New mechanistic aspects on the catalytic transformation of vinylthiiranes to mono and disubstituted 3,6-dihydro-1,2-dithiins by tungsten pentacarbonyl monoacetonitrile

Article abstract of DOI:10.1016/S0040-4020(02)00398-8

Various alkyl and aryl mono- and disubstituted 3,6-dihydro-1,2-dithiins have been synthesised from their corresponding vinylthiiranes exploiting Adams' tungsten pentacarbonyl monoacetonitrile catalytic transformation. New conclusions pertaining to the rate determining step, the sensitivity of the process to precursor sterics and electronics, and the nature of various reaction intermediates are highlighted.

Full text of DOI:10.1016/S0040-4020(02)00398-8

TETRAHEDRON
Pergamon
Tetrahedron 58 <2002) 4517±4527
New mechanistic aspects on the catalytic transformation of
vinylthiiranes to mono and disubstituted 3,6-dihydro-1,2-dithiins
by tungsten pentacarbonyl monoacetonitrile
David W. Lupton and Dennis K. Taylorp
Department of Chemistry, University of Adelaide, Adelaide 5005, Australia
Received 29 January 2002; revised 4 March 2002; accepted 4 April 2002
AbstractÐVarious alkyl and aryl mono- and disubstituted 3,6-dihydro-1,2-dithiins have been synthesised from their corresponding
vinylthiiranes exploiting Adams' tungsten pentacarbonyl monoacetonitrile catalytic transformation. New conclusions pertaining to the
rate determining step, the sensitivity of the process to precursor sterics and electronics, and the nature of various reaction intermediates
are highlighted. q 2002 Published by Elsevier Science Ltd.
1. Introduction
compounds. After some preliminary work our synthetic
efforts became concentrated on the use of vinyl thiiranes
as precursors to 1,2-dithins as pioneered by Adams, due to
the high reported yields and mild reaction conditions.
In connection with our ongoing studies on the ring-opening
of 3,6-disubstituted 1,2-dioxines 1 by stabilised phosphor-
ous ylides with the ®nal construction of highly substituted
cyclopropanes 2, we were stimulated to prepare an analo-
gous range of 3,6-dihydro-1,2-dithiins of type 3.1 While
there is signi®cant interest in 1,2-dithiins of type 3 within
the scienti®c community their preparation is not yet routine,
with much of that interest residing in chemistry parallel to
their synthesis. In work aimed at exploring the chemistry of
singlet sulfur, the synthesis of 1,2-dithiins by Diels±Alder
chemistry has been used as a model to monitor the forma-
tion of this reactive elemental form of sulfur.2 More
recently, Adams and his colleagues investigated novel
methods to desulfurise petroleum feedstock which led to
the synthesis of 1,2-dithiins along with other sulfur con-
taining heterocycles such as thiiranes and thietanes.3
Finally, the bioactivity of 1,2-dithiins has been investigated
with various natural products isolated. For instance, unsatu-
rated 1,2-dithiins as typi®ed by thiarubin A and B display
signi®cant antiviral,4 antibacterial5 and antifungal5a,6
activities. Furthermore, dihydro-1,2-dithiins have been
isolated from various foods such as the garlic bush,7 onions8
and cooked asparagus.9 Our interest in 1,2-dithiins lay in
their possible use as synthetic building blocks and as such
we were interested in an ef®cient route to these compounds.
Although the work discussed above is not centred around
synthesis, the use of singlet sulfur and also the desulfuri-
sation of thiiranes provided possible routes to these
The catalytic cycle proposed by Adams3 for the conversion
of vinylthiiranes 4 into 1,2-dithiins 3 initially involves the
formation of the vinylthiirane W<CO)5 complex 5 through
the displacement of the acetonitrile ligand by 4, step a,
Scheme 1. It has been tentatively proposed that formation
of zwitterionic intermediate 6, step b, occurs via the inter-
mediacy of 7, and not as a direct result of a bimolecular
nucleophilic addition mechanism in which the sulfur atom
of the second thiirane adds to the b-carbon of complex 5.
This proposition was based on the observation that methyl
substitution on the vinyl group of 4 led to acceleration of
reaction rate through stabilisation of the resultant carbo-
cation 7. Intramolecular cyclisation of 6 with concomitant
loss of the 1,3-butadiene 8, step c, affords intermediates of
type 9 which have been characterised crystallographically.3c
Final displacement of the 1,2-dithiin from 9 by vinyl
Keywords: vinylthiiranes; 1,2-dithiins; tungsten pentacarbonyl monoaceto-
nitrile.
p
Corresponding author. Tel.: 161-8-8303-5494; fax: 161-8-8303-4358;
e-mail: dennis.taylor@adelaide.edu.au
0040±4020/02/$ - see front matter q 2002 Published by Elsevier Science Ltd.
PII: S0040-4020<02)00398-8

4518
D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
Scheme 1.
thiirane 4, step d, completes the catalytic cycle, thus,
reforming intermediate 5. While Adams has shown that
the reaction is successful for a range of H and Me-substi-
tuted vinylthiiranes 4, the ability of the transformation to
`carry' a diverse range of substituents <H, alkyl and aryl)
both in the 3- and 6-positions of the forming 1,2-dithiin is
yet to be demonstrated. Furthermore, by utilising aryl
substituted vinylthiiranes for the synthesis of 3- and
6-substituted 1,2-dithiins, aspects of the reaction mecha-
nism pertaining to the formation and stability of inter-
mediates of type 5 and 7 can be investigated. This report,
therefore, highlights a range of new mechanistic ®ndings in
regard to the tungsten catalysed synthesis of mono and
disubstituted 3,6-dihydro-1,2-dithiins from their respective
precursor vinylthiiranes and also extends the overall
synthetic scope of Adams' new catalytic transformation.
the transformation of each individual isomer into the corre-
sponding 1,2-dithiin should be able to be conveniently
monitored by 1H NMR.
The most direct route to the requisite vinylthiiranes <10a±h)
was from the corresponding vinyloxiranes10 <11a±h), which
were in turn prepared by the action of various sulfur ylides
on the appropriate aldehyde, as summarised in Table 1.
