Conjugate additions to alkylidene bis(sulfoxides)

Article abstract of DOI:10.1002/asia.201000904

A general study on the conjugate addition of anionic nucleophiles to alkylidene bis(sulfoxides) is presented. Alkoxides gave high yielding and diastereoselective addition reactions, which could be influenced by solvents and the counteranion. Azides provided an interesting entry into sulfinyl-substituted triazoles. Organometallics, mainly copper reagents, proved also to be valuable nucleophiles, and complete inversion of the stereoselectivity was achieved in the addition reaction with the latter. Modelizations provide a rationale for the observed diastereoselectivity. Taking the Michael: Alkylidene bis(sulfoxides) are exquisite Michael acceptors toward a broad series of nucleophiles, and in most cases the reaction affords the adducts in high yields with complete diastereoselectivity.

Full text of DOI:10.1002/asia.201000904

DOI: 10.1002/asia.201000904
Conjugate Additions to Alkylidene Bis(Sulfoxides)
Franck Brebion,[a] Guillaume Vincent,[a] Saloua Chelli,[a] Ophꢀlie Kwasnieski,[a]
Francisco Najera,[a] Bꢀnꢀdicte Delouvriꢀ,[a] Ilan Marek,[b] Etienne Derat,[a]
Jean-Philippe Goddard,[a] Max Malacria,*[a] and Louis Fensterbank*[a]
On the occasion of the 150th anniversary of Department of Chemistry, The University of Tokyo
Abstract: A general study on the conjugate addition of anionic nucleophiles to al-
kylidene bis(sulfoxides) is presented. Alkoxides gave high yielding and diastereo-
selective addition reactions, which could be influenced by solvents and the coun-
Keywords: azides · bis(sulfoxides) ·
teranion. Azides provided an interesting entry into sulfinyl-substituted triazoles.
diastereoselectivity · ketene equiva-
Organometallics, mainly copper reagents, proved also to be valuable nucleophiles,
lent · Michael addition
and complete inversion of the stereoselectivity was achieved in the addition reac-
tion with the latter. Modelizations provide a rationale for the observed diastereo-
selectivity.
Introduction
Chiral ketene equivalents have attracted the attention of
synthetic chemists. While mainly used in cycloaddition reac-
ACHTUNGTRENNUNG
tions,[1] the bis(sulfinyl) congeners,[2,3] which are generally
based on simple dioxygenated (S,S)-acetal templates A or B
(Scheme 1),[4] or bis-p-tolyl analogs C,[5] have proven to be
versatile partners in these reactions. They could, however,
also be involved in other key transformations, such as Mi-
chael additions,[6] as recently illustrated by Podlech and
Wedel with monosubstituted substrates of type A[7] and B[8]
[a] Dr. F. Brebion, Dr. G. Vincent, Dr. S. Chelli, Dr. O. Kwasnieski,
Dr. F. Najera, Dr. B. Delouvriꢀ, Dr. E. Derat, Dr. J.-P. Goddard,
Prof. M. Malacria, Prof. L. Fensterbank
UPMC University Paris 06
Scheme 1. Most commonly used alkylidene bis(sulfoxides).
Institut Parisien de Chimie Molꢀculaire (UMR CNRS 7201)
4 place Jussieu, C. 229, 75005 Paris (France)
Fax : (+33)1-44-27-73-60
E-mail: louis.fensterbank@upmc.fr
and also demonstrated by our research group with deriva-
tives of type 1.[9]
max.malacria@upmc.fr
[b] Prof. I. Marek
The Mallat Family Laboratory of Organic Chemistry
Schulich Faculty of Chemistry and The Lise Meitner-MinerVa
Center for Computational Quantum Chemistry
Technion-Israel Institute of Technology
Haifa 32 000 (Israel)
In this study, we bring some additional elements on the
reactivity of conjugate additions[10] to 1, as well as rational-
ize the stereochemical outcome of the addition of organo-
metallic reagents.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201000904.
Chem. Asian J. 2011, 6, 1825 – 1833
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1825

FULL PAPERS
Results and Discussion
of alkylidene bis(sulfoxides) over simple vinyl sulfoxides in
terms of efficiency in the addition step and diastereoselec-
tivity.[10,16] It was also interesting to check if the same reac-
tivity and stereoselectivity could apply to charged nucleo-
philes. Thus, we examined malonates, ester enolates, and
other carbanions.[9a–c] Herein, we focus on new anionic nu-
cleophiles and extend our findings with alkoxides and orga-
nometallics.
Preparation of 1 and Structural Data
Although, we began to look at asymmetric oxidation pro-
cesses to prepare precursors 1,[11] we have relied on follow-
ing the bis
previously described.[12] It should also be noted that bis-
(sulfinyl) alcohols 3 can be obtained in moderate to very
good diastereoselectivities, and that the synthesis of bis-
(sulfinyl) methane 2 has recently been optimized.[13]
ACHTUNGTRENNUNG(sulfinyl) methane route (Scheme 2), which we
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
Alkoxides as Nucleophiles
Alkoxides proved to be versatile nucleophiles towards pre-
cursor 1a (Table 1). Optimized conditions involved adding
1.5 equivalents of the alkoxide at À458C and quenching the
Table 1. Addition of alkoxides on 1a.
Scheme 2. Synthesis of alkylidene bis(sulfoxides) 1.
Entry Alkoxide Solvent
Product Yield [%] d.r.
Abs
conf.
Most precursors of type 1 have been obtained in satisfac-
tory overall yields, and isolated as solids, thus allowing X-
ray structure determination of 1a (R=Ph) and 1b (R=iPr).
