Brook/Elimination/Aldol Reaction Sequence for the Direct One-Pot Preparation of Difluorinated Aldols from (Trifluoromethyl)trimethylsilane and Acylsilanes

Article abstract of DOI:10.1002/adsc.201500849

A methodology allowing the one-pot preparation of difluorinated aldols directly from Ruppert-Prakash reagent, acyltrimethylsilanes and aldehydes is reported. The process, initiated by a catalytic amount of an ammonium salt, involves the addition of (trifluoromethyl)trimethylsilane to the acylsilane, followed by a Brook rearrangement and elimination of a fluoride anion that promotes the subsequent aldol reaction. An efficient racemic reaction catalyzed by tetrabutylammonium difluorotriphenylsilicate is described, as well as our first efforts towards an asymmetric version.

Full text of DOI:10.1002/adsc.201500849

COMMUNICATIONS
DOI: 10.1002/adsc.201500849
Brook/Elimination/Aldol Reaction Sequence for the Direct One-
Pot Preparation of Difluorinated Aldols from (Trifluoromethyl)-
trimethylsilane and Acylsilanes
MØlanie Decostanzi,a JØrØmy Godemert,b Sylvain Oudeyer,b Vincent Levacher,b
Jean-Marc Campagne,a and Eric Leclerca,*
a
Institut Charles Gerhardt, UMR 5253, CNRS-UM2-UM1-ENSCM 8, rue de l’Ecole Normale, 34296 Montpellier Cedex 5,
France
Fax: (+33)-4-6714-4322; e-mail: eric.leclerc@enscm.fr
Normandie Univ, COBRA, UMR 6014 & FR3038; Univ Rouen; INSA Rouen; CNRS. IRCOF, 1 Rue Tesni›re, F-76821
b
Mont-Saint-Aignan Cedex, France
Received: September 11, 2015; Revised: November 24, 2015; Published online: February 9, 2016
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201500849.
Abstract: A methodology allowing the one-pot
preparation of difluorinated aldols directly from
Ruppert–Prakash reagent, acyltrimethylsilanes and
aldehydes is reported. The process, initiated by
a catalytic amount of an ammonium salt, involves
the addition of (trifluoromethyl)trimethylsilane to
the acylsilane, followed by a Brook rearrangement
and elimination of a fluoride anion that promotes
the subsequent aldol reaction. An efficient racemic
reaction catalyzed by tetrabutylammonium difluor-
otriphenylsilicate is described, as well as our first ef-
forts towards an asymmetric version.
improve the binding of a molecule with a receptor, in-
crease its lipophilicity, modulate its pKa or inhibit its
metabolic degradation.[3] The number of fluorinated
drugs has consequently increased, as well as the need
for methodologies that enable the efficient prepara-
tion of fluorinated synthons. Great achievements in
the fields of fluorination and trifluoromethylation re-
actions have been realized over the past twenty
years.[4] The use of reactive fluorinated building
blocks, such as fluoroenol ethers or fluoroenolates, is
nevertheless a nice alternative.[5] Indeed, this ap-
proach allows an access to fluoro- or difluoromethy-
lene-containing molecules without the limitations of
direct fluorination reactions. Moreover, the second
functional group which is generally introduced
through such reactions might serve for further syn-
thetic elaboration. Aldol reactions with difluoroeno-
lates are highly representative of this strategy since
a difluoromethylene group and a carbonyl function
Keywords: aldol reaction; Brook rearrangement;
fluorine; organocatalysis
While long ignored by organic and medicinal chem- can be introduced in a single step. The preparation of
ists, fluorinated molecules have increasingly attracted a,a-difluoro-b-hydroxy ketones or esters through
their attention since half a century. The ever-growing either Mukaiyama aldol or Reformatsky-like reac-
number of fluorinated drugs and agrochemicals re- tions is well documented.[6,7] However, Mukaiyama re-
leased on the market every year reflects this situa- actions are often hampered by the poor stability of
tion.[1] This popularity arises from the unique proper- trimethylsilyl fluoroenol ethers, while Reformatsky
