Asymmetric synthesis of sulfinyl-substituted arene chromium tricarbonyl complexes

Article abstract of DOI:10.1039/a907590i

The synthesis of (SRS_s[(phenylsulfinyl)benzene] chromium tricarbonyl 5 and (SRS_s)-[(p-tolylsulfiny))benzene] chromium tricarbonyl 6 is achieved via a nucleophilic displacement reaction between the anion derived from (benzene) chromium tricarbonyl 9 and a suitable sulfinate ester. Replacing the sulfinate ester with a chiral sulfinyl-transfer reagent allows the isolation of the non-racemic sulfinyl-substituted complexes with good enantioselectivities (ee 80-89%) under optimised conditions. The use of Kagan's cyclic sulfite methodology for the synthesis of an enantiomerically pure tert-butylsulfinyl complex is unsuccessful, but results in the identification of a novel fragmentation - isomerisation process of the intermediate sulfinate. The Royal Society of Chemistry 1999.

Full text of DOI:10.1039/a907590i

View Article Online / Journal Homepage / Table of Contents for this issue
Asymmetric synthesis of sulfinyl-substituted arene chromium
tricarbonyl complexes
Stephen G. Davies,*a Tracey Loveridge,a M. Fatima C. C. Teixeiraa and John M. Cloughb
a Dyson Perrins Laboratory, University of Oxford, South Parks Road, Oxford, UK OX1 3QY
b ZENECA Agrochemicals, Jealott’s Hill Research Station, Bracknell, Berkshire,
UK RG12 6EY
Received (in Cambridge, UK) 20th September 1999, Accepted 6th October 1999
The synthesis of (SRSS)-[(phenylsulfinyl)benzene] chromium tricarbonyl 5 and (SRSS)-[(p-tolylsulfinyl)benzene]
chromium tricarbonyl 6 is achieved via a nucleophilic displacement reaction between the anion derived from
(benzene) chromium tricarbonyl 9 and a suitable sulfinate ester. Replacing the sulfinate ester with a chiral sulfinyl-
transfer reagent allows the isolation of the non-racemic sulfinyl-substituted complexes with good enantioselectivities
(ee 80–89%) under optimised conditions. The use of Kagan’s cyclic sulfite methodology for the synthesis of an
enantiomerically pure tert-butylsulfinyl complex is unsuccessful, but results in the identification of a novel
fragmentation–isomerisation process of the intermediate sulfinate.
carbonyl and a suitable chiral sulfinyl-transfer reagent, and
Introduction
report herein our studies towards this goal. Part of this work
The use of chiral arene chromium tricarbonyl complexes in
has been the subject of a preliminary communication.8
asymmetric synthesis is well established. However, the relative
paucity of methods for obtaining such complexes in homo-
Results and discussion
chiral form has reduced the scope of this methodology. In the
past, homochiral complexes have been accessed via chemical1
or enzymic kinetic resolutions2 or by the diastereoselective
complexation of a di- or tri-substituted benzene derivative
carrying a stereogenic centre on one of the side-chains.3 Since
1,2-differentially substituted arene chromium tricarbonyl
complexes are chiral and have been shown to undergo highly
stereoselective reactions, particular attention has recently been
focused on their asymmetric synthesis, either by the attachment
of a chiral auxiliary to the complexed arene ring in order to
accomplish diastereoselective ortho-deprotonation,4 or by the
use of a homochiral base to achieve enantioselective ortho-
deprotonation.5 However, almost all the reported routes to
enantiomerically pure arene chromium tricarbonyl complexes
require the presence of a side-chain which cannot subsequently
be cleaved, rendering them substrate specific. We envisaged that
a novel combination of sulfoxide and arene chromium tri-
carbonyl chemistries, via the introduction of a sulfoxide sub-
stituent onto the complexed arene ring, would allow access to a
homochiral complex. Furthermore, the sulfoxide functionality
was anticipated to induce regio- and stereoselective ortho-
deprotonation, whilst being a latent source of its ipso anion,
thus being removable or replaceable at a subsequent stage.
In a series of excellent papers Thomas and co-workers have
reported the diastereoselective oxidation of an ortho-
substituted alkylthio complex using dimethyldioxirane to
achieve the first documented synthesis of an alkylsulfinyl-
substituted arene chromium tricarbonyl complex.6 In an
extension of this methodology these workers have recently pub-
lished the direct asymmetric oxidation of (methylthiobenzene)
chromium tricarbonyl, employing Kagan’s modified Sharpless
reagent, to give the required sulfoxide with 83% enantiomeric
excess (ee). However, despite the efficacy of this methodology in
achieving an asymmetric oxidation of this complex it is com-
pletely ineffective for other alkylthio or arylthio substituents.7
We envisaged an alternative simple approach to the synthesis of
homochiral sulfinyl-substituted arene chromium tricarbonyl
complexes, via a nucleophilic displacement reaction between
the anion derived by deprotonation of (benzene) chromium tri-
The methyl benzene-, methyl toluene-p- and methyl 1,1-
dimethylethanesulfinates 1, 2, and 3 were all prepared in
reasonable to good yields according to literature procedures,9
whilst (benzene) chromium tricarbonyl 4 was synthesised by
the thermolysis of benzene with chromium hexacarbonyl in a
mixture of dibutyl ether and THF (10:1).
With these starting materials in hand the synthesis of a
sulfinyl-substituted arene chromium tricarbonyl complex was
attempted. Complex 4 was dissolved in THF and cooled to
Ϫ78 ЊC before being treated with butyllithium and stirred for 2
h, in order to effect aryl deprotonation.10 The resulting brown
solution was then slowly added to a solution of 1 in THF at
Ϫ78 ЊC and the reaction mixture stirred at this temperature for
14 h. Work-up and crystallisation led to the isolation of 5 as
yellow–orange needles in good yield (71%), representing a sig-
nificant improvement on the 55% yield achieved by Thomas
and co-workers via dimethyldioxirane oxidation of the corre-
sponding sulfide complex.7 The synthesis of 6 was achieved in
an analogous manner to give the novel complex as yellow–
orange needle-like crystals in good yield (64%) (Scheme 1).
Scheme 1 Reagents and conditions: i, BuLi (1.1 equiv.), Ϫ78 ЊC; ii, 1 or
2 (1.3 equiv.), Ϫ78 ЊC.
J. Chem. Soc., Perkin Trans. 1, 1999, 3405–3412
This journal is © The Royal Society of Chemistry 1999
3405

View Article Online
In the synthesis of both 5 and 6, the slow inverse addition
protocol described was found to be vital in order to obtain good
yields. Alternative procedures were found to afford significant
quantities of a very polar by-product which appeared to
account for the reduced yields observed.11
It was anticipated that this methodology could also be
applied to the synthesis of the tert-butylsulfinyl complex 7.