These structures established an eclipsed lone pair conforma-
tion for the sulfinyl moiety syn to the R group and a s-cis
conformation for the other sulfinyl group;[9a] this is consis-
tent with Tietzeꢂs calculations on simpler vinyl sulfoxide sys-
tems.[14] Assuming that the same conformational bias would
operate in solution, we could anticipate highly diastereose-
lective Michael additions from the si face, thereby avoiding
any interaction of the nucleophile with the tolyl groups that
lead to adducts D. It should be mentioned that Podlech has
recently revisited the diastereoselection model for the addi-
tions on substrates of type A and B and has proposed that
the stereochemical outcome is controlled by the stabilizing
1
2
3
MeOLi
MeONa THF
MeOLi
THF
4a
4a
4a
94
90
100
>98:2 (S)
>98:2 (S)
82:18 (S)
THF/
MeOH[a]
THF
4
5
EtOLi
EtONa
4b
4b
85
89
96:4
(S)
THF
>98:2 (S)
[a] A THF/MeOH (1.2:1) mixture was used.
reaction at the same temperature to provide high yields and
diastereoselectivities of adducts 4. Sodium alkoxides gave
complete stereoselectivity (Table 1, entries 2 and 5) as deter-
mined from the 1H NMR spectrum of the crude product,
whereas in some cases a lithium salt resulted in slightly
eroded diastereoselectivity (entry 4). Different results were
obtained by varying the solvent, as illustrated in entry 3 with
a tetrahydrofuran (THF)/MeOH (1.2:1) mixture. In this
case, a noticeable reduction in the stereoselectivity (entry 3)
was observed. Unfortunately, the reaction in pure methanol
could not proceed because 1a is poorly soluble in this reac-
tion medium. While solvent effects have already been ob-
served in the addition of nucleophiles on simple vinyl sulf-
oxides by Pyne et al.,[16a] we suspect that methanol might
favor hydrogen-bonding interactions with 1a, thus altering
the conformations in the reaction system (Scheme 3).
interaction between the forming lone pair at the bisACTHNUTRGNEUG(N sulfinyl)
[15]
À
center and the S O s* of an antiperiplanar S=O group.
Preliminary reports confirmed that amines[9a] could add
efficiently and with high to complete diastereoselectivity to
a series of substrates 1, following the proposed stereoinduc-
tion of Scheme 3. While these initial findings provided a
new entry into the preparation of enantiopure b-aminoalco-
hols and a-aminoacids, they also illustrated the clear benefit
Adduct 4a was derivatized into the previously described
alcohol 5 following a two-step sequence, which involved a
Pummerer rearrangement and subsequent reduction of the
type-E thioester. Correlation with literature data established
an (S) configuration for 5.[9a] An enantiomeric excess of
97.5% was measured by chiral GC for (S)-5, thereby con-
firming a highly diastereoselective conjugate addition pro-
cess and
a
nonracemizing Pummerer rearrangement
(Scheme 4). We deduced by analogy the same (S) absolute
configuration for all products 4.
Scheme 3. Proposed stereochemical outcome.
1826
www.chemasianj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 1825 – 1833

Conjugate Additions
Scheme 4. Derivatization and determination of the absolute configuration
of 4a. TFAA=trifluoroacetic anhydride.
By using these optimized conditions, we could exemplify
this addition reaction. High yields of bisACTHNUTRGNE(NUG sulfinyl) ethers 4
were observed as well as complete diastereoselectivity
(Table 2). A notable exception was the addition of propargyl
alcohol; this resulted in a much lower yield (entry 6). A
simple change in the reaction conditions, such as adding an
excess amount of propargyl alcohol (2 equiv) significantly
Scheme 5. Meerwein–Ponndorf–Verley reaction and reversibility of the
alkoxide additions.
alkyl-substituted
bisACHTUNGTRENNUNG(sulfinyl)
Table 2. Addition of alkoxides: yields and diastereoselectivity.
compounds of type 6, which so
far have been little described.
Methyl-substituted product is
known.[18] However, Trost and
Bridges have shown that simple
alkylation of the bis(phenylsul-
finyl) anion with an allyl bro-
mide or a primary alkyliodide
requires relatively harsh condi-
tions, which leads to sulfinic
acid elimination products.[19]
Also, in our hands this alkyla-
tion reaction has proved to be
troublesome so far (yields
<50%).[20]
Entry
Precursor
R1
R2
Product
Yield [%]
d.r.
Abs conf.
1
2
3
4
5
6
7
1b
1b
1b
1b
1a
1a
1a
iPr
iPr
iPr
iPr
Ph
Ph
Ph
Me
Et
Bn
propargyl
allyl
4d
4e
4 f
4g
4h
4i
94
93
>98:2
>98:2
>98:2
>98:2
>98:2
>98:2
>98:2
(S)
(S)
(S)
(S)
(S)
(S)
(S)
84
64[a]
96
propargyl
propargyl
15[b,c]
64[d]
4i
[a] 4 equiv of alkoxide were used and 30% of 1b was recovered. [b] 5 equiv of alkoxide were used. [c] 85% of
1a was recovered. [d] 2 equiv of alkoxide and 2 equiv of propargylic alcohol were used.
Further experiments gave us
improved the outcome (64% yield at 70% conversion,
entry 7). Presumably, protonation of the bis(sulfinyl) anion
by excess propargyl alcohol ensures the irreversibility of the
addition and results in an improved yield.
a better insight into this reaction and suggested the reversi-
ble nature of the addition of the alkoxides (Scheme 5).
