A straightforward enantioselective route to dialkyl sulfoxides based upon two carbon-for-carbon substitution reactions on the sulfinyl group

Article abstract of DOI:10.1021/jo025974j

Benzyl p-bromophenyl sulfoxide 1 was obtained on a multigram scale and in an enantiomerically pure form by an enantioselective catalytic oxidation, using tert-butyl hydroperoxide in the presence of chiral titanium complexes. Some mechanistic and stereochemical features of interest were studied in this process. Compound 1 was then subjected to two different substitution reactions with Grignard reagents, which caused two sequentially stereocontrolled carbon-for-carbon displacements, leading to chiral nonracemic dialkyl sulfoxides.

Full text of DOI:10.1021/jo025974j

A Str a igh tfor w a r d En a n tioselective Rou te to Dia lk yl Su lfoxid es
Ba sed u p on Tw o Ca r bon -for -Ca r bon Su bstitu tion Rea ction s on th e
Su lfin yl Gr ou p
Maria Annunziata M. Capozzi,†,‡ Cosimo Cardellicchio,§ Francesco Naso,*,†,§ and
Valerio Rosito†
Consiglio Nazionale delle Ricerche, Istituto di Chimica dei Composti OrganoMetallici, Sezione di Bari
and Dipartimento di Chimica, Universita` di Bari, via Orabona 4, 70126 Bari, Italy
naso@area.ba.cnr.it
Received May 27, 2002
Benzyl p-bromophenyl sulfoxide 1 was obtained on a multigram scale and in an enantiomerically
pure form by an enantioselective catalytic oxidation, using tert-butyl hydroperoxide in the presence
of chiral titanium complexes. Some mechanistic and stereochemical features of interest were studied
in this process. Compound 1 was then subjected to two different substitution reactions with Grignard
reagents, which caused two sequentially stereocontrolled carbon-for-carbon displacements, leading
to chiral nonracemic dialkyl sulfoxides.
SCHEME 1
Chiral nonracemic sulfoxides are valuable intermedi-
ates1 and have relevant applications in asymmetric
synthesis.1,2 In our recent work,3-5 we have reported a
synthetic useful route to sulfoxides in high enantiomeric
purity by using the stereocontrolled reaction between
Grignard reagents and suitable sulfinyl compounds, with
the displacement of a carbanionic moiety. This approach
was especially convenient for the preparation of dialkyl
sulfoxides, a class of compounds for which a general and
straightforward route appeared particularly needed. Our
work was characterized by two significant moments: in
a first approach,3,4 we synthesized aryl alkyl or dialkyl
sulfoxides by a two-step procedure involving the prepara-
tion of the chiral sulfinyl precursor by an enantioselective
oxidation, followed by the synthesis of the final sulfoxide
through a Grignard reagent promoted stereocontrolled
substitution of a carbanionic leaving group (LG) (Scheme
1).
SCHEME 2
In a second approach,5 we obtained various chiral
nonracemic dialkyl sulfoxides by a two-displacement
sequence starting from menthyl (S)-p-bromobenzene-
sulfinate. In particular, alkyl Grignard reagents pro-
moted the stereocontrolled displacement of the menthox-
ide ion in the first step, which was then followed by the
displacement of the p-bromophenyl leaving group on the
resulting p-bromophenyl sulfoxide.
The procedure of Scheme 2 represents a useful syn-
thetic route to dialkyl sulfoxides based upon a carbon-
for-oxygen substitution, followed by a carbon-for-carbon
substitution. In principle, since the sulfinate was ob-
tained through a diastereomeric separation by crystal-
lization, it appeared that the importance of our procedure
could be significantly increased by finding a suitable
starting compound, accessible by a direct enantioselective
oxidation and susceptible to transformation into the final
sulfoxide by two consecutive displacements of carbanionic
leaving groups (Scheme 3).
† Dipartimento di Chimica, Universita` di Bari.
‡ Present address: Istituto di “Produzioni e Preparazioni Alimen-
tari”, Facolta` di Agraria, Universita` degli Studi di Foggia, via Napoli
25, 71100 Foggia.
§ Consiglio Nazionale delle Ricerche, Istituto di Chimica dei Com-
posti OrganoMetallici.
(1) Drabowicz, J .; Kiełbasinski, P.; Mikołajczyk, M. Synthesis of
Sulphoxides. In The Chemistry of Sulphones and Sulphoxides; Patai,
S., Rappoport, Z., Stirling, C., Eds.; J ohn Wiley and Sons: New York,
1988; pp 233-378. Walker, A. J . Tetrahedron: Asymmetry 1992, 3,
961-998.
(2) Carren˜o, M. C. Chem. Rev. 1995, 95, 1717-1760.
(3) (a) Cardellicchio, C.; Fracchiolla, G.; Naso, F.; Tortorella, P.
