The reduction of α-silyloxy ketones using phenyldimethylsilyllithium

Article abstract of DOI:10.1039/a709114a

Phenyldimethylsilyllithium reacts with acyloin silyl ethers RCH(OSiMe₃)COR 8 to give regiodefined silyl enol ethers RCH=C(OSiMe₂Ph)R 9, and hence by hydrolysis ketones RCH₂COR 10. The yields can be high but are usually moderate. The mechanism of this reduction is established to involve a Brook rearrangement (Scheme 6) rather than a Peterson elimination (Scheme 1). Although the mechanism appears to be the same in each case, the stereochemistries of the silyl enol ethers 9 are opposite in sense in the aromatic series (R = Ph, Scheme 7) and the aliphatic series (R = cyclohexyl, Scheme 8), with the major aromatic silyl enol ether being the thermodynamically less stable isomer E-PhCH=C(OSiMe₂Ph)Ph E-9aa, and the major aliphatic silyl enol ether being the thermodynamically more stable isomer Z-c-C₆H₁₁CH= C(OSiMe₂Ph)-c-C₆H₁₁ Z-9ba. This is a consequence of anomalous anti-Felkin attack in the aromatic series. The reaction with the silyl ether Bu^tCH(OSiMe₃)COPh 13b is normal in giving Z-Bu^tCH= C(OSiMe₂Ph)Ph Z-38 (Scheme 11), but reduction of the silyl ether 8a with lithium aluminium hydride is also anti-Felkin giving with high selectivity the meso diol PhCH(OH)CH(OH)Ph 39. The reaction between Phenyldimethylsilyllithium and the acyloin silyl ether 8d (R = Bu^t) does not give the ketone Bu^tCH₂COBu^t, but gives instead the anti-Felkin meso diol Bu^tCHOHCHOHBu^t 40 also with high selectivity (Scheme 12). Silyllithium and some related reagents react with trifluoromethyl ketones 46 and 48 to give α,α-difluoro silyl enol ethers 47 and 49 (Scheme 14).

Full text of DOI:10.1039/a709114a

View Article Online / Journal Homepage / Table of Contents for this issue
The reduction of á-silyloxy ketones using phenyldimethylsilyllithium
Ian Fleming,*,a Richard S. Robertsa and Stephen C. Smithb
a Department of Chemistry, Lensfield Road, Cambridge UK CB2 1EW
b Zeneca Agrochemicals, Jealott’s Hill, Bracknell, Berkshire UK RG42 6ET
Phenyldimethylsilyllithium reacts with acyloin silyl ethers RCH(OSiMe3)COR 8 to give regiodefined silyl
enol ethers RCH᎐C(OSiMe Ph)R 9, and hence by hydrolysis ketones RCH COR 10. The yields can be
᎐
2
2
high but are usually moderate. The mechanism of this reduction is established to involve a Brook
rearrangement (Scheme 6) rather than a Peterson elimination (Scheme 1). Although the mechanism
appears to be the same in each case, the stereochemistries of the silyl enol ethers 9 are opposite in sense in
the aromatic series (R ؍ Ph, Scheme 7) and the aliphatic series (R ؍ cyclohexyl, Scheme 8), with the major
aromatic silyl enol ether being the thermodynamically less stable isomer E-PhCH᎐C(OSiMe Ph)Ph E-9aa,
᎐
2
and the major aliphatic silyl enol ether being the thermodynamically more stable isomer Z-c-C H CH᎐
᎐
6
11
C(OSiMe2Ph)-c-C6H11 Z-9ba. This is a consequence of anomalous anti-Felkin attack in the aromatic
series. The reaction with the silyl ether ButCH(OSiMe )COPh 13b is normal in giving Z-ButCH᎐
᎐
3
C(OSiMe2Ph)Ph Z-38 (Scheme 11), but reduction of the silyl ether 8a with lithium aluminium hydride is
also anti-Felkin giving with high selectivity the meso diol PhCH(OH)CH(OH)Ph 39. The reaction between
phenyldimethylsilyllithium and the acyloin silyl ether 8d (R ؍ But) does not give the ketone ButCH2COBut,
but gives instead the anti-Felkin meso diol ButCHOHCHOHBut 40 also with high selectivity (Scheme 12).
Silyllithium and some related reagents react with trifluoromethyl ketones 46 and 48 to give á,á-difluoro
silyl enol ethers 47 and 49 (Scheme 14).
limitations, which we report here. Our principal conclusions are
Introduction
that it is not generally high yielding, and that the mechanism is
different from that shown in Scheme 1. In our work we have
used the more easily made phenyldimethylsilyllithium rather
than trimethylsilyllithium, and have not therefore had to use
HMPA as an unavoidable co-solvent which may change the
mechanism. It is also possible that the mechanism changes with
specific substrates like Corey’s aldehyde 1, and the intermediate
5 would not be on the direct path that our reactions follow.
In their aphidicolin synthesis, Corey et al. reduced an
α-trimethylsilyloxy aldehyde 1 with trimethylsilyllithium to an
aldehyde 2 (Scheme 1).1 The mechanism that they suggested,
OSiMe3
CHO
i, ii, iii, iv
CHO
1
2
Results and discussion
i
iv
Our work is divided into four parts: (1) a study of the synthetic
usefulness of the reaction, for which we used the trimethylsilyl
ethers of acyloins and some of their other derivatives, (2) a
study of the mechanism of the reaction, for which we used the
acyloin silyl ethers 8a and 8b, (3) a study of the stereochemistry
of the reaction to explain the remarkable difference in the
geometries of the major silyl enol ether products 9 between the
aromatic and the aliphatic series, and finally (4) a brief study of
a related reaction between the silyllithium reagent and trifluoro-
methyl ketones 46 and 48, as a route to α,α-difluoromethyl
ketones.
SiMe3
O
O–
O–
SiMe3
OSiMe3
SiMe3
OSiMe3
3
4
6
ii
iii
OH
SiMe3
OSiMe3
The scope and limitations of the reaction
5
Our results on the general reaction with a small range of ‘sym-
metrical’ acyloin silyl ethers 8 and two ‘unsymmetrical’ acyloin
silyl ethers 13 are summarised in Scheme 2. At this stage, we
made no effort to isolate the intermediate silyl enol ethers 9 and
14, but hydrolysed the crude mixtures and isolated the ketones
10 and 15 directly. Some yields are very acceptable (15a; 92%),
but the general run of yields was not encouraging. The reason
appears to be that the silyllithium reagent attacks the silyl group
some of the time, in a reaction shown in the preceding paper to
be easy.3 In consequence, we detected the starting acyloin in the
product mixture in all these reactions.
Scheme 1 Reagents: i, Me3SiLi, HMPA; ii, work-up and chromatog-
raphy; iii, LDA, THF, HMPA; iv, H3Oϩ
supported by the isolation of the intermediate 5, involved the
1,4-oxygen-to-oxygen transfer of the silyl group 3→4 followed
by the β-silylalkoxide elimination 4→6, often referred to as a
Peterson elimination, a name that we find convenient and will
use here.
This sequence seemed to us to have been undervalued as a
method for reducing α-hydroxycarbonyl compounds, and as a
regiocontrolled method for making silyl enol ethers. Having
already provided one other example of the reaction taking
place in high yield,2 we undertook a study of its scope and
In the hope that varying the leaving group might make the
reaction higher yielding, we tried reaction with the acyloin
derivatives 16 (Scheme 3). Some of these yields, notably
J. Chem. Soc., Perkin Trans. 1, 1998
1215

View Article Online
of a silyllithium reagent on an aldehyde or ketone giving an
O
O
OSiMe2R
O
α-silyl alkoxide is a well used route to a α-silyl alcohols.5 The to
and fro shift of a silyl group in 1,2- and 1,3-diol systems is a
common problem in functional group protection,6 and the
Peterson elimination is the thoroughly established syn stereo-
specific second step of the reaction that is properly called
Peterson olefination.7
i
ii
iii
R
R
R
R
R
R
R
R
OH
7a R = Ph
b R = c-C6H11
c R = 2-furyl
d R = But
e R = Pri
R = nC5H11
OSiMe3
8a 92%
9
10a 59%
b
c
d
e
96%
79%
90%
90%
b
c
d
e
f
59%
59%
0%
43%
53%
However, it has also been established by
Hudrlik that 1,2-diols with a 1-silyl group 21 and 22 undergo
stereospecific 1,2-elimination by a Brook rearrangement fol-
lowed by, or concerted with, the anti expulsion of the β-hydroxy
group, summarised as a concerted reaction in 23→25 and
24→26, and that this pathway is faster in general than the more
direct seeming Peterson reaction in the same substrate (Scheme
5).8 Similarly Reich showed that his reactions 17→20, with
f 100%
f
O
Ph
Ph
iv, v
O
O
OSiMe2R
O
R1
R2
R1
R1
ii
iii
11
OH
OH
Ph
Ph
Ph
SiMe3
SiMe3
OSiMe3
R2
R2
OSiMe3
13a R1 = Ph, R2 = Me, 92% 14
b R1 = H, R2 = But, 65%
15a 92%
b 53%
OH
OH
21
22
vi, vii
Ph
CN
12
i
i
OH
OH
Scheme 2 Reagents: i, Me3SiCl, (Me3Si)2NH, CH2Cl2; ii, PhMe2SiLi,
THF; iii, NH4Cl, H2O; iv, MeLi; v, NaH, Me3SiCl; vi, LDA; vii,
ButCHO
O–
SiMe3
Me3Si
O–
23
25
24
26
O
O
i, ii
R
R
R
R
OSiMe3
X
7
10
a R = Ph
16 a R = Ph
b R = Ph
X = OH
X = ONa
a
a
a
30%
15%
92%
b 70%
b 50%
OSiMe3
X = O2CPh
Scheme 5 Reagents: i, KH, Et2O
c R = c-C6H11 X = OSiMe2But
d R = c-C6H11 X = O2CPh
better nucleofugal groups, were also stereospecifically anti.4
Although the formation of a pentacovalent intermediate would
be unexceptional, it is not clear whether the Brook rearrange-
ment involves a separate carbanion intermediate. It is however
known that in the absence of an adjacent leaving group, Brook
rearrangement can be consummated by stereospecific proton-
ation, with inversion of configuration when the silyl group is
benzylic9 and with retention of configuration when the
substituent α to the silyl group is alkyl.10 These pathways are
equally understandable with or without a carbanion inter-
mediate. We suggest that a pentacovalent siliconate is the only
‘carbanion’, and that there is unlikely to be another inter-
mediate either in the elimination or in the protonation pathway.
Hudrlik’s and Reich’s work suggested that the mechanism of
the reactions we had been studying might more likely be that
shown in Scheme 6, by way of the Brook rearrangement
e R = c-C6H11 X = OCO2Et
b 70%
b 46%
f
R = c-C6H11 X = OSO2Tol
Scheme 3 Reagents: i, PhMe2SiLi, THF; ii, NH4Cl, H2O
that from the benzoate 16b (92%), but regrettably not from
all, were better than those from the corresponding reactions
with the trimethylsilyl ethers 8 and 13.
Likewise, Reich et al. observed a 77% yield for the generation
of a silyl enol ether 20 (R1 = Ph, R2 = Me) by a similar reaction
of phenyldimethylsilyllithium with the α-phenylthio ketone 19,
but on the whole they obtained higher yields of silyl enol ethers
20 from the reaction of organolithium reagents with the corre-
sponding acyl silanes 17, either by a one-pot reaction or by
isolation of the α-silyl alcohol 18 and a subsequent base-
catalysed Brook rearrangement and elimination sequence
(Scheme 4).4
PhMe2Si O–
O
O
O
OSiMe2Ph
HO SiMe2R
i
R
R
i, ii
R
R
R
R1
SiMe2R
R1
R2
YPh
Ph
R
OSiMe3
OSiMe3
27
YPh
17
SPh
19
8
28
18
Scheme 6 Reagents: i, PhMe2SiLi, THF
iii
OSiMe2R
coupled with desilylative β-elimination (27 arrows) which we
call the Brook rearrangement route, in contrast to the Peterson
route of Scheme 1 as used by Corey, and also by us in our
earlier paper.2
iv
v
R1
R2
77%
Y = S, Se
40–94%
20
We tested which mechanism obtains by carrying out the reac-
tion of phenyldimethylsilyllithium on the trimethylsilyl ether
8a, and this time isolating the intermediate silyl enol ethers
(Scheme 7). If the Peterson route were to be followed the prod-
ucts would be trimethylsilyl enol ethers, but if the Brook
rearrangement route of Scheme 6 were followed, the products
would be phenyldimethylsilyl enol ethers. We found that the
Scheme 4 Reagents: i, R2M, Ϫ78 ЊC; ii, NH4OAc–MeOH; iii, base; iv,
R2M, Ϫ78→0 ЊC; v, PhMe2SiLi, THF
The mechanism of the reaction
The mechanism in the Corey paper is, of course, entirely
reasonable, with each step having good precedent. The attack
1216
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Table 1 NMR chemical shifts for the silyl enol ethers 9aa and 9ab. Authentic samples were prepared by reaction of ketone 10a with LDA and the
appropriate silyl chloride.
Authentic
trimethylsilyl enol
ethers 9ab
Authentic
dimethylphenylsilyl
enol ethers 9aa
Reaction products
8a ϩ PhMe2SiLi
E-9ab
Z-9ab
E-9aa
Z-9aa
Major
Minor
Proportion
δH(vinyl)
δH(SiMe)
δC(C-1)
δC(C-2)
δC(SiMe)
10
6.18
0.30
151.68
111.49
0.47
90
6.22
0.14
150.97
110.61
0.76
10
6.11
0.52
151.48
111.97
Ϫ0.94
90
6.16
0.34
150.90
110.94
Ϫ0.87
76
6.10
0.51
151.37
111.83
Ϫ1.05
24
6.15
0.35
≈150
110.82
Ϫ0.95
in the higher yielding reaction in which the mixture was
quenched with phenyldimethylsilyl chloride. We also carried
out molecular modelling calculations for the ground state-
energies of the trimethylsilyl enol ethers E-9ab and Z-9ab using
the MM2 force field and Monte Carlo minimisation, which
confirmed our expectations, the global minimum energies being
81.2 and 72.3 kJ molϪ1, respectively. We proved which isomer
was Z in the phenyldimethylsilyl case by an NOE difference
experiment on a mixture of both isomers. There was a clear
enhancement in intensity of the signal from the vinyl proton
upon irradiation of the signal from the silyl methyl group, and
vice versa in only one of the isomers—that to which we assign
the structure E-9aa. We assigned configurations to the corres-
Ph
1
2
Ph
OSiMe2Ph
28.5% or 40%
1.5% or 11%
0%
E-9aa
OSiMe2Ph
1
2
Ph
O
Ph
i
Ph
Ph
Z-9aa
or i, ii
Ph
1
OSiMe3
2
Ph
Ph
8a
OSiMe3
E-9ab
OSiMe3
1
2
Ph
SiMe2Ph
PhMe2Si
0%
Z-9ab
H
O
O
H
Scheme 7 Reagents: i, PhMe2SiLi, THF; ii, PhMe2SiCl
H
H
H
H
silyl enol ethers produced were indeed the phenyldimethylsilyl
enol ethers E-9aa and Z-9aa, in 30% yield and in a ratio of
95:5, and not the trimethylsilyl enol ethers 9ab. The low yield is
probably a result of the silyllithium reagent reacting with the
silyl enol ether to give the disilane and the lithium enolate, since
we find that this reaction does take place when carried out
deliberately, as discussed in more detail in the preceding paper.3
This hypothesis also explains why Reich’s yields using alkyl-
and aryl-lithium reagents on acylsilanes were higher, because
his reagents cleave the silyl enol ethers more slowly than our
silyllithium reagent does. In agreement with this hypothesis, we
obtained a better yield (51%) of the silyl enol ethers E-9aa and
Z-9aa by adding phenyldimethylsilyl chloride to the mixture
after the reaction was over. The ratio of these silyl enol ethers
was now 78:22, either because the enolate lost its configur-
ational identity to some extent, or because the Z-enolate was
desilylated more rapidly than the E-enolate. In conclusion, the
mechanism is that shown in Scheme 6, by way of the Brook
rearrangement.