Thus, treating adduct 4e with lithium diisopropylamide
(LDA) at room temperature provided a mixture of 6 and 7,
thus suggesting that retroaddition took place, liberating lithi-
um ethoxide, which triggers the Meerwein–Ponndorf–Verley
sequence.
AHCTUNGTRENNUNG
While the addition of sodium ethoxide to 1b at room tem-
perature resulted in the formation of 4e in high yield
(Table 2, entry 2), lithium ethoxide did not provide the same
outcome under similar conditions (Scheme 5). Instead, two
new products 6 and 7 were isolated; these products presum-
ably resulted from a Meerwein–Ponndorf–Verley[17] reaction.
Thus, hydride conjugate delivery to 1b with lithium ethoxide
would generate 6. The resulting ethanal by-product would
be deprotonated in the basic medium, and the correspond-
ing enolate would also add in a conjugate manner to provide
aldehyde 7 as a major diastereomer. We also found that the
presence of an excess amount of ethanol does not prevent
enolate formation, but results in a slight increase in the
yield of 6.
Trapping of the Bis
Aforementioned, the alkylation of bis
ficult. We could also confirm this by trying to add electro-
philes to the bis(sulfinyl) anion, which results from the addi-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
tion of alkoxides. For instance, the trapped product was not
detected with methyl iodide. Another attempt consisted of
bubbling oxygen through the reaction mixture to hydroxy-
late the bisACTHNUTRGNEUG(N sulfinyl) position, and this should give a transi-
ent intermediate that rapidly evolves into a carboxylate de-
rivative. However, no discrete product was isolated with
these attempts. Finally, ourselves and others have shown
that addition of nucleophiles armed with a leaving group,
thus allowing an intramolecular electrophilic trapping, could
Optimization of the reaction conditions, which would pre-
sumably involve the use of an aluminum alkoxide from an
alcohol leading to a nonenolizable carbonyl derivative might
provide an alternative route for the preparation of primary
Chem. Asian J. 2011, 6, 1825 – 1833
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1827

L. Fensterbank et al.
FULL PAPERS
give high yields of cyclopropyl products as single diastereo-
mers,[9a,c,e] which originates from attack of the nucleophile
from the si face.[21]
Organometallics
Organometallics were obvious candidates to explore.[25] Not
surprisingly, lithium or Grignard reagents favor a 1,2-type
addition, which corresponds to a thiophilic attack and after
displacement of the sulfinyl group[26] leads to a vinyl sulfox-
ide carbanion and the corresponding p-tolyl sulfoxide deriv-
ative with an inversion in configuration. It should be noted
that a slight erosion in the enantiomeric excess was ob-
served with the (R)-methyl-p-tolyl sulfoxide (Table 3, en-
tries 1–3). It is known that vinyl sulfoxide lithium carbanions
isomerize to give, after protonation, the corresponding (E)-
vinyl sulfoxides.[27] Indeed, compounds (E)-11a and (E)-11b
were isolated in good yields.
Azides
In the same vein, aiming at the formation of triazoles, we
also examined the addition of azides. While sodium azide
did not afford any addition product with 1a, presumably for
solubility reasons, tetrabutylammonium azide led to triazole
9a in good yield, the structure of which was confirmed by
X-ray crystallography analysis (Scheme 6).[22a] This transfor-
mation, which formally corresponds to a Huisgen-type cy-
Table 3. Addition of lithium reagents and Grignards.
Entry
Precursor
Organometallic (equiv)
(E)-11 [%]
12 [%]
1
2
3
4
5
6
1a
1a
1a
1a
1a
1b
MeMgCl, 1
MeMgBr, 1
MeMgI, 1.1
AHCTUNGRTENN(GUN nBu)2Mg, 1
MeLi, 1.2
MeLi, 1.2
11a, 85
11a, 83
11a, 31
11a, 75
11a[d]
12a, 63[a]
12a, 82[b]
12a, 43[c]
12b, 83
12a[d]
11b[d]
12a[d]
[a] [a]D =À144.2 (c=1.25, acetone). Literature data for (R)-methylsul-
foxyde is [a]D =+146.0 (c=2.0, acetone). [b] [a]D =À137.0 (c=1.25, ace-
tone). [c] [a]D =À132.0 (c=0.98, acetone). [d] Observed by 1H NMR
spectroscopy on the crude product but not isolated.
Therefore, we turned our attention to softer organometal-
lics and initially examined the reactivity of zinc reagents.
Our initial endeavors focused on the use of the commercial-
ly available diethylzinc reagent solution (Scheme 7) with al-
kylidene 1a. At À308C with 6 equivalents of reagent, no re-
action took place, however, running the reaction at 08C pro-
vided adducts 13 in good yield (84%) as a 7.5:1 mixture of
13aM/13am stereoisomers.[28] In sharp contrast, isobutyli-
dene 1b proved to be unreactive toward diethyl zinc.
Scheme 6. Formation of triazoles.
cloaddition has some precedent with other Michael accept-
ors to an acetylenic sulfoxide,[23] however, to the best of our
knowledge it remains unknown with vinyl sulfoxides. The
putative intermediate 8 was observed in the NMR spectrum
of the crude product but sulfinic acid elimination during
flash chromatography on silica gel took place. Alkyl precur-
sor 1b gave the Evans–Mislow product 10 as a single diaste-
reomer in good yield.[24] Presumably, with alkyl derivatives,
the basic medium would trigger isomerization to the allyl
Because of these moderate results in terms of diastereose-
lectivity and reactivity, we then turned our attention to
copper-based derivatives.[25c–f] Our preliminary findings with
copper reagents (RLi/CuX, 1:1) or cuprates (RLi/CuX, 2:1)
show that these reagents efficiently add to alkylidene bis(-
bisACHTUNGTRENNUNG(sulfinyl) derivative, thereby setting the stage for an
Evans–Mislow rearrangement. The tert-butyl substrate 1c
did not follow the Evans–Mislow pathway and it could also
be operative, thus leading to triazole 9c in good yield. An-
other limitation in this reaction was observed with diene
precursor 1d, which did not afford any adduct in significant
yield.