Tetrahedron 1999, 55, 525-532. (b) Capozzi, M. A. M.; Cardellicchio,
C.; Fracchiolla, G.; Naso, F.; Tortorella, P. J . Am. Chem. Soc. 1999,
121, 1, 4708-4709.
(4) Capozzi, M. A. M.; Cardellicchio, C.; Naso, F.; Tortorella, P. J .
Org. Chem. 2000, 65, 2843-2846.
(5) Capozzi, M. A. M.; Cardellicchio, C.; Naso, F.; Spina, G.;
Tortorella, P. J . Org. Chem. 2001, 66, 5933-5936.
Since the first carbanionic leaving group was already
at hand (i.e., the p-bromophenyl group), the attention had
to be focused on the second one, which had to show either
10.1021/jo025974j CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/18/2002
J . Org. Chem. 2002, 67, 7289-7294
7289

Capozzi et al.
SCHEME 3
substrate especially suitable for obtaining dialkyl sul-
foxides by a two-step substitution procedure. At this point
we had to face the production of 1 through an enantio-
selective oxidation of the benzyl p-bromophenyl sulfide.
Oxid a tion of Ben zyl p-Br om op h en yl Su lfid e. Syn -
th etic, Mech a n istic, a n d Ster eoch em ica l Asp ects.
Among the various types of enantioselective oxidation of
sulfides,8 we focused our attention on the reaction with
hydroperoxides in the presence of a catalytic amount of
chiral titanium complexes, a procedure originally intro-
duced by Modena et al.9 and by Kagan et al.10 Summing
up the information now available on this process, we can
distinguish two main classes of reactions:
SCHEME 4
(A) The sulfoxide is produced in high enantiomeric
purity with the formation of only low amounts of the
corresponding sulfone. Examples of these truly enantio-
selective oxidations were described by Modena,9 Kagan,10
Bolm,11 and their co-workers, and by us.3b
(B) High amounts of sulfone are produced in the
oxidation process. Examples of these processes were
reported by Uemura12 and Imamoto.13 The oxidation with
a chiral hydroperoxide, reported by Adam,14 was also
accompanied by a large quantity of sulfone. In these
papers, it was proved that a kinetic resolution process
occurred during the oxidation of the sulfoxide to sulfone,
with the enrichment of the enantiomeric purity of the
remaining sulfoxide. High levels of sulfone were required
in order to recover a sulfoxide with high enantiomeric
purity. As a consequence of this overoxidation, the
isolated yield of sulfoxide could not be high.
Keeping in mind this mechanistic dichotomy, we
undertook the series of oxidation experiments reported
in Table 1, with the aim of obtaining the sulfoxide 1,
possibly through a reaction following the type-A mech-
anism. At the outset, we oxidized benzyl p-bromophenyl
sulfide with tert-butyl hydroperoxide (TBHP) in the
presence of a complex between Ti(O-i-Pr)4, water and (R)-
1,1′-bi-2-naphthol (BINOL) as a chiral ligand (Table 1,
entries 1-4), at room temperature according to our
previously reported procedure,3b which works particularly
well in the oxidation of (alkylthio)- or (arylthio)meth-
ylphosphonates. In contrast with the results obtained
with these substrates, after 29 h, we obtained the benzyl
p-bromophenyl sulfoxide 1 in high ee (up to 95%), but in
low isolated yield (34%), due to the formation of a large
amount of sulfone (Table 1, entry 1). On the other hand,
decreasing the reaction time led to low amounts of
sulfone, but the corresponding sulfoxide was produced
in low ee (Table 1, entry 2). Furthermore, we found a
depletion of ee upon changing the solvent from carbon
a higher or a lower leaving group ability in respect to
the p-bromophenyl moiety. The difficulty of finding a
second carbanionic leaving group with this peculiar
feature was coupled with the need of obtaining the
suitable starting sulfoxide, once found, by an enantiose-
lective oxidation.
Now, we wish to report our successful experimental
execution of this synthetic plan together with several
relevant aspects which emerged during our efforts for the
production of the appropriate starting material with
special reference to the enantioselective oxidation step.
Resu lts a n d Discu ssion
Th e Sea r ch for th e Su bstr a te. Upon scanning the
literature background on possible candidates as carban-
ionic leaving groups for a carbon-for-carbon substitution,
it appeared to us that a carbanionic leaving group worth
testing was the benzyl moiety. Indeed, a ligand exchange
process was observed by Durst et al.6 when n- or tert-
butyllithium was reacted with benzyl phenyl sulfoxide
to give n- or tert-butyl phenyl sulfoxide (yield 40-50%).
The stereochemical course of the benzyl group displace-
ment was not reported. Another relevant result derived
from our previous investigations3-5 was represented by
the observation that synthetically useful ligand exchange
processes, as a general rule, required Grignard reagents,
instead of organolithium compounds. Under these par-
ticular conditions, the reactions could be driven toward
a complete displacement of a carbanionic leaving group,
with limited amounts of side products and full stereo-
chemical control.
