Enantioselective Addition of Dialkylzinc Reagents to N-(Diphenylphosphinoyl) Imines Promoted by 2-Azanorbornylmethanols

Article abstract of DOI:10.1021/jo9718096

A set of new β-amino alcohols 2, with the 2-azanorbornyl framework, has been prepared and evaluated as promoters for the enantioselective addition of dialkylzinc reagents to N-(diphenylphosphinoyl) imines 1. By variation of the substitution pattern in the ligand, ee's up to 92% could be obtained. Although a stoichiometric amount of the ligand was used, about 90% of it could be recovered during the workup. Amino alcohol 2b, that gave the best enantioselectivities in the stoichiometric reaction, was also applied in a catalytic process, and ee's up to 85% were achieved using 0.25 equiv of the ligand, which is the highest ee obtained so far using that catalytic amount of the ligand. Addition products 3 could be converted into the free amines 4 without racemization by acidic hydrolysis. The utility of ligands 2 as catalysts in the addition of diethylzinc to benzaldehyde has also been investigated, and ee's up to 75% were achieved.

Full text of DOI:10.1021/jo9718096

2530
J . Org. Chem. 1998, 63, 2530-2535
En a n tioselective Ad d ition of Dia lk ylzin c Rea gen ts to
N-(Dip h en ylp h osp h in oyl) Im in es P r om oted by
2-Aza n or bor n ylm eth a n ols
David Guijarro, Pedro Pinho, and Pher G. Andersson*
Department of Organic Chemistry, Uppsala University, Box 531, S-751 21 Uppsala, Sweden
Received September 30, 1997
A set of new â-amino alcohols 2, with the 2-azanorbornyl framework, has been prepared and
evaluated as promoters for the enantioselective addition of dialkylzinc reagents to N-(diphenylphos-
phinoyl) imines 1. By variation of the substitution pattern in the ligand, ee’s up to 92% could be
obtained. Although a stoichiometric amount of the ligand was used, about 90% of it could be
recovered during the workup. Amino alcohol 2b, that gave the best enantioselectivities in the
stoichiometric reaction, was also applied in a catalytic process, and ee’s up to 85% were achieved
using 0.25 equiv of the ligand, which is the highest ee obtained so far using that catalytic amount
of the ligand. Addition products 3 could be converted into the free amines 4 without racemization
by acidic hydrolysis. The utility of ligands 2 as catalysts in the addition of diethylzinc to
benzaldehyde has also been investigated, and ee’s up to 75% were achieved.
In tr od u ction
The carbon-carbon bond-forming reaction is one of the
most fundamental operations for constructing organic
molecules. Addition of organometallic reagents to car-
bonyl compounds is among the most common reaction for
this purpose, and its asymmetric version is particularly
useful. Among various organometallic compounds, dior-
ganozinc reagents serve as excellent alkyl nucleophiles,
and the enantioselective addition of these reagents to
prochiral aldehydes has been studied extensively.1 These
asymmetric addition reactions are catalyzed by chiral
â-amino alcohols often prepared from the corresponding
R-amino acids. Proline has showed to be especially useful
as precursor for the preparation of highly effective
catalysts.2,3
Chiral bicyclic â-amino alcohols have also been suc-
cessfully used as catalysts for the above-mentioned
reaction. By variation of the bicyclic structure, several
highly enantioselective catalysts have been prepared.4
Moderate to good ee’s were attained in the alkylation of
aromatic and aliphatic aldehydes catalyzed by 2-aza-
norbornylmethanols.5a,b These sterically constrained
â-amino alcohols and their bicyclo[2.2.1] ring system may
F igu r e 1.
block the approach of the attacking species to one of the
enantiotopic faces of the aldehyde. Finally, using a set
of ligands containing the octahydro-cyclopenta[b]pyrrole
system, ee’s up to 99% were achieved.5c
In the last years, we have been interested in the use
of chiral ligands with the 2-azanorbornyl skeleton in
asymmetric synthesis. Some bicyclic analogues of proline
were applied as ligands for the Cu-catalyzed asymmetric
allylic oxidation of olefins.6 The rigid structure of these
unnatural amino acids led to ee’s up to 65% in the
oxidation of cyclic olefins, the highest observed so far
using an amino acid as chiral ligand. In another project,
we have recently studied the addition of dialkylzinc
reagents to N-(diphenylphosphinoyl) imines in the pres-
ence of stoichiometric or catalytic amounts of some new
chiral aziridino alcohols,7 and high enantioselectivities
were attained (ee up to 94%). One drawback with this
system is its low activity when used in catalytic amounts.8
Therefore, we found it interesting to test some bicyclic
â-amino alcohols 2 (Figure 1), having the 2-azanorbornyl
framework, in the above-mentioned addition reaction.
(1) For a review, see: (a) Noyori, R.; Kitamura, M. Angew. Chem.,
Int. Ed. Engl. 1991, 30, 49. (b) Soai, K.; Niwa, S. Chem. Rev. 1992,
92, 833.
(2) (a) Soai, K.; Ookawa, A. J . Chem. Soc., Chem. Commun. 1986,
412. (b) Ookawa, A.; Soai, K. J . Chem. Soc., Perkin Trans. 1 1987,
1465.
(3) (a) Soai, K.; Ookawa, A.; Kaba, T.; Ogawa, K. J . Am. Chem. Soc.
1987, 109, 7111. (b) Corey, E. J .; Hannon, F. J . Tetrahedron Lett. 1987,
28, 5233. (c) Corey, E. J .; Yuen, P.; Hannon, F. J .; Wierda, D. A. J .
Org. Chem. 1990, 55, 784. (d) Shi, M.; Satoh, Y.; Makihara, T.; Masaki,
Y. Tetrahedron: Asymmetry 1995, 6, 2109.
(4) (a) Kitamura, M.; Suga, S.; Kawai, K.; Noyori, R. J . Am. Chem.
