Enantioselective addition of dialkylzinc reagents to N-(diphenylphosphinoyl)imines catalyzed by β-aminoalcohols with the prolinol skeleton

Article abstract of DOI:10.1016/j.tetasy.2007.11.006

Several β-aminoalcohols with the prolinol framework are shown to be very efficient catalysts for the enantioselective addition of dialkylzinc reagents to N-(diphenylphosphinoyl)imines. The use of 0.5 equiv of the catalyst leads to the expected addition products in good yields and with ee up to 94% in a reaction time of only 4 h at room temperature. This ee is the highest value reported so far using 0.5 equiv of an aminoalcohol as a promoter. High enantioselectivities are obtained in the addition of dialkylzincs to both aromatic and aliphatic imines. The amount of the catalyst can be reduced to 0.25 equiv with a slight decrease in the ee. A very interesting effect of the addition rate and temperature on the enantioselectivity was also observed.

Full text of DOI:10.1016/j.tetasy.2007.11.006

Available online at www.sciencedirect.com
Tetrahedron: Asymmetry 18 (2007) 2828–2840
Enantioselective addition of dialkylzinc reagents to
N-(diphenylphosphinoyl)imines catalyzed by b-aminoalcohols
with the prolinol skeleton
Raquel Almansa, David Guijarro* and Miguel Yus*
Departamento de Quı´mica Orga´nica, Facultad de Ciencias and Instituto de Sı´ntesis Orga´nica (ISO), Universidad de Alicante,
Apdo. 99, 03080 Alicante, Spain
Received 26 October 2007; accepted 7 November 2007
Dedicated to the memory of Professor Al Meyers
Abstract—Several b-aminoalcohols with the prolinol framework are shown to be very efficient catalysts for the enantioselective addition
of dialkylzinc reagents to N-(diphenylphosphinoyl)imines. The use of 0.5 equiv of the catalyst leads to the expected addition products in
good yields and with ee up to 94% in a reaction time of only 4 h at room temperature. This ee is the highest value reported so far using
0.5 equiv of an aminoalcohol as a promoter. High enantioselectivities are obtained in the addition of dialkylzincs to both aromatic and
aliphatic imines. The amount of the catalyst can be reduced to 0.25 equiv with a slight decrease in the ee. A very interesting effect of the
addition rate and temperature on the enantioselectivity was also observed.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
cesses.5 Concerning the carbon nucleophiles, dialkylzinc
reagents are very useful since organozinc reagents8 bearing
several functional groups can be easily prepared,9 and can
lead to polyfunctionalized organic compounds. However,
the reaction of N-phosphinoylimines with dialkylzincs is
very slow, leading to low yields of addition products in very
long reaction times. An improvement of both reaction rate
and yield has been achieved by using some additives, such
as b-aminoalcohols,7,10 iminoalcohols,10j hydroxyoxazo-
lines11 and copper,12 zirconium13 or nickel14 complexes.
Concerning chiral b-aminoalcohols, there is a trend to
believe that a rigid backbone in the ligand is generally
required to obtain high enantioselectivity.7,10c,e–g However,
the synthesis of structurally rigid aminoalcohols often in-
volves a multistep process, which makes the procedure
inconvenient and expensive. The development of easily
accessible and inexpensive chiral ligands is thus very inter-
esting. We have prepared several aminoalcohols with the
prolinol skeleton and various substituents at different posi-
tions of the pyrrolidine ring and the carbinol carbon atom.
Herein we report our results on their use as catalysts for the
enantioselective addition of dialkylzinc reagents to both
aromatic and aliphatic N-(diphenylphosphinoyl)imines.15
By fine tuning of the ligand structure, an enantioselectivity
of up to 94% has been achieved, which is the highest
In recent years, there has been great interest in the develop-
ment of methodologies for the asymmetric preparation of
optically active amines, since they are important com-
pounds which are extensively used as resolving agents,1
starting materials for the preparation of biologically active
substances2 and chiral auxiliaries in asymmetric synthesis.3
The addition of organometallic reagents to imines is a valu-
able method for the synthesis of primary and secondary
amines.4,5 However, it presents some problems due to the
low electrophilic character of the C@N bond and to the
tendency of enolizable imines to undergo a-deprotonation
instead of addition. The low electrophilicity of the imine
can be overcome by introducing an electron-withdrawing
group attached to the nitrogen atom. Amongst the acti-
vated imines, N-phosphinoylimines6 are very attractive
since the phosphinoyl group can be easily removed from
the addition products under mild reaction conditions lead-
ing to the free amines.7 N-Phosphinoylimines have found a
variety of synthetic applications, including asymmetric pro-
*
Corresponding authors. Tel.: +34 965 903548; fax: +34 965 903549
(M.Y.); e-mail addresses: dguijarro@ua.es; yus@ua.es
0957-4166/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.11.006

R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
2829
reported so far using 0.5 equiv of an aminoalcohol as a
promoter of the same reaction.
O
P
O
P
Ph
Ph
Ph
Ph
i, ii
N
HN
+
R'2Zn
R
H
R
R'
5a: R = Ph
5b: R = 4-ClC6H4
3a: R = Ph
3b: R = 4-ClC6H4
4
2. Results and discussion
3c: R = 4-MeOC6H4
3d: R = 2-naphthyl
3e: R = Me2CHCH2
5c: R = 4-MeOC6H4
5d: R = 2-naphthyl
5e: R = Me2CHCH2
We decided to start by testing N-benzyl-L-prolinol 1c
(Fig. 1) as a promoter of the addition of dialkylzincs to
N-(diphenylphosphinoyl)benzaldimine. The reasons that
led us to choose this ligand were (a) the prolinol substruc-
ture can be found in ligands with the 2-azanorbornylmeth-
anol framework 2 (Fig. 1) that have been successfully
applied as catalysts for the same reaction,10c,e,g (b) ligands
with a prolinol skeleton have given good enantioselectivi-
ties in several asymmetric processes16 and (c) ligand 1 is
commercially available and inexpensive and can be easily
prepared in two steps from the readily accessible natural
aminoacid L-proline.
3f: R =
c-C6H11
5f: R = c-C6H11
Scheme 1. Reagents and conditions: (i) ligand 1 (0.25–1 equiv), toluene, T;
(ii) NH4Cl (aq).
58% (Table 1, entry 4). However, when the addition of
diethylzinc was performed at room temperature, the
expected addition product was obtained after only 4 h in
89% yield and 92% ee, which was slightly higher than the
one reported for the rigid ligand
2
(R4 = PhCH2,
R5 = H; 91% ee).10c Since the reaction was fast under these
conditions, we thought about reducing the amount of
diethylzinc. Setting up the reaction with 3 equiv and main-
taining the addition rate, the same ee of 92% was obtained
in 4 h, although the yield decreased to 60% (Table 1, entry
6). Since we had observed before that the addition rate was
very important for the enantioselectivity, we repeated the
last experiment with addition times of 10 min and 2 h. In
the first case, the reaction was completed in 4 h and the
addition product was isolated in 79% yield and 92% ee
(Table 1, entry 7). Lengthening the addition time to 2 h
caused a decrease in both yield and ee (Table 1, entry 8).
When the reaction under the conditions of entry 7 was
set up at 50 °C (oil bath temperature) instead of room tem-
perature, it was complete in only 1 h, affording the expected
product in 96% yield and 90% ee (Table 1, entry 9). We
were pleased by this result, since, to the best of our knowl-
edge, this is the fastest addition reaction of diethylzinc to
imine 3a that has ever been reported using an aminoalcohol
as a promoter.
R3
R2
N
N
R1
H
OH
OH
Ph
1a: R1 = Me
1b: R1 = Pri
1h: R2 = R3 = Me
1i: R2 = R3 = Ph
1c: R1 = PhCH2
1j: R2 = Me, R3 = H
1k: R2 = H, R3 = Me
1l: R2 = Ph, R3 = H
1m: R2 = H, R3 = Ph
1d: R1 = (1-naphthyl)CH2
1e: R1 = (2-naphthyl)CH2
1f: R1 = (
R)-1-phenylethyl
1g: R1 = (
S
)-1-phenylethyl
R5
N
R5
OH
N
N
OH
OH
R4
Ph
Ph
1n
1o
2
Figure 1.
In the first experiment, diethylzinc (3 equiv) was added
dropwise to a solution of benzaldimine 3a (Scheme 1)
and ligand 1c (1 equiv) in toluene at 0 °C over ca. 10 min.
The reaction was stirred allowing the temperature to rise
to room temperature and, after 24 h, a disappointing 74%
ee was obtained (Table 1, entry 1), which was lower than
the one reported for the analogous ligand with 2-azanor-
bornylmethanol skeleton 2 (R4 = PhCH2, R5 = H; 91%
ee).10c,17 However, when diethylzinc was slowly added over
3 h using a syringe pump, the ee increased to 86% (Table 1,
entry 2). Gratifyingly surprised by this result, we decided to
screen the reaction conditions further. Since yields were
only moderate in both cases, we repeated the reaction using
9 equiv of diethylzinc instead of 3 equiv and the yield
improved to 84% with almost the same enantioselectivity
as before, with the time needed for completion of the reac-
tion being much shorter (8 h; Table 1, entry 3).
In all cases, although a stoichiometric amount of ligand 1c
was used, up to 85% of it could be recovered by column
chromatography and reused without loss of chiral induc-
tion. For instance, ligand 1c, which was isolated from the
reaction of entry 7 in Table 1, was used as a promoter of
another addition reaction of diethylzinc to imine 3a and
the expected addition product 5aa was obtained in 78%
yield and 92% ee.
The possibility of using a substoichiometric amount of the
ligand was then investigated. We were delighted to see that
we obtained the same result using either 0.5 or 1 equiv of
ligand 1c (compare entries 7 and 10 in Table 1). The ee
of 92% is almost equal to the highest ee reported so far
using this amount of an aminoalcohol (93%)10j and our
reaction was much faster (4 h instead of 48 h). With
0.25 equiv of the ligand, the reaction time increased to
20 h and the ee decreased to 80%. However, when the reac-
tion temperature was 50 °C instead of 25 °C, the same
results were achieved and the reaction was finished after
only 3 h (compare entries 11 and 12 in Table 1). It is worth
Next, we studied the effect of temperature. When the addi-
tion of diethylzinc (9 equiv in 3 h) to imine 3a in the pres-
ence of ligand 1c (1 equiv) was stirred for 24 h at 0 °C, the
reaction did not go to completion and the ee decreased to