Thus, treatment of aldehydes <12a±e) with the appropriate
2. Results
2.1. Construction of the requisite vinylthiiranes (10a±h)
Table 1. Preparation of vinyloxiranes <11a±h)
Entry
Aldehyde Sulfur ylide
Product Yield <%)a cis/transb
In order to accumulate signi®cant mechanistic information
whilst limiting the study to a practical size we decided to
investigate Adams' catalytic 1,2-dithiin transformation
utilising the eight distinctly different vinylthiiranes
<10a±h) depicted below. In particular, the incorporation of
various aryl substituents on the vinylthiiranes was deemed
highly desirable as this should allow for conclusions
pertaining to the rate determining step, the processes sensi-
tivity to precursor sterics and electronics, and the nature of
various reaction intermediates to be made. Furthermore, the
preparation of various vinylthiiranes as mixtures of their cis
and trans isomers was considered fortuitous as this should
effectively double the mechanistic information attainable as
1
2
3
4
5
6
7
8
12a
12b
12c
12d
12a
12e
12e
12e
13a
13a
13a
13a
13b
13d
13c
13a
11a
11b
11c
11d
11e
11f
71
91
78
81
81
75c
80c
58
45:55
25:75
41:59
45:55
±
47:53
±
29:71
11g
11h
See Section 5 for general procedures.
Isolated yield except where noted.
a
b
cis/trans ratios were determined from the coupling constants between the
protons about the oxirane moiety; cis,4.5 Hz, trans,3.0 Hz.
Unpuri®ed yield; purity estimated to be ,95% by H NMR.
c
1

D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
Table 2. Preparation of vinylthiiranes <13a±h)
4519
Entry
Vinyloxirane
Vinylthiirane <%)a
Recovered oxirane <%)a
Diene <%)a
1
2
3
4
5
6
7
8
11a
11b
11c
11d
11e
11f
10a <43)
10b <40)
10c <44)
10d <±)
10e <39)
10f <65)
10g <67)
10h <46)
20
15
20
±
15
±
11
15
12
73
22
5
11g
11h
±
±
±
23
See Section 5 for general procedures.
Isolated yield except where noted.
a
sulfonium salt <13a±d) under basic conditions afforded
vinyloxiranes <11a±g) in excellent yields with cis/trans
ratios consistent with the literature.11 Vinyloxiranes <11f,
g) were utilised crude as attempted puri®cation led to facile
decomposition, while in the case of vinyloxirane 11h only
moderate yields were attained due to decomposition in the
neat form.
diene. Vinylthiirane 10d was too unstable to be isolated
from the reaction mixture, undergoing exclusive sulfur
extrusion, Table 2, entry 4. Indeed, electron withdrawing
substituents have been found to facilitate the loss of sulfur
from analogous vinylthiiranes.12 Due to the instability of all
vinylthiiranes it was considered desirable to quench the
reaction prior to completion, thus maximising the yield of
both the vinylthiirane and recovered starting material.
Furthermore, in order to avoid sulfur extrusion following
puri®cation, all vinylthiiranes were stored as chloroform
solutions.
2.2. Treatment of vinylthiiranes with tungsten
pentacarbonyl monoacetonitrile
With the majority of the requisite vinylthiiranes success-
fully prepared, it was now time to verify whether tungsten
pentacarbonyl monoacetonitrile would catalyse conversion
to the appropriate 1,2-dithiin. The data collated within Table
3 indicates that all vinylthiiranes undergo smooth conver-
sion into the 1,2-dithiins with excellent yields attainable.
Vinylthiiranes 10a and 10g <entries 1, 6) as well as 10e
and 10f <entries 4, 5) afford the same 1,2-dithiins, 14a and
14d, respectively, in approximately the same yields indi-
cating ¯exibility in the position of the thiirane moiety
with respect to the vinyl grouping. In addition, vinylthiirane
10h affords two 1,2-dithiins <14e and 14f) due to the pre-
cursor vinylthiirane containing two inequivalent vinyl
groupings, entry 7.
The preparation of the requisite vinylthiiranes <10a±h) was
readily achieved by exposure of the vinyloxiranes to an
aqueous inorganic thiocyanate solution Table 2.10b±d
Isolated yields were moderate due to the extrusion of sulfur,
thus, forming appreciable quantities of the corresponding
Table 3. Preparation of 1,2-dithiins <14a±g)
Having successfully synthesised a series of new 1,2-dithiins,
attempts were now made to elucidate the reaction's
sensitivity to precursor sterics and electronics through an
investigation of the reaction's kinetics. Rate data was
Entry
Vinylthiirane
1,2-Dithiin
Yield <%)a Reaction time <h)
1
2
3
4
5
6
7
10a
10b
10c
10e
10f
14a
14b
14c
14d
14d
14a
14e, 14f
87
80
82
79
75
85
24
20
26
30
38
34
30
1
obtained from H NMR experiments in which 1.4 mol%
of catalyst transformed 0.573 mmol of substrate using
CDCl3 <0.7 ml) as solvent, except where noted. All experi-
ments were performed at 218C with the consumption of
thiirane calculated by comparison to hexamethylbenzene
as an internal standard. Table 4 contains a summary of the
rates determined for each thiirane. In the cases of vinyl-
thiiranes 10a and 10c, the rate of consumption of the
cis and trans isomers were conveniently monitored by
10g
10h
40, 43
See Section 5 for a general procedure.
a
Isolated yield taking into account the stoichiometry of the reaction, i.e.
2 equiv. of vinylthiirane are necessary to afford 1 equiv. of 1,2-dithiin and
the corresponding butadiene.

4520
D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
Table 4. Rate constants determined for various vinylthiiranes
observing the difference in peak area of the doublet at d
4.2 ppm, corresponding to the cis vinylthiirane, to the multi-
plet at d 3.7 ppm, corresponding to the combined cis and
trans vinylthiirane concentration. Similarly, the consump-
tion of cis and trans 10f could be followed by observing the
area of the doublet of doublets at d 3.74 ppm, cis isomer
only, and the pentet at d 2.91 ppm, trans isomer only. In the
case of vinylthiiranes 10b and 10e, determining the rates for
the individual isomers was not possible due to overlaying
precursor/product resonances, consequently the rates were
determined as an average of the two isomers.