Scheme 7. Addition of diethylzinc.
1828
www.chemasianj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 1825 – 1833

Conjugate Additions
sulfoxides) in good yields with high and identical stereose-
lectivity, as generally only one diastereomer was isolated
from these reactions (Table 4, entries 1–4). We also found
that the reactions had to be carried out under very anhy-
drous conditions, as hydroxy anions, which could originate
from the presence of water also add and give the corre-
Table 4. Addition of copper reagents.
Scheme 8. Competitive experiment.
Entry
precursor, R
R’Cu (equiv)
Product
Yield [%]
70[a,b]
96
as well as for 14b.[22c] From these data, we deduced the con-
figurations for other adducts. All these stereochemical deter-
minations mark a sharp contrast with the previous cases and
are consistent with an attack from the re face with copper
reagents.
It is noteworthy that the diastereoselectivity in the addi-
tion between the ethylcopper and diethylzinc reagents to 1a
is opposite (13aM/13am ratio shifts from 0:1 to 7.5:1, see
Scheme 7 and Table 4, entry 4). A possible explanation for
this would be that in the case of the reaction with the zinc
reagent, a radical Michael addition is involved, as proposed
recently by different research groups.[29] The free ethyl radi-
cal would follow the diastereoselectivity of a simple nonor-
ganometallic nucleophile (Figure 1).
1
2
3
4
5
6
7
8
1a, Ph
1a, Ph
1a, Ph
1a, Ph
1b, iPr
1b, iPr
1b, iPr
1b, iPr
MeCu, 1
MeCu, 2
nBuCu, 2
EtCu, 2
MeCu, 1
MeCu, 2
Me2CuLi, 1
Me2CuLi, 2
nBuCu, 2
MeCu, 2
MeCu, 2
nBuCu, 2
14a
14a
15a
13am
14b
14b
14b
14b
15b
14e
14f
85
80
42[c,d]
96
73[a]
96
9
1b, iPr
100
85
10
11
12
1e, nBu
1f, CH2OTBS
1f, CH2OTBS
83[b]
52[d]
15f
[a] Around 20% of 1a was recovered. [b] Traces of methylate addition
product of type 4. [c] Around 20% of 1b was recovered. [d] Bis
alcohol of type 3 was isolated (10–15%).
ACHTUNGTRENUN(NG sulfinyl)
sponding bisACHTUNGTRENNUNG(sulfinyl) alcohol 3. We found that cuprates,
probably due to their increased basicity, resulted in less
clean reactions, for example, primary alkyl derivatives (1e
and 1 f), which bear allylic hydrogens can suffer from
Evans–Mislow-type rearrangements. Therefore, we used
copper reagents with these substrates. In some cases, the ad-
dition of minor amounts of methylate, presumably originat-
ing from the oxidation of methyllithium, were observed
(products of type 4).
By examining the results in Table 4, it appears that higher
yields are obtained when two equivalents of the copper or
cuprate reagent are used instead of one (compare entries 1
and 2, and 5–8). We also excluded a radical mechanism in
these additions, as when methylcopper was added to a mix-
ture of 1b and isopropyliodide, no addition product of the
isopropyl radical that could be generated under these condi-
tions[29a,b] was observed and 14a is obtained in high yield. To
gain further insight into this reactivity, a competitive experi-
ment was run (Scheme 8). Thus, an equimolar mixture of 1b
and cyclohexenone was treated with one equivalent of the
methylcuprate reagent, however, only adduct 14b was isolat-
ed, thus suggesting much faster addition on the alkylidene
bisACHTUNGTRENNUNG(sulfinyl) substrate compared to that of cyclohexenone.
As a control experiment, the methylcuprate was added effi-
ciently to cyclohexenone to afford 16.
Importantly, adducts 13–15 were obtained as single diaste-
reomers (Table 4). The relative configuration of adducts 14a
and 14e was determined by derivatization.[9a] In the case of
14a,[22b] it was confirmed by X-ray crystallography analysis,
Figure 1. X-Ray structures of 14a and 14b.
Chem. Asian J. 2011, 6, 1825 – 1833
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1829

L. Fensterbank et al.
FULL PAPERS
To account for this inversion in stereoselectivity, we ini-
tially invoked a chelate of the sulfinyl groups with the
copper salt.[9a] All our experiment attempts using different
copper salts to stabilize a copper chelate either failed by
sulfur or oxygen complexation. To understand more precise-
ly if indeed a chelate effect can explain the selectivity, we
ran quantum mechanics calculations on different models.[30]
First, we choose to compare the reactivity between a copper
and cuprate reagent (Scheme 9). The copper model uses a
simple CuMe to describe the organometallic substrate, while
the cuprate model uses two Me2CuLi subunits, following
previous calculations by Nakamura;[31] these results are sum-
marized in Table 5. For the copper model, the barrier for
the observed experimental configuration is around 40 kcal
molÀ1, while the barrier for the other (not observed) config-
uration is much lower (21 kcalmolÀ1). A similar barrier was
observed (around 21 kcalmolÀ1) with the cuprate model,
comparing the not observed and the experimental configura-
tion. Therefore, we can exclude these models for further in-
vestigations of the mechanism.