2830
R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
Table 1. Enantioselective addition of diethylzinc to N-(diphenylphosphinoyl)imine 3a in the presence of aminoalcohol 1c: Optimization of the reaction
conditions
Entry
Equiv of Et2Zn
Equiv of 1c
T (°C)
Add. timea (h)
Time (h)
Product 5aa
Yieldb (%)
eec (%)
Configurationd
1
2
3
4
5
6
7
8
9
3
3
9
9
9
3
3
3
3
3
3
3
1
1
1
1
1
1
1
1
1
0–25
0–25
0–25
0
25
25
25
25
50
25
0.2
3
3
3
3
1
0.2
2
0.2
0.2
0.2
0.2
24
24
8
24
4
4
4
4
1
52
48
84
45
89
60
79
48
96
79
76
75
74
86
84
58
92
92
92
76
90
92
80
80
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
10
11
12
0.5
0.25
0.25
4
20
3
25
50
a Period of time during which the dropwise addition of diethylzinc to the solution of 1c and 3a was performed.
b Isolated yield after column chromatography (silica gel, pentane/acetone) based on the starting imine 3a. All isolated compounds were P95% pure (GC
and/or 300 MHz 1H NMR).
c Enantiomeric excess determined by HPLC using a ChiralCel OD-H column.
d The absolute configuration of the major enantiomer was determined by comparison of the specific rotation of the free primary amine with the one
reported in the literature.
noting that, to the best of our knowledge, this is the fastest
reaction of this type reported so far using 0.25 equiv of an
aminoalcohol as the catalyst with the ee obtained close to
the highest ee reported in the literature for that amount
of ligand (85%).10c
50 °C led to an improved yield (74%) in only 1 day with
a very slight decrease in the ee (Table 2, entry 3). Diisoprop-
ylzinc was as efficient as diethylzinc giving good yield and
enantioselectivity in a reaction time of only 4 h (Table 2,
entry 4). The reaction with dibutylzinc also required a
longer reaction time to reach completion affording the
addition product in 64% yield and 90% ee (Table 2, entry
5). The same reaction at 50 °C improved both the reaction
rate and yield with almost the same enantioselectivity
(Table 2, entry 6).
The versatility of our procedure concerning the dialkylzinc
reagent was also studied. Dimethyl-, diisopropyl- and di-
butylzinc were used as nucleophiles for the addition to
imine 3a under the reaction conditions of entry 10 in Table
1. As previously reported,7,10c dimethylzinc turned out to
be much less reactive than diethylzinc. The use of 9 equiv
was necessary to obtain a moderate yield of the addition
product in 3 days at room temperature, but a 90% ee was
obtained (Table 2, entry 2). Setting up the reaction at
Next, we decided to investigate the scope of this reaction by
testing some other imine substrates 3b–f (Scheme 1). All
imines 3 were prepared from the corresponding aldehydes
according to the literature procedures.12g,18 Ligand 1c
Table 2. Enantioselective addition of dialkylzinc reagents to N-(diphenylphosphinoyl)imines 3 in the presence of N-benzyl-L-prolinol 1ca: Preparation of
compounds 5
Entry
Imine
R0
Equiv
T (°C)
Time (h)
Product
eec (%)
No.
Yieldb (%)
Configurationd
1
2
3
4
5
6
7
8
9
3a
3a
3a
3a
3a
3a
3b
3c
3d
3e
3f
Et
3
9
9
3
3
3
3
3
3
3
3
25
25
50
25
25
50
25
25
25
25
25
4
72
24
4
24
4
4
12
4
5aa
5ab
5ab
5ac
5ad
5ad
5b
79
51
74
80
64
86
80
71
86
60
36
92
90
88
90
90
88
90
92
90
58
90
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
Me
Me
Pri
Bun
Bun
Et
Et
Et
Et
Et
5c
5d
10
11
4
4
5e
5f
a All reactions were performed by the dropwise addition of the dialkylzinc reagent over ca. 10 min to a stirred solution of imine 3 (0.5 mmol) and ligand 1c
(0.25 mmol) in anhydrous toluene (3 mL) under argon and stirring was continued for the time indicated.
b Isolated yield after column chromatography (silica gel, pentane/acetone) based on the starting imine 3. All isolated compounds 5 were P95% pure (GC
and/or 300 MHz 1H NMR).
c Enantiomeric excess determined by HPLC using a ChiralCel OD-H column or a Chiralpak AD column.
d The absolute configuration of the major enantiomer was determined by comparison of the specific rotation of the free primary amine with that reported
in the literature.

R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
2831
(0.5 equiv) was used as a promoter of the addition of dieth-
i
ii or iii
ylzinc to imines 3b–f under the reaction conditions of entry
10 in Table 1 and the results are listed in Table 2. As can be
seen, all imines derived from aromatic aldehydes gave very
good ees regardless of the electronic nature of the substitu-
ents on the aromatic ring (Table 2, entries 7–9). Although
the reaction with imine 3c, which has an electron-donating
group at the para position of the benzene ring, was slower
(reaction time of 12 h instead of 4 h), a very high ee of 92%
was observed in the addition product 5c (Table 2, entry 8).
Our ligand 1c could also effectively catalyze the addition of
diethylzinc to aliphatic imines 3e and 3f. The ee obtained
with imine 3e, derived from isobutyraldehyde, was only
moderate (58%, Table 2, entry 10), but we were pleased
to see that the ee of the addition product to imine 3f,
derived from cyclohexanecarbaldehyde, was of the same
level as those obtained with the aromatic imines (90%,
Table 2, entry 11).
CO2R
CO2H
CO2R
N
N
H
N
R1
H2
Cl
8
9a: R = Me
9b
10b,d-g
: R = Et
b: R = Et, R1 = Pri
iv
1
N
R1
d
: R = Et, R = (1-naphthyl)CH2
OH
e: R = Et, R1 = (2-naphthyl)CH2
f: R = Me, R1 = (R)-1-phenylethyl
1b,d-g
1
g
: R = Me, R = (S)-1-phenylethyl
Scheme 2. Reagents and conditions: (i) 2 M HCl in EtOH, reflux (for
compound 9b); (ii) PriBr, K2CO3, 18-crown-6 (cat.), NaI, MeCN, reflux,
55% (two steps) (for compound 10b); (iii) (1-naphthyl)CH2Cl (for
compound 10d) or (2-naphthyl)CH2Br (for compound 10e) or
PhCH(Me)Br (for compounds 10f and 10g), Et3N, CHCl3, reflux, 64%
(for 10d, two steps), 74% (for 10e, two steps) and 70% (for 10f + 10g, 1H
NMR ratio 10f:10g = 1:1); (iv) LiAlH4, Et2O, 0 °C, 68%, 49%, 33%, 82%
and 85%, respectively.
We also tried to use the activated ketimine 6 (Fig. 2) as the
substrate. The reaction of this imine with diethylzinc
(3 equiv) in the presence of aminoalcohol 1c (0.5 equiv) at
room temperature gave the reduction product 7a (Fig. 2)
exclusively. We assumed that the reduction process took
place via a b-hydride transfer from diethylzinc to the imine
carbon atom. Thus, we decided to test dimethylzinc, which
lacks hydrogen atoms at the b position. The use of this
nucleophile (9 equiv) led to the formation of the expected
addition product 7b (Fig. 2) after stirring at room temper-
ature for 24 h in 40% yield, but with a disappointing 18%
ee.
alkyl halide in the presence of a base, giving esters 10b,
10d and 10e in 55%, 64% and 74% yields, respectively.
The final reduction with LiAlH4 afforded the expected N-
alkylated prolinols 1b, 1d and 1e. On the other hand, com-
mercially available L-proline methyl ester hydrochloride 9a
was alkylated at nitrogen by reaction with 1-bromo-1-
phenylethane in the presence of triethylamine, giving a
mixture of the diastereomeric esters 10f and 10g in an over-
1
all yield of 70% (1:1 ratio by H NMR). Aminoesters 10f
and 10g could be separated by flash chromatography and
reduction with LiAlH4 afforded aminoalcohols 1f and 1g
in 82% and 85% yield, respectively. To determine the abso-
lute stereochemistry of the sec-phenethyl substituent of
compound 10f, its ester moiety was transformed into a
diphenylcarbinol centre by reaction with an excess of
PhMgBr. X-ray diffraction analysis of the obtained amino-
alcohol showed an (R) stereochemistry for the substituent
at nitrogen (Fig. 3).19
O
P
Ph
CO2Et
O
P
Ph
CO2Et
Ph
Ph
N
HN
R
6
7a: R = H
7b: R = Me
Compounds 1h and 1i, which have a tertiary carbinol cen-
tre, were synthesized by the reaction of the commercially
available N-benzyl-L-proline ethyl ester 11 with the corres-
ponding Grignard reagents (Scheme 3). The two epimers 1j
and 1k were prepared from N-benzyl-L-prolinol 1c, which
was converted into aldehyde 12 by a Swern oxidation
(Scheme 4). The addition of MeMgBr to 12 in the presence
of CeCl3 gave a mixture of products 1j and 1k, which could
be separated by flash chromatography. Following the same
procedure, but using PhMgBr instead of MeMgBr, the two
epimers 1l and 1m were obtained and could be separated by
flash chromatography. The absolute stereochemistry of li-
gand 1j was determined by comparison of its physical
and spectroscopic data with that reported in the litera-
ture.20 The absolute stereochemistry of the carbinol site
of ligand 1m could be determined by X-ray diffraction anal-
ysis (Fig. 4).21
Figure 2.
As described above, by proper choice of the reaction con-
ditions, N-benzyl-L-prolinol was shown to be a catalyst
for the addition of dialkylzinc reagents to N-(diphenyl-
phosphinoyl)aldimines and even more effective than the
analogous bicyclic ligand 2. It is worth noting that the
enantiomer of ligand 1c is also commercially available,
which provides the opportunity of preparing both enantio-
mers of the final amine products.
Having established that compound 1c with the prolinol
skeleton was an efficient catalyst for this reaction, we
thought about studying the effect of different ligand substit-
uents on the enantioselectivity. Figure 1 shows all the
substituted prolinols 1 which have been tested as catalysts
for the addition of diethylzinc to imine 3a. N-Methyl pro-
linol 1a was also commercially available. Aminoalcohols
1b, 1d and 1e were prepared from L-proline as described
in Scheme 2. L-Proline ethyl ester hydrochloride 9b was
alkylated at nitrogen by reaction with the corresponding
Commercially available aminoacid 13 was used as the start-
ing material for the synthesis of ligand 1n (Scheme 5). The
transformation of 13 into its ethyl ester hydrochloride
followed by benzylation of the nitrogen atom afforded

2832
R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
i, ii
iii
CO2Et
N
N
CO2H
N
OH
H
Ph
Ph
13
14
1n
Scheme 5. Reagents and conditions: (i) 2 M HCl in EtOH, reflux; (ii)
PhCH2Br, Et3N, CHCl3, reflux, 80% (two steps); (iii) LiAlH4, Et2O, 0 °C,
76%.
i
ii
CO2Me
N
N
CO2Me
N
H
OH
Ph
Ph
15
16
1o
Scheme 6. Reagents and conditions: (i) PhCH2Br, Et3N, CHCl3, reflux,
30%; (ii) LiAlH4, Et2O, 0 °C, 92%.
With all aminoalcohols 1 in hand, we decided to test them
as catalysts for the addition of diethylzinc to benzaldimine
3a using 0.5 equiv of ligand 1 under the reaction conditions
of entry 10 in Table 1. The results of these reactions are
collected in Table 3. It seems that the enantioselectivity
improves with the size of the nitrogen substituent. The ee
increased upon going from methyl (Table 3, entry 1) to iso-
propyl (Table 3, entry 2). The presence of an aromatic ring
in R1 seems to be important, since ligand 1c, which has an
N-benzyl group, gave 92% ee (Table 3, entry 3). Substitu-
tion of the phenyl group by a 2-naphthyl group afforded
Figure 3. X-ray structure of the diphenylcarbinol derived from 10f.
R2
R2
i
CO2Et
N
N
OH
Ph
Ph
Table 3. Enantioselective addition of diethylzinc to N-(diphenylphosphi-
noyl)benzaldimine 3a in the presence of aminoalcohols 1a
1h: R2 = Me
11
2
1i
: R = Ph
Entry Ligand Time (h)
Product 5aa
Yieldb (%) eec (%) Configurationd
Scheme 3. Reagents and conditions: (i) MeMgBr (for compound 1h) or
PhMgBr (for compound 1i), THF, ꢀ78 °C to rt, 99% and 50%,
respectively.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1a
1b
1c
1d
1e
1f
18
4
4
4
4
43
80
79
70
85
66
80
45
30
73
35
76
54
86
62
62
84
92
80
90
60
94
60
20
94
66
88
12
90
60
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
O
R
R
i
ii
N
N
N
N
+
6
4
H
H
OH
H
OH
OH
1g
1h
1i
Ph
Ph
Ph
Ph
24
48
4
24
7
24
4
6
1c
12
1j: R = Me
1l: R = Ph
1k: R = Me
1m: R = Ph
1j
1k
1l
Scheme 4. Reagents and conditions: (i) ClCOCOCl, DMSO, Et3N,
CH2Cl2, ꢀ78 °C; (ii) MeMgBr (for compounds 1j and 1k) or PhMgBr
(for compounds 1l and 1m), CeCl3, Et2O, ꢀ78 °C, 81% (for 1j + 1k, two
steps, 1H NMR ratio 1j:1k = 1:3) and 77% (for 1l + 1m, two steps, 1H
NMR ratio 1l:1m = 3:7).
1m
1n
1o
a All reactions were performed by dropwise addition of Et2Zn (3 equiv)
over ca. 10 min to a stirred solution of imine 3a (0.5 mmol) and ligand 1
(0.25 mmol) in anhydrous toluene (3 mL) under argon at room tem-
perature and stirring was continued for the time indicated.
b Isolated yield after column chromatography (silica gel, pentane/acetone)
based on the starting imine 3a. All isolated compounds were P95% pure
(GC and/or 300 MHz 1H NMR).
c Enantiomeric excess determined by HPLC using a ChiralCel OD-H
column.
d The absolute configuration of the major enantiomer was determined by
comparison of the specific rotation of the free primary amine with that
reported in the literature.
aminoester 14 in 80% yield. Treatment of 14 with LiAlH4
led to aminoalcohol 1n in 76% yield.
Finally, ligand 1o was prepared from the cis-5-methyl pro-
line methyl ester 15 (Scheme 6), which was synthesized in a
sequence of several steps from methyl N-Boc-(S)-pyroglu-
tamate according to the literature.22 N-Benzylation of 15
followed by reduction with LiAlH4 afforded the expected
ligand 1o.

