Chiral bronsted acid catalysis for enantioselective hosomi-sakurai reaction of imines with allyltrimethylsilane

Article abstract of DOI:10.1021/ol200595b

The chiral Bronsted acid (1b or 1c) has been shown to initiate the Hosomi-Sakurai reaction of imines with excellent enantioselectivities. The combined Bronsted acid system has been developed to offer a new class of chiral Bronsted acid catalysis. The pres

Full text of DOI:10.1021/ol200595b

ORGANIC
LETTERS
2011
Vol. 13, No. 8
2126–2129
Chiral Brønsted Acid Catalysis for
Enantioselective HosomiꢀSakurai
Reaction of Imines with
Allyltrimethylsilane
Norie Momiyama,* Hayato Nishimoto, and Masahiro Terada*
Department of Chemistry, Graduate School of Science, Tohoku University, Sendai,
980-8578, Japan
momiyama@m.tohoku.ac.jp; mterada@m.tohoku.ac.jp
Received March 5, 2011
ABSTRACT
The chiral Brønsted acid (1b or 1c) has been shown to initiate the HosomiꢀSakurai reaction of imines with excellent enantioselectivities. The
combined Brønsted acid system has been developed to offer a new class of chiral Brønsted acid catalysis. The present system proceeds through
regeneration of the chiral Brønsted acid by proton transfer from additional Brønsted acid to silylated chiral Brønsted acid, a newly elucidated
mechanism for the role of the additional Brønsted acid.
Chiral Brønsted acid catalysis has been explosively
widespread in asymmetric synthesis over the past decade.1
Our laboratory and others have focused on the develop-
ment of chiral binaphthol-derived monophosphoric acid
as the chiral Brønsted acid catalyst in a variety of organic
transformations.2 During the development of chiral
Brønsted acidcatalyzedenantioselective reactions, seminal
advances have been made in the reaction of highly reactive
nucleophiles3 including dienes, enolsilanes and enols,
enamine, enamides and enecarbamates, cyanide, furanes,
indoles and pyrroles, diazoacetates, azidos, and allyl,
alkenyl, and alkynyl boronates, etc.1,2,4 However, its
ability to deliver both the reactivity and stereoselectivity
over a range of less nucleophilic substrates has been an
enduring challenge in chiral Brønsted acid catalysis. This
report describes our investigation regarding the use of
allyltrimethylsilanes5 as stable but potential reagents in
the chiral Brønsted acid catalysis. We developed a new
(4) For references of each reaction, see the Supporting Information.
(5) (a) Hosomi, A. Acc. Chem. Res. 1988, 21, 200–206. (b) For other
representative reviews, see the Supporting Information.
(1) (a) Akiyama, T. Chem. Rev. 2007, 107, 5744–5758. (b) Hydrogen
Bonding in Organic Synthesis; Pihko, P. M., Ed.; Wiley-VCH: Weinheim,
2009.
(2) (a) Terada, M. Chem. Commun. 2008, 4097–4112. (b) Terada, M.
Synthesis 2010, 1929–1982.
(6) (a) Ishihara, K.; Nakamura, S.; Kaneeda, M.; Yamamoto, H.
J. Am. Chem. Soc. 1996, 118, 12854–12855. (b) Hasegawa, A.; Naganawa,
Y.; Fushimi, M.; Ishihhara, K.; Yamamoto, H. Org. Lett. 2006, 8,
3175–3178. Akiyama, T.; Tamura, Y.; Itoh, J.; Morita, H.; Fuchibe, K.
Synlett 2006, 141–143. (d) Cheon, C. H.; Yamamoto, H. J. Am. Chem.
Soc. 2008, 130, 9246–9247. (e) Zamfir, A.; Tsogoeva, S. B. Org. Lett. 2010,
12, 188–191. (f) For other examples, see the Supporting Information.
(3) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36, 66–
77.
r
10.1021/ol200595b
Published on Web 03/24/2011
2011 American Chemical Society