Proving this simple point was a little more difficult than that
bald statement sounds. We prepared authentic samples of the
silyl enol ethers 9aa and 9ab by treating benzyl phenyl ketone
10a either with sodium hydride or LDA and either phenyl-
dimethylsilyl chloride or trimethylsilyl chloride. These reactions
were selective for the formation of the thermodynamically pre-
ferred Z-ethers, but we were able to detect and characterise (1H
NMR and 13C NMR) the minor E-isomers. We proved that
these were the thermodynamically preferred isomers by allow-
ing a mixture of the E- and Z-silyl enol ethers 9aa to stand in
deuteriochloroform for 5 days, after which time the E-isomer
had completely isomerised to Z-9aa. This equilibration meant
that any ratios of Z- and E-isomers that we measured were
somewhat dependent upon the time between isolating the silyl
enol ethers and taking their NMR spectra. This might also
contribute to the change in ratio of the silyl enol ethers E- and
Z-9aa obtained from the silyl ether 8a in the direct reaction and
E-9aa
Z-9aa
ponding trimethylsilyl enol ethers by analogy, supported by the
observation that the major isomers at equilibrium in both series
had the signals from the vinyl protons downfield and the signals
from the silyl methyl protons upfield of the signals from the
minor isomers. The signals in the 1H and 13C NMR spectra of
both isomers of the silyl enol ethers 9aa were identical with the
signals from the product of the reaction 8a→E-9aa ϩ Z-9aa,
and spiking the product mixture with a mixture of the tri-
methylsilyl enol ethers E-9ab and Z-9ab added four signals to
make a total of eight signals from vinyl carbons. The diagnostic
peaks in the NMR spectra of each of the four silyl enol ethers
are listed in Table 1.
We have therefore shown that this version of Corey’s reaction
takes the Brook rearrangement pathway, but it is not clear that
this will always be the case, since Brook rearrangement is well
known to be much easier when there is a phenyl group stabilis-
ing the carbanion. We therefore examined the corresponding
reaction of the acyloin silyl ether 8b, with cyclohexyl groups in
place of the phenyl groups. As before, we found that the silyl
enol ethers produced were the phenyldimethylsilyl enol ethers
E-9ba and Z-9ba, in 57% yield and in a ratio of 20:80, and not
the trimethylsilyl enol ethers E-9bb and Z-9bb (Scheme 8).
Again the products were those expected of the Brook rearrange-
ment pathway and not the Peterson.
As in the aromatic series, we prepared authentic samples of
the silyl enol ethers 9ba and 9bb by treating the ketone 10b with
lithium tetramethylpiperidide and either phenyldimethylsilyl
chloride or trimethylsilyl chloride. These reactions were not
completely selective for the formation of the E- and Z-silyl enol
ethers 9ba and 9bb, with a further complication that these com-
pounds were susceptible to equilibration, particularly when
J. Chem. Soc., Perkin Trans. 1, 1998
1217

View Article Online
Table 2 NMR chemical shifts for the silyl enol ethers 9ba and 9bb
Authentic
trimethylsilyl enol
ethers 9bb
Authentic
dimethylphenylsilyl
enol ethers 9ba
Reaction products
8b ϩ PhMe2SiLi
E-9bb
Z-9bb
E-9ba
Z-9ba
Major
Minor
Proportiona
δH(vinyl)
δH(SiMe)
δC(C-1)
δC(C-2)
δC(SiMe)
20
4.34
0.16
154.55
112.70
0.57
49
4.27
0.18
153.40
112.30
0.69
6
4.34
0.41
154.34
112.68
Ϫ0.75
31
4.27
0.45
153.37
112.36
Ϫ0.80
73
4.25
0.43
153.33
112.37
Ϫ0.82
18
4.29
0.40
154.29
112.70
Ϫ0.74
a The E and Z proportions are the percentages of the total silyl enol ether content, the remainder being the regioisomer 29.
cis cyclohexyl group. We assigned the stereochemistry to
the phenyldimethylsilyl enol ethers by analogy, supported by
C6H11-c
1
2
c-C6H11
OSiMe2Ph
11%
the observation that the major isomers at equilibrium in each
case had the signals from their vinyl protons upfield and their
E-9ba
OSiMe2Ph
1
silyl methyl groups downfield. The signals in the H and 13C
1
NMR spectra of both isomers of the silyl enol ethers 9ba pre-
pared from the ketone 10b were identical with the signals from
the products of the reaction 8b→E-9ba ϩ Z-9ba, and spiking
the product mixture with a mixture of the trimethylsilyl enol
ethers E-9bb and Z-9bb added three signals to make a total of
seven signals from vinyl carbons, two of the eight peaks being
coincident. The diagnostic peaks in the NMR spectra of each
of the four silyl enol ethers are listed in Table 2. Thus the reac-
tion of phenyldimethylsilyllithium with the silyl ether 8b gave
the silyl enol ethers 9ba in 57% yield with an E:Z ratio of
20:80.
2
c-C6H11
O
C6H11-c
46%
i
c-C6H11
Z-9ba
C6H11-c
C6H11-c
OSiMe3
1
2
c-C6H11
c-C6H11
8b
OSiMe3
0%
0%
E-9bb
OSiMe3
1
2
C6H11-c
Z-9bb
The stereochemistry
It is noteworthy that, whereas the major product in the aliphatic
series is the thermodynamically more stable silyl enol ether
Z-9ba, the major product in the aromatic series is the less stable
silyl enol ether E-9aa. The stereochemical outcome in the ali-
phatic series is what one would expect—attack on the ketone 8b
can be expected to be governed by the Cornforth modification
of Cram’s rule, or its Felkin–Anh equivalent, hereinafter
referred to as Felkin, to give predominantly the intermediate
30, which will undergo anti elimination, following the prece-
dents of Hudrlik and of Reich, to give the silyl enol ether Z-9ba
(Scheme 9). To explain the anomalous result in the aromatic
series, however, either the attack on the ketone 8a had given the
anti-Felkin product 31, which undergoes anti elimination, or
the Felkin product 32 had given the silyl enol ether E-9aa by
unprecedented syn elimination. It is not clear why the ketone 8a
should not obey Felkin’s rule—chelation is no more likely in
this ketone than in the aliphatic ketone 8b. Elimination with syn
stereochemistry, on the other hand, is not unreasonable,
because it is, in a sense, the corollary of the inversion of con-
figuration seen in Brook rearrangements followed by proton-
ation when the silyl group is attached to a benzylic position.9
However, Reich and his co-workers reported that the α-
phenylthio acyl silane 17 (R = Me, R1 = Ph, Y = S) reacted with
both methyllithium and phenyllithium to give largely the silyl
enol ethers 20 (R = Me, R1 = Ph, R2 = Me or Ph) with the same
E-geometry. We would, if syn elimination were taking place in
the benzyl series, have expected the reaction with phenyllithium
to give the opposite stereochemistry. They also carried out the
reaction of phenyldimethylsilyllithium on the methyl ketone 19,
in a reaction which is analogous to our reaction 8b→Z-9ba, and
obtained the Z-silyl enol ether 20 (R = R1 = Ph, R2 = Me) pre-
sumably by way of a Felkin intermediate and anti elimination.
The missing experiment in their series was the reaction of the
ketone 33 with phenyldimethylsilyllithium, which ought, if syn
elimination were taking place in the benzyl series, to have given
the E-isomer of the silyl enol ether 35, by analogy with our
reaction 8a→E-9aa. We have therefore carried out this reaction,
OSiMe2Ph
OSiMe3
29a
29b
Scheme 8 Reagents: i, PhMe2SiLi, THF
they were being distilled, giving the regioisomeric silyl enol
ethers 29a and 29b, respectively. We again carried out molecular
modelling calculations for the trimethylsilyl enol ethers E-9bb
and Z-9bb, which had global minimum energies of 91.3 and
80.5 kJ molϪ1, respectively. We proved which isomer was which
in the trimethylsilyl series by 2D COSY and NOESY spectra
using a mixture of E-9bb, Z-9bb and 29b. In the isomer
Me
Si
Me
Si
Me
Me
H1
Me
Me
O
O
H1
H2
H3
H3 H2
Z-9bb
E-9bb
Dashed lines are weak enhancements. The structures drawn are similar
to calculated minimum-energy conformations.
to which we assigned the structure E-9bb, we saw strong cross
peaks between the signals from the vinyl protons and the signals
from the silyl methyl groups, and between both cyclohexyl
methine protons. In the other isomer, we saw a weak cross peak
between the signal from the vinyl proton and the signal from
the silyl methyl groups, and a definitive cross peak between
the signal from the vinyl proton and the methine signal of the
1218
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
We assigned stereochemistry to this pair of products from the
O
OSiMe2Ph
C6H11-c
observations that the major isomer was identical with the major
(98.5:1.5) product from the silylation of the corresponding
ketone, and there was an NOE enhancement in the signal from
the tert-butyl protons upon irradiation of the signal from the
silyl methyl protons only in the major isomer. Thus the stereo-
chemical outcome in this case was the normal pattern of Felkin
control followed by anti elimination, still leaving the stereo-
chemical result with the ketone 8a anomalous. Reduction of the
silyl ether 8a using lithium aluminium hydride gave very largely
(95:5) the meso diol 39, which is also the anti-Felkin product
(Scheme 12). Similarly, in our one failure in Scheme 2, the
i
c-C6H11
c-C6H11
C6H11-c
OSiMe3
Z-9ba
8b
–O SiMe2Ph
c-C6H11
C6H11-c
OSiMe3
Felkin
anti
30
anti
Felkin
PhMe2Si O–
Ph
anti
O
OH
i, ii
i
Ph
Ph
Ph
O
Ph
Ph
Ph
OSiMe3
Ph
OSiMe3
OH
Ph
Ph
31
OSiMe2Ph
OSiMe3
8a
39 73% 94:6
E-9aa
–O SiMe2Ph
8a
i
Ph
O
OH
Ph
iii, iv
syn
Felkin
But
But
But
But
OSiMe3
32
OSiMe3
OH
40 21% 94:6
8d
Scheme 9 Reagents: i, PhMe2SiLi, THF
Scheme 12 Reagents: i, LiAlH4; ii, Rochelle’s salt, MeOH; iii, PhMe2-
O
SiLi, THF; iv, NH4Cl, H2O
OSiMe2Ph
Ph
i
Ph
Ph
Ph
product that we obtained from the reaction of phenyldimethyl-
silyllithium on the silyl ether 8d was largely (95:5) the
meso diol 40, which is the result from attack in the anti-Felkin
sense followed by Brook rearrangement–protonation with
retention of configuration. Thus, when both groups are phenyl
or tert-butyl, the Felkin rule breaks down.
SPh
35
33
–O SiMe2Ph
Ph
Ph
Felkin
anti
SPh
In conclusion, acyloin silyl ethers can be reduced to silyl enol
ethers using phenyldimethylsilyllithium. The pathway is that
shown in Scheme 6, and the yield is less than useful because
some of the silyllithium reagent attacks the silyl group, return-
ing the starting acyloin. A further complication is that the
product silyl enol ether is also attacked by the silyllithium
reagent to give the lithium enolate.
34
Scheme 10 Reagents: i, PhMe2SiLi, THF
and obtained the silyl enol ether 35 with an E:Z ratio of <1:99
(Scheme 10), the major product corresponding to anti elimin-
ation from the Felkin intermediate 34. Hence in all four
examples in Reich’s series, the stereochemistry appears to be
attack on the carbonyl group in the Felkin sense followed by
anti elimination, regardless of whether the intermediate oxy-
anion is benzylic as in 34 or not.
Since the elimination step was always anti, we tried to verify
whether or not the initial attack on the ketone 8a was under
Felkin control, since anti-Felkin attack followed by anti elimin-
ation would have led to the observed E-silyl enol ether 9aa. We
therefore looked at the stereochemistry of reaction of phenyl-
dimethylsilyllithium with the silyl ether 13b in which one of the
phenyl groups had been replaced by another large group, the
tert-butyl moiety. We obtained the silyl enol ethers 38 in 53%
yield and with an E:Z ratio of 12:88 (Scheme 11).
The reaction of the silyllithium reagent with trifluoromethyl
ketones
Related to all the work described above is the possibility of
making difluorinated silyl enol ethers 41, by treating trifluoro-
methyl ketones 45 with the silyllithium reagent. These interest-
ing synthons11 have been made (Scheme 13) by the other
obvious combinations of nucleophile and electrophile: gener-
ation of the zinc enolate from the chlorodifluoro ketone 42 in
Ref. 11
Ref. 12
Ref. 13
O
+
+
Zn
R3SiCl
R
R
CF2Cl
SiR3
CF3
42
O
anti
F–
R3SiCF3
Ph
+
+
Felkin
PhMe2Si O–
But
anti
OSiR3
But
OSiMe2Ph
F
i
43
12%
Ph
R
O
E-38
Z-38
OSiMe3
F
O
But
RMgX
+
Ph
RLi or
36
41
R3Si
OSiMe3
–O SiMe2Ph
OSiMe2Ph
44
13b
anti
i
But
But
?
Ph
O
Ph
Felkin
88%
R3SiLi
+
OSiMe3
R
CF3
37
45
Scheme 11 Reagents: i, PhMe2SiLi, THF
Scheme 13
J. Chem. Soc., Perkin Trans. 1, 1998
1219

View Article Online
the presence of a silyl chloride,12 reaction of trifluoromethyl
and other fluorinated ‘anions’ with acyl silanes 43,13 and reac-
tion of organometallic ‘anions’ RϪ with trifluoromethylacyl
silanes 44.14 Both the phenyldimethylsilyllithium reagent and
trifluoromethyl ketones 45 are easy to make, and should react
with ease, since the silyllithium reagent is exceptionally nucleo-
philic and trifluoromethyl ketones are exceptionally electro-
philic ketones. Accordingly, we have investigated the missing
possibility.
Phenyldimethylsilyllithium reacted rapidly with (trifluoro-
acetyl)benzene 46 at Ϫ78 ЊC. The colour was completely dis-
charged by only half an equivalent of the ketone. Presumably
the silyl enol ether 47a was being cleaved by the silyllithium
reagent as fast as it was being formed (Scheme 14). We were
strate some potential for these compounds. They both reacted
with benzaldehyde in a Mukaiyama-aldol reaction to give the
difluorinated β-hydroxy ketone 53 (Scheme 15).
OSiR3
O
OH
i
F
Ph
Ph
Ph
+ OHCPh
F
F
F
47a R3 = Me2Ph
b R3 = Ph2But
53 38% from 47a
31% from 47b
Scheme 15 Reagents: i, TiCl4, CH2Cl2
Experimental
Compounds 7a, 7c and 11 were available commercially. Light
petroleum refers to the fraction bp 40–60 ЊC. Ether refers to
diethyl ether. Standard work-up with a salt (e.g. ammonium
chloride) refers to the procedure of quenching the reaction mix-
ture with, typically, an equal volume of an aqueous solution of
the salt (saturated unless stated otherwise). The mixture was
then extracted with a comparable volume of ether (3×), the
combined organic fractions were washed with brine, dried
(MgSO4) and evaporated under reduced pressure.