We can notice from these calculations that when lithium
is complexed between the two oxygen atoms of the bis(sulf-
oxide) (cuprate case, in the con-
Scheme 9. Stereoselectivity model.
figuration not observed experi-
mentally), the structure is much
more stable. To elucidate the
Table 5. Relative energies for addition of copper and cuprate reagents calculated at the B3LYP/LACVP(d)
level, including ZPC. See Scheme 9 for a description of the chemical system.
mechanism and the stereoselec- Model system
Reactant
Transition State
Product
tivity, this kind of chelation
must be taken into account
(leading to structures A-X and
A-X’). As it is known that cup-
rates are monomeric in a tetra-
hydrofuran solution,[32] we de-
Exp.
Non Exp
Exp.
21.43
0.00
0.00
0.12
44.16
20.77
39.56
21.37
À24.72
À27.26
À2.27
Cuprate
Copper
Non Exp.
À24.40
cided to use only one cuprate to study the reactivity and
add specific solvent effects, namely, by adding a molecule of
dimethylether or tetrahydrofuran. Results are summarized
in Figure 2. It can be seen that the gap between the two con-
figurations is reduced and the barriers are lower (ca. 15 kcal
molÀ1) than in our previous models. Without taking into ac-
Figure 2. Energetic profiles for the addition or organocuprate reagent on alkylidene bis(sulfoxides), calculated at the B3LYP/LACVP(d) level. Energies
are in kcalmolÀ1. All three value sets corresponding to each calculated structure are given in the following order, from top to bottom: without ligand,
with dimethyether, with THF. Italic: without entropy and bold: with entropy.
1830
www.chemasianj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 1825 – 1833

Conjugate Additions
count entropy, the not observed configuration is still fa-
vored, with or without one solvent molecule complexed to
lithium. When entropy is included, we observed a selectivity
inversion. Barriers are slightly favored: 15.44 vs. 16.75 kcal
molÀ1 without solvent, 16.43 vs. 17.22 kcalmolÀ1 with dime-
thylether and 15.46 vs. 15.76 kcalmolÀ1 with tetrahydrofuran
(compare TSAB-X vs. TSAB-Xꢁ in Figure 2). We can attri-
bute this entropy effect to the fact there is one more degree
of freedom in the experimental configuration, as there is no
bonding between copper and lithium (each of them being
on the opposite face of the alkylidene bis(sulfoxides)).
After these encouraging results showed that the stereose-
lectivity can be different depending of the presence or ab-
sence of solvent, we decided to include a second molecule
of solvent, namely, tetrahydrofuran (Figure 3).
diastereoisomer (B-2THF) and a barrier of 20.82 kcalmolÀ1
for the other. Also, the product is more stable when the lith-
ium is on the opposite face of copper, as in this case tetrahy-
drofuran molecules can interact well with the cation to sta-
bilize it.
In summary, DFT calculations suggest that a copper re-
agent is not likely to be the alkylating agent in solution. A
cuprate model is found to be more favored. This could be
corroborated by the fact that better yields are obtained
when two equivalents of copper reagent are used. Lithium
chelation of the alkylidene bis(sulfoxides) together with ex-
plicit solvent effect are found to be necessary to account for
the observed diastereoselectivity.
Conclusions
In conclusion, we have shown
that alkylidene bis(sulfoxides)
of type 1 are remarkable Mi-
chael acceptors towards vari-
ous anionic nucleophiles. Very
good yields and complete ste-
reoselectivities are observed in
most cases with heteronucleo-
philes as well as copper re-
agents, the latter yielding re-
versed facial selectivity, which
could be rationalized by DFT
calculations.
Experimental Section
General Procedure for the Addition
of Sodium Alkoxides to Alkylidene
Bis(Sulfoxides)
At 08C, NaH (60% suspension in
mineral oil, 1.5 equiv) was added to a
solution of alcohol (1.5 equiv) in
THF (4.5 mLmmolÀ1). The solution
Figure 3. Energetic profiles for the addition of organocuprate reagent on alkylidene bis(sulfoxides) with two
THF molecules, calculated at the B3LYP/LACVP(d) level. Italic values correspond to energies with ZPC in-
was stirred for 5 min, cooled to
cluded. Bold values also include entropy. Energies are in kcalmolÀ1
.
À458C, and a solution of alkylidene
bis(sulfoxide)
1 (1 equiv) in THF
(6.5 mLmmolÀ1) was canulated. The
reaction mixture was stirred at
Results with two tetrahydrofuran molecules are summar-
ized in Figure 3. The starting alkylidene-cuprate complex
(A-2THF) that gives finally the observed product is more
stable by approximately 6 kcalmolÀ1. It should be noted that
A-2THF can be seen as a CuI or as a CuIII complex, as two
mesomeric structures can be drawn to describe the copper-
alkylidene interaction.[31] The stability of A-2THF is mainly
due to the fact that the second tetrahydrofuran molecule
can be close to the lithium atom in that case, whereas when
both lithium and copper are on the same side of the alkyli-
dene bis(sulfoxides), some unfavorable steric interaction is
developed with tetrahydrofuran. As a result, we computed a
barrier of 16.61 kcalmolÀ1 for the experimentally observed
À458C for 14 h and quenched with an aqueous saturated solution of
NH4Cl. Then it was diluted with CH2Cl2, washed with water and brine,
dried over MgSO4, filtered, and concentrated in vacuo. The residue was
purified by silica gel chromatography to afford the corresponding product
4.
