Organocatalytic trifluoromethylation of imines using phase-transfer catalysis with phenoxides. A general platform for catalytic additions of organosilanes to imines

Article abstract of DOI:10.1039/c0cc05777k

A new approach to additions of silicon nucleophiles to imines was developed. The method is based on the phase-transfer of phenoxides by ammonium catalysts, overcoming the inability of amide adducts in promoting the reactions.

Full text of DOI:10.1039/c0cc05777k

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 1428–1430
www.rsc.org/chemcomm
COMMUNICATION
Organocatalytic trifluoromethylation of imines using phase-transfer
catalysis with phenoxides. A general platform for catalytic additions
of organosilanes to iminesw
Luca Bernardi,* Eugenio Indrigo, Salvatore Pollicino and Alfredo Ricci
Received 24th December 2010, Accepted 21st March 2011
DOI: 10.1039/c0cc05777k
A new approach to additions of silicon nucleophiles to imines
was developed. The method is based on the phase-transfer of
phenoxides by ammonium catalysts, overcoming the inability of
amide adducts in promoting the reactions.
Addition of silicon-based nucleophiles to carbonyl compounds
is arguably one of the pillars of organic synthesis. Activation
of the silicon reagent with a Lewis base catalyst is one of
the most viable strategies for the promotion of these
transformations.1 These reactions are thought to proceed
through an autocatalytic cycle triggered by the Lewis base,
wherein the anionic adduct generated in the initiation step
serves as the actual promoter. However, whereas alkoxides
formed upon addition to aldehydes and ketones are able to
carry out the autocatalytic cycle, the corresponding amide
adducts derived from imines do not always feature sufficient
silicon affinity to activate the silicon reagent.2 A stoichiometric
amount of the Lewis base catalyst is thus required.
Herein we present a new approach to the addition of silicon-
based nucleophiles to imines relying on phase-transfer catalysis
(PTC, Scheme 1).3 This reaction manifold, here developed for
the trifluoromethylation of imines, is amenable to various
silicon nucleophile additions. Our approach is based on the
phase-transfer extraction of a stoichiometric promoter, an
insoluble metal phenoxide, by a quaternary ammonium salt
(eventually chiral). The catalyst ammonium phenoxide is
regenerated in the organic layer (toluene) after each catalytic
cycle through a phase-transfer process, with concomitant
release of the amide adduct. The inability of the anionic amide
adduct in promoting the autocatalytic cycle is thus overcome,
rendering possible the use of ammonium salts in substoichio-
metric amounts. The choice of phenoxides as stoichiometric
promoters was dictated by their higher polarisability and
lipophilicity, compared to other popular anionic Lewis bases
such as fluorides4 and hydroxides.5 These features should
secure a good affinity with lipophilic ammonium salts,6
Scheme 1 TMS = trimethylsilyl, Pg = protecting group.
facilitating the phase-transfer process. Besides, phenoxides
have proven very useful as catalysts in various additions of
silicon reagents to carbonyl compounds.7
Trifluoromethylation of imines8 was chosen as a case study
for different reasons: (i) it represents one of the most direct
routes to a-(trifluoromethyl)amines 3, key building blocks
in pharmaceutical and agro-chemistry;9 (ii) due to the low
acidity of trifluoromethane and due to the instability of the
corresponding anion,10 trimethyl(trifluoromethyl)silane 2a,
Ruppert–Prakash reagent,11 is mostly used in nucleophilic
trifluoromethylation reactions; (iii) Lewis base activation
forming hypervalent silicon species12 weakening the Si–CF3
bond is generally necessary in trifluoromethylation with 2a due
to its low reactivity,8 i.e. Lewis acid activation of the electro-
philic partner is not sufficient; (iv) overstoichiometric amounts
of Lewis-base promoters, such as fluoride salts, are mostly
employed in trifluoromethylation of imines;13 (v) the few catalytic
protocols reported5,14–17 (of which only one asymmetric)5 still bear
considerable limitations: some of these are in fact limited
to highly reactive substrates (azirines),14 proceed with low
efficiency,15 or rely on a highly coordinating solvent such as
DMF,2b,16 which assists in the formation of hypervalent
Department of Organic Chemistry ‘‘A. Mangini’’,
University of Bologna, V. Risorgimento 4, 40136 Bologna, Italy.