b
Entry
Vinylthiirane
Solvent
Isomera
105 rate <s21
)
1
10a
CDCl3
CDCl3
CDCl3
CDCl3
C6D6
cis
trans
cis
trans
cis
trans
cis
trans
mix
mix
cis
trans
mix
cis
trans
±
7.71
11.00
4.91
5.92
5.66
6.79
1.86
2.31
0.00
20.70
5.81
10.20
3.65
0.75
0.43
2.26
2
3c
4c
5c
6c
C6D6
7c
OC<CH3)2
OC<CH3)2
CD3CN
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
8c
9c
10
11
12
13
14d
15d
16d
10b
10c
A plot of the variation in 2-phenyl-3-vinylthiirane 10a
concentration against time is shown in Fig. 1 and is typical
for the systems investigated. Data manipulation13 indicated
that the reaction was only ®rst order over the ®rst 10% of the
reaction course, Fig. 2. Therefore, all rate constants except
where noted within Table 4, were calculated before 10%
completion. The occurrence of ®rst order kinetics only in
the initial stages of the reaction can be contributed to
catalyst inhibition by the product.13 In order to test this
hypothesis, the reaction was repeated using a stoichiometric
quantity of catalyst. A plot of the variation in 2-phenyl-3-
vinylthiirane 10a concentration against time was again
generated and is presented in Fig. 3, while Fig. 4 represents
a plot of ln <A/0A) against time. The reaction now proceeded
10e
10f
10g
See Section 5 for a general procedure.
a
Isomers were assigned as discussed. Rates were not determined for the
individual isomers of vinylthiiranes <10b and 10e) and were determined
for the complete mixture.
b
Rate constants were calculated using the modulus of the slope of
ln <A/0A) plotted against time and are for reactions performed with
1.4 mol% catalyst at 218C and were determined only for the ®rst 10%
of the overall reaction.
Reactions performed with dilution factor of two compared to other
entries.
Rate constant obtained over the entire course of the reaction.
c
d
Figure 1. Consumption of cis <D) and trans <W) 2-phenyl-3-vinylthiirane 10a plotted against time with 1.4 mol% W<CO)5<MeCN) in CDCl3 at 218C.
Figure 2. Plot of ln <A/0A) against time for the ®rst 10% of the reaction of 10a with 1.4 mol% W<CO)5<MeCN) in CDCl3 at 218C with an equation of the
trendline, for the two isomers, displayed within the key.

D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
4521
Figure 3. Consumption of 2-phenyl-3-vinylthiirane 10a plotted against time in seconds with a stoichiometric quantity of W<CO)5<MeCN) performed in CDCl3
at 218C temperature.
with ®rst order kinetics until 85% completion after which,
presumably, the 1,2-dithiin concentrations is suf®cient to
inhibit the catalyst.
the side of the acetonitrile bound catalyst and the reaction
is completely inhibited. To further characterise the rate
determining step, the reaction rates associated with various
p-substituted aryl thiiranes were then considered, entries 1,
2 and 10±12, Table 4. Comparison of the sp value of the
substituent to the log <k/kH) indicates rate enhancement with
electron donating groups and retardation with electron with-
drawing groups. This trend is consistent with a rate deter-
mining step of associative coordination by the vinylthiirane
in which donation of a lone pair of electrons from sulfur is
promoted by electron donating aryl substituents. Although
the magnitude of the slope of the plot could be used to
derive conclusions as to the nature of the transition state,
this was considered inappropriate at this stage due to the
limitation of the data set.
Thiiranes 10f and 10g did not display catalytic inhibition
and were found to be ®rst order over the entire course of the
reaction, the rates being depicted within Table 4, entries
14±16. Finally, rates for the conversion of 10a were also
determined in a range of solvents under identical reaction
conditions in order to probe for solvent effects, entries 3±9.
3. Discussion
3.1. Solvent and substituent effects
The effect of solvent polarity on the conversion of 2-phenyl-
3-vinylthiirane 10a into 1,2-dithiin 14a was investigated
using various deuterated solvents, Table 4 entries 3±9. No
signi®cant effect on the reaction rate was observed with
change in solvent. This result concurs with the previously
proposed rate determining step of co-ordination.3 Further-
more, this result eliminates the possibility of simultaneous
co-ordination and ring-opening, Scheme 1, step a0. A second
observation from these experiments was the lack of reaction
when employing acetonitrile as solvent. This result suggests
an equilibrium between the acetonitrile bound catalyst and
the vinylthiirane bound catalyst. Thus, when employing
acetonitrile as the solvent, the equilibrium exists solely on
3.2. Involvement of zwitterionic species
In the work of Adams the formation of a zwitterionic inter-
mediate of type 7 prior to nucleophilic addition of a second
vinylthiirane was proposed.3c,d This pathway is preferred
due to kinetic observations pertaining to different methyl
substituted thiiranes and also differences in reaction rate
of alternative tungsten catalysts.3c In order to probe the
proposed zwitterionic mechanism we believed that the
synthesis of a divinylthiirane in which the alternative zwit-
terions would have signi®cantly different stability, thus
determining the obtained product, was warranted. To this
end, vinylthiirane 10h was synthesised and its conversion to
Figure 4. Plot of the ln <A/0A) against time for the ®rst 85% of the reaction of 10a with a stoichiometric quantity of W<CO)5<MeCN) in CDCl3 at 218C with an
equation of the trendline within the key.

4522
D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
Scheme 2.
1,2-dithiins 14e and 14f investigated, Scheme 2. If zwitter-
ionic species were involved in the reaction, as proposed
by Adams, then benzylic stabilisation would favour inter-
mediate 15a over 15b, thus, leading to 1,2-dithiin 14e as the
major product, following nucleophilic `attack' and cyclisa-
tion. Alternatively, if neutral intermediates were involved
then equivalence between 1,2-dithiins 14e and 14f would be
expected with, perhaps, slight predominance of 14f due to
steric considerations.
quickly than the trans, Table 4, entries 14 and 15. These
differences in rate can be attributed to differences in steric
crowding around the catalyst during formation of inter-
mediate 5. Since the sulfur atom of the thiirane has two
lone pairs of electrons through which it can coordinate,
two isomeric complexes of intermediate 5 are possible,
Scheme 3. These alternative isomers were observed in the
work of Adams by using 13C enriched NMR studies.3c
Although coordination by the cis thiiranes gives the steri-
cally more favourable intermediate 5a, it also affords the
unfavourable intermediate 5b. The trans isomer on the other
hand gives the two intermediates 5c and 5d, presumably of
moderate stability. The faster reaction rate for the trans
isomers of 10a and 10c implies that the increased proba-
bility of coordination, two orientations <5c and 5d) com-
pared to one <5a), outweighs the greater stability of 5a. In
the case of thiirane 10f the greater stability of the equivalent
isomer 5a must outweigh the effect of an increased number
of orientations through which coordination can occur.
