Asymmetric synthesis of chiral primary amines by transfer hydrogenation of N -(tert -Butanesulfinyl)ketimines

Article abstract of DOI:10.1021/jo101057s

(Figure presented) The diastereoselective reduction of (R)-N-(tert- butanesulfinyl)ketimines by a ruthenium-catalyzed asymmetric transfer hydrogenation process in isopropyl alcohol, followed by desulfinylation of the nitrogen atom, is an excellent method to prepare highly enantiomerically enriched α-branched primary amines (up to >99% ee) in short reaction times (1-4 h). (1S,2R)-1-Amino-2-indanol has been shown to be a very efficient ligand to perform this transformation. Ketimines bearing either an aryl or a heteroaryl group and an alkyl group as substituents of the iminic carbon atom are very good substrates for this process. The reduction of a dialkyl ketimine could also be achieved, affording the expected amine with moderate optical purity (69% ee). Some amines which are precursors of very interesting biologically and pharmacologically active compounds have been prepared in excellent yields and enantiomeric excesses.

Full text of DOI:10.1021/jo101057s

pubs.acs.org/joc
Asymmetric Synthesis of Chiral Primary Amines by Transfer
Hydrogenation of N-(tert-Butanesulfinyl)ketimines§
ꢀ
David Guijarro,* Oscar Pablo, and Miguel Yus*
´
ꢀ
Departamento de Quımica Organica, Facultad de Ciencias and Instituto de Sıntesis Organica (ISO),
ꢀ
´
Universidad de Alicante, Apdo. 99, 03080 Alicante (Spain).
dguijarro@ua.es; yus@ua.es
Received May 31, 2010
The diastereoselective reduction of (R)-N-(tert-butanesulfinyl)ketimines by a ruthenium-catalyzed
asymmetric transfer hydrogenation process in isopropyl alcohol, followed by desulfinylation of the
nitrogen atom, is an excellent method to prepare highly enantiomerically enriched R-branched
primary amines (up to >99% ee) in short reaction times (1-4 h). (1S,2R)-1-Amino-2-indanol has
been shown to be a very efficient ligand to perform this transformation. Ketimines bearing either an
aryl or a heteroaryl group and an alkyl group as substituents of the iminic carbon atom are very good
substrates for this process. The reduction of a dialkyl ketimine could also be achieved, affording the
expected amine with moderate optical purity (69% ee). Some amines which are precursors of very
interesting biologically and pharmacologically active compounds have been prepared in excellent
yields and enantiomeric excesses.
Introduction
(b) addition of organometallic reagents to imines,5 (c) asym-
metric reduction of ketimines,2,5e,6 and (d) enzyme-mediated
methods like transamination between carbonyl compounds
and amines2 and kinetic resolution of racemic amines.2,7 In
recent years, the reduction of compounds containing CdN
bonds has emerged as a powerful way of synthesizing chiral
amines. Due to the low electrophilic character of the CdN
bond, the nitrogen atom of these precursors is normally
bonded to an activating group, such as aryl, alkoxy, amino,
phosphinyl, sulfonyl, or sulfinyl. The latter is especially
interesting because its inherent chirality makes it act as a
chiral auxiliary apart from being an activating group for the
Chiral amines are present in many natural products and
biologically active compounds.1 R-Branched amines are con-
stituents of pharmaceutical compounds, and several thera-
peutic applications have been found for them.2 Optically
active amines have also extensively been used as chiral auxi-
liaries3 and resolving agents4 in asymmetric synthesis. Due to
all of these features, considerable efforts have been made by
chemists in order to design efficient procedures for their
stereoselective preparation. Among the reported methodol-
ogies to produce enantiomerically enriched primary amines,
we can find (a) resolution of racemic amines by crystalli-
zation of their diastereomeric salts with carboxylic acids,2
(5) For reviews on the asymmetric nucleophilic addition to imines, see:
(a) Enders, D.; Reinhold, U. Tetrahedron: Asymmetry 1997, 8, 1895–1946.
(b) Bloch, R. Chem. Rev. 1998, 98, 1407–1438. (c) Kobayashi, S.; Ishitani, H.
Chem. Rev. 1999, 99, 1069–1094. (d) Vilaivan, T.; Bhanthumnavin, W.;
Sritana-Anant, Y. Curr. Org. Chem. 2005, 9, 1315–1392. (e) Weinreb, S. M.;
Orr, R. K. Synthesis 2005, 1205–1227.
(6) (a) Johansson, A. Contemp. Org. Synth. 1995, 2, 393–407. For
catalytic enantioselective hydrogenation of imines, see: (b) Moessner, C.;
Bolm, C. Angew. Chem., Int. Ed. 2005, 44, 7564–7567 and references cited
therein. (c) Cheruku, P.; Church, T. L.; Andersson, P. G. Chem. Asian J.
2008, 3, 1390–1394.
§
ꢀ
Dedicated to Professor Carmen Najera on the occasion of her 60th birthday.
(1) (a) Scott, J. W. In Asymmetric Synthesis; Morrison, J. D., Ed.;
Academic Press: New York, 1985; Vol. 5, Chapter 1, pp 27-29.
(b) Lawrence, S. A. Amines: Synthesis, Properties and Applications;
Cambridge University Press: Cambridge, UK, 2004.
