intermediate rather than 47. Two inductively electron-
withdrawing substituents at the allylic position will make the
intermediate metal alkylidene less reactive; Hoye and Zhao42
estimated that a single allylic alkoxy group slowed the form-
ation of a 5-membered ring by RCM by almost an order of
magnitude relative to an unsubstituted system. However, alkene
50 can be prepared by the homodimerisation of vinyl dioxolane
48 (though the reaction is slow), consistent with the utility of
both vinyl dioxolane and acrolein diethyl acetal in kinetic cross-
metathesis reactions with terminal alkenes,43 and indicating that
alkylidene 49 is sufficiently reactive to undergo intermolecular
metallocyclobutanation. Alkylidene 46 would therefore be
expected to be a competent (if sluggish) intermediate in an
intramolecular metallocyclobutanation reaction.44
Fig. 4 Related RCM reactions from the literature.
Terminal allyl groups are highly reactive in both RCM and
reversible cross metathetical homodimerisation reactions.
Ethers 33 and 34, and acetals 35–39 can all react via a common
mechanism in which rapid formation and turnover of a reactive
new alkylidene from a terminal allyl group precedes rapid six-
membered ring closure. Acrylate esters 40 and 41 can react by
the same mechanism.
Presumably, methallyl ether 14 fails to react for steric
reasons, while acrylate ester 15 cannot react with the catalyst to
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
532

View Article Online
spectrometers. High resolution mass spectrometry measure-
ments were carried out either on the Micromass LCT or
the Kratos Concept 1H spectrometers using peak matching
to suitable reference peaks, depending on the technique used.
Thin Layer Chromatography (TLC) was performed on pre-
coated aluminium silica gel plates supplied by E. Merck,
A.G. Darmstadt, Germany (Silica gel 60 F254, thickness
0.2 mm, Art. 1.05554) or on precoated plastic silica gel plates
supplied by Macherey-Nagel (Polygram® SIL G/UV254, thick-
ness 0.25 mm, Art. 805 023) or on precoated glass plates sup-
plied by Merck (Silica gel 60 F254, art. 1.05715). Visualisation
was achieved by UV light and/or potassium permanganate
stain. Flash column chromatography was performed using
silica gel (Fluorochem, Silica gel 60, 40–63µ, Art. 02050017) or
using a Biotage flash chromatography system. A Transsonic
T460/H sonicator was used for sonicated reactions. Ozone was
generated by a 500M Fischer Technology ozone generator.THF
was dried by refluxing with benzophenone over sodium wire
until a deep purple color developed and persisted, then distilled
and collected by dry syringe as required. Dichloromethane was
dried by refluxing with calcium hydride; it was distilled and
collected by dry syringe as required. All other chemicals were
used as received without any further purification. Where
required, solvents were degassed by bubbling argon or nitrogen
through them for at least 30 minutes.
form a competent metal alkylidene complex; the alkoxy-
carbonyl group is known to have an extremely deleterious effect
on the competence of the derived alkylidene. At the start of
the study, there were no examples of RCM reactions in which
fluorine atoms were located next to one of the alkenyl groups.
Now, Blechert45 and Grubbs44 have shown examples of cross
metathesis reactions that involve perfluoroalkyl-substituted
alkene components, though they are not reactive substrates for
the reaction. Indeed, perfluoroalkyl alkenes are considerably
less reactive than acrolein dialkyl acetals in cross metathesis
reactions. The key step in both reactions will involve metallo-
cyclobutanation of the first-formed metal alkylidene across the
perfluoroalkyl alkene and not the formation of a perfluoroalkyl
alkylidene complex which would be strongly deactivated by the
powerful inductive effect of the perfluoroalkyl group. Meta-
thesis reactions of perfluoroalkyl alkenes remain an area where
significant improvements in catalyst activity are required.46
Our outcomes show that RCM can be used to cyclise systems
like 17 smoothly to build highly-functionalised oxacycles con-
taining a CF2 centre, particularly when the second generation
catalysts are used, and that conventional dihydroxylation con-
ditions oxidise electron-deficient alkenes such as 13 effectively
and stereoselectively.
(Pivaloyloxy)acetaldehyde 9f
Trimethylacetyl chloride (200 mmol, 24.9 mL) was added
dropwise to a cold (0 ЊC) solution of Z-2-butene-1,4-diol
(100 mmol, 8.2 mL) and pyridine (200 mmol, 16.2 mL). The
mixture was allowed to warm to room temperature and stirred
for 18 hours. The reaction mixture was quenched with water
(100 mL) and extracted with DCM (3 × 100 mL). The com-
bined organic extracts were washed successively with HCl
(50 mL of a 1 M solution) and sodium hydrogencarbonate
(50 mL of a saturated aqueous solution), dried (MgSO4), fil-
tered and concentrated under reduced pressure to afford a pale
yellow oil (26.03 g) which was taken up in DCM (250 mL). The
solution was cooled to Ϫ78 ЊC and ozone was bubbled through
it until complete consumption of the starting alkene (TLC), by
which time a persistent blue color had developed in the solu-
tion. Nitrogen was bubbled through the solution to remove the
excess ozone and the reaction mixture was quenched with tri-
phenylphosphine (110 mmol, 28.85 g), allowed to warm to
room temperature and concentrated under reduced pressure to
afford a yellow oil (55.02 g), from which part of the triphenyl-
phosphine oxide crystallised. The remaining oil was decanted
from the solid and purified by short column chromatography
(50% diethyl ether in light petroleum) to afford the desired
aldehyde 9f as a colourless oil (10.24 g, 71% over two steps). Rf
(50% diethyl ether in light petroleum) 0.35; νmax (film)/cmϪ1
The target sugars have great potential for the construction of
probes and inhibitors of sugar processing enzymes, work which
continues in our laboratories.
Experimental
NMR spectra were recorded on Bruker ARX-250, Bruker
DPX-300, Bruker AC-300 or Bruker DRX-400 spectrometers.
1H and 13C NMR spectra were recorded using the deuterated
solvent as the lock and the residual solvent as the internal refer-
ence. 19F NMR spectra were recorded relative to chloro-
trifluoromethane as the external standard. The multiplicities of
the spectroscopic data are presented in the following manner:
app. = apparent, s = singlet, d = doublet, t = triplet, pent. =
pentet, q = quartet, m = multiplet and br = broad. The appear-
ance of complex signals is indicated by app. Homocouplings
(H–H, F–F) are given in Hertz and specified by J; the nuclei
involved in heteronuclear couplings are defined with the
observed nucleus given first. Unless stated otherwise, all refer
2975s (H–CO), 2874m (C–H), 2719w (H–CO), 1734s (C᎐O);
᎐
δH (250 MHz, CDCl3) 9.51 (1H, s, CHO), 4.57 (2H, s, H-1Ј),
1.20 (9H, s, ϪC(CH3)3); δC (63 MHz, CDCl3) 196.3, 178.3, 68.9,
39.1, 27.5. Spectral data were in agreement with those reported
by Grubbs et al.47
3
to J couplings. Carbohydrate numbering is used for the prod-
ucts of dihydroxylation reactions to simplify the reading of
the NMR data. Chemical ionisation (CI) mass spectra were
recorded on a Micromass Prospec or a Kratos Concept 1H
spectrometer using ammonia as the reagent gas. Electron
impact (EI) spectra were recorded on a Kratos MS-80, a
Micromass Prospec or a Kratos Concept 1H spectrometer. Fast
atom bombardment (FAB) spectra were recorded on a Kratos
Concept 1H spectrometer at about 7 kV using xenon and
m-nitrobenzyl alcohol as the matrix. GC-MS was carried out
on a Perkin Elmer TurboMass spectrometer fitted with a
Zebron ZB-5 column (30 m × 0.25 µm) running a 20–350 ЊC
ramp over 27 minutes. Electrospray (ES) mass spectra were
recorded on a Micromass LCT or a Micromass Quattro LC
General procedure for the preparation of difluoro homoallylic
alcohols 10a–10d
3-Bromo-3,3-difluoroprop-1-ene 8 (7.5 mmol, 0.76 mL) and the
aldehyde 9 (5 mmol) were added successively to a sonicated
suspension of indium powder (7.5 mmol, 0.86 g) in DMF
(15 mL). The mixture was sonicated for 3 to 5 hours at room
temperature. The reaction mixture was then quenched with HCl
(20 mL of a 1 M aqueous solution) and extracted with di-
chloromethane (3 × 25 mL). The combined organic extracts
were washed with brine (20 mL), dried (MgSO4), filtered and
concentrated under reduced pressure to afford pale yellow oils
which were purified by column chromatography.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
533

View Article Online
2,2-Difluoro-1-phenylbut-3-en-1-ol 10a
petroleum) 0.28; νmax (film)/cmϪ1 3431s br (O–H), 3064w
(Ar–H), 3031w (᎐C–H), 2872m (C–H), 1652w (C᎐C); δ (300
᎐
᎐
H
As above from benzaldehyde (5.0 mmol, 0.51 mL) with soni-
cation for 4 hours and purification by column chromatography
(15% diethyl ether in light petroleum) to afford 10a as a colour-
less oil (0.52 g, 57%, 100% by GC). Rf (15% diethyl ether in
light petroleum) 0.28; νmax (film)/cmϪ1 3410s br (O–H), 2980w
MHz, CDCl3) 7.40–7.28 (5H, m, –C6H5), 6.09–5.91 (1H, m,
H-4), 5.72 (1H, dt, Jtrans 17.6, 2J 2.6, 4JH–F 2.6, H-5a), 5.52 (1H,
d, Jcis 11.0, H-5b), 4.57 (2H, s, –OCH2Ph), 4.10–3.98 (1H, m,
2
2
H-2), 3.70 (1H, dd, J 9.9, J 3.3, H-1a), 3.59 (1H, dd, J 9.9,
J 7.0, H-1b), 2.76 (1H, d, J 5.2, –OH ); δC (75 MHz, CDCl3)
(᎐C–H), 1647w (C᎐C); δ (300 MHz, CDCl ) 7.44–7.35 (5H, m,
᎐
᎐
2
3
H
3
137.3, 129.9 (t, JC–F 25.5), 128.4, 127.9, 127.7, 120.9 (t, JC–F
9.5), 119.1 (t, 1JC–F 243.2), 73.6, 72.2 (t, 2JC–F 29.8), 68.6; δF (282
MHz, CDCl3) Ϫ107.3 (1F, dt, 2J 252.8, JF–H 10.3, 10.3), Ϫ111.3
–C6H5), 5.94–5.77 (1H, m, H-3), 5.59 (1H, d, Jtrans 17.3, H-4a),
5.47 (1H, d, Jcis 11.0, H-4b), 4.89 (1H, td, JH–F 10.6, J 4.1, H-1),
2.60–2.56 (1H, m, –OH ); δC (63 MHz, CDCl3) 136.4, 129.8 (t,
2JC–F 25.9), 129.1, 128.6, 128.0, 122.0 (t, 3JC–F 9.2), 120.0 (t, 1JC–F
244.4), 76.3 (t, 2JC–F 29.8); δF (282 MHz, CDCl3) Ϫ107.9 (1F, dt,
(1F, dt, J 252.8, JF–H 11.3, 11.3); [HRMS (ES, [M ϩ Na]ϩ)
2
Found: 251.0872. Calc. for C12H14F2O2Na: 251.0860]; m/z (EI)
228 (30%, Mϩ), 121 (100, Mϩ–OBn).