Reaction of 4 with 3, in an analogous manner to the synthesis
of 5, gave low conversion to the required product as indicated
by TLC analysis of the crude reaction mixture. Hence, the reac-
tion was allowed to warm to 0 ЊC over 2.5 h before work-up.
1
Unfortunately, H NMR spectroscopic analysis of the crude
product showed the presence of only a trace of 7. Column
chromatography allowed isolation of returned starting material
4, followed by 7 as a yellow oil in low yield (3%) which had an
1
identical H NMR spectrum to that described in the literature
(Scheme 2).7 This disappointing result has been attributed to
Scheme 2 Reagents and conditions: i, BuLi (1 equiv.), Ϫ78 ЊC; ii, 3 (1.4
equiv.), Ϫ78 to 0 ЊC.
the steric bulk of both 3 and the reacting anion of 4 which
appears to prevent close approach of the anion to the sulfur
centre precluding reaction. Increasing the reaction temperature
did not improve this situation, resulting only in an increased
amount of an unidentified, very polar by-product.
With the racemic complexes 5, 6 and 7 in hand it was
anticipated that the above methodology could be extended
to the synthesis of the corresponding homochiral complexes,
by employing a suitable chiral sulfinyl-transfer reagent in place
of the racemic sulfinate ester used previously.
For the preparation of (SS)-5 the Evans’ sulfinyl transfer
reagent (4R,5S,RS)-8 derived from (1S,2R)-norephedrine was
identified as a candidate worthy of investigation due to its
efficacy in the synthesis of homochiral sulfoxides, achieved via a
clean SN2 displacement reaction at the sulfur stereocentre on
treatment with an appropriate nucleophile. Hence, (4R,5S,RS)-
8 was synthesised in good yield according to the literature
procedure.12 Reaction of (phenyllithium) chromium tricarbonyl
9 with (4R,5S,RS)-8 (1.2 equiv.) in THF at Ϫ78 ЊC over a 19 h
period led to isolation of “(SS)”-5 in 78% yield but with an ee of
just 8% as determined by 1H NMR spectroscopic studies
employing the chiral solvating agent (S)-(ϩ)-N-(3,5-dinitro-
benzoyl)-α-methylbenzylamine13 (Scheme 3).
Scheme 4
(SS)-5 in preference to the required reaction with (4R,5S,RS)-8,
resulting in racemisation at the sulfur centre of the newly
formed complex (path A) and reflecting the stability of 9.
Furthermore, the initial step in this sequence generates an
oxazolidinone anion 10. This in turn may react with (4R,5S,RS)-
8, epimerising the sulfur centre of the sulfinyl-transfer reagent
to give (4R,5S,SS)-11 (path B). Obviously, reaction of this with
the anion of 4 will also form the unwanted enantiomer of 5
(path C). This path B/path C racemisation route was identi-
fied by Evans and used to explain the reduced enantiomeric
purities of the sulfoxides obtained on reaction of the related
sulfinyl-transfer reagent (4S)-4-benzyl-3-[(S)-p-tolylsulfinyl]-
isoxazolidin-2-one with organolithium reagents.12
With these ideas in mind it was apparent that alteration of
the reaction conditions should allow isolation of (SS)-5 of
improved enantiomeric purity. Thus, a cooled (Ϫ100 ЊC) THF
solution of the anion of 4 was added rapidly to (4R,5S,RS)-8
(10 equiv.) in THF at the same temperature. On completion of
the addition the reaction was quenched immediately using
1
saturated aq. ammonium chloride and worked up. H NMR
analysis of the crude material indicated 68% conversion to
product and the diastereomeric purity of 8 to be eroded to 94%,
validating path B and therefore path C as a route to racemis-
ation in this synthesis. Column chromatography allowed the
components of the mixture to be separated, the first fraction
affording
a white solid which on crystallisation gave
(4R,5S,RS)-8 as white needles (72%). The second fraction
yielded “(SS)”-5 (51%) with a much improved ee of 80%
(Scheme 5). Unfortunately, this enantiomeric purity could not
be reliably improved upon by crystallisation.14
Scheme 3 Reagents and conditions: i, BuLi (1 equiv.), Ϫ78 ЊC; ii,
(4R,5S,RS)-8 (1.2 equiv.), Ϫ78 ЊC.
Despite the identification of path B as a route to racemis-
ation the lower enantiomeric purity of the sulfoxide, compared
with that of the sulfinyl-transfer reagent, indicated that a sec-
ond racemisation pathway must also be in operation. Hence, a
Ϫ100 ЊC solution of (phenyllithium) chromium tricarbonyl 9
was rapidly added to a solution of 5 (ee 39%) in THF at
Ϫ100 ЊC. Immediate quenching furnished “(SS)”-5 of only
This unexpected and disappointing result focused attention
on the identification of possible routes for the racemisation
occurring in this reaction. Two such pathways were recognised
as feasible, as illustrated in Scheme 4. Initial reaction between
anion 9 and (4R,5S,RS)-8 must generate the homochiral com-
plex (SS)-5. However, any unchanged 9 may now react with
3406
J. Chem. Soc., Perkin Trans. 1, 1999, 3405–3412

View Article Online
Scheme 8 Reagents and conditions: i, BuLi (1.1 equiv.), Ϫ78 ЊC; ii,
(2S,SS)-13 (0.4 equiv.), Ϫ78 ЊC to rt.
Scheme 5 Reagents and conditions: i, BuLi (1 equiv.), Ϫ78 ЊC; ii,
(4R,5S,RS)-8 (10 equiv.), Ϫ100 ЊC.
low yield achieved in the synthesis of rac-7, it was assumed that
steric encumbrance once again accounted for the lack of
reactivity seen.
It was considered that reversing the order of addition of the
two nucleophilic reagents should circumvent this problem.
Thus, treatment of a cooled (Ϫ78 ЊC) THF solution of (2R,5S)-
trans-5-methyl-4,4-diphenyl-1,3,2-dioxathiolane 2-oxide 14
with the anion of 4 (0.9 equiv.) gave a yellow solid shown to
comprise a mixture of the regioisomers (2S,RS)-15 and (2S,SS)-
16 in the ratio 1:4.5, together with 50% unchanged starting
material by 1H NMR spectroscopy (Scheme 9). Attempted
Scheme 6 Reagents and conditions: i, 9 (0.6 equiv.), Ϫ100 ЊC.
6% enantiomeric excess, as determined by 1H NMR chiral
shift studies (Scheme 6). This result served to confirm the
operation of path A in the racemisation process. Indeed, the
rapid racemisation observed made the obtention of a scalemic
sulfoxide by the above methodology all the more remarkable.