Soc. 1986, 108, 6071. (b) Smaardijk, A. A.; Wynberg, H. J . Org. Chem.
1987, 52, 135. (c) Oppolzer, W.; Radinov, R. N. Tetrahedron Lett. 1988,
29, 5645. (d) Tanaka, K.; Ushio, H.; Suzuki, H. J . Chem. Soc., Chem.
Commun. 1989, 1700.
(6) So¨dergren, M. J .; Andersson, P. G. Tetrahedron Lett. 1996, 37,
7577.
(7) (a) Andersson, P. G.; Guijarro, D.; Tanner, D. Synlett 1996, 727.
(b) Andersson, P. G.; Guijarro, D.; Tanner, D. J . Org. Chem. 1997, 62,
7364.
(5) (a) Nakano, H.; Kumagai, N.; Kabuto, C.; Matsuzaki, H.; Hongo,
H. Tetrahedron: Asymmetry 1995, 6, 1233. (b) Nakano, H.; Kumagai,
N.; Matsuzaki, H.; Kabuto, C.; Hongo, H. Tetrahedron: Asymmetry
1997, 8, 1391. (c) Wilken, J .; Kossenjans, M.; Gro¨ger, H.; Martens, J .
Tetrahedron: Asymmetry 1997, 8, 2007.
(8) For earlier results on this subject, see ref 7b.
S0022-3263(97)01809-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/28/1998

Enantioselective Addition of Dialkylzinc Reagents
J . Org. Chem., Vol. 63, No. 8, 1998 2531
Sch em e 1a
Sch em e 3a
Sch em e 2a
Sch em e 4a
glyoxylate and (S)-1-phenylethylamine according to a
literature procedure.10 It turned out to be crucial to
deactivate the silica gel with Et3N before purifying the
adduct 5 by flash chromatography in order to get good
yields, presumably due to decomposition on the otherwise
slightly acidic silica. Pd-catalyzed hydrogenolysis of 5
led to the NH ester 6. The optical purity of 6 was
determined to be 98% by HPLC analysis of the N-benzoyl
ester (see Experimental Section). N-Alkylation of 6
furnished the N-protected esters 7. Treatment of 6 with
MeI in refluxing acetonitrile afforded the N-methyl ester
7a . Reaction of 6 with ethyl or isopropyl bromide in the
presence of K2CO3 and a catalytic amount of 18-crown-6
yielded the corresponding N-ethyl or N-isopropyl esters
7b or 7c, respectively. In the case of 7c, NaI (2 equiv)
was added to the reaction mixture in order to facilitate
the substitution reaction by in situ generation of i-PrI.
The esters 7 were then converted into the ligands 2a -c
by reduction with LiAlH4.
For the synthesis of ligands 2d -g, the N-benzyl amino
ester 8 was required (Scheme 3). It was prepared by
reaction of the NH amino ester 6 with benzyl bromide in
acetonitrile in the presence of K2CO3. Reduction of 8 with
LiAlH4 resulted in ligand 2d . The rest of the ligands,
2e-g, were synthesized by addition of the corresponding
Grignard reagents to the ester 8 in THF. The addition
of i-PrMgBr to 8 gave only a low yield (20%) of the adduct
2f, due to the formation of several unidentified byprod-
ucts.
One of the features of those ligands that prompted us to
choose them was the fact that both enantiomers are
equally available. This is obviously not the case when
using proline or camphor derived ligands, where the
unnatural stereoisomer is much more expensive than the
other one.9 The bicyclic ring system of ligands 2 can be
constructed via an aza-Diels-Alder reaction between
cyclopentadiene and an iminium ion derived from (S)-1-
phenylethylamine and ethyl glyoxylate.10 Since the
amine used as chiral auxiliary is readily available in both
enantiomeric forms, either of the two enantiomeric
ligands can be prepared.
In this paper, we present the application of some new
bicyclic â-amino alcohols 211 (Figure 1) to the addition of
dialkylzinc reagents to N-(diphenylphosphinoyl) imines
(Scheme 1). By fine-tuning of the ligand structure, ee’s
up to 92% could be obtained. Full experimental details
about the preparation of the ligands and the addition
reaction are given. The utility of some of the ligands 2
as catalysts in the same reaction has also been studied.
Acidic hydrolysis of the addition products 3 afforded the
free amines 4 without loss of enantiomeric purity (Scheme
1). Ligands 2 were also evaluated as catalysts for the
well studied addition of diethylzinc to benzaldehyde,
leading to ee’s up to 75%.
The bicyclic amino alcohols 2 were tested as promoters
for the addition of dialkylzinc reagents to N-(diphenyl-
phosphinoyl) imines 1 (Scheme 4), which were prepared
from the corresponding aldehydes according to the lit-
erature.12 Toluene, a nonpolar solvent, was chosen as
reaction medium in order to maximize the rate difference
between the catalyzed and the noncatalyzed reaction. As
a general procedure, the dialkylzinc reagent (3 equiv) was
added dropwise to the stirred solution of the imine 1 and
the ligand 2 (1 equiv) in dry toluene, under Ar, at 0 °C.
Resu lts a n d Discu ssion
The common precursor for all the ligands was the
Diels-Alder adduct 5 (Scheme 2). It was prepared by a
diastereoselective aza-Diels-Alder reaction between cy-
clopentadiene and an iminium ion derived from ethyl
(9) For example, the unnatural D-proline is 24 times more expensive
than the natural L-proline.
(10) (a) Stella, L.; Abraham, H.; Feneau-Dupont, J .; Tinant, B.;
Declercq, J . P. Tetrahedron Lett. 1990, 31, 2603. (b) Waldmann, H.;
Braun, M. Liebigs Ann. Chem. 1991, 1045.
(11) For the preparation of the enantiomer of ligand 2g and its
evaluation in the enantioselective addition of Et2Zn to aldehydes, see
ref 5b.