2J 246.8, JF–H 10.6, 10.6), Ϫ109.3 (1F, dt, J 246.8, JF–H 10.6,
2
10.6); m/z (ES) 207 (100%, [M ϩ Na]ϩ). Spectral data were in
agreement with those reported by Burton et al.16
1-Acetoxy-3,3-difluoropent-4-en-2-ol 10e
3-Bromo-3,3-difluoroprop-1-ene 8 (2.2 mmol, 0.22 mL) and
(acetoxy)acetaldehyde (2.0 mmol, 204 mg) were added to a
sonicated suspension of indium (2.2 mmol, 230 mg) in DMF
(4 mL). The mixture was sonicated at room temperature for
4.5 h, quenched with water (10 mL) and extracted with diethyl
ether (3 × 20 mL). The combined organic extracts were washed
with brine (20 mL), dried (MgSO4), filtered and concentrated
under reduced pressure to afford a yellow oil (373 mg). Purifi-
cation by column chromatography (40% diethyl ether in
light petroleum) afforded 10e as a colourless oil (233 mg,
65%, 98% by GC). Rf (40% diethyl ether in light petroleum)
2,2-Difluoro-1-phenylbut-3-en-1-ol 10a
As above from furfuraldehyde (5.0 mmol, 0.41 mL) to afford
10b as a pale yellow oil (1.24 g, 95%, 100% by GC); Rf (20%
diethyl ether in light petroleum) 0.27; νmax (film)/cmϪ1 3415s br
(O–H), 2923m (C–H), 1652w (C᎐C); δ (300 MHz, CDCl ) 7.41
᎐
H
3
(1H, dd, J 1.8, 4J 0.8, H-3Ј), 6.42 (1H, d, J 3.3, H-5Ј), 6.37 (1H,
dd, J 3.3, 1.8, H-4Ј), 6.04–5.87 (1H, m, H-2), 5.70 (1H, dtd,
Jtrans 17.3, 4JH–F 2.2, 2J 0.7, H-4a), 5.51 (1H, dd, Jcis 11.0, 2J 0.7,
3
H-4b), 4.88 (1H, td, JH–F 9.9, J 6.6, H-1), 2.98 (1H, d, J 6.6,
3
–OH ); δC (75 MHz, CDCl3) 149.5 (t, JC–F 3.1), 142.9, 129.5
0.29; νmax (film)/cmϪ1 3458s (O–H), 2974s (᎐C–H), 2875m
2
3
1
᎐
(t, JC–F 25.7), 121.8 (t, JC–F 9.3), 118.7 (t, JC–F 244.7), 110.6,
4
2
(C–H), 1718s (C᎐O), 1654 (C᎐C); δ (300 MHz, CDCl3)
᎐
᎐
H
109.5 (t, JC–F 1.7), 70.2 (t, JC–F 31.7); δF (282 MHz, CDCl3)
Ϫ108.8 (t, JF–H 11.4, 11.4); [HRMS (CI, [M ϩ NH4]ϩ) Found:
192.084434. Calc. for C8H12NO2F2: 192.083614]; m/z (CI) 192
(52%, [M ϩ NH4]ϩ), 174 (62), 157 (28), 139 (100).
3
4
6.01–5.89 (1H, m, H-4), 5.73 (1H, dt, Jtrans 17.3, JH–F 2.5,
3
2
H-5a), 5.55 (1H, d, Jcis 11.0, H-5b), 4.27 (1H, dd, JH–H 12.0,
3JH–H 3.3, H-1a), 4.17 (1H, dd, JH–H 12.0, JH–H 7.4, H-1b),
4.09–3.97 (1H, m, H-2), 3.08 (1H, br s, –OH ), 2.08 (3H, s,
–CH3); δC (75 MHz, CDCl3) 171.4, 129.4 (t, 2JC–F 25.4), 121.6 (t,
3JC–F 9.6), 118.9 (dd, 1JC–F 244.2, 243.0), 71.9 (t, 2JC–F 29.7), 63.5
(tdd, 3JC–F 4.5, 2.8), 20.7; δF (282 MHz, CDCl3) Ϫ107.6 (1F, dt,
2
3
2,2-Difluoro-1-furan-2Ј-ylbut-3-en-1-ol 10b
As above from furfuraldehyde (5.0 mmol, 0.41 mL) to afford
10b as a pale yellow oil (1.24 g, 95%, 100% by GC); Rf (20%
diethyl ether in light petroleum) 0.27; νmax (film)/cmϪ1 3415s br
2J 254.3, JF–H 8.9, 8.9), Ϫ111.8 (1F, dt, J 254.3, JF–H 11.4,
11.4); [HRMS (ES, [M ϩ Na]ϩ) Found: 203.0488, calc. for
C7H10O3F2Na: 203.0496]; m/z (ES) 203 (100, [M ϩ Na]ϩ).
3
2
3
(O–H), 2923m (C–H), 1652w (C᎐C); δ (300 MHz, CDCl ) 7.41
᎐
H
3
(1H, dd, J 1.8, 4J 0.8, H-3Ј), 6.42 (1H, d, J 3.3, H-5Ј), 6.37 (1H,
dd, J 3.3, 1.8, H-4Ј), 6.04–5.87 (1H, m, H-2), 5.70 (1H, dtd,
Jtrans 17.3, 4JH–F 2.2, 2J 0.7, H-4a), 5.51 (1H, dd, Jcis 11.0, 2J 0.7,
Attempted preparation of 10f in DMF; preparation of 10f and 22
3
H-4b), 4.88 (1H, td, JH–F 9.9, J 6.6, H-1), 2.98 (1H, d, J 6.6,
3-Bromo-3,3-difluoro-prop-1-ene 8 (5.5 mmol, 0.56 mL) and
aldehyde 9f (5.0 mmol, 0.721 g) were added successively to a
sonicated suspension of indium (5.5 mmol, 0.632 g) in DMF
(10 mL) at room temperature. The mixture was sonicated for
2 hours, then usual work-up and flash column chromatography
afforded alcohols 10f (0.643 g, 58%)* and 22 (0.149 g, 13%, GC
retention time 10.86 min), Rf (20% diethyl ether in light petrol-
3
–OH ); δC (75 MHz, CDCl3) 149.5 (t, JC–F 3.1), 142.9, 129.5
2
3
1
(t, JC–F 25.7), 121.8 (t, JC–F 9.3), 118.7 (t, JC–F 244.7), 110.6,
4
2
109.5 (t, JC–F 1.7), 70.2 (t, JC–F 31.7); δF (282 MHz, CDCl3)
Ϫ108.8 (t, JF–H 11.4, 11.4); [HRMS (CI, [M ϩ NH4]ϩ) Found:
192.084434. Calc. for C8H12NO2F2: 192.083614]; m/z (CI) 192
(52%, [M ϩ NH4]ϩ), 174 (62), 157 (28), 139 (100).
eum) 0.14, δH (300 MHz, CDCl ) 5.98–5.85 (1H, m, CH᎐CH ),
᎐
3
2
5.65 (1H, app. dt, Jtrans 17.4, 4JH–F2.3, CH᎐CH H ), 5.45 (1H, d,
3,3-Difluorooct-1-en-4-ol 10c
᎐
a
b
Jcis 11.1, CH᎐CH H ), 4.20 (1H, dd, 2J 11.8, J 3.8, CH H OH),
᎐
a
b
a
b
As above from valeraldehyde (5.0 mmol, 0.53 mL) to afford 10c
as a colourless oil (0.47 g, 57%); Rf (10% diethyl ether in
light petroleum) 0.22; νmax (film)/cmϪ1 3425s br (O–H), 1990w
2
4.10 (1H, dd, J 11.8, J 7.0, CHaHbOH), 4.03–3.92 (1H, m,
CHOPiv), 1.11 (9H, s, t-Bu) (the OH signal was not observed);
2
δC (75 MHz, CDCl3) 170.9, 129.6 (t, JC–F25.1), 121.3 (t,
(᎐C–H), 2915m (C–H), 1650w (C᎐C); δ (300 MHz, CDCl3)
᎐
᎐
H
3JC–F9.5), 119.0 (t, JC–F243.9), 71.7 (t, JC–F29.9), 63.3 (t, JC–F
1
2
3
4
6.02–5.88 (1H, m, H-2), 5.71 (1H, dtd, Jtrans 17.3, JH–F 4.8,
2J 1.1, H-1a), 5.53 (1H, d, Jcis 10.7, H-1b), 3.82–3.69 (1H, m,
H-4), 1.95 (1H, d, J 5.5, –OH ), 1.65–1.25 (6H, m, H-5, H-6 and
H-7), 0.91 (3H, t, J 7.0, H-8); δC (75 MHz, CDCl3) 129.7 (t,
2JC–F 26.0), 121.9 (dd, 1JC–F 246.1, 244.3), 121.5 (t, 3JC–F 9.6), 74.5
4.25), 38.6, 26.9, 26.8; δH (282 MHz, CDCl3) Ϫ107.5 (1F, dt,
2
2J 253.3, JH–F 9.9), Ϫ111.5 (1F, dt, J253.3, JH–F11.4); m/z (EI)
223 (3%, [M ϩ H]ϩ), 179 (10), 167 (10), 145 (70), 103 (100).
Further characterisation was not obtained for this unwanted
side product. *Characterisation of 10f follows.
2
3
(dd, JC–F 30.2, 25.6), 28.3 (dd, JC–F 4.3, 1.0), 27.7, 22.6, 14.0;
δF (282 MHz, CDCl3) Ϫ108.7 (1F, dt, 2J 248.7, JF–H 10.5, 10.5),
2
3,3-Difluoro-1-pivaloxyloxy-pent-4-en-2-ol 10f
Ϫ112.1 (1F, dt, J 248.7, JF–H 10.6, 10.6); m/z (ES) 187 (100%,
[M ϩ Na]ϩ). Spectral data were in agreement with those
reported by Seyferth et al.15
A mixture of 9f (7.21 g, 50.0 mmol), 3-bromo-3,3-difluoro-
prop-1-ene 8 (5.10 mL, 50.0 mmol) and indium powder (5.74 g,
50.0 mmol) in water (100 mL) was sonicated at room temper-
ature for 2 hours and then stirred vigorously overnight. The
reaction mixture was quenched with HCl (50 mL of a 1 M
solution) and extracted with diethyl ether (3 × 100 mL). The
combined organic extracts were washed with brine (50 mL),
1-Benzyloxy-3,3-difluoropent-4-en-2-ol 10d
As above from benzyloxyacetaldehyde (5.0 mmol, 0.70 mL)
with sonication for 6 hours to afford 10d as a colourless oil
(0.99 g, 96%, 100% by GC); Rf (30% diethyl ether in light
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
534

View Article Online
dried (MgSO4), filtered and concentrated under reduced pres-
sure to leave a yellow oil (10.31 g). Purification by column
chromatography (20% diethyl ether in light petroleum) afforded
10f as a pale yellow oil (9.09 g, 82%, 100% by GC, retention
time 10.78 min); Rf (20% diethyl ether in light petroleum) 0.27;
light petroleum) 0.43; νmax (film)/cmϪ1 3085w (᎐C–H), 2985w
᎐
(᎐C–H), 2924m (C–H), 2871m (C–H), 1649w (C᎐C); δ (300
᎐
᎐
H
4
MHz, CDCl3) 7.44 (1H, dd, J 1.5, J 0.8, H-3), 6.44 (1H, d,
J 2.9, H-5), 6.39 (1H, dd, J 2.9, 1.5, H-4), 6.13–5.95 (1H, m,
H-3Ј), 5.92–5.79 (1H, m, OCH2CH ), 5.67 (1H, dtd, Jtrans 17.3,
2J 2.6, 4JH–F 0.8, H-4Јa), 5.50 (1H, d, Jcis 11.0, H-4Јb), 5.30–5.19
νmax (film)/cmϪ1 3458s br (O–H), 2974s (᎐C–H), 2875m (C–H),
᎐
1718s (C᎐O), 1651w (C᎐C); δ (300 MHz, CDCl3) 6.07–5.90
(2H, m, OCH CH᎐CH ), 4.64 (1H, t, JH–F 9.2, H-1Ј), 4.10 (1H,
᎐
2 2
2 2
᎐
᎐
H
4
2
(1H, m, CH᎐CH ), 5.70 (1H, dtd, Jtrans, 17.3, JH–F 2.2, J 0.8,
dd, J 12.9, J 5.1, OCHaHbCH), 3.93 (1H, dd, J 12.9, J 6.3,
᎐
2
3
CH᎐CH H ), 5.55 (1H, d, J 11.0, CH᎐CH H ), 4.30–4.18
OCHaHbCH); δC (75 MHz, CDCl3) 148.6 (dd, JC–F 4.8, 1.4),
᎐
᎐
a
b
cis
a
b
2
3
(2H, m, CH2OPiv), 4.07–3.97 (1H, m, CHOH), 2.85 (1H,
br s, –OH ), 1.19 (9H, s, –C(CH3)3); δC (75 MHz, CDCl3)
143.2, 133.6, 130.0 (t, JC–F 25.2), 121.0 (t, JC–F 9.3), 118.3 (t
1JC–F 244.4), 118.2, 110.7 (t, 4JC–F 1.1), 110.5, 75.9 (t, 2JC–F 31.7),
70.6; δF (282 MHz, CDCl3) Ϫ107.7 (1F, dt, 2J 250.5, JF–H 10.5),
2
3
1
179.3, 129.8 (t, JC–F 25.4), 121.8 (t, JC–F 9.6), 119.2 (t, JC–F
2
3
2
243.9), 72.4 (t, JC–F 29.7), 63.6 (dd, JC–F 4.2, 3.1), 39.1, 27.3;
δF (282 MHz, CDCl3) Ϫ107.8 (1F, dt, 2J 253.0, 3JF–H 10.2, 10.2),
Ϫ105.0 (1F, dt, J 250.5, JF–H 9.9); [HRMS (ES, [M ϩ Na]ϩ)
Found 237.0709, calc. for C11H12O2F2Na: 237.0703]; m/z (ES)
Ϫ111.6, (1F, dt, 2J 253.0, JF–H 12.7, 12.7); [HRMS (FAB,
237 (100%, [M ϩ Na]ϩ).
