Synthesis in ionic liquids only: Access to α-oxo-γ-thio-esters via Mukaiyama coupling

Article abstract of DOI:10.1016/j.tetlet.2014.01.024

Ionic liquids are solvents general enough to conduct a multi-step process in organic synthesis. We show that both the preparation of starting materials (thioacetals and enoxysilane) as well as their coupling can be realized in such medium.

Full text of DOI:10.1016/j.tetlet.2014.01.024

Tetrahedron Letters 55 (2014) 1353–1356
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis in ionic liquids only: access to
via Mukaiyama coupling
a-oxo-c-thio-esters
Khouloud Jebri a,b, Marie-Rose Mazières a, Stéphanie Ballereau a, Taïcir Ben Ayed b,
Jean-Christophe Plaquevent a, Michel Baltas a, , Frédéric Guillen a,
⇑
⇑
a Université Paul Sabatier, CNRS-UMR 5068, SPCMIB, 118 route de Narbonne, F-31062 Toulouse Cedex 9, France
b Institut National des Sciences Appliquées et de Technologie (INSAT), Université de Carthage, Tunis, Tunisia
a r t i c l e i n f o
a b s t r a c t
Article history:
Ionic liquids are solvents general enough to conduct a multi-step process in organic synthesis. We show
that both the preparation of starting materials (thioacetals and enoxysilane) as well as their coupling can
be realized in such medium.
Received 13 November 2013
Revised 20 December 2013
Accepted 7 January 2014
Available online 13 January 2014
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Ionic liquid
Mukaiyama reaction
a-Oxo c-thio-ester
Thioacetal
Enoxysilane
Ethyl pyruvate
Owing to their unique properties, ionic liquids (ILs) nowadays
In the first part we summarize the background of this study,
which was initiated because of the puzzling role of the solvent
during the construction of ulosonic derivatives by Mukaiyama
reactions. In the second part are depicted our preliminary results
regarding the SILO approach.
are widely examined as eco-compatible solvents for organic
synthesis.1 Many successful applications have been disclosed,
including modern aspects such as enantioselective,2 catalytic,3 or
multicomponent reactions.4 Nevertheless one can observe that in
most of these reports only one single model reaction is generally
studied (and generally carried out on few model compounds); to
the best of our knowledge, no multi-step or all-step syntheses in
ILs have been described so far.5 We believe that time has come
to this type of research, that is the search for processes in which
no other solvents than ILs are used even for multi-step syntheses
or for the preparation of starting materials. With this ambition in
mind, we recently embarked mainly on two approaches, that is
peptide6 and Mukaiyama couplings, calling this approach ‘synthe-
sis in ILs only’ (SILO).
From the ulosonic puzzle. . .
Our interest for the construction of bioactive materials in
the ulosonic series, via Mukaiyama reaction of ethyl 2-(tri-
methylsilyloxy)acrylate with epoxyaldehydes such as 1,7 led
us to search for alternative media as solvents for this transfor-
mation. Indeed, the role of the solvent in these studies
appeared rather puzzling. For example it has been shown in
our laboratory that
a Lewis acid as catalyst in aqueous
In this Letter we describe our results in the field of Mukaiyama
medium (sodium dodecyl sufate, SDS) led to aldol derivatives
without any oxirane-ring opening (Scheme 1),8 while chlorinated
solvents mainly afforded products arising from ring expansion.7
In the latter case, even the workup procedure could exert
strong influence, giving in some cases access to fluorinated
intermediates such as compound 2 as depicted in Scheme 1.9
In order to avoid any fluorinated products, it would be worthwhile
to find either catalyst-free conditions or to replace BF3ÁEt2O by
another catalyst. Unfortunately, only this Lewis acid could give
substantial amount of adducts when employed in molecular
solvents.10
couplings leading to
a-oxo c-thio-esters, where both partners of
the Mukaiyama reaction, that is thioacetals and pyruvate enoxysi-
lane, are prepared in ionic liquids. Their coupling is also realized in
such a solvent. Thus, the full synthetic pathway does not require
any other solvents than ionic liquids, consisting in an unprece-
dented approach in organic synthesis.