O
OSiR3
i or ii or iii
or iv or v
F
F
F
Ph
Ph
F
F
46
47a R3 = Me2Ph, (ii) 15%, (iii) 19%
b R3 = Ph2But, (iv) 51%, (v) 49%
OSiMe2Ph
i, vi
Ph
Ph
Ph
F
tert-Butyl(diphenyl)silyllithium
F
Lithium shot (0.4 g in mineral oil, 55 mmol) was stirred rapidly
for 15 min in dry hexane (20 cm3) under argon. The hexane was
removed and the lithium re-suspended in dry THF (30 cm3).
The flask was cooled to 0 ЊC in an ice-bath and the lithium
surface activated by addition of chlorotrimethylsilane (0.01
cm3, 0.08 mmol). tert-Butylchloro(diphenyl)silane (5.0 cm3,
19.2 mmol) was added and the mixture was stirred rapidly for
7 h at 0 ЊC to give tert-butyl(diphenyl)silyllithium as a red–
brown solution. The molarity of the solution was determined by
the Gilman double titration method (typically 0.3 mol dmϪ3).
49 37%
O
O
i, vii
Ph
F
F
F
F
F
48
50 59% (as 2,4-DNP)
HO
iv
ButPh2SiH
F
F
+
The acyloins 7
F
Typically, the ester and an equal volume of dry ether were
added dropwise to a stirred suspension of finely divided sodium
(2.1 molar equivalents) and trimethylsilyl chloride (2.1 molar
equivalents) in dry ether. After 5 min a vigorous reaction
started and the ester was added at a rate to maintain reflux; the
mixture was refluxed under argon until a white precipitate had
replaced the sodium (typically 1–3 d). The mixture was poured
into enough water to dissolve the precipitate, acidified with
dilute hydrochloric acid (2 mol dmϪ3) and the organic layer
separated. The aqueous layer was washed twice with ether, the
organic layers combined and stirred with methanol for 1 h.
Evaporation under reduced pressure and further purification
gave the acyloin. The following compounds were prepared by
this method.
51 60%
52 76%
Scheme 14 Reagents: i, PhMe2SiLi; ii, PhMe2SiMgMe; iii, PhMe2SiLi,
ZnBr2; iv, Ph2ButSiMgMe; v, Ph2ButSiLi, ZnBr2; vi, PhMe2SiCl; vii,
NH4Cl, H2O
unable to trap the resultant enolate with silyl halides, acetyl
chloride, benzyl bromide or crotonaldehyde. However, we did
isolate the silyl enol ether 49 in 37% yield when we treated the
superficially more problematic trifluoromethyl ketone 48 with
an excess of the silyllithium reagent, and quenched with phenyl-
dimethylsilyl chloride. Working up with a proton source
instead, did give the difluoromethyl ketone 50 in up to 59%
yield, but we were unable to trap the intermediate enolate in any
other way.
A mixed phenyldimethylsilyl(methyl)magnesium reagent did
give us the silyl enol ether 47a from (trifluoroacetyl)benzene,
again in low yield (15%), as did a zinc bromide-mediated reac-
tion with the silyllithium reagent (19%). Changing to a more
hindered silyl group, in the hope of isolating a more stable silyl
enol ether, gave us a better yield—we obtained the silyl enol
ether 47b in 51% yield from a mixed tert-butyldiphenyl-
silyl(methyl)magnesium reagent and in 49% yield from the
zinc bromide-mediated reaction with tert-butyldiphenylsilyl-
lithium. In contrast, the ketone 48 gave the alcohol 51, together
with the silane 52, by unprecedented methyl transfer from the
mixed reagent.
1,2-Dicyclohexyl-2-hydroxyethanone 7b. Isolated as a pale
yellow solid [6.44 g, 63% from ethyl cyclohexanecarboxylate
(13.07 g)] after distillation, bp 172–182 ЊC at 22 mmHg; mp 39–
42 ЊC (lit.,15 mp 44–45 ЊC); Rf (EtOAc–light petroleum, 5:95)
0.25; νmax(film)/cmϪ1 3475 (br, OH) and 1702 (C᎐O); δ (250
᎐
H
MHz; CDCl3) 4.1 (1 H, m, CHOH), 3.4 (1 H, d, J 6.5, OH), 2.5
(1 H, m, COCH) and 1.8–1.0 (21 H, m, CHCHOH and
10 × CH2).
2,2,5,5-Tetramethyl-4-hydroxyhexan-3-one 7d. Isolated as
needles [4.11 g, 24% from methyl trimethylacetate (26.55 cm3)]
after distillation (Kugelrohr, 140 ЊC at 17 mmHg), mp 80–
81.5 ЊC (from toluene at Ϫ78 ЊC) (lit.,16 mp 80–81 ЊC, bp
85–87 ЊC at 20 mmHg); Rf (EtOAc–light petroleum, 20:80)
0.41; νmax(Nujol)/cmϪ1 3473 (OH) and 1698 (C᎐O); δ (250
In conclusion, several experiments, including some not
reported here, have shown that silyl enol ethers can be prepared
from trifluoromethyl ketones, but the reaction needs more study
before it is reliable enough to be used in synthesis. With one
reaction, we used the silyl enol ethers 47a and 47b to demon-
᎐
H
MHz; CDCl3) 4.20 (1 H, d, J 10.5, CHOH), 2.27 (1 H, d,
J 10.5, CHOH), 1.20 (9 H, s, COCMe3) and 1.00 (9 H, s,
CHOHCMe3).
2,5-Dimethyl-4-hydroxyhexan-3-one 7e. Isolated as an oil [3.0
1220
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
g, 41% from methyl isopropanoate (16.2)] after distillation and
chromatography (SiO2, EtOAc–light petroleum, 10:90); bp 72–
74 ЊC at 15 mmHg (lit.,17 55–57 ЊC at 3 mmHg); Rf (EtOAc–
light petroleum, 15:85) 0.36; νmax(film)/cmϪ1 3480 (OH) and
Ϫ0.1; m/z (EI) 201 (85%, M Ϫ Me), 145 (100, M Ϫ Me2CHCO)
and 73 (85, SiMe3) (Found: Mϩ Ϫ Me, 201.1311. C11H24-
O2Si Ϫ Me requires M, 201.1311).
7-(Trimethylsilyloxy)dodecan-6-one 8f.22 Isolated as an oil
[5.80 g, 100% from acyloin 7f (4.27 g)]; Rf (EtOAc–light petrol-
1707 (C᎐O); δ (250 MHz; CDCl ) 4.22 (1 H, d, J 2, CHOH),
᎐
H
3
2.82 (1 H, septet, J 7, COCHMe2), 2.15 (1 H, d, septet, J 2 and
7, CHOHCHMe2), 1.2–1.0 (9 H, m, 3 × Me) and 0.68 (3 H, d,
J 7, Me).
eum, 10:90) 0.54; νmax(film)/cmϪ1 1716 (C᎐O) and 1252
᎐
(SiMe3); δH(250 MHz; CDCl3) 3.95 (1 H, t, J 6.5, CHOSi), 2.48
(2 H, t, J 7, CH2CO), 1.5 (4 H, m, 2 × CH2), 1.4–1.2 (10 H, m,
5 × CH2), 0.9–0.8 (6 H, m, 2 × CH2Me), 0.32 (5 H, s, SiMe3)
and 0.23 (4 H, s, SiMe3).
7-Hydroxydodecan-6-one 7f. Colourless oil [12.04 g, 60%
from methyl hexanoate (29.42 cm3, 200 mmol)] after distil-
lation, bp 141–146 ЊC at 17 mmHg (lit.,17 105–107 ЊC at 3
mmHg); Rf (EtOAc–light petroleum, 10:90) 0.25; νmax(film)/
2-Phenyl-2-(trimethylsilyloxy)acetonitrile 12
cmϪ1 3480 (br, OH) and 1711 (C᎐O); δ (250 MHz; CDCl ) 4.15
Benzaldehyde (3.6 cm3, 36 mmol) was added to a stirred sus-
pension of anhydrous zinc() iodide (0.33 g, 0.10 mmol) in tri-
methylsilyl cyanide (5.3 cm3, 40 mmol) and dry acetonitrile (20
cm3) under argon. The initial exothermic reaction was cooled in
an ice-bath. The mixture was stirred for 1 h at room temper-
ature, diluted with hexane (25 cm3) and shaken with ice–water
(100 cm3). The aqueous layer was washed with hexane (3 × 40
cm3), the combined organic extracts washed with cold, satur-
ated sodium bisulfite solution (60 cm3) and dried (MgSO4). The
solvent was evaporated under reduced pressure and the crude
product distilled (Kugelrohr, 180 ЊC at 20 mmHg) (lit.,23 64 ЊC
at 0.5 mmHg) to give the silyl ether (6.3 g, 85%) as a pale yellow
liquid; Rf (EtOAc–light petroleum, 20:80) 0.56; νmax(film)/cmϪ1
1602 (Ph); δH(270 MHz; CDCl3) 7.5–7.3 (5 H, m, Ph), 5.50 (1 H,
s, CHOSi) and 0.24 (9 H, s, SiMe3).
᎐
H
3
(1 H, m, CHOH), 3.50 (1 H, d, J 4, OH), 2.44 (1 H, t, J 7.5,
CHAHBCO), 2.42 (1 H, t, J 7.5, CHAHBCO), 1.8–1.2 (12 H, m,
6 × CH2), 1.60 (2 H, quintet, J 7.5, CH2) and 0.87 (6 H, t, J 6.5,
2 × Me).
The silyl ethers 8
Following Cossy and Pale,18 typically, hexamethyldisilazane
(1.3 molar equivalents) and trimethylsilyl chloride (1.3 molar
equivalents) were added to a stirred solution of the acyloin 7
(0.1–0.4 mol dmϪ3 solution in dry CH2Cl2) under argon at 0 ЊC.
The solution was stirred at room temperature until no starting
material was visible by TLC. The mixture was cooled to 0 ЊC,
filtered through a plug of Florosil, evaporated under reduced
pressure, the residue taken up in hexane, filtered again and
re-evaporated to give the silyl ether. The following silyl ethers
were prepared by this method.
1,2-Diphenyl-2-(trimethylsilyloxy)propan-1-one 13a
1,2-Diphenyl-2-(trimethylsilyloxy)ethanone 8a. Isolated as a
pale yellow solid [1.35 g, 92% from benzoin 7a (1.09 g)], mp 79–
80 ЊC (lit.,19 78–79.5 ЊC); Rf (EtOAc–light petroleum, 50:50)
Methyllithium (1.4 mol dmϪ3 in ether, 10 cm3) was added over
25 min to a stirred solution of benzil 11 (2.96 g, 14 mmol) in
THF at Ϫ78 ЊC under argon and stirred for 90 min. Standard
work-up with ammonium chloride followed by chromato-
graphy (SiO2, EtOAc–light petroleum, 10:90) gave 1,2-diphenyl-
2-hydroxypropan-1-one (2.36 g, 75%) as needles [from light
petroleum (bp 60–80 ЊC)], mp 65–67 ЊC (lit.,24 65–66 ЊC); Rf
(EtOAc–light petroleum, 10:90) 0.18; νmax(Nujol)/cmϪ1 3455
0.68; νmax(Nujol)/cmϪ1 1688 (C᎐O), 1596 (Ph) and 1578 (Ph);
᎐
δH(250 MHz; CDCl3) 8.1–7.9 (2 H, m, PhCO o-H), 7.5–7.2
(8 H, m, other Ph), 5.83 (1 H, s, CHOSi) and 0.11 (9 H, s,
SiMe3).
1,2-Dicyclohexyl-2-(trimethylsilyloxy)ethanone 8b. Isolated
as an oil [2.45 g, 96% from acyloin 7b (1.92 g)]; Rf (EtOAc–light
(br, OH), 1677 (C᎐O), 1597 and 1577 (Ph); δ (250 MHz;
᎐
H
petroleum, 10:90) 0.64; νmax(film)/cmϪ1 1706 (C᎐O) and 1250
CDCl3) 7.67 (2 H, dd, J 8 and 1.5, PhCO o-H), 7.5–7.2 (8 H, m,
other Ph), 4.76 (1 H, s, OH) and 1.91 (3 H, s, Me). The acyloin
(0.71 g, 3.2 mmol) in dry THF (10 cm3) was added to a stirred
suspension of hexane-washed sodium hydride (60% dispersion
in oil, 0.12 g, 3.2 mmol) in dry THF at 0 ЊC under argon and
was stirred for 45 min. Trimethylsilyl chloride (0.52 cm3, 4.1
mmol) was added, forming a white precipitate, and the mixture
stirred for 2 h. The mixture was filtered though a pad of Florisil
and the solvent evaporated under reduced pressure. Chromato-
graphy (Al2O3, EtOAc–light petroleum, 10:90) gave the silyl
ether (0.77 g, 81%) as a solid, mp 50–52 ЊC (lit.,25 51–52 ЊC); Rf
(EtOAc–light petroleum, 15:85) 0.73; νmax(Nujol)/cmϪ1 1679
᎐
(SiMe3); δH(250 MHz; CDCl3) 3.78 (1 H, d, J 5.5, CHCOSi),
2.74 (1 H, m, CHCOCHOSi), 1.8–1.0 (21 H, m, CHCHOSi and
10 × CH2) and 0.07 (9 H, s, SiMe3); δC(CDCl3 215.5, 82.9, 45.4,
41.3, 29.6, 29.1, 28.9, 27.9, 26.3, 26.3, 26.0, 25.8, 25.6, 25.5 and
0.2; m/z (EI) 281 (25%, M Ϫ Me), 185 (100, M Ϫ c-C6H11CO),
103 (90, H C᎐OSiMe ), 95 (100), 83 (50, c-C H ) and 73
᎐
2
3
6
11
(95, SiMe3) (Found: Mϩ, 296.2176. C17H32O2Si requires M,
296.2171).
1,2-Di(2-furyl)-2-(trimethylsilyloxy)ethanone 8c. Isolated as
a brown solid [0.875 g, 79% from furoin 7c (0.814 g)], mp 42–
44 ЊC (lit.,20 44–45 ЊC); Rf (EtOAc–light petroleum, 20:80) 0.33;
νmax(Nujol)/cmϪ1 3015 (vinyl C᎐H), 1683 (C᎐O) and 1250
(C᎐O), 1597 (Ph), 1580 (Ph) and 1254 (SiMe ); δ (250 MHz;
᎐
᎐
3
H
(SiMe3); δH(250 MHz; CDCl3) 7.59 (1 H, dd, J 1 and 0.5,
furanoyl H-5), 7.41 (1 H, dd, J 4 and 0.5, furanoyl H-3), 7.36
(1 H, dd, J 1.5 and 1, furyl H-5), 6.51 (1 H, dd, J 3.5 and 1.5,
furyl H-3), 6.3 (2 H, m, furyl H-4 and furanoyl H-4),
5.73 (1 H, s, CHOSi) and 0.12 (9 H, s, SiMe3).
CDCl3) 7.91 (2 H, dd, J 7.5 and 1.5, PhCO o-H), 7.50 (2 H, dd,
J 7 and 1.5, PhCOSi o-H), 7.5–7.2 (6 H, m, other Ph), 1.80 (3 H,
s, Me) and 0.07 (9 H, s, SiMe3).