3
[M ϩ H]ϩ) Found: 223.11454, calc. for C10H17O3F2: 223.11458];
m/z (ES) 245 (100, [M ϩ Na]ϩ).
4-Allyloxy-3,3-difluorooct-1-ene 11c
Alcohol 10c (1.5 mmol, 246 mg), allyl bromide (1.8 mmol,
0.16 mL), sodium hydroxide (10.5 mmol, 0.55 mL of a 50 wt%
aqueous solution) and tetra-n-butylammonium hydrogen-
sulfate (75 µmol, 26 mg) were treated as described above. The
mixture was quenched with ammonium chloride (10 mL a
saturated aqueous solution) and extracted with diethyl ether
(3 × 15 mL). The combined organic extracts were washed with
water (10 mL), dried (MgSO4), filtered and concentrated under
reduced pressure to afford crude ether 11c as a pale yellow oil
(301 mg, 98%, 96% by GC); Rf (5% diethyl ether in light petrol-
General procedure for the preparation of difluorohomoallylallyl
ethers 11a–11d
Mixtures of difluorohomoallylic alcohols 11a–11d, allyl bromide
(1.2 eq.), sodium hydroxide (7 eq. of a 50 wt% aqueous solution)
and tetra-n-butylammonium hydrogensulfate (5 mol% based on
the alcohol) were stirred at 0 ЊC for 30 minutes, allowed to warm
to room temperature and stirred overnight at this temperature.
The reaction mixtures were quenched with a saturated aqueous
solution of ammonium chloride and extracted with diethyl ether.
The combined organic extracts were washed with water, dried
(MgSO4), filtered and concentrated under reduced pressure to
afford pale yellow oils, which were used for RCM without further
purification.
eum) 0.62; νmax (film)/cmϪ1 2990w (᎐C–H), 2871s (C–H), 1649w
᎐
(C᎐C); δ (300 MHz, CDCl ) 6.04–5.81 (2H, m, CH᎐CH ), 5.65
᎐
᎐
H
3
2
(1H, dddd, Jtrans 17.6, 4JH–F 2.9, 1.8, 2J 1.1, H-1a), 5.47 (1H, dt,
4
Jcis 11.0, JF–H 2.2, H-1b), 5.28–5.13 (2H, m, OCH CH᎐CH ),
᎐
2
2
4.24 (1H, dd, 2J 12.4, J 5.0, OCHaHbCH), 4.03 (1H, dd, 2J 12.4,
J 6.2, OCHaHbCH), 3.50–3.41 (1H, m, H-4), 1.54–1.22 (6H, m,
H-5, H-6 and H-7), 0.87 (3H, t, J 7.7, H-8); δC (75 MHz,
1-Allyloxy-2,2-difluoro-1-phenylbut-3-ene 11a
2
1
Alcohol 10a (1.5 mmol, 276 mg), allyl bromide (1.8 mmol,
0.16 mL), sodium hydroxide (10.5 mmol, 0.55 mL of a 50 wt%
aqueous solution) and tetra-n-butylammonium hydrogen-
sulfate (75 µmol, 26 mg) were treated as described above. The
mixture was quenched with ammonium chloride (10 mL of a
saturated aqueous solution) and extracted with diethyl ether
(3 × 15 mL). The combined organic extracts were washed with
water (10 mL), dried (MgSO4), filtered and concentrated under
reduced pressure to afford crude ether 11a as a pale yellow oil
(330 mg, 100%, 99% by GC); Rf (5% diethyl ether in light
CDCl3) 134.6, 130.1 (t, JC–F 25.7), 121.0 (dd, JC–F 245.9,
244.2), 120.5 (t, JC–F 9.6), 117.4, 80.6 (dd, JC–F 30.5, 26.0),
73.5, 29.6 (dd, JC–F4.5, 1.1), 27.8, 22.6, 14.0; δF (282 MHz,
3
2
3
2
CDCl3) Ϫ102.6 (1F, dt, J 250.5, JF–H 10.2, 10.2), Ϫ108.8 (1F,
ddd, 2J 250.5, JF–H 14.0, 6.4); [HRMS (ES, [M ϩ Na]ϩ) Found:
227.1219. Calc. for C11H18OF2Na: 227.1223]; m/z (ES) 227
(100%, [M ϩ Na]ϩ).
2-Allyloxy-1-benzyloxy-3,3-difluoropent-4-ene 11d
petroleum) 0.49; νmax (film)/cmϪ1 2985m (᎐C–H), 2867m (C–H),
Alcohol 10d (4.4 mmol, 993 mg), allyl bromide (6.5 mmol,
0.56 mL), sodium hydroxide (30.5 mmol, 1.60 mL of a 50 wt%
aqueous solution) and tetra-n-butylammonium hydrogen-
sulfate (220 µmol, 74 mg) were treated as described above. The
mixture was quenched with ammonium chloride (25 mL of a
saturated aqueous solution) and extracted with diethyl ether
(3 × 25 mL). The combined organic extracts were washed with
water (25 mL), dried (MgSO4), filtered and concentrated under
reduced pressure to afford crude ether 11d as a pale yellow oil
(1.16 g, 100%, 100% by GC); Rf (5% diethyl ether in light
petroleum) 0.24; νmax (film)/cmϪ1 3066w (Ar–H), 3029 m
᎐
1651w (C᎐C); δ (300 MHz, CDCl ) 7.38–7.35 (5H, m, –C H ),
᎐
H
3
6
5
6.07–5.93 (2H, m, H-3 and OCH2CH ), 5.57 (1H, d, Jtrans 17.3,
H-4a), 5.45 (1H, d, Jcis 11.0, H-4b), 5.31–5.18 (2H, m, OCH2-
2
CH᎐CH ), 4.59 (1H, t, J
10.3, H-1), 4.10 (1H, ddd, J 13.0,
᎐
2
H–F
J 6.3, 4J 1.1, H-1Јa), 3.92 (1H, ddd, 2J 13.0, J 4.8, 4J 1.1, H-1Јb);
δC (75 MHz, CDCl3) 134.8, 134.0, 130.2 (t, JC–F 25.4), 128.8,
128.6 (t, JC–F 1.1), 128.2, 120.9 (t, JC–F 9.3), 119.0 (t, JC–F
2
4
3
1
2
244.2), 117.7, 82.1 (t, JC–F 30.0), 70.5; δF (282 MHz, CDCl3)
Ϫ104.3 (1F, dt, 2J 248.6, JF–H 10.2, 10.2), Ϫ109.0 (1F, dt,
2J 248.6, JF–H 11.4, 11.4); [HRMS (CI, [M ϩ NH4]ϩ) Found:
242.135869. Calc. for C13H18N1O1F2: 242.135646]; m/z (CI) 242
(16%, [M ϩ NH4]ϩ), 147 (100).
(᎐C–H), 2920m (C–H); 2870m (C–H), 1649w (C᎐C); δ (300
᎐
᎐
H
MHz, CDCl3) 7.40–7.29 (5H, m, –C6H5), 6.10–5.88 (2H, m, H-4
and H-2Ј), 5.70 (1H, d, Jtrans 17.3, H-5a), 5.50 (1H, d, Jcis 11.0,
H-5b), 5.32 (1H, dt, Jtrans 17.1, 2J 1.5, J 1.5, OCH CH᎐CH H ),
᎐
2
a
b
1-Allyloxy-2,2-difluoro-1-(2Ј-furyl)but-3-ene 11b
5.22 (1H, d, Jcis 10.3, OCH CH᎐CH H ), 4.57 (2H, s, –OCH -
᎐
2
a
b
2
Alcohol 10b (1.5 mmol, 261 mg), allyl bromide (1.8 mmol,
0.16 mL), sodium hydroxide (10.5 mmol, 0.55 mL of a 50 wt%
aqueous solution) and tetra-n-butylammonium hydrogen-
sulfate (75 µmol, 26 mg) were treated as described above. The
mixture was quenched with ammonium chloride (10 mL of a
saturated aqueous solution) and extracted with diethyl ether
(3 × 15 mL). The combined organic extracts were washed with
water (10 mL), dried (MgSO4), filtered and concentrated under
reduced pressure to afford crude ether as a pale yellow oil
11b (319 mg, 99%, 97% by GC); Rf (5% diethyl ether in
Ph), 4.27 (2H, d, J 5.5, OCH2CH), 3.88–3.79 (1H, m, H-2), 3.73
(1H, dt, 2J 9.9, J 1.5, 4JH–F 1.5, H-1a), 3.59 (1H, dd, 2J 9.9, J 7.5,
2
H-1b); δC (75 MHz, CDCl3) 138.0, 134.5, 130.3 (t, JC–F 25.4),
3
1
128.5, 127.7, 127.6, 120.6 (t, JC–F 9.6), 119.5 (t, JC–F 244.7),
117.6, 79.8 (dd, 2JC–F 29.1, 28.0), 73.6, 73.4, 69.6 (dd, 3JC–F 5.4,
2.5); δF (282 MHz, CDCl3) Ϫ103.8 (1F, dt, J 254.3, JF–H 10.2,
2
2
10.2), Ϫ109.1 (1F, dt, J 254.3, JF–H 11.1, 11.1); [HRMS (EI,
Mϩ) Found: 268.12754. Calc. for C15H18F2O2: 268.12749]; m/z
(ES) 268 (25%, Mϩ), 227 (58, Mϩ–C3H5), 177 (49, Mϩ–CH2Ph),
91 (100, CH2Phϩ).
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
535

View Article Online
General procedure for the preparation of dihydropyrans 12a–12d
Purification by column chromatography (20% diethyl ether in
light petroleum) afforded 12d as a colourless oil (350 mg, 97%).
Rf (20% diethyl ether in light petroleum) 0.30; νmax (film)/cmϪ1
Solutions of ethers 11a–11d and Grubbs’ catalyst benzylidene
bis-(tricyclohexylphosphino)dichloro ruthenium (5 mol%) in
degassed DCM were stirred for 24 hours at room temperature,
then oxygen was bubbled through the reaction mixture for
15 minutes. Concentration under reduced pressure left black
oils, which were purified by column chromatography to afford
the desired difluorinated dihydropyrans 12a–12d.
3058w (Ar–H), 2977m (᎐C–H), 2865m (C–H), 1641w (C᎐C);
᎐
᎐
δH (300 MHz, CDCl3) 7.33–7.27 (5H, m, –C6H5), 6.27 (1H, ddd,
J 10.7, JH–F 3.7, 1.8, H-4), 5.95–5.86 (1H, m, H-5), 4.67 (1H,
2
2
d, J 12.1, –OCHaHbPh), 4.59 (1H, d, J 12.1, –OCHaHbPh),
4.37–4.18 (2H, m, H-6), 3.98–3.94 (1H, m, H-2), 3.90 (1H,
dd, 2J 10.9, J 2.4, –CHaHbOBn), 3.71 (1H, dd, 2J 10.9, J 7.9, –
3
CHaHbOBn); δC (75 MHz, CDCl3) 137.8, 135.7 (t, JC–F 9.0),
3,3-Difluoro-2-phenyl-3,6-dihydro-2H-pyran 12a
128.5, 127.9, 121.4 (t, 2JC–F 28.0), 113.9 (dd, 1JC–F 243.0, 235.7),
2
76.8 (t, JC–F 30.5), 73.8, 67.2, 65.2; δF (282 MHz, CDCl3)
Ether 11a (0.75 mmol, 167 mg) and Grubbs’ catalyst (75 µmol,
62 mg) were treated as described above in DCM (15 mL). Puri-
fication by column chromatography (5% diethyl ether in light
petroleum) afforded 12a as a colourless oil (118 mg, 81%). Rf
(5% diethyl ether in light petroleum) 0.30; νmax (film)/cmϪ1
2
4
Ϫ105.6 (1F, dddd, J 273.4, JF–H 17.8, 8.9, JF–H 8.9), Ϫ107.7
(1F, ddd, J 273.4, JF–H 6.4, 3.8); [HRMS (ES, [M ϩ Na]ϩ)
Found: 263.0853. Calc. for C13H14N1O2F2Na: 263.0860]; m/z
(ES) 263 (100%, [M ϩ Na]ϩ).
