C.J. Carrasco et al. / Catalysis Communications 84 (2016) 134–136
135
can be explained by the solvent coordination to the metal centre that
compete with the (S,S)-HLiPr coordination, limiting the formation of
the Mo-catalyst responsible of the asymmetric process. From the differ-
ent solvents evaluated, chlorinated solvent, such as Cl2CH2 (entry 6),
1,2-dichloroethane (entry 7) and Cl3CH (entry 8) were showed to be
the most efficient, affording 32, 18 and 38% enantiomeric excess, respec-
tively. The use of these polar non-coordinating solvents facilitates the
interaction of the precursor ligand HLiPr with the metal centre rendering
the process asymmetric. Following these results, Cl3CH was selected as
the reaction solvent for further improvements of the reaction condi-
tions. Asymmetric sulfoxidation was found to be strongly dependent
on the selected temperature. The increase of the temperature from 0
to 25 °C reduced the asymmetric induction, affording the chiral sulfox-
ide only in 10% ee (entry 9). On the other hand, lowering the reaction
temperature to −30 °C did not provide any benefit (entry 10) mainly
due to a significant reduction of the conversion (37%). Oxidation of
PhMeS clearly proceeded via a metal catalyzed mechanism since negli-
gible conversion was observed after 1 h at 0 °C in the absence of molyb-
denum (entry 11). For the sake of comparison, tert-butyl hydroperoxide
was also used as oxidant under the same conditions. The catalytic activ-
ity decreased notably with this oxidant (entry 12) and the reaction pro-
ceeds selectively to the sulfoxide as a racemic mixture, showing a non-
asymmetric process. The [PPh4]Br salt was a necessary additive to im-
prove the activity of the catalyst for the asymmetric sulfoxidation. The
phosphonium cation presumably behaves as counterion of the anionic
oxidodiperoxidomolybdenum species, formed by the in-situ reaction
between [MoO(O2)2(H2O)n] and (S,S)-HLiPr, helping to increase the sol-
ubility of the Mo-catalyst in the reaction medium. In all the reactions de-
scribed above (entries 1–12), 0.5 equivalents of [PPh4]Br respect to
molybdenum was used as additive, giving the characteristic yellow
color of the oxidodiperoxido-Mo(VI) species dissolved in the reaction
mixture. Conversely, no coloration of Cl3CH phase was observed when
the reaction was done in the absence of [PPh4]Br, indicating a poor
transfer of the Mo-catalyst from the aqueous solution. In the latter
case (entry 13), a significant worse conversion of the sulfide (38%)
than the reactions performed with the phosphonium salt was observed.
Finally, an excess of the phosphonium salt (2 equivalents respect to Mo)
was also showed to slightly improve the catalytic efficiency (entry 14),
since an increase was observed both in the conversion of the sulfide
(94%) and in the enantioselectivity (40%). Therefore, the [PPh4]Br salt
also seems to act as efficient phase transfer catalyst accelerating the re-
action between reagents dissolved in the immiscible solvents H2O (i.e.
H2O2) and chloroform (i.e. organic sulfide). In order to gain information
about the nature of the Mo-catalyst, the stoichiometric 2:1 reaction of
[Mo(O)(O2)2(H2O)n] with the sodium salt of (S,S)-HLiPr was carried
Scheme 1. Mo-catalyzed asymmetric oxidation of prochiral sulfides with H2O2 in the
presence of [MoO(O2)2(H2O)n], (S,S)-HLiPr and [PPh4]Br.
prepared as described elsewhere and referred to in this work, for the
purpose of simplicity, as [Mo(O)(O2)2(H2O)n] [35]. The catalyst was
prepared in situ by mixing [MoO(O2)2(H2O)n] with (S,S)-HLiPr and
tetraphenylphosphonium bromide (see supplementary data). Table 1
reports selected results from the reactions carried out in a microreactor
(1 mL of solvent) with a 1:1:0.025 ratio of PhMeS:oxidant:Mo-complex.
