Diastereoselective aldolization of α-aminonitriles. Diastereoselective synthesis of β-amino alcohols and β,γ-diamino alcohols

Article abstract of DOI:10.1021/jo025872t

Aldolization performed by addition of lithiated N-benzyl-N-tert-butylaminoacetonitrile to aldehydes provides diastereomerically pure anti-β-hydroxy-α-aminonitriles. They are transformed into syn,anti-protected β,γ-diamino alcohols by a two-step procedure, involving addition of a Grignard reagent and reduction. The cleavage of the N-tert-butyl group is achieved by a simple acidic treatment. The application of this methodology to chiral, nonracemic aldehydes is studied. Starting from D-isopropylideneglyceraldehyde, an anti, anti, syn, anti-(2R,3S,4S,5R,6R)-diaminotriol is prepared in acceptable yield and with a good level of diastereoselectivity.

Full text of DOI:10.1021/jo025872t

Dia ster eoselective Ald oliza tion of r-Am in on itr iles.
Dia ster eoselective Syn th esis of â-Am in o Alcoh ols a n d â,γ-Dia m in o
Alcoh ols
Eric Leclerc, Emmanuel Vrancken, and Pierre Mangeney*
Laboratoire de chimie des organoe´le´ments, UMR 7611, Universite´ Pierre et Marie Curie 4, place J ussieu,
tr. 44-45 2e`me e´t., 75252 Paris Cedex 05, France
mangeney@ccr.jussieu.fr
Received April 29, 2002
Aldolization performed by addition of lithiated N-benzyl-N-tert-butylaminoacetonitrile to aldehydes
provides diastereomerically pure anti-â-hydroxy-R-aminonitriles. They are transformed into syn,-
anti-protected â,γ-diamino alcohols by a two-step procedure, involving addition of a Grignard reagent
and reduction. The cleavage of the N-tert-butyl group is achieved by a simple acidic treatment.
The application of this methodology to chiral, nonracemic aldehydes is studied. Starting from
D-isopropylideneglyceraldehyde, an anti, anti, syn, anti-(2R,3S,4S,5R,6R)-diaminotriol is prepared
in acceptable yield and with a good level of diastereoselectivity.
SCHEME 1
In tr od u ction
R-Aminonitriles 1 are very popular bifunctional syn-
thons that have found numerous synthetic applications.1
It was shown in the late 1950s that such compounds, if
they bear an R-hydrogen, are easily deprotonated in the
R-position by strong bases, leading to metalated R-ami-
nonitriles 1, which may undergo various and efficient
nucleophilic reactions.2 Among these reactions, the ad-
dition to aldehydes is documented but has been mainly
used for the preparation of compounds where the nitrile
functionality is removed after reaction via either a
reductive decyanation, leading to amines, or a retro-
Strecker reaction, leading to carbonyl compounds.1,3 The
metalated R-aminonitriles act in this case either as
masked acyl anions or as R-aminocarbanions. Because
of this, little is known about the stereochemical course
of the aldol-type reaction performed by the addition of
these organometallic reagents to aldehydes. This lack of
information contrasts strongly with the number of results
described with metalated arylacetonitriles.4 The synthetic
utility of such a reaction may provide access to function-
alized aldols 2 with the possible control of the relative
and absolute configuration of two stereocenters. The
preparation of these aldols, which are precursors of
â-hydroxy-R-amino acids, an important class of amino
acids, is generally performed by a diastereoselective
Strecker-type reaction.5 In this way, syn-aldols are
obtained as the major products. We were interested in
the development of a diastereo- and enantioselective
access to â,γ-diamino alcohols 36 that may provide access
to all the possible diastereomers. These stereotriads
might be obtained from the aldols 2 by a sequence of
reactions involving the addition of an organometallic
reagent to the nitrile group followed by an in situ
reduction of the intermediate imine,7 as shown in Scheme
1. We report herein the synthesis of compounds of type
3, which resulted in the development of a diastereose-
* Corresponding author. Tel: (33) 1 44275567. Fax: (33) 1 44277567.
(1) Enders, D.; Shilvock, J . P. Chem. Soc. Rev. 2000, 29, 359-373.
(2) (a) Hauser, C. R.; Taylor, H. M.; Ledford, T. G. J . Am. Chem.
Soc. 1959, 82, 1786-1789. (b) Taylor, H. M.; Hauser, C. R. J . Am.
Chem. Soc. 1959, 82, 1790-1792. (c) Taylor, H. M.; Hauser, C. R. J .