The synthesis of (RS)-6 was attempted in a manner analog-
ous to that for (SS)-5, employing (1S,2R,5S,RS)-(ϩ)-menthyl
toluene-p-sulfinate 12 as the chiral sulfinyl-transfer reagent. As
previously, duplication of the reaction conditions used for
the synthesis of its racemic counterpart, replacing 2 with
(1S,2R,5S,RS)-(ϩ)-12, afforded “(RS)”-6 with a low ee of 37%.
However, rapid quenching of the reaction between (phenyl-
lithium) chromium tricarbonyl 9 and (1S,2R,5S,RS)-(ϩ)-12 (9.5
equiv.) at Ϫ100 ЊC afforded “(RS)”-6 in good yield (77%) and
enantiomeric purity (ee 89%) (Scheme 7). (1S,2R,5S,RS)-(ϩ)-
Scheme 7 Reagents and conditions: i, BuLi (1 equiv.), Ϫ78 ЊC; ii,
(1S,2R,5S,RS)-(ϩ)-12 (10 equiv.), Ϫ100 ЊC.
Scheme 9 Reagents and conditions: i, BuLi (1.1 equiv.), Ϫ78 ЊC; ii,
(2R,5S)-14 (1.1 equiv.), Ϫ78 ЊC.
12 was isolated diastereomerically pure, indicating that in this
instance the sulfinyl-transfer reagent is unlikely to be implicated
in any racemisation pathway. A path A-type racemisation
process is proposed as the only viable mechanism for the
observed racemisation and was confirmed by the reaction of
the anion of 4 with 6 (ee 68%) at Ϫ100 ЊC. Allowing reaction
between these two substrates for 5 min before quenching and
work-up afforded 6 in essentially racemic form (2% ee).
For the attempted synthesis of a homochiral tert-butyl-
sulfinyl chromium tricarbonyl complex, the cyclic sulfite meth-
odology developed by Kagan15 was considered to be the most
appealing, as simple alteration of the order of addition of
the two nucleophilic reagents should allow access to both
enantiomers of the required complex.
The route to (SS)-7 was anticipated to be straightforward,
involving the intermediacy of the known compound (2S,SS)-3-
hydroxy-3,3-diphenylpropyl 1,1-dimethylethanesulfinate 13.15
Hence, a solution of the anion of 4 (2.3 equiv.) in THF at
Ϫ78 ЊC was added over 40 min to (2S,SS)-13 in THF at the
same temperature, and the resultant solution allowed to warm
to ambient temperature over 16 h. However, inspection of the
1H NMR spectrum of this crude material indicated only the
presence of starting materials (Scheme 8). By analogy with the
crystallisation from hexane–dichloromethane (2:1) led to the
isolation of the minor diastereomer (2S,RS)-15 as a yellow
powder (15% yield). Further crystallisation of the mother
liquors afforded a second crystalline fraction as orange rods
(13% yield) shown to be a 1.3:1 mixture of (2S,RS)-15 and
1
(2S,SS)-16 by H NMR spectroscopic analysis. Calculation of
the combined yield of compound 15 showed it to have been
isolated in greater than 100% mass balance from the original
mixture, consistent with an in situ regioisomerisation to the
more stable regioisomer (2S,RS)-15. This is particularly sur-
prising in the light of the observation by Kagan that the
corresponding (2S,RS)-1,1-dimethylethanesulfinate analogue
decomposed at room temperature to give diphenylacetone.
The regioisomeric assignment was confirmed by the 1H
NMR spectra of (2S,RS)-15 and (2S,SS)-16. That of (2S,RS)-15
demonstrated a distinctive multiplet at δ 5.18–5.15 [HOCH-
(CH3)] due to coupling of this hydrogen with both the adjacent
methyl and hydroxy hydrogens, which is only possible in this
regioisomer. By contrast, a simple quartet (δ 5.59, 1H, q, J 6.4,
CH3CH) was observed for the corresponding proton in the
spectrum for (2S,SS)-16. It is reasonable to assume that
the anion of 4 can be regarded as a bulky nucleophile; hence,
J. Chem. Soc., Perkin Trans. 1, 1999, 3405–3412
3407

View Article Online
the sense of regioselectivity observed in the cleavage of the
cyclic sulfite is in agreement with the model proposed by
Kagan.15
With the pure complex (2S,RS)-15 in hand the preparation of
(SS)-7 was attempted. Treatment of a THF solution of (2S,RS)-
15 with tert-butylmagnesium chloride (2.2 equiv.) furnished a
yellow crystalline solid shown to comprise a single compound
Scheme 12
1
for their observations (Scheme 12).19 The driving force for this
isomerisation process is obviously the formation of the strong
sulfur–oxygen bond in the sulfonyl group. Rearrangement of
the intermediate anion 18 to the sulfone 17 may occur via
homolytic fission of the C–O bond, followed by a novel radical–
radical anion-pair isomerisation, in an analogous fashion to the
Wittig rearrangement of ethers with alkyllithium reagents.
Alternatively, protonation of the anion on work-up may be
followed by the standard sulfinate ester–sulfone rearrangement
which, if this is correct, would appear to proceed more rapidly
for the complexed sulfinate than for its uncomplexed counter-
part (benzhydryl toluene-p-sulfinate undergoes isomerisation
only at elevated temperatures in acetic acid20). However, the
precise course of this rearrangement remains open to
speculation.
It was predicted that (2S,SS)-16 would not undergo an
analogous fragmentation reaction on treatment with tert-
butylmagnesium chloride due to the inability of a methyl group
to stabilise an adjacent carbanion. Hence, reaction of a mixture
of (2S,RS)-15 and (2S,SS)-16 with tert-butylmagnesium
chloride was expected to afford the required homochiral com-
plex (RS)-7 together with the sulfone 17 and to obviate the need
to separate (2S,RS)-15 and (2S,SS)-16. Thus, a solution of
(2S,RS)-15 and (2S,SS)-16 (1.3:1) in THF was treated with tert-
butylmagnesium chloride (2.4 equiv.) to give a yellow oil shown
to comprise 7 and 17 in the ratio 1:2.7. Column chromato-
graphy on silica gel allowed separation of these two com-
ponents, which were both isolated in low yield (17 24% yield;
7 9% yield). However, 1H NMR spectroscopic analysis of
“(SS)”-7 using the chiral solvating agent (S)-(ϩ)-N-(3,5-
dinitrobenzoyl)-α-methylbenzylamine showed it to have an ee
of just 3% (Scheme 13).