With this background, we assumed that benzyl p-
bromophenyl sulfoxide 17 could be a feasible substrate
on which two different carbanionic leaving groups could
be sequentially substituted. Indeed, in preliminary ex-
periments, we observed that the reaction of 1 with 1.5
equiv of alkyl Grignard reagent in THF at room temper-
ature caused the displacement of only the benzyl group,
yielding the alkyl p-bromophenyl sulfoxide, without the
displacement of the p-bromophenyl moiety (Scheme 4).
Since we had already shown that alkyl p-bromophenyl
sulfoxides could be also subjected to a stereocontrolled
reaction with alkyl Grignard reagents to displace the
p-bromophenyl moiety,4,5 we inferred that 1 could be a
(8) Bolm, C.; Mun˜iz, K.; Hildebrand, J . P. Oxidations of Sulfides.
In Comprehensive Asymmetric Catalysis; J acobsen, E. N., Pfalz, A.,
Yamamoto, H., Eds.; Springer: Berlin, 1999; pp 697-710.
(9) Bortolini, O.; Di Furia, F.; Licini, G.; Modena, G.; Rossi, M.
Tetrahedron Lett. 1986, 27, 6257-6260. Di Furia, F.; Modena, G.;
Seraglia, R. Synthesis 1984, 325-327.
(10) Brunel, J . M.; Kagan, H. B. Synlett 1996, 404-406. Zhao, S.
H.; Samuel, O.; Kagan, H. B. Tetrahedron 1987, 43, 5135-5144.
Pitchen, P.; Dun˜ach, E.; Deshmukh, M. N.; Kagan, H. B. J . Am. Chem.
Soc. 1984, 106, 8188-8193.
(11) Bolm, C.; Dabard, O. A. G. Synlett 1999, 360-362.
(12) Komatsu, N.; Hashizume, M.; Sugita, T.; Uemura, S. J . Org.
Chem. 1993, 58, 7624-7626. Komatsu, N.; Hashizume, M.; Sugita, T.;
Uemura, S. J . Org. Chem. 1993, 58, 4529-4533.
(13) Yamanoi, Y.; Imamoto, T. J . Org. Chem. 1997, 62, 8560-8564.
(14) Adam, W.; Korb, M. N.; Roschmann, K. J .; Saha-Mo¨ller, C. R.
J . Org. Chem. 1998, 63, 3423-3428.
(6) Durst, T.; LeBelle, M. J .; Van Der Elzen, R.; Tin, K.-C. Can. J .
Chem. 1974, 52, 761-766.
(7) Degani, J .; Tiecco, M.; Tundo, A. Gazz. Chim. Ital. 1962, 92,
1213-1220.
7290 J . Org. Chem., Vol. 67, No. 21, 2002

Enantioselective Route to Dialkyl Sulfoxides
TABLE 1. En a n tioselective Oxid a tion of Ben zyl p-Br om op h en yl Su lfid e by Hyd r op er oxid es in th e P r esen ce of Ch ir a l
Tita n iu m Ca ta lysts
entry
ligand
(R)-BINOL
(R)-BINOL
(R)-BINOL
(R)-BINOL
(R,R)-DET
(S,S)-HB
(S,S)-HB
(S,S)-HB
(S,S)-HB
(S,S)-HB
(S,S)-HB
watera
solvent
oxidant
time (h)
yieldb (%)
eec (%) (config)
1
2
3
0.5
0.5
0.5
0
CCl4
CCl4
CH2Cl2
CCl4
CH2Cl2
CCl4
CH2Cl2
toluene
n-hexane
CCl4
n-hexane
n-hexane
n-hexane
n-hexane
n-hexane
n-hexane
n-hexane
TBHP
TBHP
TBHP
TBHP
CHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
TBHP
29
6
34
42
33
69
46
63
76
74
81e
77
85f
84
63
71
66
45g
51h
95 (R)
46 (R)
47 (R)
0
24
22
22
13
22
84
72
17
48
48
48
49
48
48
72
4
5d
6
0
58 (R)
96 (R)
73 (R)
89 (R)
>98 (R)
>98 (R)
>98 (R)
>98 (S)
88 (R)
69 (R)
48 (R)
- -
0.5
0.5
0.5
0.5
0
0
0
0
0
7
8
9
10
11
12
13
14
15
16
17
(R,R)-HB
5 (S,S)-HB + 1 (R,R)-HB
3 (S,S)-HB + 1 (R,R)-HB
2 (S,S)-HB + 1 (R,R)-HB
(R,S)-HB
0
0
0
1 (S,S)-HB + 1 (R,S)-HB
58 (R)
a
b
d
Water/sulfide molar ratio. Yields refer to pure isolated sulfoxide 1. c Determined by HPLC (see text). Reagents ratio: Ti(O-i-Pr)4:
DET: i-Pr2NEt: substrate: oxidant 3:6:3:10:9.5. Reaction temperature for this entry -20 °C. e Isolated sulfone 7%. f Isolated sulfone 5%.