(2) Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; Kesseler, M.;
€
Sturmer, R.; Zelinski, T. Angew. Chem., Int. Ed. 2004, 43, 788–824.
(3) Whitesell, J. K. Chem. Rev. 1989, 89, 1581–1590.
(4) Jacques, J.; Collet, A.; Wilen, S. H. Enantiomers, Racemates and
Resolution; John Wiley & Sons: New York, 1981.
(7) Ditrich, K. Synthesis 2008, 2283–2287.
DOI: 10.1021/jo101057s
r
Published on Web 07/09/2010
J. Org. Chem. 2010, 75, 5265–5270 5265
2010 American Chemical Society

Guijarro et al.
JOCArticle
imine functionality and a protecting group for the nitrogen
atom of the reduction product. Furthermore, the sulfinyl
group can easily be removed under mild acidic conditions8
for further transformations. The tert-butanesulfinyl group is
particularly interesting because it has shown high levels of
asymmetric induction in a variety of processes. Since the ini-
tial studies carried out by Ellman, N-(tert-butanesulfinyl)-
imines have consolidated as excellent substrates for the
preparation of chiral primary amines.9 The diastereoselec-
tive reduction of N-(tert-butanesulfinyl)ketimines has been
achieved using several boranes,9b,c,10 sodium or lithium
borohydrides,9b,c,e,10,11 aluminum hydrides,9c,10,11 and diethyl-
zinc in the presence of Ni(acac)2.12
Among the different approaches to perform the reduction
of a carbon-heteroatom double bond, the transfer hydro-
genation protocol presents several advantages because it is
operationally simple, normally uses low catalyst loadings,
and avoids the handling of hazardous chemicals such as
metallic hydrides or molecular hydrogen. Moreover, volatile
reaction side products, such as acetone or carbon dioxide,
are formed, which facilitates the isolation of the reduction
products. The asymmetric transfer hydrogenation has widely
been applied to the synthesis of chiral secondary alcohols by
reduction of ketones.13 However, there are only a limited
number of examples of stereoselective preparation of amines
by this methodology. The asymmetric reduction of imines
bearing alkyl,13d aryl,13c,d benzyl,5c,13c,d phosphinyl,2,13d or
sulfonyl13b,d groups on the nitrogen atom and endocyclic
imines5c,13b-d by transfer hydrogenation has been published,
with the catalysts being ruthenium,14 rhodium, or iridium
complexes with chiral ligands such as monotosylated dia-
mines, β-aminoalcohols, diphosphines, and N-heterocyclic
carbenes. Some organocatalysts have also been employed in
the reduction of N-aryl- and endocyclic imines.13c,d How-
ever, to the best of our knowledge, the asymmetric transfer
hydrogenation has never been applied to the reduction of sul-
finylimines. As part of our research on the use of these useful
imines in asymmetric synthesis,15 herein we report our results
on the preparation of highly enantiomerically enriched ami-
nes by a ruthenium-catalyzed transfer hydrogenation of
N-(tert-butanesulfinyl)imines.16
Results and Discussion
The use of β-aminoalcohols as ligands for asymmetric
catalysis has been one of the subjects of our research activi-
ties during the last several years.17 Since one of the most suc-
cessful applications of these compounds has been as ligands
for ruthenium complexes used as catalysts for the transfer
hydrogenation of ketones,13 we decided to explore the pos-
sibility of extending this reduction methodology to N-(tert-
butanesulfinyl)imines. After screening several aminoalcohols,16
we found that (1S,2R)-1-amino-2-indanol18,19 3 (20 mol %)
gave a complex by reaction with [RuCl2(p-cymene)]2 (5 mol %)
that was able to catalyze the reduction of the imine derived
from acetophenone 1a (Table 1) in isopropyl alcohol at room
temperature and in the presence of KOH (50 mol %). The
yield of the reduction product 2a was quite low (31%), but we
were pleased to see that the ee was excellent (98%, Table 1,
entry 1). The high enantioselectivity obtained encouraged us
to do an optimization study of the reaction conditions since,
to the best of our knowledge, this was the first time that the
asymmetric transfer hydrogenation of an acyclic imine using
a β-aminoalcohol as a chiral ligand and isopropyl alcohol as
a hydrogen source had been achieved.20,21 Moreover, the
possibility of performing the reductions in isopropyl alcohol
was very attractive because it has been proven to be a
convenient solvent for industrial scale processes.22,23
(15) (a) Almansa, R.; Guijarro, D.; Yus, M. Tetrahedron: Asymmetry
2008, 19, 603–606. (b) Almansa, R.; Guijarro, D.; Yus, M. Tetrahedron:
Asymmetry 2008, 19, 2484–2491. (c) Almansa, R.; Guijarro, D.; Yus, M.
Tetrahedron Lett. 2009, 50, 3198–3201. (d) Almansa, R.; Guijarro, D.; Yus,
M. Tetrahedron Lett. 2009, 50, 4188–4190. (e) Almansa, R.; Collados, J. F.;
Guijarro, D.; Yus, M. Tetrahedron: Asymmetry 2010, 21, doi: 10.1016/j.
tetasy.2010.03.046.
(8) See, for instance: (a) Sun, X.; Wang, S.; Sun, S.; Zhu, J.; Deng, J.