À
ties of the fluorine atom and of the C F bond. The reactions suffer from the drawbacks associated with
fluorine atom is the second smallest (van der Waals the use of stoichiometric amounts of organometallic
radius is 1.47 Š) of the periodic table and exhibits the reagents (basic and strictly anhydrous conditions, met-
strongest electronegativity (4.0 on the Pauling scale). allic wastes, …). In this context, the trifluoroacetate
À
The C F bond is consequently short (1.35 Š) and has release strategy initially developed by Colby and
a very high dissociation energy of 105.4 kcalmolÀ1.[2] Wolf, which enables the mild in situ generation of a di-
Thanks to these properties, the introduction of fluo- fluoroenolate from a stable precursor, was a very in-
rine onto a bioactive molecule has often major conse- teresting approach.[8] We were on our part intrigued
quences on its pharmacodynamic or pharmacokinetic by the long known Brook rearrangement/fluoride
properties. Indeed, the presence of a fluorine atom or elimination process, which proved highly helpful for
a fluorinated group in the appropriate position can the preparation of difluoroenoxysilanes (Scheme 1).[9]
526
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2016, 358, 526 – 531

COMMUNICATIONS
Brook/Elimination/Aldol Reaction Sequence
Table 1. Optimization of the one-pot Brook rearrangement/
fluoride elimination/aldol reaction sequence.
Entry
n
Temp.
Method[a]
3:4:PhCHO[b]
1
2
3
4
5
1
1
2
2
2
À408C
À408C
À408C
À408C
À608C
A
B
A
B
B
51:49:0
6:12:82
56:44:0
100:0:0
66:34:0
Scheme 1. Brook/elimination sequence.
This sequence was appealing since it allowed a mild
in situ generation of a difluoroenoxysilane, while gen-
erating a fluoride anion as the sole by-product. We
thus envisioned that such a process could be applied
to a one-pot/one-catalyst aldol reaction directly from
acylsilanes. We wish to report herein our efforts in
[a]
Method A: the PhCHO is introduced at once. Method
B: a THF solution of PhCHO is added in 15 min with
a syringe pump.
[b]
1
Measured by H NMR of the crude mixture.
this area that led to the development of a Brook rear- obtained in the same amount, a simple optimization
rangement/fluoride elimination/aldol reaction se- of the reaction procedure was expected to solve this
quence.
issue. Disappointingly, the addition of benzaldehyde
Inspired by our results in the development of a simi- in 15 min using a syringe pump to a THF solution of
lar sequence leading to monofluorinated aldols from CF3TMS and CH3COTMS led to a very poor conver-
b,b-difluoro-a-(trimethylsilyl) alcohols,[10] we focused sion of benzaldehyde (Table 1, entry 2). A result simi-
our interest on the nice approach reported by Portella lar to entry 1 was obtained when using an excess of
for the preparation of difluoroenoxysilanes. This di- (trifluoromethyl)trimethylsilane reagent and acetyltri-
fluorotriphenylstannate-promoted addition of (tri- methysilane (Table 1, entry 3). Combining the slow
fluoromethyl)trimethylsilane to acylsilanes was the addition of benzaldehyde with the use of an excess of
starting point of our study. We followed the simple difluoroenoxysilane precursors eventually allowed
reasoning that the fluoride anion that is generated in a full conversion of benzaldehyde and a total selectiv-
the process could itself act as a catalyst for the aldol ity in favour of the aldol product (Table 1, entry 4).
reaction. Indeed, Portellaꢁs methodology was appa- Lowering the temperature at À608C was deleterious
rently willingly restricted to the preparation of di- since trifluoromethyl carbinol was again detected in
fluoroenoxysilanes, even if the latter were afterwards the reaction mixture, whatever the rate of addition of
subjected to a Lewis acid-promoted Mukiyama aldol the benzaldehyde solution (Table 1, entry 5).