Upon synthesising vinylthiirane 10h it was transformed into
1,2-dithiins 14e and 14f under the usual conditions. The two
isomers formed were distinguished by the resonance due to
the hydrogen in the 3-position, and from their relative areas,
the ratio of the 1,2-dithiins could be determined. The more
prevalent 1,2-dithiin was 14f, Table 3, entry 7, thus, indi-
cating that the reaction was not preceeding via a charged
intermediate and was slightly sensitive to steric effects. This
result is in stark contrast to the aforementioned observations
of Adams.3c
The effect of steric bulk on reaction rate can also be
observed by comparing the various phenyl and methyl
substituted thiiranes utilised. In changing the hydrogen
substitution b to the thiirane moiety into a methyl group,
Table 4, entries 1, 2 and 13±16, respectively, a decrease in
reaction rate is observed. This result is contrary to the obser-
vations of Adams and Perrin who noted rate enhancement
with addition of a methyl group to either the position a, or
b, to the thiirane. Their rate enhancement was attributed to
stabilisation of charged intermediate 7 by the methyl group
on the double bond. Ignoring the fact that these arguments
pertained to the non rate determining ring-opening, it
would seem likely, in any respect, that the stability of the
allylic cation derived from methyl substitution a, or b, or on
the thiirane would be similar. Similarly when thiiranes 10f
and 10g, Table 4, entries 14±16, are compared then it is
noted that methyl substitution, this time on the thiirane
ring, results in retardation of the reaction rate. This latter
result concurs with the work of Adams and Perrin who
observed that stabilisation of the thiirane ring by methyl
substituents leads to a slower reaction rate.3c However,
their explanation once again invokes chemistry not
associated with the rate determining step but rather the
subsequent ring-opening.
3.3. Steric effects on the rate determining step
Entries 1, 2, 11 and 12 of Table 4 demonstrate that the trans
isomers of 10a and 10c reacted more rapidly than the cis
isomers, while the cis isomer of thiirane 10f reacted more
Scheme 3.

D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
4523
Scheme 4.
We believe that the increase in steric bulk upon substitution
of a methyl group directly onto the thiirane ring, or onto the
vinylic double bond, results in a decrease in the reaction rate
due to unfavourable steric interactions, Scheme 4. Further-
more, the degree of change appears to correlate with the
proximity of the steric bulk to the thiirane ring, namely,
the rate decrease is much larger when methyl substitution
occurs on the ring, 4.5 fold, compared to the more distal
vinylic position, 2.5 fold. Apparently, contradicting this
conclusion is the observation of faster reaction rates for
thiiranes 10a and 10e, compared to thiiranes 10f and 10g,
the later of which are terminal thiiranes which, presumably,
you might expect to give rise to less steric crowding.
However, this result appears to pertain to the stability of
the formed complex rather than the kinetics of association,
and is discussed below.
Alternately, if Michael attack by 10e is more favourable
than 10g then the complete conversion to 1,2-dithiin 14a
and diene 8a, Path A, would be followed by the conversion
of thiirane 10g to 1,2-dithiin 14a and diene 8a, Path D, and,
thus, equal quantities of 1,2-dithiins 14a and 14d would be
observed. Similarly, the unexpected paths would yield 14a
initially followed by 14d, Path C, or 14a and 8a as the only
products, Path B. These scenarios are depicted within
Scheme 5, and represent the extremes that may be expected.
The experiment was set up as described, and monitored by
1H NMR as for the kinetic studies, although in this case over
19 h. The results of the experiment are shown in Fig. 5. As
can be seen the more `slower' reacting thiirane 10g gives
rise to the major product, 14a, while the formation of the
minor 1,2-dithiin proceeds after a lag time to give 14d
derived from the more `rapidly' reacting thiirane 10e.
Although this result appears counter intuitive it can,
however, be rationalised by considering previously
discussed results. Firstly, the thiirane 10e can be considered
to be in equilibrium between the catalyst bound and
unbound forms, as demonstrated by the lack of reaction in
acetonitrile. Secondly, thiirane 10g reacts after a lag time
indicating that the thiirane bound catalyst has a compara-
tively long lifetime. Coupled to this we observe no catalytic
inhibition with thiirane 10g, thus, indicating that the
1,2-dithiin does not compete for the catalyst, therefore the
thiirane binds with a higher af®nity than the 1,2-dithiin.
Since catalytic inhibition is observed with thiirane 10e it
is therefore possible to conclude that thiirane 10g forms a
thermodynamically more stable complex with the catalyst
than does thiirane 10e. This is presumably due to less steric
bulk around the thiirane ring. Hence, although the thiirane
10e binds more rapidly to the catalyst, it is in equilibrium
with the unbound form, thus allowing the catalyst to form
3.4. Catalytic inhibition
As discussed previously the reaction rates of thiiranes 10f
and 10g, Table 4, entries 14±16, are lower than expected
based on the grounds of sterics alone. Furthermore, these
substrates are the only substrates to display ®rst order
kinetics over the entire reaction. In order to clarify the
implications of these results, and to further characterise
the nature of the rate determining step, a reaction in which
a mixture of thiiranes 10e and 10g was transformed into
1,2-dithiins 14d and 14a, respectively, was undertaken. It
was expected that the faster reacting thiirane 10e would bind
to the catalyst more quickly and subsequently be `attacked'
by either thiirane 10e or 10g, Path A or B, Scheme 5, in the
Michael fashion to eventually furnish the 1,2-dithiin 14d as
the major product along with both diene 8a and 8b.
More favourable attack by 10g, Path B, would result in
the formation of 1,2-dithiin 14d and diene 8b exclusively.
Scheme 5.

4524
D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
Figure 5. Plot of the conversion of vinylthiirane 10 h to 1,2-dithiins 14a and 14d by W<CO)5<NCMe) in CDCl3 at 218C against time.
the more stable association with thiirane 10g. As a result the
catalyst is sequestered away from the rapidly reacting thi-
irane, resulting in the formation of 1,2-dithiin 14a as the
major product due to Michael attack by both thiiranes,
Scheme 5 Paths C and D. These results are summarised in
Scheme 6.
last observation could possibly be exploited in the future in
the reaction of thiirane mixtures in which one thiirane
provides the `backbone' and the other the second sulfur
atom, thus allowing for a more atomically ef®cient synthetic
tool.