2
3
2990w (᎐C–H), 2865m (C–H), 1648w (C᎐C); δ (300 MHz,
᎐
᎐
H
CDCl3) 7.53–7.38 (5H, m, –C6H5), 6.30 (1H, ddd, J 10.3, 4JH–F
3.3, 1.9, H-4), 6.05 (1H, br app. t, J, JH–F10.3, H-5), 4.70 (1H,
dd, 3JH–F 18.7, 2.6, H-2Ј), 4.52–4.29 (2H, m, H-6); δC (75 MHz,
5-(Benzyloxy)methyl-4,4-difluoro-2,3-dihydroxytetrahydro-
pyrans 13d
3
3
A solution of NMO (0.66 mmol, 0.078 g) in water (0.1 mL) was
added to a cold (0 ЊC) solution of 12d (0.33 mmol, 0.068 g) in
acetone (0.35 mL) and tert-butanol (0.35 mL) in a screw-
capped vial. The mixture was stirred for 10 minutes and
osmium tetroxide (0.007 mmol, 0.1 mL of a 2.5 wt% solution in
tert-butanol, 2 mol%) was added. The reaction mixture was
stirred at 0 ЊC for 2 hours, allowed to warm to room temper-
ature and stirred at this temperature for 7 days. It was then
quenched with sodium sulfite (3.5 mmol, 0.44 g), diluted with
water (3 mL), stirred for an additional 30 minutes and extracted
with ethyl acetate (3 × 5 mL). The combined organic extracts
were dried (MgSO4), filtered and concentrated under reduced
pressure to afford 13d, a separable mixture of cis- and trans-
diastereoisomers, as a yellow oil. Column chromatography
(40% ethyl acetate in light petroleum) afforded in order of
elution: trans (( )-2S,3S,5R) 13d as a powder (0.02 g, 31%)
Rf (40% ethyl acetate in light petroleum) 0.25; GC retention
time 20.49 min; mp 101–102 ЊC; Found: C, 57.22; H, 6.12. Calc.
for C13H16F2O4: C, 56.93; H, 5.88%. δH (400 MHz, CDCl3)
CDCl3) 135.2 (dd, JC–F 9.6, 8.5), 133.8 (t, JC–F 1.1), 128.7,
128.2, 127.8, 122.3 (dd, 2JC–F 30.5, 26.6), 113.7 (dd, 1JC–F 245.6,
2
4
234.3), 78.9 (dd, JC–F 31.1, 25.4), 65.9 (t, JC–F 1.1); δF (282
MHz, CDCl3) Ϫ102.2 (1F, dddd, 2J 274.3, JF–H 18.7, 8.8, 4JF–H
7.6), Ϫ106.5 (1F, br d, J 274.3); [HRMS (ES, [M ϩ Na]ϩ)
2
Found: 219.0596. Calc. for C11H10N1OF2Na: 219.0598]; m/z
(EI) 196 (30%, Mϩ), 105 (74), 90 (100), 77 (40).
3,3-Difluoro-2-(2Ј-furyl)-3,6-dihydro-2H-pyran 12b
Ether 11b (0.75 mmol, 161 mg) and Grubbs’ catalyst (37 µmol,
31 mg) were treated as described above in DCM (15 mL). Puri-
fication by column chromatography (10% diethyl ether in light
petroleum) afforded 12b as a colourless oil (97 mg, 69%). Rf
(10% diethyl ether in light petroleum) 0.35; νmax (film)/cmϪ1
3079w (᎐C–H), 2986w (᎐C–H), 2925m (C–H), 1647w (C᎐C);
᎐
᎐
᎐
4
δH (300 MHz, CDCl3) 7.47 (1H, dd, J 1.8, J 0.7, H-3Ј), 6.55
(1H, d, J 3.3, H-5Ј), 6.41 (1H, dd, J 3.3, 1.8, H-4Ј), 6.29 (1H, dt,
J 10.3, JH–F 2.6, H-4), 6.07–5.98 (1H, m, H-5), 4.84 (1H, dd,
JH–F 14.0, 5.9, H-2), 4.37–4.32 (2H, m, H-6); δC (75 MHz, CDCl3)
147.1, 143.2, 135.2 (t, JC–F 9.3), 121.8 (dd, JC–F 29.1, 27.4),
113.2 (dd, JC–F 243.3, 237.7), 110.5, 110.4 (t, JC–F 1.7), 73.2
2
4.66 (1H, d, one half of an AB q, J 12.0, OCHaCHbPh), 4.57
3
2
(1H, d, one half of an AB q, 2J 12.0, OCHaCHbPh), 4.11
1
4
3
3
(1H, ddd, J 3.3, JH–Feq 6.9, JH–Fax 4.6, H-3), 4.02 (1H, ddd,
3JH–Fax 24.3, J 7.8, 2.5, H-5), 3.99–3.92 (1H, m, H-2), 3.89 (1H,
ddd, 2J 11.0, J 5.0, 5JH–F2.2, H-1eq), 3.84 (1H, ddd, 2J 11.0, J 2.5,
4JH–F1.5, H-6eq), 3.67 (1H, dd, 2J 11.0, J 7.8, H-6ax), 3.55 (1H, t, 2J,
J11.0, H-1ax), 3.15(1H, br, OH), 2.58(1H, brs, OH);δC (75MHz,
2
4
(dd, JC–F 36.6, 31.1), 65.2 (t. JC–F2.0); δF (282 MHz, CDCl3)
2
4
Ϫ102.9 (1F, dddd, app. d pent., J 274.5, JF–H 14.0, 10.2, JF–H
5.1), (Ϫ103.4)-(Ϫ103.5) and (Ϫ104.4)-(Ϫ104.5) (1F, m includ-
ing 2J 274.5); [HRMS (ES, [M ϩ Na]ϩ) Found: 209.0387. Calc.
for C9H8O2F2Na: 209.0390]; m/z (ES) 209 (100%, [M ϩ Na]ϩ).
1
CDCl3) 137.5, 128.5, 127.9 (2 signals), 117.9 (dd, JC–F254.9,
250.1), 73.8, 73.2 (dd, 2JC–F27.5, 21.6), 69.5 (dd, 2JC–F34.7, 20.3),
66.8 (d, 3JC–F4.7), 66.1 (d, 3JC–F4.7), 65.4; δF (376 MHz, CDCl3)
2-Butyl-3,3-difluoro-3,6-dihydro-2H-pyran 12c
2
2
Ϫ116.2 (1F, dd, J 260.5, JF–H6.9), Ϫ124.5 (1F, dd, J 260.5,
JF–H24.3); m/z (EI) 274 (10%, Mϩ), 107 (60, [OCH2Ph]ϩ), 91
(100, PhCH2ϩ) and cis (( )-2R,3R,5R) 13d as a powder (0.02 g,
31%), mp 96–97 ЊC; Rf (40% ethyl acetate in light petroleum)
0.11; GC retention time 20.26min; δH (400 MHz, CDCl3)
Ether 11c (0.75 mmol, 153 mg) and Grubbs’ catalyst (37 µmol,
31 mg) were treated as described above in DCM (15 mL). Puri-
fication by column chromatography (5% diethyl ether in light
petroleum) afforded 12c as a colourless oil (73 mg, 56%). Rf (5%
diethyl ether in light petroleum) 0.43; νmax (film)/cmϪ1 2980w
2
4.65 (1H, d, one half of an AB q, J 12.0, OCHaCHbPh), 4.58
(᎐C–H), 2890m (C–H), 1640w (C᎐C); δ (300 MHz, CDCl3)
᎐
᎐
H
(1H, d, one half of an AB q, 2J 12.0, OCHaCHbPh), 4.15
6.22 (1H, ddd, J 10.3, JH–F 3.3, 1.9, H-4), 5.94–5.85 (1H, m,
2
5
(1H, ddd, J 12.9, J 2.0, JH–F2.2, H-1eq), 4.02–3.97 (1H, m,
3
H-5), 4.30–4.07 (2H, m, H-6), 3.52 (1H, ddt, JH–F 17.8, 9.2,
2
H-2), 3.89 (1H, br dd, J 10.8, J 2.0, H-6ax), 3.90–3.70 (1H, br
d, contains JH–Fax 21.5, H-3*), 3.75 (1H, br dd, J 10.8, J 7.7,
J 3.6, H-2), 1.82–1.31 (6H, m, H-1Ј, H-2Ј and H-3Ј), 0.91 (3H, t,
3
2
J 7.2, H-4Ј); δC (75 MHz, CDCl3) 135.4 (t, 3JC–F 9.3), 122.0 (dd,
3
H-6eq), 3.80–3.65 (1H. br m, contains JH–Fax 21.5, H-5*), 3.65
1
2
2JC–F 29.7, 27.4), 114.6 (dd, JC–F 243.6, 234.6), 77.4 (dd, JC–F
(1H, br dd, 2J 12.9, J 1.5, H-1ax), 3.15 (1H, br, OH), 2.35
(1H, br OH); δC (75 MHz, CDCl3) 137.4, 128.5, 128.0, 127.9,
4
31.1, 26.0), 65.2 (t, JC–F 2.0), 27.6, 26.6, 22.6, 14.0; δF (282
2
4
MHz, CDCl3) Ϫ106.4 (1F, dddd, J 271.5, JF–H9.0, 8.2, JF–H
1
2
117.6 (t, JC–F252.1, 250.1), 78.6 (dd, JC–F27.4, 22.8), 73.9,
7.6), Ϫ108.0 (1F, br d, J 271.5,); [HRMS (ES, [M ϩ Na]ϩ)
2
2
3
70.3, 70.1 (dd, JC–F39.4, 19.2), 69.1(d, JC–F7.1), 66.6 (d,
Found: 199.0906. Calc. for C9H14OF2Na: 199.0910]; m/z (ES)
3JC–F7.8), 65.4; δF (376 MHz, CDCl3) Ϫ114.6 (1F, dt, J 253.3,
2
199 (100%, [M ϩ Na]ϩ).
JF–H5.8), Ϫ129.1 (1F, dt, 2J 253.5, JF–H 21.5); [HRMS (EI, Mϩ)
274.10163. Calc. for C13H16F2O4: 274.10167]; m/z (EI) 274 (3%,
Mϩ), 107 (60, [OCH2Ph]ϩ), 91 (100, PhCH2ϩ). *Signals for H-3
and H-5 in the 1H NMR spectrum are very broad because both
containalargediaxialH–Fcoupling. Thoughwewerenotableto
2-(Benzyloxy)methyl-3,3-difluoro-3,6-dihydro-2H-pyran 12d
Ether 11d (1.5 mmol, 403 mg,) and Grubbs’ catalyst (75 µmol,
62 mg,) were treated as described above in DCM (30 mL).
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
536

View Article Online
assign these signals accurately using 2D NMR methods,
integration of the 1D NMR spectrum yields the correct
number of protons, an appropriate HSQC pattern was obtained
and the presence of a 7.7 Hz coupling (to H-6) could be
detected in the 1H{19F} NMR spectrum for the signal assigned
as H-5.
