This document is the Accepted Manuscript version of a Published Work that appeared in final form in
Organic Letters, copyright © American Chemical Society after peer review and technical editing by the
publisher. To access the final edited and published work see http://pubs.acs.org/doi/pdf/10.1021/acs.orglett.5b02139

Hydrogenative Kinetic Resolution of Vinyl Sulfoxides
Joan R. Lao,† Héctor Fernández-Pérez,† and Anton Vidal-Ferran*,†,‡
†Institute of Chemical Research of Catalonia (ICIQ), Av. Països Catalans, 16, 43007 Tarragona, Spain
‡Catalan Institution for Research and Advanced Studies (ICREA), P. Lluís Companys, 23, 08010 Barcelona, Spain

ABSTRACT: Enantiopure sulfoxides are valuable precursors of organosulfur compounds with a broad application in organic and
pharmaceutical chemistry. An unprecedented strategy for obtaining highly enantioenriched sulfoxides based on a hydrogenative
kinetic resolution using Rh-complexes of phosphine-phosphites ligands as catalysts is reported. After optimization, highly efficient
conditions for the kinetic resolution of racemic sulfoxides have been identified. This methodology has been applied to a set of racemic aralkyl or aryl vinyl sulfoxides and allowed the isolation of both recovered and reduced product in excellent yields and enantioselectivities (up to 99% and 97% ee, respectively; 16 examples).
Optically pure sulfoxides are a valuable family of chiral
compounds which have proven to be highly efficient chiral
ligands1 as well as useful intermediates in the synthesis of
relevant biologically active compounds.1a,2 Among the approaches that asymmetric catalysis offers, kinetic resolution
(KR) of racemic sulfoxides3 should be considered an appealing method for the preparation of two optically pure sulfoxides
in only one synthetic step, provided that some requisites are
fulfilled. First and foremost, it is necessary that efficient enantioselective catalysts working at low catalyst loadings are
available and, second, that starting materials and products can
be isolated in good yields and enantiomerically enriched
forms.4
Whilst the reported nonenzymatic KRs on racemic sulfoxides are mainly based on oxidative transformations (Scheme 1,
(a)),3,5 nonoxidative KRs, including reductive transformations
(Scheme 1, (b)), have received much less attention. Moreover,
nonoxidative KRs have normally offered unsatisfactory stereoselectivities, with the exception of enzymatic6 transformations and hydrogenative dynamic kinetic resolutions (DKR)
of allyl sulfoxides.7 There are a few studies reporting reductive
KR of vinyl sulfoxides with optically active reagents,8 however, to the best of our knowledge, there are no previous reports
on the KR of vinyl sulfoxides via asymmetric hydrogenation.9

Our group recently reported the highly enantioselective hydrogenation of a structurally diverse set of substrates mediated
by phosphine-phosphite (POP)10 ligands. The high catalytic
activities achieved with our ligands promted us to address the
challenge of hydrogenatively resolving racemic vinyl sulfoxides (Scheme 1, (c)), whose resolved products have found
broad applicability in catalytic asymmetric synthesis.1,2 Herein
we describe our results, which include the catalyst optimization studies and the application of the lead-catalyst to the
highly efficient hydrogenative KR of an array of racemic
aralkyl or aryl vinyl sulfoxides. At the onset of our study, we
chose phenyl vinyl sulfoxide rac-1a as model substrate. The
reaction conditions and the results of this initial screening are
summarized in Table 1.
Table 1. Ligand screening for the KR of rac-1aa

entry

L1

25

22 (S)

73 (R)

2

L2

16

7 (R)

37 (S)

3

L3

79

99 (S)

28 (R)

4
a

conv,
%b

1

Scheme 1. Kinetic resolution strategies for racemic sulfoxides

ligand

L4

81
b

ee of 1a,
ee of 2a,
%;c (config.)d %;c (config.)d

85 (R)
1

20 (S)
c

[Substrate] = 0.2 M. Determined by H NMR. Determined by
HPLC on chiral stationary phases. daThe absolute configuration
was assigned by comparison with reported data.