Synlett 2005, 2776–2780. (b) Kosciolowicz, A.; Rozwadowska, M. D. Tetra-
hedron: Asymmetry 2006, 17, 1444–1448. (c) Denolf, B.; Mangelinckx, S.;
Toernroos, K. W.; De Kimpe, N. Org. Lett. 2006, 8, 3129–3132.
(9) (a) Ellman, J. A.; Owens, T. D.; Tang, T. P. Acc. Chem. Res. 2002, 35,
984–995. (b) Ellman, J. A. Pure Appl. Chem. 2003, 75, 39–46. (c) Zhou, P.;
Chen, B.-C.; Davis, F. A. Tetrahedron 2004, 60, 8003–8030. (d) Lin, G.-Q.;
Xu, M.-H.; Zhong, Y.-W.; Sun, X.-W. Acc. Chem. Res. 2008, 41, 831–840.
(16) Preliminary communication: Guijarro, D.; Pablo, O.; Yus, M.
Tetrahedron Lett. 2009, 50, 5386–5388.
(17) (a) Almansa, R.; Guijarro, D.; Yus, M. Tetrahedron: Asymmetry
2007, 18, 896–899. (b) Almansa, R.; Guijarro, D.; Yus, M. Tetrahedron:
Asymmetry 2007, 18, 2828–2840. (c) Almansa, R.; Guijarro, D.; Yus, M.
Tetrahedron: Asymmetry 2008, 19, 1376–1380.
ꢀ
(e) Ferreira, F.; Botuha, C.; Chemla, F.; Perez-Luna, A. Chem. Soc. Rev.
2009, 38, 1162–1186. (f) Robak, M. T.; Herbage, M. A.; Ellman, J. A. Chem.
Rev. 2010, 110, 3600–3740.
(18) Both enantiomers of cis-1-amino-2-indanol have been extensively
used as ligands for the asymmetric transfer hydrogenation of ketones. See,
for instance: (a) Palmer, M.; Walsgrove, T.; Wills, M. J. Org. Chem. 1997, 62,
5226–5228. (b) Blacker, A. J.; Mellor, B. J. Patent WO 9842643A1, 1998.
(c) Wills, M.; Palmer, M.; Smith, A.; Kenny, J.; Walsgrove, T. Molecules
2000, 5, 4–18. (d) Palmer, M. J.; Kenny, J. A.; Walsgrove, T.; Kawamoto,
A. M.; Wills, M. J. Chem. Soc., Perkin Trans. 1 2002, 416–427. (e) Hansen,
K. B.; Chilenski, J. R.; Desmond, R.; Devine, P. N.; Grabowski, E. J. J.;
Heid, R.; Kubryk, M.; Mathre, D. J.; Varsolona, R. Tetrahedron: Asymme-
try 2003, 14, 3581–3587. (f) Sun, X.; Manos, G.; Blacker, J.; Martin, J.;
Gavriilidis, A. Org. Process Res. Dev. 2004, 8, 909–914. (g) Mogi, M.; Fuji,
K.; Node, M. Tetrahedron: Asymmetry 2004, 15, 3715–3717. (h) Wu, X.; Li,
X.; McConville, M.; Saidi, O.; Xiao, J. J. Mol. Catal. A 2006, 247, 153–158.
(i) Sun, X.; Gavriilidis, A. Org. Process Res. Dev. 2008, 12, 1218–1222.
(19) A rhodium complex bearing N-tosyl-(1S,2R)-1-amino-2-indanol as a
ligand has been used as a catalyst for the transfer hydrogenation of an
endocyclic imine. See ref 18b.
(10) (a) Chelucci, G.; Baldino, S.; Chessa, S.; Pinna, G. A.; Soccolini, F.
Tetrahedron: Asymmetry 2006, 17, 3163–3169. (b) Zhao, C.-H.; Liu, L.;
Wang, D.; Chen, Y.-J. Eur. J. Org. Chem. 2006, 2977–2986. (c) Dutheuil, G.;
Couve-Bonnaire, S.; Pannecoucke, X. Angew. Chem., Int. Ed. 2007, 46, 1290–
1292. (d) Han, Z.; Koenig, S. G.; Zhao, H.; Su, X.; Singh, S. P.; Bakale, R. P.
Org. Process Res. Dev. 2007, 11, 726–730. (e) Liu, Z.-J.; Liu, J.-T. Chem.
Commun. 2008, 5233–5235. (f) Martjuga, M.; Shabashov, D.; Belyakov, S.;
Liepinsh, E.; Suna, E. J. Org. Chem. 2010, 75, 2357–2368.
(11) (a) Peltier, H. M.; Ellman, J. A. J. Org. Chem. 2005, 70, 7342–7345.
(b) Colyer, J. T.; Andersen, N. G.; Tedrow, J. S.; Soukup, T. S.; Faul, M. M.
J. Org. Chem. 2006, 71, 6859–6862. (c) Grajewska, A.; Rozwadowska, M. D.
Tetrahedron: Asymmetry 2007, 18, 557–561. (d) Tanuwidjaja, J.; Peltier,
H. M.; Ellman, J. A. J. Org. Chem. 2007, 72, 626–629. (e) Denolf, B.;
Leemans, E.; De Kimpe, N. J. Org. Chem. 2007, 72, 3211–3217.