General Procedure for the Addition of Lithium and Grignards reagents to
Alkylidene Bis(Sulfoxides)
At À788C, a solution of the organometallic reagent (1 equiv) was added
to
a solution of alkylidene bis(sulfoxide) 1 (1 equiv) in THF
(10 mLmmolÀ1). These reagents are used from the commercial solutions:
CH3MgBr (3m) in diethyl ether, CH3MgCl (3m) in THF, CH3MgI (0.5m)
in toluene, Bu2Mg (1m) in THF, CH3Li (1.6m) in diethyl ether. After
5 min at À788C, the reaction mixture turned to an orange-brown color
and an aqueous saturated solution of ammonium chloride was added.
Chem. Asian J. 2011, 6, 1825 – 1833
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1831

L. Fensterbank et al.
FULL PAPERS
The crude mixture was extracted with dichloromethane and the organic
phase was successively washed with water and brine, dried over magnesi-
um sulfate. The solvents were removed under vacuum and the crude was
purified by silica gel chromatography (pentane/ethyl acetate from 80:20
to 0:100). Two fractions were obtained containing the vinyl sulfoxide 11
and the sulfoxide 12.
recent use of bis-p-tolylsulfinyl substrates in organocatalysis, see: I.
Fernꢄndez, V. Valdivia, M. Pernꢆa Leal, N. Khiar, Org. Lett. 2007, 9,
2215–2218; d) For recent applications of chiral bis-p-tolylsulfinyl
substrates as ligands for platinum, see: R. Martinez Mallorquin, S.
Chelli, F. Brebion, L. Fensterbank, J.-P. Goddard, M. Malacria, Tet-
rahedron: Asymmetry 2010, 21, 1695–1700.
[6] An initial example of highly diastereoselective Michael addition has
been reported by Aggarwal in the study of the epoxidation of a sub-
strate of type B (R=Ph) with H2O2, NaOH, see: V. K. Aggarwal,
J. K. Barrell, J. M. Worrall and R. Alexander, J. Org. Chem. 1998,
63, 7128–7129.
General Procedure for the Addition of Copper and Cuprate Reagents on
Alkylidene Bis(Sulfoxides)
(MeCu, LiI) and (Me2CuLi, LiI) were prepared according the following
procedure: MeLi (1.6m in diethyl ether, 1 equiv for the organocopper
and 2 equiv for the cuprate) was slowly added to a THF (2 mLmmolÀ1
)
[7] T. Wedel, J. Podlech, Synlett 2006, 2043–2046.
[8] T. Wedel, J. Podlech, Org. Lett. 2005, 7, 4013–4015.
suspension of CuI (1 equiv) at 08C. Stirring was continued for 30 min at
this temperature. The solution became yellow. (EtCu, LiI) and (nBuCu,
LiI) were prepared according the following procedure: EtLi (0.5m in
benzene/cyclohexane 9:1, 1 equiv) or nBuLi (2.5m in hexanes, 1 equiv)
was slowly added to a THF (2 mLmmolÀ1) suspension of CuI (1 equiv) at
À508C; stirring was continued for 30 min at this temperature.
[9] a) F. Brebion, B. Delouvriꢀ, F. Nꢄjera, L. Fensterbank, M. Malacria,
J. Vaissermann Angew. Chem. 2003, 115, 5500–5503; Angew. Chem.
Int. Ed. 2003, 42, 5342–5345; Angew. Chem. Int. Ed. 2003, 42, 5342–
5345; b) F. Brebion, J.-P. Goddard, C. Gomez, L. Fensterbank, M.
Malacria, Synlett 2006, 713–716; c) F. Brebion, J.-P. Goddard, L.
Fensterbank, M. Malacria, Synthesis 2005, 2449–2452; d) J.-P. God-
dard, C. Gomez, F. Brebion, S. Beauviꢇre, L. Fensterbank, M. Mal-
acria, Chem. Commun. 2007, 2929–2931; e) A. Abramovitch, L.
Fensterbank, M. Malacria, I. Marek, Angew. Chem. 2008, 120, 6971–
6974; Angew. Chem. Int. Ed. 2008, 47, 6865–6868; f) For Michael
conjugate radical addition, see: F. Brebion, M. Vitale, L. Fenster-
bank, M. Malacria, Tetrahedron: Asymmetry 2003, 14, 2889–2896.
[10] For important references on the Michael addition to vinylsulfoxides,
see: a) M. C. CarreÇo, Chem. Rev. 1995, 95, 1717–1760; b) G. Han-
Addition onto Alkylidene Bis(Sulfoxides)
At À788C a solution of alkylidene bis(sulfoxide) 1 (1 equiv) in THF
(4 mLmmolÀ1) was canulated into a solution of the copper or cuprate re-
agent (1 or 2 equiv) in THF (2 mLmmolÀ1). After 30 min at À788C, the
reaction mixture was allowed to warm to À308C for another 30 min, then
it was diluted in Et2O and quenched with an aqueous saturated solution
of ammonium chloride. The organic layer was successively washed with
water and brine, dried over MgSO4, and concentrated in vacuo. The resi-
due was purified by silica gel chromatography to afford the correspond-
ing compounds 13, 14, or 15.
quet, F. Colobert, S. Lanners, G. Solladiꢀ, Arkivoc 2003,
401; c) P. Łyz˙wa, A. Jankowiak, M. Kwiatkowska, M. Mikołajczyk,
ACHTUNGTRENNUNG(vii), 328–
´
P. Kiełbasinski, A. Betz, P.-A. Jaffres, A.-C. Gaumont, M. Gulea,
All the experimental details and the characterization data of the new
compounds are reported in the Supporting Information.
Tetrahedron Lett. 2007, 48, 351 and references therein.