6,6-Difluoro-3-ethoxy-5-(benzyloxy)methyl-4-oxa-1,7-octadiene
17d
A solution of alcohol 10d (4.4 mmol, 1.00 g), acrolein diethyl
acetal (22 mmol, 3.4 mL) and PPTS (0.44 mmol, 0.125 g) in dry
toluene (12.5 mL) was stirred at 30–40 ЊC under reduced pres-
sure (60 mmHg) for 14 hours to remove ethanol by azeotropic
distillation. The reaction mixture was quenched with sodium
carbonate (13 mL of a saturated aqueous solution) and
extracted with diethyl ether (3 × 10 mL). The combined organic
extracts were washed with brine (5 mL), dried (MgSO4), filtered
and concentrated under reduced pressure to leave a pale yellow
oil. Purification by column chromatography (5% diethyl ether
in light petroleum) afforded 17d (an inseparable 1 : 1 mixture of
diastereoisomers) as a clear oil (0.85 g, 63%), Rf 0.4; GC reten-
1-(Benzyloxy)methyl-3,3-difluoro-2-(methallyloxy)pent-4-ene 14
A mixture of 10d (2.0 mmol, 456 g), methallyl chloride (2.2
mmol, 0.22 mL), sodium hydroxide (14.0 mmol, 0.75 mL of a
50 wt% aqueous solution), tetra-n-butylammonium hydro-
gensulfate (100 µmol, 34 mg) and tetra-n-butylammonium
iodide (100 µmol, 37 mg) was stirred at 0 ЊC for 30 minutes,
allowed to warm to room temperature and stirred overnight at
this temperature. The reaction mixture was quenched with
ammonium chloride (10 mL of a saturated aqueous solution)
and extracted with diethyl ether (3 × 20 mL). The combined
organic extracts were washed with water (15 mL), dried
(MgSO4), filtered and concentrated under reduced pressure
to afford 14 as a pale yellow oil (0.56 g, 99%), which was
used without further purification. Rf (5% diethyl ether in
light petroleum) 0.57; νmax (film)/cmϪ1 3065w (Ar–H), 3030m
(᎐C–H), 2916m (C–H), 2808m (C–H), 1654w (C᎐C); δ (300
tion times 16.18, 16.21min; νmax (film)/cmϪ1 2977m (᎐C–H),
᎐
2934w (C–H), 1652w (C᎐C); δ (300 MHz, CDCl3) mixture:
᎐
H
6.08–5.80 (2H, m, –CH᎐CH ), 5.72–5.10 (4H, m, –CH᎐CH ),
᎐
᎐
2
2
4.61–4.47 (2H, m, –OCH2Ph), 4.13–3.98 (1H, m, CF2CH ),
3.74–3.46 (4H, m, –CHCH2O, OCH2CH3), 1.20–1.18 (3H,
m, CH2CH3); δC (376 MHz, CDCl3) mixture: 138.9, 138.3,
2
130.6 (app. q, JC–F 19.8, 2 signals), 128.9, 128.3, 128.2
3
(2 signals), 128.0 (2 signals), 121.0 (t, JC–F 9.5), 120.7 (t,
1
1
3JC–F 9.5), 119.8 (t, JC–F 243.5), 119.5 (t, JC–F 252.5), 119.2,
᎐
᎐
H
MHz, CDCl3) 7.41–7.28 (5H, m, –C6H5), 6.05 (1H, ddt,
119.0, 103.4, 102.8, 77.0 (t, 2JC–F 29.3), 76.0 (t, 2JC–F 29.3), 73.9
3
2
Jtrans 17.3, JH–F 12.9, Jcis 11.0, H-4), 5.72 (1H, d, Jtrans 17.3,
(2 signals), 72.8 (t, JC–F 28.8, 2 signals), 70.0 (br), 69.1 (br),
H-5a), 5.51 (1H, d, Jcis 11.0, H-5b), 5.04 (1H, s, C(Me)᎐
61.7, 61.4, 15.6, 15.5; δF (376 MHz, CDCl3) diastereoisomer 1:
᎐
3
CH H ), 4.95 (1H, s, C(Me)᎐CH H ), 4.58 (2H, s, –OCH Ph),
Ϫ104.2 (1F, dt, 2J 252.1, JF–H 9.0, 9.0), Ϫ109.3 (1H, ddd,
᎐
a
b
a
b
2
3
4.18 (2H, s, H-1Ј), 3.90–3.80 (1H, m, H-2), 3.75 (1H,
2J 252.1, JF–H 11.3, 8.1), diastereoisomer 2: Ϫ104.3 (1F, dt,
2
4
3
3
ddd, J 10.5, J 2.9, JH–F 1.8, H-1a), 3.61 (1H, ddd, 2J 10.5,
2J 253.7, JF–H 9.6), Ϫ109.4 (1F, dt, 2J 253.7, JF–H 11.3);
[HRMS (EI, Mϩ) Found: 312.15383. Calc. for C17H22F2O3:
312.15370]; m/z (EI, 20eV) 312 (20%, Mϩ), 311 (20, [M–H]ϩ),
285 (100, [M–CH᎐CH ]ϩ).
J 7.0, 4JH–F 1.1, H-1b), 1.79 (3H, s, –CH3); δC(63 MHz, CDCl3)
2
142.3, 138.4, 130.8 (t, JC–F 25.2), 128.8, 128.1, 128.0, 120.8 (t,
3JC–F 9.7), 119.9 (t, 1JC–F 244.9), 113.4, 80.3 (t, 2JC–F 28.7), 76.7,
᎐
2
3
74.0, 70.0 (dd, JC–F 5.6, 2.5), 19.9; δF (282 MHz, CDCl3)
2
3
Ϫ103.3 (1F, dt, J 254.3, JF–H 10.2, 10.2), Ϫ108.9 (1F, ddd,
6,6-Difluoro-3-ethoxy-5-(pivaloyloxy)methyl-4-oxa-1,7-
octadiene 17f
2J 254.3, JF–H 12.1, 9.5); [HRMS (ES, [M ϩ Na]ϩ) Found:
3
305.1326, calc. for C16H20O2F2Na: 305.1329]; m/z (ES) 305
A solution of alcohol 10f (85.0 mmol, 18.90 g), acrolein diethyl
acetal (425 mmol, 65 mL) and PPTS (8.5 mmol, 2.14 g) in dry
toluene (250 mL) was stirred at 30–40 ЊC under reduced pres-
sure (60 mmHg) for 9 hours to remove ethanol by azeotropic
distillation. The reaction mixture was quenched with sodium
carbonate (250 mL of a saturated aqueous solution) and
extracted with diethyl ether (3 × 200 mL). The combined
organic extracts were washed with brine (100 mL), dried
(MgSO4), filtered and concentrated under reduced pressure to
leave a pale yellow oil (29.71 g). Purification by column chrom-
atography (5% diethyl ether in light petroleum) afforded an
inseparable diastereoisomeric mixture (1 : 1) of the desired
acetal as a colourless oil 17f (14.67 g, 56%) along with starting
alcohol 10f (36.9 mmol, 8.20g), which was reacted with acrolein
diethyl acetal (184.4 mmol, 28.0 mL) and PPTS (3.7 mmol,
0.93 g) in toluene (100 mL) as described above. After work-up
and purification, products of the two different batches were
combined to afford an inseparable diastereoisomeric mixture
(1 : 1) of the desired acetal 17f as a pale yellow oil (18.92 g, 73%
total yield, 100% by GC). Rf (5% diethyl ether in light petrol-
(100%, [M ϩ Na]ϩ).
2-(Acryloyloxy)-1-(benzyloxymethyl)-2,2-difluorobut-3-ene 15
Freshly-distilled acryloyl chloride (4.7 mmol, 0.38 mL) was
added dropwise to a cold (0 ЊC) solution of 10d (4.7 mmol, 1.06
g) and triethylamine (4.7 mmol, 0.65 mL) in DCM (20 mL)
under a nitrogen atmosphere. The reaction mixture was stirred
at 0 ЊC for 2 hours and at room temperature for 16 hours. The
reaction mixture was quenched with water (20 mL) and
extracted with DCM (3 × 20 mL). The combined organic
extracts were dried (MgSO4), filtered and concentrated under
reduced pressure to leave a pale yellow oil (1.33 g). Purification
by column chromatography (10% diethyl ether in light petrol-
eum) afforded 15 as a colourless oil (0.66 g, 50%, 100% by GC).
Rf (10% diethyl ether in light petroleum) 0.43; νmax(film)/cmϪ1
3065w (Ar–H), 3032w (Ar–H), 2960m (᎐C–H), 2932m (C–H),
᎐
2872m (C–H), 1738s (C᎐O); δ (300 MHz, CDCl3); 7.37–7.28
᎐
H
2
(5H, m, –C6H5), 6.49 (1H, dd, Jtrans 17.3, J 1.1, C(᎐O)CH᎐
᎐
᎐
CHaHb), 6.18 (1H, dd, Jtrans 17.3, Jcis 10.3, C(᎐O)CH᎐CH ),
᎐
᎐
2
6.01–5.84 (2H, m, C(᎐O)CH᎐CH H and H-3), 5.71 (1H, dt,
eum) 0.25; νmax (film)/cmϪ1 2977m (᎐C–H), 2934w (C–H), 1735s
᎐
᎐
᎐
a
b
4
Jtrans 17.3, JH–F 2.2, H-4a), 5.56–5.45 (2H, m, H-1 and H-4b),
(C᎐O), 1652w (C᎐C); δ (300 MHz, CDCl ) mixture: 6.15–5.28
᎐ ᎐
H 3
4.58 (1H, d, 2J 12.1, –OCHaHbPh), 4.50 (1H, d, 2J12.1,
(6H, m, CH᎐CH , CH᎐CH ), 5.13–5.08 (1H, m, CH(OEt)O),
᎐
᎐
2
2
2
–OCHaHbPh), 3.81 (1H, dd, J 11.4, J 3.1, –CHaHbOBn), 3.71
4.37–4.23 (1H, m, CHCH2), 4.18–4.03 (2H, m, CHCH2), 3.77–
3.47 (2H, m, –OCH2CH3), 1.21 (9H, s, –C(CH3)3), 1.24–1.18
(3H, m, –OCH2CH3); δC (63 MHz, CDCl3) mixture: 178.3,
(1H, dd, 2J 11.4, J 7.5, –CHaHbOBn); δC(75 MHz, CDCl3)
2
164.8, 137.5, 132.3, 129.8 (t, JC–F 25.4), 128.5, 127.8. 127.6,
127.5, 121.6 (t, JC–F 9.6), 118.0 (dd, JC–F 245.3, 243.6), 73.2,
3
1
2
2
135.0, 130.3 (t, JC–F 25.4), 130.2 (t, JC–F 25.7), 121.3, 119.5,
2 2
2
3
71.9 (dd, JC–F 31.7, 29.4), 66.9 (t, JC–F 3.4); δF (282 MHz,
CDCl3) Ϫ105.8 (1F, dt, 2J 254.3, 3JF–H 2.7, 2.7), Ϫ110.1 (1F, dt,
second half of an AB quartet, 2J 254.3, 3JF–H 3.2, 3.2); [HRMS
(FAB, [M ϩ H]ϩ) Found: 283.11459, calc. for C15H17O3F2:
283.11458; m/z (EI) 282 (30%, Mϩ), 227 (15, M–H C᎐
119.1, 103.1, 102.5, 75.0 (t, JC–F 28.7), 74.4 (t, JC–F 30.3),
63.0, 61.6, 39.0, 27.4, 15.4, 15.3; δF (282 MHz, CDCl3)
diastereoisomer 1: Ϫ103.1 (1F, dt, 2J 254.6, JF–H 8.9, 8.9),
3
2
3
Ϫ108.9 (1H, ddd, J 254.6, JF–H 14.0, 6.4), diastereoisomer 2:
2
3
Ϫ103.2 (1F, dt, J 252.8, JF–H 10.2, 10.2), Ϫ109.2 (1F, ddd,
᎐
2
CHCHO), 175 (50, M–OCH2Ph), 107 (49, OCH2Phϩ), 91 (100,
2J 252.8, JF–H 14.0, 7.6). These acetals were taken on without
3
CH2Phϩ).
further purification.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
537

View Article Online
Ethyl 4,4-difluoro-2,3,4-trideoxy-6-O-benzyloxy-DL-glycero-
hex-2-enopyranoside 18d and 19d; procedure with Grubbs’
catalyst
refluxed for an additional 2 days and concentrated under
reduced pressure to leave a dark brown oil (20.03 g). Purifi-
cation by column chromatography (10% diethyl ether in light
petroleum) afforded an inseparable mixture (1 : 1) of the cis-
and trans- dihydropyran diastereoisomers 18f and 19f as a pale
yellow oil (11.47 g, 66%, 100% by GC-MS); Rf (10% diethyl
ether in light petroleum) 0.29; νmax (film)/cmϪ1 2977m (C–H),
A mixture of diastereoisomeric acetals 17d (5.2 mmol, 1.63 g)
and Grubbs’ catalyst (0.26 mmol, 0.22 g) in dry degassed DCM
(30 mL) was refluxed for 3 days under a nitrogen atmosphere.
The reaction mixture was concentrated under reduced pressure
to leave a dark brown oil (1.88 g). Purification by column
chromatography (5% diethyl ether in light petroleum) afforded
a mixture (1 : 1) of cis-18d and trans-19d (as a 1 : 1 mixture of
diastereoisomers) as a clear oil (1.05 g, 68%, 100% by GC-MS);
careful column chromatography (5% ether in light petroleum)
afforded (in order of elution) cis-18d (0.51 g, 35%), Rf 0.28, GC
2935w (᎐C–H), 1735s (C᎐O), 1657w (C᎐C); δ (250 MHz,
᎐
᎐
᎐
H
CDCl3) cis–trans-mixture 6.11–6.05 (1H, m, H-3), 5.95–5.85
(1H, m, H-2), 5.10–5.01 (1H, m, H-1), 4.43–3.34 (5H, m,
–OCH2CH3, H-6 and H-5), 1.20–1.15 (12H, m, –OCH2CH3 and
–C(CH3)3); δc (63 MHz, CDCl3) cis–trans-mixture: 178.0, 136.6
3
3
2
(t, JC–F 9.4), 134.0 (t, JC–F 9.2), 124.7 (t, JC–F 25.9), 124.2 (t,
2JC–F 21.9), 113.7 (dd, 1JC–F 243.6, 236.5), 113.4 (dd, 1JC–F 245.2,
235.5), 96.8, 93.5, 73.4 (t, JC–F 27.7), 68.6 (t, JC–F 27.0), 64.6,
61.1, 38.9, 27.2, 15.2; δF (235 MHz, CDCl3) diastereoisomer 1:
retention time 15.8 mins., νmax (film)/cmϪ1 2977m (᎐C–H),
᎐
2
2
2934w (C–H), 1652w (C᎐C); δ (400 MHz, CDCl3) 7.25–7.08
᎐
H
(5H, m, Ph), 6.19 (1H, dd, J 10.4, 1.3, ᎐CHCH), 6.00 (1H,
᎐
2
3
Ϫ104.3 (1F, ddd, J 279.0, JF–H 11.9, 5.3), Ϫ107.6 (1F, ddt,
3
4
dddd, J 10.4, 7.7, JH–F 6.9, 3.0, J 1.8, ᎐CHCF ), 5.23–5.10
᎐
3
4
2J 279.0, JF–H 14.6, 5.4, JF–H 5.4) diastereoisomer 2: Ϫ110.1
2
2
(1H, m, CHOEt), 4.70 (1H, d, J 12.1, OCHaHbPh), 4.64 (1H,
2
3
2
3
(1F, dd, J 279.3, JF–H 21.2), Ϫ113.9 (1F, ddd, J 279.3, JF–H
9.3, 2.6); [HRMS (EI, [M ϩ Na]ϩ) Found: 277.12509, calc. for
C13H19F2O4:277.12514]; m/z (EI) 277 (16%, Mϩ Ϫ H), 249 (5,
Mϩ Ϫ C2H5), 233 (13, Mϩ Ϫ OEt), 205 (50), 176 (80), 134 (100).