approach, chiral Brønsted acid catalysis by a combined
Brønsted acid system, in which additional Brønsted acid
can facilitate proton transfer to silylated chiral Brønsted
acid (Scheme 1a) that is NOT the conventional mode
previously reported for silicon nucleophiles (Scheme 1b).6
acid 1b in toluene and ethyl acetate (entries 6 and 7). In
both cases, the reactions afforded higher enantioselectivity
that was observed when the reaction was conducted with
20 mol % of 1b in toluene (entry 2 vs 6) or in ethyl acetate
(entry 3 vs 7). This outcome suggested that (i) the sense of
the enantioselectivity is dominated by chiral phosphoric
acid 1b, (ii) the present catalysis does not allow chiral
phosphoric acid 1b to regenerate, and (iii) silylated phos-
phoric acid catalyzes the reaction quite slowly which
reduces enantioselectivity, indicating that the present reac-
tion proceeds via chiral Brønsted acid initiated silyl Lewis
acid catalysis among previously reported mechanisms.7
Scheme 1. Role of Additional Brønsted Acid in Combined
Brønsted Acid System
Table 1. Initial Survey of the Reaction of Allyltrimethylsilane
Because of the specific character as well as the low
nucleophilicity of allyltrimethylsilane, the proper choice
of Brønsted acidity, neither too strong nor too weak, has to
be of primary importance.7,8 The chiral phosphoric acids
possessing an electron-withdrawing group were chosen as
the chiral Brønsted acid catalyst in view of their relatively
strong acidity. Initial investigation using 20 mol % catalyst
loading revealed that while 20 mol % of 1a, with a 3,5-bis-
trifluoromethylphenyl group, was able to initiate the reac-
tion of N-acyl imine 2aa, enantioselectivity was very low
(Table 1, entry 1). In contrast, the reaction with 1b,
possessing a pentafluorophenyl group, afforded the homo-
allylamine 4aa with moderate enantioselectivity (entry 2).
Next, a survey of appropriate solvents for this reaction was
undertaken in the presence of 20 mol % of 1b. The use of
solvents other than toluene or ethyl acetate led to either a
significant decline in enantioselectivity or no reaction
(entries 2 and 3). To improve the chemical yields, the
reactions were examined at higher temperatures and/or
extended reaction times. When the reaction was conducted
at 40 °C for 2 days, no effects was observed on either
chemical yield or enantioselectivity (entry 4). In sharp
contrast, continuing the reaction for 7 days resulted in a
considerable improved in the chemical yield, accompanied
by a significant decrease in enantioselectivity (entry 5).
It is evident that the inferior enantioselectivity obtained
with extended reaction time may be due to the distinctive
pathways facilitated by different catalyst species during the
reaction. The details associated with the catalyst species
responsible for enantioselectivity were investigated by
employing stoichiometric amounts of chiral phosphoric
entrya 1 (mol %)
2
solvent temp (°C) yieldb (%) eec (%)
1
2
1a (20)
1b (20)
1b (20)
1b (20)
1b (20)
2aa toluene
2aa toluene
2aa EtOAc
2aa toluene
2aa toluene
30
30
30
40
40
30
30
30
30
30
20
19
34
18
56
80
95
77
93
94
5 (R)
60 (R)
50 (R)
58 (R)
45 (R)
68 (R)
55 (R)
97 (R)
95 (R)
98 (R)
3
4
5d
6
1b (100) 2aa toluene
1b (100) 2aa EtOAc
1b (100) 2ab toluene
1b (100) 2ab EtOAc
7
8
9
10
1c (100)
2ab EtOAc
a The reactions were conducted with 2a and 3 for 2 days in the
presence of 1. b Isolated yield. c Determined by chiral HPLC analysis.
d The reaction was conducted for 7 days.
Encouraged by this implementation of a chiral 1b in-
itiated HosomiꢀSakurai reaction, we conducted a survey
of structures with an acyl group using stoichiometric 1b to
improve the enantioselectivity.9 From this survey, the 3,5-
bis-tert-butylbenzoyl group (imine 2ab) surfaced as an
attractive substituent that afforded promising levels of
enantioselection (entries 8 and 9). In terms of the chemical
yield, reaction in ethyl acetate gave better results than that
in toluene (entry 9). Furthermore, the replacement of
binaphthyl with octahydrobinaphthyl as the catalyst led
to an increase in enantioselectivity (98% ee) with no
detrimental effect on the yield (entry 10).
(7) (a) Olah, G. A.; Laali, K.; Farooq, O. J. Org. Chem. 1984, 49,
4591–4594. (b) Davis, A. P.; Jaspars, M. J. Chem. Soc., Chem. Commun.
1990, 1176–1178. (c) Hollis, T. K.; Bosnich, B. J. Am. Chem. Soc. 1995,
117, 4570–4581.
(8) For recent examples, see: (a) Kampen, D.; List, B. Synlett 2006,
2589–2592. (b) Kampen, D.; Ladepeche, A.; Claben, G.; List., B. Adv.
Synth. Catal. 2008, 350, 962–966.
(9) For detailed experimental results, see the Supporting
Information.
Org. Lett., Vol. 13, No. 8, 2011
2127