(98% yield) by H NMR spectroscopy (Scheme 10). However,
Scheme 10 Reagent: i, tBuMgCl (2.2 equiv.).
none of the observed signals corresponded to those seen in
1
the H NMR spectrum of rac-7. Instead, two triplets and a
doublet were observed in the region δ 5.56–4.94. In addition,
a distinctive one-proton singlet at δ 5.33 and two aromatic
multiplets [δ 7.61–7.57 (4H, m, Ph), 7.39–7.36 (6H, m, Ph)] were
noted. IR spectroscopy showed two strong absorptions at 1329
and 1145 cmϪ1, indicative of a sulfone functional group, in add-
ition to the two very strong absorption bands at 1996 and 1937
cmϪ1 corresponding to the stretching modes of the CO
ligands.16 On this basis the compound was tentatively assigned
as benzhydryl (phenyl) chromium tricarbonyl sulfone 17
although the alternative [(benzhydrylsulfinyl)benzene] chrom-
ium tricarbonyl structure could not be discounted.
Confirmation of the structure of 17 was achieved by sub-
jecting this complex to standard oxidative decomplexation
conditions to give benzhydryl phenyl sulfone, shown to be iden-
1
tical to an independently synthesised, authentic sample by H
and 13C NMR spectroscopy.
The generation of 17 is postulated to occur via a fragment-
ation of the initially generated alkoxide anion as illustrated
(Scheme 11), facilitated by the ability of the benzhydryl group
Scheme 11
to stabilise the negative charge in the resultant carbanion 18. It
is noteworthy that Kagan and co-workers did not witness this
fragmentation process, as in all cases they reported quantitative
conversion of the intermediate sulfinate to the required sulf-
oxide. It is possible that the steric bulk at the sulfur centre in
(2S,RS)-15, due to the presence of the (benzene) chromium
tricarbonyl fragment, reduces the propensity for reaction
between the sulfinate and the second equivalent of the Grignard
reagent present, resulting in the alternative reaction pathway
observed being favoured.
The rearrangement of sulfinate esters to sulfones has been
the subject of extensive investigation.18 Darwish and McLaren
have studied the rearrangement of benzhydryl 2,6-dimeth-
ylbenzenesulfinate to the corresponding sulfone under a range
of conditions and propose an ion-pair mechanism to account
Scheme 13 Reagent: i, tBuMgCl (2.4 equiv.).
3408
J. Chem. Soc., Perkin Trans. 1, 1999, 3405–3412

View Article Online
In conclusion, it has been shown that the synthesis of the
sulfinyl-substituted chromium tricarbonyl complexes 5 and 6
can be readily achieved in good yield by a simple nucleophilic
displacement reaction between the anion of 4 and a suitable
sulfinate ester. However, the synthesis of the tert-butyl analogue
7 using this methodology was much less successful. Synthesis of
the non-racemic material (SS)-5 and (RS)-6 was achieved with
reasonable success employing (4R,5S)-4-methyl-5-phenyl-3-
[(R)-phenylsulfinyl]isoxazolidin-2-one and (1S,2R,5S,RS)-(ϩ)-
menthyl toluene-p-sulfinate as the chiral sulfinyl transfer
reagents [(SS)-5 ee 80%; (RS)-6 ee 89%]. Identification of the
racemisation pathways in operation enabled rational exper-
imental design in order to optimise the ees obtained. In
contrast, all attempts to achieve an asymmetric synthesis of 7
using Kagan’s cyclic sulfite methodology failed. Instead a novel
fragmentation–isomerisation process was observed.
thermally jacketed 10 cm cell. [α]D-Values are given in units of
10Ϫ1 deg cm2 gϪ1. Mp were recorded on a Gallenkamp hot-stage
apparatus and are uncorrected.
Preparation of (benzene) chromium tricarbonyl 423
A solution of benzene (17.5 ml, 0.196 mol) and chromium
hexacarbonyl (4.02 g, 18.3 mmol) in dibutyl ether (120 ml)–
THF (10 ml) was rigorously degassed and heated at reflux
under nitrogen for 48 h. After cooling, the resulting yellow solu-
tion was filtered through alumina (Grade I), washing initially
with petrol (100 ml) to remove the dibutyl ether, followed by
dichloromethane (200 ml) to yield a yellow solution. Evapor-
ation of the solvents in vacuo gave the title compound 4 as a
yellow solid (2.35 g, 60%); δH (300 MHz) 5.33 (6H, s, ArH).
General procedure A: Synthesis of (SRS)-sulfinyl-substituted
arene chromium tricarbonyl complexes
Experimental
General
To a solution of (benzene) chromium tricarbonyl 4 in THF at
Ϫ78 ЊC was added butyllithium dropwise. The resultant brown
solution was stirred for 2 h before its addition over 40 min, via
cannula, to a solution of the required sulfinyl-transfer reagent
in THF at Ϫ78 ЊC. The mixture was stirred at this temperature
for 14 h and then quenched by the addition of saturated aq.
ammonium chloride (20 ml) and warmed to room temperature.
The product was extracted with ethyl acetate (3 × 20 ml), fil-
tered through a plug of magnesium sulfate and silica, and the
solvent removed in vacuo.
All reactions involving air- or moisture-sensitive reagents were
carried out under an atmosphere of dry nitrogen unless stated
otherwise. All experiments and purifications involving (arene)
chromium tricarbonyl complexes were performed using
standard vacuum line and Schlenk tube techniques,21 with all
solvents being deoxygenated prior to use. THF and diethyl
ether were distilled, under an atmosphere of nitrogen, from
sodium benzophenone ketyl, and dichloromethane from cal-
cium hydride, also under nitrogen. Dibutyl ether was passed
through a short plug of alumina (Grade I) prior to distillation
from calcium hydride under a nitrogen atmosphere. ‘Petroleum
ether’ (petrol) refers to that fraction which boils in the range
40–60 ЊC and was re-distilled before use. Chromium hexa-
carbonyl was sublimed under reduced pressure before use.
Butyllithium was used as a 1.3–2.1 M solution in hexane, and
tert-butylmagnesium chloride as a 2 M solution in diethyl ether.
Silica gel column chromatography was performed on Kieselgel
60 according to the guidelines of Still et al.22 TLC was carried
out using Camlab Polygram SIL G/UV254, with a 0.25 mm
Preparation of (SRS)-[(phenylsulfinyl)benzene] chromium tri-
carbonyl 5. A solution of 4 (879 mg, 4.10 mmol) in THF (30 ml)
was treated with butyllithium (3.47 ml, 4.51 mmol) and 1 (860
mg, 5.53 mol) in THF (10 ml) according to General Procedure
A to give a yellow–orange oil. Crystallisation from hexane–ethyl
acetate (1:1) yielded the title compound 5 as yellow–orange
needles (978 mg, 71%), mp 86–89 ЊC (Found: C, 52.98; H, 2.67.