Am. Chem. Soc. 1959, 82, 1960-1965
(3) (a) Stork, G., J acobson, R. M., Levitz, R Tetrahedron Lett. 1979,
20, 771-774. (b) Husson, H.-P, Royer, J . Chem. Soc. Rev. 1999, 28,
383. (c) Enders, D.; Lotter, H. Tetrahedron Lett. 1982, 23, 639-642.
(4) (a) Carlier, P. R.; Moon Lo, K. J . Org. Chem. 1994, 59, 4053-
4055. (b) Carlier, P. R.; Moon Lo, K.; Lo, M. M.-C.; Williams, I. D. J .
Org. Chem. 1995, 60, 7511-7517. (c) Carlier, P. R.; Moon Lo, K.; Lo,
M. M.-C.; Lo, P. C.-K.; Lo, C. W.-S. J . Org. Chem. 1997, 62, 6316-
6321. Mecozzi, T., Petrini, M., Profeta, R. J . Org. Chem. 2001, 66,
8264-8267. Carlier, P. R., Madura, J . D. J . Org. Chem. 2002, 67,
3832-3840.
(5) (a) Zandbergen, P., Brussee, J . Van Der Gen, A.Tetrahedron:
Asymmetry 1992, 3, 769-774. (b) Cainelli, G., Giacomini, D., Trere´,
A, Galletti, P. Tetrahedron: Asymmetry 1995, 6, 1593-1600. (c)
Chakraborty, T. K., J ayaprakash, S. Tetrahedron Lett. 1997, 38, 8899-
8902. (d) Badorrey, R., Cativiela, C., Diaz-de-Villegas, M. D., Galvez,
J . A. A. Tetrahedron: Asymmetry 2000, 11, 1015-1025. Davis, F. A.,
Srirajan, V., Fanelli, D. L., Portonovo, P. J . Org. Chem. 2000, 65, 7663-
7666.
(6) (a) Cromwell, N.H., Tsou, K. C. J . Org. Chem. 1950, 15, 1219-
1223. (b) Klein, J . L., Combret, J . C. Bull. Soc. Chim. Fr. 1983, 28-
32.
(7) Brussee, J .; Dofferhoff, F.; Kruse, C. G.; Van Der Gen, A.
Tetrahedron 1990, 46, 1653-1658.
10.1021/jo025872t CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/14/2002
8928
J . Org. Chem. 2002, 67, 8928-8937

Diastereoselective Aldolization of R-Aminonitriles
SCHEME 2
TABLE 1. Ad d ition of Meta la ted
substituents on R-aminonitriles 6-9 was examined via
addition of the lithiated derivatives to cyclohexylcarbox-
aldehyde (entries 1-4, Table 2).
Diben zyla m in oa ceton itr ile 4 to Ald eh yd es
entry
R
M
solvent product anti/ syna yield (%)b
The dimethylpyrrole derivative 6, utilized by Un-
eyama,11 and four other R-aminonitriles (7, 8, 9a , and
9b) bearing bulky N-substituents were prepared as
shown in Scheme 3. It is noteworthy that the tert-butyl
derivatives 9a ,b were the most easily obtained of all these
bulky substrates. The aldolization performed with the
lithiated aminonitrile 6 in THF occurred in good yield
but with poor selectivity (entry 1). This result points out
the importance of the presence of a sp3 nitrogen in the
aminonitrile, probably due to steric factors. The use of
Et2O as a solvent slightly increased the diastereoselec-
tivity, but led to the formation of a byproduct resulting
from the addition of LDA to the nitrile, thus lowering
the yield. Aminonitriles 7 and 8 gave the corresponding
aldols in poor yields (entries 2 and 3), with a large
amount of the starting material being recovered. The
1
2
3
4
5
6
7
Ph
Mes
crotyl Li
n-Bu Li
c-Hex Li
Li
Li
THF
THF
THF
THF
THF
THF
THF+
TMEDA
Et2O
5a
5b
5c
5d
5e
5f
50/50
63/37
71/29
70/30
80/20
91/09
80/20
75
68
70
70
65
65
70
t-Bu
c-Hex Li
Li
5e
8
9
10
c-Hex Li
5e
5e
5e
80/20
80/20
80/20
50
67
37
c-Hex MgBrc THF
c-Hex ZnBrc THF
Determined by 1H NMR of the crude mixture. Reported
yields are weight recoveries of the combined syn- and anti-
stereoisomers. The byproduct results from the addition of LDA to
4 to give the corresponding amide. c Two hours at -40 °C and then
0 °C.
a
b
lective aminoacetonitrile aldol reaction and the diaste-
reoselective synthesis of some syn,anti-â,γ-diamino alco-
hols.