5. Experimental
4. Conclusions
Solvents were dried by appropriate methods wherever
needed. All organic extracts were dried over anhydrous
magnesium sulfate. Thin-layer chromatography <TLC)
used aluminum sheets silica gel 60 F254 <40£80 mm2)
from Merck. Melting points were taken on a Reichert
Thermovar Ko¯er apparatus and are uncorrected. Infrared
spectra were recorded on a ATI Mattson Genesis Series
FTIR spectrophotometer as nujol mulls unless otherwise
We have successfully applied the methodology of Adams to
the synthesis of various alkyl and aryl mono and 3,6-disub-
stituted 1,2-dithiins with consistently good yields, thus,
further demonstrating the robust nature of this methodology.
Furthermore, through our synthetic endeavours we have
been able to present ®ndings that concur with Adams
proposal of the rate determining step being the associative
coordination of the thiirane moiety to tungsten, whilst
noting a high degree of sensitivity to thiirane sterics as
indicated by the different kinetics of the various phenyl or
methyl substituted derivatives. In regard to the Michael
attack on a neutral or zwitterionic intermediate we are
able to report that the reaction appears to proceed via a
neutral species which is then ring-opened by a second
vinyl thiirane. Finally, we have observed subtle differences
in the thiiranes af®nity for the catalyst and there ability to
sequester and thus retard the reaction of other thiiranes. This
1
stated. H and 13C NMR spectra were recorded in CDCl3
solution on a Varian INOVA <600 MHz) or on a Varian
Gemini 2000 instrument, TMS <0 ppm) and CDCl3
<77.0 ppm) as internal standards unless otherwise speci®ed.
All yields reported refer to isolated material judged to
be homogeneous by TLC and NMR spectroscopy unless
otherwise speci®ed. The following materials were
purchased from Aldrich and used without further puri®-
cation; Aldehydes <12a±e), methyl sul®de, ethyl sul®de,
dimethyl sulfate, diethyl sulfate, allyl bromide.
Scheme 6.

D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
4525
5.1. Materials
prepared using this methodology. All physical and spectro-
scopic data were consistent with the literature. 10b10e and
10c10e where prepared from oxirane 11b and 11c, respec-
tively. 11d was treated as above, however, the only iso-
latable product was 4-[<E/Z)-1,3-butadienyl]benzonitrile14
as a result of desulfurisation.
5.1.1. Preparation of 2-phenyl-3-vinyloxiranes (11a±e).
A typical procedure.11b A heterogeneous mixture of allyl
bromide <4.84 g, 0.04 mol), methyl sul®de <2.79 g,
0.045 mol) and water <3.75 ml) was stirred vigorously for
20 h at ambient temperature. Excess methyl sul®de was
removed under vacuum from the resulting brown solution
and the appropriate benzaldehyde <2.65 g, 0.025 mol) in
isopropanol <7.5 ml) added. To this solution, sodium
hydroxide <1.6 g, 0.04 mol) in water <3.25 ml), was added
dropwise over 30 min. The reaction was monitored by TLC
and after 5 h was complete. The produced methyl sul®de
was removed in vaccuo and the solution extracted with ether
<3£30 ml) and dried <MgSO4) to yield the crude product.
Puri®cation by ¯ash chromatography <5% ethyl acetate/
hexanes) yielded a cis and trans mixture of the title
compound, 11a. Refer to Table 1 for yields and cis/trans
ratios of all oxiranes prepared using this methodology.
Physical and spectroscopic data was consistent with the
literature for all compounds prepared using this method-
ology. 11b,11b 11c11b and 11d11b were prepared from
5.1.4. 2-Phenyl-3-[(E/Z)-1-propenyl]thiirane (10e). The
title compound was prepared as a colourless oil from
oxirane 11e using the approach for thiirane 10a. Rf 0.35
<5% dichloromethane/hexane); IR 1684, 1602, 1493 cm21
;
<cis E-10e); 1H NMR <CDCl3, 600 MHz) d 1.60 <dd, Jˆ1.8,
6.6 Hz, 3H), 3.80 <dd, Jˆ7.2, 9.6 Hz, 1H), 4.21 <d, Jˆ
7.2 Hz, 1H), 5.15±4.84 <m, 1H), 5.94±5.88 <m, 1H),
7.39±7.21 <m, 5H); <trans E-10e); 1H NMR <CDCl3,
600 MHz) d 1.73 <dd, Jˆ1.8, 6.6 Hz, 3H), 3.54 <dd, Jˆ
5.4, 9.0 Hz, 1H), 3.71 <d, Jˆ5.4 Hz, 1H), 5.26 <ddq,
Jˆ1.8, 9.0, 15.6 Hz, 1H), 5.94±5.88 <m, 1H), 7.39±7.21
1
<m, 5H); <cis Z-10e); H NMR <CDCl3, 600 MHz) d 1.82
<m, 3H), 4.02 <ddd, Jˆ1.2, 7.2, 10.2 Hz, 1H), 4.26 <d,
Jˆ7.2 Hz, 1H), 5.16 <tq, Jˆ2.4, 10.2 Hz, 1H), 5.62 <ddq,
Jˆ1.2, 6.6, 10.2 Hz, 1H), 7.39±7.21 <m, 5H); <trans Z-10e);
1H NMR <CDCl3, 600 MHz) d 1.82 <m, 3H), 3.73 <d,
Jˆ5.4 Hz, 1H), 3.77 <ddd, Jˆ0.6, 5.4, 9.6 Hz, 1H), 5.15±
4.84 <m, 1H), 5.71 <ddq, Jˆ0.6, 7.2, 10.2, 1H), 7.39±7.21
<m, 5H); 13C NMR <CDCl3) d 13.29, 13.36, 17.82, 18.01,
38.15, 40.88, 42.50, 42.99, 43.32, 44.13, 45.91, 52.59,
126.37, 126.88, 126.93, 127.34, 127.39, 127.53, 127.97,
128.47, 128.59, 129.16, 129.23, 129.73, 130.34, 131.10,
131.18, 131.95, 135.48, 135.59, 138.62. The titled mixture
has unstable and decomposed over several days. Conse-
quently 10e was used immediately.
p-methoxybenzaldehyde,
p-chlorobenzaldehyde
and
p-cyanobenzaldehyde, respectively. 11e11c was prepared
from benzaldehyde and the salt of crotyl bromide and
methyl sul®de using the approach for 11a. 11h11d was
prepared from cinnamaldehyde using the approach for 11a.