(12) (a) J ennings, W. B.; Lovely, C. J . Tetrahedron 1991, 47, 5561.
(b) Krzyzanowska, B.; Stec, W. J . Synthesis 1982, 270.

2532 J . Org. Chem., Vol. 63, No. 8, 1998
Guijarro et al.
Ta ble 1. Ad d ition Rea ction of R2Zn to 1 P r om oted by
th e Liga n d s 2
ligand
entry imine (equiv)
yield
R
product (%)a ee (%)b configc
1
2
3
4
5
6
7
8
9
1a
1a
1a
1a
1a
1a
1b
1a
1a
1a
1a
2a (1)
2b (1)
2c (1)
2d (1)
2d (0.25) Et
2d (0.10) Et
2d (1)
2e (1)
2e (1)
2f (1)
2g (1)
Et
3a
3a
3a
3a
3a
3a
3b
3a ′
3a
3a
3a
50
43
59
63
46
38
65
32
65
52
33
75
92
85
91
85
68
92
83
88
43
16
S
S
S
S
S
S
S
S
S
S
S
Et
Et
Et
F igu r e 2.
Et
Med
Et
Sch em e 5a
10
11
Et
Ete
a
Isolated yield after flash chromatography (silica gel, pentane/
b
acetone). Determined by HPLC analysis on a chiral column
(ChiralCel OD-H). c Determined by comparison of the optical
rotation of the free amine with the data given in the literature
d
(see Experimental Section). Six equiv of Me2Zn were used, and
the reaction was run for 6 days at room temperature. e The
reaction was run for 4 days at room temperature.
the workup in a typical experiment. Addition products
3 were converted into the free amines 4 without racem-
ization by acidic hydrolysis (see Experimental Section).
According to our earlier observations,7b a transition
state like the one depicted in Figure 2 could be used to
rationalize the stereochemical outcome of the addition
reactions. The steric hindrance brought about by the
methylene bridge of the ligand (C7, Figure 2) determines
the position of ZnB and, therefore, the configuration of
ZnA. Coordination of the imine to both zinc atoms (N-
ZnA and O-ZnB) would lead to a bicyclic transition state
like the one shown. The presence of a small R2 group in
the ligand seems to be very important in order to get high
selectivities. The steric hindrance between one of the R2
groups of the ligand and the Ar group of the imine carbon
would disfavor the formation of the well-ordered bicyclic
transition state, leading to a decrease in the enantiose-
lectivity, which has been observed experimentally (com-
pare entry 4 with 9-11 in Table 1). The transfer of one
of the ethyl groups of ZnB to the imine would give the
addition products with the observed stereochemistry.
Prompted by the good enantioselectivities obtained
with ligand 2d , we decided to explore its application as
a catalyst in the same transformation. The addition of
Et2Zn to 1a was run with 0.25 and 0.10 equiv of the
ligand 2d (Table 1, entries 5 and 6) under identical
reaction conditions as for the stoichiometric process.
Again, a ligand acceleration effect14 was apparent, since
both the yield and the ee dropped down when the amount
of ligand was reduced. However, the ee’s obtained when
using substoichiometric amounts of the ligand are very
promising. Only a small loss of enantioselectivity was
found with 0.25 equiv of 2d (compare entries 4 and 5).
As a matter of fact, the ee given in entry 5 is the highest
value reported so far with that amount of ligand.15 The
value of 68% obtained when using 10 mol % of the ligand,
although still moderate, is also very promising.
After allowing the reaction mixture to slowly reach rt, it
was stirred until completion according to TLC (2 days).
It was then quenched with saturated aqueous NH4Cl, and
extractive workup afforded the secondary amines 3
(Scheme 4, Table 1).
The yields and ee’s obtained are summarized in Table
1. Both imines 1a and 1b showed to be good substrates
for the addition reaction (Table 1, entries 4 and 7). The
effect of varying the substituents R1 and R2 (Figure 1)
on the selectivity was explored. In accordance to our
previous observations with the aziridino alcohol ligands,7
a primary alcohol on the side chain led to a very high
level of enantioselection. Enantiomeric excesses of 91
and 92% were obtained when ligand 2d was used as
promoter for the addition of Et2Zn to 1a and 1b, respec-
tively (Table 1, entries 4 and 7). The ee decreased upon
increasing the size of the R2 substituent in the ligand
(compare entry 4 with 9-11). When R2 ) Ph, the
reaction was much slower than in the other cases. Only
33% yield of the addition product was isolated after 4
days at room temperature. The ee dropped down to 16%,
probably due to competition between the rate of the
catalyzed and the noncatalyzed reaction. In the case of
the other ligands, the catalyzed reaction was faster than
the noncatalyzed one,13 indicating a ligand acceleration
effect.14
Concerning the substituent on the nitrogen, R1, a more
sterically demanding alkyl group seems to result in a
beneficial effect (Table 1, entries 2-4). The best results
were obtained using either the N-ethyl (2b) or the
N-benzyl ligand (2d ), which gave the addition products
in 92 and 91% ee, respectively (entries 2 and 4).
Me2Zn turned out to be much less reactive than Et2Zn.
When Me2Zn was used as nucleophile, only 32% yield of
the addition product was obtained after 6 days of reaction
time (Table 1, entry 8). However, a high level of
asymmetric induction was still observed (83% ee with
ligand 2e, entry 8).
Ligands 2 were also evaluated as catalysts for the
addition of Et2Zn to benzaldehyde (Scheme 5). In a
typical experiment, Et2Zn (2.2 equiv) was added to the
solution of the chiral ligand 2 (0.05 equiv) in dry toluene,
under Ar, at 0 °C. After 20 min, PhCHO was added
dropwise, and the reaction was stirred at 0 °C for 24 h.
After hydrolysis with saturated aqueous NH4Cl and
In all cases, when a stoichiometric amount of the chiral
ligand was used, up to 90% of it could be recovered during
(13) The rate of the addition of Et2Zn to 1a promoted by 2d was
estimated to be 2.5 times the rate of the reaction in the absence of the
ligand.