1
selectivity appeared to be high by H NMR (see note c,
Table 2), but with such low conversion, one must be
cautious. This lack of efficiency could not be attributed
to anion formation, as complete deprotonation was
confirmed by deuteriolysis. Finally, the best result was
obtained with 9a , which provided the corresponding aldol
in good yield and high selectivity (entry 4). Several
aldehydes were then tested (Table 2, entries 5-8) with
9a , and all of them were found to react cleanly with high
selectivity (in all cases only one diastereomer was
Resu lts a n d Discu ssion
The first part of this study was an examination of the
stereochemical course of the addition of metalated un-
substituted R-aminonitriles 1 to aldehydes. The first
attempts were performed with N,N-dibenzylaminoaceto-
nitrile 48 via deprotonation with LDA at low temperature
and quenching of the resulted anion with several alde-
hydes (Scheme 2, Table 1).
As can be seen in Table 1, the addition of lithiated N,N-
dibenzyl aminonitrile to several aldehydes in THF was
not diastereoselective (entries 1-5). High diastereose-
lectivity was only achieved with the bulky pivalaldehyde
(entry 6). The use of TMEDA as a cosolvent or Et2O
rather than THF (entries 5, 7 and 8) did not change the
diatereoselectivity. Additionally, a lower temperature
(-100 °C) or transmetalation with MgBr2 or ZnBr2
(entries 9 and 10) had no effect. The anti relative
configuration of the major diastereomer was assigned on
the basis of an X-ray analysis of the crystalline aldol
5f.9,10 Last, we have shown that the aldolization was not
under thermodynamic control, since deprotonation of the
pure syn aldol 5f with LDA in THF gave 5f completely
unchanged after several hours of stirring. Since the
diastereoselectivity of the aldol reaction seems to be
sensitive to steric factors, the influence of other nitrogen
1
observed by H NMR of the crude mixture).
All the aldols 10b-f were crystalline compounds and
could be purified by recrystallization. The anti relative
configuration was determined by X-ray analysis of 10c.12
To rule out any π interaction in the transition state, we
have checked that the addition of lithiated 9b to cyclo-
hexyl carboxaldehyde provided selectively the corre-
sponding anti-aldol 10g in good yield (Table 2, entry 9).
As was shown above, the selectivity appears to be
kinetically controlled under our reaction conditions.
Therefore, a flat six-membered transition state4,13 involv-
ing a N-lithiated R-cyano anion14 (Scheme 4) can be
postulated in order to rationalize the anti-selectivity of
the aldol reaction. Obviously, higher anti-selectivity is
expected by increasing the size of the R3 group of the
aldehyde. Indeed, using the R-aminonitrile 4, the best
result was obtained with pivalaldehyde.
More difficult is the rationalization for the increase in
(11) Katagiri, T.; Irie, M.; Uneyama, K. Org. Lett. 2000, 2, 2423-
2427
(8) Guillaume, D., Aitken, D. J ., Husson, H. P. Synlett, 1991, 747-
749.
(12) Crystallographic data for the strustural analysis have been
deposited with the Cambridge Crystallographic Data Center, CCDC
No. 178048.
(13) Ishiguro, M.; Ikeda, N.; Yamamoto, H. J . Org. Chem. 1982, 47,
2225-2227.
(14) Enders, D.; Kirchhoff, J .; Gerdes, P.; Mannes, D.; Raabe, G.;
Runsink, J .; Boche, G.; Marsch, M.; Ahlbrecht, H.; Sommer, H. Eur.
J . Org. Chem. 1998, 63, 3-72. Fleming, F. F., Shook, B. C. Tetrahedron
2002, 58, 1-23.
(9) Crystallographic data for the strustural analysis have been
deposited with the Cambridge Crystallographic Data Center, CCDC
No. 178177.
(10) The vicinal H2-H3 coupling constants in the 1H NMR were
found to be higher for all the minor (syn) diastereomers than the
corresponding ones observed for the major (anti) diastereomers; see
ref 4.
J . Org. Chem, Vol. 67, No. 25, 2002 8929

Leclerc et al.