3,3-Dimethyl-1-phenyl-2-(trimethylsilyloxy)butan-1-one 13b
Following Hünig and Wehner,25 n-butyllithium (1.27 mol dmϪ3
in hexane, 8.7 cm3) was added dropwise to a stirred solution of
diisopropylamine (1.6 cm3, 11 mmol) in dry THF (10 cm3)
under argon at 0 ЊC. After 20 min, the solution was cooled to
Ϫ78 ЊC, silyl ether 12 (2.05 g, 10 mmol) in dry THF (5 cm3) and
trimethylacetaldehyde (1.08 cm3, 10 mmol) were added, main-
taining the temperature below Ϫ60 ЊC. The mixture was
allowed to warm over 2 h, precipitating lithium cyanide. When
the temperature reached Ϫ5 ЊC, dichloromethane (25 cm3) and
water (5 cm3) were added. The organic layer was washed with
saturated ammonium chloride solution (2 × 30 cm3), dried
(MgSO4) and evaporated under reduced pressure. Chrom-
atography (SiO2, EtOAc–hexane, 2.5:97.5) gave the silyl ether
(1.72 g, 65%) as prisms, mp 38–41 ЊC; Rf (EtOAc–hexane, 5:95)
2,2,5,5-Tetramethyl-4-(trimethylsilyloxy)hexan-3-one 8d. Iso-
lated as a pale yellow solid [0.96 g, 86% from acyloin 7d (0.79
g)], mp 35–37 ЊC (lit.,21 37–39 ЊC); Rf (EtOAc–light petroleum,
20:80) 0.66; νmax(Nujol)/cmϪ1 1716 (C᎐O) and 1252 (SiMe );
᎐
3
δH(250 MHz; [2H6]DMSO) 4.36 (1 H, s, CHOSi), 1.13 (9 H, s,
COCMe3) 0.91 (9 H, s, SiOCCMe3) and 0.13 (9 H, s, SiMe3).
2,5-Dimethyl-4-(trimethylsilyloxy)hexan-3-one 8e. Isolated as
an oil [1.34 g, 90% from acyloin 7e (1.00 g)]; Rf (EtOAc–light
petroleum, 50:50) 0.73; νmax(film)/cmϪ1 1712 (C᎐O) and 1252
᎐
(SiMe3); δH(250 MHz; CDCl3) 3.82 (1 H, d, J 5.5, CHOSi), 3.00
(1 H, septet, J 7, COCHMe2), 1.97 (1 H, m, CHCHMe2), 1.05
(6 H, d, J 7, COCHMe2), 0.87 (3 H, d, J 7, CHCHMeAMeB),
0.83 (3 H, d, J 7, CHCHMeAMeB) and 0.08 (9 H, s, SiMe3);
δC(CDCl3) 216.5, 83.1, 35.1, 31.7, 19.2, 19.0, 18.9, 17.4 and
J. Chem. Soc., Perkin Trans. 1, 1998
1221

View Article Online
0.48; νmax(film)/cmϪ1 1691 (C᎐O), 1597 (Ph), 1578 (Ph), 1252
(15%, Mϩ), 245.1 (100, M Ϫ c-C6H11), 223 (18, M Ϫ COPh),
᎐
(SiMe3), 1110 (Si᎐O) and 841 (SiMe3); δH(270 MHz; CDCl3)
8.05 (2 H, dd, J 7.5 and 1, o-H), 7.5 (1 H, m, p-H), 7.4 (2 H, dt,
J 1 and 7.5, m-H), 4.53 (1 H, s, CHOSi), 0.92 (9 H, s, CMe3)
and 0.05 (9 H, s, SiMe3); δC(CDCl3) 202.1, 138.0, 132.9, 129.6,
128.4, 83.7, 35.8, 27.0 and 0.0.
217.1 (60, M Ϫ c-C6H11CO), 206 (25, M Ϫ PhCOOH), 105
(100, PhCO), 83 (50, c-C6H11) and 77 (20, Ph) (Found: Mϩ,
328.2014. C21H28O3 requires M, 328.2016).
Ethyl 1,2-dicyclohexyl-2-oxoethyl carbonate 16e
Methyllithium (1.4 mol dmϪ3 in ether, 3.75 cm3) was added to
a stirred solution of acyloin 7b (1.05 g, 5.0 mmol) in dry THF
(5 cm3) at Ϫ78 ЊC under argon, forming a white precipitate.
After 15 min, the mixture was warmed to 0 ЊC, ethyl chloro-
formate (0.50 cm3, 5.3 mmol) was added and the solution
stirred for 6 h, and kept overnight. Standard work-up with
sodium hydrogen carbonate and chromatography (SiO2,
EtOAc–light petroleum, 8:92) gave the carbonate (0.92 g, 61%)
as needles, mp 59–60 ЊC (from MeOH); Rf (EtOAc–light petrol-
eum, 8:92) 0.31; νmax(Nujol)/cmϪ1 1746 (C᎐O) and 1717 (C᎐O);
Benzoin, sodium salt 16a
Benzoin 7a (0.50 g, 2.36 mmol) in dry THF (10 cm3) was added
to a stirred suspension of hexane-washed sodium hydride (94
mg of a 60% suspension in oil) in THF (2 cm3) at 0 ЊC and used
without further purification.
1,2-Diphenyl-2-oxoethyl benzoate 16b
Pyridine (0.45 cm3, 5.6 mmol), benzoyl chloride (0.65 cm3, 5.6
mmol), benzoin 7a (1.00 g, 4.7 mmol) and dry dichloromethane
(25 cm3) were stirred for 90 min at 0 ЊC and then 90 min at room
temperature, under argon. More benzoyl chloride (0.65 cm3, 5.6
mmol) was added and the mixture stirred for a further 90 min.
The solution was washed with hydrochloric acid (3 mol dm3, 50
cm3) and saturated sodium hydrogen carbonate solution (2 × 50
cm3), dried (MgSO4) and evaporated under reduced pressure to
give a crude white solid which was recrystallised to give the
benzoyl ester (0.658 g, 44%) as needles, mp 121–123 ЊC (from
MeOH) (lit.,26 123–124 ЊC); Rf (EtOAc–light petroleum, 50:50)
0.50; νmax(Nujol)/cmϪ1 1715 (ester C᎐O), 1695 (ketone C᎐O),
᎐
᎐
δH(400 MHz; CDCl3), 4.89 (1 H, d, J 4, CHCOCO2Et), 4.20
(1 H, dq, J 14.5 and 7, OCHAHBMe), 4.17 (1 H, dq, J 14.5 and
7, OCHAHBMe), 2.5 (1 H, tt, J 11 and 3.5, cyclohexoyl-CHCO),
2.0–1.1 (21 H, m, cyclohexyl-H) and 1.30 (3 H, t, J 7, OCHA-
HBMe); δC(CDCl3) 209.7, 154.9, 84.3, 64.4, 47.2, 39.0, 29.7,
29.4, 27.6, 26.9, 26.3, 26.0, 25.9, 25.8, 25.7, 25.3 and 14.2; m/z
(EI) 296 (10%, Mϩ), 214 (100, M Ϫ cyclohexene), 241 (M Ϫ
COC6H11), 137 (70), 111 (45, COC6H11), 95 (65) and 83 (C6H11)
(Found: Mϩ, 296.1987. C17H28O4Si requires M, 296.1987).
᎐
᎐
1598 (Ph) and 1583 (Ph); δH(250 MHz; CDCl3) 8.12 (2 H, dd,
J 8 and 1, PhCO2 o-H), 8.00 (2 H, dd, J 8 and 1.5, PhCOCH
o-H), 7.6–7.3 (11 H, m, other Ph) and 7.09 (1 H, s, OCHCO).
1,2-Dicyclohexyl-2-(p-tolylsulfonyloxy)ethanone 16f
Following Julia and Maumy,27 toluene-p-sulfonyl chloride (0.36
g, 1.9 mmol) was added over 1 h to acyloin 7b (0.31 g, 1.5
mmol) in dry pyridine (2 cm3) at 0 ЊC, the flask stoppered and
kept overnight at 4 ЊC, forming needles. The mixture was
poured into cold water (5 cm3) to give a precipitate, the solid
was extracted with ether (4 × 5 cm3), the organic extracts
washed with dilute hydrochloric acid (2 mol dmϪ3, 4 × 5 cm3),
saturated sodium hydrogen carbonate solution (5 cm3), water
(2 × 5 cm3), dried (MgSO4) and evaporated under reduced pres-
sure to give the tosylate (0.48 g, 86%) as needles, mp 102–
103.5 ЊC [from light petroleum (bp 60–80 ЊC)]; Rf (EtOAc–light
1,2-Dicyclohexyl-2-(tert-butyldimethylsilyloxy)ethanone 16c
tert-Butyldimethylsilyl trifluoromethanesulfonate (1.47 cm3, 6.4
mmol) was added to a stirred solution of acyloin 7b (0.96 g, 4.2
mmol) and 2,6-lutidine (1.0 cm3, 8.4 mmol) in dry dichloro-
methane (5 cm3) at 0 ЊC under argon and stirred for 40 min.
Standard work-up with sodium hydrogen carbonate, washing
the combined organic extracts with dilute hydrochloric acid
(3 mol dmϪ3, 2 × 20 cm3) prior to drying, gave a residue which
was heated (Kugelrohr, 120 ЊC at 17 mmHg) to remove a
low-boiling impurity. The residual oil was taken up in hexane,
filtered, and the solution evaporated under reduced pressure to
give the silyl ether as an oil (1.27 g, 88%); Rf (EtOAc–light
petroleum, 10:90) 0.31; νmax(Nujol)/cmϪ1 1714 (C᎐O), 1599
᎐
(Ph), 1369 (OSO2) and 1176 (OSO2); δH(400 MHz; CDCl3) 7.77
(2 H, d, J 8.5, o-H), 7.32 (2 H, d, J 8.5, m-H), 4.65 (1 H, d, J 5.5,
CHOSO2), 2.55 (1 H, tt, J 11 and 3, cyclohexyl-CHCO), 2.43
(3 H, s, tolyl Me) and 1.9–0.9 (21 H, m, cyclohexyl-H);
δC(CDCl3) 208.9, 145.0, 133.5, 129.8, 128.0, 87.5, 46.7, 39.6,
29.1, 27.7, 27.4, 25.9, 25.8, 25.7, 25.6, 25.3, and 21.7; m/z (EI)
378 (15%, Mϩ), 296 (20, M Ϫ cyclohexene), 266 (80, M Ϫ
C6H11CHO), 223 (10, M Ϫ SO2Tol), 139 (30), 123 (40), 111 (80,
C6H11CO), 91 (40, C6H4Me) and 83 (100, C6H11) (Found: Mϩ,
378.1864. C21H30O4S requires M, 378.1864).
petroleum, 10:90) 0.70; νmax(film)/cmϪ1 1721 (C᎐O), 1706
᎐
(C᎐O), 1450 and 1252 (SiMe But); δ (250 MHz; CDCl ) 3.82
᎐
2
H
3
(1 H, d, J 6, CHOSi), 2.75 (1 H, m, cyclohexyl CHCO), 1.8–1.0
(21 H, m, cyclohexyl H), 0.91 (9 H, s, SiCMe3) and Ϫ0.01 (6 H,
s, SiMe2); δC(CDCl3) 215.5, 82.9, 45.4, 41.7, 29.7, 29.4, 28.6,
27.8, 26.3, 26.2, 26.2, 26.1, 26.0, 25.8, 25.6, 18.2, Ϫ4.4 and
Ϫ4.9; m/z (EI) 338 (10%, Mϩ), 323 (55, M Ϫ Me), 281 (65,
M Ϫ But), 227 (100, M Ϫ C5H11CO) and 115 (10, SiMe2But)
(Found: Mϩ, 338.2652. C20H38O2Si requires M, 338.2641).
1,3-Dipheny-2-(phenylthio)propan-1-one 33
Chalcone (2.08 g, 10 mmol) in methanol (10 cm3) was added
with stirring to palladium (5%/C, 0.16 g) suspended in meth-
anol (60 cm3) under hydrogen at 1 atmosphere. After stirring for
24 h, the hydrogen was replaced with argon, the mixture filtered
through Celite and evaporated under reduced pressure to give
1,3-diphenylpropan-1-ol28 (2.02 g, 95%) as an oil; Rf (EtOAc–
light petroleum, 10:90) 0.15; νmax(film)/cmϪ1 3362 (br, OH),
1602 (Ph) and 1585 (Ph; δH(250 MHz; CDCl3) 7.4–7.1 (10 H, m,
Ph), 4.68 (2 H, dd, J 7.5 and 5.5, PhCHOH), 2.8–2.6 (2 H, m,
PhCH2CH2), 2.2–1.9 (2 H, m, PhCH2CH2) and 1.95 (1 H, m,
OH). Jones reagent29 [2.2 cm3 of a solution prepared from
chromium trioxide (26.7 g, 26.7 mmol), concentrated sulfuric
acid (23 cm3) and made up to 100 cm3 with water] was added
dropwise to a stirred solution of the alcohol in acetone (20
cm3). Standard work-up with water gave 1,3-diphenylpropan-1-
one (1.54 g, 89%) as plates, mp 68–69 ЊC (from MeOH–light
petroleum) (lit.,30 70–71 ЊC); Rf (EtOAc–light petroleum,
1,2-Dicyclohexyl-2-oxoethyl benzoate 16d
Pyridine (0.37 cm3, 4.5 mmol) and benzoyl chloride (0.53 cm3,
4.5 mmol) were added to a stirred solution of acyloin 7b (0.85 g,
3.8 mmol) in dry dichloromethane (25 cm3) under argon at 0 ЊC
and stirred for 3 d. Standard work-up with sodium hydrogen
carbonate, washing the combined organic extracts with dilute
hydrochloric acid (3 mol dmϪ3, 3 × 50 cm3) prior to drying, and
chromatography (SiO2, EtOAc–light petroleum, 8:92) gave the
benzoate (0.66 g, 54%) as needles, mp 73.5–76 ЊC (from
MeOH); Rf (EtOAc–light petroleum, 10:90) 0.39; νmax(Nujol)/
cmϪ1 1717 (C᎐O), 1602 (Ph) and 1584 (Ph); δ (400 MHz;
᎐
H
CDCl3) 8.07 (2 H, dd, J 7.5 and 1.5, o-H), 7.57 (1 H, tt, J 7.5
and 1.5, p-H), 7.45 (2 H, t, J 7.5, m-H), 5.27 (1 H, d, J 4,
PhCO2CH), 2.59 (1 H, tt, J 11.5 and 3, cyclohexyl-CHCO), 2.0
(2 H, m, cyclohexyl-H), 1.9–1.55 (10 H, m, cyclohexyl-H) and
1.5–1.1 (9 H, m, cyclohexyl-H); δC(CDCl3) 209.8,
166.1, 133.2, 129.8, 129.7, 128.5, 81.7, 47.5, 39.3, 30.1, 29.3,
27.7, 27.5, 26.3, 26.1, 26.0, 25.9, 25.7 and 25.4; m/z (EI) 328
10:90) 0.37; νmax(Nujol)/cmϪ1 1681 (C᎐O), 1595 (Ph) and 1580
᎐
(Ph); δH(200 MHz; CDCl3) 7.97 (2 H, dd, 7 and 1.5, PhCO
1222
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
o-H), 7.6–7.2 (8 H, m, other Ph), 3.32 (1 H, t, J 7.5, COCHA-
HBCHCHD), 3.30 (1 H, t, J 7.5, COCHAHBCHCHD), 3.08 (1 H,
t, J 7.5, COCHAHBCHCHD) and 3.06 (1 H, t, J 7.5, COCHA-
HBCHCHD). n-Butyllithium (1.6 mol dmϪ3 in hexane, 3.6 cm3)
was added dropwise to a stirred solution of diisopropylamine
(0.80 cm3, 5.7 mmol) in dry THF (20 cm3) under argon at 0 ЊC,
and was stirred for 30 min. The solution was cooled to Ϫ78 ЊC,
1,3-diphenylpropan-1-one (1.09 g, 5.2 mmol) in dry THF
(10 cm3) was added and the solution stirred for a further 90 min
at Ϫ78 ЊC. Diphenyl disulfide (1.35 g, 6.2 mmol) in dry THF
(15 cm3) was added over 15 min at 0 ЊC. The mixture was stirred
for 4 h, warming to room temperature. Standard work-up
with sodium hydrogen carbonate and chromatography (SiO2,
EtOAc–light petroleum, 7:93) gave the ketone 33 (0.71 g, 39%)
as white needles, mp 84–86.5 ЊC (from MeOH) (lit.,31 83–85 ЊC);
Rf (EtOAc–light petroleum, 10:90) 0.32; νmax(Nujol)/cmϪ1 1668
hydrolysed by stirring with dilute hydrochloric acid (3 mol
dmϪ3, 2 cm3) and methanol (30 cm3) for 2 h. Evaporation under
reduced pressure and chromatography (SiO2, EtOAc–hexane,
20:80) gave starting silyl ether 8a (60 mg, 31% recovery) and
ketone 10a (157 mg, 51% from silyl ether 8a). Silyl enol ethers
9aa: Rf (EtOAc–light petroleum, 2:98) 0.33; νmax(film)/cmϪ1
1630 (C᎐C), 1600 (Ph) and 1592 (Ph); δ (250 MHz; CDCl )
᎐
H
3
Z-9aa: 7.7–7.2 (15 H, m, Ph and SiPh), 6.16 (1 H, s, vinyl H)
and 0.34 (6 H, s, SiMe2); E-9aa: 7.7–7.2 (13 H, m, Ph and SiPh),
7.00 (2 H, dd, J 7.5 and 1.5, C᎐CHPh o-H), 6.11 (1 H, s, vinyl
᎐
H) and 0.52 (6 H, s, SiMe2); δC(CDCl3) Z-9aa (partial): 150.90
(C-1), 139.45 (ipso C), 110.82 (C-2) and Ϫ0.95 (SiMe2); E-9aa:
151.4, 137.5, 137.3, 136.7, 133.5, 129.8, 129.1, 128.8, 128.3,
128.1, 128.0, 127.9, 127.7, 125.7 and Ϫ0.94 (SiMe2); m/z (EI)
330 (90%, Mϩ), 135 (100, SiMe2Ph) (Found: Mϩ, 330.1438.