reaction.[11] We thus investigated a one-pot reaction
Based on the conditions devised in Table 1, entry 4,
between (trifluoromethyl)trimethylsilane, acetyltrime- a quick survey of onium pair catalysts was conducted
thylsilane and benzaldehyde that was expected to di- (Table 2). para-Bromobenzaldehyde was used as the
rectly afford the corresponding difluorinated aldol.
electrophile and an acidic cleavage of the trimethyl-
We first studied this reaction using tetrabutylammo- silyl ether was performed after the reaction. This
nium difluorotriphenylsilicate (TBAT) as the catalyst, study was meant to assess the influence of the nature
a commercially available, silicate version of the cata- of the Lewis base on the course of the reaction. The
lyst used by Portella.[9c] In a first attempt, an equimo- efficiency of TBAT as a catalyst was confirmed since
lar mixture of the three reagents was subjected to 5a could be isolated in 62% yield after deprotection
10 mol% of TBAT in THF at À408C. After three of the intermediate trimethylsilyl ether (Table 2,
hours of reaction and complete consumption of ben- entry 1). In contrast, the use of tetrabutylammonium
zaldehyde according to TLC monitoring, a 51:49 mix- fluoride (TBAF) led to a complex mixture, from
ture of the expected aldol 3 and trifluoromethyl carbi- which only small amounts of trifluoromethylated
nol 4 was obtained (Table 1, entry 1). Although unsat- product 3 and of its corresponding free alcohol could
isfactory, this result was nonetheless encouraging. be identified (Table 2, entry 2). This result was not
Indeed, we were initially worried about the relative surprising since Portella had demonstrated that TBAF
reactivities of acetyltrimethylsilane and benzaldehyde. was not an appropriate promoter for the difluoro-
Since the aldol and the trifluoromethyl carbinol were enoxysilane generation and led to many side-pro-
Adv. Synth. Catal. 2016, 358, 526 – 531
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
527

COMMUNICATIONS
MØlanie Decostanzi et al.
Table 2. Catalyst survey.
Entry
XÀ
Yield[a]
À
1
Ph3SiF2
FÀ
62%
nd
16%
54%
2
3[b]
4
(p-MeO)C6H4OÀ
AcOÀ
[a]
The reaction is performed using method B: a THF solu-
tion of PhCHO is added in 15 min with a syringe pump.
The catalyst was prepared in situ by anion metathesis
from (n-Bu)4N+,BrÀ and (p-MeO)C6H4ONa in THF for
1 h.
[b]
ducts.[9c] Disappointingly, the use of tetrabutylammo-
nium (para-methoxy)phenoxide led to a low 16% iso-
lated yield of 5a. The crude mixture was mainly con-
stituted of unreacted aldehyde and its trifluoromethy-
lation product (Table 2, entry 3). In contrast, tetrabu-
tylammonium acetate appeared as an efficient pro-
moter since the same reaction led this time to a 54%
yield of 5a, with no traces of unreacted aldehyde or
of trifluoromethylcarbinol (Table 2, entry 4).
The scope of the reaction was next examined using
TBAT as the catalyst under the optimized conditions
(Scheme 2).
The compounds were obtained in overall yields
ranging from 38 to 72%. One should keep in mind
that these yields reflect a four-step process (addition
of CF3TMS to the acylsilane, Brook/elimination se-
quence, aldol reaction and deprotection) with, there-
fore, an average yield of 79–92% for each step. Our
one-pot sequence is thus fairly efficient and nicely
competes with Portellaꢁs sequential method.[11] Aro-
matic aldehydes appeared as suitable substrates for
this reaction, albeit the yield decreased using cinnam-
aldehyde (38%). Moreover, only a complex mixture
was obtained from isobutyraldehyde. Other acylsi-
Scheme 2. Scope of the one-pot aldol reaction.