TABLE 2. In flu en ce of th e Nitr ogen Su bstitu en ts
amino-
nitrile
yield
(%)
entry
1
R
conditions
-80 °C, 1 h
product
anti/ syna
6
c-Hex
10a -1
67/33 in THF
75/25 in Et2O
>95/5c
86b
63b
20c
50c
84
86
81
90
79
2
3
4
5
6
7
8
9
7
8
c-Hex
c-Hex
c-Hex
n-Bu
n-Hex
t-Bu
-80 to 0 °C, 1 h
-80 to 0 °C, 1 h
10a -2
10a -3
10b
10c
10d
10e
10f
10g
>95/5c
9a
9a
9a
9a
9a
9b
-80 to -40 °C, 2 h
-80 to -40 °C, 2 h
-80 to -40 °C, 2 h
-80 to -40 °C, 2 h
-80 to -40 °C, 2 h
-80 to -40 °C, 2 h
>95/5
>95/5
>95/5
>95/5
Ph
c-Hex
>95/5
>95/5
73
a
b
Determined by 1H NMR of the crude mixture. Reported yields are weight recoveries of the combined syn- and anti-stereoisomers.
c Aldol/starting material ratio in the crude mixture estimated by 1H NMR. The anti/syn ratio was also estimated by 1H NMR of this crude
mixture.
SCHEME 3. P r ep a r a tion of Am in on itr iles 6-9
bearing a stereogenic center in the R-position were tested.
Two diastereomers were obtained in a 1/1 ratio with the
homochiral or racemic17 O-TBDMS mandelic aldehyde 11
(entry 1). Poor selectivity was observed with both the
racemic O-TBDMS lactic aldehyde 12 (62/38, entry 2) and
the D-isopropyleneglyceraldehyde 13 (69/31, entry 3). We
postulated that the anti-selectivity obtained with the
achiral aldehydes was preserved. It meant, therefore,
that the aldolization with these chiral aldehydes did not
occur with good facial selectivity.18 In all cases, the two
diastereomers were inseparable by chromatography. But
despite the low selectivity obtained with the protected
glyceraldehyde, it was observed that the pure major
diastereomer 14 could be isolated in reasonable yield
(40%) by simple recrystallization of the crude mixture.
The anti,anti-relationship (2S,3R,4R) of this compound
was determined by X-ray analysis.19
The modest, but significant, selectivity in the formation
of 14 might be rationalized by the addition of the lithiated
ketene imine species to the aldehyde 13 according to a
(16) As suggested by one of our reviewers, one may involve a
transition state where the two N-substituents are roughly in the plane
of the keteniminate, with the t-Bu group exo and the benzyl endo (see
structure below). Such conformation would reduce steric congestion
in the transition state energy, thereby negating the effect of any
ground-state destabilization.
the selectivity observed with the aminoacetonitrile 9a ,b,
where the nitrogen is a stereogenic center subject to fast
equilibration. Indeed, it is very likely that the addition
to the aldehyde occurs from the side of the less bulky
nitrogen substituent, i.e. the benzyl group. Accordingly,
the same selectivity should be obtained using 4 or 9a ,b.
It is known that the presence of a bulky group on the
nitrogen decreases the inversion barrier and leads to a
flattened pyramid.15 Therefore, the global steric influence
of the nitrogen substituents is increased, leading to better
selectivity.16 Such a hypothesis would also account for a
decrease of the aldolization rate (a few minutes at -80
°C with 4 and 2 h at -40 °C with 9a ,b).
(17) These two experiments were run in order to perform a pseudo-
Hoffmann test due to the possible but improbable configurational
stability of lithiated R-aminonitriles. Hoffmann, R.; Lanz, J .; Metter-
nich, R.; Tarara, G.; Hoppe, D. Angew. Chem., Int. Ed. Engl. 1987, 26,
1145-1146.
(18) (a) Similar results were obtained with configurationally stable
allenyl zinc reagents: Poisson, J . F.; Normant, J . F. J . Org. Chem. 2000,
65, 6553-6560. (b) For a review on Cram’s rule see: Mengel, A.; Reiser,
O. Chem. Rev. 1999, 1191-1223.
(19) Crystallographic data for the structural analysis have been
deposited with the Cambridge Crystallographic Data Center, CCDC
No. 178047.
Since complete control of the diastereoselectivity of the
aminoacetonitrile aldol reaction was achieved with 9a ,
we then examined the same reaction with chiral alde-
hydes. As reported in Table 3, three aldehydes (11-13)
(15) (a) Saunders, M.; Yamada, F. J . Am. Chem. Soc. 1963, 85, 1882.
(b) Lehn, J .-M. Nitrogen Inversion. Top. Curr. Chem. 1970, 15, 311-
377.