C22H22OSi requires M, 330.1440).
(C᎐O), 1594 (Ph) and 1580 (Ph); δ (270 MHz; CDCl ) 7.82
᎐
H
3
(2 H, dd, J 7 and 1.5, PhCO o-H), 7.50 (1 H, dt, J 1.7 and 7,
PhCO p-H), 7.4–7.1 (12 H, m, other Ph), 4.69 (1 H, dd, J 8.5
and 6, COCHSPh), 3.41 (1 H, dd, J 14 and 8.5, PhCHAHB) and
3.14 (1 H, dd, J 14 and 6, PhCHAHB).
1,2-Diphenyl-1-(trimethylsilyloxy)ethene 9ab
Method A. Following the procedure for the synthesis of 9aa
(Method A), 1,2-diphenylethanone 10a (0.79 g, 4.0 mmol) in
dry THF (10 cm3) was treated with lithium tetramethylpiper-
idide [n-butyllithium (1.6 mol dmϪ3 in hexane, 2.50 cm3),
2,2,6,6-tetramethylpiperidine (0.74 cm3, 4.4 mmol) and dry
THF (17 cm3), 20 min, 0 ЊC, under argon] for 1 h, and silylated
with dry triethylamine (0.85 cm3, 6.1 mmol) and trimethylsilyl
chloride (0.61 cm3, 4.9 mmol) to give starting ketone 10a (133
mg, 17% recovery) and the crude silyl enol ethers 9ab (0.62 g,
58%) (Z-9ab:E-9ab, 87:13).
The silyl enol ethers 9, 35 and 38
1,2-Diphenyl-1-[dimethyl(phenyl)silyloxy]ethene 9aa
Method A. n-Butyllithium (1.6 mol dmϪ3 in hexane, 2.5 cm3)
was added dropwise to a stirred solution of diisopropylamine
(0.56 cm3, 4.0 mmol) in dry THF (17 cm3) under argon at 0 ЊC.
After 20 min, the solution was cooled to Ϫ78 ЊC and 1,2-
diphenylethanone 10a (0.65 g, 3.3 mmol) in dry THF (10 cm3)
was added dropwise. After 1.5 h, dry triethylamine (0.85 cm3,
6.1 mmol) and dimethyl(phenyl)silyl chloride (0.83 cm3, 5.0
mmol) were added, the mixture warmed to room temperature
and stirred for 30 min and the solution was diluted with light
petroleum (100 cm3). Following standard work-up with sodium
hydrogen carbonate, distillation (Kugelrohr, 180 ЊC at 3 mmHg)
failed to purify the product further, giving the silyl enol ethers
9aa (173 mg, 16%) (Z-9aa:E-9aa, 88:12), contaminated with
the starting ketone (65 mg, 10% recovery) and 1,2-diphenyl-
1,1,2,2-tetramethyldisiloxane (76 mg).
Method B. Following the procedure for the synthesis of 9aa
(Method B), 1,2-diphenylethanone 10a (0.42 g, 2.2 mmol) was
treated with hexane-washed sodium hydride (60% suspension in
oil, 0.20 g) in dry THF (10 cm3) for 3 h, and silylated with dry
triethylamine (0.46 cm3, 3.2 mmol) and trimethylsilyl chloride
(0.40 cm3, 3.2 mmol). Distillation of the residue (Kugelrohr,
160 ЊC at 3 mmHg) (lit.,32 132–136 ЊC at 1 mmHg) gave the silyl
enol ether 9ab (0.41 g, 70%) (Z-9ab:E-9ab, >99:1) as a pale
yellow oil, contaminated with starting ketone (15 mg, 4%
recovery). 9ab; νmax(film)/cmϪ1 1632 (C᎐C), 1600 (Ph), 1252
᎐
(SiMe3) and 846 (SiMe3); δH(250 MHz; CDCl3); Z-9ab: 7.69
(2 H, dd, J 1.5 and 7, o-H), 7.63 (2 H, dd, J 1.5 and 7, other
o-H), 7.6–7.1 (6 H, m, other Ph), 6.22 (1 H, s, vinyl-H) and 0.14
(9 H, s, SiMe3); E-9ab: 7.7–7.0 (8 H, m, Ph), 7.10 (2 H, m,
Method B. 1,2-Diphenylethanone 10a (0.42 g, 2.2 mmol) in
dry THF (4 cm3) was added to a stirred suspension of hexane-
washed sodium hydride (60% suspension in oil, 0.20 g) in dry
THF (6 cm3) under argon and refluxed for 3 h, before cooling to
room temperature. Dry triethylamine (0.46 cm3, 3.2 mmol) and
dimethyl(phenyl)silyl chloride (0.55 cm3, 3.2 mmol) were added,
forming a white precipitate. The mixture was stirred for 15 min
and diluted with ether (30 cm3). Standard work-up with sodium
hydrogen carbonate gave crude silyl enol ether 9aa (0.62 g, 85%)
(Z-9aa:E-9aa, >99:1) as a pale yellow oil, contaminated with
1,2-diphenyl-1,1,2,2-tetramethyldisiloxane (0.25 g).
C᎐CHPh o-H), 6.18 (1 H, s, vinyl-H) and 0.30 (9 H, s, SiMe );
᎐
3
δC(CDCl3) Z-9ab (partial): 150.97 (C-1), 139.7 (ipso-C), 110.61
(C-2) and 0.76 (SiMe3); E-9ab (partial): 151.68 (C-1), 111.49
(C-2), and 0.47 (SiMe3). All other aromatic peaks between 140
and 125 ppm were left unassigned.
OSiMe2Ph
OSiMe2Ph
Method C. Dimethyl(phenyl)silyllithium3 (1.28 mol dmϪ3 in
THF, 0.32 cm3) was added dropwise to a stirred solution of silyl
ether 8a (114 mg, 0.40 mmol) in dry THF (5 cm3) at Ϫ78 ЊC,
under argon and the mixture stirred for 20 min at Ϫ78 ЊC and
for 1 h at room temperature. The solution was diluted with
hexane (15 cm3) followed by standard work-up with water.
Chromatography (SiO2, EtOAc–light petroleum, 2:98) of the
residue gave the silyl enol ethers 9aa (44 mg, 30%) (Z-9aa:
E-9aa, 5:95) as a colourless oil.
2
5
6
3
1
6
3
4
1
4
4
2
5
3'
29a
9ba
1,2-Dicyclohexyl-1-[dimethyl(phenyl)silyloxy]ethene 9ba
Method A. Following the procedure for the synthesis of 9aa
(Method A), 1,2-dicyclohexylethanone 10b (0.41 g, 2.0 mmol)
in dry THF (5 cm3) was treated with lithium tetramethylpiperid-
ide [n-butyllithium (1.6 mol dmϪ3 in hexane, 1.38 cm3), 2,2,6,6-
tetramethylpiperidine (0.37 cm3, 2.2 mmol) and dry THF
(8 cm3), 30 min, 0 ЊC under argon] for 3 h at Ϫ78 ЊC, and was
silylated with dry triethylamine (0.42 cm3, 3.0 mmol) and
dimethyl(phenyl)silyl chloride (0.40 cm3, 2.4 mmol). Distillation
(Kugelrohr, 170–200 ЊC at 16 mmHg) gave the silyl enol
ethers 9ba and regioisomer 2-cyclohexyl-1-cyclohexylidene-1-
[dimethyl(phenyl)silyloxy]ethane 29a (67 mg, 10%) (Z-9ba:
E-9ba:29a, 31:6:63) as an oil.
Method D. Dimethyl(phenyl)silyllithium3 (0.85 mol dmϪ3 in
THF, 2.30 cm3) was added dropwise to a stirred solution of silyl
ether 8a (454 mg, 1.6 mmol) in dry THF (2 cm3) at Ϫ78 ЊC,
under argon and the mixture stirred for 15 min. Dimethyl-
(phenyl)silyl chloride (0.42 cm3, 2.5 mmol) was added, the
mixture warmed to room temperature and stirred for 30 min.
Following addition of cold hexane (25 cm3), standard work-up
with ice–water and distillation (Kugelrohr, 180 ЊC at 3 mmHg)
gave the crude silyl enol ethers 9aa (Z-9aa:E-9aa, 22:78) as
a pale yellow oil, contaminated with starting silyl ether 8a
and 1,2-diphenyl-1,1,2,2-tetramethyldisiloxane. Crude 9aa was
Method B. Following the procedure for the synthesis of 9aa
(Method C), silyl ether 8b (238 mg, 0.80 mmol) in dry THF
J. Chem. Soc., Perkin Trans. 1, 1998
1223

View Article Online
(1 cm3) was treated at Ϫ78 ЊC with dimethyl(phenyl)silyl-
lithium3 (1.25 mol dmϪ3 in THF, 0.77 cm3) for 5 h. Standard
work-up with water gave the crude silyl enol ethers 9ba and 29a
(Z-9ba:E-9ba:29a, 73:18:9). Distillation (Kugelrohr, 160 ЊC
at 14 mmHg) gave the silyl enol ethers 9ba and 29a (153 mg,
56%) (Z-9ba:E-9ba:29a, 45:5:60) as an oil; νmax(film)/cmϪ1
1,3-Diphenyl-1-[dimethyl(phenyl)silyloxy]prop-1-ene 35
Dimethyl(phenyl)silyllithium3 (1.2 mol dmϪ3 in THF, 0.35 cm3)
was added dropwise to a stirred solution of ketone 33 (101 mg,
0.32 mmol) in dry THF (5 cm3) at Ϫ78 ЊC, under argon, the
mixture stirred for 20 min, warmed to room temperature and
stirred for 1 h. The mixture was quenched with water (10 cm3),
extracted with hexane (10 cm3), washed with aqueous sodium
hydroxide (5% w/v, 2 × 10 cm3), brine (20 cm3), dried (MgSO4)
and evaporated under reduced pressure. Chromatography
(SiO2, EtOAc–light petroleum, 2:98) gave the silyl enol ether 35
(63 mg, 57%) (Z-35:E-35, >99:1) as an oil; Rf (ether–light
1661 (C᎐C), 1250 (SiMe ), 1118 (SiPh) and 841 (SiMe ); δ (400
᎐
2
2
H
MHz; CDCl3) Z-9ba: 7.6 (2 H, m, m-H), 7.4 (3 H, m, o- and
p-H), 4.27 (1 H, d, J 9.5, H-2), 2.4–0.8 (22 H, m, cyclohexyl CH
and CH2) and 0.45 (6 H, s, SiMe2); E-9ba: 7.6 (2 H, m, m-H), 7.4
(3 H, m, o- and p-H), 4.34 (1 H, d, J 9.5, H-2), 2.4–0.8 (22 H, m,
cyclohexyl-CH and CH2) and 0.41 (6 H, s, SiMe2); 29a 7.6 (2 H,
m, m-H), 7.4 (3 H, m, o- and p-H), 2.18 (2 H, t, J 6, H-3), 2.03
(2 H, t, J 6, H-3Ј), 1.87 (2 H, d, J 7, H-5), 1.9–0.8 (12 H, m,
cyclohexyl-CH2), 1.49 (1 H, m, H-6), 1.45 (4 H, m, H-4)
and 0.43 (6 H, s, SiMe2); δC(CDCl3) Z-9ba (partial): 153.37
(C-1), 138.42 (ipso-C), 133.35 (o-C), 129.53 (p-C), 127.72
(M-C), 112.36 (C-2), 44.32 (C-3 or C-5), 34.43 (C-5 or C-3)
and Ϫ0.80 (SiMe2); E-9ba (partial): 154.34 (C-1), 138.42
(ipso-C), 133.51 (o-C), 129.39 (p-C), 127.62 (m-C), 112.68
(C-2), 39.65 (C-3 or C-5), 35.81 (C-5 or C-3) and Ϫ0.75
(SiMe2); 29a (partial): 140.39 (C-1), 138.47 (ipso-C), 133.41
(o-C), 129.53 (p-C), 127.75 (m-C), 119.37 (C-2), 39.16 (C-5),
35.74 (C-6) and Ϫ0.96 (SiMe2). All other cyclohexyl carbon
signals between 35 and 21 ppm were left unassigned; m/z (EI)
342 (15%, Mϩ), 135 (30, SiMe2Ph), 97 (50, CH2C6H11 from 29a)
and 83 (65, C6H11) (Found: Mϩ, 342.2377. C22H34OSi requires
M, 342.2379).
petroleum, 2:98) 0.31; νmax(film)/cmϪ1 1645 (C᎐C), 1599 (Ph),
᎐
1581 (Ph); δH(400 MHz; CDCl3) Z-35: 7.65 (2 H, dd, J 1.5 and
7.5, Ph o-H), 7.50 (2 H, dd, J 1.5 and 7.5, other Ph o-H), 7.5–7.2
(11 H, m, Ph), 5.39 (1 H, t, J 7, vinyl H), 3.46 (2 H, d, J 7, CH2)
and 0.44 (6 H, SiMe2); δC(CDCl3) 149.69, 141.43, 138.88,
137.25, 133.50, 129.86, 128.43, 128.36, 128.05, 127.85, 127.69,
125.83, 125.80, 110.27, 32.41 and Ϫ0.82; m/z (EI) 344 (5%,
Mϩ), 253 (5, M Ϫ CH2Ph), and 135 (100, SiMe2Ph) (Found:
Mϩ, 344.1597. C23H24OSi requires M, 334.1596).