As indicated in Table 1, both activity and selectivity were
highly dependent on the POP ligand used. Rhodium complexes derived from ligands L1 and L2 afforded the hydrogenated product 2a with 3773% ee, though conversions were
very low and ee values for the recovered starting material 1a
poor (see entries 1 and 2 in Table 1). In contrast, rhodium
complexes of the new ligands L3 and L4 displayed an opposite trend with high conversions and excellent enantioselectivities for 1a (up to 99% ee) (see entries 3 and 4 in Table 1).
Therefore, these results clearly identified L1 and L3 as the
optimal ligands for this chemistry with the stereogenic phosphite group being the principal stereochemical director (opposite absolute configurations for sulfoxides 1a and 2a are obtained depending on the configuration of the phosphite group:
compare entries 1 with 2 for L1 and L2, or entries 3 with 4 for
L3 and L4 in Table 1, respectively).
Next, we proceeded to optimize the reaction conditions with
POP ligands L1 and L3 in a range of different solvent mixtures.11 The assayed reaction conditions and results are shown
in Table 2.
Table 2. Solvent optimization using ligands L1 or L3a

enantiomer) and the corresponding hydrogenated product 2a
(see entry 6 in Table 2).
With the optimal catalyst in hand, we then attempted to
broaden the substrate scope to a set of structurally diverse
vinyl sulfoxides (rac-1ah). In order to maximize the yield for
both the recovered and hydrogenated products (1ah and
2ah, respectively), specific reaction conditions for 1ah and
for 2ah were investigated.13 The results and optimized reaction conditions are listed in Table 3.
Table 3. Substrate scope of the hydrogenative KR of racemic vinyl sulfoxides rac-1ah.a

entry

R

t, min

conv,
%b

product, isol. yield,
%;14
ee, %;c (config.)d

se

1f

Ph
(rac-1a)

60

54

1a, 74, 99 (S)

30

30

44

2a, 62, 86 (R)

35

p-Me-Ph
(rac-1b)

80

55

1b, 74, 98 (S)

25

10

35

2b, 58, 95 (R)

55

o-F-Ph
(rac-1c)

15

55

1c, 80, 99 (S)

56

8

42

2c, 80, 97 (R)

148

m-F-Ph
(rac-1d)

30

61

1d, 54, 98 (S)

25

10

39

2d, 76, 92 (R)

44

p-F-Ph
(rac-1e)

30

54

1e, 70, 98 (S)

33

15

44

2e, 56, 90 (R)

40

p-MeO-Ph
(rac-1f)

120

64

1f, 64, 99 (S)

26

30

39

2f, 76, 88 (R)

29

p-NO2-Ph
(rac-1g)

25

66

1g, 64, 99 (S)

13

10

40

2g, 72, 82 (R)

18

2
3
4
ee of ee of
1a:2a:3ab 1a, %;c 2a, %;c
(S)d
(R)d

entry

L

solvent

t, h

conv,
%b

1

L1

Cy

2

56

44:34:12

99

80

2

L1

MeOH

2

10

90:6:4

9

73

3

L1

MeTHFe

2

44

56:42:2

65

83

4

L1

Toluene

2

56

44:52:4

92

76

5

L1

CH2Cl2

2

25

75:25:0

22

73

6

L3

Cy

1

54

46:54:0

99

72

7

L3

MeOH

1

79

21:79:0

99

28

8

L3

MeTHFe

1

60

40:60:0

99

64

L3

Toluene

L3

CH2Cl2

9
10

1
1

58
68

42:58:0
32:68:0

99
99

64
45

a, b, c, d

See notes a, b, c and d in Table 1. e MeTHF ≡ 2-methyltetrahydrofuran.

According to these results, a mixture of cyclohexane and
CH2Cl2 was identified as the optimal solvent for rac-1a as it
provided the highest ee values for sulfoxides 1a and 2a (see
entries 1 and 6 in Table 2). Unfortunately, the rhodium complex derived from ligand L1 led to the formation of significant
amounts of ethyl phenyl sulfide 3a as byproduct arising from
the overreduction of the starting material (see entry 1 in Table
2).12 This phenomenon was also observed for L1 in all the
mixtures of solvents tested (see entries 1 to 4 in Table 2).
However, we were pleased to find that this side reaction was
completely eliminated by using the rhodium complex derived
from the POP ligand L3. Moreover, this catalyst provided at
54% conversion the recovered vinyl sulfoxide 1a in a perfect
enantioselectivity (up to 99% ee in favor of the (S)-