R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
2833
almost the same enantioselectivity (Table 3, entry 5),
whereas a 1-naphthylmethyl substituent at the nitrogen
atom gave a lower ee (80%; Table 3, entry 4). The introduc-
tion of a methyl group on the a-carbon atom of the
benzylic substituent at nitrogen also affected the enantio-
selectivity. The stereochemistry of this new asymmetric
carbon atom has proven to be very important. The (S)-1-
phenylethyl substituent of ligand 1g led to an increase of
the ee to 94% (Table 3, entry 7), whereas the (R)-1-phenyl-
ethyl substituent gave an ee of 60% (Table 3, entry 6), lower
than the one obtained with the unsubstituted benzyl group.
The ee of 94% obtained in the reaction catalyzed by 1g is,
to the best of our knowledge, the highest ee reported so far
using 0.5 equiv of an aminoalcohol as a promoter of this
reaction.
ring, were tested as catalysts. No additional benefits were
achieved by the introduction of the methyl group at C2,
since the ee (90%, Table 3, entry 14) was slightly lower than
the one obtained with N-benzyl-L-prolinol 1c. The intro-
duction of the methyl group at C5 was detrimental to the
enantioselectivity, giving only 60% ee (Table 3, entry 15).
The results shown in Table 3 could be rationalized assum-
ing that the addition reaction takes place through a transi-
tion state like the one depicted in Figure 5 for the addition
of dimethylzinc to imine 3a promoted by ligand 1c.23 The
configuration of the stereocentre on the pyrrolidine ring
and the coordination between ZnA and the nitrogen atom
of the ligand would force the benzyl group to be in the
position indicated in Figure 5. Another molecule of di-
methylzinc (Me2ZnB) would be coordinated to the oxygen
atom of the ligand and the oxygen atom of the imine would
coordinate to ZnA. The transfer of one methyl group from
Me2ZnB to the carbon of the imine would render the (R)-
enantiomer of the addition product, which is the major
one in all the cases.
As was previously observed,7,10c a decrease in the enantio-
selectivity was obtained upon increasing the size of the R2
substituent (compare entries 3, 8 and 9 in Table 3). More-
over, the reactions promoted by the ligands with a tertiary
alcohol, 1h and 1i, were much slower (reaction times of 24
and 48 h, respectively). Since it had been reported that the
introduction of a stereogenic centre at the carbinol site
could improve the enantioselectivity of these reactions,10e
we tested aminoalcohols 1j and 1k. A comparison of entries
10 and 11 in Table 3 shows the importance of the stereo-
chemistry at this carbinol centre: ligand 1j, which has an
(R)-configuration, gave product 5aa in good yield and with
94% ee in a fast reaction, whereas the ligand with a (S)-car-
binol carbon 1k needed 24 h to give a low yield of the addi-
tion product in only 66% ee. The same trend was observed
for the pair of epimeric ligands 1l and 1m (Fig. 4), having a
phenyl group at the carbinol site, concerning both reaction
rate and enantioselectivity, the ees being lower than the
ones obtained with the ligands 1j and 1k bearing a methyl
group at the carbinol carbon atom.
Figure 5.
For the ligands with a primary hydroxymethyl moiety, 1a–
g, 1n and 1o, the presence of an aromatic ring on the R1
substituent seems to be beneficial for the enantioselectivity
(Table 3, entries 3, 5, 7 and 14). p–p interactions between
the aromatic ring of the R1 substituent and one of the
phenyl groups on the phosphorus atom could stabilize
the depicted transition state to some extent, which could
also explain the higher enantioselectivities observed. The
ee decreased when H1 of the ligand (Fig. 5) was changed
by a methyl group (compare entries 3 and 6 in Table 3).
The steric hindrance introduced by this methyl group could
make coordination to the imine more difficult, leading to a
lower enantioselectivity. The change of H2 (Fig. 5) to a
methyl group (ligand 1g) afforded the highest enantioselec-
tivity obtained with this kind of ligands (Table 3, entry 7).
The introduction of a methyl group at C2 of the pyrroli-
dine ring (ligand 1n) did not cause any detrimental effect
on the enantioselectivity (compare entries 3 and 14 in Table
3). However, a cis methyl group at C5 decreased the ee.
This could be due to the steric hindrance between this
methyl group and the methyl group on ZnA, which would
Figure 4. X-ray structure of ligand 1m.
To determine if further steric bulkiness close to the nitrogen
atom would improve the enantioselectivity, aminoalcohols
1n, having a methyl group at C2 of the pyrrolidine ring,
and 1o, bearing a methyl group at C5 of the pyrrolidine

2834
R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
disfavour the formation of the well-ordered transition
state.
the sample. Commercially available compounds 1a {Al-
19
drich, 96%, ½aꢁD ¼ ꢀ49:5 (c 5.0, MeOH)}, 1c {Aldrich,
20
20
99%, ½aꢁD ¼ ꢀ72:2 ðneatÞ}, 11 {Aldrich, 97%, ½aꢁD
¼
The change of H3 (Fig. 5) by a methyl group slightly
improves the enantioselectivity (compare entries 3 and 10
in Table 3). A phenyl group instead of H3 gave only a slight
decrease in the ee. Thus, it seems that the effect caused by a
substituent at this position (R2 in Fig. 1) on the enantio-
selectivity is small, probably because this substituent would
point away from the imine. However, changing H4 (Fig. 5)
with a substituent (ligands 1k and 1m) led to a decrease of
the ee. Bigger the substituent at this position, lower the ee
(compare entries 3, 11 and 13 in Table 3). The repulsive
interaction between this substituent (R3 in Fig. 1) and both
the aromatic ring of the phosphinoyl group of the imine
and one methyl group of Me2ZnB would disfavour the for-
mation of the depicted transition state. When the ligand
has a tertiary carbinol site (1h and 1i), the substituent
located at R3 seems to be responsible for the decrease
observed in the enantioselectivity, since the ee values are
very similar to the ones obtained with ligands 1k and 1m
(compare entry 8 with 11 and 9 with 13 in Table 3).
ꢀ62:0 ðneatÞ} and 13 (Aldrich, P98%) were used as re-
ceived. Optical rotations were measured on a Perkin–Elmer
341 polarimeter. HPLC analyses were performed at 25 °C
on a JASCO apparatus, equipped with a PU-2089 Plus
pump, a MD-2010 Plus detector and an AS-2059 Plus
automatic injector. The HRMS (EI) were performed by
the Technical Services of the University of Alicante on a
Finnigan MAT 95S apparatus. The X-ray diffraction anal-
yses were carried out by the Technical Services of the Uni-
versity of Alicante on a Bruker CCD-Apex instrument
equipped with an X-ray tube with a Mo anode and a
KRYOFLEX low temperature equipment.
4.2. Synthesis of L-proline ethyl ester hydrochloride 9b
SOCl2 (3.5 mL, 52 mmol) was added dropwise to a solution
of L-proline (5.0 g, 43 mmol) in EtOH (56 mL) at 0 °C.
When the addition was complete, the mixture was refluxed
for 6 h. After cooling to room temperature, the solvent was
evaporated under reduced pressure. The resulting residue
was redissolved in Et2O and the solvents were evaporated
again. This process was repeated twice to remove EtOH
completely. The resultant L-proline ethyl ester hydrochlo-
ride 9b was obtained in 98% yield and was used in the
alkylation step without further purification.
3. Conclusions
In conclusion, we have reported that several b-aminoalco-
hols with the prolinol skeleton are very efficient catalysts
for the addition of dialkylzinc reagents to N-(diphenyl-
phosphinoyl)imines. Fast enantioselective addition reac-
tions can be achieved using 0.5 equiv of the ligand. Our
results show the importance of carefully screening the reac-
tion conditions to obtain high levels of enantioselectivity.
A proper choice of the addition rate and the temperature
turned out to be crucial for obtaining high ee values. More-
over, it is possible to improve both the reaction rates and
yields by setting up the experiments at 50 °C with only a
very slight decrease in the enantioselectivity. The selectivity
can be improved upon by introduction of a stereogenic cen-
tre at either the carbinol carbon atom or at the substituent
on the nitrogen atom of the ligand. The absolute configura-
tion of this stereogenic centre determines if the enantio-
selectivity increases or decreases. Another important
conclusion that can be drawn from our results is that a
rigid structure in the ligand is not always a requisite to
obtain high enantioselectivities in this reaction.
4.2.1. L-Proline ethyl ester hydrochloride 9b.24 Colourless
20
oil; Rf 0.32 (ethyl acetate; deactivated silica gel); ½aꢁD
¼
ꢀ30:0 (c 3.3, EtOH); m (film) 3404 (NH), 1757 (C@O),
1231 cmꢀ1 (CO); dH 1.33 (3H, t, J = 7.2 Hz, Me), 2.00–
2.28, 2.39–2.51 (3H and 1H, respectively, 2m,
CH2CH2CH), 3.47–3.70 (2H, m, CH2N), 4.29 (2H, q,
J = 7.2 Hz, CH2O), 4.46–4.58 (1H, m, CH); dC 13.8
(Me), 23.4, 28.5 (CH2CH2CH), 45.9 (CH2N), 59.1
(CH2O), 62.6 (CH), 168.6 (CO); m/z 144 (M+ꢀ35.5,
<1%), 143 (2), 70 (100).
4.3. Preparation of N-isopropyl L-proline ethyl ester 10b
Ester 9b (3.0 g, 21 mmol), PriBr (20 mL, 210 mmol), NaI
(6.3 g, 42 mmol) and 18-crown-6 (576 mg, 2.4 mmol) were
dissolved in anhydrous MeCN (117 mL). K2CO3 (4.4 g,
32 mmol) was added and the reaction mixture was refluxed
for 3 days. The solvent was evaporated, after which water
(20 mL) was added and the resulting mixture was extracted
with CH2Cl2 (3 ꢂ 50 mL). The combined organic layers
were dried over Na2SO4. After filtration and evaporation
of the solvents, the crude residue was purified by flash chro-
matography (silica gel, pentane/Et2O), giving product 10b
in 55% yield.
4. Experimental
4.1. General
For general experimental information, see Ref. 14. Imines 3
were prepared according to the literature procedures.12g,18
When mentioned, purification by flash chromatography
on deactivated silica gel means that, before adding the reac-
tion crude to the column, the latter was eluted with a mix-
ture of 5% triethylamine in hexane until the eluent coming
from the column was basic according to the pH paper.
When mentioned, an Rf value measured on deactivated sil-
ica gel means that the TLC plate was eluted with a mixture
of 5% triethylamine in hexane and dried before applying
4.3.1. (S)-Ethyl 1-isopropylpyrrolidine-2-carboxylate 10b.
20
Yellow oil; Rf 0.34 (ethyl acetate); ½aꢁD ¼ ꢀ54:9 (c 1.2,
CHCl3); m (film) 1731 (C@O), 1162 cmꢀ1 (CO); dH 1.08
(6H, d, J = 6.4 Hz, Me2CH), 1.27 (3H, t, J = 7.1 Hz,
MeCH2), 1.75–2.18 (4H, m, CH2CH2CH), 2.57–2.64,
2.76–2.84 (1H each, 2m, CH2N), 3.06–3.16 (1H, m,
CHMe), 3.39–3.43 (1H, m, CHCO), 4.18 (2H, q,
J = 7.1 Hz, CH2O); dC 14.1 (MeCH2), 20.0, 21.4 (Me2CH),