3,3-Dimethyl-1-phenyl-1-[dimethyl(phenyl)silyloxy]but-1-ene 38
Method A. 3,3-Dimethyl-1-phenylbutan-1-one 15b (1.00 g,
5.7 mmol) in dry THF (3 cm3) was added dropwise to a stirred
suspension of hexane-washed sodium hydride (60% suspension
in oil, 0.50 g, 12.5 mmol) under argon at room temperature and
stirred for 4.5 h. Dry triethylamine (1.2 cm3, 8.6 mmol) and
dimethyl(phenyl)silyl chloride (1.5 cm3, 9.0 mmol) were added,
forming a white precipitate, the mixture stirred for 30 min,
diluted with hexane (30 cm3), washed with saturated sodium
hydrogen carbonate solution (2 × 40 cm3), dried (MgSO4) and
evaporated under reduced pressure. Kugelrohr distillation (bath
temperature 160 ЊC at 15 mmHg) gave the silyl enol ether 38
(0.75 g, 42%) (Z-38:E-38, >99:1) as an oil, contaminated with
disiloxane (0.25 g).
OSiMe3
OSiMe3
2
5
6
3
1
6
3
1
4
4
2
5
3'
4
29b
9bb
Method B. Following the procedure for the synthesis of 9aa
(Method A), silyl ether 13b (0.44 g, 1.67 mmol) in dry THF
(5 cm3) was treated with dimethyl(phenyl)silyllithium3 (1.0 mol
dmϪ3 in THF, 2.0 cm3) for 4.5 h at Ϫ78 ЊC and distilled (Kugel-
rohr, 120 ЊC at 0.5 mmHg) to give the silyl enol ether 38
1,2-Dicyclohexyl-1-(trimethylsilyloxy)ethene 9bb
Following the procedure for the synthesis of 9aa (Method A),
1,2-dicyclohexylethanone 10b (0.41 g, 2.0 mmol) in dry THF
(5 cm3) was treated with lithium tetramethylpiperidide [n-butyl-
lithium (1.6 mol dmϪ3 in hexane, 1.38 cm3), 2,2,6,6-tetra-
methylpiperidine (0.37 cm3, 2.2 mmol) and dry THF (8 cm3), 30
min, 0 ЊC, under argon] for 2.5 h at Ϫ78 ЊC and was silylated
with dry triethylamine (0.42 cm3, 3.0 mmol) and trimethylsilyl
chloride (0.30 cm3, 2.4 mmol). Distillation (Kugelrohr, 170 ЊC
at 14 mmHg) gave the silyl enol ethers 9bb and the regioisomer
2-cyclohexyl-1-cyclohexylidene-1-(trimethylsilyloxy)ethane 29b
(215 mg, 38%) (Z-9bb:E-9bb:29b, 49:21:30) as an oil, con-
(Z-38:E-38, 88:12); νmax(film)/cmϪ1 1645 (C᎐C), 1600 (Ph),
᎐
1253 (SiMe2), 1119 (SiPh) and 832 (SiMe2); δH(250 MHz;
CDCl3) Z-38: 7.45 (2 H, dd, J 7.5 and 1.5, o-H), 7.4–7.1 (8 H,
m, other Ph), 4.80 (1 H, s, vinyl-H), 1.14 (9 H, s, CMe3) and 0.31
(6 H, s, SiMe2); E-38: 7.5–7.0 (10 H, m, Ph), 5.15 (1 H, s, vinyl-
H) and 0.31 (6 H, s, SiMe2); δC(CDCl3) Z-38: 148.63, 140.82,
137.60, 133.49, 129.51, 127.67, 127.57, 127.30, 127.15, 121.47,
31.67, 30.65 and Ϫ0.53; E-38 (partial): 148.04, 137.77, 128.42,
122.96, 31.76 and Ϫ0.92; m/z (EI) 310 (20%, Mϩ), 295 (70,
M Ϫ Me), 135 (70, SiMe2Ph) and 69 (100) (Found: Mϩ,
310.1757. C20H26OSi requires M, 310.1753).
taminated with starting ketone (53 mg, 13% recovery); νmax
-
(film)/cmϪ1 1661 (C᎐C), 1250 (SiMe ) and 841 (SiMe ); δ (500
᎐
3
3
H
MHz; CDCl3) Z-9bb: 4.28 (1 H, d, J 9, H-2), 2.24 (1 H, m, H-3),
1.9–0.8 (12 H, m, 6 × cyclohexyl-CH2), 1.85 (1 H, or 2 H, m,
H-5 or equatorial H-6), 1.73 (2 H, or 1 H, m, equatorial H-6 or
H-5), 1.59 (2 H, m, equatorial H-4), 1.10 (2 H, m, axial H-6),
0.98 (2 H, m, axial H-4) and 0.18 (9 H, s, SiMe3); E-9bb: 4.35 (1
H, d, J 9.5, H-2), 2.28 (1 H, m, H-5), 2.04 (1 H, m, H-3), 1.9–0.8
(16 H, m, 6 × cyclohexyl CH2), 1.62 (2 H, m, equatorial H-4),
1.05 (2 H, m, axial H-4) and 0.16 (9 H, SiMe3); 29b 2.15 (2 H, t,
J 6, H-3), 2.06 (2 H, t, J 6, H-3Ј), 1.97 (2 H, d, J 7, H-5), 1.9–0.8
(12 H, m, cyclohexyl-CH2), 1.49 (1 H, m, H-6), 1.45 (4 H, m,
H-4) and 0.16 (9 H, s, SiMe3); δC(CDCl3) Z-9bb (partial): 153.40
(C-1), 112.30 (C-2), 44.35 (C-3 or C-5), 34.47 (C-5 or C-3) and
0.69 (SiMe3); E-9bb (partial): 154.55 (C-1), 112.40 (C-2), 39.56
(C-3 or C-5), 35.81 (C-5 or C-3) and 0.57 (SiMe3); 29b (partial):
140.37 (C-1), 119.20 (C-2), 39.47 (C-5), 35.81 (C-6) and 0.62
(SiMe3). All cyclohexyl peaks between 139 and 126 ppm
were left unassigned; m/z (EI) 280 (75%, Mϩ), 237 (55), 197
(55, M Ϫ C6H11), 183.1 (35, M Ϫ CH2C6H11), 143 (50) and 73
(100, SiMe3) (Found: Mϩ, 280.2226. C17H32OSi requires M,
280.2222).
Silyl enol ether 38 was hydrolysed by stirring in methanol
(10 cm3) and dilute hydrochloric acid (3 mol dmϪ3, 2 drops) for
48 h, the solvent evaporated under reduced pressure and the
residue chromatographed (SiO2, EtOAc–light petroleum, 5:95)
to give ketone 15b (155 mg, 53% from silyl ether 13b) as an oil.
The reaction between dimethyl(phenyl)silyllithium and carbonyl
compounds 7, 8, 13 and 16
Typically, dimethyl(phenyl)silyllithium3 (1–2 molar equivalents
in THF) was added to the carbonyl compound (0.1–1 mol
dmϪ3) in dry THF under argon at Ϫ78 ЊC. The mixture was
stirred for 1–24 h (the colour of the silyl lithium reagent often
discharging by this time), followed by standard work-up with
ammonium chloride. The residue was stirred overnight in
methanol and hydrochloric acid (3 mol dmϪ3) to hydrolyse the
silyl ethers 8 to their corresponding acyloins 7. The solvent was
re-evaporated under reduced pressure and the residue chrom-
atographed (SiO2, EtOAc–light petroleum, 10:90) to give the
products. The following compounds were treated in this way.
1224
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Reaction with benzoin 7a. Dimethyl(phenyl)silyllithium (0.90
mol dmϪ3 in THF, 1.36 cm3) and benzoin 7a (108 mg, 0.51
mmol) in THF (0.5 cm3) gave after 6 h at Ϫ78 ЊC, 15 h at room
temperature and chromatography (SiO2, CH2Cl2–light petrol-
eum, 50:50), ketone 10a (30 mg, 30%).
Reaction with silyl ether 8a. Dimethyl(phenyl)silyllithium
(1.04 mol dmϪ3 in THF, 0.65 cm3) and silyl ether 8a (0.16 g,
0.57 mmol) in THF (0.5 cm3) gave, after 1.5 h at Ϫ78 ЊC and
chromatography (SiO2, EtOAc–light petroleum, 10:90), ketone
10a (66 mg, 59%).
Reaction with silyl ether 8b. Dimethyl(phenyl)silyllithium
(0.85 mol dmϪ3 in THF, 0.61 cm3) and silyl ether 8b (128 mg,
0.43 mmol) in THF (2 cm3) gave, after 8 h at Ϫ78 ЊC and
chromatography (SiO2, EtOAc–light petroleum, 4:96), ketone
10b (47 mg, 59%).
Reaction with silyl ether 8c. Dimethyl(phenyl)silyllithium
(0.67 mol dmϪ3 in THF, 1.04 cm3) and silyl ether 8c (154 mg,
0.58 mmol) in THF (1.5 cm3) gave, after 8 h in the dark (warm-
ing to room temperature) and chromatography (SiO2, EtOAc–
light petroleum, 20:80), ketone 10c (60 mg, 59%) and 1,2-di-
(2-furyl)ethanedione (14 mg, 12%).
Reaction with silyl ether 8d. Dimethyl(phenyl)silyllithium (0.9
mol dmϪ3 in THF, 2.40 cm3) and silyl ether 8d (174 mg, 0.71
mmol) in dry toluene–THF (1.5 cm3, 50:50) gave, after 1.5 h at
Ϫ78 ЊC, 20 h at room temperature and chromatography (SiO2,
EtOAc–light petroleum, 5:95), a compound {43 mg, tentatively
assigned as 4-[dimethyl(phenyl)silyloxy]-2,2,5,5-tetramethyl-
hexan-3-ol}; Rf (EtOAc–light petroleum, 20:80) 0.38. The
compound decomposed on standing for several days, the
residual oil being a 35:65 mixture of 1,2-diphenyl-1,1,2,2-
tetramethyldisiloxane, identical to an authentic sample, and
diol 40 (25 mg, 21%) (meso:dl, 96.5:3.5).
Reaction with silyl ether 8e. Dimethyl(phenyl)silyllithium
(0.85 mol dmϪ3 in THF, 7.0 cm3) and silyl ether 8e (497 mg, 2.3
mmol) in THF (5 cm3) gave, after 20 h (warming to 10 ЊC),
ketone 10e (43% by gas chromatography).
Reaction with silyl ether 8f. Dimethyl(phenyl)silyllithium
(1.20 mol dmϪ3 in THF, 0.84 cm3) and silyl ether 8f (229 mg,
0.84 mmol) in THF (2 cm3) gave, after 20 h (warming to 10 ЊC)
and chromatography (SiO2, EtOAc–light petroleum, 3:97),
ketone 10f (77 mg, 53%) and 6-[dimethyl(phenyl)silyl]dodecan-
6-ol (6 mg, 6%).
Reaction with silyl ether 13a. Dimethyl(phenyl)silyllithium
(1.14 mol dmϪ3 in THF, 1.8 cm3) and silyl ether 13a (205 mg,
0.69 mmol) in THF (1 cm3) gave, after 1.5 h at Ϫ78 ЊC, 5 h at
0 ЊC and chromatography (SiO2, EtOAc–light petroleum,
5:95), ketone 15a (119 mg, 83%).
Reaction with sodium salt 16a. Dimethyl(phenyl)silyllithium
(0.71 mol dmϪ3 in THF, 4.0 cm3) and crude salt 16a in THF
(10 cm3) gave after 20 h (warming to room temperature) and
chromatography (SiO2 plate, run in 5 × CH2Cl2–light petrol-
eum, 50:50), ketone 10a (69 mg, 15%) and benzil 11 (75 mg,
15%).
Reaction with benzoate ester 16b. Dimethyl(phenyl)silyl-
lithium (1.08 mol dmϪ3 in THF, 1.64 cm3) and ester 16b (187
mg, 0.59 mmol) in THF (1 cm3) gave, after 2 h at Ϫ78 ЊC, 6 h at
0 ЊC and chromatography (SiO2, EtOAc–light petroleum,
15:85), ketone 10a (106 mg, 92%).
Reaction with silyl ether 16c. Dimethyl(phenyl)silyllithium
(0.90 mol dmϪ3 in THF, 0.81 cm3) and silyl ether 16c (207 mg,
0.61 mmol) in THF (1 cm3) gave, after 4 h at Ϫ78 ЊC and
chromatography (SiO2, EtOAc–light petroleum, 4:96), ketone
10b (89 mg, 70%).
0.61 mmol) in THF (1 cm3) gave, after 1.5 h at Ϫ78 ЊC, 2 h at
0 ЊC and chromatography (SiO2, EtOAc–light petroleum,
5:95), ketone 10b (88 mg, 70%).
Reaction with tosylate 16f. Dimethyl(phenyl)silyllithium (0.6
mol dmϪ3 in THF, 0.25 cm3) and tosylate 16f (55 mg, 0.145
mmol) in THF (0.25 cm3) gave, after 12 h (warming to room
temperature) and chromatography (SiO2, EtOAc–light petrol-
eum, 5:95), ketone 10b (14 mg, 46%).
The following compounds were prepared by these methods.
1,2-Diphenylethanone 10a. Isolated as a solid, mp 52–54 ЊC
(lit.,33 56–57 ЊC); Rf (CH2Cl2–light petroleum, 50:50) 0.31;
νmax(Nujol)/cmϪ1 1685 (C᎐O); δ (250 MHz; CDCl ) 8.0 (2 H,
᎐
H
3
dd, J 7 and 1.5, PhCO o-H), 7.7–7.2 (8 H, m, other Ph) and 4.28
(2 H, s, CH2).
1,2-Dicyclohexylethanone 10b. Isolated as an oil;34 Rf
(EtOAc–light petroleum, 4:96) 0.30; νmax(film)/cmϪ1 1707
(C᎐O); δ (250 MHz; CDCl ) 2.27 1 H, cyclohexyl COCH),
᎐
H
3
2.27 (2 H, d, J 7, COCH2), 1.9–1.5 (11 H, m, cyclohexyl H) and
1.0–0.8 (2 H, m, cyclohexyl H); δC(CDCl3) 213.7, 51.1, 48.3,
33.6, 33.3, 28.3, 26.2, 26.0, 25.8, 25.6.
1,2-Di-(2-furyl)ethanone 10c. Isolated as an oil (lit.,35 mp 20–
23 ЊC); Rf (EtOAc–light petroleum, 20:80) 0.25; νmax(film)/cmϪ1
3131 (vinyl-C᎐H), 1677 (C᎐O) 1592 (C᎐C) and 1569 (C᎐C);
᎐
᎐
᎐
δH(250 MHz; CDCl3) 7.60 (1 H, m, furoyl-H-5), 7.33 (1 H, m,
furyl-H-5), 7.25 (1 H, d, J 3.5, furoyl-H-3), 6.54 (1 H, dd, J 3.5
and 1.5, furoyl-H-4), 6.33 (1 H, dd, J 3.5 and 1.5, furyl-H-4),
6.25 (1 H, d, J 3.5, furyl-H-3) and 4.14 (2 H, s, CH2).