(14) For a review on this subject see: Berrisford, D. J .; Bolm, C.;
Sharpless, K. B. Angew. Chem., Int. Ed. Engl. 1995, 34, 1059.
(15) For previous work on the catalytic addition of R2Zn to N-
(diphenylphosphinoyl) imines, see: (a) Soai, K.; Hatanaka, T.; Miyaza-
wa, T. J . Chem. Soc., Chem. Commun. 1992, 1097. (b) Reference 7.

Enantioselective Addition of Dialkylzinc Reagents
J . Org. Chem., Vol. 63, No. 8, 1998 2533
Ta ble 2. Ad d ition Rea ction of Et2Zn to 9 Ca ta lyzed by
th e Liga n d s 2
favor the formation of the other enantiomer (entries 6
and 7) as R2 becomes larger.
entry
ligand
yield (%)a
ee (%)b
configc
Ligands 2b-e showed to be less catalytically active
than 2a , 2f and 2g. After 24 h at 0 °C, a low conversion
was achieved when those ligands were employed as
catalysts. It was necessary to stir the reactions for an
additional 24 h at room temperature in order to achieve
full conversion of the starting material.
1
2
3
4
5
6
7
2a
62
59
63
59
47
81
46
24
48
53
75
24
28
49
R
S
S
S
S
R
R
2bd
2cd
2d d
2ed
2f
2ge
a
Con clu sion s
Isolated yield after flash chromatography (silica gel, pentane/
Et2O). b Determined by HPLC analysis on a chiral column (Chiral-
Cel OD-H). c Determined by comparison of the optical rotation with
In this paper, we have shown that 2-azanorbornyl-
methanols 2 can effectively be used as promoters for the
enantioselective addition of dialkylzinc reagents to N-
(diphenylphosphinoyl) imines 1. High enantioselectivi-
ties were obtained in both the stoichiometric and the
catalytic processes. It should be noted that these type
of ligands can be further refined, i.e. by the introduction
of only one R2 group on the hydroxymethyl carbon, thus
providing an additional chiral center. These studies are
currently under investigation.
d
that given in the literature (see Experimental Section). The
reaction was run for 1 day at 0 °C and one more day at room
temperature. e Values given according to the data reported in the
literature (ref 5b).
Exp er im en ta l Section
For general experimental information, see ref 16. Unless
otherwise noted, final product solutions were dried over
MgSO4, filtered, and evaporated. Flash chromatography was
performed on silica gel (Matrex 60A, 37-70 µm). When
mentioned, deactivated silica gel means that it was treated
with 5% Et3N in pentane, and the column was eluted with
the same solvent mixture until the coming eluent was basic
according to pH paper. TLC analysis was performed on
precoated TLC plates, SIL G-60 UV254, which were purchased
from Macherey-Nagel. When mentioned, deactivated silica gel
means that the TLC plate was eluted with 5% Et3N in pentane
and dried before applying the sample. Unless otherwise
mentioned, [R] values were measured in CHCl3. HPLC
analysis were carried out on a chiral column (ChiralCel OD-
H), using a 254 nm UV detector and a flow rate of 0.5 mL/
min. 1H and 13C NMR spectra were recorded at 400 and 100.4
MHz, respectively. Et2Zn and Me2Zn were purchased from
Aldrich Co. Imines 1 were prepared according to a literature
procedure.12 PhCHO was purchased from Aldrich Co. and
distilled before use.
E t h yl (1S,3R,4R)-2-[(S)-1-P h en ylet h yla m in o]-2-a za -
bicyclo[2.2.1]h ep t-5-en e-3-ca r boxyla te (5). Compound 5
was prepared via an aza-Diels-Alder reaction between cyclo-
pentadiene and an iminium ion derived from ethyl glyoxylate
and (S)-1-phenylethylamine, following a literature procedure.10
Purification by flash chromatography (deactivated silica gel,
pentane/Et2O: 95/5 to 80/20) gave 60% yield. All the physical
and spectroscopic data for compound 5 were in complete
agreement with the reported data for its enantiomer,10 except
for the sign of the optical rotation.
F igu r e 3.
extractive workup, the expected 1-phenyl-1-propanol was
isolated with the yields and ee’s shown in Table 2.
As it was the case in the addition to imines 1, ligand
2d showed the highest level of chiral induction, giving
an ee of 75%. The enantioselectivity dropped upon
decreasing the steric bulk of the substituent at nitrogen
(Table 2, entries 2-4). A change in the absolute config-
uration of the product was observed when R1 ) Me (entry
1). These results could be rationalized assuming that A
and B (Figure 3) are two possible transition states
involved in the addition reaction. When R1 is a large
group, complex A is favored over B due to the steric
interaction between R1 and the H of the aldehyde. As a
result, the S enantiomer of the addition product is
preferably formed. When the size of R1 is reduced, the
participation of B becomes important and, consequently,
the ee is lowered (compare entries 2-4). In the case of
R1 ) Me, it seems that B begins to be preferred over A,
therefore changing the configuration of the major enan-
tiomer of the addition product (entry 1).