R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
2835
23.4, 30.0 (CH2CH2CH), 50.0 (CHMe), 52.1 (CH2N), 60.2
(CHCO), 62.7 (CH2O), 175.2 (CO); m/z 185 (M+, 2%), 112
(100), 70 (42); HRMS: M+ found 185.1428, C10H19NO2
requires 185.1416.
4.4.3. (2S)-Methyl 1-[(R)-1-phenylethyl]pyrrolidine-2-carb-
oxylate 10f. Colourless oil; Rf 0.52 (hexane/diethyl ether:
20
1:1); ½aꢁD ¼ ꢀ60:0 (c 1.0, CHCl3); m (film) 3085, 3058, 3026,
1629, 1493 (HC@C), 1733 (C@O), 1160 cmꢀ1 (CO); dH
1.40 (3H, d, J = 6.8 Hz, MeCH), 1.75–2.13 (4H, m,
CH2CH2CH), 2.47–2.53, 3.12–3.17 (1H each, 2m, CH2N),
3.43–3.47 (1H, m, CHCO), 3.45 (3H, s, MeO), 3.63 (1H,
q, J = 6.6 Hz, CHPh), 7.19–7.34 (5H, m, ArH); dC 21.6
(MeCH), 23.5, 30.2 (CH2CH2CH), 51.3 (CH2N), 51.9
(MeO), 63.3, 63.8 (2 ꢂ CHN), 127.1, 127.8 (2C), 128.0
(2C), 144.3 (ArC), 175.4 (CO); m/z 233 (M+, 2%), 218
(10), 175 (13), 174 (100), 105 (88), 103 (11), 79 (11), 77
(13), 70 (74); HRMS: M+ found 233.1407, C14H19NO2
requires 233.1416.
4.4. Synthesis of N-substituted L-proline esters 10d–g.
General procedure
A solution of the L-proline ester hydrochloride 9a (for 10f
and 10g) or 9b (for 10d and 10e) (7 mmol), Et3N (2.3 mL,
16 mmol) and the corresponding alkylating agent [7 mmol,
(1-naphthyl)CH2Cl for 10d, (2-naphthyl)CH2Br for 10e or
PhCH(Me)Br for 10f and 10g] in CHCl3 (27 mL) was
refluxed for 6 h. After cooling to room temperature, an
aqueous 1 M NaOH solution (27 mL) was added and the
mixture was extracted with CH2Cl2 (3 ꢂ 10 mL). The com-
bined organic layers were dried over MgSO4. After filtra-
tion and evaporation of the solvents, the crude residue
was purified by flash chromatography (silica gel, hexane/
acetone), giving products 10d and 10e in 64% and 74%
yield (from L-proline), respectively. In the reaction between
9a and PhCH(Me)Br, a 1:1 mixture of epimers 10f and 10g
was obtained, which could be separated by flash chroma-
tography (silica gel, hexane/acetone) and were isolated in
pure form in 13% and 20% yield, respectively. The corres-
ponding physical, spectroscopic and analytical data for
compounds 10d–g follow.
4.4.4. (2S)-Methyl 1-[(S)-1-phenylethyl]pyrrolidine-2-car-
boxylate 10g. Colourless oil; Rf 0.57 (hexane/diethyl
20
ether: 1:1); ½aꢁD ¼ ꢀ95:4 (c 1.1, CHCl3); m (film) 3078,
3061, 3026, 1630, 1492 (HC@C), 1733 (C@O), 1158 cmꢀ1
(CO); dH 1.39 (3H, d, J = 6.6 Hz, MeCH), 1.71–1.79,
1.82–1.94, 2.01–2.12 (1H, 2H and 1H, respectively, 3m,
CH2CH2CH), 2.55–2.61, 2.96–3.01 (1H each, 2m, CH2N),
3.30 (1H, dd, J = 9.3 and 4.2 Hz, CHCO), 3.68 (3H, s,
MeO), 3.74 (1H, q, J = 6.8 Hz, CHPh), 7.22–7.34 (5H,
m, ArH); dC 22.2 (MeCH), 23.1, 29.9 (CH2CH2CH), 50.5
(CH2N), 51.5 (MeO), 61.7, 62.8 (2 ꢂ CHN), 127.0, 127.5
(2C), 128.1 (2C), 143.5 (ArC), 175.6 (CO); m/z 233 (M+,
3%), 175 (14), 174 (100), 105 (86), 103 (12), 79 (11), 77
(14), 70 (64); HRMS: M+ found 233.1430, C14H19NO2
requires 233.1416.
4.4.1. (S)-Ethyl 1-(1-naphthylmethyl)pyrrolidine-2-carboxyl-
ate 10d. Yellow oil; Rf 0.70 (hexane/diethyl ether: 2:1);
20
½aꢁD ¼ ꢀ26:9 (c 1.1, CHCl3); m (film) 3072, 3065, 3044,
1597, 1510 (HC@C), 1740 (C@O), 1179 cmꢀ1 (CO); dH
1.22 (3H, t, J = 7.1 Hz, Me), 1.68–1.88, 1.91–2.03, 2.09–
2.21 (2H, 1H and 1H, respectively, 3m, CH2CH2CH),
2.40 (1H, q, J = 8.3 Hz, 1 ꢂ CH2CHHN), 2.88–2.95 (1H,
m, 1 ꢂ CH2CHHN), 2.28 (1H, dd, J = 6.4 and 2.2 Hz,
CHCO), 3.82, 4.45 (1H each, 2d, J = 12.5 Hz each,
CH2Ar), 4.10 (2H, q, J = 7.1 Hz, CH2O), 7.27–7.58,
7.74–7.84 (5H and 2H, respectively, 2m, ArH); dC 14.2
(Me), 22.9, 29.4 (CH2CH2CH), 53.4, 57.2 (2 ꢂ CH2N),
60.4 (CH2O), 66.1 (CHCO), 124.9, 125.0, 125.6 (2C),
125.9, 127.2, 128.0, 128.2, 132.5, 133.7 (ArC), 174.2 (CO);
m/z 283 (M+, 1%), 211 (12), 210 (70), 142 (13), 141 (100),
115 (17); HRMS: M+ found 283.1570, C18H21NO2 requires
283.1572.
4.4.5. Determination of the absolute configuration of com-
pound 10f. PhMgBr (1.89 mmol, 0.63 mL of a 3 M solu-
tion in Et2O) was added dropwise to a stirred solution of
compound 10f (97 mg, 0.42 mmol) in anhydrous THF
(2 mL) under Ar at room temperature. After stirring over-
night at the same temperature, water (10 mL) was added
and the resulting mixture was extracted with ethyl acetate
(3 ꢂ 10 mL). The combined organic layers were dried over
Na2SO4. After filtration and evaporation of the solvents,
the crude residue was purified by flash chromatography
(silica gel, hexane/ethyl acetate), giving the corresponding
tertiary alcohol in 91% yield. The white solid obtained
was recrystallized in hexane and the crystals obtained were
studied by X-ray diffraction analysis, giving the (R)-config-
uration for the sec-phenethyl substituent at the nitrogen
atom (Fig. 3).19
4.4.2. (S)-Ethyl 1-(2-naphthylmethyl)pyrrolidine-2-carboxyl-
ate 10e. Yellow oil; Rf 0.65 (hexane/diethyl ether: 2:1);
20
½aꢁD ¼ ꢀ55:4 (c 1.0, CHCl3); m (film) 3082, 3053, 1606,
4.4.6. {(2S)-1-[(R)-1-Phenylethyl]pyrrolidin-2-yl}diphenyl-
methanol. White solid; Rf 0.26 (hexane/ethyl acetate:
1508 (HC@C), 1743 (C@O), 1178 cmꢀ1 (CO); dH 1.18
(3H, t, J = 7.1 Hz, Me), 1.69–2.19 (4H, m, CH2CH2CH),
2.40 (1H, q, J = 8.5 Hz, 1 ꢂ CH2CHHN), 2.98–3.11 (1H,
m, 1 ꢂ CH2CHHN), 3.27 (1H, dd, J = 6.2 and 2.5 Hz,
CHN), 3.68 (1H, d, J = 12.8 Hz, 1 ꢂ CHHAr), 3.99–4.15
(3H, m, CH2O and 1 ꢂ CHHAr), 7.39–7.53, 7.73–7.81
(3H and 4H, respectively, 2m, ArH); dC 14.1 (Me), 22.9,
29.2 (CH2CH2CH), 53.2, 58.8 (2 ꢂ CH2N), 60.3 (CH2O),
65.3 (CHCO), 125.4, 125.7, 127.3, 127.5 (2C), 127.6,
127.7, 132.6, 133.2, 136.1 (ArC), 174.0 (CO); m/z 283
(M+, 1%), 211 (13), 210 (77), 142 (13), 141 (100), 115
(17); HRMS: M+ found 283.1573, C18H21NO2 requires
283.1572.
20
9:1); mp 111 °C (hexane); ½aꢁD ¼ þ54:0 (c 0.5, CHCl3); m
(KBr) 3457 (OH), 3098, 3065, 3031, 1602, 1495 (HC@C),
1088 cmꢀ1 (CO); dH 1.23 (3H, d, J = 6.8 Hz, Me), 1.48–
1.72, 1.81–1.90 (3H and 1H, respectively, 2m,
CH2CH2CH), 2.50–2.55, 2.71–2.77 (1H each, 2m,
CH2CH2N), 3.35 (1H, q, J = 6.8 Hz, CHMe), 4.30 (1H,
dd, J = 5.3 and 3.5 Hz, CH2CHN), 5.13 (1H, s, OH),
7.08–7.30 (11H, m, 11 ꢂ ArH), 7.61, 7.78 (2H each, 2d,
J = 8.2 Hz each, 4 ꢂ ArH); dC 9.4 (Me), 24.8, 30.3, 46.7
(CH2CH2CH2N), 56.2 (CHMe), 67.0 (CHN), 77.4 (CO),
125.4 (4C), 126.1, 126.4, 126.5, 127.4 (2C), 127.9 (2C),
128.0 (2C), 128.2 (2C), 143.8, 146.8, 148.6 (ArC); m/z 339