C15H10CrO4S requires C, 53.26; H, 2.98%); νmax (CHCl3)/cmϪ1
᎐
1986, 1921 vs (C᎐O), 1040 w (S᎐O); δ (300 MHz) 7.75–7.73
᎐
᎐
H
(2H, m, Ph), 7.56–7.53 (3H, m, Ph), 5.88 (1H, d, J 6.5, ArH2),
5.43 (1H, d, J 6.4, ArH6), 5.39 (1H, t, J 6.2, ArH4), 5.30 (1H, t,
J 6.4, ArH3), 5.17 (1H, t, J 6.3, ArH5); δC (125 MHz) 230.4
[Cr(CO)3], 144.4 (Ph Ci), 132.0 (Ph Cp), 129.6, 124.4 (Ph Co,
Cm), 113.9 (Ar C1), 93.4, 89.9, 89.0, 88.7, 87.8 (Ar C2–6); m/z (CI,
NH3) 356 (MNH4ϩ, 6%), 339 (10), 323 (7), 203 (100).
coating of silica gel containing fluorescent indicator UV254
Visualisation was effected by UV and palladium() chloride in
acidic methanol. H NMR spectra were measured in deuterio-
.
1
chloroform unless otherwise stated, using residual chloroform
as an internal standard (δ 7.27), on either a Bruker WH 300
(300 MHz) or AM 500 (500 MHz) instrument. Coupling con-
stants (J ) are all given in Hertz and first-order approximations
are used throughout. A Bruker AMX 500 (125 MHz) machine
was used to record 13C NMR spectra, which were obtained in
deuteriochloroform unless specified otherwise. The following
numbering system has been utilised for the complexed arene
ring proton and carbon resonances, and ArЈ used to designate
resonances due to previously complexed aryl rings.
Preparation of (SRS)-[(p-tolylsulfinyl)benzene] chromium tri-
carbonyl 6. A solution of 4 (200 mg, 0.93 mmol) in THF (5 ml)
was treated with butyllithium (642 µl, 1.03 mmol) and 2 (196
mg, 1.16 mmol) in THF (10 ml) according to General Pro-
cedure A to give a yellow–orange oil. Crystallisation from
petrol–ethyl acetate (2:1) yielded the title compound 6 as
yellow–orange needles (210 mg, 64%); mp 133–136 ЊC (Found:
C, 54.42; H, 3.56. C16H12CrO4S requires C, 54.54; H, 3.43%);
νmax (CHCl3)/cmϪ1 1985, 1921 vs (C᎐O); δ (300 MHz) 7.63
᎐
᎐
H
(2H, d, J 8.1, Ph), 7.35 (2H, d, J 8.1, Ph), 5.90 (1H, d, J 6.5,
ArH2), 5.40–5.36 (2H, m, ArH4, H6), 5.31 (1H, t, J 6.0, ArH3),
5.16 (1H, t, J 6.3, ArH5), 2.42 (3H, s, CH3); δC (125 MHz) 230.9
[Cr(CO)3], 143.3, 141.6 (Ph Ci, Cp), 130.7, 125.1 (Ph Co, Cm),
114.7 (Ar C1), 93.7, 90.2, 89.5, 89.2, 88.4 (Ar C2–6), 21.9 (CH3);
m/z (CI, NH3) 353 (MHϩ, 13%), 337 (50), 217 (100).
Mass spectra were recorded by the Dyson Perrins Laboratory
analytical department on either a VG MICROMASS ZAB-1F
or VG MASSLAB VG 20–250 machine, employing either
chemical impact, electron impact or fast atom bombardment
techniques. Elemental microanalyses were performed by the
Dyson Perrins Laboratory analytical department. Unless
otherwise stated, all IR spectra were recorded for samples as
chloroform solutions in 0.1 mm sodium chloride cells on a
Perkin-Elmer 1750 FT spectrophotometer. Optical rotations
were recorded using a Perkin-Elmer 241 polarimeter with a
Preparation of (SRS)-[(tert-butylsulfinyl)benzene] chromium
tricarbonyl 7. A solution of 4 (119 mg, 0.56 mmol) in THF (5
ml) was treated with butyllithium (350 µl, 0.56 mmol) and 3
(0.104 g, 0.77 mmol) in THF (10 ml) according to General
Procedure A except that the solution was allowed to warm to
0 ЊC over 2.5 h before quenching with saturated aq. ammonium
chloride and reaction work-up under standard conditions to
yield a yellow oil. Purification by column chromatography on
silica gel [EtOAc–petrol (1:1)] yielded initially returned starting
J. Chem. Soc., Perkin Trans. 1, 1999, 3405–3412
3409

View Article Online
material 4 (28 mg, 24% recovery), followed by the title com-
pound 7 as a yellow oil (5 mg, 3%); δH (300 MHz) 5.88 (1H, d,
J 6.3, ArH2), 5.45–5.40 (2H, m, ArH4, H6), 5.34 (1H, t, J 6.2,
ArH3), 5.20 (1H, t, J 6.2, ArH5), 1.25 [9H, s, C(CH3)3].
The above procedure was repeated with the following
modifications.
A solution of 4 (39 mg, 0.18 mmol) in THF (2 ml) was
treated with butyllithium (137 µl, 0.19 mmol) according to
General Procedure A before cooling to Ϫ100 ЊC and fast
addition of the resultant anion to a solution of (1S,2R,5S,RS)-
(ϩ)-12 (507 mg, 1.72 mmol) in THF (20 ml) at Ϫ100 ЊC. The
reaction mixture was stirred for 5 min before being quenched by
the addition of saturated aq. ammonium chloride (20 ml).
Reaction work-up according to General Procedure A afforded
a mixture of a yellow oil and a white solid comprising 77% of 6
as determined by 1H NMR spectroscopic analysis. Column
chromatography on silica gel [petrol–EtOAc (2:1)] led to the
isolation of two fractions, the first (1S,2R,5S,RS)-(ϩ)-12 as
white needles (413 mg, 82% recovery); [α]D23 ϩ203 (c 1.05,
CH3COCH3) {lit.,24 (Ϫ)-menthyl toluene-p-sulfinate [α]D21 Ϫ201
(c 2.0, CH3COCH3)}. The second fraction gave the title
compound 6 as a yellow solid (50 mg, 77%). 1H NMR spectro-
scopic studies using (S)-(ϩ)-N-(3,5-dinitrobenzoyl)-α-methyl-
benzylamine as a chiral solvating agent showed the product
(RS)-6 to have an ee of 89%.