⇑
Corresponding authors.
E-mail addresses: baltas@chimie.ups-tlse.fr (M. Baltas), guillen@chimie.
ups-tlse.fr (F. Guillen).
0040-4039/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2014.01.024

1354
K. Jebri et al. / Tetrahedron Letters 55 (2014) 1353–1356
Scheme 1. Construction of ulosonic structures by Mukaiyama couplings. Molecular view of isopropylidene derivative of compound 2 in the solid state (thermal ellipsoids at
50% probability); hydrogen atoms are omitted for clarity excepted on asymmetric carbons. Crystallographic data (excluding structure factors) have been deposited with the
Cambridge Crystallographic Data Centre as supplementary publication no. CCDC-969490.
ILs have already been used either as solvents or promoters for
Mukaiyama couplings; for example it has been shown that aldoli-
zations could be performed at room temperature in the absence of
any activation.11 Also the role of the IL structure was examined and
it was found that better results were observed for cations bearing a
long carbon chain (omim vs hmim) and for chloride anions with re-
spect to hexafluorophosphates.11 More recently, Gaumont et al.
showed that Mukaiyama aldol reactions leading to dihydropyra-
nones were promoted either by ionic liquids or ionic additives.12
In this study, thiazolinium salts were the best catalysts. Interest-
ingly, only the Mukaiyama process was observed without any het-
ero Diels-Alder cycloaddition.12 Inspired by these results, we
examined ionic liquids as solvents for the same type of transforma-
tions from the cis-epoxyaldehyde depicted in Scheme 1. New
materials were obtained as traces that were not identified.
Although not satisfactory from a preparative point of view, this re-
sult incited us to focus on this IL strategy since full conversion was
observed without the need for any catalyst. The major aims were to
find catalyst-free conditions (or to avoid the use of BF3ÁEt2O), to
improve the overall yield of the transformations, and eventually
to realize all steps in ILs.
example CuCl, CuBr, CuI, Pd(acac)2 et Co(acac)3 were successfully
used in acetonitrile.14 Also, Sc(OTf)3 was immobilized in imidazo-
lium ILs such as [bmim][PF6] and [bmim][BF4], and used as catalyst
for the preparation of thioacetals with better performance than in
molecular solvents such as dichloromethane (faster reaction rate
as well as better yields).15 We used this last method to obtain
the targeted thioacetals (Table 1). We selected two commercial
hydrophobic ionic liquids,16 [bmim][PF6] and [emim][NTf2], as
model solvents for this study, for their low melting point, viscosity,
and hygroscopy. We did not consider hydrophilic ILs possessing
halide anions that are known as non-innocent species because of
their nucleophilic and reducing properties.
A series of para-substituted benzaldehydes was transformed
into the corresponding thioacetals with good performance. The
reaction is rather fast for electroenriched aromatic rings and gives
76–99% yield after purification by column chromatography. Reac-
tion rate is lower for electron poor compounds as well as for the
dimethylamino substituted substrate, for which complexation
of the Lewis acid by the nitrogen lone pair occurs. Even a more
sterically demanding compound such as 1-naphtaldehyde was
converted to the corresponding thioacetal in 65% yield. Aliphatic
and conjugated starting materials are also conveniently protected
by this method. As previously mentioned, we verified that the
transformation was less efficient in organic molecular solvents.
To ionic liquid solution?
We focused on a strategy recently published by Bolm,13 who de-
scribed the synthesis of
a-oxo-c-thio-esters by reaction of ethyl
Table 1
pyruvate enoxysilane onto thioacetals. Aldolizations were carried
out at 0 °C in dichloromethane, in the presence of scandium triflate
as catalyst, with best results as high as 62%.13 We thus embarked
on a project in which both precursors as well as the reaction itself
would be realized in ionic liquids (Fig. 1).
Synthesis of thioacetals in ILs in the presence of Sc(OTf)3
Sc(OTf)3 (2 mol%),
EtSH (2.2 eq.)