2836
R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
(M+ꢀ18, 5%), 182 (23), 174 (68), 105 (100), 77 (28), 70 (42).
HRMS: M+ꢀ18 found 339.1945, C25H25N requires
339.1987.
4.5.4. {(2S)-1-[(R)-1-Phenylethyl]pyrrolidin-2-yl}methanol
1f. Colourless oil; Rf 0.24 (ethyl acetate); ½aꢁD ¼ þ5:0 (c
20
1.0, CHCl3); m (film) 3405 (OH), 3091, 3065, 3025, 1609,
1489 (HC@C), 1040 cmꢀ1 (CO); dH 1.40 (3H, d,
J = 6.8 Hz, Me), 1.66–1.76, 1.83–1.93 (3H and 1H, respec-
tively, 2m, CH2CH2CH), 2.55–2.61 (1H, m, 1 ꢂ CHHN),
2.67 (1H, br s, OH), 2.89–3.03 (3H, m, 1 ꢂ CHHN, CHCO
and 1 ꢂ CHHO), 3.13 (1H, dd, J = 10.4 and 2.0 Hz,
1 ꢂ CHHO), 3.75 (1H, q, J = 6.6 Hz, CHMe), 7.22–7.37
(5H, m, ArH); dC 17.7 (Me), 24.4, 29.4 (CH2CH2CH),
50.5 (CH2N), 60.9, 62.0 (2 ꢂ CHN), 63.4 (CH2O), 127.1,
127.4 (2C), 128.3 (2C), 144.8 (ArC); m/z 205 (M+, <1%),
190 (12), 175 (10), 174 (75), 106 (10), 105 (100), 103 (11),
79 (12), 77(14), 70 (82); HRMS: M+ found 205.1475,
C13H19NO requires 205.1467.
4.5. Reduction of esters 10b,d–g. General procedure
A solution of amino ester 10b,d–g (12.6 mmol) in anhy-
drous Et2O (25 mL) was added dropwise for ca. 10 min
to a stirred suspension of LiAlH4 (1.7 g, 37.8 mmol) in
anhydrous Et2O (50 mL) under Ar at 0 °C. The reaction
mixture was stirred for 3 h, allowing the temperature to rise
to room temperature. Then, the reaction was quenched fol-
lowing a literature procedure.25 Evaporation of the sol-
vents gave the expected aminoalcohols 1b,d–g in pure
form in 68%, 49%, 33%, 82% and 85% yield, respectively.
The corresponding physical, spectroscopic and analytical
data for compounds 1b,d–g follow.
4.5.5. {(2S)-1-[(S)-1-Phenylethyl]pyrrolidin-2-yl}methanol
20
1g. Colourless oil; Rf 0.22 (ethyl acetate); ½aꢁD ¼ ꢀ20:0
4.5.1. [(2S)-1-Isopropylpyrrolidin-2-yl]methanol 1b.26 Yel-
(c 1.1, CHCl3); m (film) 3386 (OH), 3091, 3065, 3018,
1602, 1502 (HC@C), 1032 cmꢀ1 (CO); dH 1.44 (3H, d,
J = 6.9 Hz, Me), 1.54–1.89 (4H, m, CH2CH2CH), 2.34–
2.42 (1H, m, 1 ꢂ CHHN), 2.86–3.00 (2H, m, 1 ꢂ CHHN
and CHCO), 3.11 (1H, br s, OH), 3.39 (1H, dd, J = 10.4
and 2.3 Hz, 1 ꢂ CHHO), 3.65 (1H, dd, J = 10.4 and
4.2 Hz, 1 ꢂ CHHO), 3.78 (1H, q, J = 6.9 Hz, CHMe),
7.22–7.40 (5H, m, ArH); dC 21.9 (Me), 24.0, 29.3
(CH2CH2CH), 51.0 (CH2N), 60.1, 60.8 (2 ꢂ CHN), 63.7
(CH2O), 127.0, 127.7 (2C), 128.1 (2C), 142.8 (ArC); m/z
205 (M+, <1%), 175 (11), 174 (80), 105 (100), 103 (11),
79 (12), 77 (14), 70 (77); HRMS: M+ found 205.1451,
C13H19NO requires 205.1467.
20
low oil; Rf 0.28 (ethyl acetate); ½aꢁD ¼ ꢀ27:2 (c 1.0,
CHCl3); m (film) 3384 (OH), 1043 cmꢀ1 (CO); dH 0.99,
1.11 (3H each, 2d, J = 6.5 Hz each, 2 ꢂ Me), 1.65–1.90
(4H, m, CH2CH2CH), 2.52–2.60 (1H, m, 1 ꢂ CH2CHHN),
2.85–3.32 (3H, m, 1 ꢂ CH2CHHN and 2 ꢂ CH), 3.13 (1H,
br s, OH), 3.33 (1H, dd, J = 10.4 and 2.8 Hz, 1 ꢂ CHHO),
3.53 (1H, dd, J = 10.4 and 4.5 Hz, 1 ꢂ CHHO); dC 16.6,
22.2 (2 ꢂ Me), 24.2, 29.0 (CH2CH2CH), 47.7 (CHMe),
49.9 (CH2N), 60.0 (CH2O), 63.2 (CHCH2); m/z 143 (M+,
3%), 142 (12), 128 (18), 113 (10), 112 (100), 70 (82).
4.5.2. [(2S)-1-(1-Naphthylmethyl)pyrrolidin-2-yl]methanol
1d.27 Yellow oil; Rf 0.80 (hexane/ethyl acetate: 1:1);
20
½aꢁD ¼ ꢀ30:6 (c 1.1, CHCl3); m (film) 3429 (OH), 3071,
4.6. Preparation of ligands 1h and 1i
3057, 3045, 1597, 1510 (HC@C), 1045 cmꢀ1 (CO); dH
1.56–1.75, 1.77–1.88, 1.92–1.99 (2H, 1H and 1H, res-
pectively, 3m, CH2CH2CH), 2.31–2.43 (1H, m, 1 ꢂ
CH2CHHN), 2.69 (1H, br s, OH), 2.77–2.87 (1H, m,
CHN), 2.87–2.95 (1H, m, 1 ꢂ CH2CHHN), 3.42 (1H, d,
J = 10.8 Hz, 1 ꢂ CHHO), 3.64 (1H, dd, J = 10.8, 2.7 Hz,
1 ꢂ CHHO), 3.78, 4.40 (1H each, 2d, J = 12.9 Hz each,
CH2Ar), 7.37–7.54 (4H, m, 4 ꢂ ArH), 7.76, 7.84, 8.17
(1H each, 3d, J = 8.2 Hz each, 3 ꢂ ArH); dC 21.1, 23.8
(CH2CH2CH), 55.2, 57.1 (2 ꢂ CH2N), 62.5 (CH2O), 65.1
(CHN), 123.9, 125.3, 125.7, 125.8, 126.2, 127.0, 128.1,
128.7, 132.2, 135.2 (ArC); m/z 241 (M+, 1%), 210 (45),
142 (13), 141 (100), 115 (14).
MeMgBr (15.0 mmol, 5 mL of a 3 M solution in THF, for
1h) or PhMgBr (15.0 mmol, 5 mL of a 3 M solution in
Et2O, for 1i) was added dropwise to a solution of the com-
mercially available aminoester 11 (1.2 g, 5.0 mmol) in
anhydrous THF (25 mL) under Ar at ꢀ78 °C and the reac-
tion was stirred overnight allowing the temperature to rise
to room temperature. Then, water (25 mL) was added and
the resulting mixture was extracted with Et2O (2 ꢂ 50 mL).
The combined organic layers were washed with brine
(10 mL) and dried (MgSO4). After filtration and evapora-
tion of the solvents, compound 1h was obtained in pure
form in quantitative yield. Compound 1i was purified by
crystallization with hexane, giving a 50% yield. The corre-
sponding physical and spectroscopic data for compounds
1h and 1i follow.
4.5.3. [(2S)-1-(2-Naphthylmethyl)pyrrolidin-2-yl]methanol
1e.27 White solid; Rf 0.62 (hexane/ethyl acetate: 1:1);
20
mp 80 °C; ½aꢁD ¼ ꢀ34:4 (c 1.0, CHCl3); m (KBr) 3333
(OH), 3057, 3043, 3014, 1598, 1508 (HC@C), 1046 cmꢀ1
(CO); dH 1.65–2.03 (5H, m, CH2CH2CH and OH), 2.29–
2.38 (1H, m, 1 ꢂ CH2CHHN), 2.75–2.82 (1H, m, CHN),
2.95–3.01 (1H, m, 1 ꢂ CH2CHHN), 3.45 (1H, dd, J = 7.1
and 0.6 Hz, 1 ꢂ CHHO), 3.72 (1H, dd, J = 7.1 and
3.6 Hz, 1 ꢂ CHHO), 3.51, 4.12 (1H each, 2d, J = 13.0 Hz
each, CH2Ar), 7.42–7.49, 7.72–7.84 (3H and 4H, respec-
tively, 2m, ArH); dC 23.6, 27.9 (CH2CH2CH), 54.7, 58.9
(2 ꢂ CH2N), 62.0 (CH2O), 64.5 (CHN), 125.8, 126.2,
127.15, 127.2, 127.8 (2C), 128.2, 132.8, 133.5, 137.0
(ArC); m/z 241 (M+, 3%), 210 (38), 142 (18), 141 (100),
115 (16).
4.6.1. 2-[(2S)-1-Benzylpyrrolidin-2-yl]propan-2-ol 1h.28
20
Yellow oil; Rf 0.30 (hexane/ethyl acetate: 9:1); ½aꢁD
¼
ꢀ42:1 (c 2.5, CH2Cl2); m (film) 3446 (OH), 3093, 3071,
3022, 1602, 1494 (HC@C), 1071 cmꢀ1 (CO); dH 1.17, 1.26
(3H each, 2s, 2 ꢂ Me), 1.65–1.94 (4H, m, CH2CH2CH),
2.37–2.45 (1H, m, 1 ꢂ CH2CHHN), 2.65 (1H, br s, OH),
2.75 (1H, dd, J = 5.1 and 3.6 Hz, CHN), 2.86–2.93 (1H,
m, 1 ꢂ CH2CHHN), 3.59, 4.14 (1H each, 2d, J = 13.9 Hz
each, CH2Ar), 7.21–7.37 (5H, m, ArH); dC 25.25, 25.3
(2 ꢂ Me), 27.9, 28.8 (CH2CH2CH), 55.5, 63.2 (2 ꢂ CH2N),
72.8 (C), 73.0 (CH), 126.9, 128.2 (2C), 128.4 (2C), 140.6
(ArC); m/z 219 (M+, <1%), 161 (17), 160 (100), 91 (86).