Attempted preparation of (SS)-[(phenylsulfinyl)benzene]
chromium tricarbonyl 5. A solution of 4 (598 mg, 2.79 mmol) in
THF (15 ml) was treated with butyllithium (1.75 ml, 2.80
mmol) and a solution of (4R,5S,RS)-8 (965 mg, 3.34 mol) in
THF (25 ml) according to General Procedure A except that
the solution was stirred at Ϫ78 ЊC for 19 h before quenching
with saturated aq. ammonium chloride (40 ml). The product
was extracted with diethyl ether (3 × 20 ml), filtered through a
plug of magnesium sulfate and silica, and the solvent removed
in vacuo to yield a yellow–orange oil. Purification by column
chromatography on silica gel [petrol–EtOAc (2:1)] gave an
orange solid, which was recrystallised from diethyl ether to give
1
5 as orange crystals (733 mg, 78%). H NMR spectroscopic
studies using (S)-(ϩ)-N-(3,5-dinitrobenzoyl)-α-methylbenzyl-
amine as a chiral solvating agent showed the product (SS)-5 to
have an ee of 8%.
The above procedure was repeated with the following
modifications.
Treatment of (RS)-[(p-tolylsulfinyl)benzene] chromium tri-
carbonyl 6 with (phenyllithium) chromium tricarbonyl 9. In an
analogous manner to the treatment of 5 with (phenyllithium)
chromium tricarbonyl 9 a solution of 4 (17 mg, 0.079 mmol) in
THF (5 ml) was treated with butyllithium (43 µl, 0.069 mmol)
and (RS)-6 (10 mg, 0.028 mmol, ee 68%) in THF (2 ml) at
Ϫ100 ЊC. The reaction mixture was stirred for 5 min before
being quenched by the addition of saturated aq. ammonium
chloride (10 ml) and work-up according to General Procedure
A to afford a yellow oil. 1H NMR spectroscopic studies
using (S)-(ϩ)-N-(3,5-dinitrobenzoyl)-α-methylbenzylamine as
a chiral solvating agent on the crude reaction mixture showed
the ee of 5 to have been reduced to 2%.
A solution of 4 (61 mg, 0.29 mmol) in THF (10 ml) was
treated with butyllithium (182 µl, 0.29 mmol) according to
General Procedure A before cooling to Ϫ100 ЊC and fast
addition via cannula of the resultant anion to a solution of
(4R,5S,RS)-8 (824 mg, 2.85 mmol) in THF (30 ml) at Ϫ100 ЊC.
The reaction mixture was quenched by the addition of satur-
ated aq. ammonium chloride (30 ml) as soon as the addition
of the anion was complete. Reaction work-up according to
General Procedure A afforded a mixture of a yellow oil and
a white solid comprising 68% of 5 and the sulfinyl-transfer
1
reagent 8 (de 94%) as determined by H NMR spectroscopic
analysis. Column chromatography on silica gel [petrol–EtOAc
(2:1)] led to the isolation of two fractions, the first a white solid
which, on recrystallisation, yielded (4R,5S,RS)-8 as white
needles (592 mg, 72% recovery). The second fraction gave the
Attempted preparation of (SS)-[(tert-butylsulfinyl)benzene]
chromium tricarbonyl 7. A solution of 4 (50 mg, 0.23 mmol) in
THF (2 ml) was treated with butyllithium (153 µl, 0.25 mmol)
and a solution of 13 (34 mg, 0.10 mmol) in THF (5 ml) accord-
ing to General Procedure A except that the resultant solution
was allowed to warm to ambient temperature over 16 h before
being quenched to give, after standard work-up, a mixture of
a yellow oil and white crystals. Analysis of the 1H NMR
spectrum of the crude product indicated only the presence of
starting materials.
1
title compound 5 as a yellow solid (49 mg, 51%). H NMR
spectroscopic studies using (S)-(ϩ)-N-(3,5-dinitrobenzoyl)-α-
methylbenzylamine as a chiral solvating agent showed the
product (SS)-5 to have an ee of 80%.
Treatment of (SS)-[(phenylsulfinyl)benzene] chromium tricar-
bonyl 5 with (phenyllithium) chromium tricarbonyl 9. A solution
of 4 (71 mg, 0.332 mmol) in THF (2 ml) was treated with butyl-
lithium (186 µl, 0.30 mmol) according to General Procedure A
before cooling to Ϫ100 ЊC and fast addition of the resultant
anion 9 to a solution of (SS)-5 (65 mg, 0.19 mmol, ee 39%)
in THF (2 ml) at Ϫ100 ЊC. The reaction mixture was stirred for
30 s before being quenched by the addition of saturated aq.
ammonium chloride (5 ml) and work-up according to General
Preparation of (2S,RS)-2-hydroxy-1,1-diphenylpropyl(benzene)
chromium tricarbonyl sulfinate 15 and (2S,SS)-1-hydroxy-1,1-
diphenylpropan-2-yl(benzene) chromium tricarbonyl sulfinate 16
A solution of 4 (255 mg, 1.19 mmol) in THF (5 ml) was treated
with butyllithium (818 µl, 1.31 mmol) and
a solution
1
Procedure A. H NMR spectroscopic studies using (S)-(ϩ)-N-
of (2R,5S)-trans-5-methyl-4,4-diphenyl-1,3,2-dioxathiolane 2-
oxide 14 (346 mg, 1.26 mmol) in THF (15 ml) according to
General Procedure A except that the solution was stirred at
Ϫ78 ЊC for 20 h before quenching. Standard work-up led to the
isolation of a yellow solid, which was shown to comprise a
mixture of (2S,RS)-15 and (2S,SS)-16 in the ratio 1:4.5,
(3,5-dinitrobenzoyl)-α-methylbenzylamine as a chiral solvating
agent on the crude reaction mixture showed the ee of 5 to have
been reduced to 6%.
Attempted preparation of (RS)-[(p-tolylsulfinyl)benzene]
chromium tricarbonyl 6. In an analogous manner to the prepar-
ation of (SS)-5 a solution of 4 (251 mg, 1.17 mmol) in THF
(10 ml) was treated with butyllithium (730 µl, 1.17 mmol)
and (1S,2R,5S,RS)-(ϩ)-menthyl toluene-p-sulfinate 12 (407 mg,
1.38 mmol) in THF (10 ml) at Ϫ78 ЊC. Work-up and column
chromatography on silica gel [petrol–EtOAc (2:1)], followed by
crystallisation from diethyl ether, afforded the title compound 6
as yellow crystals (230 mg, 56%). 1H NMR spectroscopic
studies using (S)-(ϩ)-N-(3,5-dinitrobenzoyl)-α-methylbenzyl-
amine as a chiral solvating agent showed the product (RS)-6 to
have an ee of 37%.
1
together with 50% unchanged starting material by H NMR
spectroscopy. Crystallisation from hexane–dichloromethane
(2:1) yielded (2S,RS)-15 as a yellow powder (86 mg, 15%).
Further crystallisation of the mother liquors afforded a second
crystalline fraction as orange rods (77 mg, 13%) shown to be a
1.3:1 mixture of 15 and 16 by 1H NMR spectroscopic analysis.