(12) Xiao, X.; Wang, H.; Huang, Z.; Yang, J.; Bian, X.; Qin, Y. Org. Lett.
2006, 8, 139–142.
€
(20) Backvall has reported the non-stereoselective ruthenium-catalyzed
transfer hydrogenation of imines in refluxing isopropyl alcohol in the
€
presence of K2CO3: Wang, G.-Z.; Backvall, J.-E. J. Chem. Soc., Chem.
Commun. 1992, 980–982.
(21) Andersson has reported the synthesis of chiral aziridines by a
ruthenium-catalyzed asymmetric transfer hydrogenation of strained 2H-
azirines in isopropyl alcohol using a β-aminoalcohol as a chiral ligand:
Roth, P.; Andersson, P. G.; Somfai, P. Chem. Commun. 2002, 1752–1753.
(13) (a) Noyori, R.; Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97–102.
(b) Palmer, M. J.; Wills, M. Tetrahedron: Asymmetry 1999, 10, 2045–2061.
(c) Wang, C.; Wu, X.; Xiao, J. Chem. Asian J. 2008, 3, 1750–1770. (d) Wills,
M. In Modern Reduction Methods; Andersson, P. G., Munslow, I. J., Eds.;
Wiley-VCH: Weinheim, Germany, 2008; pp 271-296. (e) Malacea, R.; Poli,
R.; Manoury, E. Coord. Chem. Rev. 2010, 254, 729–752.
(14) Kitamura, M.; Noyori, R. In Ruthenium in Organic Synthesis;
Murahashi, S.-I., Ed.; Wiley-VCH: Weinheim, Germany, 2004; pp 3-52.
5266 J. Org. Chem. Vol. 75, No. 15, 2010

Guijarro et al.
JOCArticle
TABLE 1. Optimization of the Reaction Conditionsa
base
product
entry
Ru dimer (mol %)
3 (mol %)
base
(mol %)
4 A MS (g)
T (°C)
tb (h)
yieldc (%)
eed (%)
1
2
3
4
5
6
7
8
9
10
11
12f
13g
5
5
5
5
5
5
5
3
8
5
3
5
5
20
20
20
20
10
20
10
6
16
10
6
10
10
KOH
50
50
50
50
50
25
25
15
40
25
15
25
25
25
25
25
25
25
25
25
25
25
40
40
40
40
15
16
19
20
22
24
20
30
5
2
4
3
2
31e
76
74
95
97
97
98
84
99
99
67
83
96
98
99
99
>99
>99
>99
>99
99
99
98
96
98
t-BuOK
t-BuOK
t-BuOK
t-BuOK
t-BuOK
t-BuOK
t-BuOK
t-BuOK
t-BuOK
t-BuOK
t-BuOK
t-BuOK
0.2
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
97
aThe solution of imine 1a (0.9 mmol) in i-PrOH (9 mL) was added to a solution of the ruthenium catalyst [prepared by refluxing a mixture of [RuCl2(p-
cymene)]2, aminoalcohol 3, and 4 A molecular sieves in i-PrOH (2 mL)] at the temperature indicated. Then, the base (as a 0.1 M solution in i-PrOH) was
added and the reaction was stirred at the same temperature for the time indicated. bTime for the transfer hydrogenation reaction. cIsolated yield of amine
2a after acid-base extraction based on the starting imine 1a. Compound 2a was g95% pure (300 MHz 1H NMR). dDetermined for the corresponding
benzamide by HPLC using a ChiralCel OD-H column. The R-enantiomer was the major one in all cases. eAcetophenone, t-BuSONH2 and
1-phenylethanol were also obtained in the crude reaction mixture. fA solution of the imine 1a (0.9 mmol) in i-PrOH (5 mL) was used. gA solution of
the imine 1a (0.9 mmol) in i-PrOH (18 mL) was used.
The detection of acetophenone, t-BuSONH2 and 1-phe-
nylethanol (the latter resulting from the reduction of
acetophenone) in the crude reaction mixture suggested to
us that a hydrolysis of the sulfinylimine was occurring, which
was certainly the reason for the low yield obtained. Since we
had used anhydrous isopropyl alcohol and a non-aqueous
workup procedure was carried out (simply filtration through
silica gel to remove the ruthenium complex and evaporation
of the solvent), we suspected that the water that was forming
after the deprotonation with KOH was responsible for the
decomposition of the imine. Therefore, we decided to change
the base to t-BuOK, and we were glad to see that the yield of
the reduction product improved to 76%, keeping the excel-
lent ee value (Table 1, entry 2). To be sure that all traces
of water had been removed, we repeated the reaction in the
presence of 0.2 g of activated 4 A molecular sieves, but this
additive did not seem to cause any improvement (Table 1,
entry 3). However, when 0.5 g of 4 A molecular sieves was
used, a much higher yield of 2a (95%) was obtained (Table 1,
entry 4).
not cause any detriment to the ee (compare entries 4 and 5
in Table 1). When the amount of t-BuOK was reduced to
25 mol %, keeping the amounts of Ru complex and ligand
constant, the result did not vary (compare entry 4 with 6 and
5 with 7 in Table 1). After these observations, we decided to
use the ratio Ru dimer/3/t-BuOK = 1:2:5 in further tests.