lanes (i-PrCOSiMe3 and PhCH2CH2COSiMe3) were The latter affords the difluoroenoxysilane as well as
successfully used (products 5b, 6b, 5c, 6c and 13c). an ammonium fluoride that can promote the aldol re-
Disappointingly, product 6d, resulting from the reac- action. Two pathways are possible for closing the
tion with benzoyltrimethylsilane and benzaldehyde, cycle, depending on the initial catalyst. If the latter is
could never be obtained as a pure sample, despite TBAT, the resulting ammonium alkoxide might be si-
a good conversion. Unexpectedly, this aldol was, in lylated with TMSF to release the product and an am-
our hands, moderately stable and always underwent monium fluoride that can catalyze the first step.[12] If
degradation during purifications.
the reaction is catalyzed by tetrabutylammonium ace-
The mechanism and the catalytic cycle of the reac- tate or aryloxide, silylation of the ammonium alkox-
tion are depicted on Scheme 3. As mentioned above, ide by the ROSiMe3 species produced in the first step
the addition of (trifluoromethyl)trimethylsilane to the can occur and regenerate the catalyst.[13] In both
acylsilane is triggered by the catalyst and is followed cases, the possibility that the ammonium alkoxide re-
by the Brook rearrangement/elimination sequence. sulting from the aldol reaction would itself catalyze
528
asc.wiley-vch.de
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2016, 358, 526 – 531

COMMUNICATIONS
Brook/Elimination/Aldol Reaction Sequence
Figure 1. Chiral ammonium salts tested during the survey.
efficient catalysts for the racemic reaction. Starting
from a small family of N-benzylammonium chlorides
derived from Cinchona alkaloids (Figure 1), several
acetate and phenoxide salts were thus prepared and
tested as catalysts in our reaction. They were generat-
ed in situ from the corresponding chloride salts,
through an anion metathesis reaction with sodium
acetate or with the corresponding sodium phenoxi-
Scheme 3. Postulated catalytic cycle (BE=Brook rearrange-
ment/elimination; AR=aldol reaction).
the addition of the Ruppert reagent cannot be ruled de.[16c] Our first goal was to determine the appropriate
out.[10]
counter-anion, using commercially available
We afterwards decided to explore the possibilities QN1+,ClÀ as the standard pre-catalyst (Table 3). Sur-
of developing an asymmetric version of this reaction. prisingly, using the acetate counter-ion was in this
Indeed, although meaningless for the enoxysilane case unproductive (Table 3, entries 1 and 2). We thus
generation, the use of a chiral ammonium cation turned back to ammonium phenoxides and the use of
would allow us to induce enantioselectivity during the QN1+,(p-CF3)C6H4OÀ as the catalyst allowed the for-
aldol reaction. An efficient asymmetric version of our mation of 13a in average yield and low enantioselec-
reaction would provide an access to enantioenriched tivity (Table 3, entry 3). We then tried to slightly mod-
difluoro aldols derived from aliphatic ketones, while ulate the basicity of this anion (Table 3, entries 4 and
the advanced literature methods are limited to aro- 5), which has a slight, but sensitive influence on the
matic difluoroenolates.[8h] Cooperative chiral ion pair- reactivity. The use of a simple phenoxide brought no
ing catalysis, i.e., processes involving the active partic- improvement, while the more basic (p-MeO)C6H4OÀ
ipation of the anionic and cationic parts in the catalyt- led to a moderate yield and a slighty higher ee. The
ic cycle, has proved to be an efficient approach in latter was retained for the rest of the study. A quick
terms of reaction scope and stereoselectivity.[14] As solvent survey demonstrated that apolar or non-basic
such, chiral ammonium fluorides have been extensive- solvents were inappropriate (Table 3, entries 6 and 7).