5
6
7
8
9
10
11
12
13
14
15

g

16

g

Bn
(rac-1h)

240
25

67
33

h

11

h

13

1h, 62, 99 (R)
2h, 56, 80 (S)

a, b, c

See notes a, b and c in Table 1. d The absolute configuration
of 1a,b and 2a,b,fh was established by comparison with reported
optical rotations. The absolute configuration of 1ch and 2ce
was tentatively assigned by analogy with the stereochemical
outcome of the reactions leading to 1a,b and 2a,b,fh.12 eaThe
selectivity factor (s)4 was determined by the equation s 
krel(fast/slow)  ln[1C(1ee2)]/ln[1C(1ee2)]. faThis result has
been already shown in Table 2. g Solvent ratio used was
Cy:CH2Cl2 (2.6:1). h The opposite R or S prefixes in 1h and 2h
arise from different priorities in the CIP rules.

As illustrated in Table 3, different substitution patterns on
the aryl groups of vinyl sulfoxides rac-1ce (o-F, m-F and p-F
substitution, respectively) were well tolerated to furnish sulfoxides 2ce in 5680% isolated yield14 with 9097% ee, and
the recovered vinyl sulfoxides 1ce in 5480% isolated yield14
with very high enantioselectivities (9899% ee; see entries
510 in Table 3). The results obtained for the ortho-

substituted substrate rac-1c were the best among all the substrates assessed (see entries 5 and 6 in Table 3). Plots of ee
values of resolved products against conversion displayed that
the highest enantioselectivities for such compounds were
achieved in the range of 4060% conversion (see Figure 1,
which corresponds to the KR of substrate rac-1c), demonstrating the high efficiency of this KR to provide unreacted starting
material and hydrogenated product in high yields and ee’s.

vations strongly suggest that the CC bonds of (R)-1g and (S)1g are coordinated to the phosphino group in a cis fashion, as
this data is practically coincident with that reported in the
literature for cis-coordinated CC and phosphorus groups in
related rhodium complexes. 15 Furthermore, intense crosspeaks between the olefinic HC proton trans to HC and
aromatic protons of the diphenylphosphino group in NOESY
correlation experiments were observed for bound (S)-1g, thus
confirming the previous structural assignment.12 This coordination mode of the CC double bond places the R group of the
sulfoxide in close proximity to the phosphite group, which
accounts for this group being the principal stereochemical
director in the KR. By comparing the results of these coordination studies and the configuration of the resolved products, a
tentative reaction pathway for the stereochemical outcome of
the KR is proposed in Figure 2.

Figure 1. Ee values (%) of 1c and 2c vs. conversion (%).

Electronic effects were also studied by the examination of
para-substituted substrates rac-1b,e,f,g (p-Me, p-F, p-MeO, pNO2, respectively). Regardless of the electronic nature of the
substituent on the aromatic ring, the substrates were hydrogenated leading to both unreacted and reduced sulfoxides with
high enantioselectivities (from 82 to 99% ee) in 5676% isolated yields14 (see entries 3,4 and 914 in Table 3). However,
an electron withdrawing group at the para-position increased
the reaction rate (compare entries 9, 10 and 13, 14 with entries
3, 4 and 11, 12 in Table 3). The lead catalytic system was also
capable of efficiently resolving benzyl vinyl sulfoxide (rac1h): Hydrogenation of rac-1h for 4 h provided (R)-1h in 62%
isolated yield14 with perfect enantioselectivity (99% ee, see
entry 15 in Table 3), while optimal reaction conditions for the
hydrogenated product (S)-2h led to its isolation in 56% isolated yield14 and 80% ee (see entry 16 in Table 3). In order to
demonstrate the practicality of this KR method, experiments at
the mmol scale were performed for racemic substrates rac1a,g to afford products 1a,g and 2a,g with the same efficiency
than catalytic experiments.12
To shed light on the favored stereodifferentiating routes, we
studied the coordination of (R)-1g and (S)-1g to the
[Rh(POP)]+ complex of the lead ligand (L3). We pursued the
in situ preparation of [Rh(1g)(L3)]BF4 by hydrogenation of
[Rh(nbd)(L3)]BF4 in 1,2- dimethoxyethane followed by the
addition of 1.1 equiv. of (R)-1g or (S)-1g. Based on related
literature precedents,7,9 we hypothesized that the KR proceeds
via hydrogenation of the CC bond with chelating assistance
of the oxygen atom ofthe sulfoxide group. Examination of the
NMR data in solution indicated that both (R)-1g and (S)-1g
coordinate to the [Rh(POP)]+ motif. The coordination of (S)1g led to the formation of a stable complex at rt, as evidenced
by the sharpness of the vinylic signals in the 1H NMR spectrum.12 The formation of the substrate-catalyst adduct [Rh((R)1g)(L3)]+ was evidenced by the appearance of broad vinylic
signals in the 1H NMR spectrum, which sharpened up upon
recording the spectra at a lower temperature (253 K). With
regard to the geometry of the complexes between (R)-1g or
(S)-1g and the [Rh(POP)]+ motif, cross-peaks only between
the olefinic protons and the phosphino group in heteronuclear
1
H-31P correlation experiments were observed.12 These obser-