R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
2837
4.6.2. [(2S)-1-Benzylpyrrolidin-2-yl]diphenylmethanol 1i.29
(Fig. 4).21 The corresponding physical, spectroscopic and
White solid; Rf 0.60 (hexane/ethyl acetate: 9:1); mp
121 °C (hexane); ½aꢁD ¼ þ87:5 (c 1.0, CHCl3); m (KBr)
analytical data for compounds 1k–m follow.
20
3435 (OH), 3085, 3051, 3025, 1607, 1493 (HC@C),
1099 cmꢀ1 (CO); dH 1.55–1.85, 1.88–2.02 (3H and 1H,
respectively, 2m, CH2CH2CH), 2.29–2.42, 2.87–2.93 (1H
each, 2m, CH2CH2N), 3.01, 3.21 (1H each, 2d,
J = 12.6 Hz, CH2Ph), 3.97 (1H, dd, J = 4.8 and 4.5 Hz,
CHN), 4.92 (1H, s, OH), 6.90–7.41 (11H, m, 11 ꢂ ArH),
7.59, 7.73 (2H each, 2d, J = 7.3 Hz each, 4 ꢂ ArH); dC
24.3, 29.9 (CH2CH2CH), 55.7, 60.7 (2 ꢂ CH2N), 70.8
(CHN), 78.1 (CO), 125.7 (2C), 125.8 (2C), 126.4, 126.5,
127.0, 128.2 (2C), 128.25 (2C), 128.3 (2C), 128.7 (2C),
139.8, 146.8, 148.2 (ArC); m/z 343 (M+, <1%), 161 (13),
160 (100), 91 (51).
4.8.1. (S)-1-[(2S)-1-Benzylpyrrolidin-2-yl]ethanol 1k. Yel-
low oil; Rf 0.51 (hexane/ethyl acetate: 1:1); ½aꢁD ¼ ꢀ12:6
20
(c 0.7, CHCl3); m (film) 3415 (OH), 3085, 3061, 3027,
1602, 1495 (HC@C), 1074 cmꢀ1 (CO); dH 1.17 (3H, d,
J = 6.2 Hz, Me), 1.49–1.58, 1.62–1.81, 1.87–2.00 (1H, 2H
and 1H, respectively, 3m, CH2CH2CH), 2.38–2.47 (1H,
m, 1 ꢂ CH2CHHN), 2.66–2.72 (1H, m, CHN), 2.84–2.91
(1H, m, 1 ꢂ CH2CHHN), 3.36–3.45 (1H, m, CHO), 3.56,
3.98 (1H each, 2d, J = 13.2 Hz each, CH2Ph), 3.58 (1H,
br s, OH), 7.21–7.36 (5H, m, ArH); dC 20.2 (Me), 24.5,
28.0 (CH2CH2CH), 53.9, 62.0 (2 ꢂ CH2N), 69.8 (CHN),
70.6 (CHO), 127.1, 128.4 (2C), 128.7 (2C), 139.7 (ArC);
m/z 205 (M+, <1%), 161 (13), 160 (100), 91 (90). HRMS:
M+ found 205.1416, C13H19NO requires 205.1467.
4.7. Oxidation of N-benzylprolinol 1c to the corresponding
aldehyde 12
4.8.2.
(R)-[(2S)-1-Benzylpyrrolidin-2-yl](phenyl)methanol
DMSO (2.1 mL, 37.0 mmol) was added to a solution of
oxalyl chloride (1.5 mL, 17.0 mmol) in anhydrous CH2Cl2
(95 mL) at ꢀ78 °C for ca. 10 min. The reaction mixture
was stirred for 15 min and then a solution of aminoalco-
hol 1c (3 g, 15.7 mmol) in anhydrous CH2Cl2 (25 mL)
was added over a period of 10 min. The reaction mix-
ture was stirred for 15 min and triethylamine (9 mL,
68.0 mmol) was added over a period of 10 min. The
reaction mixture was stirred allowing the temperature to
reach room temperature and then it was washed with
brine (100 mL). The aqueous phase was extracted with
CH2Cl2 (3 ꢂ 100 mL). The combined organic layers were
dried over MgSO4. After filtration and evaporation of
the solvent, the crude aldehyde 12 was used in the next
step, due to decomposition when its purification was
attempted.
1l.30 Yellow oil; Rf 0.75 (hexane/ethyl acetate: 1:1);
20
½aꢁD ¼ ꢀ69:2 (c 1.3, CHCl3); m (film) 3442 (OH), 3085,
3058, 3027, 1615, 1495 (HC@C), 1116 cmꢀ1 (CO); dH
1.20–1.43, 1.56–1.79 (1H and 3H, respectively, 2m,
CH2CH2CH), 2.33 (1H, q, J = 8.5 Hz, 1 ꢂ CH2CHHN),
2.85–2.91 (1H, m, CHCO), 3.00–3.06 (1H, m,
1 ꢂ CH2CHHN), 3.46, 4.19 (1H each, 2d, J = 13.0 Hz
each, CH2Ph), 3.73 (1H, br s, OH), 4.90 (1H, d,
J = 2.9 Hz, CHO), 7.21–7.39 (10H, m, ArH); dC 23.1,
23.9 (CH2CH2CH), 54.6, 58.2 (2 ꢂ CH2N), 69.1 (CHN),
70.1 (CHO), 125.4 (2C), 126.7, 127.1, 128.1 (2C), 128.4
(2C), 128.7 (2C), 139.0, 141.5 (ArC); m/z 267 (M+, <1%),
161 (12), 160 (100), 91 (70).
4.8.3.
(S)-[(2S)-1-Benzylpyrrolidin-2-yl](phenyl)methanol
1m.30 White solid; Rf 0.67 (hexane/ethyl acetate: 1:1);
20
mp 75 °C (hexane); ½aꢁD ¼ þ101:8 (c 1.1, CHCl3); m
4.8. Synthesis of ligands 1j–m. General procedure
(KBr) 3150 (OH), 3085, 3056, 3027, 1489 (HC@C),
1070 cmꢀ1 (CO); dH 1.71–1.83, 1.89–2.04 (3H and 1H,
respectively, 2m, CH2CH2CH), 2.35–2.47, 2.93–3.00 (1H
each, 2m, CH2CH2N), 3.06–3.12 (1H, m, CHN), 3.35,
3.66 (1H each, 2d, J = 12.9 Hz each, CH2Ph), 4.40 (1H,
d, J = 4.5 Hz, CHO), 7.20–7.40 (10H, m, ArH); dC 24.2,
29.3 (CH2CH2CH), 54.2, 61.2 (2 ꢂ CH2N), 70.1 (CHN),
75.2 (CHO), 126.1 (2C), 127.0, 127.1, 128.2 (2C), 128.3
(2C), 128.6 (2C), 139.4, 143.7 (ArC); m/z 267 (M+, <1%),
216 (12), 161 (13), 160 (100), 91 (74).
MeMgBr (45 mmol, 15 mL of a 3 M solution in THF, for
1j and 1k) or PhMgBr (45 mmol, 15 mL of a 3 M solution
in Et2O, for 1l and 1m) was added dropwise to a suspension
of dry CeCl3 (10.0 g, 45 mmol) in Et2O (56 mL) at ꢀ78 °C.
After stirring for 1 h at that temperature, aldehyde 12
(2.8 g, 15 mmol) in Et2O (28 mL) was added. The reaction
mixture was allowed to reach room temperature overnight.
The solvent was removed under reduced pressure and the
residue was dissolved in CH2Cl2 (200 mL) and washed with
water (100 mL). The aqueous phase was extracted with
CH2Cl2 (3 ꢂ 100 mL). The combined organic layers were
dried over MgSO4. The residue obtained after filtration
and evaporation of the solvent was a mixture of epimers
1j/1k (81% combined yield, 1:3 ratio by 1H NMR) or
4.9. Preparation of N-benzyl aminoester 14
Compound 14 was prepared from the commercially avail-
able aminoacid 13 by the same sequence esterification–N-
alkylation described above for the synthesis of aminoesters
10d–g. The overall yield of 14 was 80%. Its corresponding
physical, spectroscopic and analytical data follow.
1
1l/1m (77% combined yield, 3:7 ratio by H NMR). The
two pairs of epimers could be separated by flash chroma-
tography (deactivated silica gel, hexane/ethyl acetate),
affording pure compounds 1j, 1k, 1l and 1m in 26%, 40%,
20% and 26% yield, respectively. The absolute stereochem-
istry of ligand 1j was determined by comparison of its
physical and spectroscopic data with the ones reported in
the literature.20 Compound 1m was recrystallized in hexane
and the crystals obtained were studied by X-ray diffraction
analysis, giving the (S)-configuration for the carbinol site
4.9.1. (S)-Ethyl 1-benzyl-2-methylpyrrolidine-2-carboxylate
14. Yellow oil; Rf 0.69 (hexane/diethyl ether: 4:1);
20
½aꢁD ¼ þ42:8 (c 1.0, CHCl3); m (film) 3085, 3062, 3027,
1604, 1494 (HC@C), 1724 (C@O), 1179 cmꢀ1 (CO); dH
1.30 (3H, t, J = 7.1 Hz, MeCH2), 1.40 (3H, s, MeC), 1.71–
1.89, 2.20–2.29 (3H and 1H, respectively, 2m, CH2CH2C),
2.69–2.75, 2.80–2.86 (1H each, 2m, CH2CH2N), 3.51, 3.83