Scheme 2. Combined Brønsted Acid Catalysis for Catalytic Enanioselective Process
With optimal imines and chiral phosphoric acid in hand,
we focused on transforming this reaction into a catalytic
asymmetric process. There are two plausible approaches
that the regeneration of chiral Brønsted acid could be
initiated by additional Brønsted acid as shown in Scheme
2: (a) attack on silane in the β-cation intermediate (cycle A)
and (b) proton transfer to silylated chiral phosphoric acid
(cycle B). The screening of the additional Brønsted acid is
summarized in Table 2. The use of phenol 5a and car-
boxylic acid 5b was expected to have an effect in cycle A
that would resemble previously reported additional
Brønsted acid in the reaction of silyl enolates, trapping
silane on a siloxocarbenium ion intermediate.6 However,
phenol 5a and carboxylic acid 5b did not prove to be
practical solutions for improvement of the product yield
(entries 2 and 3). Commercially available phosphoric acid
5c did not increase the product yield without loss of
enantioselectivity (entry 4). In contrast, although biphe-
nol-derived phosphoric acid 5da decreased the enantios-
electivity, the reaction did achieve high conversion with
moderate enantioselectivity (entry 5). We envisioned that
the substituents at the 3,30-position in 5d could be the key
to facilitate the regeneration of chiral phosphoric acid 1c
such that loss of enantioselectivity might be prevented.
After screening of substituents at the 3,30-position, biphenol-
derived phosphoric acid 5db, with its 2,6-dimethylphenyl
group, was fortuitously found to meet the criteria and
afforded the homoallylamine 4ab in good yield with high
enantioselectivity (entry 6).
To gain insight into the mode of additional Brønsted
acid 5db in a combined Brønsted acid system, the
following experiments were conducted. Treatment of
imine 2ab with allylsilane 3 in the presence of 20 mol %
of 1c under the same conditions as in Table 2, entry 1,
yielded the product 4ab and concomitantly formed the
expected silylated chiral phosphoric acid, and then a
further 80 mol % of 5db was added (Scheme 3). This
yielded the product 4ab in 81% yield with 92% ee, an
improvement on the product yield in Table 2, entry 1,
and the same as that in Table 2, entry 6. This experiment
clearly demonstrated that biphenol-derived phosphoric
acid 5db is able to involve proton transfer to regenerate
chiral phosphoric acid 1c from the silylated phosphoric
acid and that the reaction in the combined system might
proceed through cycle B in Scheme 2.
Table 2. Screening of Additional Brønsted Acid
entrya
5
yieldb (%)
eec (%)
Scheme 3. Control Experiment for Proton Transfer to Silylated
Chiral Phosphoric Acid
1
2
3
4
5
6
22
29
45
42
75
83
95
94
93
50
50
92
5a
5b
5c
5da
5db
a The reactions were conducted with 2ab and 3 at 30 °C in the
presence of 1c or 5. b Isolated yield. c Determined by chiral HPLC
analysis.
The scope of the reaction using chiral phosphoric acid 1c
was evaluated by investigating a range of imines 2 with a
3,5-di-tert-butyl benzoyl group on the imine nitrogen
2128
Org. Lett., Vol. 13, No. 8, 2011