E-mail: luca.bernardi2@.unibo.it; Fax: +39 051 2093654;
Tel: +39 0512093631
w Electronic supplementary information (ESI) available: Experimental
details, background experiments, product characterisation, NMR spectra,
HPLC traces and deprotection of (+)-3o. See DOI: 10.1039/c0cc05777k
c
1428 Chem. Commun., 2012, 48, 1428–1430
This journal is The Royal Society of Chemistry 2012

View Online
Table 2 Scope of the catalytic trifluoromethylation reactiona
Table 1 Representative optimisation resultsa
Entry
M
Equiv. Yieldb (%) Entry
M
Equiv. Yieldb (%)
Yieldb (%)
1
K
K
K
K
K
0.2
2
2
1.1
5
22
o5
60
7
8
9
Na 1.1
Cs 1.1
Na 0.2
Na 1.1
Na 1.1
Na 1.1
80
30
30
10
>95
>95
Entry
1
R
Pg
A or B
2c
3
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1o
1p
1q
1r
1s
Ph
Ts
Ts
Ts
Ts
Ts
Ts
Ts
Ts
Ts
Ts
Ts
SO2PMP
SO2PMP
SO2Mes
SO2Mes
P(O)Ph2
P(O)Ph2
Cbz
A
A
A
A
A
A
A
B
3a (97)
3b (85)
3c (98)
3d (80)
3e (80)
3f (68)
3g (85)
3h (91)
3i (76)
3j (59)
3k (78)
3l (90)
3m (39)
3n (81)
3o (75)
3p (71)
3q (51)
3r (75)
3s (53)
4
5
6
72
o10
10c
11d
12d,e
2-BrC6H4
4-ClC6H4
4-MeOC6H4
2-Naphthyl
1-Naphthyl
2-Thienyl
Cy
(CH3)2CH
Ph(CH2)2
CH3(CH2)2
Ph
Cy
Ph
Cy
Ph
Cy
Ph
Cy
Li 1.1
35
a
Reagents and conditions: imine 1a (0.10 mmol), 2a (2 M in THF,
0.30 mmol), TBAB (0.020 mmol), toluene (1.0 mL), then PhOM
(equiv.), 16–20 h. Determined by 1H NMR spectroscopy using
c
dibenzyl as internal standard. Without catalyst. MS
b
d
9
B
B
B
5 A,
e
10
11
12
13
14
15
16
17
18
19
10 mol% TBAB, 2a 0.15 mmol, addition of PhONa at 0 1C. Neat
TMSCF3 was used. Ts = p-toluenesulfonyl.
A
Bc
A
Bc
A
Bc
Bd
Be
silicon species but often hampers asymmetric processes.
Furthermore, imines derived from linear unbranched
aldehydes, or bearing readily removable protecting groups at
nitrogen such as carbamoyl, have never been used in these
catalytic protocols for their instability.