2
3
d, J 12.1, OCHaHbPh), 4.10 (1H, dddd, JH–F 17.8, 5.0, J 7.8,
2
3.0, CF2CH ), 4.01 (1H, dq, J 9.5, J 7.0, OCHaHbCH3), 4.00
(1H, dd, 2J 12.3, J 3.0, CHaHbOBn), 3.84 (1H, dd, 2J 12.3, J 7.8,
CHaHbOBn), 4.01 (dq, 2J 9.5, J 7.0, OCHaHbCH3), 1.12 (3H, t,
3
J 7.0, CH3); δC (75 MHz, CDCl3) 138.0, 136.4 (t, JC–F9.5),
128.5, 127.8, 127.7, 127.4 (t, JC–F28.6), 113.7 (dd, JC–F242.7,
236.7), 96.8, 75.4 (dd, JC–F30.0, 25.3), 73.7, 67.3 (d, JC–F5.1),
64.4, 15.1; δF (376 MHz, CDCl3) Ϫ105.2 (1F. br d, one half of
Ethyl 6-O-benzyl-4,4-difluoro-4-deoxy-ꢀ-DL-ribo-pyranoside 20
2
1
2
3
A solution of NMO (3.5 mmol, 0.41 g) in water (1 mL) was
added to a cold (0 ЊC) solution of cis-18d (1.76 mmol, 0.50 g) in
acetone (2 mL) and tert-butanol (2 mL). The mixture was
stirred for 10 minutes and osmium tetroxide (0.035 mmol, 0.36
mL of a 2.5 wt% solution in tert-butanol, 2 mol%) was added.
The reaction mixture was stirred at 0 ЊC for 2 hours, allowed to
warm to room temperature and stirred at this temperature for
10 days. It was then quenched with sodium sulfite (7 mmol, 0.88
g), diluted with water (7 mL), stirred for an additional 30 min-
utes and extracted with ethyl acetate (3 × 7 mL). The combined
organic extracts were dried (MgSO4), filtered and concentrated
under reduced pressure to leave a yellow oil. Purification by
column chromatography (50% ethyl acetate in light petroleum)
afforded the desired pyranoside 20 as a pale yellow oil (0.428 g,
65%, 100% by GC) as a low melting solid, Rf 0.38; GC retention
time 21.27min; Found: C, 56.90; H, 6.01%. Calc. for
C15H20F2O5: C, 56.60; 6.29%. δH(400 MHz, CDCl3); 4.71 (1H,
d, J 8.0, H-1), 4.63 (1H, d, one half of an AB q, 2J 12.1,
CHaCHbPh), 4.57 (1H, d, one half of an AB q, 2J 12.1,
2
an AB q, J 278.3), Ϫ106.7 (1F. dddd, one half of an AB q,
2J 278.3, JH–F17.8, 6.9, 3.0); [HRMS (EI) Found: 284.12243.
3
Calc. for C15H18F2O3: 284.12240]; m/z (EI, 40eV) 284 (18%,
[M ϩ H]ϩ), 145 (40), 91 (100, [CH2Ph]ϩ): and trans-19d (0.50 g,
33%), Rf 0.25, GC retention time 16.1 mins., νmax (film)/cmϪ1
2977m (᎐C–H), 2934w (C–H), 1652w (C᎐C); δ (400 MHz,
᎐
᎐
H
CDCl3) δH (400 MHz, CDCl3) 7.30–7.10 (5H, m, Ph), 6.18 (1H,
3
dd, J 10.3, 3.0, ᎐CHCH), 6.00 (1H, br t, J 10.3, J
10.3,
᎐
H–F
᎐CHCF ), 5.16 (1H, t, J 3.0, 4J 3.0, CHOEt), 4.70 (1H, d,
᎐
2
2J 12.1, OCHaHbPh), 4.66 (1H, d, J 12.1, OCHaHbPh), 4.51
2
3
(1H, ddt, JH–F 22.8, 3.0, J 7.9, 3.0, CF2CH ), 4.01 (1H, dd,
2J 11.1, J 3.0, CHaHbOBn), 3.96 (1H, dq, J 9.6, J 7.0, OCHa-
2
2
HbCH3), 3.81 (1H, dd, J 11.1, J 7.9, CHaHbOBn), 3.66 (1H,
dq, 2J 9.6, J 7.0, OCHaHbCH3), 1.15 (3H, t, J 7.0, CH3);
3
δC (75 MHz, CDCl3) 138.1, 133.7 (t, JC–F8.7), 128.4, 127.7,
127.5, 124.3 (dd, 2JC–F30.0, 26.1), 113.4 (dd, 1JC–F244.1, 233.9),
2
3
93.4, 73.4, 69.7 (dd, JC–F29.9, 23.9), 66.9 (d, JC–F6.2), 64.7,
15.1; δF (376 MHz, CDCl3) Ϫ109.3 (1F, dd, one half of an AB
3
CHaCHbPh), 4.24 (1H, ddd, JH–F 25.1, J 7.5, 2.5, H-5), 4.20–
2
3
q, J 279.0, JH–F22.5), Ϫ113.2 (1F. ddd, one half of an AB q,
4.13 (1H, br m, H-3), 4.10–3.88 (3H, m [including 4.03 (1H, dq,
2J 9.6, J 7.1, OCHaCHbCH3) and 3.92 (1H, br dd, 2J 11.1, J 2.5,
H-6a)], OCHaCHbCH3, H-6a, OH), 3.79–3.46 (4H, m [includ-
2J 279.0, JH–F10.3, JH–F 3.0); [HRMS (EI) Found: 284.12248.
Calc. for C15H18F2O3: 284.12240]; m/z (EI, 40eV) 284 (18%, [M
ϩ H]ϩ), 145 (40), 91 (100, [CH2Ph]ϩ).
3
3
2
ing 3.76 (1H, dd, J 11.1, J 7.5, H-6b)]), OCHaCHbCH3, H-2,
H-6b and OH), 1.30 (3H, t, J 7.1, CH3); δC (75 MHz, CDCl3)
1
137.8, 128.5, 127.8, 127.7, 117.7 (dd, JC–F256.1, 248.9), 99.7,
Ethyl 4,4-difluoro-2,3,4-trideoxy-6-O-benzyloxy-DL-glycero-
hex-2-enopyranoside 18d and 19d; procedure with Nolan catalyst
2
2
71.7 (dd, JC–F29.2, 22.1), 70.0 (dd, JC–F34.2, 21.0), 69.8 (d,
3JC–F4.5), 65.8, 15.1; δF (282 MHz, CDCl3) Ϫ118.7 (1F, dd,
2J 259.3, JF–H5.9), Ϫ124.3 (1F, dd, 2J 259.3, JF–H25.1); m/z (EI)
318 (40%, Mϩ), 300 (10, M Ϫ H2O), 272 (21, M Ϫ 2H2O), 91
(100, PhCH2ϩ).
A mixture of diastereoisomeric acetals 17d (0.43 mmol, 0.135 g)
and Nolan catalyst 1,3-(bis(mesityl)-2-imidazolidinylidene)
dichloro(phenylmethylene) (tricyclohexylphosphino) ruthen-
ium (0.022 mmol, 0.20 g) in dry degassed DCM (2.5 mL) was
refluxed for 3 hours under a nitrogen atmosphere. The reaction
mixture was treated as described previously to afford a mixture
(1 : 1) of cis-18d and trans-19d (as a 1 : 1 mixture of diastereo-
isomers) as a clear oil (0.091 g, 75%, 100% by GC-MS). Data
for 17d as described previously.
Ethyl 6-O-benzyl-4,4-difluoro-4-deoxy-ꢀ-DL-lyxo-pyranoside 21
Was prepared as for 20 from 19d (0.7 mmol, 0.20 g), NMO
(1.4 mmol, 0.165 g), osmium tetroxide (0.014 mmol, 0.15 mL of
a 2.5% w/v solution in tert-butanol), tert-butanol (0.7 mL) and
water (0.35 mL). The reaction required 8 days to reach comple-
tion. The usual work-up and purification afforded 21 (0.151 g,
68%, 100% by GC) as a powder, mp 77–78 ЊC; Rf (50% ethyl
acetate in light petroleum) 0.36; GC retention time 21.13min;
Found: C, 56.77; H, 6.47. Calc. for C15H20F2O5: C, 56.60; H,
6.29%. δH(400 MHz, CDCl3) 4.63 (1H, s), H-1), 4.66 (1H, d,
one half of an AB q, 2J 12.1, CHaCHbPh), 4.59 (1H, d, one half
of an AB q, 2J 12.1, CHaCHbPh), 4.15–3.99 (2H, m, [including
4.10 (1H, ddd, 3JH–F 24.7, J 7.6, 2.5, H-5)], H-5 and H-3), 3.99–
Ethyl 4,4-difluoro-2,3,4-trideoxy-6-O-pivaloyloxy-DL-glycero-
hex-2-enopyranoside 18f and 19f
A mixture of acetal 17f (62 mmol, 19.00 g) and Grubbs’ catalyst
(1.55 mmol, 1.28 g) in dry degassed DCM (1.2 L) was refluxed
for 3 days under a nitrogen atmosphere after which time some
more Grubbs’ catalyst (1.55 mmol, 1.28 g) was added as a solu-
tion in dry degassed DCM (10 mL). The reaction mixture was
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
538

View Article Online
3.91 (2H, m, [including 3.94 (1H, dd, 2J 10.9, J 2.5, H-6a)], H-2
was separated and extracted further with diethyl ether in a con-
2
and H-6a), 3.82 (1H, dq, J 9.8, J 7.0, OCHaCHbCH3), 3.78
tinuous extractor for 4 days. The combined organic extracts
were dried (MgSO4), filtered and concentrated under reduced
pressure to leave a yellow oil (2.85 g). Purification by column
chromatography (80% ethyl acetate in light petroleum) afforded
29 as colourless oil, which solidified on standing (1.71 g, 62%).
Rf (80% ethyl acetate in light petroleum) 0.27; Mp 92–93 ЊC;
(Found C, 42.28; H, 6.22; C8H14F2O3 requires: C, 42.11; H,
6.18%); νmax(KBr)/cmϪ1 3392s br (O–H), 2978m (C–H), 2934m
(C–H); δH (250 MHz, DMSO-d6) 6.16 (1H, d, J 4.8, –OH ), 5.41
(1H, d, J 6.9, –OH ), 5.01 (1H, t, J 6.0, –OH ), 4.71 (1H, d, J 7.8,
H-1), 4.11–3.96 (3H, m, H-3, H-5 and –OCHaHbCH3), 3.91–
3.83 (1H, m, H-6a), 3.77–3.63 (2H, m, –OCHaHbCH3 and
H-6b), 3.52–3.43 (1H, m, H-2), 1.33 (3H, t, J 7.1 –OCH2CH3;
(1H, dd, 2J 10.9, J 7.6, H-6b), 3.58 (1H, dq, 2J 9.8, J 7.0, OCHa-
CHbCH3), 3.34–3.23 (1H, m, C(3)OH ), 2.70–2.58 (1H, m,
C(2)OH ), 1.28 (3H, t, J 7.0, CH3); δC (75 MHz, CDCl3) 137.8,
1
128.4, 127.8, 127.7, 118.0 (dd, JC–F252.1, 250.8), 99.3, 73.6,
2
2
70.7, 69.4 (dd, JC–F28.6, 22.6), 67.2 (dd, JC–F20.6, 18.6), 66.4
3
(d, JC–F5.0), 63.9, 14.8; δF (282 MHz, CDCl3) Ϫ114.8 (1F, d,
1J 251.2, JF–H5.9), Ϫ124.3 (1F, dt, J 251.2, JF–H24.7); [HRMS
1
(EI, Mϩ) Found: 318.12761. Calc. for C15H20F2O5: 318.12788];
m/z (EI) 318 (40%, Mϩ), 300 (10, M Ϫ H2O), 272 (21,
M Ϫ 2H2O), 91 (100, PhCH2ϩ).