Isolated sulfone 31%. Isolated sulfone 28%.
g
h
tetrachloride to methylene chloride (Table 1, entry 3), and
a racemic sulfoxide was obtained when no water was
added (Table 1, entry 4). All these aspects showed a close
similarity with the results reported by others12 for a
type-B enantioselective oxidation with the same chiral
ligand, even if different reaction conditions were used.
Thus, considering all the results obtained with BINOL
as a chiral ligand, only the experiment of entry 1
appeared to be of some synthetic interest, which in-
creased significantly when we found that the product 1
having a 95% ee could be easily recrystallized to yield
the enantiomerically pure compound.
The absolute configuration of 1 was determined with
NMR techniques, by adding (R)-(methoxy)phenylacetic
acid, and analyzing the pattern of the methylene signals
of an enantiomerically enriched mixture.5 The configu-
ration of the enantiomer, which was obtained in the
reaction with the (R)-BINOL as a chiral ligand, was found
to be (R).
Then, we investigated the oxidation of the same sulfide
with cumene hydroperoxide (CHP) in the presence of a
complex between Ti(O-i-Pr)4 and (+)-diethyl tartrate
(DET) in several reaction conditions. The best results
(Table 1, entry 5, 46% yield, 58% ee) were obtained in
conditions similar to those reported in a recent paper.15
Ti(O-i-Pr)4, DET and the sulfide were heated to 40 °C
and then the mixture was cooled to -20 °C and i-Pr2-
NEt and CHP were added.
The subsequent chiral ligand tested in our investiga-
tion was represented by (S,S)- or (R,R)-1,2-diphenyl-1,2-
ethanediol (stilbene diol, or hydrobenzoin, HB). This
compound, which could be obtained in high enantiomeric
purity by an enantioselective dihydroxylation,16 is cur-
rently commercially available. Diols structurally related
to HB had been used as ligands in a hydroperoxide
oxidation of sulfide,17 catalyzed by a stoichiometric
amount of a titanium complex (55-81% yields, 43-84%
ee). Recently, Rosini et al.18 have reported an enantio-
selective TBHP-oxidation of aryl alkyl sulfides in the
presence of 5% of a 1:2 complex between Ti(O-i-Pr)4 and
HB (55-74% yields and 22f99% ee). The reaction was
performed at 0 °C, in CCl4 and in the presence of a large
amount of water (1:1 molar ratio with the substrate). A
2:1 molar ratio of oxidant/sulfide was added, and the
reaction was quenched after 2 h, to prevent overoxidation
of both the sulfide and the chiral ligand. This system was
used also to oxidize benzyl aryl sulfide (aryl ) phenyl,
p-tolyl, p-anisyl, 60-73% yields; 92f99% ee).
In our experiments, we used a 2.5% molar ratio of a
1:2 catalyst prepared from Ti(O-i-Pr)4 and HB, in the
presence of water. The substrate was in a 2:1 molar ratio
in respect to water and 1.1 to 1 molar ratio in respect to
the oxidant. After 13 h at room temperature, using CCl4
as solvent, 1 was obtained in moderate yields and in high
enantiomeric excess (96% ee, Table 1, entry 6).
Lower enantiomeric purities (73-89% ee) were ob-
tained when toluene or methylene chloride were em-
ployed as solvents (Table 1, entries 7 and 8), whereas
the use of n-hexane gave the best results (Table 1, entry
9). During the reaction performed in this solvent, a white
crystalline solid was observed to separate. After 3 days
at room temperature, the white solid, collected and
analyzed, was found to be enantiopure 1, with a small
(16) McKee, B. H.; Gilheany, D. G.; Sharpless, K. B. Org. Synth.
1991, 70, 47-53.
(17) Yamamoto, K.; Ando, H.; Shuetake, T.; Chikamatsu, H. J .
Chem. Soc., Chem. Commun. 1989, 754-755.
(18) Donnoli, M. I.; Superchi, S.; Rosini, C. J . Org. Chem. 1998, 63,
9392-9395.
(15) Cotton, H.; Elebring, T.; Larsson, M.; Li, L.; Sorensen. H.; von
Unge, S. Tetrahedron: Asymmetry 2000, 11, 3819-3825.
J . Org. Chem, Vol. 67, No. 21, 2002 7291

Capozzi et al.
amount of sulfone. A simple recrystallization (ethanol)
yielded the desired compound in high yield (81%). Also
for these reactions, when the (S,S)-HB was used as a
chiral ligand, the absolute configuration of 1 was found
to be (R). Surprisingly, the addition of water, whose role
was crucial in almost every oxidation procedures of
sulfides,3b,9-12 did not affect the enantioselectivity of our
reaction with HB as a chiral ligand. Indeed, similar high
enantiomeric purities were obtained both in the presence
or in the absence of water (Table 1, entries 6, 9-11).