(2S,RS)-15 mp 167 ЊC (decomp.) (Found: C, 58.80; H, 3.83.
C24H20CrO6S requires C, 59.01; H, 4.13%); [α]D23 Ϫ43.0 (c 0.47,
CH3COCH3); νmax (CHCl3)/cmϪ1 1996, 1932 vs (C᎐O); δ (300
᎐
᎐
H
MHz) 7.68–7.66 (2H, m, Ph), 7.45–7.29 (8H, m, Ph), 5.49 (1H,
d, J 6.3, ArH), 5.44 (1H, t, J 6.2, ArH) 5.18–5.15 [1H, m,
3410
J. Chem. Soc., Perkin Trans. 1, 1999, 3405–3412

View Article Online
HOCH(CH3)], 5.10 (1H, d, J 6.6, ArH), 4.88 (1H, t, J 6.3,
ArH), 4.70 (1H, t, J 6.2, ArH), 3.71 (1H, br s, OH), 1.21 [3H, d,
J 6.2, HOCH(CH3)]; δC (125 MHz) 228.9 [Cr(CO)3], 135.9,
133.1 (Ph Ci), 132.2, 131.9, 129.0, 128.6, 127.9, 127.6 (Ph Co,
Cm, Cp), 101.4 (Ar C1), 95.1, 94.7, 94.6, 84.8, 84.7, (Ar C2–6),
obscured (CPh2), 69.7 [HOCH(CH3)], 19.2 (CH3); m/z (FABϩ)
489 (MHϩ, 46%), 405 (100).
using (S)-(ϩ)-N-(3,5-dinitrobenzoyl)-α-methylbenzylamine as
a chiral solvating agent showed the product 7 to have an ee of
3%.
Acknowledgements
We are grateful to the LINK Asymmetric Synthesis
Programme, to ZENECA Agrochemicals for a studentship
(to T. L.) and to the Portuguese Junta Nacional de Investi-
gacao Cientifica e Technologica (JINCT) for support (to
M. F. C. C. T.).
(2S,SS)-16 δH (300 MHz) 7.69 (2H, d, J 7.7, Ph Ho), 7.54 (2H,
d, J 7.6, Ph Ho), 7.44 (2H, t, J 7.6, Ph Hm), 7.36–7.30 (3H, m, Ph
Hm, Hp) 7.24 (1H, obs t, Ph Hp), 5.59 (1H, q, J 6.4, CH3CH),
5.41–5.37 (2H, m, ArH), 5.11 (1H, t, J 6.4, ArH), 4.93 (1H, t,
J 6.4, ArH), 4.46 (1H, d, J 6.5, ArH2 or H6), 2.81 (1H, s, OH),
1.43 (3H, d, J 6.4, CH3).
References
Treatment of (2S,RS)-2-hydroxy-1,1-diphenylpropyl(benzene)
chromium tricarbonyl sulfinate 15 with tert-butylmagnesium
chloride
1 (a) A. Solladié-Cavallo, G. Solladié and E. Tsamo, J. Org. Chem.,
1979, 44, 4189; (b) A. Alexakis, P. Mangeney, I. Marek, F. Rose-
Munch, E. Rose, A. Semra and F. Robert, J. Am. Chem. Soc., 1992,
114, 8288; (c) S. G. Davies and C. L. Goodfellow, J. Chem. Soc.,
Perkin Trans. 1, 1989, 192; L. A. Bromley, S. G. Davies and C. L.
Goodfellow, Tetrahedron: Asymmetry, 1991, 2, 139.
2 K. Nakamura, K. Ishihara, A. Ohno, M. Uemura, H. Nishimura
and Y. Hayashi, Tetrahedron Lett., 1990, 31, 3603; S. Top,
G. Jaouen, J. Gillois, C. Baldoli and S. Maiorana, J. Chem. Soc.,
Chem. Commun., 1988, 1284.
3 M. Uemura, T. Minami, K. Hirotsu and Y. Hayashi, J. Org. Chem.,
1989, 54, 469; M. Uemura, T. Kobayashi, T. Minami and
Y. Hayashi, Tetrahedron Lett., 1986, 27, 2479; M. Uemura,
T. Kobayashi, K. Isobe, T. Minami and Y. Hayashi, J. Org. Chem.,
1986, 51, 2859. See also ref. 16.
4 Y. Kondo, J. R. Green and J. Ho, J. Org. Chem., 1991, 56, 7199;
J. Aubé, J. A. Heppert, M. L. Milligan, M. J. Smith and P. Zenk,
J. Org. Chem., 1992, 57, 3563; Y. Kondo, J. R. Green and J. Ho,
J. Org. Chem., 1993, 58, 6182; A. Alexakis, T. Kanger, P. Mangeney,
F. Rose-Munch, A. Perrotey and E. Rose, Tetrahedron: Asymmetry,
1995, 6, 47; S. G. Davies and W. E. Hume, J. Chem. Soc., Chem
Commun., 1995, 251; J. W. Han, S. U. Son and Y. K. Chung, J. Org.
Chem., 1997, 62, 8264.
5 D. A. Price, N. S. Simpkins, A. M. MacLeod and A. P. Watt, J. Org.
Chem., 1994, 59, 1961; Tetrahedron Lett., 1994, 35, 6159; E. P.
Kundig and A. Quattropani, Tetrahedron Lett., 1994, 35, 3497;
M. Uemura, Y. Hayashi and Y. Hayashi, Tetrahedron: Asymmetry,
1994, 5, 1427; H. G. Schmalz and K. Schellhaas, Tetrahedron Lett.,
1995, 36, 5515; A. Alexakis, T. Kanger, P. Mangeney, F. Rose-
Munch, A. Perrotey and E. Rose, Tetrahedron: Asymmetry, 1995, 6,
2135; M. J. Siwet and J. R. Green, Chem. Commun., 1996, 2359;
R. A. Ewin, A. M. MacLeod, D. A. Price, N. S. Simpkins and
A. P. Watt, J. Chem. Soc., Perkin Trans. 1, 1997, 401;
A. Quattropani, G. Bernardinelli and E. P. Kunig, Helv. Chim. Acta,
1999, 82, 90.
6 A. Perez-Encabo, S. Perrio, A. M. Z. Slawin, S. E. Thomas, A. T.
Wierzchleyski and D. J. Williams, J. Chem. Soc., Chem. Commun.,
1993, 1059; J. Chem. Soc., Perkin Trans. 1, 1994, 629.
7 S. L. Griffiths, S. Perrio and S. E. Thomas, Tetrahedron: Asymmetry,
1994, 5, 545, 1847.