ly studied as organocatalysts in which the ammonium However, the use of a strongly polar solvent such as
part is able to bring chiral information and stabilize DMF resulted in a complete erosion of the enantiose-
the development of a negative charge, while the fluo- lectivity (Table 3, entry 8). Finally, N-benzylquinidini-
ride part acts as a (Lewis or Brønsted) base that acti- um, cinchoninium and cinchonidinium were also
vates a pro-nucleophile.[15] However, the handling and tested under the standard conditions and a modest
storage of ammonium fluoride catalysts is often diffi- 24% ee was obtained in the best case (Table 3, en-
cult due to their hygroscopic character. Moreover, tries 9–11).[18] The feasibility of such an asymmetric
TBAF failed to promote our racemic reaction, ruling one-pot/one-catalyst aldol reaction has, however,
out the use of chiral ammonium fluorides (Table 2). been demonstrated and an extensive screening of
The use of a less basic, softer anion such as hydrogen onium pair catalysts has now to be performed.
bifluoride, acetate or phenoxide is often a nice alter-
In summary, a one-pot preparation of difluorinated
native to produce chiral catalysts that are more stable aldols directly from (trifluoromethyl)trimethylsilane,
and more easily handled.[16,17] Regarding our one-pot acyltrimethylsilanes and aldehydes was developed.
reaction, tetrabutylammonium acetate and, to a much The process is promoted by a catalytic amount of an
smaller extent, (para-methoxy)phenoxide appeared as ammonium salt and generates TMSF as the sole by-
Adv. Synth. Catal. 2016, 358, 526 – 531
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
529

COMMUNICATIONS
MØlanie Decostanzi et al.
Table 3. Preliminary study for an asymmetric version of the
one-pot Brook/elimination/aldol reaction sequence.
Experimental Section
General Procedure for the TBAT-Catalyzed Brook
Rearrangement/Fluoride Elimination/Aldol Reaction
Sequence
To a solution of CF3TMS (296 mL, 2 mmol, 2 equiv.) in THF
(2 mL) at À408C were added acylsilane (2 mmol, 2 equiv.)
and TBAT (54 mg, 0.1 mmol, 0.1 equiv.). To this mixture
was added over a period of 15 min a solution of aldehyde
(1 mmol, 1 equiv.) in THF (1 mL). The solution was stirred
for 1h 30min at À408C and then a saturated aqueous solu-
tion of NH4Cl was added. The layers were separated and
the aqueous layer was extracted with Et2O. The combined
organic layers were washed with brine, dried over MgSO4,
filtered and concentrated under vacuum. The crude product
was then diluted in MeOH (2 mL) and a solution of 10%
aqueous HCl (5 mL) was added at 08C. The solution was
stirred for 10 min at this temperature and neutralized with
a saturated solution of NaHCO3. The layers were separated
and the aqueous layer was extracted with Et2O. The com-
bined organic layers were washed with brine, dried over
MgSO4, filtered and concentrated under vacuum. The crude
product was purified by flash column chromatography (pen-
tane/Et2O 90:10) to give the expected difluorinated aldol.
Entry XÀ[a]
R4N+ Solvent Yield [%] ee[c]
[%]
1
AcOÀ
QN1+ THF
QN1+ THF
QN1+ THF
QN1+ THF
–
–
49
38
40
–
–
–
10
12
16
–
2[b]
3
AcOÀ
(p-CF3)C6H4OÀ
C6H5OÀ
4
5
6
7
8
9
10
11
(p-MeO)C6H4OÀ QN1+ THF
(p-MeO)C6H4OÀ QN1+ PhCH3
(p-MeO)C6H4OÀ QN1+ CH2Cl2
(p-MeO)C6H4OÀ QN1+ DMF
(p-MeO)C6H4OÀ QD1+ THF
(p-MeO)C6H4OÀ CN1+ THF
(p-MeO)C6H4OÀ CD1+ THF
–
–
52
42
20
54
2[d]
24[d]
6[d]
12
[a]
The catalyst was prepared in situ by anion metathesis
from R4N+,ClÀ and NaX in THF for 1 h, unless otherwise
indicated.
The catalyst was prepared and isolated prior to reaction
by anion metathesis from QN1+,ClÀ and AcOAg in
CH2Cl2.[19]
Measured prior to any chromatographic purification by
chiral HPLC using a Daicel Chiralpakꢂ IB column.
The enantioselectivity is reversed compared to the other
entries.