Cbz
a
Method A: 1 (0.25 mmol), 2a (2 M in THF, 0.375 mmol), TBAB
(0.025 mmol), MS 5 A, toluene (2.5 mL), then PhONa (0.275 mmol,
added at 0 1C), 16–20 h RT. Method B: 1 (0.20 mmol), 2a (2 M in
THF, 0.80 mmol), TBAB (0.040 mmol), MS 5 A, toluene (2.0 mL),
The trifluoromethylation of N-tosyl imine 1a with TMSCF3
2a (3 equiv., commercial solution 2 M in THF) in toluene as
solvent was studied first, using tetra-n-butyl ammonium
bromide (TBAB, 20 mol%) as catalyst (Table 1). Initial experi-
ments were aimed at assessing the correctness of our assump-
tions in envisaging the catalytic cycle depicted in Scheme 1. The
inability of the amide adduct to promote a subsequent addition
was first confirmed by using a substoichiometric amount of
ammonium phenoxide (20 mol%, in situ generated from TBAB
and PhOK)18 which gave product 3a without catalyst turnover,
i.e. the reaction stopped after the initiation step (Table 1,
entry 1). Importantly, excess solid PhOK was unable to pro-
mote the reaction in the absence of TBAB, presumably for its
insolubility in toluene (Table 1, entry 2). To our delight, the
combination of 20 mol% TBAB as catalyst and 2 equiv. PhOK
produced instead the trifluoromethylated product 3a in 60%
yield (Table 1, entry 3).19 Counter-intuitively, an increase in
yield was observed using a smaller excess of PhOK (1.1 instead
of 2 equiv., Table 1, entry 4). Apparently, PhOK is responsible
for a parasitic decomposition of imine 1a, as confirmed by the
reaction with a large excess of PhOK, which hardly produced
any trifluoromethylated product 3a (Table 1, entry 5). It was
found at this stage that the phase-transfer exchange is by far
the rate limiting step of the whole process, as in most solid–
liquid PTC reactions.20 To improve mass transfer kinetics,
different alkaline metal phenoxides were tested, the sodium
derivative outperforming the other salts (Table 1, entries 4,
6–8). After having verified the inactivity of the amide adduct
and of solid PhONa in the reaction (Table 1, entries 9 and 10),
a screening of different additives was undertaken.21 Activated
5 A molecular sieves gave a very positive effect. It was finally
found that using these drying agents and adding PhONa at
0 1C, it was possible to obtain the trifluoromethylated adduct
3a in nearly quantitative yield even with lower amounts of the
catalyst (10 mol%) and TMSCF3 2a (1.5 instead of 3 equiv.,
b
then PhONa (0.44 mmol), ꢀ55 1C, 16–20 h. Isolated yield after
c
d
e
chromatography on silica gel. At ꢀ30 1C. At RT. At 0 1C. Mes:
2,4,6-trimethylphenyl; PMP: p-methoxyphenyl; Cy: cyclohexyl.
Table 1, entry 11). The use of neat 2a gave identical results
compared to its THF solution (Table 1, entry 12), indicating
that THF did not influence the reaction.
We then explored possible variations at the imine partner
(Table 2). Various N-tosyl imines 1a–g derived from aromatic
or heteroaromatic aldehydes reacted under the optimised
conditions (method A in Table 2) furnishing the adducts
3a–g in generally good yields (Table 2, entries 1–7). We next
turned our attention to the more challenging imines derived
from aliphatic enolisable aldehydes. Competing tautomerisa-
tion giving the enamide, especially when derived from primary
un-branched aldehydes, prevented their generalised use in
most catalytic trifluoromethylations reported so far.5,14–17
Nevertheless, a slight modification to our protocol, involving
the in situ generation of the imines at low temperature from the
corresponding a-amido sulfones 1h–k,22 sacrificing additional
equivalents of TMSCF3 2a and PhONa (method B in Table 2),
allowed the obtainment of the adducts 3h–k in satisfactory
yields, even with imines derived from primary aldehydes
(Table 2, entries 8–11). Different protecting groups at the
nitrogen atom of the imines were also tolerated by this system.
Besides various sulfonyl derivatives (Table 2, entries 12–15),
synthetically more attractive diphenylphosphinoyl and Cbz
(carbobenzyloxy) groups could be in fact employed, with both
aromatic and aliphatic imines (Table 2, entries 16–19).
The general applicability of this new reaction manifold was
challenged using other silicon nucleophiles 2b–e in the reaction
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1428–1430 1429

View Online
3 Asymmetric Phase-Transfer Catalysis, ed. K. Maruoka,
Wiley-VCH, Weinheim, 2008.
4 Due to low affinity, in situ generation of fluorides is achieved only
from ammonium salts with hydrogensulfate as a counteranion, see:
T. Ooi, K. Doda and K. Maruoka, Org. Lett., 2001, 3, 1273.
5 Indeed, an autocatalytic cycle is operative even when excess KOH
is used in the reaction: H. Kawai, A. Kusuda, S. Nakamura,
M. Shiro and N. Shibata, Angew. Chem., Int. Ed., 2009, 48,
6324.