Eth yl (1S,3R,4R)-2-Aza bicyclo[2.2.1]h ep ta n e-3-ca r box-
yla te (6). A solution of the Diels-Alder adduct 5 (7.869 g,
29.0 mmol) in absolute ethanol (50 mL) was stirred under a
hydrogen pressure of 100 psi at room temperature for 48 h in
the presence of 5% Pd-C (1.574 g, 20 wt %). Pd-C was
removed by filtration through Celite and evaporation of the
solvent yielded 4.809 g (98%) of the pure NH amino ester 6:
Once more, the preference for the S enantiomer was
diminished when increasing the steric bulk of the R2
group of the ligand (Table 2, entries 4-7). This could be
explained assuming that, as the size of R2 is increased,
steric interactions between one of the R2 groups and the
Et2ZnB moiety would force ZnB to change its position from
below to above the five-membered ring, provoking thus
a change in the configuration of ZnA (complex C, Figure
3). The competition between complexes A and C would
make the ee to drop (compare entries 4 and 5) and even
Rf 0.53 (pentane/acetone: 1/1; deactivated silica gel); [R]24
)
D
-28.2 (c ) 1.26); IR (neat, cm-1) 3310, 1728 and 1206; 1H NMR
δ 1.21 (1 Η, br d, J ) 9.8 Hz), 1.25 (3 H, t, J ) 7.1 Hz), 1.34-
1.66 (5 H, m), 2.18 (1 H, br s), 2.59, 3.27, 3.50 (1 H each, 3 br
s), 4.15 (2 H, q, J ) 7.1 Hz); 13C NMR δ 14.2, 28.4, 31.1, 35.7,
41.7, 56.2, 61.0, 63.6, 174.5; MS (EI) m/z (rel intensity) 169
(M+, 8%), 131 (26), 73 (23), 69 (100), 57 (25), 55 (40). Anal.
(16) Bedekar, A. V.; Koroleva, E. B.; Andersson, P. G. J . Org. Chem.
1997, 62, 2518.

2534 J . Org. Chem., Vol. 63, No. 8, 1998
Guijarro et al.
Calcd for C9H15NO2‚0.1H2O:17 C, 63.21; H, 8.96; N, 8.19.
Found: C, 62.98; H, 9.06; N, 8.11.
resultant mixture was extracted with ethyl acetate (3 × 20
mL). The combined organic layers were washed with brine
and dried. Flash chromatography (pentane/Et2O: 99/1 to 95/
5) gave 407 mg (78%) of the N-benzyl ester 8: Rf 0.53
(pentane/Et2O: 1/1); [R]25 ) -0.9 (c ) 1.06), [R]25 ) -8.1
The NH ester 6 was N-benzoylated by reaction with benzoyl
chloride in the presence of Et3N. Analysis by HPLC, using
20% i-PrOH in hexane as eluent and a flow rate of 0.4 mL/
min, gave an optical purity of 98%. The retention times for
the two enantiomers of the benzamide were 15.0 (minor) and
22.9 min (major).
Eth yl (1S,3R,4R)-2-Meth yl-2-a za bicyclo[2.2.1]h ep ta n e-
3-ca r boxyla te (7a ). The solution of the ester 6 (339 mg, 2.0
mmol) and MeI (0.21 mL, 3.4 mmol) in MeCN (10 mL) was
refluxed under Ar for 19 h. Solvent was evaporated, and the
resulting residue was dissolved in CH2Cl2 and washed with
saturated aqueous NaHCO3. The aqueous phase was ex-
tracted twice with CH2Cl2 and the combined organic layers
were dried. Removal of the solvent by rotary evaporation gave
D
365
(c ) 1.06); IR (neat, cm-1) 1742, 1155; 1H NMR δ 1.13 (3 H, t,
J ) 7.1 Hz), 1.24 (1 H, dt, J ) 9.4, 1.4 Hz), 1.30-1.44, 1.61-
1.71, 1.92-2.05 (2, 1 and 2 H, respectively, 3 m), 2.50-2.54 (1
H, m), 2.67 (1 H, s), 3.31-3.33 (1 H, m), 3.72, 3.76 (1H each,
2 d, J ) 12.9 Hz each), 4.00 (2 H, q, J ) 7.1 Hz), 7.18-7.37 (5
H, m);13C NMR δ 14.2, 22.4, 29.3, 36.6, 42.4, 55.5, 59.6, 60.2,
69.9, 126.8, 128.1, 129.0, 139.4, 173.5; MS (EI) m/z (rel
intensity) 259 (M+, <1%), 186 (39), 158 (32), 91 (100), 65 (21).
Anal. Calcd for C16H21NO2: C, 74.10; H, 8.16; N, 5.40.
Found: C, 73.89; H, 8.17; N, 5.37.
Red u ction of Ester s 7 a n d 8. Gen er a l P r oced u r e. To
a stirred suspension of LiAlH4 (111 mg, 2.8 mmol) in dry THF
(6 mL), under Ar, cooled to 0 °C, the solution of the corre-
sponding amino ester 7-8 (1.4 mmol) in dry THF (3 mL) was
added dropwise during ca. 10 min. The reaction mixture was
stirred for 2 h at 0 °C and then quenched following a literature
procedure.18
a
residue that was purified by flash chromatography
(pentane/acetone: 4/1), to afford 123 mg (34%) of the N-Me
ester 7a : Rf 0.54 (pentane/acetone: 1/1); [R]25 ) +13.2 (c )
D
1
1.21, CH2Cl2); IR (neat, cm-1) 1740, 1179; H NMR δ 1.25 (3
Η, t, J ) 7.1 Hz), 1.22-1.35 (3 H, m), 1.56-1.67, 1.78-1.92 (1
and 2 H, respectively, 2 m), 2.39 (3 H, s), 2.44 (1 H, s), 2.53-
2.56 (1 H, m), 3.30 (1 H, br s), 4.09-4.21 (2 H, m); 13C NMR δ
14.3, 21.6, 29.2, 37.2, 37.8, 42.4, 60.5, 62.2, 71.3, 173.4; MS
(EI) m/z (rel intensity) 183 (M+, 2%), 131 (24), 73 (21), 69 (100),
57 (21), 55 (36). Anal. Calcd for C10H17NO2: C, 65.54; H, 9.35;
N, 7.64. Found: C, 65.26; H, 9.35; N, 7.55.