Figure 2. Tentative reaction pathways for the hydrogenative KR
of racemic vinyl sulfoxides in the Rh-L3 complexes with the CC
and P groups coordinated in a cis fashion (aR or S configurations
of the product have been established assuming the highest CIP’s
priority for the R group).

In summary, we have developed a highly efficient hydrogenative KR of vinyl sulfoxides mediated by rhodium complexes of POP ligand L3, which are responsible for the differentiation of the reaction rates of the two enantiomers of the
starting material toward hydrogenation. This KR method is an
unprecedented approach for preparing optically active vinyl
sulfoxides and their hydrogenated products in notable yields
and high enantioselectivities (up to 99% ee). The easy availability of racemic vinyl sulfoxides, together with the excellent
catalytic profile of the catalyst derived from L3, makes the
herein described synthetic methodology a valuable synthetic
entry for chiral sulfoxides. Further studies on the application
of this synthetic methodology, together with mechanistic investigations, are underway in our laboratory and will be reported in due course.

ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS publications website at DOI: 10.1021/acs.orglett.5b02139.
Experimental procedures, spectral data, and determination of
the enantiomeric excess. (PDF)

AUTHOR INFORMATION
Corresponding Author
* E-mail: avidal@iciq.cat

Notes
The authors declare no competing financial interest.

ACKNOWLEDGMENT
The authors thank MINECO (CTQ2014-60256-P and Severo
Ochoa Excellence Accreditation SEV-2013-0319) and the ICIQ
Foundation for financial support. J.-R. L. thanks the ICIQ Foundation for a pre-doctoral fellowship. Dr. P. Etayo (ICIQ) is
acknowledged for discussions at the beginning of the project.