Ethyl cis-4,4-difluoro-2,3,4-trideoxy-ꢁ-DL-glycero-hex-2-
enopyranoside 27 and ethyl trans-4,4-difluoro-2,3,4-trideoxy-ꢀ-
DL-glycero-hex-2-enopyranoside 28
1
δC (63 MHz, DMSO-d6) 119.2 (dd, JC–F 256.6, 247.4), 100.2,
2
2
72.7 (dd, JC–F 29.0, 21.9), 70.4 (dd, JC–F 32.3, 19.6), 69.6
3
3
(d, JC–F 6.1), 64.6, 58.2 (d, JC–F 4.6), 15.4; δF (235 MHz,
DMSO-d6) Ϫ117.8 (1F, dd, 2J 249.5, 3JF–H 6.6), Ϫ123.6 (1F, ddt,
DIBAL-H (117.8 mmol, 97.4 mL of a 20 wt%. solution in hex-
anes) was added dropwise to a cold (Ϫ78 ЊC) solution of cis-18f
and trans-19f (47.1 mmol, 13.12 g) in dry DCM (500 mL). After
completion of the addition the mixture was stirred at Ϫ78 ЊC
for 1 hour and at 0 ЊC for an additional hour. The mixture was
quenched carefully with water (200 mL) and HCl (150 mL of a
1 M solution). The phases were separated and the aqueous layer
was extracted with DCM (3 × 230 mL). The combined organic
extracts were dried (MgSO4), filtered and concentrated under
reduced pressure to afford a mixture (1 : 1) of alcohols 27 and
28 as a yellow oil (9.25 g). Purification by column chromato-
graphy (30% diethyl ether in light petroleum) allowed the separ-
ation of the two diastereoisomers. In order of elution, trans-28
was obtained as a pale yellow oil (3.75 g, 41%, 100% by GC); Rf
(30% diethyl ether in light petroleum) 0.27; νmax (film)/cmϪ1
2J 249.5, JF–H 26.5, 4.6, JF–H 4.6); [HRMS (FAB, [M ϩ H]ϩ)
Found: 229.08879, calc. for C15H8O5F2: 229.08876]; m/z (FAB)
229 (41%, [M ϩ H]ϩ), 211 (11), 183 (37). An analytical sample
was recrystallised by vapour diffusion (diethyl ether–light
petroleum) to afford colourless needles, which were used to
obtain an X-ray crystal structure of 29.
3
4
Ethyl 4,4-difluoro-4-deoxy-ꢀ-DL-lyxo-pyranoside 30
Was prepared exactly as for 29, on the same scale and with the
same reaction times and work-up (including continuous extrac-
tion). The combined organic extracts were dried (MgSO4), fil-
tered and concentrated under reduced pressure to leave a yellow
oil (2.85 g). Purification by column chromatography (80%
ethyl acetate in light petroleum) afforded 30 as a pale yellow oil
(1.70 g, 62%). Rf (80% ethyl acetate in light petroleum) 0.24;
νmax(film)/cmϪ1 3396s br (O–H), 2978m (C–H), 2934m (C–H);
δH (250 MHz, DMSO-d6) 5.50 (1H, d, J 8.0, –OH ), 5.17 (1H, d,
J 4.4, –OH ), 5.03 (1H, t, J 6.0, –OH ), 4.89 (1H, s, H-1), 4.02–
3.79 (5H, m, H-2, H-4, H-5, H-6a, –OCHaHbCH3), 3.76–3.56
(2H, m, H-6b, –OCHaHbCH3), 1.32 (3H, t, J 7.1, –OCH2CH3);
3430s br (O–H), 2978m (᎐C–H), 1664w (C᎐C); δ (250 MHz,
᎐
᎐
H
CDCl3) 5.96 (1H, dd, J 10.3, 3JH–F 3.0, H-3), 5.82–5.74 (1H, m,
4
H-2), 4.93 (1H, t, J 3.0, J 3.0, H-1), 4.14–4.00 (1H, m, H-6a),
3.83–3.62 (3H, m H-6b, –OCHaHbCH3 and H-5), 3.47–3.34
(1H, m, –OCHaHbCH3), 1.81 (1H, br s, –OH ), 1.06 (3H, t,
3
J 7.0, –OCH2CH3); δC (63 MHz, CDCl3) 133.9 (t, JC–F 9.4),
124.6 (dd, 2JC–F 30.5, 25.9), 114.0 (dd, 1JC–F 244.1, 235.0), 93.7,
70.8 (dd, 2JC–F 31.0, 23.9), 65.1, 59.6 (d, 3JC–F 6.6), 15.4; δF (235
MHz, CDCl3) Ϫ109.4 (1F, dd, 2J 279.0, 3JF–H 21.9), Ϫ114.0 (1F,
1
δC (63 MHz, DMSO-d6) 118.8 (dd, JC–F 258.1, 244.4), 99.6,
71.5 (dd, 2JC–F 29.0, 22.9), 70.4 (d, 3JC–F 7.6), 67.0 (t, 2JC–F 19.1,
2
3
3
ddd, J 279.0, JF–H 9.3, 2.7); [HRMS (EI, [M–H]ϩ) Found:
193.06756, calc. for C8H11F2O3:193.06763]; m/z (EI) 193 (2%,
Mϩ–H), 163 (42, Mϩ–CH2OH), 149 (100, Mϩ–OEt): followed
by cis-27, pale yellow oil (3.56 g, 39%); Rf (30% diethyl ether in
light petroleum) 0.19; νmax (film)/cmϪ1 3426s br (O–H), 2979 m
19.1), 62.8, 58.0 (d, JC–F 5.6), 15.0; δF (235 MHz, DMSO-d6)
2
2
3
Ϫ112.7 (1F, d, J 242.8), Ϫ128.6 (1F, dt, J 242.8, JF–H 24.5,
24.5); [HRMS (FAB, [M Ϫ H]Ϫ) Found: 227.07300, calc. for
C8H13O5F2: 227.07311]; m/z (ES) 227 (100%, [M Ϫ H]Ϫ).
(᎐C–H), 1165w (C᎐C); δ (250 MHz, CDCl3) 5.98 (1H, dd,
Ethyl 4,4-difluoro-4-deoxy-ꢁ-DL-lyxo-pyranoside triacetate 31
᎐
᎐
H
J 10.3, 3JH–F 1.4, H-3), 5.88–5.79 (1H, m, H-2), 5.05–5.00 (1H,
m, H-1), 3.86–3.72 (4H, m, H-6, –OCHaHbCH3 and H-5), 3.56–
3.44 (1H, m, –OCHaHbCH3), 2.28 (1H, br s, –OH ), 1.10 (3H,
Acetic anhydride (0.5 mL, 5 mmol) was added to a solution of
pyranoside 30 (228 mg, 1 mmol) in pyridine (5 mL). The reac-
tion mixture was stirred at room temperature for 17 hours,
poured onto ice–water (10 mL) and extracted with DCM (3 ×
10 mL). The combined organic extracts were washed succes-
sively with sodium hydrogen carbonate (10 mL of a saturated
aqueous solution) and water (10 mL), dried (MgSO4), filtered
and concentrated under reduced pressure to leave a pale yellow
oil (262 mg). Purification by column chromatography (10% di-
ethyl ether in light petroleum) afforded triacetate 31 as a pale
yellow solid (251 mg, 72%). Rf 0.26 (10% diethyl ether in light
petroleum); mp 52–53 ЊC; (Found C, 47.59; H, 5.78; C14-
H20F2O8 requires: C, 47.46; H, 5.69%); νmax(KBr)/cmϪ1 2988m
3
t, J 7.1, –OCH2CH3); δC (63 MHz, CDCl3) 135.9 (t, JC–F 9.7,
2
1
9.7), 125.2 (t, JC–F 28.5, 28.5), 114.1 (dd, JC–F 263.0, 242.6),
2
3
96.4, 76.3 (dd, JC–F 30.8, 25.7), 64.8, 60.2 (d, JC–F5.1), 15.4;
δF (235 MHz, CDCl3) Ϫ102.0 (1F, ddd, J 279.3, JF–H 11.3,
6.0), Ϫ108.7 (1F, ddd, J 279.3, JF–H 10.6, 4.0); [HRMS (EI,
[M–H]ϩ) Found: 193.06762, calc. for C8H11F2O3:193.06763];
m/z (EI) 193 (21%, Mϩ Ϫ H), 163 (42, Mϩ Ϫ CH2OH), 149
(100, Mϩ Ϫ OEt).
2
3
2
3
Ethyl 4,4-difluoro-4-deoxy-ꢀ-DL-ribo-pyranoside 29
A solution of NMO (24.0 mmol, 2.81 g) in water (3 mL) was
added to a cold (0 ЊC) solution of cis-18f (12.0 mmol, 2.33 g) in
acetone (12 mL) and tert-butanol (12 mL). The mixture was
stirred for 10 minutes and osmium tetroxide (0.24 mmol, 2.9
mL of a 2.5 wt% solution in tert-butanol) was added. The reac-
tion mixture was stirred at 0 ЊC for 2 hours, allowed to warm to
room temperature and stirred at this temperature for 5 days. It
was then quenched with sodium sulfite (48.0 mmol, 6.05 g),
diluted with water (50 mL), stirred for an additional 30 minutes
and extracted with ethyl acetate (3 × 50 mL). The aqueous layer
(C–H), 1755s (C᎐O); δ (400 MHz, CDCl ) 5.49 (1H, ddd, 3J
᎐
H
3
H–F
22.4, 6.0, J 4.0, H-3), 5.31–5.26 (1H, m, H-2), 4.88 (1H, s, H-1),
4.50 (1H, dd, 2J 11.9, J 3.3, H-6a), 4.38 (1H, dd, 2J 11.9, J 7.6,
H-6b), 4.16 (1H, ddd, 3JH–F 22.4, J 7.6, 3.3, H-5), 3.78 (1H, dq,
2J 9.8, J 7.1, –OCHaHbCH3), 3.60 (1H, dq, 2J 9.8, J 7.1,
–OCHaHbCH3), 2.07 (3H, s, – CH3CO), 2.05 (3H, s, CH3CO),
2.02 (3H, s, CH3CO), 1.20 (3H, t, J 7.1, –OCH2CH3); δC (63
MHz, CDCl3) 170.9, 170.5, 169.7, 116.3 (dd, 1JC–F 260.4, 248.7),
97.9, 69.4 (t, 3JC–F 8.9), 68.9 (dd, 2JC–F 28.5, 22.9), 66.5 (dd, 2JC–F
22.1, 17.0), 60.7, 60.6, 21.1, 21.0, 20.8, 15.2; δF (235 MHz,
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
539

View Article Online
CDCl3) Ϫ115.9 (1F, d, 2J 248.1), Ϫ129.0 (1F, dt, 2J 248.1, 3JF–H
22.4); [HRMS (FAB, [M ϩ H]ϩ) Found: 355.12041, calc. for
C14H21O8F2: 355.12045]; m/z(FAB) 355 (50%, [M ϩ H]ϩ), 309
(65), 249 (100).
difluorinated anyhdrosugar related to our targets, see: F. Oberdorfer,
R. Haeckel and G. Lauer, Synthesis, 1998, 201–206. The synthesis is
based upon a low yielding reaction of -galactal, and an inefficient
fluorination with DAST.
10 S. Gaisser, C. J. Martin, B. Wilkinson, R. M. Sheridan, R. E. Lill,
A. J. Weston, S. J. Ready, C. Waldron, G. D. Crouse, P. F. Leadlay
and J. Staunton, Chem. Commun., 2002, 618–619 explicitly raises the
issue of conformational flexibility in highly deoxygenated sugars,
and the relationship to specificity of biosynthetic enzymes.
11 U. M. Unligil and J. M. Rini, Curr. Opin. Struct. Biol., 2000, 10,
510–517. For recent theroetical work, see: I. Tvaroska, I. Andre and
J. P. Carver, J. Am. Chem. Soc., 2000, 122, 8762–8776.
12 T. Taguchi, Y. Kodama and M. Kanazawa, Carbohydr. Res., 1993,
249, 243–252.