(1 S ,3 R ,4 R )-3 -(H y d r o x y m e t h y l )-2 -m e t h y l -2 -a z a -
bicyclo[2.2.1]h ep ta n e (2a ). Evaporation of the solvent
yielded 150 mg (76%) of pure 2a : Rf 0.08 (CHCl3/MeOH: 95/
5); [R]25 ) -27.7 (c ) 1.03); IR (neat, cm-1) 3384; 1H NMR δ
D
1.19-1.36, 1.54-1.64, 1.71-1.76, 1.86-1.98 (3, 1, 1, and 2 H,
respectively, 4 m), 2.13-2.16 (1 H, m), 2.35 (3 H, s), 3.19 (1 H,
br s), 3.32 (1 H, br s), 3.36 (1 H, dd, J ) 10.8, 5.4 Hz), 3.47 (1
H, dd, J ) 10.8, 5.8 Hz); 13C NMR δ 21.5, 29.4, 36.6, 37.7,
40.9, 62.1, 64.5, 70.5; MS (EI) m/z (rel intensity) 141 (M+, 2%),
83 (26), 71 (25), 69 (100), 57 (32), 55 (59). Anal. Calcd for
C8H15NO: C, 68.04; H, 10.71; N, 9.92. Found: C, 68.38; H,
10.78; N, 9.58.
Alk yla tion of th e NH Ester 6. P r ep a r a tion of P r od -
u cts 7b,c. Gen er a l P r oced u r e. The NH amino ester 6 (339
mg, 2.0 mmol), the corresponding alkyl bromide (10.0 mmol),
and 18-crown-6 (53 mg, 0.2 mmol) were dissolved in dry MeCN
(10 mL). Anhydrous K2CO3 (335 mg, 2.4 mmol) was added,
and the reaction mixture was refluxed for 24 h. Solvent was
evaporated, water was added, and the resultant mixture was
extracted with ethyl acetate (3 × 20 mL). The combined
organic layers were washed with brine and dried.
Eth yl (1S,3R,4R)-2-Eth yl-2-a za bicyclo[2.2.1]h ep ta n e-3-
ca r boxyla te (7b). Flash chromatography (pentane/acetone:
95/5 to 6/1) yielded 317 mg (80%) of the N-ethyl ester 7b: Rf
0.25 (pentane/acetone: 4/1); [R]25D ) -11.7 (c ) 1.08); IR (neat,
cm-1) 1748, 1174; 1H NMR δ 1.01 (3 H, t, J ) 7.3 Hz), 1.23 (3
H, t, J ) 7.1 Hz), 1.21-1.34, 1.56-1.67, 1.79-1.89 (3, 1 and 2
H, respectively, 3 m), 2.47-2.50 (1 H, m), 2.50 (1 H, br s), 2.56
(2 H, q, J ) 7.3 Hz), 3.43 (1 H, br s), 4.14 (2 H, q, J ) 7.1 Hz);
13C NMR δ 14.21, 14.22, 22.1, 29.2, 36.5, 42.6, 45.8, 60.3, 60.5,
70.3, 174.0; MS (EI) m/z (rel intensity) 197 (M+, 6%), 131 (34),
124 (51), 96 (65), 69 (100), 68 (36). Anal. Calcd for
( 1 S ,3 R ,4 R ) -2 -E t h y l -3 -( h y d r o x y m e t h y l ) -2 -a z a -
bicyclo[2.2.1]h ep ta n e (2b). Purification by flash chroma-
tography (deactivated silica gel, pentane/acetone/MeOH: 100/
100/5 and then acetone MeOH: 9/1) afforded 130 mg (60%) of
the expected product 2b: Rf 0.22 (acetone/MeOH: 95/5;
deactivated silica gel); [R]25D ) -14.4 (c ) 1.00); IR (neat, cm-1
)
1
3384; H NMR δ 1.06 (3 H, t, J ) 7.2 Hz), 1.19-1.34, 1.54-
1.62, 1.71-1.76, 1.81-1.89 (3, 1, 1 and 1 H, respectively, 4
m), 2.02-2.06, 2.14-2.17 (1 H each, 2 m), 2.52, 2.65 (1H each,
2 dq, J ) 12.0, 7.2 Hz each), 3.35 (1 H, dd, J ) 10.7, 4.5 Hz),
3.35-3.38 (2 H, m), 3.44 (1 H, dd, J ) 10.7, 6.3 Hz); 13C NMR
δ 14.4, 21.7, 29.2, 36.0, 41.0, 45.1, 58.6, 64.4, 69.8; MS (EI)
m/z (rel intensity) 155 (M+, 11%), 124 (35), 96 (100), 95 (27),
69 (62), 68 (59), 67(29). Anal. Calcd for C9H17NO‚H2O:17 C,
62.39; H, 11.05; N, 8.08. Found: C, 62.67; H, 10.99; N, 8.09.
(1S ,3R ,4R )-3-(H y d r o x y m e t h y l)-2-is o p r o p y l-2-a za -
bicyclo[2.2.1]h ep ta n e (2c). The residue obtained after
evaporation of the solvent was chromatographed (deactivated
silica gel, pentane/acetone: 1/1 and then pentane/
acetone/MeOH: 100/100/5 to 0/100/5) and 190 mg (80%) of 2c
C
11H19NO2: C, 66.97; H, 9.71; N, 7.10. Found: C, 66.85; H,
9.76; N, 7.04.
E t h yl
(1S,3R,4R)-2-Isop r op yl-2-a za b icyclo[2.2.1]-
h ep ta n e-3-ca r boxyla te (7c). NaI (600 mg, 4.0 mmol) was
added to the reaction mixture before refluxing it, to facilitate
the substitution reaction by in situ generation of i-PrI.