REFERENCES
(1) For selected recent reviews, see: (a) Fernández, I.; Khiar, N.
Chem. Rev. 2003, 103, 3651. (b) Fernández, I. and Khiar, N. In Organosulfur Chemistry in Asymmetric Synthesis; Toru, T. and Bolm,
C., Eds.; WILEY-VCH Verlag GmbH & Co. KGaA: Weinheim,
2008; p 265. (c) Trost, B. M.; Rao, M. Angew. Chem., Int. Ed. 2015,
54, 5026. (d) Sipos G.; Drinkel, E. E.; Dorta, R. Chem. Soc. Rev.
2015, 44, 3834.
(2) For a recent review, see: Legros, J.; Dehli, J. R.; Bolm, C. Adv.
Synth. Catal. 2005, 347, 19.
(3) For example, see: Kagan, H. B. In Organosulfur Chemistry in
Asymmetric Synthesis; Toru, T. and Bolm, C., Eds.; WILEY-VCH
Verlag GmbH & Co. KGaA: Weinheim, 2008; p 1.
(4) Keith, J. M.; Larrow, J. F.; Jacobsen, E. N. Adv. Synth. Catal.
2001, 343, 5.
(5) The following article has been selected as an oxidative method
considering oxidative as all transformations in which the formal
oxidation state of the S-atom increases: Wang, J.; Frings, M.; Bolm,
C. Chem. - Eur. J. 2014, 20, 966.
(6) For selected examples on enzymatically catalyzed KRs, see for
example: (a) Boyd, D. R.; Sharma, N. D.; King, A. W. T.; Shepherd,
S. D.; Allen, C. C. R.; Holt, R. A.; Luckarift, H. R.; Dalton, H. Org.
Biomol. Chem. 2004, 2, 554. (b) Boyd, D. R.; Sharma, N. D.; Byrne,
B. E.; Haughey, S. A.; Kennedy, M. A.; Allen, C. C. R. Org. Biomol.
Chem. 2004, 2, 2530.
(7) Though DKRs are not the same as the aimed transformations in
this manuscript (KRs), the following reference on DKR of allyl sulfoxides has been included due to the high enantioselectivities
achieved (up to 90% ee): Dornan, P. K.; Kou, K. G. M.; Houk, K. N.;
Dong, V. M. J. Am. Chem. Soc. 2014, 136, 291.
(8) (a) Mikolajczyk, M.; Para, M. J. Chem. Soc. D 1969, 1192. (b)
Mikolajczyk, M.; Drabowicz, J. Phosphorous Sulfur Relat. Elem.
1976, 1, 301. (c) Drabowicz, J.; Pacholczyk, M. Phosphorous Sulfur
Relat. Elem. 1981, 10, 233. (d) Annunziata, R.; Borgogno, G.;
Montanari, F.; Quici, S.; Cucinella, S. J. Chem. Soc., Perkin Trans. 1
1981, 113.
(9) To best of our knowledge, the only reported transformation,
which is close to the chemistry herein described (see: Ando, D.;
Bevan, C.; Brown, J. M.; Price, D. W. J. Chem. Soc., Chem. Commun.
1992, 592.), involves asymmetric hydrogenation of vinyl sulfoxides
and kinetic resolution of one vinyl sulfone.
(10) For selected references on the application of POP ligands in
iridium-mediated hydrogenations, see: (a) Núñez-Rico, J. L.;
Fernández-Pérez, H.; Benet-Buchholz, J.; Vidal-Ferran, A.
Organometallics 2010, 29, 6627. (b) Núñez-Rico, J. L.; Vidal-Ferran,
A. Org. Lett. 2013, 15, 2066. (c) Núñez-Rico, J. L.; Fernández-Pérez,
H.; Vidal-Ferran, A. Green Chem. 2014, 16, 1153. For those on
rhodium-mediated hydrogenation, see: (d) Donald, S. M. A.; VidalFerran, A.; Maseras, F. Can. J. Chem. 2009, 87, 1273. (e) FernándezPérez, H.; Donald, S. M. A.; Munslow, I. J.; Benet-Buchholz, J.;
Maseras, F.; Vidal-Ferran, A. Chem. - Eur. J. 2010, 16, 6495. (f)
Etayo, P.; Núñez-Rico, J. L.; Fernández-Pérez, H.; Vidal-Ferran, A.
Chem. - Eur. J. 2011, 17, 13978. (g) Núñez-Rico, J. L.; Etayo, P.;
Fernández-Pérez, H.; Vidal-Ferran, A. Adv. Synth. Catal. 2012, 354,
3025. (h) Fernández-Pérez, H.; Benet-Buchholz, J.; Vidal-Ferran, A.
Chem. - Eur. J. 2014, 20, 15375.
(11) A certain amount of CH2Cl2 was kept in order to ensure the
complete solubility of both rhodium precursor and ligand.
(12) See the Supporting Information for further details.
(13) The reaction conditions were optimized by modifying the reaction times, (controlling in this way the value of conversion), in an

optimal compromise amongst enantiopurity and amounts of unreacted
starting material and product.
(14) Yields are calculated with respect to the 50 mol % amount of
starting material that was subjected to KR.
(15) (a) Bircher, H; Bender, B. R.; Philipsborn, W. v. Magn. Reson. Chem. 1993, 31, 293. (b) Gridnev, I. D.; Higashi, N.; Asakura,
K.; Imamoto, T. J. Am. Chem. Soc. 2000, 122, 7183. (c) Suárez, A.;
Méndez-Rojas, M. A.; Pizzano, A. Organometallics 2002, 21, 4611.

5