[11] a) S. Colonna, V. Pironti, J. Drabowicz, F. Brebion, L. Fensterbank,
M. Malacria, Eur. J. Org. Chem. 2005, 1727–1730; b) Vanadium-cat-
alyzed asymmetric oxidation has furnished promising results, M.
Vitale, L. Fensterbank, M. Malacria, unpublished results, see also:
c) J. Skarzewski, E. Ostrycharz, R. Siedlecka, Tetrahedron: Asymme-
try 1999, 10, 3457–3461; d) C. Bolm, F. Bienewald, Angew. Chem.
1995, 107, 2883–2885; Angew. Chem. Int. Ed. Engl. 1995, 34, 2640–
2642.
Acknowledgements
We thank UPMC, CNRS, IUF, ANR (grant BLAN0309, Radicaux Verts)
for funding. We thank Patrick Herson, Lise-Marie Chamoreau for the X-
ray structure determination of 9a, 14a, and 14b, and Prof. Eiichi Naka-
mura for insightful discussion.
[12] B. Delouvriꢀ, F. Nꢄjera, L. Fensterbank, M. Malacria, J. Organomet.
Chem. 2002, 643–644, 130–135.
[13] a) Ref. [9c]; b) M. Reggelin, H. Weinberger, V. Spohr, Adv. Synth.
Catal. 2004, 346, 1295–1306.
[14] a) L. F. Tietze, A. Schuffenhauer, P. R. Schreiner, J. Am. Chem. Soc.
1998, 120, 7952–7958; b) For similar directing effect of the sulfinyl
moiety, see also: G. Sklute, I. Marek, J. Am. Chem. Soc. 2006, 128,
4642–4649.
[1] a) J.-L. Reymond, P. Vogel, Tetrahedron: Asymmetry 1990, 1, 729–
736; b) W. R. Roush, B. B. Brown, J. Org. Chem. 1992, 57, 3380–
3387; c) J. Martynow, M. Dimitroff, A. G. Fallis, Tetrahedron Lett.
1993, 34, 8201–8204; d) R. Lꢃpez, J. C. Carretero, Tetrahedron:
Asymmetry 1991, 2, 93–96.
[15] T. Wedel, T. Gehring, J. Podlech, E. Kordel, A. Bihlmeier, W. Klop-
per, Chem. Eur. J. 2008, 14, 4631.
[2] For a review on bis(sulfoxides), see: B. Delouvriꢀ, L. Fensterbank,
F. Nꢄjera, M. Malacria, Eur. J. Org. Chem. 2002, 3507–3525.
[3] For recent reviews on the use of sulfoxides in asymmetric synthesis,
see: a) I. Fernꢄndez, N. Khiar, Chem. Rev. 2003, 103, 3651–3705;
b) H. Pellissier, Tetrahedron 2006, 62, 5559–5601; c) M. C. CarreÇo,
G. Hernꢄndez-Torres, M. Ribagorda, A. Urbano, Chem. Commun.
2009, 6129–6144.
[4] a) N. Katagiri, S. Ise, N. Watanabe, C. Kaneko, Chem. Pharm. Bull.
1990, 38, 3242–3248; b) V. K. Aggarwal, M. Lightowler, S. D. Lin-
dell, Synlett 1992, 730–732; c) V. K. Aggarwal, Z. Gꢅltekin, R. S.
Grainger, H. Adams, P. L. Spargo, J. Chem. Soc. Perkin Trans. 1
1998, 2771–2781; d) V. K. Aggarwal, R. S. Grainger, H. Adams and
P. L. Spargo, J. Org. Chem. 1998, 63, 3481–3485; e) V. K. Aggarwal,
S. Roseblade and R. Alexander, Org. Biomol. Chem. 2003, 1, 684–
691; f) see also: O. De Lucci, D. Fabbri, V. Lucchini, Synlett 1991,
565–568 and Ref. [2] for other references; see also: g) S. Yoshida,
H. Yorimitsu, K. Oshima, Org. Lett. 2009, 11, 2185; h) T. Kobatake,
S. Yoshida, H. Yorimitsu, K. Oshima Angew. Chem. 2010, 122,
2390–2393; Angew. Chem. Int. Ed. 2010, 49, 2340–2343.
[16] Additions of amines to simple vinylsulfoxides are slower and less se-
lective, see: a) S. G. Pyne, P. Bloem, S. L. Chapman, C. E. Dixon, R.
Griffith, J. Org. Chem. 1990, 55, 1086–1093; b) D. J. Abbott, S. Co-
lonna, J. M. Stirling, J. Chem. Soc. Perkin Trans. 1 1976, 492–498.
[17] M. B. Smith, J. March in Advanced Organic Chemistry, 5th ed.,
Wiley, 2001, New York, p. 1199.
[18] a) N. Khiar, I. Fernandez, F. Alcudia, Tetrahedron Lett. 1993, 34,
123–126; b) see also: Ref. [9d].
[19] a) B. M. Trost, A. J. Bridges, J. Org. Chem. 1975, 40, 2014–2016;
b) For an application of the sulfinic acid elimination in the diaste-
reoselective synthesis of vinylsulfoxides, see: F. Brebion, J.-P. God-
dard, L. Fensterbank, M. Malacria, Org. Lett. 2008, 10, 1917–1920.
[20] C. Gomez, J.-P. Goddard, M. Malacria, L. Fensterbank, unpublished
results.
[21] For epoxidation, see: Ref. [6].
[22] The following CCDC files contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif: a) 9a: CCDC 777833; b) 14a:
CCDC 779148; c) 14b: CCDC 779149.
[5] a) Y. Arai, S. Kuwayama, Y. Takeuchi, T. Koizumi, Synth. Commun.