Table 3. Scope of Chiral Phosphoric Acid Catalyzed Hosomiꢀ
Scheme 4. Enantio- and Diastereoselective Crotylation Reac-
tion of Acyl Imine 2ab
Sakurai Reaction
entrya
R
condition
yieldb (%)
eec (%)
1
2
Ph
A
B
A
B
A
B
A
B
A
B
A
B
A
B
94
83
70
58
88
54
92
76
92
75
84
80
70
66
98
92
94
82
96
93
97
87
95
86
92
75
96
83
3
2-MeC6H4
3-MeC6H4
4-MeC6H4
4-ClC6H4
which pentafluorophenyl groups on binaphthyl and/or the
bulkier tert-dibutyl benzoyl group on the imino nitrogen
could be playing an organizing role in this reaction.
In summary, we have demonstrated a chiral Brønsted
acid catalyzed highly enantio- and syn diastereoselective
HosomiꢀSakurai reaction of imines with allyl- and
crotyltrimethylsilane.11,12 The pentafluorophenyl intro-
duced chiral phosphoric acid exhibits robust activity to
induce stereoselectivities in the case of N-acyl imine 2ab. In
particular, a combined Brønsted acid system with chiral
phosphoric acid 1c and additional Brønsted acid 5db was
shown to be an enantioselective catalytic process. The
present system proceeds through regeneration of the chiral
Brønsted acid by proton transfer from additional Brønsted
acid to chiral silylated phosphoric acid, which is a newly
discovered function of biphenol-derived phosphoric acid
and a completely different mode than the previously
developed reaction of silicon nucleophiles. The combined
Brønsted acid system described herein is an innovative and
new entry into chiral Brønsted acid catalysis that should
find wide application in enantioselective synthesis using
stable silicon reagents.
4
5
6
7
8
9
10
11
12
13
14
4-CF3C6H4
4-MeOC6H4
a The reactions were conducted with 2 and 3 under condition A or B.
Condition A: 100 mol % of 1c. Condition B: 20 mol % of 1c and 80 mol %
of 5db. b Isolated yield. c Determined by chiral HPLC analysis.
(Table 3). Good yields and excellent enantiomeric excesses
were observed for each of the products formed when using
aromatic imines as the electrophile. Unfortunately, alipha-
tic imines proved not to be synthesized when the substi-
tuent on the imine nitrogen was a 3,5-di-tert-butyl benzoyl
group. It should be noted that although the reaction in the
combined system provided desired adducts in slightly
lower yields and enantioselectivities, high enantioselectiv-
ities were obtained in all cases even with 80 mol % of
additional Brønsted acid 5db in the reaction media. Further-
more, the crotylation reaction of (E)-crotyltrimethylsilane
(E/Z = 84/16) completed at 30 °C, afforded the syn adduct
7ab with high enantioselectivity (syn/anti = 93/7, 98% ee
(syn)) (Scheme 4). While the reaction of (Z)-crotyltrimethyl-
silane (E/Z = 1/>99) resulted in significantly lower con-
version at 30 °C, an increase in conversion was observed
when the reaction was run at 40 °C, providing syn adduct
7ab with good diastereo- and high enantioselectivity (syn/
anti = 84/16, 94% ee (syn)) (Scheme 4). This correlation of
crotyl geometry and product stereochemistry suggests that
the reaction proceeds via an acyclic transition-state,10 in
Acknowledgment. Support was provided by JSPS via
Grant-in-Aid for Young Scientists B (No. 21750087) and
MEXT via Grant-in-Aid for Scientific Research on Priority
Areas (No. 20037005 “Chemistry of Concerto Catalysis”).
Supporting Information Available. Detailed experimen-
tal procedures and characterization data for new com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(11) For representative example of enantioselective reaction of imi-
nes with allyltrimethylsilane, see: (a) Nakamura, K.; Nakamura, H.;
Yamamoto, Y. J. Org. Chem. 1999, 64, 2614–2615. (b) Fang, X.;
Johannsen, M.; Yao, S.; Gathergood, N.; Hazell., R. G.; Jørgensen,
K. A. J. Org. Chem. 1999, 64, 4844–4849. (c) Ferraris, D.; Young, B.;
Cox, C.; Dudding, T.; Drury, W. J.; Ryzhkov, L.; Taggi, A. E.; Lectka,
T. J. Am. Chem. Soc. 2002, 124, 67–77.
(10) (a) Yamamoto, Y.; Yatagai, H.; Naruta, Y.; Maruyama, K. J.
Am. Chem. Soc. 1980, 102, 7107–7109. (b) Yamamoto, Y.; Yatagai, H.;
Ishihara, Y.; Maeda, N.; Maruyama, K. Tetrahedoron 1984, 40, 2239–
2246. (c) Bottoni, A.; Costa, A. L.; Tommaso, D. D.; Rossi, I.;
Tagliavini, E. J. Am. Chem. Soc. 1997, 119, 12131–12135.
(12) For allylation of other examples using allylsilanes, see the
Supporting Information.
Org. Lett., Vol. 13, No. 8, 2011
2129