Since Rosini et al.18 had demonstrated the need for water
in the oxidation of methyl p-tolyl sulfide in the presence
of a Ti(O-i-Pr)4/HB complex, a different mechanism
should be considered operating in the oxidation leading
to 1. The reaction in n-hexane was also used for the
oxidation of 25 g of benzyl p-bromophenyl sulfide (see
Experimental Section). A 24.06 g amount of white solid
precipitated from the reaction mixture. After a recrys-
tallization (ethanol), 22.5 g (85% yield) of enantiopure 1
could be obtained. No particular difficulty should arise
performing the oxidation on a larger scale.
effect on the chemical shift of the above signals. This
result could suggest a catalytic species with the HB
directly bound to the metallic center and without any
significant participation of water molecules.
From a mechanistic point of view, nonlinear effects
(NLEs) connecting the enantiomeric purity of the chiral
ligands and the enantiomeric purity of the products have
been also considered important.20 For example, the
oxidation with formation of high amounts of sulfone was
characterized by a positive NLE,12 i.e. asymmetric am-
plification, whereas Modena oxidation had a negative
NLE,20 i.e. an asymmetric depletion. On the other hand,
the Kagan procedures showed mainly a negative or a
negligible NLE, but also cases of slightly positive values
were observed, due to the different ways of producing the
catalyst.20
In the TBHP-oxidation of benzyl p-bromophenyl sulfide
in the presence of the complex Ti(O-i-Pr)4/HB, we found
a positive NLE (Table 1, entries 13-15) after 2 days.
Indeed, HB of lower enantiomeric purity could drive the
formation of a sulfoxide toward higher values of enan-
tiomeric excess.
Different results were found by using meso-HB, either
pure or mixed with (S,S)-HB. When the reaction was
performed only with meso-HB (Table 1, entry 16), large
amounts of sulfone were produced after 2 days (31%) and
racemic 1 was isolated in a 45% yield. When a 1:1
mixture of (S,S)-HB and (R,S)-HB was used, a sulfoxide
of 58% ee was obtained, once again with a large amount
of sulfone (isolated yield 51% for the sulfoxide 1 and 28%
for the corresponding sulfone, Table 1, entry 17). In
contrast with the results of entry 11, in these conditions,
the ee of the sulfoxide was found to increase during the
reaction. The values observed were 8% after 3 h and 17%
after 6 h. The final value (51%) was measured after 3
days. The high amount of sulfone and the evolution of
the ee during the reaction time suggest that a kinetic
resolution step is also operating as a consequence of the
addition of the meso-ligand.
On the basis of the results presented above, we can
conclude that using (S,S)- or (R,R)-HB as a chiral ligand,
in conditions largely different from those reported by
others,18 high ee values and high yields can be obtained
in the production of 1. Furthermore, the mechanistically
important conclusion that can be reached is that the
enantioselection is only due to the sulfoxide oxidation
step, without any significant contribution of the kinetic
resolution of the resulting sulfoxide. However, the latter
process appears always impending, as shown by the
significant changes observed upon addition of meso-HB.
Undoubtedly, further work is required to better clarify
the various factors influencing the two mechanisms,
including the role of the water and the significance of
the sign of the NLE. However, due to the main aim of
the present work, i.e. the synthesis of dialkyl sulfoxides
from a general precursor, for the moment it was not
considered of interest to expand further the mechanistic
study.
The oxidation procedure was tested also with (R,R)-
HB (entry 12). Also in this case, a white solid precipi-
tated, which, after recrystallization, yielded (S)-1 having
the usual high ee values (>98%).
We are inclined to consider that this process is a type-A
oxidation. Such a conclusion is based first of all upon the
fact that the amount of sulfone produced was rather
small (5-7%). Furthermore, when the white solid ob-
tained during the reaction performed in n-hexane was
analyzed at the early stages of oxidation (4 h), it was
found to present already a high enantiomeric purity (ee
93%). This observation suggests that mainly the direct
oxidation of the sulfide is the responsible for the high
degree of enantioselectivity, whereas the kinetic resolu-
tion in the subsequent oxidation of the sulfoxide to
sulfone, if any, plays a minor role.
In an attempt to gain further insight into the mecha-
nism, the formation of the titanium complexes in the
enantioselective oxidation reactions was also monitored
by 1H NMR. In this respect, it is worth noting that, when
the spectra of the catalyst used in the Kagan-Modena
oxidation had been recorded,19 several signals were
detected, which could be associated to a large number of
different titanium/diethyl tartrate species in solution.