[b]
Acknowledgements
[c]
[d]
This work was supported by the Ecole Nationale SupØrieure
de Chimie de Montpellier (PhD grant for M.D.) and by the
Centre National de la Recherche Scientifique, which we grate-
fully acknowledge.
product. The anionic part of the ammonium salt has
to be carefully chosen so that it efficiently activates
(trifluoromethyl)trimethylsilane. An efficient racemic
reaction was devised, using commercially available
TBAT as the catalyst, leading directly to difluorinated
aldols in good overall yield. Compared to the previ-
ous sequential methods (preparation of the difluoro-
enoxysilane followed by Mukaiyama aldol reaction),
this one-pot/one-catalyst method is fairly efficient.
Despite the fact that acylsilanes are not always easily
accessible, this approach allows the preparation of di-
fluoro aldols derived from aliphatic ketones, which is
rarely the case for classical methods.[6,8] A preliminary
study of an asymmetric version is also reported, with
a first result (42% yield, 24% ee) that is of course un-
satisfactory. If asymmetric catalysis appears possible
for this reaction, the determination of an efficient
chiral ion pair is still under investigation in our labo-
ratory and results in this area will be reported in due
course.
References
[1] J. Wang, M. Sµnchez-Roselló, J. L. AceÇa, C. del Pozo,
A. E. Sorochinsky, S. Fustero, V. A. Soloshonok, H.
Liu, Chem. Rev. 2014, 114, 2432–2506.
[2] a) A. Bondi, J. Phys. Chem. 1964, 68, 441–451; b) J. C.
Biffinger, H. W. Kim, S. G. DiMagno, ChemBioChem
2004, 5, 622–627; and references cited therein; c) D.
OꢁHagan, Chem. Soc. Rev. 2008, 37, 308–319.
[3] a) J.-P. BØguØ, D. Bonnet-Delpon, Bioorganic and Me-
dicinal Chemistry of Fluorine, John Wiley & Sons Inc.,
Hoboken, 2008; b) S. Purser, P. R. Moore, S. Swallow,
V. Gouverneur, Chem. Soc. Rev. 2008, 37, 320–330;
c) P. Kirsch, Modern Fluoroorganic Chemistry, Wiley-
VCH, Weinheim, 2004; d) H. J. Bçhm, D. Banner, S.
Bendels, M. Kansy, B. Kuhn, K. Müller, U. Obst-
Sander, M. Stahl, ChemBioChem 2004, 5, 637–643.
[4] T. Liang, C. N. Neumann, T. Ritter, Angew. Chem.
2013, 125, 8372–8423; Angew. Chem. Int. Ed. 2013, 52,
8214–8264.
[5] For a review on fluorinated enol ethers, see: M. Deco-
stanzi, J.-M. Campagne, E. Leclerc, Org. Biomol.
Chem. 2015, 13, 7351–7380.
[6] For selected examples of difluoro aldol reactions, see:
a) F.-M. Liao, Y.-L. Liu, J.-S. Yu, F. Zhou, J. Zhou, Org.
Biomol. Chem. 2015, 13, 8906–8911; b) J.-S. Yu, Y.-L.
530
asc.wiley-vch.de
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2016, 358, 526 – 531

COMMUNICATIONS
Brook/Elimination/Aldol Reaction Sequence
Liu, J. Tang, X. Wang, J. Zhou, Angew. Chem. 2014,
126, 9666–9670; Angew. Chem. Int. Ed. 2014, 53, 9512–
9516; c) P. Zhang, C. Wolf, Angew. Chem. 2013, 125,
8023–8027; Angew. Chem. Int. Ed. 2013, 52, 7869–7873;
d) Y. Liu, J. Zhou, Acta Chim. Sinica 2012, 70, 1451–
1456; e) Y.-L. Liu, J. Zhou, Chem. Commun. 2012, 48,
1919–1921; f) L. Wu, J. Fluorine Chem. 2011, 132, 367–
372; g) W. J. Chung, S. C. Ngo, S. Higashiya, J. T.