The reduction of the amount of the Ru dimer to 3 mol %
slowed the transfer hydrogenation reaction and led to a lower
yield of 2a, although the stereoselectivity was still excellent
(compare entries 7 and 8 in Table 1). When 8 mol % of the Ru
complex was used, the reaction was much faster than the one
with 5 mol % of it, with the yield and ee of the amine 2a being
almost the same in both cases (compare entries 7 and 9 in
Table 1). For this reason, we did not consider it necessary to
use more than 5 mol % of the Ru dimer as the catalyst pre-
cursor. The effect of an increase of the reaction temperature
was also studied, and we found that a quantitative yield of
the amine 2a was obtained in only 2 h with a very slight decrease
in the ee when the transfer hydrogenation was performed at
40 °C (Table 1, entry 10). This fast reaction encouraged us to try
to reduce the amount of the ruthenium precatalyst to 3 mol %,
but an important reduction in yield and a lower enantioselec-
tivity were obtained (Table 1, entry 11). Finally, the influence of
the reaction concentration was investigated, and it was ob-
served that neither a higher (entry 12) nor a lower (entry 13)
concentration improved the results in comparison with the ones
obtained under the conditions of entry 10. After all of these
tests, we chose conditions of entry 10 as the optimum ones be-
cause they gave a good combination of short reaction time and
excellent yield and enantioselectivity.
We next tried to find the optimum proportion between the
reagents. The reduction of the Ru dimer/3 ratio to 1:2 did
(22) See, for instance: (a) Reference 18e. (b) Blacker, J.; Martin, J. In
Asymmetric Catalysis on Industrial Scale; Blaser, H. U., Schmidt, E., Eds.;
Wiley-VCH: Weinheim, Germany, 2004; pp 201-216.
(23) The use of tert-butanesulfinamide as a precursor of the sulfinylimines
might appear as an economical drawback for possible industrial scale
applications of our methodology. However, the cost of the process could
be diminished taking into account that an efficient procedure for the
recycling of the sulfinyl chiral auxiliary has been recently reported:
Wakayama, M.; Ellman, J. A. J. Org. Chem. 2009, 74, 2646–2650.
J. Org. Chem. Vol. 75, No. 15, 2010 5267

Guijarro et al.
JOCArticle
TABLE 2. Ruthenium-Catalyzed Asymmetric Transfer Hydrogenation of Imines 1a
product
entry
imine
1a
1b
R1
R2
tb (h)
no.
2a
2b
2c
yieldc (%)
eed (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Ph
Ph
Ph
Ph
Ph
Me
Et
2
4
3
99
98
94
95
97
97h
99
98
97
97
99
96
95
81
92
91
82
97
98
97
98
98
1c
1d
Pr
CH2Cl
(E)-CHdCHPh
99
5
2d
99e
98
1ef
0 .5
2.5
3
2eg
2f g
2g
1f
1g
1h
1i
1j
1k
1l
4-BocNHC6H4
4-MeOC6H4
3-ClC6H4
4-ClC6H4
4-CF3C6H4
4-NO2C6H4
2-naphthyl
2-furyl
Me
Me
Me
Me
Me
Me
Me
Me
Me
---
98
95
>99
>99
99
97
98
99
97
2
2h
2i
2j
2k
2.5
1.5
1
2
3
8
3.5
5
24
3
2.5
3
2l
1m
1n
2m
2n
2-thienyl
---
---
1og
1pg
1qi
2og
2pg
2q
93
98
69
---
Cy
Ph
Me
Me
Me
Me
ent-1ag
ent-1ag,k
ent-1rg,k
ent-2ag
ent-2ag
ent-2rg
84j
98j
>99j
Ph
3-MeOC6H4
aThe solution of imine 1 (0.9 mmol) in i-PrOH (9 mL) was added to a solution of the ruthenium catalyst [prepared by refluxing a mixture of [RuCl2(p-
cymene)]2 (0.045 mmol), aminoalcohol 3 (0.09 mmol), and 4 A molecular sieves (0.5 g) in i-PrOH (2 mL)] at 40 °C. Then, t-BuOK (0.225 mmol, as a 0.1 M
solution in i-PrOH) was added, and the reaction was stirred at the same temperature for the time indicated. bTime for the transfer hydrogenation
reaction. cIsolated yield of amine 2 after acid-base extraction based on the starting imine 1. All compounds 2 were g95% pure (300 MHz 1H NMR).
e
dDetermined for the corresponding benzamide by HPLC using a ChiralCel OD-H column. The R-enantiomer was the major one in all cases. The
absolute configuration of this product was S due to the priority order of the groups according to the CIP rules. fThis reaction was performed at 20 °C.
gFor the structure of this compound, see Chart 1. hCompound 2f was isolated as its dihydrochloride. iThis reaction was performed at 80 °C using 10 mol
% of Ru dimer, 20 mol % of ligand 3, and 50 mol % of t-BuOK. jThe S-enantiomer was the major one in this case. kThe (SS)-imine ent-1 and (1R,2S)-1-
amino-2-indanol (the enantiomer of ligand 3) were used in this reaction.