Purification by flash chromatography (pentane/acetone: 99/1
to 9/1) afforded 346 mg (82%) of the N-isopropyl ester 7c: Rf
0.37 (pentane/acetone: 4/1); [R]25 ) +3.0 (c ) 1.09), [R]25
were isolated: Rf 0.34 (pentane/acetone: 1/1; deactivated silica
1
gel); [R]25 ) -24.6 (c ) 1.02); IR (neat, cm-1) 3356; H NMR
D
D
302
) -36.0 (c ) 1.09); IR (neat, cm-1) 1750, 1174; 1H NMR δ 0.97,
1.09 (3 H, each, 2 d, J ) 6.2 Hz each), 1.25 (3 H, t, J ) 7.1
Hz), 1.22-1.38, 1.59-1.69, 1.78-1.97 (3, 1 and 2 H, respec-
tively, 3 m), 2.42-2.45 (1 H, m), 2.55 (1 H, septet, J ) 6.2
Hz), 2.63 (1 H, s), 3.60-3.63 (1 H, m), 4.17 (2 H, q, J ) 7.1
Hz); 13C NMR δ 14.2, 21.8, 22.2, 22.8, 29.6, 36.1, 43.2, 50.8,
58.2, 60.4, 70.5, 174.5; MS (EI) m/z (rel intensity) 211 (M+,
4%), 138 (86), 131 (31), 110 (89), 69 (100), 55 (28). Anal. Calcd
for C12H21NO2: C, 68.21; H, 10.02; N, 6.63. Found: C, 68.08;
H, 10.08; N, 6.60.
δ 1.06, 1.09 (3H each, 2 d, J ) 6.3 Hz each), 1.19-1.37, 1.56-
1.65, 1.79-1.91 (3, 1 and 2 H, respectively, 3 m), 2.21-2.25 (2
H, m), 2.66 (1 H, septet, J ) 6.3 Hz), 3.39 (1 H, dd, J ) 10.6,
3.6 Hz), 3.43 (1 H, dd, J ) 10.6, 6.6 Hz), 3.52 (1 H, br s), 3.92
(1 H, br s); 13C NMR δ 22.2, 22.5, 23.1, 28.6, 35.6, 42.0, 50.6,
58.7, 64.9, 68.8; MS (EI) m/z (rel intensity) 169 (M+, 2%), 138
(88), 110 (100), 96 (25), 68 (61), 67 (34). Anal. Calcd for
C
10H19NO‚0.6H2O:17 C, 66.70; H, 11.31; N, 7.78. Found: C,
67.10; H, 11.27; N, 7.41.
(1 S ,3 R ,4 R )-2 -B e n z y l -3 -(h y d r o x y m e t h y l )-2 -a z a -
b icyclo[2.2.1]h ep t a n e
(2d ).
Flash
chromatography
Eth yl (1S,3R,4R)-2-Ben zyl-2-a za bicyclo[2.2.1]h ep ta n e-
3-ca r boxyla te (8). Anhydrous K2CO3 (335 mg, 2.4 mmol) was
added to the solution of 6 (339 mg, 2.0 mmol) and benzyl
bromide (0.29 mL, 2.4 mmol) in MeCN (10 mL) at room
temperature, and the reaction mixture was stirred for 32 h.
Then, solvent was evaporated, water was added, and the
(pentane/acetone: 4/1) afforded 225 mg (74%) of the amino
alcohol 2d : Rf 0.20 (pentane/acetone: 1/1); [R]25 ) -0.6 (c )
D
1.06), [R]25302 ) -80.7 (c ) 1.06); IR (neat, cm-1) 3384; 1H NMR
δ 1.17 (1 H, br d, J ) 9.6 Hz), 1.25-1.37, 1.56-1.66, 1.74-
1.80, 1.95-2.05 (2, 1, 1 and 1 H, respectively, 4 m), 2.17-2.22
(17) The product was very hygroscopic.
(18) Micovic, V. M.; Mihailovic, M. L. J . Org. Chem. 1953, 18, 1190.

Enantioselective Addition of Dialkylzinc Reagents
J . Org. Chem., Vol. 63, No. 8, 1998 2535
(2 H, m), 2.50 (1 H, br s), 3.21 (1 H, br s), 3.24 (1 H, dd, J )
10.7, 3.9 Hz), 3.29 (1 H, dd, J ) 10.7, 5.6 Hz), 3.67, 3.71 (1H
each, 2 d, J ) 13.1 Hz each), 7.20-7.39 (5 H, m); 13C NMR δ
22.1, 29.6, 36.3, 41.7, 54.9, 59.1, 64.1, 69.0, 127.0, 128.3, 128.8,
139.9; MS (EI) m/z (rel intensity) 217 (M+, <1%), 186 (50), 158
(39), 91 (100), 65 (21). Anal. Calcd for C14H19NO‚0.3H2O:17
C, 75.50; H, 8.87; N, 6.29. Found: C, 75.50; H, 8.83; N, 6.33.
Ad d ition of Gr ign a r d Rea gen ts to 8. P r ep a r a tion of
Am in o Alcoh ols 2e-g. Gen er a l P r oced u r e. The solution
of the amino ester 8 (1.5 mmol) in dry THF, under Ar, was
cooled to -30 °C, and the solution of the corresponding
Grignard reagent (3.75 mmol) in dry Et2O (4 mL) was added
dropwise. The reaction was stirred for 7 h, allowing the
temperature to rise to room temperature. Then, water was
added, and the mixture was extracted with ethyl acetate (3 ×
20 mL). After drying and evaporation of the solvent, the
residue was purified by flash chromatography.
product 2g. All the physical and spectroscopic data of 2g were
in complete agreement with the data reported for its
enantiomer,5b except for the sign of the optical rotation.
Ad d ition Rea ction of Dia lk ylzin c Rea gen ts to 1 P r o-
m oted by th e Liga n d s 2. Gen er a l P r oced u r e. The phos-
phinoyl imine 1 (76 mg, 0.25 mmol) and the amino alcohol 2
(0.25 mmol) were dissolved in dry toluene (1.5 mL) under Ar,
and the mixture was stirred for 10 min at room temperature.