6 Chiral ammonium p-nitrophenoxides have been obtained by
mixing the halide salt with 4-NO2C6H4OK in CH3OH, see:
(a) E. J. Corey, X. Feng and M. C. Noe, J. Am. Chem. Soc.,
1997, 119, 12414; (b) T. B. Poulsen, L. Bernardi, J. Aleman,
J. Overgaard and K. A. Jørgensen, J. Am. Chem. Soc., 2007,
129, 441. For in situ formation, see: (c) B. Lygo, C. Beynon,
C. Lumley, M. C. McLeod and C. E. Wade, Tetrahedron Lett.,
2009, 50, 3363.
Scheme 2 Reagents and conditions: 1a (0.20 mmol), 2b–e (0.30 mmol),
TBAB (0.020 for 2d,e, 0.040 mmol for 2b,c), toluene (2.0 mL), ꢀ30 1C
(2b,e), 0 1C (2d), 0 1C–RT (2c), MS 5 A, PhONa (0.22 mmol), 18–24 h.
7 (a) H. Nagao, Y. Kawano and T. Mukaiyama, Bull. Chem. Soc.
Jpn., 2007, 80, 2406; (b) A. Capperucci, C. Tiberi, S. Pollicino and
A. Degl’Innocenti, Tetrahedron Lett., 2009, 50, 2808;
(c) M. Hatano, E. Takagi and K. Ishihara, Org. Lett., 2007, 9,
4527; (d) H. Zhao, B. Qin, X. Liu and X. Feng, Tetrahedron, 2007,
63, 6822. For silicon activation by a betaine, see: (e) D. Uraguchi,
K. Koshimoto, S. Miyake and T. Ooi, Angew. Chem., Int. Ed.,
2010, 49, 5567.
8 (a) G. K. S. Prakash and A. K. Yudin, Chem. Rev., 1997, 97, 757;
(b) J.-M. Ma and D. Cahard, Chem. Rev., 2004, 104, 6119;
(c) A. D. Dilman and V. V. Levin, Eur. J. Org. Chem., 2011, 831.
9 (a) W. K. Hagmann, J. Med. Chem., 2008, 51, 4359; (b) M. Sani,
A. Volonterio and M. Zanda, ChemMedChem, 2007, 2, 1693.
10 B. R. Langlois and T. Billard, Synthesis, 2003, 185.
11 (a) L. Ruppert, K. Schlich and W. Volbach, Tetrahedron Lett.,
1984, 25, 2195; (b) G. K. S. Prakash, R. Krishnamurti and
G. A. Olah, J. Am. Chem. Soc., 1989, 111, 393.
12 (a) N. Maggiarosa, W. Tyrra, D. Naumann, N. V. Kirij and
Y. L. Yagupolskii, Angew. Chem., Int. Ed., 1999, 38, 2252;
(b) A. Kolomeitsev, G. Bissky, E. Lork, V. Movchun,
E. Rusanov, P. Kirsch and G.-V. Roschentaler, Chem. Commun.,
1999, 1017; (c) S. Rendler and M. Oestreich, Synthesis, 2005, 1727;
(d) Y. Orito and M. Nakajima, Synthesis, 2006, 1391.
13 See for example: (a) V. A. Petrov, Tetrahedron Lett., 2000, 41,
6959; (b) G. K. S. Prakash, M. Mandal and G. A. Olah, Angew.
Chem., Int. Ed., 2001, 40, 589; (c) J.-C. Blazejewski, E. Anselmi and
M. P. Wilmshurst, Tetrahedron Lett., 1999, 40, 5475; (d) S. Mizuta,
N. Shibata, T. Sato, H. Fujimoto, S. Nakamura and T. Toru,
Synlett, 2006, 267; (e) Y. Kawano and T. Mukaiyama, Chem. Lett.,
2005, 34, 894; (f) A. D. Dilman, D. E. Arkhipov, V. V. Levin,
P. A. Belyakov, A. A. Korlyukov, M. I. Struchkova and
V. A. Tartakovsky, J. Org. Chem., 2008, 73, 5643.
with imine 1a. As shown in Scheme 2, transformations closely
related to the trifluoromethylation, such as perfluoroalkyl-
ation and perfluorophenylation with 2b and 2c giving 3t and
3u, were readily performed with good results under standard
conditions. It was however even more exciting to observe that
also distinct silicon reagents, such as trimethylsilylacetonitrile
2d and the trimethylsilyl enol ether 2e, could be successfully
employed, affording the corresponding Mannich adducts 3v
and 3w in excellent yields.