2838
R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
(1H each, 2d, J = 13.4 Hz each, CH2Ph), 4.13–4.24 (2H, m,
CH2O), 7.19–7.36 (5H, m, ArH); dC 14.4 (MeCH2), 21.6
(MeC), 21.5, 37.7 (CH2CH2C), 51.3, 53.6 (2 ꢂ CH2N),
60.1 (CH2O), 67.5 (CMe), 126.6, 128.1 (2C), 128.3 (2C),
140.4 (ArC), 175.4 (CO); m/z 247 (M+, <1%), 175 (13),
174 (100), 91 (57); HRMS: M+ꢀEt found 218.1178,
C15H21NO2 requires 218.1181.
1494 (HC@C), 1075 cmꢀ1 (CO); dH 1.07 (3H, d,
J = 6.0 Hz, Me), 1.33–1.45, 1.68–1.87 (1H and 3H, respec-
tively, 2m, CH2CH2CH), 2.51 (1H, br s, OH), 2.77–2.84,
2.86–2.92 (1H each, 2m, 2 ꢂ CHN), 3.21–3.28 (2H, m,
CH2O), 3.63, 3.82 (1H each, 2d, J = 13.5 Hz each, CH2Ph),
7.21–7.35 (5H, m, ArH); dC 20.6 (Me), 26.8, 32.8
(CH2CH2CH), 56.8 (CH2N), 60.5, 65.3 (2 ꢂ CHN), 62.5
(CH2O), 127.0, 128.2 (2C), 128.8 (2C), 139.4 (ArC); m/z
205 (M+, <1%), 175 (14), 174 (100), 91 (85).
4.10. Reduction of aminoester 14. Synthesis of ligand 1n
The reduction of 14 was carried out in the same way as for
compounds 10b,d–g. The isolated yield of 1n was 76%. Its
corresponding physical, spectroscopic and analytical data
follow.
4.13. Addition of dialkylzinc reagents to imines 3 catalyzed
by ligands 1. Preparation of compounds 5. General procedure
The dialkylzinc reagent (1.5 mmol) was added dropwise
over ca. 10 min to a stirred solution of imine 3 (0.5 mmol)
and ligand 1 (0.25 mmol) in anhydrous toluene (3 mL)
under argon at room temperature. After stirring for the
time indicated in Tables 2 and 3, the reaction was hydro-
lyzed with an aqueous saturated solution of NH4Cl
(5 mL). Water (5 mL) was added and the mixture was
extracted with ethyl acetate (3 ꢂ 20 mL). The combined
organic layers were washed with brine (5 mL) and then
dried (Na2SO4). After filtration and evaporation of the sol-
vents, the crude residue was purified by column chroma-
tography (silica gel, pentane/acetone), to give products 5
in the yields and enantiomeric excesses indicated in Tables
2 and 3. Compounds 5aa,7 5ab,7 5ad,14 5b,7 5c,10j 5e33 and
5f33 were characterized by comparison of their physical and
spectroscopic data with the ones reported in the literature.
These products were analyzed by HPLC on a ChiralCel
OD-H column using a 254 nm UV detector, 10% i-PrOH
in hexane as eluent and a flow rate of 0.5 mL/min or on
a Chiralpak AD column using a 254 nm UV detector,
20% i-PrOH in hexane as eluent and a flow rate of
1.0 mL/min. The retention times were 13.7 (R) and 18.0
(S) for 5aa (OD-H column), 14.8 (R) and 19.6 (S) for 5ab
(OD-H column), 21.2 (R) and 26.0 (S) for 5ac (OD-H col-
umn, 5% i-PrOH in hexane as eluent), 11.8 (R) and 19.8 (S)
for 5ad (OD-H column, 5% i-PrOH in hexane as eluent),
12.2 (R) and 14.3 (S) for 5b (AD column), 11.1 (R) and
13.8 (S) for 5c (AD column), 9.3 (R) and 12.9 (S) for 5d
(AD column), 7.6 (S) and 11.0 (R) for 5e (AD column),
18.2 (S) and 22.3 (R) for 5f (AD column, 10% i-PrOH in
hexane as eluent). The absolute configuration of the major
enantiomer of 5aa was determined by hydrolysis of it7 and
comparison of the specific rotation of the free amine ob-
tained with the reported data.7 The absolute configuration
of the major enantiomer of 5ab–ad was tentatively assigned
according to the order of elution of the two enantiomers in
the HPLC analysis on the analogy of product 5aa. For
addition products 5b–d, the absolute configuration of the
major enantiomer was tentatively assigned according to
the HPLC data described in the literature for similar com-
pounds under the same conditions.33 The retention times of
the two enantiomers of compounds 5e and 5f have already
been described.33 The corresponding physical and spectro-
scopic data for compounds 5ac and 5d follow.
4.10.1. [(2S)-1-Benzyl-2-methylpyrrolidin-2-yl]methanol 1n.
20
Yellow oil; Rf 0.24 (ethyl acetate); ½aꢁD ¼ ꢀ35:7 (c 1,
CHCl3); m (film) 3423 (OH), 3085, 3062, 3027, 1602, 1495
(HC@C), 1050 cmꢀ1 (CO); dH 0.99 (3H, s, Me), 1.51–
1.78, 2.07–2.16 (3H and 1H, respectively, 2m, CH2CH2C),
2.40–2.48, 2.88–2.94 (1H each, 2m, CH2CH2N), 3.21, 3.78
(1H each, 2d, J = 12.8 Hz each, CH2Ph), 3.32, 3.44 (1H
each, 2d, J = 10.4 Hz each, CH2O), 7.21–7.36 (5H, m,
ArH); dC 17.4 (Me), 21.8, 35.3 (CH2CH2C), 51.3, 52.5
(2 ꢂ CH2N), 63.6 (C), 65.1 (CH2O), 127.0, 128.4 (2C),
128.6 (2C), 139.9 (ArC); m/z 205 (M+, <1%), 175 (13),
174 (100), 91 (80). HRMS: M+ found 205.1464,
C13H19NO requires 205.1467.
4.11. Preparation of N-benzyl aminoester 16
The cis-5-methyl proline methyl ester 15 was prepared
according to a literature procedure.31 Benzylation of 15
was performed in the same way as for the synthesis of com-
pounds 10d–g. The isolated yield of 16 was 30%. Its corre-
sponding physical, spectroscopic and analytical data
follow.
4.11.1. (2S,5S)-Methyl 1-benzyl-5-methylpyrrolidine-2-carb-
oxylate 16. Yellow oil; Rf 0.26 (hexane/diethyl ether: 4:1);
20
½aꢁD ¼ ꢀ11:5 (c 1.2, CHCl3); m (film) 3085, 3062, 3027,
1604, 1494 (HC@C), 1747 (C@O), 1197 cmꢀ1 (CO); dH
1.18 (3H, d, J = 5.8 Hz, MeCH), 1.50–1.65, 1.81–1.98
(1H and 3H, respectively, 2m, CH2CH2CH), 2.62–2.73,
3.23–3.28 (1H each, 2m, 2 ꢂ CHN), 3.47 (3H, s, MeO),
3.62, 3.94 (1H each, 2d, J = 13.7 Hz each, CH2Ph), 7.20–
7.31 (5H, m, ArH); dC 19.5 (MeCH), 27.7, 32.2
(CH2CH2CH), 51.5 (MeO), 56.5 (CH2N), 59.5, 66.0
(2 ꢂ CHN), 126.9, 127.9 (2C), 129.5 (2C), 137.7 (ArC),
174.9 (CO); m/z 233 (M+, < 1%), 175 (14), 174 (100), 91
(77). HRMS: M+ found 233.1406, C14H19NO2 requires
233.1416.
4.12. Reduction of aminoester 16. Synthesis of ligand 1o
The reduction of 16 was carried out in the same way as for
compounds 10b,d–g. The isolated yield of 1o was 92%. Its
corresponding physical and spectroscopic data follow.
4.13.1. N-(2-Methyl-1-phenylpropyl)-P,P-diphenylphosphi-
nic amide 5ac14. White solid; Rf 0.38 (ethyl acetate); mp
4.12.1. [(2S,5S)-1-Benzyl-5-methylpyrrolidin-2-yl]methanol
20
20
1o.32 Yellow oil; Rf 0.52 (ethyl acetate); ½aꢁ ¼ þ39:2 (c
166 °C; ½aꢁD ¼ þ14:6 (c 1.0, CHCl3) 90% ee; m (KBr)
1.0, CHCl3); m (film) 3417 (OH), 3091, 3061D, 3027, 1596,
3443 (NH), 3088, 3060, 3028, 1503 (HC@C), 1165 cmꢀ1

R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
2839
Oxford, 1991; (b) Erdik, E. Organozinc Reagents in Organic
Synthesis; CRC Press: Boca Raton, 1996.
(P@O); dH 0.82, 1.01 (3H each, 2 d, J = 6.9 Hz each,
2 ꢂ Me), 1.90–2.11 (1H, m, CHMe), 3.33 (1H, m, NH),
3.80–3.99 (1H, m, CHN), 7.00–7.11, 7.15–7.54, 7.59–7.73,
7.77–7.91 (2H, 9H, 2H and 2H, respectively, 4 m, ArH);
dC 19.2, 19.3 (2 ꢂ Me), 35.7 (d, J = 4.3 Hz, CHMe), 61.3
(CN), 126.8, 126.9, 128.0, 128.1, 128.2, 128.4, 128.5,
131.7, 131.8, 131.9, 132.6, 132.7, 142.9 (d, J = 3.9 Hz)
(ArC); m/z 349 (M+, <1%), 307 (22), 306 (100), 201 (50).
9. (a) Rozema, M. J.; Sidduri, A.; Knochel, P. J. Org. Chem.
1992, 57, 1956–1958; For a review, see: (b) Knochel, P.; Singer,
R. D. Chem. Rev. 1993, 93, 2117–2188; (c) Vettel, S.; Vaupel,
A.; Knochel, P. Tetrahedron Lett. 1995, 36, 1023–1026; (d)
Vettel, S.; Vaupel, A.; Knochel, P. J. Org. Chem. 1996, 61,
7473–7481; (e) Langer, F.; Schwink, L.; Devasagayaraj, A.;
Chavant, P.-Y.; Knochel, P. J. Org. Chem. 1996, 61, 8229–
8243; (f) Micouin, L.; Knochel, P. Synlett 1997, 327–328.
10. See, for instance: (a) Soai, K.; Hatanaka, T.; Miyazawa, T. J.
Chem. Soc., Chem. Commun. 1992, 1097–1098; (b) Suzuki, T.;
Narisada, N.; Shibata, T.; Soai, K. Tetrahedron: Asymmetry
1996, 7, 2519–2522; (c) Guijarro, D.; Pinho, P.; Andersson, P.
G. J. Org. Chem. 1998, 63, 2530–2535; (d) Jimeno, C.; Vidal-
4.13.2.
N-[1-(2-Naphthyl)propyl]-P,P-diphenylphosphinic
amide 5d.14 White solid; Rf 0.29 (ethyl acetate); mp
20
105 °C; ½aꢁD ¼ þ27:5 (c 1.0, CHCl3) 86% ee; m (KBr) 3127
(NH), 3040, 3026, 3012, 1598 (HC@C), 1114 cmꢀ1 (P=O);
dH 0.81 (3H, t, J = 7.4 Hz, Me), 1.88–2.17 (2H, m, CH2),
3.27–3.46 (1H, m, NH), 4.21–4.34 (1H, m, CHN), 7.21–
7.30, 7.32–7.57, 7.65–7.93 (2H, 8H and 7H, respectively,
3m, ArH); dC 10.6 (Me), 32.3 (d, J = 4.0 Hz, CH2), 57.3
(CN), 124.5, 125.5, 125.7, 126.1, 127.6, 127.8, 128.2,
128.3, 128.4, 128.5, 131.65, 131.7, 131.75, 131.8, 131.9,
132.5, 132.6, 132.7, 133.2, 140.7 (d, J = 5.4 Hz) (ArC);
m/z 386 (M++1, 2%), 385 (M+, 6), 357 (25), 356 (100),
202 (16), 201 (89), 185 (13), 184 (83), 154 (13), 77 (17).
`
Ferran, A.; Moyano, A.; Pericas, M. A.; Riera, A. Tetrahe-
dron Lett. 1999, 40, 777–780; (e) Brandt, P.; Hedberg, C.;
Lawonn, K.; Pinho, P.; Andersson, P. G. Chem. Eur. J. 1999,
`
5, 1692–1699; (f) Jimeno, C.; Reddy, K. S.; Sola, L.; Moyano,
`
A.; Pericas, M. A.; Riera, A. Org. Lett. 2000, 2, 3157–3159;
(g) Pinho, P.; Andersson, P. G. Tetrahedron 2001, 57, 1615–
1618; (h) Zhang, H.-L.; Zhang, X.-M.; Gong, L.-Z.; Mi, A.-
Q.; Cui, X.; Jiang, Y.-Z.; Choi, M. C. K.; Chan, A. S. C. Org.
Lett. 2002, 4, 1399–1402; (i) Beresford, K. J. M. Tetrahedron
Lett. 2002, 43, 7175–7177; (j) Zhang, H.-L.; Jiang, F.; Zhang,
X.-M.; Cui, X.; Gong, L.-Z.; Mi, A.-Q.; Jiang, Y.-Z.; Wu, Y.-
D. Chem. Eur. J. 2004, 10, 1481–1492; (k) Beresford, K. J. M.
Tetrahedron Lett. 2004, 45, 6041–6044.
Acknowledgements
11. (a) Zhang, X.; Lin, W.; Gong, L.; Mi, A.; Cui, X.; Jiang, Y.;
Choi, M. C. K.; Chan, A. S. C. Tetrahedron Lett. 2002, 43,
1535–1537; (b) Zhang, X.-M.; Zhang, H.-L.; Lin, W.-Q.;
Gong, L.-Z.; Mi, A.-Q.; Cui, X.; Jiang, Y.-Z.; Yu, K.-B. J.
Org. Chem. 2003, 68, 4322–4329.
´
This work was generously supported by the Direccion Gen-
eral de Ensenanza Superior (DGES) of the current Spanish
Ministerio de Educacion y Ciencia (MEC; Grant No.
CSD2007-00006) and the Generalitat Valenciana (GV/
2007/036). R.A. thanks the Spanish Ministerio de Educa-
cion y Ciencia for a predoctoral fellowship. We also thank
˜
´
12. (a) Shi, M.; Wang, C.-J. Adv. Synth. Catal. 2003, 345, 971–
973; (b) Boezio, A. A.; Charette, A. B. J. Am. Chem. Soc.
´
´
2003, 125, 1692–1693; (c) Boezio, A. A.; Pytkowicz, J.; Coˆte,
MEDALCHEMY S.L. for a gift of chemicals.
A.; Charette, A. B. J. Am. Chem. Soc. 2003, 125, 14260–
´
14261; (d) Coˆte, A.; Boezio, A. A.; Charette, A. B. Angew.
Chem., Int. Ed. 2004, 43, 6525–6528; (e) Wang, M.-C.; Liu,
L.-T.; Hua, Y.-Z.; Zhang, J.-S.; Shi, Y.-Y.; Wang, D.-K.
Tetrahedron: Asymmetry 2005, 16, 2531–2534; (f) Kim, B.-S.;
Kang, S.-W.; Kim, K. H.; Ko, D.-H.; Chung, Y.; Ha, D.-C.
Bull. Korean Chem. Soc. 2005, 26, 1501–1502; (g) Desrosiers,
References
1. Jacques, J.; Collet, A.; Wilen, S. H. In Enantiomers, Race-
mates and Resolution; John Wiley & Sons: New York, 1981.
2. Moser, H.; Rihs, G.; Santer, H. Z. Naturforsch. 1982, 376,
451–462.
3. Whitesell, J. K. Chem. Rev. 1989, 89, 1581–1590.
4. (a) Volkmann, R. A.. In Comprehensive Organic Synthesis;
Trost, B. M., Fleming, I., Shreiber, S. L., Eds.; Pergamon
Press: Oxford, 1991; Vol. 1, Chapter 1.12, (b) Kleinman, E.
F.; Volkmann, R. A.. In Comprehensive Organic Synthesis;
Trost, B. M., Fleming, I., Heathcock, C. H., Eds.; Pergamon
Press: Oxford, 1991; Vol. 2, Chapter 4.3, (c) Bloch, R. Chem.
Rev. 1998, 98, 1407–1438.
´
J.-N.; Coˆte, A.; Boezio, A. A.; Charette, A. B.; Denmark, S.
E.; Montgomery, J. I. Org. Synth. 2006, 83, 5–17.
13. (a) Porter, J. R.; Traverse, J. F.; Hoveyda, A. H.; Snapper,
M. L. J. Am. Chem. Soc. 2001, 123, 984–985; (b) Porter, J. R.;
Traverse, J. F.; Hoveyda, A. H.; Snapper, M. L. J. Am.
Chem. Soc. 2001, 123, 10409–10410.
14. Almansa, R.; Guijarro, D.; Yus, M. Tetrahedron 2007, 63,
1167–1174.
15. Preliminary communication: Almansa, R.; Guijarro, D.; Yus,
M. Tetrahedron: Asymmetry 2007, 18, 896–899.
16. For some recent examples, see: (a) Xu, Q.; Wang, H.; Pan, X.;
Chan, A. S. C.; Yang, T.-k. Tetrahedron Lett. 2001, 42, 6171–
6173; (b) Zhong, G.; Fan, J.; Barbas, C. F. Tetrahedron Lett.
2004, 45, 5681–5684; (c) Liu, X.; Fox, J. M. J. Am. Chem.
Soc. 2006, 128, 5600–5601; (d) Wang, G.-y.; Liu, X.-y.; Zhao,
G. Synlett 2006, 1150–1154; (e) Li, Y.; Liu, X.; Yang, Y.;
Zhao, G. J. Org. Chem. 2007, 72, 288–291.
17. Note that the absolute configuration of the major enantiomer
in Ref. 10c is (S), opposite to the one obtained in our case,
due to the fact that the prolinol substructure found in ligand 2
(R4 = PhCH2, R5 = H) is enantiomeric with respect to our
ligand 1.
5. For reviews on the asymmetric nucleophilic addition to
imines, see: (a) Enders, D.; Reinhold, U. Tetrahedron:
Asymmetry 1997, 8, 1895–1946; (b) Kobayashi, S.; Ishitani,
H. Chem. Rev. 1999, 99, 1069–1094; (c) Charette, A. B.;
´
Boezio, A. A.; Coˆte, A.; Moreau, E.; Pytkowicz, J.; Desro-
siers, J.-N.; Legault, C. Pure Appl. Chem. 2005, 77, 1259–
1267; (d) Vilaivan, T.; Bhanthumnavin, W.; Sritana-Anant,
Y. Curr. Org. Chem. 2005, 9, 1315–1392.
6. For a review on the use of N-phosphinoylimines in stereo-
selective synthesis, see: Weinreb, S. M.; Orr, R. K. Synthesis
2005, 1205–1227.
7. See, for instance: Andersson, P. G.; Guijarro, D.; Tanner, D.
J. Org. Chem. 1997, 62, 7364–7375.
18. (a) Krzyzanowska, B.; Stec, W. J. Synthesis 1982, 270–273;
(b) Jennings, W. B.; Lovely, C. J. Tetrahedron 1991, 47, 5561–
5568.
8. (a) Knochel, P.. In Comprehensive Organic Synthesis; Trost,
B. M., Fleming, I., Shreiber, S. L., Eds.; Pergamon Press:

2840
R. Almansa et al. / Tetrahedron: Asymmetry 18 (2007) 2828–2840
19b
19. (a) Crystal data (excluding structure factors deposited at the
Cambridge Crystallographic Data Centre as Supplementary
Publication Number CCDC 665373): C25H27NO, M =
using the program SAINT
and the integrated intensities
were corrected for Lorentz-polarization effects with SAD-
ABS.19c The structure was solved by direct methods19d and
refined to all 1465 unique F 2o by full matrix least squares.19d
All the hydrogen atoms were placed at idealized positions and
refined as rigid atoms. Final wR2 = 0.1113 for all data and
182 parameters; R1 = 0.0457 for 1079 Fo > 4r(Fo).
22. Pei, Z.; Li, X.; Longenecker, K.; Von Geldern, T. W.;
Wiedeman, P. E.; Lubben, T. H.; Zinker, B. A.; Stewart, K.;
Ballaron, S. J.; Stashko, M. A.; Mika, A. K.; Beno, D. W. A.;
Long, M.; Wells, H.; Kempf-Grote, A. J.; Madar, D. J.;
McDermott, T. S.; Bhagavatula, L.; Fickes, M. G.; Pireh, D.;
Solomon, L. R.; Lake, M. R.; Edalji, R.; Fry, E. H.; Sham,
H. L.; Trevillyan, J. M. J. Med. Chem. 2006, 49, 3520–3535.
23. This proposal is based on the results obtained in theoretical
calculations performed by P. G. Andersson and coworkers
for the addition of dialkylzinc reagents to N-(diphenylphos-
phinoyl)benzaldimine in the presence of 2-azanorbornylmeth-
anols. See Ref. 10e.
24. Nazarova, L. S.; Rozonov, Y. B.; Likhosherstov, A. M.;
Morozova, T. V.; Skoldinov, A. P.; Kaverina, N. V.; Markin,
V. A. Khim.-Farmatsevticheskii Zh. 1984, 18, 1445–1448;
Chem. Abstr 1984, 102, 185050.
25. Micovic, V. M.; Mihailovic, M. L. J. Org. Chem. 1953, 18,
1190–1200.
26. Komatsu, Y.; Shima, K.; Naka, T.; Akahoshi, F. Jpn. Kokai
Tokkyo Koho, JP 2006282602, 2006; Chem. Abstr. 2006, 145,
438531.
27. Kashihara, H.; Suzuki, M.; Ohara, Y. PCT Int. Appl., WO
9417079, 1994; Chem. Abstr. 1994, 123, 168993.
28. Sibi, M. P.; Zhang, R.; Manyem, S. J. Am. Chem. Soc. 2003,
125, 9306–9307.
˚
˚
357.48; monoclinic, a = 8.8977(7) A, b = 9.2003(8) A, c =
3
˚
˚
12.2670(10) A, b = 92.182(2)°; V = 1003.47(14) A ; space
group P2(1); Z = 2; Dc = 1.183 Mg mꢀ3
; k = 0.71073 A;
˚
l = 0.071 mmꢀ1; F(000) = 384; T = 24 1 °C. Data collec-
tion was performed on a Bruker Smart CCD diffractometer,
based on three x-scan runs (starting = ꢀ34°) at values /
= 0°, 120°, 240° with the detector at 2h = ꢀ32°. For each of
these runs, 606 frames were collected at 0.3° intervals and 20 s
per frame. An additional run at / = 0° of 100 frames was
collected to improve redundancy. The diffraction frames were
19b
integrated using the program SAINT
and the integrated
intensities were corrected for Lorentz-polarisation effects with
SADABS.19c The structure was solved by direct methods19d
and refined to all 2463 unique F 2o by full matrix least
squares.19d All the hydrogen atoms were placed at idealised
positions and refined as rigid atoms. Final wR2 = 0.1300 for
all data and 246 parameters; R1 = 0.0424 for 2051 Fo >
4r(Fo); (b) SAINT Version 6.02A: Area-Detector Integration
Software; Siemens Industrial Automation: Madison, WI,
1995.; (c) Sheldrick, G. M. SADABS: Area-Detector Absorp-
tion Correction, Go¨ttingen University, 1996.; (d) SHELX97
[includes SHELXS97, SHELXL97 and CIFTAB]—Programs for
Crystal Structure Analysis (Release 97-2). Sheldrick, G. M.,
Institut fur Anorganische Chemie der Universita¨t,
¨
¨
Tammanstrasse 4, D-3400 Go¨ttingen, Germany, 1998.
20. Weber, B.; Schwerdtfeger, J.; Fro¨hlich, R.; Go¨hrt, A.; Hoppe,
D. Synthesis 1999, 1915–1924.
21. Crystal data (excluding structure factors deposited at the
Cambridge Crystallographic Data Centre as Supplementary
Publication Number CCDC 665374): C18H21NO, M= 267.36;
29. Liu, L.-y.; Sun, J.; Liu, N.; Chang, W.-x.; Li, J. Tetrahedron:
Asymmetry 2007, 18, 710–716.
30. Calvez, O.; Chiaroni, A.; Langlois, N. Tetrahedron Lett.
1998, 39, 9447–9450.
˚
˚
˚
monoclinic, a = 18.450(7) A, b = 6.679(3) A, c = 12.670(5) A,
3
˚
b = 104.011(9)°; V = 1514.8(11) A ; space group C2; Z = 4;
Dc = 1.172 Mg mꢀ3
;
k = 0.71073 A;
l = 0.072 mmꢀ1
;
˚
F(000) = 576; T = 24 1 °C. Data collection was performed
on a Bruker Smart CCD diffractometer, based on three x-
scan runs (starting = ꢀ34°) at values / = 0°, 120°, 240° with
the detector at 2h = ꢀ32°. For each of these runs, 606 frames
were collected at 0.3° intervals and 20 s per frame. An
additional run at / = 0° of 100 frames was collected to
improve redundancy. The diffraction frames were integrated
31. Compound 15 was prepared following methods A1, A2 and
A3 as described in Ref. 22 for the preparation of the
corresponding ethyl ester.
32. Overberger, C. G.; David, K. H.; Moore, J. A. Macromol-
ecules 1972, 5, 368–372.
´
33. Coˆte, A.; Boezio, A. A.; Charette, A. B. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 5405–5410.