Welch, Tetrahedron Lett. 2004, 45, 5403–5406; h) G.
Blond, T. Billard, B. R. Langlois, Chem. Eur. J. 2002, 8,
2917–2922; i) H. Amii, T. Kobayashi, Y. Hatamoto, K.
Uneyama, Chem. Commun. 1999, 1323–1324; j) J. A.
Weigel, J. Org. Chem. 1997, 62, 6108–6109; k) K. Iseki,
Y. Kuroki, D. Asada, M. Takahashi, S. Kishimoto, Y.
Kobayashi, Tetrahedron 1997, 53, 10271–10280; l) M.
Yamana, T. Ishihara, T. Ando, Tetrahedron Lett. 1983,
24, 507–510.
1993, 58, 6675–6680; e) F. Jin, Y. Xu, W. Huang, J.
Chem. Soc. Chem. Commun. 1993, 814–816; f) F. Jin, B.
Jiang, Y. Xu, Tetrahedron Lett. 1992, 33, 1221–1224.
[10] M. Decostanzi, A. Van Der Lee, J.-M. Campagne, E.
Leclerc, Adv. Synth. Catal. 2015, 357, 3091–3097.
[11] For sequential Brook rearrangement/fluoride elimina-
tion, Mukaiyama aldol reactions, see: a) O. Lefebvre,
T. Brigaud, C. Portella, J. Org. Chem. 2001, 66, 4348–
4351; b) D. Saleur, T. Brigaud, J.-P. Bouillon, C. Portel-
la, Synlett 1999, 432–434.
[12] For silylation with TMSF, see: E. Nakamura, M. Shimi-
zu, I. Kuwajima, J. Sakata, K. Yokoyama, R. Noyori, J.
Org. Chem. 1983, 48, 932–945.
[13] For silylation with ArOTMS species, see : A. Claraz, S.
Oudeyer, V. Levacher, Tetrahedron: Asymmetry 2013,
24, 764–768.
[14] For a review on asymmetric catalysis with chiral ion
pairs, see: J.-F. Bri›re, S. Oudeyer, V. Dalla, V. Levach-
er, Chem. Soc. Rev. 2012, 41, 1696–1707.
[15] For selected examples, see: a) S. Mizuta, N. Shibata, S.
Akiti, H. Fujimoto, S. Nakamura, T. Toru, Org. Lett.
2007, 9, 3707–3710; b) M. B. Andrus, J. Liu, Z. Ye, J. F.
Cannon, Org. Lett. 2005, 7, 3861–3864; c) S. Colonna,
H. Hiemstra, H. Wynberg, J. Chem. Soc. Chem.
Commun. 1978, 238.
[16] For selected examples, see: a) A. Claraz, G. Landelle,
S. Oudeyer, V. Levacher, Eur. J. Org. Chem. 2013,
7693–7696; b) A. Claraz, S. Oudeyer, V. Levacher Adv.
Synth. Catal. 2013, 355, 841–846; c) L. Bernardi, E. In-
drigo, S. Pollicino, A. Ricci, Chem. Commun. 2012, 48,
1428–1430; d) R. P. Singh, B. M. Foxman, L. Deng, J.
Am. Chem. Soc. 2010, 132, 9558–9560; e) D. Peng, H.
Zhou, X. Liu, L. Wang, S. Chen, X. Feng, Synlett 2007,
2448–2450; f) T. Tozawa, H. Nagao, Y. Yamane, T. Mu-
kaiyama, Chem. Asian J. 2007, 2, 123–134; g) T. Ooi, S.
Takada, K. Doda, K. Maruoka, Angew. Chem. 2006,
118, 7768–7770; Angew. Chem. Int. Ed. 2006, 45, 7606–
7608.
[7] For selected examples of Reformatsky reactions, see:
a) T. Poisson, M.-C. Belhomme, X. Pannecoucke, J.
Org. Chem. 2012, 77, 9277–9285; b) M. Fornalczyk, K.