SEt
SEt
R
O
Ionic Liquid, rt
R
Aldehyde
X
Ionic liquid
Reaction time (yield)
Synthesis of thioacetals
X = H
[bmim][PF6]
[emim][NTf2]
10 min (78%)
10 min (99%)
O
X
Thioacetals are generally prepared by condensation of a parent
carbonyl compound with thiols in the presence of a Lewis acid. For
X = OMe
X = Me
[bmim][PF6]
[emim][NTf2]
[bmim][PF6]
[emim][NTf2]
[bmim][PF6]
[emim][NTf2]
[bmim][PF6]
[bmim][PF6]
[emim][NTf2]
10 min (98%)
10 min (85%)
60 min (76%)
20 min (84%)
105 min (79%)
>16 ha (55%)
60 min (48%)
240 min (25%)
240 min (33%)
X = NO2
catalysis
SR'
in
R
X = COMe
X = NMe2
O
R
SR'
catalysis
in
ionic liquid
Mukaiyama
adduct
electrophilic
masked aldehyde
ionic
O
[bmim][PF6]
[emim][NTf2]
60 min (65%)
45 min (82%)
liquid
O
ethyl
OEt
pyruvate
in ionic
liquid
OTMS
Ph
[bmim][PF6]
[emim][NTf2]
[bmim][PF6]
15 min (99%)
20 min (82%)
30 min (89%)
O
nucleophilic
enoxysilane
t-BuCHO
a
Reaction was not complete after 16 h.
Figure 1. General strategy.

K. Jebri et al. / Tetrahedron Letters 55 (2014) 1353–1356
1355
We believe that the use of [bmim][PF6] and [emim][NTf2] as sol-
Sc(OTf)3
(2 mol%)
SEt
vents, which are hydrophobic ILs, has the advantage to remove
water from the reaction mixture, thus preventing any subsequent
hydrolysis of the formed thioacetal. Although significant differ-
ences in yield and/or reaction rates could be observed between
reactions run in [emim][NTf2] and in [bmim][PF6] with several
substrates, no clear trend could be established.
O
HO
SEt
SEt
OTBDPS
EtSH (2.2 eq.)
O
TBDPSO
[emim][NTf2],
rt
1
68%
Scheme 3. Thioacetalization of 1 in IL in the presence of Sc(OTf)3.
We also examined CuBr, a cheaper and less toxic catalyst than
Sc(OTf)3, in the same transformation on benzaldehyde. However
in these conditions the reaction requires a longer duration and
the yield is significantly lower (Scheme 2). This can be easily ratio-
nalized using HSAB theory: Cu(I), which is a much softer acid than
Sc(III), interacts not only with the oxygen of the carbonyl group
(hard base), but also with the sulfur atom of ethanethiol (soft base),
thus slowing down the thioacetalization reaction.
Finally, the epoxyaldehyde 1 was used as substrate in the
thioacetalization reaction in ionic liquid. However, concomitant
nucleophilic opening of the epoxide by ethanethiol was consis-
tently observed, giving access to a highly functionalized carbohy-
drate derivative in rather good yield and in pure diastereomeric
form (Scheme 3).
O
O
BSA (1.2 eq.)
OEt
OEt
Ionic Liquid,
100°C, 4h
O
OTMS
in [NBu4][Br]
in [omim][NTf2]
in [emim][NTf2]
62 %
48 %
38 %
Scheme 4. Synthesis of ethyl pyruvate enoxysilane in ILs.
SEt O
OEt
Ar
Sc(OTf)3
(0.1 eq.)
SEt
SEt
O
O
O
A
Synthesis of ethyl pyruvate enoxysilane
+
Ar
OEt
O
[emim][NTf2],
0°C, 5h
OTMS
3 eq.
OEt
Ar
This compound has been prepared in 1992, starting from ethyl
pyruvate.17 More recently, Smietana and Mioskowski described a
new method for the preparation of enoxysilanes in ILs.18 In the
presence of bis-trimethylsilylacetamide (BSA), aldehydes and ke-
tones are transformed into their corresponding silyl enol ether in
ionic medium such as [NBu4][Br]. The reaction is carried out at
the mp of the IL (ca 100 °C) and requires heating for 4 h to give
the target enoxysilanes in 75–90% yield. We decided to adapt this
method to ethyl pyruvate, and were delighted to obtain the corre-
sponding enoxysilane in rather good yield (Scheme 4, 62% after
purification by bulb-to-bulb distillation). This method presents
some advantages like neutral conditions (no need to add base such
as DMAP), simplicity, and atom economy.