To a solution of (2S,RS)-15 (44 mg, 0.09 mmol) in THF (5 ml)
at ambient temperature was added tert-butylmagnesium
chloride (100 µl, 0.20 mmol) dropwise. The solution was stirred
for 14 h at room temperature before quenching of the reaction
with saturated aq. ammonium chloride (10 ml). The product
was extracted using ethyl acetate (2 × 10 ml), filtered through
a plug of magnesium sulfate and silica, and the solvent evap-
orated to yield a yellow crystalline solid tentatively assigned
as benzhydryl (phenyl) chromium tricarbonyl sulfone 17 (39 mg,
98%). A portion was recrystallised from hexane–dichloro-
methane (2:1) to give an analytically pure sample as fine yellow
needles, mp 186 ЊC (Found: C, 59.48; H, 3.42. C22H16CrO5S
requires C, 59.46; H, 3.63%); [α]D25 0.0 (c 0.47, CHCl3);
νmax (CHCl3)/cmϪ1 1996, 1937 vs (C᎐O), 1329, 1145 s (O᎐S᎐O);
᎐
᎐
᎐ ᎐
δH (300 MHz) 7.61–7.57 (4H, m, Ph), 7.39–7.36 (6H, m, Ph),
5.56 (2H, d, J 6.5, ArH2, H6), 5.47 (1H, t, J 6.2, ArH4), 5.33
(1H, s, Ph2CH), 4.94 (2H, t, J 6.4, ArH3, H5); δC (125 MHz)
228.9 [Cr(CO)3], 132.5 (Ph Ci), 130.1, 128.9 (Ph Co, Cm), 129.1
(Ph Cp), 101.9 (Ar C1), 94.1 (Ar C4), 93.8, 85.7 (Ar C2,6, C3,5),
76.7 (Ph2CH); m/z (CI, NH3) 462 (MNH4ϩ, 51%), 326 (20), 167
(100).
Decomplexation of benzhydryl (phenyl) chromium tricarbonyl
sulfone 17
A solution of 17 (5 mg, 0.011 mmol) in diethyl ether (5 ml) was
exposed to atmospheric oxygen and sunlight for 3 days. Fil-
tration through alumina (Grade V) and removal of the solvent
in vacuo gave a pale yellow solid. Crystallisation from ethyl
acetate–hexane (2:1) led to the isolation of white, needle-like
crystals (3 mg, 86%) shown to be benzhydryl phenyl sulfone by
comparison with an authentic sample; νmax (CHCl3)/cmϪ1 1309,
1148 s (O᎐S᎐O); δ (300 MHz) 7.63 (2H, d, J 7.8, Ph Ho), 7.56–
8 S. G. Davies, T. Loveridge and J. M. Clough, J. Chem. Soc., Chem.
Commun., 1995, 817.
᎐ ᎐
H
9 J.-H. Youn and R. Herrmann, Tetrahedron Lett., 1986, 27, 1493;
H. F. Herbrandson, R. T. Dickerson and J. Weinstein, J. Am. Chem.
Soc., 1956, 78, 2576; M. Mikolajcyzk and J. Drabowicz, Synthesis,
1974, 124.
7.31 (13H, m, Ph), 5.29 (1H, s, CHPh2); selected data for δC (125
ϩ
MHz; CD2Cl2) 76.5 (CHPh2); m/z (CI, NH3) 326 (MNH4
,
27%), 167 (100).
10 R. J. Card and W. S. Trahanovsky, J. Org. Chem., 1980, 45, 2555,
Attempted preparation of (SS)-[(tert-butylsulfinyl)benzene]
chromium tricarbonyl 7 from 15/16
2560.
11 Isolation of this material was achieved by washing the silica frit with
1
methanol. However, H NMR spectroscopic analysis resulted in a
To a solution of 15/16 (33 mg, 0.068 mmol, 1.3:1) in THF
(2 ml) at room temperature was added tert-butylmagnesium
chloride (80 µl, 0.16 mmol) dropwise. The reaction mixture was
stirred for 45 min before quenching by the addition of saturated
aq. ammonium chloride (5 ml). The product was extracted with
ethyl acetate (2 × 5 ml), filtered through a plug of magnesium
sulfate and silica, and the solvent evaporated to give a yellow
complex spectrum with no identifiable peaks.
12 D. A. Evans, M. M. Faul, L. Colombo, J. J. Bisaha, J. Clardy and
D. Cherry, J. Am. Chem. Soc., 1992, 114, 5977.
13 M. Deshmukh, E. Dunach, S. Juge and H. B. Kagan, Tetrahedron
Lett., 1984, 25, 3467; 1985, 26, 402.
14 On one occasion three recrystallisations of “(SS)”-5 (ee 67%) from
hexane–toluene (2:1) furnished mother liquors with an enhanced ee
of 91%.
15 F. Rebière, O. Samuel, L. Ricard and H. B. Kagan, J. Org. Chem.,
1991, 56, 5991.
16 L. E. Orgel, Inorg. Chem., 1962, 1, 25.
17 F. G. Bordwell and B. M. Pitt, J. Am. Chem. Soc., 1955, 77, 572;
B. M. Trost and D. P. Curran, Tetrahedron Lett., 1981, 22, 1287;
C. Y. Meyers, W. S. Matthews, G. J. McCollum and J. C. Branca,
Tetrahedron Lett., 1974, 1105.
1
oil. Analysis of the H NMR spectrum of the crude material
indicated the presence of 7 and 17 in the ratio 1:2.7. Column
chromatography on silica gel [petrol–EtOAc (1.5:1)] led to the
isolation of 2 fractions; the first a yellow solid comprising 17
(7 mg, 24%). The second fraction afforded the title compound 7
as a yellow oil (2 mg, 9%). 1H NMR spectroscopic studies
J. Chem. Soc., Perkin Trans. 1, 1999, 3405–3412
3411

View Article Online
18 For a review of work published in this area see: S. Braverman, in The
Chemistry of Sulfoxides and Sulfones, ed. S. Patai, Z. Rappoport and
C. J. M. Stirling, Wiley and Sons, Chichester, 1988, p. 665.
19 D. Darwish and R. McLaren, Tetrahedron Lett., 1962, 1231.
20 E. Ciuffarin, M. Isola and A. Fava, J. Am. Chem. Soc., 1968, 90,
3594.
22 W. C. Still, M. Kahn and A. Mitra, J. Org. Chem., 1978, 43,
2923.
23 C. A. L. Mahaffy and P. L. Pauson, Inorg. Synth., 1979, 19,
154.
24 K. K. Andersen, Tetrahedron Lett., 1962, 93.
21 D. F. Shriver and M. A. Dreszdon, The Manipulation of Air
Sensitive Compounds, Wiley, New York, 1986, 2nd edn.
Paper 9/07590I
3412
J. Chem. Soc., Perkin Trans. 1, 1999, 3405–3412