Singh, A. M. Stuart, Org. Biomol. Chem. 2012, 10,
3332–3342; c) R. J. Kloetzing, T. Thaler, P. Knochel,
Org. Lett. 2006, 8, 1125–1128; d) K. Sato, A. Tarui, T.
Kita, Y. Ishida, H. Tamura, M. Omote, A. Ando, I. Ku-
madaki, Tetrahedron Lett. 2004, 45, 5735–5737; e) M.
Braun, A. Vonderhagen, D. Waldmüller, Liebigs Ann.
1995, 1447–1450; f) T. T. Curran, J. Org. Chem. 1993,
58, 6360–6363; g) T. Ishihara, T. Yamanaka, T. Ando,
Chem. Lett. 1984, 1165–1168.
[8] For examples of generation of difluoroenolate through
mild release of a trifluoroacetate, see: a) C. Xie, Y.
Dai, H. Mei, J. Han, V. A. Soloshonok, Y. Pan, Chem.
Commun. 2015, 51, 9149–9152; b) H. Mei, C. Xie, J. L.
AceÇa, V. A. Soloshonok, G.-V. Rçschenthaler, J. Han,
Eur. J. Org. Chem. 2015, 6401–6412; c) C. Xie, L. Wu,
J. Zhou, H. Mei, V. A. Soloshonok, J. Han, Y. Pan, J.
Fluorine Chem. 2015, 172, 13–21; d) W. Li, Y. Zhu, Y.
Duan, M. Zhang, C. Zhu, Adv. Synth. Catal. 2015, 357,
1277–1282; e) C. Xie, L. Wu, J. Han, V. A. Soloshonok,
Y. Pan, Angew. Chem. Int. Ed. 2015, 54, 6019–6023;
f) C. Xie, L. Wu, H. Mei, V. A. Soloshonok, J. Han, Y.
Pan, Org. Biomol. Chem. 2014, 12, 7836–7843; g) C.
Xie, L. Wu, H. Mei, V. A. Soloshonok, J. Han, Y. Pan,
Tetrahedron Lett. 2014, 55, 5908–5910; h) P. Zhang, C.
Wolf, Angew. Chem. 2013, 125, 8023–8027; Angew.
Chem. Int. Ed. 2013, 52, 7869–7873; i) P. Zhang, C.
Wolf, J. Org. Chem. 2012, 77, 8840–8844; j) C. Han,
E. H. Kim, D. A. Colby, J. Am. Chem. Soc. 2011, 133,
5802–5805; k) J. P. John, D. A. Colby, J. Org. Chem.
2011, 76, 9163–9168.
[17] For a review on catalysis with ammonium phenoxides,
see: J. Godemert, S. Oudeyer, V. Levacher, Chem-
CatChem 2016, 8, 74–85.
[18] SDE (self-disproportionation of enantiomers) tests
were conducted to ensure the accuracy of the ee values.
As mentioned in Table 3, the ee values were measured
prior to any purification but no discrepancy was ob-
served when measured after column chromatography.
A sublimation test (see the Supporting Information for
details) was also performed and no SDE occurred. The
study is inspired from the one conducted in ref.[8e] For
a review on SDE, see: V. A. Soloshonok, C. Roussel,
O. Kitagawa, A. E. Sorochinsky Chem. Soc. Rev. 2012,
4180–4188.
[9] a) L. Wu, J. Fluorine Chem. 2011, 132, 367–372; b) I.
Fleming, R. S. Roberts, S. C. Smith, J. Chem. Soc.
Perkin Trans. 1 1998, 1215–1228; c) T. Brigaud, P. Dous-
sot, C. Portella, J. Chem. Soc. Chem. Commun. 1994,
2117–2118; d) P. Doussot, C. Portella, J. Org. Chem.
[19] D. Peng, H. Zhou, X. Liu, L. Wang, S. Chen, X. Feng,
Synlett 2007, 2448–2450.
Adv. Synth. Catal. 2016, 358, 526 – 531
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
531