The presence of this product 7 coming from an unknown
rearrangement was unexpected and it intrigued us. Indeed,
to the best of our knowledge, such sulfone rearrangement
had never been observed previously and could lead to inter-
esting gem-disulfone adducts. This lead us to undergo a
complete study on this reaction by starting optimising the
reaction conditions in order to obtain a complete selectivity
in favour of the rearranged product (Table 1).
this catalyst, the cleanest reaction (68% conversion, 40%
NMR spectroscopic yield and 88% ee in 8 h) was in water.
However, this reactivity was limited to this example, since
no reaction at all was observed with isovaleraldehyde as
donor. All these disappointing results lead us to turn our at-
tention to the use of our recently developed aminal–pyrroli-
dine catalysts, which already showed excellent reactivity on
1,1-bisACHTUNGTRNEUNG
(sulfone)vinylenes.[11] To our delight, the three already
published catalysts 9a–9c lead to a cleaner reaction in an
impressive short reaction time (entries 5–7). A short solvent
screening using 10 mol% of the best catalyst 9a and de-
creasing the amount of aldehyde to five equivalents, indicat-
ed that the reaction performed best in toluene or chloro-
form with less then 20% self-aldol product formed (entries 8
and 9). Full conversion was obtained and the elimination
compound 3 was observed in less then 15% using these sol-
vents (entries 8–10).[12] Finally, decreasing the temperature
to À108C considerably slowed down the reaction, while in-
creasing the ee to 81%. To keep an excellent reactivity, a
temperature of 158C was then used in the reaction and
other aldehydes were tested to confirm the generality of this
reaction (Table 2).
Table 1. Catalysts and conditions screening for the addition of aldehyde
to 1,2 bis sulfone 6.[a]
t [h] Conv[b] [%] Yield[c] [%] ee[d] [%]
Interestingly, using catalyst 9a, this rearrangement was
observed with different Michael donors. Indeed, by using
different linear aldehydes, good yields and enantioselectivi-
ties could be obtained in an impressively short time. The
more bulky the aldehyde, the better the yield and enantiose-
lectivity. Indeed, using iPr or tBu as R1 groups, no traces of
the elimination product was observed, while ee up to 87%
could be reached (entries 4 and 5). In contrast, when using
propionaldehyde as Michael donor (R1 =Me), poor conver-
sions were obtained, probably due to the formation of a
large amount of sulfinic acid by the elimination pathway
(result not shown in the table). Changing the sulfone group
to SO2Me instead of SO2Ph also lead to an increased reac-
tivity and excellent yield, while lower ee was observed. This
lower enantioselectivity indicates that the more bulky
SO2Ph must interact strongly with the catalyst in the transi-
tion state, leading to the higher ee. In contrast, the reaction
with the SO2tBu group lead to a slow reaction rate and to a
messy mixture (entry 7). Finally, this process is equally effi-
cient, in terms of yields, when an a-disubstituted aldehyde is
used, while leading to lower enantioselectivity (entry 8). Al-
though low (ee=30%), the ee is still higher then the one al-
Cat Solvent
1
2
3
4
5
6
7
8
9
8a
8b
8c
8d
9a
9b
9c
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
7
24
3
9
3
45
<10
100
25
100
100
100
100
100
100
traces
nd
32
nd
56
26
36
84
83 (71)
77 (49)
92
nd
0
nd
70
43
64
72
70
81
3.5
3
6
4
9a[e] CHCl3
9a[e] toluene
10 9a[e] toluene (À108C) 45
[a] Reactions were performed by using catalyst (20 mol%) and aldehyde
(10 equiv) with bis(sulfone)vinylene (0.1 mmol) in solvent (0.2 mL).
ACHTUNGTRENNUNG
When using 1,4-dioxane as solvent, three equivalents of water as additive
was needed to allow the reaction. [b] Determined by 1H NMR spectros-
1
copy. [c] Determined by H NMR spectroscopy. Isolated yields are shown
in brackets. nd= not determined. [d] Determined by super fluid chroma-
tography. [e] Catalyst (10 mol%) and aldehyde (5 equiv) were used.
The first attempts using commercially available catalysts
were rather disappointing. Indeed, only catalyst 8c lead to
complete conversion, but, unfortunately, gave a racemate
(entry 3). In contrast, diphenylprolinol silyl ether 8a, usually
a powerful catalyst for Michael additions,[8] showed excellent
enantiocontrol (92% ee), but only 45% conversion
(entry 1). Furthermore, with all these catalysts, the reaction
was really messy, leading to many byproducts (such as the
product arising from the sulfinic acid elimination 3). The ab-
solute configuration of the rearranged product was ascribed
by comparison with known compounds arising from the
ready observed in the case of 1,1-bis
Surprisingly, when using the E starting material 2 in DMF
(in order to increase the solubility of bis(sulfone)vinylene),
87% of the Michael adduct 7g could be obtained (entry 9).
The Michael addition of ketones to bis(sulfone)vinylenes
(sulfone)vinylene.[4c]
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
represents an even more challenging task in enamine cataly-
sis. Indeed, using catalyst 9a, the addition of cyclohexanone
A
E
to 1,1-bis
mono- and di-addition product and product arising from the
degradation of the bis(sulfone)vinylene, with poor enantio-
control (Scheme 2). Only recently, Lu et al. described the
addition of six-membered ring ketones to this disulfone,
using a primary amine cinchonidine derivative.[13]
ACHTUNGTREN(NUNG sulfone)vinylene leads to a complex mixture of
in terms of chemical yield failed using catalyst 8a.[9] The use
of chloroform, acidic additives in water (PhCOOH, AcOH),
or 5% ethanol in water as recently used in our group,[10]
lead to no, or dirty, reaction (result not presented). With
ACHTUNGTRENNUNG
11110
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11109 – 11113