2,5-Dimethylhexan-3-one 10e.36 Isolated as an oil; Rf (EtOAc–
light petroleum, 10:90) 0.49; νmax(film)/cmϪ1 1709 (C᎐O);
᎐
δH(250 MHz; CDCl3) 2.55 (1 H, septet, J 7.5, COCHMe2), 2.31
(2 H, d, J 7.5, COCH2CHMe2), 2.15 (1 H, nonet, J 7.5, COCH2-
CHMe2), 1.06 (6 H, d, J 7.5, COCHMe2) and 0.89 (6 H, d, J 7.5,
COCH2CHMe2).
Dodecan-6-one 10f.22 Isolated as an oil; Rf (EtOAc–light
petroleum, 2:98) 0.23; νmax(film)/cmϪ1 1715 (C᎐O); δ (250
᎐
H
MHz; CDCl3) 2.36 (4 H, t, J 7.5, 2 × COCH2), 1.57 (4 H,
2 × COCH2CH2), 1.4–1.2 (10 H, m, CH2), 0.90 (3 H, t, J 7.5,
Me) and 0.88 (3 H, t, J 7.5, Me).
Benzil 11. Isolated as yellow needles, mp 88–94 ЊC (lit.,37 92–
94 ЊC); Rf (CH2Cl2–light petroleum, 50:50) 0.57; νmax(Nujol)/
cmϪ1 1659 (C᎐O), 1593 (Ph) and 1578 (Ph); δ (250 MHz;
᎐
H
CDCl3) 7.95 (4 H, dd, J 8 and 1, o-H), 7.65 (2 H, tt, J 8 and 1,
p-H) and 7.53 (4 H, t, J 8, m-H).
1,2-Diphenylpropan-1-one 15a. Isolated as needles, mp 46–
48 ЊC (from hexane) (lit.,38 53–54 ЊC); Rf (ether–light petroleum,
12:88) 0.33; νmax(film)/cmϪ1 1682 (C᎐O), 1597 (Ph) and 1582
᎐
(Ph); δH(250 MHz; CDCl3) 7.97 (2 H, dd, J 7.5 and 1, PhCO
o-H), 7.5–7.1 (8 H, m, other Ph), 4.70 (1 H, q, J 7.5, CHMe)
and 1.55 (3 H, d, J 7.5, CHMe).
2,2,5,5-Tetramethylhexane-3,4-diol 40.39 Isolated as an oil; Rf
(EtOAc–light petroleum, 20:80) 0.08; νmax(film)/cmϪ1 3411 (br,
OH); δH(400 MHz; CDCl3) meso: 3.24 (2 H, s, CHOH), 1.40
(2 H, br s, CHOH) and 1.00 (18 H, s, CMe3); ( ) (partial): 3.31
(2 H, s CHOH) and 0.90 (18 H, s, CMe3); δC(CDCl3) meso: 80.4,
35.6 and 26.5; ( ) (partial): 74.9 and 25.8; m/z (EI) 156 (0.5%,
M Ϫ H2O), 117 (5, M Ϫ C4H9), 99 (15, M Ϫ H2O Ϫ C4H9), 87
(25, C4H9CHOH), 57 (100, C4H9) and 41 (50, C3H5).
1,2-Di(2-furyl)ethanedione. Isolated as yellow needles, mp
159–163 ЊC (from EtOAc–hexane) (lit.,40 164–165 ЊC); Rf
(EtOAc–light petroleum, 20:80) 0.18; νmax(Nujol)/cmϪ1 3148
(furyl-C᎐H) and 1643 (C᎐O); δ (250 MHz; CDCl ) 7.77 (2 H, s,
᎐
H
3
J 1.5, H-5), 7.64 (2 H, d, J 3.5, H-3) and 6.63 (2 H, dd,
J 3.5 and 1.5, H-4); δC(CDCl3) 176.9, 149.4, 149.3, 124.7 and
113.1.
6-[Dimethyl(phenyl)silyl]dodecan-6-ol. Isolated as an oil; Rf
(EtOAc–light petroleum, 4:96) 0.13; νmax(film)/cmϪ1 3462 (br,
OH), 1248 (SiMe2), 1112 (Si᎐Ph) and 831 (SiMe2); δH(400 MHz;
CDCl3) 7.6 (2 H, m, m-H), 7.3 (3 H, m, o- and p-H), 1.6 (4 H, m,
2 × CH2CHOH), 1.4–1.2 (15 H, m, 7 × CH2 and OH), 0.88
(3 H, t, J 6.5, CH2Me), 0.87 (3 H, t, J 6.5, CH2Me) and 0.37
Reaction with benzoate ester 16d. Dimethyl(phenyl)silyl-
lithium (1.4 mol dmϪ3 in THF, 1.16 cm3) and benzoate ester 16d
(178 mg, 0.54 mmol) in dry THF (1 cm3) gave, after 2 h at
Ϫ78 ЊC, 3 h at 0 ЊC and chromatography (SiO2, EtOAc–light
petroleum, 5:95), ketone 10b (56 mg, 50%).
Reaction with carbonate 16e. Dimethyl(phenyl)silyllithium
(1.3 mol dmϪ3 in THF, 1.40 cm3) and carbonate 16e (181 mg,
J. Chem. Soc., Perkin Trans. 1, 1998
1225

View Article Online
(6 H, s, SiMe2); δC(CDCl3) 137.3, 134.6, 129.1, 127.7, 69.1, 37.3,
37.2, 32.6, 31.8, 30.1, 23.2, 22.9, 22.6, 22.6, 14.1, 14.0 and Ϫ4.3;
m/z (EI) 320 (23%, Mϩ), 319 (55, M Ϫ H), 305 (65, M Ϫ Me),
249 (70, M Ϫ C5H11), 235 (70, M Ϫ C6H13) and 185 (65, M Ϫ
SiMe2Ph) (Found: Mϩ, 320.2535. C20H36OSi requires M,
320.2535).
enol ether 47a (Method A), trifluoromethyl ketone 46 (0.25 cm3,
1.78 mmol) was treated with dry zinc bromide (1.2 g, 5.3 mmol)
and tert-butyldiphenylsilyllithium (0.51 mol dmϪ3 in THF, 3.53
cm3) in dry THF–ether (1:2, 24 cm3) for 20 min at Ϫ78 ЊC and
90 min at 0 ЊC. Standard work-up and chromatography (SiO2,
EtOAc–light petroleum, 1:99) gave the silyl enol ether 47b (346
mg, 49%) as an oil.
1,2-Diphenylethane-1,2-diol 39
Method B. Following the procedure for the synthesis of silyl
enol ether 47a (Method B), trifluoromethyl ketone 46 (0.25
mmol, 1.78 mmol) in dry ether (15 cm3) was treated with the
reagent formed from methylmagnesium bromide (3.0 mol dmϪ3
in ether, 0.6 cm3) and tert-butyldiphenylsilyllithium (0.28 mol
dmϪ3 in THF, 6.4 cm3) for 20 min at Ϫ78 ЊC and for 50 min at
0 ЊC. Standard work-up and chromatography (SiO2, EtOAc–
light petroleum, 1:99) gave the silyl enol ether 47b (357 mg,
51%) as an oil; Rf (ether–light petroleum, 2:98) 0.38; νmax(film)/
Silyl ether 8a (208 mg, 0.73 mmol) in dry ether (10 cm3) was
stirred with a suspension of lithium aluminium hydride (28 mg,
0.74 mmol) in dry ether (40 cm3) under argon at Ϫ78 ЊC for 6 h.
The reaction was quenched at Ϫ78 ЊC with methanol (10 cm3)
and warmed to room temperature. Standard work-up with
potassium sodium tartrate and chromatography (SiO2, EtOAc–
light petroleum, 20:80) gave the diol (114 mg, 73%) [meso:( )
94:6], as needles, mp 128–133 ЊC (lit., meso41 128–130 ЊC; ( )42
cmϪ1 3072 (C᎐H), 3050 (C᎐H), 1732 (C᎐C), 1590 (Ph) and 1114
148–149 ЊC); Rf (EtOAc–light petroleum, 20:80) 0.12; νmax
-
᎐
(Nujol)/cmϪ1 3366 (br, OH) and 3314 (br, OH); δH(250 MHz;
CDCl3) meso: 7.4–7.1 (10 H, m, Ph), 4.82 (2 H, s, CHOH) and
2.23 (2 H, br s, CHOH); ( ): 7.4–7.1 (10 H, m, Ph), 4.70 (2 H, s,
CHOH) and 2.86 (2 H, br s, CHOH).
(SiPh); δH(250 MHz; CDCl3) 7.8–7.7 (6 H, m, 6 × o-H), 7.5–7.2
(9 H, m, other H) and 1.17 (9 H, s, CMe3); δC(CDCl3) 154.0 (dd,
3
1JCF 284 and 289, CF2), 137.6, 135.6, 135.6, 132.5 (dd, JCF
6
and 4, PhC᎐CF ipso-C), 130.0, 128.1, 127.6, 127.6, 127.0 (dd,
᎐
2
2JCF 5.5 and 3, C᎐CF ), 26.8 and 19.8; m/z (EI) 394 (5%, Mϩ),
᎐
2
Standard work-up procedure for fluoro compounds 47–53
353 (40, M Ϫ C3H5), 337 (95, M Ϫ But), 303 (95, M Ϫ PhCH2),
259 (60), 231 (35) and 201 (100) (Found: Mϩ, 394.1566.
C24H24F2OSi requires M, 394.1564).
Unless stated otherwise, the reaction mixture was diluted with
hexane (20 cm3), quenched with saturated potassium sodium
tartrate solution (20 cm3), the aqueous layer extracted with
hexane–ether (1:1, 3 × 20 cm3), the combined organic extracts
washed with saturated potassium sodium tartrate solution
(30 cm3), dried (MgSO4) and evaporated under reduced pressure.
1,1-Difluoro-3-phenyl-2-[dimethyl(phenyl)silyloxy]prop-1-ene 49
1,1,1-Trifluoro-3-phenylpropan-2-one 48 (0.10 cm3, 0.65 mmol)
was added to a stirred solution of dimethyl(phenyl)silyllithium3
(0.13 mol dmϪ3 in THF, 10.8 cm3) under argon at Ϫ78 ЊC and
stirred for 15 min. Chlorodimethyl(phenyl)silane (0.14 cm3,
0.85 mmol) was added and the mixture warmed to room
temperature over 20 min. Standard work-up and rapid chrom-
atography through a short column (SiO2, EtOAc–light petrol-
eum, 1.5:98.5) gave the crude silyl enol ether 49 contaminated
with 1,1,2,2-tetramethyl-1,2-diphenyldisilane, identical to an
authentic sample. Distillation (Kugelrohr, 180 ЊC at 28 mmHg)
failed to purify the product further. Yield of 49 (from 1H NMR
spectrum) 37%; Rf (ether–light petroleum, 2:98) 0.46;
1,1-Difluoro-2-phenyl-2-[dimethyl(phenyl)silyloxy]ethene 47a
Method A. Zinc bromide (4.5 g, 20 mmol) was dried under
vacuum (110 ЊC, 0.1 mmHg overnight, then 200 ЊC, 4 h),
allowed to cool and was suspended in dry THF–ether (75 cm3,
1:4) under argon. (Trifluoroacetyl)benzene 46 (0.25 cm3, 1.78
mmol) was added, the solution cooled to Ϫ78 ЊC and stirred for
15 min. Dimethyl(phenyl)silyllithium3 (1.1 mol dmϪ3 in THF,
6.6 cm3) was added dropwise over 10 min, the mixture stirred
for 20 min, warmed to 0 ЊC and stirred a further 1 h. Water
(40 cm3) was added followed by standard work-up. Distillation
(Kugelrohr, 110 ЊC increasing to 150 ЊC at 1.5 mmHg) gave the
crude silyl enol ether as an oil, contaminated with 1,2-diphenyl-
1,1,2,2-tetramethyldisilane (140 mg), identical to an authentic
νmax(CDCl3)/cmϪ1 1763 (C᎐C), 1602 (Ph), 1592 (Ph) and 1255
᎐
(SiMe2); δH(250 MHz; CDCl3) 7.6–7.2 (8 H, m, Ph), 7.16 (2 H,
4
dd, J 7.5 and 2, PhCH2 o-H), 3.32 (2 H, dd, JHF 4 and 2,
PhCH2) and 0.33 (6 H, s, SiMe2); δC(CDCl3) 153.6 (dd, 1JCF 282
1
sample. Yield of 47a by estimation of the H NMR spectrum;
4
and 276, CF2), 137.0 (t, JCF 2, PhCH2 ipso-C), 136.5, 133.2,
19%.
2
Method B. Methylmagnesium bromide (3.0 mol dmϪ3 in
ether, 0.6 cm3) was added to dimethyl(phenyl)silyllithium3 (0.51
mol dmϪ3 in THF, 3.5 cm3) at 0 ЊC under argon, and was stirred
for 20 min. The silyl(methyl)magnesium reagent was then added
dropwise over 10 min to (trifluoroacetyl)benzene 46 (0.25
mmol, 1.78 mmol) in dry ether (12 cm3) at Ϫ78 ЊC. After stir-
ring for 30 min, the solution was warmed to 0 ЊC and stirred a
further 90 min. Standard work-up and rapid chromatography
through a short column (SiO2, EtOAc–light petroleum, 1:99)
gave the crude silyl enol ether, as an oil contaminated with (di-
129.8, 128.9, 128.3, 127.7, 126.6, 112.9 (dd, JCF 42 and 15,
3
C᎐CF ), 35.0 (t, J 1, PhCH2), Ϫ1.6 and Ϫ1.7; m/z (EI) 304
᎐
2
CF
(20%, Mϩ), 135 (100, SiMe2Ph), 91 (35, PhCH2), 87 (45) and 74
(80) (Found: Mϩ, 304.1078. C17H18F2OSi requires M,
304.1095).
1,1-Difluoro-3-phenylpropan-2-one 5044
Dimethyl(phenyl)silyllithium3 (1.28 mol dmϪ3 in THF, 2.60
cm3) was added dropwise over 5 min to a stirred solution of
trifluoromethyl ketone 48 (0.25 cm3, 1.64 mmol) in dry ether
(40 cm3) at Ϫ100 ЊC, under argon. The mixture was stirred for
30 min, warming to Ϫ40 ЊC. Excess methyl iodide (0.8 cm3) was
added (ineffectually) and the mixture was allowed to warm to
room temperature. A solution of 2,4-dinitrophenylhydrazine
(0.26 g, 1.3 mmol) and concentrated sulfuric acid (0.33 cm3) in
methanol (10 cm3) was added, the mixture heated to boiling,
allowed to cool and water (30 cm3) added. The precipitate was
filtered off, the mother liquor extracted with ether (3 × 40 cm3)
and the precipitate and the organic extracts combined and
evaporated under reduced pressure. Chromatography (SiO2,
EtOAc–lightpetroleum,10:90)gave1,1-difluoro-3-phenylpentan-
2-one (2,4-dinitrophenyl)hydrazone (330 mg, 59%) as an orange
solid, mp 107–108 ЊC (from MeOH); Rf (ether–light petroleum,
1
fluoroacetyl)benzene.43 Yields by estimation of the H NMR
spectrum; 47a (15%) and (difluoroacetyl)benzene (15%). 47a; Rf
(ether–light petroleum, 2:98) 0.44; νmax(film)/cmϪ1 1722 (C᎐C),
᎐
1598 (Ph), 1256 (SiMe2), 1119 (SiPh) and 832 (SiMe2); δH(250
MHz; CDCl3) 7.8–7.3 (10 H, m, Ph) and 0.33 (6 H, s, SiMe2);
δC(CDCl3) (partial) 154.9 (t, 1JCF 175, CF2), 139.9, 136.4, 133.6,
133.2, 126.3 (dd, 2JCF 27.5 and 16, C᎐CF ) and 1.0; m/z (ϩFAB)
᎐
2
290 (55%, Mϩ), 271 (55, M Ϫ F), 209 (100) and 193 (100)
(Found: Mϩ, 290.0935. C16H16F2OSi requires M, 290.0938).