Our NMR analysis was conducted by performing the
reaction in CCl4. A sample of the reaction mixture was
taken, C6D6 (0.1 mL) was added as an internal reference,
and 1H NMR spectra of the resulting mixture were
recorded. When the Ti(O-i-Pr)4/BINOL catalyst was used,
the NMR spectra revealed the presence of a dominant
complex among the others, but some signals had a
different chemical shift depending upon the presence or
the absence of water. These spectra could suggest that
different catalytic species were operating in the two cases.
On the other hand, when the complex between Ti(O-i-
Pr)4 and HB was used, the spectra were simple and
sharp. The signals of the isopropoxy moiety shifted
upfield, as in free 2-propanol, whereas the HB signals
moved downfield with respect to the signals of the free
HB. The presence or absence of water had no significant
Rea ction betw een Ben zyl p-Br om op h en yl Su lfox-
id e (1) a n d Gr ign a r d Rea gen ts. The substitution of
the benzyl group in the benzyl p-bromophenyl sulfoxide
(19) Potvin, P. G.; Fieldhouse, B. G. Tetrahedron: Asymmetry 1999,
10, 1661-1672.
(20) Brunel, J .-M.; Luukas, T. O.; Kagan, H. B. Tetrahedron:
Asymmetry 1998, 9, 1941-1946.
7292 J . Org. Chem., Vol. 67, No. 21, 2002

Enantioselective Route to Dialkyl Sulfoxides
TABLE 2. Ster eosp ecific Su bstitu tion of Ca r ba n ion ic Lea vin g Gr ou p s by Gr ign a r d Rea gen ts on Su lfin yl Com p ou n d s
entry
substrate
R1
R2
solventa
T (°C)
productb,c
yield (%)d
ee (%)e
1
2
3
4
5
6
7
8
9
1 (R)
1 (R)
1 (R)
1 (R)
1 (R)
1 (R)
3 (S)
4 (S)
4 (S)
n-dodecyl
n-pentyl
ethyl
n-pentyl
n-dodecyl
cyclohexyl
n-pentyl
n-dodecyl
n-dodecyl
THF
-30
5
4 (S)c
3 (S)b
2 (S)b
3 (S)b
4 (S)c
5 (S)b
6 (R)c
6 (S)c
7 (R)c
53
62
67
69
87
70
62
69
91
87
95
benzene
toluene
toluene
toluene
toluene
THF
-30
-30
-30
-30
-30
-30
-30
>98
>98
>98
>98
>98
>98
>98
n-dodecyl
n-pentyl
i-propyl
THF
THF
a
b
Solvent for the substrate. Grignard reagents were prepared in THF and used in a 1.5:1 molar ratio in respect to the substrate. The
configuration was established by correlation with the known configuration of sulfoxides obtained in previous work. c Configuration attributed
assuming a stereochemical outcome of inversion of configuration as found in similar cases (see text). Isolated yield. e Determined by
d
HPLC (see text).
1 by 1.5 equiv of n-dodecylmagnesium bromide in THF
was found to occur with an 87% ee, even at low temper-
ature (Table 2, entry 1). Better results were obtained in
the reaction of the n-pentylmagnesium bromide in ben-
zene (Table 2, entry 2). A complete stereochemical control
of this carbon-for-carbon displacement was achieved
performing the reaction in toluene, at -30 °C (Table 2,
entry 3-6). The sulfoxides obtained from (R)-1 had
negative optical rotations. The absolute configurations
of the sulfoxides (-)-2 and (-)-5 were then found to be
(S), since (+)-2 and (+)-5 are known to have the (R)-
configuration.5 It is also worth noting that sulfoxide (+)-3
was obtained21 in the reaction between menthyl (S)-p-
bromobenzenesulfinate and pentylmagnesium bromide,
a reaction which we have shown to occur with inversion
of configuration.5 Consequently, the (R)-configuration
could be attributed to sulfoxide (+)-3 and the (S)-
configuration to the counterpart (-)-3. The overall pat-
tern observed is then consistent with the expected
inversion of configuration in the displacement of the
benzyl moiety. On the basis of such a pattern, it seems
correct to attribute the (S)-configuration to the sulfoxide
(-)-4.
Rea ction s betw een Alk yl p-Br om op h en yl Su lfox-
id es a n d Gr ign a r d Rea gen ts. The displacement of the
bromobenzene anion moiety with Grignard reagents was
already described in a previous work.4,5 However, for the
sake of completeness and also with the aim of presenting
here the full execution of the synthetic plan, in Table 2
we have added three further cases involving a primary
and a secondary alkyl group, leading to 6 and 7, respec-
tively (Table 2, entries 7-9), to which the configuration
was assigned on the basis of previous work.5 Both
enantiomers of 6 were obtained by choosing the suitable
sequence of substitution (Table 2, entries 7 and 8).
of dialkyl sulfoxides through a two carbon-for-carbon
substitution sequence. The starting compound can be
easily obtained in an optically pure form and on a
multigram scale by an enantioselective oxidation followed
by a simple crystallization. The method compares very
favorably with other procedures since it offers the
advantage of the high enantioselectivity of the steps upon
which it is based without presenting any problem con-
nected with separations of stereoisomers. A parallel
outcome of our investigation is represented by a series
of observations which are of relevance for the mechanistic
picture of the oxidation process occurring in the first step
of the procedure.