CHART 1
The transfer hydrogenation of ketones in isopropyl alco-
hol is a reversible process, and an unnecessarily long reaction
time can cause a reduction of the enantiomeric excess of the
desired secondary alcohol, as it has been observed by several
research groups.13a,18a,d,f,i,22b Concerned about this pro-
blem, we investigated whether a prolonged reaction time
could cause any decrease in the diastereoselectivity of our
transfer hydrogenation procedure under the optimum reac-
tion conditions. Two aliquots were taken from a reduction of
1a at reaction times of 4.5 and 24 h. No detriment in the ee
(98%) of the final amine 2a was observed in any of them,
which seems to indicate that the transfer hydrogenation of
the sulfinylimines in isopropyl alcohol is not a reversible
process.
Once the optimum reaction conditions had been establis-
hed, we explored the substrate scope, and all of the obtained
results are summarized in Table 2. Imines 1a-c, bearing
different straight chain alkyl substituents, led to the expected
primary amines 2a-c in very good yields and enantioselec-
tivities (Table 2, entries 1-3). The chloro-substituted imine
1d also gave an excellent ee in the reduction product 2d
(Table 2, entry 4). The transfer hydrogenation of imine 1e,
derived from chalcone, was very fast. It was not necessary to
raise the temperature to 40 °C since the reaction at room
temperature was finished in 30 min, affording product 2e
(Chart 1), resulting from reduction of both the imine and the
CdC bond (Table 2, entry 5). This reaction was repeated
at -40 °C in an attempt to get a selective reduction of one of
1
the unsaturated bonds. The reaction was monitored by H
NMR, and it seemed that the reduction of the CdC bond
was taking place faster than the reduction of the CdN bond.
After a reaction time of 10 min, a sample was withdrawn and
worked up, and its 1H NMR spectrum showed that the main
5268 J. Org. Chem. Vol. 75, No. 15, 2010

Guijarro et al.
JOCArticle
product was the one resulting from reduction of the olefin
(ratio of 1e/monoreduction product = 3:1). However, the
signals corresponding to the double reduction product could
be detected, too, which suggested that the selective reduction
of the CdC bond would not be possible. Further evolution of
the reaction confirmed our suspicion: after 2 h, the ratio of
1e/monoreduction product/double reduction product was
1.1:0.9:0.14.
SCHEME 1
Next, we studied the effect of several substituents on the
aromatic ring bonded to the iminic carbon atom. Almost
quantitative yields and excellent enantiomeric excesses were
obtained in the transfer hydrogenation of acetophenones
bearing either electron-releasing (Table 2, entries 6 and 7) or
electron-withdrawing groups (Table 2, entries 8-11). The
same good results were found when isomeric imines 1h and 1i
were reduced: very high yields and enantioselectivities were
obtained when the chloro substituent was either in meta or
para position to the imine functionality. However, the re-
duction of the imine derived from o-chloroacetophenone was
much slower: after the reaction was stirred at 40 °C for 8 h, a
conversion of less than 20% was observed. In the case of the
imine 1f, the Boc protecting group was also removed during
the desulfinylation process, affording the diamine 2f (Chart 1),
which was isolated as its dihydrochloride due to problems in
the purification of the free diamine.
Some other aromatic systems different from benzene
were introduced in the iminic substrate. The 2-naphthyl-sub-
stituted imine 1l was reduced with the same efficiency as the
acetophenone-derived imines (Table 2, entry 12). Our metho-
dology was shown to be equally effective for the pre-
paration of heterocyclic amines 2m and 2n (Table 2, entries 13
and 14) by reduction of imines bearing a 2-furyl or a 2-thienyl
substituent, respectively. Benzo-condensed cyclic imines 1o and
1p (Chart 1), derived from 1-indanone and 4-chromanone,
respectively, also yielded the expected amines 2o and 2p
(Chart 1) with very high optical purity (Table 2, entries 15 and
16). We also tried the reduction of the aliphatic imine 1q. As
expected, it was much less reactive than the aromatic imines: the
reaction had to be set up at 80 °C in order to get full conversion
of the imine after 24 h. However, the isolated yield of amine 2q
was good, and the 69% ee that was obtained is comparable to
the enantioselectivities reported for other transfer hydrogena-
tion processes with aliphatic imines.24 Moreover, to the best of
our knowledge, this is the first time that the asymmetric transfer
hydrogenation of an aliphatic imine using isopropyl alcohol as
hydrogen source has been reported.
transfer hydrogenation of imine 1a using ligand 3 gave the
(R)-amine 2a with a 98% ee (Table 2, entry 1), which is higher
than the one obtained for the reduction of the same imine
with the enantiomer of the ligand (1R,2S)-1-amino-2-inda-
nol (ent-3) that was 84% ee (data not shown in Table 2). The
latter ee value is equal to the one that we had obtained using
simple 2-aminoethanol as a ligand (see ref 16). This suggests
that maybe it is not necessary to use a chiral ligand to achieve
high stereoselectivity levels.25 The search for a ligand with the
appropriate substituents, even if achiral, would probably im-
prove the diastereoselectivity of the transfer hydrogenation
process. This possibility is currently being explored in our
laboratory, and the results will be reported in due course. A
theoretical study is also in progress in order to ascertain the role
of the sulfinyl group and a possible mechanism for the asym-
metric transfer hydrogenation process. The enantiomeric purity
of the (S)-amine ent-2a could be improved by preparing the
ruthenium catalyst with the ligand ent-3 (Table 2, entry 19),
reproducing the 98% ee that had been obtained in the reduction
of the (RS)-imine 1a using ligand 3. Another (SS)-imine sub-
stituted with a methoxy group, ent-1r (Chart 1), was submitted
to the transfer hydrogenation process using the enantiomer of
ligand 3, and the expected (S)-amine ent-2r (Chart 1) was
obtained in optically pure form in 97% yield (Table 2, entry 20).