(Difluoroacetyl)benzene; Rf (hexane) 0.05; νmax(film)/cmϪ1 1696
(C᎐O) and 1598 (Ph); δ (250 MHz; CDCl ) 8.15 (2 H, dd, J 7.5
᎐
H
3
and 1, o-H), 7.8–7.3 (3 H, other Ph) and 6.32 (1 H, t, 2JHF 53.5,
CHF2).
20:80) 0.40; νmax(Nujol)/cmϪ1 3290 (N᎐H), 1632 (C᎐N), 1597
᎐
1,1-Difluoro-2-phenyl-2-(tert-butyldiphenylsilyloxy)ethene 47b
Method A. Following the procedure for the synthesis of silyl
(Ph), 1549 (NO2), 1339 (NO2) and 1314 (NO2); δH(250 MHz;
CDCl3) 11.02 (1 H, br s, NH), 9.04 (1 H, d, J 2.5, NHAr H-3),
1226
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
8.34 (1 H, dd, J 9.5 and 2.5, NHAr H-5), 7.97 (1 H, d, J 9.5,
NHAr H-6), 7.4–7.3 (5 H, other Ph), 6.35 (1 H, t, JHF 55,
tion (15 cm3) was added, the aqueous layer extracted with ether
2
(4 × 15 cm3), the combined organic extracts washed with brine
(30 cm3), dried (MgSO4) and evaporated under reduced
pressure. Chromatography (SiO2, EtOAc–light petroleum,
12:88) gave the aldol 53 (72 mg, 31%); Rf (EtOAc–light petrol-
CHF2) and 3.98 (2 H, s, PhCH2); δC(CDCl3) 171.1, 147.7, 144.1,
139.6, 132.0, 130.0, 129.4, 128.6, 127.9, 122.9, 116.9, 114.5
(t, 1JCF 239, CHF2) and 60.4; m/z (EI) 350 (95%, Mϩ), 315 (80),
269 (30), 143 (55), 117 (PhCH2CN) and 91 (100 PhCH2)
(Found: Mϩ, 350.0825. C15H12F2N4O4 requires M, 350.0826).
1,1-Difluoro-3-phenylpentan-2-one (2,4-dinitrophenyl)hydra-
zone (0.41 g, 1.17 mmol), titanium trichloride (1.8 mol dmϪ3 in
water, 10 cm3) and dry 1,2-dimethoxyethane (60 cm3) were
refluxed under argon for 15 min, until no starting material was
visible by TLC. The mixture was cooled to room temperature
and water (50 cm3) was added. The aqueous layer was extracted
with ether (3 × 60 cm3), the combined organic extracts washed
with brine (100 cm3) and dried (MgSO4). The solvents were
distilled using a Vigreux column (oil-bath temperature 110 ЊC).
Chromatography of the residue (SiO2, light petroleum then
EtOAc–light petroleum, 20:80) gave the ketone 50 (90 mg,
45%) as an oil; Rf (ether–light petroleum, 15:85) 0.29;
νmax(CDCl3)/cmϪ1 1749 (C᎐O) and 1603 (Ph); δ (250 MHz;
eum, 20:80) 0.30; νmax(film)/cmϪ1 3364 (OH), 1702 (C᎐O), 1597
᎐
(Ph), and 1579 (Ph); δH(250 MHz; CDCl3) 8.05 (2 H, dd, J 8.5
and 1, PhCO o-H), 7.7–7.3 (8 H, m, other Ph), 5.39 (1 H,
ddd, 3JHH 4.5, 2JHF 18.5 and 4.5, CHOH) and 3.04 (1 H, d, J 4,
OH).
Acknowledgements
We thank the EPSRC and Zeneca Argrochemicals for a CASE
award (R. S. R.).
References
1 E. J. Corey, M. A. Tius and J. Das, J. Am. Chem. Soc., 1980, 102,
1742.
2 I. Fleming and U. Ghosh, J. Chem. Soc., Perkin Trans. 1, 1994, 257.
3 I. Fleming, R. S. Roberts and S. C. Smith, J. Chem. Soc., Perkin
Trans. 1, 1998, 1209.
4 H. J. Reich, R. C. Holtan and C. Bolm, J. Am. Chem. Soc., 1990,
112, 5609.
5 D. Wittenberg, T. C. Wu and H. Gilman, J. Org. Chem., 1959, 24,
1349; E. Vedejs, M. J. Arnost, J. M. Eustache and G. A. Krafft,
J. Org. Chem., 1982, 47, 4384; G. A. Krafft and P. T. Meinke, J. Am.
Chem. Soc., 1988, 110, 8671; M. Koreeda and S. Koo, Tetrahedron
Lett., 1990, 31, 831; A. G. M. Barrett, J. M. Hill and E. W. Wallace,
J. Org. Chem., 1992, 57, 386.
6 G. H. Dodd, B. T. Golding and P. V. Ioannou, J. Chem. Soc., Perkin
Trans. 1, 1976, 2273.
7 D. J. Ager, Org. React. (N.Y.), 1990, 38, 1.
8 P. F. Hudrlik, R. H. Schwartz and A. K. Kulkarni, Tetrahedron
Lett., 1979, 2233; P. F. Hudrlik, G. Nagendrappa, A. K. Kulkarni
and A. M. Hudrlik, Tetrahedron Lett., 1979, 2237.
9 M. S. Biernbaum and H. S. Mosher, J. Am. Chem. Soc., 1971, 93,
6221; A. G. Brook and J. D. Pascoe, J. Am. Chem. Soc., 1971, 93,
6224.
10 P. F. Hudrlik, A. M. Hudrlik and A. K. Kulkarni, J. Am. Chem.
Soc., 1982, 104, 6809; S. R. Wilson, M. S. Hague and R. N. Misra,
J. Org. Chem., 1982, 47, 747; R. J. Linderman and A. Ghannam,
J. Am. Chem. Soc., 1990, 112, 2392.
11 M. J. Tozer and T. F. Herpin, Tetrahedron, 1996, 52, 8619; J. T.
Welch, Tetrahedron, 1987, 43, 3123; Selective Fluorination in Organic
and Biological Chemistry, ed. J. T. Welch, ACS Symposium series
456, ACS, Washington DC, 1991; M. H. Gelb, J. P. Svaren and R. H.
Abeles, Biochemistry, 1985, 24, 1813; S. Thaisrivongs, D. T. Pals,
W. M. Kati, S. R. Turner, L. M. Thamasco and W. Watt, J. Med.
Chem., 1986, 29, 2080; M. H. Gelb, J. Am. Chem. Soc., 1986, 108,
3146.
12 M. Yamana, T. Ishihara and T. Ando, Tetrahedron Lett., 1983, 24,
507; T. Ishihara, T. Yamanaka and T. Ando, Chem. Lett., 1984,
1165; M. Kuroboshi and T. Ishihara, Tetrahedron Lett., 1987, 28,
6481; M. Kuroboshi and T. Ishihara, Bull. Chem. Soc. Jpn., 1990, 63,
428; K. Iseki, Y. Kuroki, D. Asada and Y. Kobayashi, Tetrahedron
Lett., 1997, 38, 1447.
13 T. Brigaud, P. Doussot and C. Portella, J. Chem. Soc., Chem.
Commun., 1994, 2117; C. Portella and B. Dondy, Tetrahedron Lett.,
1991, 32, 83; B. Dondy, P. Doussot and C. Portella, Synthesis, 1992,
995; B. Dondy and C. Portella, J. Org. Chem., 1993, 58, 6671;
P. Doussot and C. Portella, J. Org. Chem., 1993, 58, 6675; B. Dondy,
P. Doussot and C. Portella, Tetrahedron Lett., 1994, 35, 409.
14 F. Jin, B. Jiang and Y. Xu, Tetrahedron Lett., 1992, 33, 1221; F. Jin,
Y. Xu and W. Huang, J. Chem. Soc., Perkin Trans. 1, 1993, 795.
15 H. Schubert, J. Prakt. Chem., 1956, 3, 146.
᎐
H
CDCl3) 7.4–7.1 (5 H, m, Ph), 5.78 (1 H, t, 2JHF 54, CHF2) and
3.98 (2 H, s, PhCH2); δC(CDCl3) 196.8 (t, 2JCF 26.5, COCHF2),
131.3, 129.7, 128.9, 127.6, 109.8 (t, 1JCF 253, CHF2) and 43.0.
1,1,1-Trifluoro-2-methyl-3-phenylpropan-2-ol 5145
Following the procedure for the synthesis of silyl enol ether 47a
(Method B), trifluoromethyl ketone 48 (0.25 mmol, 1.64 mmol)
in dry ether (15 cm3) was treated with the reagent formed from
methylmagnesium bromide (3.0 mol dmϪ3 in ether, 0.55 cm3)
and tert-butyldiphenylsilyllithium (0.26 mol dmϪ3 in THF, 6.3
cm3) for 20 min at Ϫ78 ЊC and 50 min at 0 ЊC. Standard work-
up and chromatography (SiO2, EtOAc–light petroleum, 1:99)
gave alcohol 51 (0.20 g, 61%); Rf (EtOAc–light petroleum,
20:80) 0.33; νmax(film)/cmϪ1 3464 (br OH) and 1606 (Ph);
δH(250 MHz; CDCl3) 7.4–7.2 (5 H, m, Ph), 3.10 (1 H, d, J 14,
PhCHAHB), 2.84 (1 H, d, J 14, PhCHAHB), 1.98 (1 H, s, OH)
and 1.28 (3 H, s, Me); δC(CDCl3) 134.4, 130.8, 128.4, 127.2, 73.7
(q, J 28, CF3), 40.8 and 20.5; m/z (EI) 204 (25%, Mϩ) and 91
(100, PhCH2) (Found: Mϩ, 204.0766. C10H11F3O requires M,
204.0762).
Chromatography also gave tert-butyl(diphenyl)silane 52 (0.30
g, 76%) as an oil; Rf (ether–light petroleum, 2:98) 0.55;
νmax(CDCl3)/cmϪ1 2113 (Si᎐H), 1588 (Ph) and 1111 (SiPh);
δH(250 MHz; CDCl3) 7.87 (4 H, dd, J 5.5 and 2, o-H), 7.54 (6 H,
m, m- and p-H), 4.88 (1 H, s, SiH) and 1.28 (9 H, s, CMe3);
δC(CDCl3) 135.9, 134.2, 129.6, 128.0, 27.8 and 18.0.
2,2-Difluoro-3-hydroxy-1,3-diphenylpropan-1-one 5346
Method A. Silyl enol ether 47a was prepared following the
procedure detailed in Method B, and was worked up prior to
chromatography. The crude silyl enol ether in dry dichloro-
methane (20 cm3) was added to a stirred solution of benz-
aldehyde (0.87 cm3, 8.6 mmol) in dry dichloromethane at
Ϫ78 ЊC. Titanium tetrachloride (1.0 mol dmϪ3 in dichloro-
methane, 10.7 cm3) was added, the solution stirred for 15 min,
warmed to Ϫ15 ЊC and stirred overnight, warming to room
temperature and forming a brown precipitate. Water (100 cm3)
and saturated potassium sodium tartrate solution (100 cm3)
were added, the aqueous layer extracted with EtOAc (3 × 100
cm3), the combined organic extracts washed with brine (150
cm3), dried (MgSO4) and evaporated under reduced pressure.
Chromatography (SiO2, EtOAc–light petroleum, 12:88) gave
the aldol 53 (0.715 g, 38%), inseparable from dimethyl-
(phenyl)silanol (0.30 g), identical to an authentic sample.
Method B. Titanium tetrachloride (1.32 cm3 of a 1.0 mol
dmϪ3 solution in CH2Cl2, 1.32 mmol) was added to a solution
of pure silyl enol ether 47b (346 mg, 0.88 mmol) and benz-
aldehyde (0.11 cm3, 1.06 mmol) in dichloromethane (10 cm3) at
Ϫ78 ЊC, and the mixture was stirred overnight, warming to
room temperature. Saturated potassium sodium tartrate solu-
16 K. Rühlmann, Synthesis, 1971, 236.
17 B. B. Corson, W. L. Benson and T. T. Goodwin, J. Am. Chem. Soc.,
1930, 52, 3988.
18 J. Cossy and P. Pale, Tetrahedron Lett., 1987, 28, 6039.
19 I. Dyong and H. P. Bertram, Chem. Ber., 1973, 106, 1743.
20 H. Teichmann and V. Prey, Justus Liebigs Ann. Chem., 1970, 732,
121.
21 G. J. D. Peddle, J. Organomet. Chem., 1967, 27, 2311.
22 G. A. Molander and G. Hahn, J. Org. Chem., 1986, 51, 1135.
23 W. Lidy and W. Sundermeyer, Chem. Ber., 1973, 106, 587.
24 R. Roger, J. Chem. Soc., 1925, 127, 518.
J. Chem. Soc., Perkin Trans. 1, 1998
1227

View Article Online
25 S. Hünig and G. Wehner, Chem. Ber., 1979, 112, 2062.
26 E. P. Kohler and J. L. E. Erickson, J. Am. Chem. Soc., 1931, 53,
2301.
27 M. Julia and M. Maumy, Org. Synth., 1988, Coll. Vol. 6, 586.
28 R. J. Bushby and G. J. Ferber, J. Chem. Soc., Perkin Trans. 2, 1976,
1683.
29 A. J. Bowers, T. G. Halsall, E. R. H. Jones and A. J. Lemin, J. Chem.
Soc., 1953, 2548.
30 M. M. Shemyakin, L. A. Neiman, S. V. Zhokova, Y. S. Nekrasov,
T. J. Pehk and E. T. Lippmaa, Tetrahedron, 1971, 27, 2811.
31 A. Jonczyk, M. Ludwikow and M. Makosza, Rocz. Chem., 1977, 51,
175.
32 G. Simchen and W. Kober, Synthesis, 1976, 259.
33 Handbook of Chemistry and Physics, The Chemical Rubber
Publishing Co., Cleveland, Ohio, 52 edn., 1972.
34 D. S. Matteson and R. J. Moody, Organometallics, 1982, 1, 20.
35 D. S. MacNair, Justus Liebigs Ann. Chem., 1880, 258, 220.
36 M. G. Mayberry and J. G. Aston, J. Am. Chem. Soc., 1934, 56, 2682.
37 T.-L. Ho, Synthesis, 1972, 697.
38 V. Meyer and L. Oelkers, Ber., 1888, 21, 1295.
39 J. C. Lee Jr., C. G. Boojamra and R. H. Crabtree, J. Org. Chem.,
1993, 58, 3895.
40 B. B. Corson and R. W. McAllister, J. Am. Chem. Soc., 1929, 51,
2822.
41 G. Berti and F. Bottari, J. Org. Chem., 1960, 25, 286.
42 F. Eisenlohr and L. Hill, Ber., 1937, 70, 942.
43 S. G. Cohen, H. T. Wolosinski and P. J. Scheuer, J. Am. Chem. Soc.,
1949, 71, 3439.
44 C. Wakselman and J. Leroy, J. Chem. Soc., Chem. Commun., 1976,
611.
45 S. Sibille, E. d’Incan, L. Leport and J. Perichon, Tetrahedron Lett.,
1986, 27, 3129.
46 M. Kuroboshi and T. Ishihara, Bull. Chem. Soc. Jpn., 1990, 63, 428.
Paper 7/09114A
Received 22nd December 1997
Accepted 5th February 1998
1228
J. Chem. Soc., Perkin Trans. 1, 1998