1986, 16, 233–244; b) J. C. Carretero, J. L. Garcꢆa Ruano, L. M.
Martꢆn Cabrejas, Tetrahedron: Asymmetry 1997, 8, 409–416; c) For
1832
www.chemasianj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 1825 – 1833

Conjugate Additions
[23] a) F. Fringuelli, F. Pizzo, L. Vaccaro, J. Org. Chem. 2004, 69, 2315;
b) Z. Dong, K. A. Hellmund, S. G. Pyne, Aust. J. Chem. 1993, 46,
1431.
[24] a) F. Brebion, F. Nꢈjera, B. Delouvriꢀ, E. Lacꢉte, L. Fensterbank,
M. Malacria, Synthesis 2007, 2273–2278; see also the SPAC reaction
(Sulfoxide Piperidine Aldehyde Condensation): b) V. Guerrero-de
la Rosa, M. OrdoÇez, F. Alcudia, J. M. Llera, Tetrahedron Lett. 1995,
36, 4889–4892; c) K. Burgess, J. Cassidy, I. Henderson, J. Org.
Chem. 1991, 56, 2050–2058.
[25] For organometallics addition to vinylsulfoxides, see: a) J. P. Marino,
A. de Dios, L. J. Anna, R. Fernndez de La Pradilla, J. Org. Chem.
1996, 61, 109–117; b) J. P. Marino, A. Viso, R. Fernꢄndez de La Pra-
dilla, P. Fernꢄndez, J. Org. Chem. 1991, 56, 1349–1351. For copper
reagent additions, see: c) E. Nakamura, N. Yoshikai in The Chemis-
try of Organocopper Compounds (Ed. Z. Rappoport I. Marek),
Wiley New-York, 2009, pp. 1–23; d) S. Mori, E. Nakamura in
Modern Organocopper Chemistry (Ed.: N. Krause), Wiley-VCH,
Weinheim, 2002, pp. 315–346; e) W. Henze, T. Gꢊrtner, R.
Gschwind, J. Am. Chem. Soc. 2008, 130, 13718–13726; f) B. Breit, Y.
Schmidt, Chem. Rev. 2008, 108, 2928–2951.
2006, 45, 7600–7603; Angew. Chem. Int. Ed. 2006, 45, 7600–7603;
c) S. Bazin, L. Feray, N. Vanthuyne, M. P. Bertrand, Tetrahedron
2005, 61, 4261–4274; d) F. Denes, S. Cutri, A. Pꢀrez-Luna, F.
Chemla, Chem. Eur. J. 2006, 12, 6506–6513; e) Conjugate Addition
of Organometallic Reagents, K. Tomioka, Y. Nagaoka in Compre-
hensive Asymmetric Catalysis (Eds.: E. N. Jacobsen, A. Pfaltz, H.,
Yamamoto), Springer, New York, 1999, 3, 1105–1120; f) B. L. Ferin-
ga, R. M. Kellog, H. van der Deen, Org. Lett. 2000, 2, 1593–1595;
g) T. Cohen, H. Gibney, R. Ivanov, E. A.-H. Yeh, I. Marek, D. P.
Curran, J. Am. Chem. Soc. 2007, 129, 15405–15409.
[30] Computational methods: All the structures were calculated by DFT
using the hybrid functional B3LYP, and the double-zeta basis set,
LACVP(d), which includes effective core potential for heavier ele-
ments like Cu. In geometry optimization procedures, we use
JAGUAR 6.5, and subsequently employ GAUSSIAN03/09 for ana-
lytic frequency calculations. Reaction pathways were verified by a
scan along a given coordinate, while optimizing freely all other coor-
dinates. The data are summarized is the Supporting Information
(SI) deposited with this paper, while the text focuses on the key re-
sults.
[26] For sulfoxide-metal exchange, see Ref. [9e] and references therein.
[27] a) S. House, P. R. Jenkins, J. Fawcett, D. R. Russell, J. Chem. Soc.
Chem. Commun. 1987, 1844–1845; b) J. Fawcett, S. House, P. R. Jen-
kins, N. J. Lawrence, D. R. Russell, J. Chem. Soc. Perkin Trans. 1
[31] a) N. Yoshikai, S.-L. Zhan, E. Nakamura, J. Am. Chem. Soc. 2008,
130, 12862–12863; S. Mori, E. Nakamura, Chem. Eur. J. 1999, 5,
1534–1543; b) E. Nakamura, S. Mori, K. Morokuma, J. Am. Chem.
Soc. 1997, 119, 4900–4910; c) E. Nakamura, S. Mori, Angew. Chem.
2000, 112, 3902–3924; Angew. Chem. Int. Ed. 2000, 39, 3750–3771.
[32] A. Gerold, J. T. B. H. Jastrzebski, C. M. P. Kronenburg, N. Krause,
G. von Koten, Angew. Chem. 1997, 109, 778–780; Angew. Chem. Int.
Ed. Engl. 1997, 36, 755–757.
´
1993, 67–73; c) M. Mikołajczyk, W. Perlikowska, J. Omelanczuk,
H.-J. Cristau, A. Perraud-Darcy, J. Org. Chem. 1998, 63, 9716–9722.
[28] The determination of configuration is based by comparison with re-
sults obtained in the copper reagent series (see Table 4, line 4)
which presumably gives 13am as a major diastereomer.
[29] a) H. O. House, Acc. Chem. Res. 1976, 9, 59–67; b) K. Li, A. Alexa-
kis Angew. Chem. 2006, 118, 7762–7765; Angew. Chem. Int. Ed.
Received: December 16, 2010
Published online: March 11, 2011
Chem. Asian J. 2011, 6, 1825 – 1833
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1833