The solution was cooled to 0 °C, R2Zn (0.75 mL 1.0 M solution
in hexane, 0.75 mmol) was added dropwise, and the reaction
was stirred, allowing the temperature to rise slowly to room
temperature (3-4 h) and then for 2 days. The reaction was
quenched with saturated aqueous NH4Cl (10 mL), and the
mixture was extracted with CH2Cl2 (4 × 10 mL) and dried.
Solvents were evaporated, and the residue was purified by
flash chromatography (pentane/acetone: 6/1 to 1/1). The
resultant solid was analyzed by HPLC, using 10% i-PrOH in
hexane as eluent. The retention times are: 3a :7b 13.6 (R) and
17.7 min (S); 3a ′:7b 14.7 (R) and 18.6 min (S); 3b:7b 16.3 (R)
and 19.8 min (S). Yields and ee’s are given in Table 1. The
absolute configuration of the major isomer was determined by
hydrolysis of the reaction product7b and comparison of the
optical rotation of the obtained amine with the reported data.7b
Ad d ition Rea ction of Et2Zn to P h CHO Ca ta lyzed by
th e Liga n d s 2. Gen er a l P r oced u r e. Et2Zn (2.2 mL 1.0 M
solution in hexane, 2.2 mmol) was added dropwise to the
solution of the ligand 2 (0.05 mmol) in dry toluene (1.5 mL),
under Ar, at 0 °C. After 20 min, PhCHO (0.10 mL, 1.0 mmol)
was added dropwise, and the resulting solution was stirred
for 1 day at 0 °C. The reaction was quenched with saturated
aqueous NH4Cl (10 mL), and the mixture was extracted with
Et2O (3 × 20 mL). The combined organic extracts were washed
with water (10 mL) and brine (10 mL) and dried. Evaporation
of the solvents and flash chromatography (pentane/Et2O: 9/1)
afforded the expected product 10, that was analyzed by HPLC,
using 5% i-PrOH in hexane as eluent. The retention times
are 15.6 (R) and 17.7 min (S). Yields and ee’s are given in
Table 2. The absolute configuration of the major isomer was
determined by comparison of the optical rotation with the
reported data.19
(1S ,3R ,4R )-2-Be n zyl-3-(2-h yd r oxy-2-p r op yl)-2-a za -
bicyclo[2.2.1]h ep ta n e (2e). Flash chromatography (pentane/
acetone: 95/5) gave 311 mg (84%) of product 2e: Rf 0.40
(pentane/acetone: 4/1); mp ) 48-50 °C; [R]25 ) +31.5 (c )
D
1.09); IR (KBr, cm-1) 3482; 1H NMR δ 1.07 (1 H, br d, J ) 9.5
Hz), 1.21, 1.23 (3 H each, 2 s), 1.19-1.30, 1.57-1.66, 1.83-
1.89, 2.01-2.09 (2, 1, 1 and 1 H, respectively, 4 m), 2.01 (1 H,
s), 2.38-2.42 (1 H, m), 3.12 (1 H, br s), 3.18 (1 H, br s), 3.70,
4.02 (1H each, 2 d, J ) 14.0 Hz each), 7.22-7.38 (5 H, m); 13
C
NMR δ 22.1, 26.5, 29.8, 30.3, 36.0, 39.5, 55.5, 58.0, 71.2, 76.3,
126.8, 128.3 (4 C), 140.0; MS (EI) m/z (rel intensity) 230 (M+
- 15, 2%), 186 (56), 158 (45), 91 (100), 65 (18). Anal. Calcd
for C16H23NO: C, 78.32; H, 9.45; N, 5.71. Found: C, 78.34;
H, 9.42; N, 5.69.
(1S,3R,4R)-2-Ben zyl-3-(3-h yd r oxy-2,4-d im eth yl-3-p en -
tyl)-2-a za bicyclo[2.2.1]h ep ta n e (2f). Flash chromatography
(pentane/acetone: 99/1 to 95/5) afforded 90 mg (20%) of the
amino alcohol 2f: Rf 0.66 (pentane/Et2O: 1/1); mp ) 73-74
°C; [R]25 ) +42.5 (c ) 1.00); IR (KBr, cm-1) 3448; 1H NMR δ
D
1.021, 1.022, 1.04, 1.12 (3 H each, 4 d, J ) 7.1 Hz each), 1.04-
1.08, 1.16-1.32, 1.58-1.66, 1.92-2.11 (1, 2, 1 and 2 H,
respectively, 4 m), 2.02 (1 H, septet, J ) 7.1 Hz), 2.16 (1 H,
septet, J ) 7.1 Hz), 2.44 (1 H, br s), 2.47-2.50 (1 H, m), 3.12
(1 H, br s), 3.14 (1 H, br s), 3.64, 4.03 (1H each, 2 d, J ) 14.1
Hz each) and 7.21-7.36 (5 H, m); 13C NMR δ 18.4, 18.8, 19.1,
19.6, 22.3, 29.7, 32.0, 35.0, 35.7, 40.3, 55.3, 57.7, 70.7, 77.2,
126.7, 128.2, 128.3 and 140.1; MS (EI) m/z (rel intensity) 258
(M+ - 43, 13%), 186 (92), 158 (40), 91 (100) and 65 (13). Anal.
Calcd for C20H31NO: C, 79.68; H, 10.36; N, 4.65. Found: C,
79.57; H, 10.39; N, 4.67.
Ack n ow led gm en t. We thank the Wenner-Gren
Foundation (postdoctoral stipend to D. G.), the Swedish
Natural Science Research Council, and the Foundation
for Strategic Research (SSF) for financial support.
J O9718096
(1S ,3R ,4R )-2-Be n zyl-3-(h yd r oxyd ip h e n ylm e t h yl)-2-
a za b icyclo[2.2.1]h ep t a n e (2g). Flash chromatography
(pentane/Et2O: 95/5) yielded 261 mg (47%) of the expected
(19) Pickard, R. H.; Kenyon, J . J . Chem. Soc. 1914, 1115.