Exp er im en ta l Section
The purified reaction products were characterized by their
1H and 13C NMR spectra, recorded in CDCl3 at 500 and 125
MHz, respectively, and their mass spectra were determined
by GC/MS analysis (70 eV), when they did not decompose. The
ee values of the sulfoxides were determined by HPLC (Chiral-
cel OB-H or OD-H).
Syn th esis of Ben zyl p-Br om op h en yl Su lfoxid e (1) by
E n a n t ioselect ive Oxid a t ion of Ben zyl p -Br om op h en yl
Su lfid e w ith ter t-Bu tyl Hyd r op er oxid e in th e P r esen ce
of a Ti Ca ta lyst (gen er a l p r oced u r e). A solution of Ti(O-
i-Pr)4 (7.7 mg, 0.027 mmol) in 4 mL of the specified solvent
was added to a solution of the ligand (0.054 mmol) in 8 mL of
the solvent under a nitrogen atmosphere. If present, water
(0.54 mmol) was added at this stage. The mixture was stirred
for 1 h at room temperature. Benzyl p-bromophenyl sulfide
(0.3 g, 1.07 mmol) was then added, and the mixture was stirred
for 30 min. After this time, 0.15 mL of a commercial 80%
solution of tert-butyl hydroperoxide (1.18 mmol) was added,
and the stirring was continued for the time specified in the
Table. Then, the solvent was removed in vacuo, and the
evaporated mixture was subjected to column chromatography
(eluent ethyl acetate/petroleum ether 3:7) to yield benzyl
p-bromophenyl sulfoxide 1.
(R)-Ben zyl p-br om op h en yl su lfoxid e (1): mp 170-172
°C (ethanol) (lit.7 mp 141-142 for racemic 1). [R]D ) +79.2 (c
) 1, CHCl3). (S)-Ben zyl p-br om op h en yl su lfoxid e 1: mp
170-172 °C (ethanol). [R]D ) -80.0 (c ) 1.6, CHCl3). The ee
value was measured by HPLC (Chiralcel OD-H, hexane/2-
propanol 90:10).
Con clu sion
In the present work, we have shown that benzyl
p-bromophenyl sulfoxide can act as a general precursor
(21) Nishide, K.; Nakayama, A.; Kusumoto, T.; Hiyama, T.; Take-
hara, S.; Shoji, T.; Osawa, M.; Kuriyama, T.; Nakamura, K.; Fujisawa,
T. Chem. Lett. 1990, 623-626.
La r ge Sca le P r ep a r a t ion of (R)-Ben zyl p -Br om o-
p h en yl Su lfoxid e (1). A solution of Ti(O-i-Pr)4 (0.64 g, 2.25
J . Org. Chem, Vol. 67, No. 21, 2002 7293

Capozzi et al.
mmol) in 30 mL of n-hexane was added to a solution of (S,S)-
hydrobenzoin (0.963 g, 4.5 mmol) in 130 mL of n-hexane. The
mixture was stirred for 1 h at room temperature, under a
nitrogen atmosphere. Benzyl p-bromophenyl sulfide (25 g,
89.54 mmol) was then added, and the stirring was continued
for 30 min. The mixture was diluted with additional 230 mL
of n-hexane. At this stage, 12.5 mL of a commercial 80%
solution of tert-butyl hydroperoxide was added. The mixture
was stirred at room temperature for 48 h, and after this time,
it was subjected to centrifugation to recover 24.06 g of a
precipitated white solid. The solid collected by centrifugation
was crystallized (ethanol), obtaining 22.5 g of pure 1 (85%
yield, >98% ee).
Rea ction of Alk yl Gr ign a r d Rea gen ts w ith (R)-Ben zyl
p-Br om op h en yl Su lfoxid e (1). A solution of 6.3 mmol of
Grignard reagent in THF was dropped into a solution of 4.2
mmol of 1 in 25 mL of toluene at -30 °C and under N2. After
1.5 h, the reaction mixture was quenched with a saturated
solution of NH4Cl. The usual workup yielded a residue which
was purified by column chromatography and crystallization
(hexane).
(S)-p-Br om op h en yl eth yl su lfoxid e (2): mp 53-54 °C
(hexane) (lit.5 53-55 °C). [R]D ) -162.1 (c ) 1, CHCl3) (lit.5
[R]D ) +162.0 (c ) 1, CHCl3) for the (R)-configuration). The
ee value, measured by HPLC (Chiralcel OB-H, hexane/2-
propanol 70:30), was >98%.