Initially, several solvents were tested at 0 °C in order to evaluate their ef-
fect on the enantioselectivity of the process. The ionic liquid
[C4mim][PF6] (C4mim = 1-butyl-3-methylimidazolium) was the first
solvent to be evaluated (entry 1). In our previous work, we proved
that [C4mim][PF6] is an optimal solvent for the selective sulfoxidation
of sulfides when combined with aqueous H2O2 as oxidant and several
simple Mo-complexes, including [MoO(O2)2(H2O)n], as catalyst [33].
However, despite the reaction was selective to the formation of
PhMeSO, only an b1% of enantiomeric excess was achieved probably
due to the low solubility of (S,S)-HLiPr in the [C4mim][PF6] solvent. Insig-
nificant enantiomeric excesses were also obtained with other conven-
tional solvents, such as toluene (entry 2), MeOH (entry 3), DMF (entry
4) or acetonitrile (entry 5). The low conversion and irrelevant
enantioselectivity observed in toluene can presumably be attributed to
the low solubility of the Mo-catalyst. In polar coordinating solvents,
such as MeOH, DMF or acetonitrile, the almost null enantioselectivity
Table 1
Enantioselective oxidation of PhMeS with H2O2 to the (R)-sulfoxide with the system
[MoO(O2)2(H2O)n]/(S,S)-HLiPr/[PPh4]Br.a
Conversion Selectivity to
Selectivity to
sulfone (%)b
Sulfoxide
Entry Solvent
(%)b
sulfoxide (%)b
ee (%)c
1
2
3
4
5
6
7
8
[C4mim][PF6] 100
100
81
97
95
98
95
93
97
94
0
19
3
5
2
5
7
3
6
3
0
0
0
5
7
≤2
≤2
≤2
≤2
≤2
32
18
38
10
26
≤2
0
Toluene
MeOH
DMF
CH3CN
Cl2CH2
1,2-Cl2C2H4
Cl3CH
Cl3CH
Cl3CH
Cl3CH
20
97
54
96
93
88
87
94
37
3
out.
A
yellow powder was isolated and the formulation
Na{[Mo(O)(O2)2(H2O)]2(μ-LiPr)} was proposed on the basis of experi-
mental data and DFT calculations (see details in supplementary data).
With the purpose of confirming the activity of this compound, it was
tested as catalyst in the enantioselective sulfoxidation of PhMeS (entry
15, Table 1), under the same reaction conditions. The conversion
(93%) and ee (42%) values obtained were similar to those of entry 14,
thus confirming the Mo-catalyst nature.
9d
10e
11f
12g
13h
14i
15j
97
100
100
100
95
Cl3CH
Cl3CH
Cl3CH
Cl3CH
14
38
94
93
≤2
40
42
The amount of H2O2 employed in all the experiments collected in
Table 1 is one equivalent per substrate, because the use of two or
more equivalents of the oxidant yielded the corresponding sulfone
[33]. However, in order to analyze the possible augment of the ee yield
by kinetic resolution, different oxidant:substrate ratios were employed
under the optimized experimental conditions (entry 14, Table 1).
Fig. 1 shows the results that confirm that an excess of the oxidant im-
prove the ee of (R)-sulfoxide (at expense of the sulfoxide yield) because
the (S)-sulfoxide was oxidized faster into sulfone than the (R)-enantio-
mer. The sulfoxide concentration follows a bell-shaped curve: after hav-
ing reached a maximum at approximately oxidant:sulfide ratio of 1, the
sulfoxide concentration finally drops off again when most of the sub-
strate is consumed and the second oxidation step, to form sulfone at
93
a
Reaction conditions: catalyst [MoO(O2)2(H2O)n] 0.025 mmol, (S,S)-HLiPr 0.0125
mmol, [PPh4]Br 0.0125, PhMeS 1.0 mmol, solvent 1.0 mL, oxidant: H2O2 (30% aq),
[oxidant]:[sulfide] ratio 1:1, 1 h, T = 0 °C.
b
Determined by GC (50 μL of dodecane as the internal standard).
Determined by HPLC (see supplementary data).
T = 25 °C.
T = −30 °C.
Without Mo-catalyst.
Oxidant: tert-butyl hydroperoxide.
Without [PPh4]Br.
c
d
e
f
g
h
i
With 0.05 mmol of [PPh4]Br.
j
Na{[Mo(O)(O2)2(H2O)]2(μ-LiPr)} as catalyst (see Supplementary data).