5.1.2. Preparation of 2-methyl-3-[(E)-2-phenyl-1-ethenyl]-
oxiranes (11f, g and i). A typical procedure.11a Triethyl-
sulfonium <sulfonatooxy)ethane <6.08 g, 25 mmol) was
added to a stirred solution of cinnamaldehyde <2.64 g,
20 mmol) in dichloromethane <20 ml) followed by the
addition of 50% aqueous NaOH solution <10 ml) to give a
dark orange heterogeneous mixture. After 2.5 h the solution
was diluted with H2O <50 ml), the organic phase separated,
and the aqueous phase extracted with ether <2£50 ml). The
combined organic fractions were washed with water
<2£50 ml), dried <CaCl2), and the volatiles removed in
vacuo to afford the title compound 11f in suf®cient purity
for the next step. The title compound can be puri®ed by
chromatography <10% hexanes/dichloromethane). Refer to
Table 1 for yields and cis/trans ratios of all oxiranes
prepared using this methodology. All physical and spectro-
scopic data was consistent with the literature. 11g11a was
prepared from trimethylsulfonium <sulfonatooxy)methane
using the approach for 11f.
5.1.5. 2-Methyl-3-[(E)-2-phenyl-1-ethenyl]thiirane (10f).
The title compound was prepared as a colourless oil from
oxirane 11f using the approach for thiirane 10a. Rf 0.30
<10% dichloromethane/hexanes); IR 1678, 1599, 1495
1
cm21; <cis 10f); H NMR <CDCl3) d 1.55 <d, Jˆ7.0 Hz,
3H), 3.19 <p, 7.0 Hz, 1H), 3.74 <dd, Jˆ7.0, 9.2 Hz, 1H),
6.05 <dd, Jˆ9.2, 15.4 Hz, 1H), 6.79 <d, Jˆ15.4 Hz, 1H),
1
7.35±7.22 <m, 5H); <trans 10f); H NMR <CDCl3) d 1.58
<d, Jˆ5.4 Hz, 3H), 2.91 <p, Jˆ5.4 Hz, 1H), 3.33 <dd, Jˆ5.4,
9.2 Hz, 1H), 5.79 <dd, Jˆ9.2, 15.4 Hz, 1H), 6.71 <d, Jˆ
15.4 Hz, 1H), 7.35±7.22 <m, 5H); 13C NMR <CDCl3) d
16.92, 21.24, 36.03, 37.74, 43.61, 46.24, 125.96, 126.08,
126.16, 126.76, 127.60, 127.66, 128.50, 130.03, 132.44,
134.41, 136.37, 136.53. The titled mixture has unstable
and decomposed over several days. Consequently 10f was
used immediately.
5.1.3. Preparation of 2-phenyl-3-vinylthiiranes (10a±c).
A typical procedure.10e To a stirred solution of oxirane 11a
<1.0 g, 6.85 mmol) in dioxane <1.3 ml) potassium thiocya-
nate <0.75 g, 7.53 mmol) and tetrabutyl ammonium bromide
<22 mg, 6.85 mol) in water <1.5 ml) was added. The solution
was stirred for 6 h at which time the aqueous layer was
removed and replaced with a solution of potassium thio-
cyanate <0.75 g, 7.53 mmol) in water <1.5 ml). The hetero-
geneous solution was stirred for a further 10 h at which time
diene formation was observed by TLC. Water <5 ml) was
added and the solution extracted with ether <3£10 ml). The
organic fractions were dried <MgSO4) and solvent removed
in vacuo. Thiirane was separated from starting oxirane and
diene by ¯ash chromatography <6% dichloromethane/
hexanes). Refer to Table 2 for yields of all thiiranes
5.1.6. Preparation of 2-[(E)-2-phenyl-1-ethenyl]thiirane
(10g).10d A typical procedure. Oxirane 11g <2.18 g,
14.93 mmol) was added dropwise to a stirred solution of
potassium thiocyanate <1.66 g, 17.12 mmol) in water
<1.8 ml) over 2.5 h. After an hour the aqueous phase was
removed and fresh potassium thiocyanate <0.83 g,
8.55 mmol) in water <1.8 ml) added. The reaction was
monitored by 1H NMR and after 48 h was complete.
Dichloromethane <50 ml) and water <50 ml) where then
added, the organic phase decanted, dried <CaCl2), and
solvent removed in vacuo. Crude thiirane was puri®ed by
¯ash chromatography <45% hexane/dichloromethane) to
afford 10g as an oil. See Table 2 for the yields of all thiiranes

4526
D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
prepared using this methodology. 10i10b,c was prepared from
oxirane 11i in an identical fashion.
<CDCl3) d 28.02, 43.47, 126.91, 128.66, 129.03, 129.63,
133.72, 138.46; MS m/z <%): 230 <M1 37Cl, 2%), 228
<M1 35Cl, 8), 195 <50), 164 <35), 129 <100); HRMS of
<14c), C10H9S2Cl: calcd, 227.9868, found, 227.9837.
5.1.7. 2-[(E)-2-Phenyl-1-ethenyl]-3-vinylthiirane (10h).
The title compound was prepared as a colourless oil from
oxirane 11h using the approach for thiirane 10a. Rf 0.31
<10% dichloromethane/hexane); IR 1634, 1597, 1491
5.1.11. 3-Methyl-6-phenyl-3,6-dihydro-1,2-dithiine (14d).
The title compound was prepared as a colourless oil from
10e or 10f using the approach for the synthesis of 14a. IR
1644, 1597, 1489 cm21; 1H NMR <CDCl3, 600 MHz) d 1.5
<d, Jˆ7.2 Hz, 3H), 3.42 <m, 1H), 4.58 <m, 1H), 5.95 <ddd,
Jˆ1.9, 3.2, 11.6 Hz, 1H), 6.02 <ddd, Jˆ2.2, 4.3, 11.6 Hz,
1H), 7.37±7.27 <m, 5H); 13C NMR <CDCl3) d 20.90, 34.68,
44.37, 127.85, 128.28, 128.53, 128.92, 133.12, 139.86; MS
m/z <%): 208 <M1, 5), 164 <7), 144 <70), 129 <100); HRMS
of <14d), C11H12S2: calcd, 208.0414; found, 208.0379.