(S)-p-Br om op h en yl n -p en tyl su lfoxid e (3): mp 46-48 °C
(hexane) (lit.21 44-45 °C). [R]D ) -169.4 (c ) 1, CHCl3) and
[R]D ) -152.7 (c ) 1, CH3OH) (lit.21 [R]D ) +148 (c ) 1.01,
CH3OH) for a sulfoxide having the (R)-configuration and 98%
ee, see text). The ee value, measured by HPLC (Chiralcel OB-
H, hexane/2-propanol 90:10), was >98%.
(S)-p-Br om op h en yl cycloh exyl su lfoxid e (5): mp )
119-121 °C (hexane) (lit.5 119-120 °C). [R]D ) -158.4 (c ) 1,
CHCl3) (lit.5 [R]D ) +156.3 (c ) 1, CHCl3) for the (R)-
configuration). The ee value, measured by HPLC (Chiralcel
OB-H, hexane/2-propanol 80:20), was >98%.
The reactions of alkyl Grignard reagents with p-bromo-
phenyl n-dodecyl sulfoxide 4 were performed according to our
previous work.5
n -Dod ecyl n -p en tyl su lfoxid e (6): mp ) 70-72 °C (pen-
tane). [R]D ) -0.93 (c ) 1, CHCl3) for the (R)-configuration).
mp ) 71-73 °C (pentane). [R]D ) +0.73 (c ) 2.5, CHCl3) for
the (S)-configuration. 1H NMR (500 MHz, CDCl3) 2.74-2.69
(m, 2 H), 2.67-2.59 (m, 2 H), 1.81-1.71 (m, 4 H), 1.50-1.20
(m, 22 H), 0.92 (t, J ) 7.1 Hz, 3 H), 0.88 (t, J ) 7.0 Hz, 3 H).
13C NMR (125 MHz, CDCl3) 52.40, 52.35, 31.89, 31.00, 29.59
(2 carbons), 29.52, 29.35, 29.32, 29.19, 28.87, 22.67, 22.60,
22.29 (2 carbons), 14.10, 13.82. Anal. Calcd for C17H36OS: C,
70.77; H, 12.58. Found: C, 70.35; H, 12.42. The ee value,
measured by HPLC (Chiralcel OB-H, hexane/2-propanol
99:1), was >98%.
(R)-n -Dod ecyl 2-p r op yl su lfoxid e (7): mp ) 54-56 °C
(pentane). [R]D ) +47.5 (c ) 1, CHCl3). 1H NMR (500 MHz,
CDCl3) 2.78 (heptet, J ) 7.1 Hz, 1 H), 2.65-2.53 (m, 2 H),
1.83-1.73 (m, 2 H), 1.50-1.21 (m, 24 H), 0.88 (t, J ) 6.9 Hz,
3 H). 13C NMR (125 MHz, CDCl3) 50.11, 48.72, 31.90, 29.59,
29.52, 29.37, 29.32, 29.22, 28.97, 22.89, 22.66, 16.09, 14.65,
14.08. Anal. Calcd for C15H32OS: C, 69.16; H, 12.38. Found:
C, 69.36; H, 12.04. The ee value, measured by HPLC (Chiralcel
OD-H, hexane/2-propanol 99:1), was >98%.
Ack n ow led gm en t. This work was financially sup-
ported in part by the Ministero dell’Istruzione, Univer-
sita` e Ricerca, Rome (National Project “Stereoselezione
in Sintesi Organica. Metodologie e Applicazioni”) and
by the University of Bari.
(S)-p-Br om op h en yl n -d od ecyl su lfoxid e (4): mp 66-68
°C (hexane). [R]D ) -109.5 (c ) 1, CHCl3). 1H NMR (500 MHz,
CDCl3) 7.67-7.65 (m, 2 H), 7.50-7.45 (m, 2 H), 2.76 (t, J )
7.8 Hz, 2 H), 1.78-1.69 (m, 1 H), 1.63-1.53 (m, 1 H), 1.45-
1.18 (m, 18 H), 0.88 (t, J ) 7.1 Hz, 3 H). 13C NMR (125 MHz,
CDCl3) 143.23, 132.39, 125.64, 125.29, 57.33, 31.88, 29.57,
29.48, 29.30, 29.12, 28.62, 22.66, 21.99, 14.09. Anal. Calcd
for C18H29BrOS: C, 57.90; H, 7.83. Found: C, 58.26; H, 8.18.
The ee value, measured by HPLC (Chiralcel OD-H, hexane/
2-propanol 90:10), was >98%.
Su p p or tin g In for m a tion Ava ila ble: Additional relevant
spectral data for compounds 1-7. This material is available
free of charge via the Internet at http://pubs.acs.org.
J O025974J
7294 J . Org. Chem., Vol. 67, No. 21, 2002