The usefulness of this reduction procedure becomes clear
if we think about possible applications of the obtained
primary amines. Some of these amines are potential precur-
sors of pharmaceuticals or biologically active compounds.
For instance, the p-chloro-substituted amine 2i is a key inter-
mediate in the synthesis of the agricultural fungicide
Carpropamid26 (Scheme 1), which is useful for the control
of rice blast. The synthesis of amine 2i in 90% ee by crystal-
lization of the racemic amine with a chiral carboxylic acid has
been described,2 with the consequent loss of 50% of the yield.
With our methodology, this amine was obtained in 97%
yield in optically pure form (Table 2, entry 9). On the other
hand, the (S)-amine ent-2r is a suitable substrate for the
synthesis of Rivastigmine (Scheme 1), which is a cholinester-
ase inhibitor that is used to treat moderate dementia in
people with Alzheimer’s or Parkinson’s diseases27 and is on
Finally, we also explored the possibility of preparing the
enantiomeric amines ent-2 by application of our transfer
hydrogenation protocol to (SS)-sulfinimines. The reduction
of imine ent-1a (Chart 1) catalyzed by the same ruthenium
complex and ligand as before gave the (S)-amine ent-2a
(Chart 1) in very high yield and with an 84% ee (Table 2,
entry 18). The comparison of this result with the one obtai-
ned with the (RS)-imine 1a (Table 2, entry 1) leads to the
conclusion that the configuration of the imine plays a leading
role in determining the stereochemical outcome of the reac-
tion, but the structure of the ligand also has an influence on
the stereoselectivity. The use of ligand 3 for the reduction of
the (RS)-imines turned out to be a matched combination: the
(25) We thank the reviewers for suggesting that we comment on the
possibility of using an achiral ligand.
(26) (a) Tsuji, G.; Takeda, T.; Furusawa, I.; Horino, O.; Kubo, Y. Pestic.
Biochem. Physiol. 1997, 57, 211–219. (b) Merz, W.; Littmann, M.; Kraatz,
U.; Mannheims, C. Ger. Offen. 1998, DE 19624820 A1; Chem. Abstr. 1998,
128, 22711.
(27) (a) Spencer, C. M.; Noble, S. Drugs Aging 1998, 13, 391–411.
(b) Birks, J.; Grimley Evans, J.; Iakovidou, V.; Tsolaki, M.; Holt, F. E.
Cochrane Database Syst. Rev. 2009, CD001191, doi: 10.1002/14651858.
CD001191.pub2.
(24) See, for instance: Renat, K.; Thomas, H. R. Angew. Chem., Int. Ed.
2003, 42, 5472–5474.
J. Org. Chem. Vol. 75, No. 15, 2010 5269

Guijarro et al.
JOCArticle
the market as its tartrate salt under the brand name Exelon.
Recently, a total synthesis of (S)-Rivastigmine in three steps
from amine ent-2r has been reported.28 In that article, the
required amine ent-2r was prepared from 3-methoxyaceto-
phenone in five synthetic steps, including an enzyme-
mediated dynamic kinetic resolution, with an overall yield
of 55% and an ee of 99%. Another reported procedure for
the synthesis of amine ent-2r involves the copper-catalyzed
addition of dimethylzinc to 3-methoxy-N-(diphenylphosphi-
noyl)benzaldimine and gives the expected amine in 88%
yield and 92% ee.29 Our methodology is a very good alter-
native to these two approaches since we obtained amine
ent-2r from 3-methoxyacetophenone in only three steps in an
overall yield of 82% and with an optical purity of >99%.
During our assays to reduce different substrates, we found
some difficulties with certain imines derived from ketones
with substituents close to the carbonyl group. For instance,
when the transfer hydrogenation procedure was attempted
with the imines derived from methyl 1-naphthyl ketone,
isobutyrophenone, cyclohexyl phenyl ketone, or 2-methyl-
benzophenone, only traces of the expected amines were dete-
cted in the 1H NMR spectra of the crude reaction mixtures,
the major component of these mixtures being the corre-
sponding unaltered imines. Some attempts to reduce imines
bearing nitrogen-containing heterocyclic substituents, such
as 2- or 4-pyridyl, were unsuccessful, too. In these cases, a
possible coordination of the imine to the ruthenium complex
through those nitrogen atoms could compete with the co-
ordination of the iminic nitrogen, which would prevent the
transfer hydrogenation process.
of our knowledge, this is the first time that the asymmetric
transfer hydrogenation of an acyclic imine and using a β-
aminoalcohol as chiral ligand has been achieved in isopropyl
alcohol, which is a very interesting solvent for designing pro-
cedures applicable to the chemical industry.