1
cm21; <cis 10h); H NMR <CDCl3, 600 MHz) d 3.71 <dd,
Jˆ6.9, 8.8 Hz, 1H), 3.83 <dd, Jˆ6.9, 9.3 Hz, 1H), 5.54 <m,
3H), 6.01 <dd, Jˆ9.3, 15.5 Hz, 1H), 6.78 <d, Jˆ15.5 Hz,
1
1H), 7.39±7.21 <m, 5H); <trans 10h); H NMR <CDCl3,
600 MHz) d 3.43 <dd, Jˆ5.0, 8.4 Hz, 1H), 3.51 <dd, Jˆ
5.0, 8.9 Hz, 1H), 5.54 <m, 3H), 5.81 <dd, Jˆ9.1, 15.7 Hz,
1H), 6.74 <d, Jˆ15.7 Hz, 1H), 7.39±7.21 <m, 5H); 13C NMR
<CDCl3) d 42.55, 42.78, 43.78, 44.05, 117.68, 119.48,
125.78, 126.10, 126.17, 126.40, 127.67, 128.17, 128.46,
128.85, 132.68, 132.87, 134.41, 136.15, 136.27, 137.47;
MS m/z <%): 188 <M1, 30), 173 <20), 156 <75), 155 <100);
HRMS of <10h), C12H12S: calcd, 188.0677; found,
188.0665.
5.1.12. Preparation of 3-phenyl-6-vinyl-3,6-dihydro-1,2-
dithiine (14e) and 3-[(E)-2-phenyl-1-ethenyl]-3,6-di-
hydro-1,2-dithiine (14f). The title compounds were
prepared as colourless oils from 10h using the approach
for the synthesis of 14a. Rf0.31 <2% ethyl acetate/hexane);
1H NMR <CDCl3, 600 MHz) d 3.86 <m, 1H), 4.70 <q, Jˆ1.8,
1H), 5.29 <ddt, Jˆ0.6, 1.2, 10.2 Hz, 1H), 5.32 <dt, Jˆ1.2,
17.4 Hz, 1H), 6.01 <ddd, Jˆ1.8, 4.8, 11.4 Hz, 1H), 6.12±
6.06 <m, 2H), 7.39±7.25 <m, 5H); 13C NMR <CDCl3) d
27.78, 38.00, 40.88, 45.00, 117.48, 125.69, 126.13,
127.46, 128.09, 128.34, 128.48, 128.63, 128.69, 128.72,
129.14, 129.63, 130.78, 131.70, 135.83, 139.26; MS m/z
<%): 220 <M1, 10), 187 <40), 156 <100); HRMS of <14e),
C12H12S2: calcd, 220.0414, found, 220.0378.
5.1.8. 3-Phenyl-3,6-dihydro-1,2-dithiine (14a). A typical
procedure for the synthesis of 1,2-dithiins (14a±f). Either
vinylthiirane 10a or 10g <0.66 g, 4.07 mmol) and tungsten
pentacarbonyl monoacetonitrile <15 mg, 0.06 mmol) in
deuterated chloroform <0.5 ml) were shaken in an NMR
tube and allowed to react over 24 h after which time all
starting material had been consumed. The reaction mixture
was then concentrated in vacuo and separated by ¯ash
chromatography <5% ethyl acetate/hexane) to afford pure
1,2-dithiin as a colourless oil. Refer to Table 3 for yields
and reaction times for 1,2-dithiins prepared using this
1
14f: Rf 0.35 <2% ethyl acetate/hexane); H NMR <CDCl3,
methodology. Rf 0.45; IR 1641, 1597, 1492, 1451 cm21
;
600 MHz) d 3.14 <ddt, Jˆ2.4, 4.2, 18.0 Hz, 1H), 3.33 <ddt,
Jˆ2.4, 4.2, 15.6 Hz, 1H), 4.05 <m, 1H), 5.92 <ddt, Jˆ2.4,
4.2, 12.0 Hz, 1H), 6.07 <m, 1H), 6.29 <dd, Jˆ7.8, 15.6 Hz,
1H), 6.55 <d, Jˆ15.6 Hz, 1H), 7.39±7.21 <m, 5H); 13C NMR
<CDCl3) d 27.86, 28.91, 41.13, 48.41, 126.19, 126.33,
126.81, 127.03, 127.13, 127.56, 128.16, 128.22, 128.34,
132.65, 133.85, 135.21, 136.36, 140.45; MS m/z <%) 219
<M1, 4), 187 <30), 156 <100); HRMS of <14f), C11H12S2:
calcd, 220.0414, found, 220.0381.
1H NMR <CDCl3, 600 MHz) d 3.22 <ddt, Jˆ2.2, 4.8,
17.4 Hz, 1H), 3.47 <ddt, Jˆ2.2, 3.8, 17.4 Hz, 1H), 4.67 <p,
Jˆ2.2 Hz, 1H), 6.04 <ddt, Jˆ2.2, 3.8, 11.7 Hz, 1H), 6.17
<dddd, Jˆ2.2, 3.8, 4.8, 11.7 Hz, 1H), 7.39±7.28 <m, 5H);
13C NMR <CDCl3) d 27.97, 44.44, 126.45, 127.85, 128.28,
128.50, 129.61, 139.77; MS m/z <%), 194 <M1, 10), 33 <8),
130 <100), 115 <45); HRMS of <14a), C10H10S2: calcd,
194.0258; found, 194.0225.
5.1.9. 3-(4-Methoxyphenyl)-3,6-dihydro-1,2-dithiine (14b).
The title compound was prepared as a colourless oil from
10b using the approach for the synthesis of 14a. Rf 0.35 <3%
ethyl acetate/hexane); IR 1643, 1609, 1510 cm21; 1H NMR
<CDCl3, 600 MHz) d 3.18 <ddt, Jˆ2.4, 4.8, 17.4 Hz, 1H),
3.47 <ddt, Jˆ2.4, 3.6, 17.4 Hz, 1H), 3.79 <s, 3H), 4.65 <p,
Jˆ2.4 Hz, 1H), 6.01 <ddt, Jˆ2.4, 3.6, 11.4 Hz, 1H), 6.14
<dddd, Jˆ2.4, 3.6, 4.8, 11.4 Hz, 1H), 7.17±7.10 <m, 2H),
7.31±7.20 <m, 2H); 13C NMR <CDCl3) d 28.00, 44.03,
55.29, 114.01, 129.48, 126.36, 130.07, 131.81, 159.37;
MS m/z <%): 224 <M1, 10), 191 <67), 160 <100); HRMS
of <14b), C11H12S2O: calcd, 224.0363; found, 224.0333.
5.2. General procedure for kinetic studies
Thiirane 10a <93 mg, 0.573 mmol) and hexamethylbenzene
<13 mg, 8.02 mmol) in CDCl3 <0.7 ml) were mixed in a
5 mm NMR tube and an initial H NMR spectra taken.