13 H. Amii, T. Kobayashi, H. Terasawa and K. Uneyama, Org. Lett.,
2001, 3, 3103–3105.
14 J. M. Percy and S. Pintat, Chem. Commun., 2000, 607–608.
15 D. Seyferth, R. M. Simon, D. J. Sepelak and H. A. Klein, J. Am.
Chem. Soc., 1983, 105, 4634–4639.
16 Z. Y. Yang and D. J. Burton, J. Org. Chem., 1991, 56, 1037–1041.
17 M. Kirihara, T. Takuwa, S. Takizawa, T. Momose and H. Nemoto,
Tetrahedron, 2000, 56, 8275–8280. For reactions of related building
blocks, see : Z. G. Wang and G. B. Hammond, Chem. Commun.,
1999, 2545–2546; Z. G. Wang and G. B. Hammond, Tetrahedron
Lett., 2000, 41, 2339–2342; Y. Hanzawa, K. Inazawa, A. Kon,
H. Aoki and Y. Kobayashi, Tetrahedron Lett., 1987, 28, 659–
662. For a recent difluoronucleoside synthesis using 1-bromo-1,1-
difluoropropene, see: X. G. Zhang, H. R. Xi, X. C. Dong, J. Jin,
W. D. Meng and F. L. Qing, J. Org. Chem., 2003, 68, 9026–9033.
18 M. Schlosser and S. Strunk, Tetrahedron, 1989, 45, 2649–2664; S. T.
Patel, J. M. Percy and R. D. Wilkes, J. Org. Chem., 1996, 61, 166–
173.
4-Deoxy-4,4-difluoro-ꢀ-DL-ribo-pyranoside 32
Deuterium chloride (0.03 mL of a 35% w/v solution in D2O)
was added to a solution of ethyl glycoside 29 (0.080 g, 0.35
mmol) in D2O (0.4 mL). The reaction was heated to 80 ЊC and
stirred for 3 days. The reaction was judged to be complete by
TLC and electrospray MS, and analysed without any attempt
at purification. Major (β) anomer δH (D2O, 400 MHz) 4.52
(1H, dd, J 8.0, 5JH–F0.6, H-1), 3.75 (1H, ddd, 3JH–F7.1, 3.9, J 5.2,
H-3), 3.67 (1H, ddd, 3JH–F25.1, J 7.2, 3.2, H-5), 3.49 (1H, ddd,
2
2J 12.2, J 3.2, 1.2, H-6a), 3.34 (1H, dd, J 12.2, J 7.2, H-6b),
3.18–3.13 (1H, m, H-2); δC (D2O, 100 MHz) 117.6 (dd,
2
1JC–F255.3, 248.8), 93.1, 71.5 (dd, JC–F30.5, 22.3), 69.6 (d,
2
3
3JC–F6.2), 69.4 (dd, JC–F33.1, 20.7), 57.6 (d, JC–F4.4); δF (D2O,
376 MHz) Ϫ120.1 (1F, dd, 2JF–F257.8, 3JH–F7.1), Ϫ125.0 (1F, ddt,
2JF–F257.8, 3JH–F25.1, 3.9, 4JH–F 3.9); m/z (ESϪ MeOH) 199 (8%,
[M Ϫ H]), 171 (100, [M Ϫ HCHO]). The following significant
peaks were observed for the minor (α) anomer: δH (D2O, 376
MHz) 4.80 (1H, t, J, 5JH–F3.2, H-1), 3.89 (1H, ddd, 3JH–F25.4, 7.5,
J 3.2, H-3), 3.55–3.44 (1H, m, H-5), 3.38 (1H, dd, 2J 12.4, J 7.2,
H-6b); other signals were overlapped with those for the major
2
anomer; δF (D2O, 376 MHz) Ϫ116.6 (1F, dd, JF–F257.0,
2
3
3JH–F7.5), Ϫ125.3 (1F, br dd, JF–F257.0, JH–F25.4). The 13C
19 P. Schwab, M. B. France, J. W. Ziller and R. H. Grubbs, Angew.
Chem., Int. Ed. Engl., 1995, 34, 2039–2041.
NMR was too weak to assign reliably.
20 M. Ahmed, A. G. M. Barrett, D. C. Braddock, S. M. Cramp and
P. A. Procopiou, Tetrahedron Lett., 1999, 40, 8657–8662.
21 J. A. Marco, M. Carda, S. Rodriguez, E. Castillo and M. N.
Kneeteman, Tetrahedron, 2003, 59, 4085–4101.
22 L. Ackermann, A. Furstner, T. Weskamp, F. J. Kohl and W. A.
Herrmann, Tetrahedron Lett., 1999, 40, 4787–4790; J. K. Huang,
E. D. Stevens, S. P. Nolan and J. L. Petersen, J. Am. Chem. Soc.,
1999, 121, 2674–2678; M. Scholl, T. M. Trnka, J. P. Morgan and
R. H. Grubbs, Tetrahedron Lett., 1999, 40, 2247–2250.
23 V. S. Enev, H. Kaehlig and J. Mulzer, J. Am. Chem. Soc., 2001, 123,
10764–10765.
Acknowledgements
The authors wish to thank the Universities of Birmingham
and Leicester, the EPSRC (Project Grant GR/K84882),
GlaxoSmithKline (CASE Studentship for SP) and Synquest
(Alachua, Florida USA) for supplying 1-bromo-1,1-difluoro-
propene. We thank Emi Uneyama for carrying out the conver-
sion of 29 to 32.
24 V. VanRheenen, R. C. Kelly and D. Y. Cha, Tetrahedron Lett., 1976,
1973–1976.
25 K. T. Tan, S. S. Chng, H. S. Cheng and T. P. Loh, J. Am. Chem. Soc.,
2003, 125, 2958–2963.
26 L. A. Paquette, Synthesis, 2003, 765–774.
References
1 M. N. Namchuk, J. D. McCarter, A. Becalski, T. Andrews and S. G.
Withers, J. Am. Chem. Soc., 2000, 122, 1270–1277.
2 M. Namchuk, C. Braun, J. D. McCarter and S. G. Withers, ACS
Symp. Ser., 1996, 639, 279–293. For 5-fluoroglycoside analogues,
see: S. Howard, S. M. He and S. G. Withers, J. Biol. Chem., 1998,
273, 2067–2072; M. C. T. Hartman and J. K. Coward, J. Am. Chem.
Soc., 2002, 124, 10036–10053.
3 G. Sulzenbacher, L. F. Mackenzie, K. S. Wilson, S. G. Withers,
C. Dupont and G. J. Davies, Biochemistry, 1999, 38, 4826–4833;
K. Persson, H. D. Ly, M. Dieckelmann, W. W. Wakarchuk, S. G.
Withers and N. C. J. Strynadka, Nat. Struct. Biol., 2001, 8, 166–175;
V. M.-A. Ducros, D. L. Zechel, G. N. Murshudov, H. J. Gilbert,
L. Szabó, D. Stoll and S. G. Withers, Angew. Chem., Int. Ed., 2002,
41, 2824–2827.
27 J. K. Cha and N. S. Kim, Chem. Rev., 1995, 95, 1761–1795.
28 H. W. Kim, P. Rossi, R. K. Shoemaker and S. G. DiMagno, J. Am.
Chem. Soc., 1998, 120, 9082–9083.
29 For an extensive recent compilation of relevant NMR data,
including long-range H–F coupling constants, see: M. Michalik,
M. Hein and M. Frank, Carbohydr. Res., 2000, 327, 185–218.
30 P. Collins and M. Ferrier, Monosaccharides: Their Chemistry, and
Their Roles in Natural Products, Wiley, Chichester, 1995, 40–44.
Clearly furanose forms are unavailable to 32 because there is no
hydroxyl group at C-4.
31 M. de Rosa, A. Solladie-Cavallo and A. Scettri, Tetrahedron Lett.,
2000, 41, 1593–1596.
4 D. O. Hart, S. M. He, C. J. Chany, S. G. Withers, P. F. G. Sims,
M. L. Sinnott and H. Brumer, Biochemistry, 2000, 39, 9826–
9836.
5 S. P. Vincent, M. D. Burkart, C. Y. Tsai, Z. Y. Zhang and C. H.
Wong, J. Org. Chem., 1999, 64, 5264–5279.
6 S. G. Withers, M. D. Percival and I. P. Street, Carbohydr. Res., 1989,
187, 43–66.
7 K. Dax, M. Albert, J. Ortner and B. J. Paul, Carbohydr. Res., 2000,
327, 47–86; K. Dax and M. Albert, Top. Curr. Chem., 2001, 215,
193–275.
32 M. T. Crimmins and A. L. Choy, J. Org. Chem., 1997, 62, 7548–
7549.
33 C. Held, R. Frohlich and P. Metz, Angew. Chem., Int. Ed. Engl.,
2001, 40, 1058–1060.
34 F. Rutjes, T. M. Kooistra, H. Hiemstra and H. E. Schoemaker,
Synlett, 1998, 192–194.
35 M. T. Crimmins and B. W. King, J. Am. Chem. Soc., 1998, 120,
9084–9085.
36 G. T. Nadolski and B. S. Davidson, Tetrahedron Lett., 2001, 42, 797–
800.
8 A. Ellaghdach, R. Echarri, M. I. Matheu, M. I. Barrena, S. Castillon
and J. Garcia, J. Org. Chem., 1991, 56, 4556–4559; R. Fernandez,
M. I. Matheu, R. Echarri and S. Castillon, Tetrahedron, 1998, 54,
3523–3532; M. I. Barrena, M. I. Matheu and S. Castillon, J. Org.
Chem., 1998, 63, 2184–2188; R. Fernandez and S. Castillon,
Tetrahedron, 1999, 55, 8497–85081; M. L. Aghmiz, Y. Diaz, G. H.
Jana, M. I. Matheu, R. Echarri, S. Castillon and M. L. Jimeno,
Tetrahedron, 2001, 57, 6733–6743.
37 P. V. Ramachandran, M. V. R. Reddy and H. C. Brown, Tetrahedron
Lett., 2000, 41, 583–586.
38 A. Kirby, ‘The Anomeric Effect, and Related Stereoelectronic Effects
at Oxygen’, Springer-Verlag, Berlin, 1983, pp. 72–75.
39 A. J. Clark, Chem. Soc. Rev., 2002, 31, 1–11 discusses many of these,
and related issues.
40 G. Stork, R. Mook, S. A. Biller and S. D. Rychnovsky, J. Am. Chem.
Soc., 1983, 105, 3741–3742.
41 T. Itoh, K. Sakabe, K. Kudo, H. Ohara, Y. Takagi, H. Kihara,
P. Zagatti and M. Renou, J. Org. Chem., 1999, 64, 252–265. For
9 H. W. Chen, Z. B. Zhao, T. M. Hallis, Z. H. Guo and H. W. Liu,
Angew. Chem., Int. Ed., 2001, 40, 607–610. For the synthesis of a
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
540

View Article Online
related cyclisations of difluoroamides see : H. Nagashima, Y. Isono
and S. Iwamatsu, J. Org. Chem., 2001, 66, 315–319. The
conformational preference in amides is considerably less well
defined, as discussed in ref. 38, p. 114, meaning that amides can be
cyclised at that oxidation level, and do not require reduction to the
aminals.
Chem. Soc., 2000, 122, 3783–3784 raises may issues relevant to this
study. For a review of mechanistic work in the area, see : T. M. Trnka
and R. H. Grubbs, Acc. Chem. Res., 2001, 34, 18–29. For an
excellent recent review of factors influencing selectivity in cross-
metathesis, see: A. K. Chatterjee, T.-L. Choi, D. P. Sanders and
R. H. Grubbs, J. Am. Chem. Soc., 2003, 125, 11360–11370.
45 S. Imhof, S. Randl and S. Blechert, Chem. Commun., 2001, 1692–
1693.
42 T. R. Hoye and H. Zhao, Org. Lett., 1999, 1, 1123–1125.
43 D. J. O’Leary, H. E. Blackwell, R. A. Washenfelder, K. Miura and
R. H. Grubbs, Tetrahedron Lett., 1999, 40, 1091–1094. See also: D. J.
O’Leary, H. E. Blackwell, R. A. Washenfelder and R. H. Grubbs,
Tetrahedron Lett., 1998, 39, 7427–7430.
46 T. M. Trnka, M. W. Day and R. H. Grubbs, Angew. Chem., Int. Ed.
Engl., 2001, 40, 3441–3444.
47 D. V. McGrath and R. H. Grubbs, Organometallics, 1994, 13, 224–
235.
44 A. K. Chatterjee, J. P. Morgan, M. Scholl and R. H. Grubbs, J. Am.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 2 8 – 5 4 1
541