B
A B
Ar = Phenyl, 66 %,
/
= 74/26
= 0/100
= 0/100
= 40/60
A B
Ar = p-MeO-Phenyl, 30%,
Ar = p-Me-Phenyl, 22%,
Ar = 1-Naphthyl, 30%,
/
A B
/
A B
/
Scheme 5. Mukaiyama couplings in ILs.
Two main compounds were obtained in our experiments. We
were delighted to obtain the expected coupling adduct A starting
from benzaldehyde and naphtaldehyde derivatives, along with
the corresponding elimination product B. The latter was the only
product from p-methoxy and p-methyl congeners, while p-nitro
derivative yielded a complex and inseparable mixture, in which
both A and B were detected by NMR. One can explain the reactivity
by electronic effects, electron donating groups (MeO and Me) sta-
bilizing the carbocationic form thus facilitating the elimination
process. Nevertheless, both compounds A and B come from the
same coupling reaction, thus proving that [emim][NTf2] is a conve-
nient solvent for such a synthetic step. We are currently focusing
on conditions which could improve the chemoselectivity of this
last step.
Imidazolium based ionic liquids16 [omim][NTf2] and
[emim][NTf2] were also assessed as solvents in this reaction, and
were found to give the target enoxysilane with acceptable, if
somewhat lower, yields. This observation is crucial for our future
studies, since we plan to realize the whole synthetic pathway
(i.e. preparation of both precursors as well as the coupling
reaction) in the same ionic solvent. The better yield obtained in
[NBu4][Br] could be due to reaction of bromide ion with BSA giving
a transient highly reactive bromosilane.
Mukaiyama couplings
In conclusion, we show that ILs are solvents general enough to
conduct a multi-step process in organic synthesis. Both the prepa-
ration of starting materials (thioacetals and pyruvate enoxysilane)
and their coupling are realized in such medium. Figure 2 summa-
rizes the whole synthetic process for the best model substrate.
Finally, we focused on the coupling in ionic medium of the pre-
viously obtained pyruvate enoxysilane with thioacetals. After
checking various ionic solvents, we observed that the pyruvate
enoxysilane was not stable in the presence of BFÀ4 and PF6À anions
(presumably because of trace amounts of HF produced by the
hydrolysis of BFÀ4 or PF6À), but could be used in NTfÀ2 containing
ILs. As imidazolium cations proved to be convenient for many or-
ganic reactions, we decided to test those couplings in
[emim][NTf2]. Our main results are depicted in Scheme 5.
SEt
SEt
Sc(OTf)3 cat.
EtSH
O
SEt O
49%
Sc(OTf)3
cat.
ionic liquid
R
R
OEt
up to 99%
O
ionic liquid
CuBr (5 mol%),
O
O
SEt
BSA
ionic liquid
EtSH (2.2 eq.)
OEt
O
OEt
SEt
O
OTMS
[emim][NTf2], rt, 60 min
72%
up to 62 %
Scheme 2. Synthesis of thioacetals in IL in the presence of CuBr.
Figure 2. SILO for the preparation of a-oxo c-thio-esters.

1356
K. Jebri et al. / Tetrahedron Letters 55 (2014) 1353–1356
Plaquevent, J.-C. In Catalytic Methods in Asymmetric Synthesis: Advanced
The full synthetic pathway described herein does not require
Materials, Techniques, and Applications, Gruttadauria, M.; Giacalone F. Eds.,
Wiley, 2011; Chapter 7.
any other solvents than ionic liquids, consisting in an unprece-
dented approach in organic synthesis. The search for multi-step
processes avoiding the use of volatile organic solvents is of grow-
ing importance,19 and we believe that our approach can give new
insights in this field. Studies are currently underway to perform
the entire synthetic sequence without extraction of the intermedi-
ate compounds from the ionic liquid.