Experimental Section
General Procedure for the Asymmetric Transfer Hydrogena-
tion of Imines 1. A mixture of [RuCl2(p-cymene)]2 (28 mg, 0.045
mmol), the ligand 3 (14 mg, 0.09 mmol), 4 A molecular sieves
(0.5 g), and anhydrous i-PrOH (2 mL) was heated to 90 °C (oil
bath temperature) for 20 min. During this heating period, the
initially orange reaction mixture turned into a dark red color.
The reaction was then cooled to 40 °C, and a solution of the
imine 1 (0.9 mmol) in i-PrOH (9 mL) and t-BuOK (2.25 mL of a
0.1 M solution in i-PrOH, 0.225 mmol) was successively added.
After completion of the reaction (monitored by TLC), the
reaction mixture was passed through a small column of silica
gel, the column was washed with ethyl acetate, and the combined
organic phases were evaporated to give a residue that was
directly submitted to the desulfinylation step.
General Procedure for the Removal of the Sulfinyl Group.
Isolation of Amines 2. The crude mixture of the transfer hydro-
genation reaction was dissolved in a 1.5 M solution of HCl in
methanol (4 mL; prepared by dropwise addition of SOCl2 to
methanol at 0 °C) and stirred overnight at room temperature.
Then, the solvent was evaporated, a 2 M aqueous HCl solution
(5 mL) was added, and the mixture was extracted with ethyl
acetate (3 ꢀ 5 mL). The organic layers were discarded. The
aqueous layer was basified with a buffer solution of NH3 (1 M)/
NH4Cl (1 M) and extracted with CH2Cl2 (3 ꢀ 10 mL). The com-
bined organic phases were dried (Na2SO4). After filtration and
evaporation of the solvent, pure amines 2 were obtained in the
yields indicated in Table 2. The corresponding physical and
spectroscopic data for the representative compound 2i follow:
(R)-1-(4-Chlorophenyl)ethanamine (2i):30 Colorless oil; Rf
Conclusion
We have presented here a ruthenium-catalyzed asym-
metric transfer hydrogenation procedure that reduces N-
(tert-butanesulfinyl)imines in isopropyl alcohol in short re-
action times and with very high diastereoselectivities. The
methodology is very efficient for the reduction of imines
bearing one aromatic substituent on the iminic carbon atom.
The reduction of one aliphatic imine has been also achieved,
although the stereoselectivity was lower in this case. The dia-
stereoselectivity seems to be controlled mainly by the abso-
lute configuration of the starting imine. An important fea-
ture of this process is that, contrary to the decrease of ee that
has been observed in the transfer hydrogenation of ketones,
there is no detriment in the de of the reduction products with
extended reaction times. Since the sulfinyl group can be
easily removed from the reduction products, this protocol re-
presents a very efficient way to prepare R-branched primary
amines with very high optical purities. Both enantiomers of
tert-butanesulfinamide, needed to prepare the starting imines,
and of the ligand used are commercially available, which
allows the preparation of both enantiomers of the final
amines in highly enantiomerically enriched form. To the best
0.33 (ethyl acetate, deactivated silica gel); [R]20 þ31.0 (c 1.2,
D
CHCl3, >99% ee) {literature30 [R]20D þ18.9 (c 1.0, CHCl3, 92%
ee)}; IR (neat) 3365, 3285, 3046, 1590, 1491, 1093 cm-1 1H
;
NMR (400 MHz, CDCl3) δ 1.35 (3H, d, J = 6.7 Hz), 1.56 (2H,
br s), 4.10 (1H, q, J = 6.7 Hz), 7.24-7.31 (4H, m); 13C NMR
(100 MHz, CDCl3) δ 25.7, 50.6, 127.1, 128.5, 132.2, 146.1; m/z
155 (Mþ, 3%), 142 (25), 140 (100), 139 (14), 138 (31), 137 (11), 77
(23), 75 (18), 51 (10).
Acknowledgment. This work was generously supported
ꢀ
by the Spanish Ministerio de Educacion y Ciencia (MEC)
(Grant Nos. CONSOLIDER INGENIO 2010, CSD2007-
00006, and CTQ2007-65218) and the Generalitat Valenciana
(PROMETEO/2009/039 and GV/2007/036). O.P. thanks the
ꢀ
Spanish Ministerio de Educacion for a predoctoral fellowship
(Grant No. AP2008-00989). We also thank MEDALCHEMY
S.L. for a gift of chemicals.
Supporting Information Available: Complete experimental
procedures, characterization data, and copies of 1H and 13C
NMR spectra for imines 1 and amines 2. This material is
available free of charge via the Internet at http://pubs.acs.org.
ꢀ
(28) Mangas-Sanchez, J.; Rodrı
´
ꢀ
guez-Mata, M.; Busto, E.; Gotor-
Fernandez, V.; Gotor, V. J. Org. Chem. 2009, 74, 5304–5310.
^ ꢀ
(29) Boezio, A. A.; Pytkowicz, J.; Cote, A.; Charette, A. B. J. Am. Chem.
Soc. 2003, 125, 14260–14261.
(30) Cheemala, M. N.; Knochel, P. Org. Lett. 2007, 9, 3089–3092.
5270 J. Org. Chem. Vol. 75, No. 15, 2010