An obvious implementation of the present method is the
utilisation of chiral enantiopure ammonium salts. Experiments
were run using Cinchona alkaloid derivatives (see ESIw),
giving the lead result described in eqn (1) with hydroquinidine
derived catalyst 4. Although unsatisfactory in terms of
enantioinduction, the turnover observed even with a different
catalyst structure holds promise for future development.
ð1Þ
In summary, a new approach for silicon nucleophile
additions to imines was developed. The method, based on
the phase-transfer of PhONa, was developed for the trifluoro-
methylation of imines. Its generalisation was demonstrated by
the application to other classes of silicon reagents. Control
experiments confirmed that the catalytic cycle depicted in
Scheme 1 is indeed operative in the examples reported in
Table 2 and Scheme 2 (see ESIw). Efforts aimed at the utilisation
of chiral enantiopure ammonium salts as catalysts are underway.23
We acknowledge financial support from ‘Stereoselezione in
Sintesi Organica Metodologie e Applicazioni’ 2009.
14 C. P. Felix, N. Khatimi and A. Laurent, Tetrahedron Lett., 1994,
35, 3303.
15 D. W. Nelson, R. A. Easley and B. N. V. Pintea, Tetrahedron Lett.,
1999, 40, 25.
16 Y. Kawano, H. Fujisawa and T. Mukaiyama, Chem. Lett., 2005,
34, 422.
17 G. K. S. Prakash, R. Mogi and G. A. Olah, Org. Lett., 2006, 8,
3589.
18 An identical result was obtained using preformed TBAOPh. For a
discussion on phenoxide salts, see: R. Goddard, H. M. Herzog and
M. T. Reetz, Tetrahedron, 2002, 58, 7847, and references therein.
19 Different potassium phenoxides (4-MeOC6H4OK, 4-BrC6H4OK,
2-BrC6H4OK, 2,4,6-(CH3)3C6H2OK, 4-NO2C6H4OK), and other
PTC’s (triethylbenzyl, cetyl, tetra-n-hexyl, tetraethyl ammonium
bromides) performed similar or worse than TBAB and PhOK.
20 For an excellent discussion, see: M. Chidambaran, S. U. J.
Sonavane de la Zerda and Y. Sasson, Tetrahedron, 2007, 63,
7696.
Notes and references
1 (a) R. Noyori, K. Yokoyama, J. Sakata, I. Kuwajima,
E. Nakamura and M. Shimizu, J. Am. Chem. Soc., 1977, 99,
1265; (b) T. Ooi and K. Maruoka, Acc. Chem. Res., 2004, 37,
526; (c) S. E. Denmark and R. A. Stavenger, Acc. Chem. Res.,
2000, 33, 432; (d) J. Gawronski, N. Wascinska and J. Gajewy,
Chem. Rev., 2008, 108, 5227.
21 Other sodium phenoxides (F5C6ONa, 4-BrC6H4ONa, 4-NO2C6H4ONa)
performed worse than PhONa.
22 (a) M. Petrini, Chem. Rev., 2005, 105, 3949; (b) B. Yin, Y. Zhang
and L.-W. Xu, Synthesis, 2010, 3583.
23 During the preparation of this manuscript, catalytic generation of
a chiral fluoride-crown ether complex was reported: H. Yan,
H. B. Jang, J.-W. Lee, H. K. Kim, S. W. Lee, J. W. Jang and
C. E. Song, Angew. Chem., Int. Ed., 2010, 49, 8915.
2 (a) R. B. Lettan II and K. A. Scheidt, Org. Lett., 2005, 7, 3227;
(b) Y. Kawano, N. Kaneko and T. Mukaiyama, Bull. Chem. Soc.
Jpn., 2006, 79, 1133; (c) H. Fujisawa, E. Takahashi and
T. Mukaiyama, Chem.–Eur. J., 2006, 12, 5082.
c
1430 Chem. Commun., 2012, 48, 1428–1430
This journal is The Royal Society of Chemistry 2012