3. Durand, J.; Teuma, E.; Gomez, M. C. R. Chimie 2007, 152–177.
4. Isambert, N.; del Mar Sanchez Duque, M.; Plaquevent, J.-C.; Génisson, Y.;
Rodriguez, J.; Constantieux, T. Chem. Soc. Rev. 2011, 40, 1347–1357.
5. For one-pot syntheses in ILs, see for example: (a) Forsyth, S. A.; Gunaratne, H.
Q. N.; Hardacre, C.; McKeown, A.; Rooney, D. W. Org. Proc. Res. Dev. 2006, 10,
94–102; (b) Ono, F.; Qiao, K.; Tomida, D.; Yokoyama, C. Appl. Catal. A: General
2007, 333, 107–113; (c) For a recent review on multicomponent reactions in
ILs, see Ref. 4.
6. For part
1 of this series, see Galy, N.; Mazières, M.-R.; Plaquevent, J.-C.,
Acknowledgement
Tetrahedron Lett. 2013, 54, 2703–2705. For our previous work in the field of
peptide synthesis in ILs, see: (a) Vallette, H.; Ferron, L.; Coquerel, G.; Gaumont,
A.-C.; Plaquevent, J.-C. Tetrahedron Lett. 2004, 45, 1617–1619; (b) Vallette, H.;
Ferron, L.; Coquerel, G.; Guillen, F.; Plaquevent, J.-C. Arkivoc 2006, 200–211; (c)
Guillen, F.; Brégeon, D.; Plaquevent, J.-C. Tetrahedron Lett. 2006, 47, 1245–1248.
7. Filali, H.; Danel, M.; Baltas, M.; Ballereau, S. Carbohydr. Res. 2010, 345, 2421–
2426.
X-ray analysis was performed by Nathalie Saffon at the X-ray
diffraction Service at the Institute of Chemistry of Toulouse (ICT-
FR CNRS 2599).
8. Ruland, Y.; Noereuil, P.; Baltas, M. Tetrahedron 2005, 61, 8895–8903.
9. The relative stereochemistry of this adduct had previously been assigned by
NMR analyses,7 and recently confirmed by this X-ray structure.
Supplementary data
10. Sugisaki, C. H.; Ruland, Y.; Baltas, M. Eur. J. Org. Chem. 2003, 672–688.
11. Chen, S. L.; Ji, S. J.; Loh, T. P. Tetrahedron Lett. 2004, 45, 375–377.
12. Mercey, G.; Brégeon, D.; Baudequin, C.; Guillen, F.; Levillain, J.; Gulea, M.;
Plaquevent, J.-C.; Gaumont, A.-C. Tetrahedron Lett. 2009, 50, 7239–7241.
13. Krebs, A.; Bolm, C. Tetrahedron 2011, 67, 4055–4060.
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.tetlet.2014.
01.024.
14. Varala, R.; Nuvula, S.; Adapa, S. R. Bull. Kor. Chem. Soc. 2006, 27, 1079–1083.
15. Kamal, A.; Chouhan, G. Tetrahedron Lett. 2003, 44, 3337–3340.
16. [bmim][PF6] (99.5%), [emim][NTf2] (99.5%) and [omim][NTf2] (99.5%) were
purchased from Solvionics (Toulouse, France); [NBu4][Br] (99+%) was
purchased from Acros Organics (Geel, Belgium).
17. Sugimura, H.; Yoshida, K. Bull. Chem. Soc. Jpn 1992, 65, 3209–3211.
18. Smietana, M.; Mioskowski, C. Org. Lett. 2001, 3, 1037–1039.
19. Earle, M. J.; Noè, M.; Perosa, A.; Seddon, K. R. RSC Advances 2014, 4, 1204–1211.
References and notes
1. (a)Ionic Liquids in Synthesis; Wasserscheid, P., Welton, T., Eds., 2nd ed.; Wiley-
VCH: Weinheim, 2008; (b) Hallet, J. P.; Welton, T. Chem. Rev. 2011, 111, 3508–
3576.
2. (a) Baudequin, C.; Plaquevent, J.-C.; Audouard, C.; Cahard, D. Green Chem. 2002,
4, 584–586; (b) Gaumont, A.-C.; Génisson, Y.; Guillen, F.; Zgonnik, V.;