1
Tungsten pentacarbonyl acetonitrile <5.5 mg, 15 mmol)
was then added and the solution shaken. Spectra were
then obtained periodically <4£15, 8£30 min then 6£60
min) on a Varian-Gemini-200 instrument utilising the
preacquisition delay <PAD) function. The reaction rate
was determined by comparing the area of the thiirane
doublet at d 4.2 ppm and the doublet of doublets at d 3.52
to the area of the hexamethyl benzene signal.
5.1.10. 3-(4-Chlorophenyl)-3,6-dihydro-1,2-dithiine (14c).
The title compound was prepared as a colourless oil from
10c using the approach for the synthesis of 14a. Rf0.35
<7.5% dichloromethane/hexane); IR 1648, 1593, 1489
1
cm21; H NMR <CDCl3, 600 MHz) d 3.26 <ddt, 2.2, 4.2,
Acknowledgements
17.6 Hz, 1H), 3.40 <ddt, Jˆ2.2, 4.2, 17.6 Hz, 1H), 4.59
<m, 1H), 5.98 <ddt, Jˆ2.2, 3.7, 11.7 Hz, 1H), 6.18 <ddt,
Jˆ2.2, 4.2, 11.7 Hz, 1H), 7.32±7.25 <m, 4H); 13C NMR
This work was supported by the Australian Research
Council.

D. W. Lupton, D. K. Taylor / Tetrahedron 58!2002) 4517±4527
4527
6. Bierer, D. E.; Dener, J. M.; Dubenko, L. G.; Gerber, E. R.;
Litvak, J.; Peterli, S.; Petrli-Roth, P.; Truong, T. V.; Mao, G.;
Baurer, B. E. J. Med. Chem. 1995, 38, 2628±2648.
7. Zoghbi, M. d. G.; Ramos, L. M.; Maia, J. G. S.; da Silva, M. L.;
Luz, A. I. R. J. Agric. Food Chem. 1984, 32, 1009±1010.
8. Sinha, N. K.; Guyer, D. E.; Gage, D. A.; Lira, C. T. J. Agric.
Food Chem. 1992, 40, 842±845.
References
1. <a) Avery, T. D.; Haselgrove, T. D.; Rathbone, T. J.; Taylor,
D. K.; Tiekink, E. R. T. Chem. Commun. 1998, 333±334.
<b) Avery, T. D.; Taylor, D. K.; Tiekink, E. R. T. J. Org.
Chem. 2000, 65, 5531±5546. <c) Avery, T. D.; Greatrex,
B. W.; Taylor, D. K.; Tiekink, E. R. T. J. Chem. Soc., Perkin
Trans. 1 2000, 1319±1321. <d) Avery, T. D.; Fallon, G.;
Greatrex, B. W.; Pyke, S. M.; Taylor, D. K.; Tiekink, E. R. T.
J. Org. Chem. 2001, 66, 7955±7966.
9. Tressl, R.; Bahri, D.; Hozer, M.; Kossa, T. J. Agric. Food
Chem. 1977, 25, 459±463.
10. <a) Sander, M. Chem. Rev. 1966, 66, 297±339. <b) Vedejs, E.;
Krafft, G. A. Tetrahedron 1982, 38, 2857±2881. <c) Bouda,
H.; Borredon, M. E.; Delmasa, M.; Gaset, A. Synth. Commun.
1987, 17, 943±951. <d) Synder, H. R.; Stewart, J. M.; Ziegler,
J. B. J. Am. Chem. Soc. 1947, 69, 2672±2674. <e) Suga, K.;
Endo, T. J. Polym. Sci., Part C: Polym. Lett. 1989, 27,
381±384.
2. <a) Steliou, K.; Gareau, Y.; Harpp, D. J. Am. Chem. Soc. 1984,
106, 799±801. <b) Steliou, K.; Salama, P.; Brodeur, D.;
Gareau, Y. J. Am. Chem. Soc. 1987, 109, 926±927. <c) Tardif,
S.; Rys, A.; Abrams, C.; Abu-Yousef, I.; Leste-Lasserre, P.;
Schultz, E.; Harpp, D. Tetrahedron 1997, 53, 12225±12236.
<d) Tardif, S. L.; Williams, C. R.; Harpp, D. N. J. Am. Chem.
Soc. 1995, 117, 9067±9068.
11. <a) Mosset, P.; Gree, R. Synth. Commun. 1985, 15, 749±757.
<b) Endo, T.; Kanda, N. J. Polym. Sci., Polym. Chem. Ed.
1985, 23, 1931±1938. <c) Koizumi, T.; Nojima, Y.; Endo, T.
J. Polym. Sci., Pt. A Polym. Chem. 1993, 31, 3489±3492.
<d) Auge, J.; David, S. Tetrahedron Lett. 1983, 24,
4009±4012.
3. <a) Adams, R. D.; Yamamoto, J. H.; Holmes, A.; Baker, B. J.
Organometallics 1997, 16, 1430±1439. <b) Adams, R. D.;
Queisser, J.; Yamamoto, J. J. Am. Chem. Soc. 1996, 118,
10674. <c) Adams, R. D.; Perrin, J. L. J. Am. Chem. Soc.
1999, 121, 3984±3991. <d) Adams, R.; Perrin, J. Polyhedron
2000, 19, 521±523. <e) Adams, R. D. Aldichim. Acta 2000, 33
<2), 39±48.
12. Ketcham, R.; Shah, V. P. J. J. Org. Chem. 1963, 28, 229±230.
13. Jordan, R. B. Reaction Mechanisms of Inorganic and
Organometallic Systems; 2nd ed; Oxford University: Oxford,
1998.
4. <a) Hudson, J. B.; Graham, E. A.; Fong, R.; Finlayson, A. J.;
Towers, G. H. N. Planta Med. 1986, 52, 51±54. <b) Hudson,
J. B.; Graham, E. A.; Chan, G.; Finlayson, A. J.; Towers,
G. H. N. Planta Med. 1986, 52, 453±457.
14. Landesburg, J. M.; Katz, L. J. Organomet. Chem. 1972, 35 <2),
327±336.
5. <a) Towers, G. H. N.; Abramowski, Z.; Finlayson, A. J.;
Zucconi, A. Planta Med. 1985, 51, 225±229. <b) Constabel,
C. P.; Towers, G. H. N. Planta Med. 1989, 55, 35±37.