8930 J . Org. Chem., Vol. 67, No. 25, 2002

Diastereoselective Aldolization of R-Aminonitriles
SCHEME 4
TABLE 3. Ald oliza tion w ith Ch ir a l Ald eh yd es
SCHEME 6
The application of this three-step sequence to our
substrates, which involves the addition of a Grignard
reagent to the nitrile, methanolysis of the resulting
N-metalloimine, and then reduction, was not obvious.
Since addition of RMgX to R-aminonitriles usually results
in the displacement of the nitrile group (Bruylants
reaction),21 our first attempt was to add n-BuLi, instead
of n-BuMgBr, to the amino alcohol 10b (Scheme 7). The
addition was found to occur in THF, but not in Et2O. The
diastereomerically pure (1H NMR) â,γ-diamino alcohol
15a was then obtained, after methanolysis, by reduction
with NaBH4 in the presence of MgBr2. We have observed
that phenyl and allyl Grignard reagents added cleanly
to the nitrile in THF.22 In this case, MgBr2 is not required
to perform the reduction after methanolysis, and the
diamino alcohols 15b,c were obtained in good yield and
as single diastereomers (Scheme 7).
a
b
Determined by 1H NMR of the crude mixture. Reported
yields are weight recoveries of the combined stereoisomers. A small
amount of an undetermined product was formed during the
reaction. c Yield of pure major diastereomer recovered after re-
crystallization.
Application of the latter sequence to the chiral nonra-
cemic aldol 14, using aryl, allyl, or alkyl Grignard
reagents, led to the diamino alcohols 16 a -c in good
isolated yields and with excellent diastereocontrol (Scheme
8).
SCHEME 5
The use of crotylmagnesium bromide led to the adduct
16d , arising from γ-addition, in good yield (79%) and with
high selectivity (dr ) 9/1, Scheme 8). The compounds 16c
and 16d were purified by recrystallization, and the
relative configuration of each major diastereomer was
established by X-ray analysis.23 Syn C4-H,C5-H and anti
C5-H,C6-H relationships were observed. Therefore, the
same syn C4-H,C5-H relative configuration was postu-
lated for compounds 15a -c and 16a -b obtained by the
same alkylation-reduction sequence. The stereodeter-
mining step of Brussee’s procedure is obviously the
model involving a combination of a cyclic transition state
(C2-H, C3-H relationship) and a Felkin type addition
(C3-H, C4-H relationship, Scheme 5).20
Having examined the scope and the limitations of the
aldol reaction, we next explored the possibility of using
the sequence of alkylmetalation-reduction of the nitrile
group, reported by Brussee (Scheme 6)7 for the diaste-
reoselective access to â,γ-diamino alcohols.
(21) (a) Bruylants, P. Bull. Soc. Chim. Belg. 1924, 33, 467. (b)
Ahlbrecht, H.; Dollinger, H. Synthesis 1985, 743-748.
(22) The addition of Grignard reagents to the nitrile is surprising
in the light of the Bruylants reaction. Such chemioselectivity is
probably due to the steric effect of the t-Bu group, which prevents the
formation of the immoniun intermediate. Indeed, we have observed
that addition of BuMgBr to an aldol derived from 4 yielded exclusively
to the Bruylants’s product. Moreover, we also found that it was possible
to perform the Bruylants reaction by addition of Grignard reagents to
the immoniums derived from aldols 10d ,e as described by Couty and
Agami: Agami, C, Couty, F., Ewano, G Org. Lett. 2000, 2, 2085-2088.
(23) Crystallographic data for the structural analysis have been
deposited with the Cambridge Crystallographic Data Center, CCDC
No. 178176 for 16c and CCDC No. 178175 for 16d .
(20) The modest selectivity obtained with chiral aldehydes 12 and
13 is in accordance with the results generally obtained with these
aldehydes. Moreover, it is possible that, due to the cyclic nature of the
transition state, the attack of the lithiated ketene imine to the aldehyde
does not occur in the Bu¨rgi-Dunitz angle, which is important for the
selectivity (see ref 18b).
J . Org. Chem, Vol. 67, No. 25, 2002 8931

Leclerc et al.
SCHEME 7
SCHEME 8
SCHEME 9
reduction of the imine by NaBH4. The relative configu-
ration observed by us and by Brussee were opposite. The
anti-selectivity observed by Brussee was attributed to a
chelated transition state involving the silyl ether moiety
(Scheme 9). In our case, such a transition state involving
chelation with a very bulky amino group is unlikely, and
it leads to the opposite relative configuration that is
observed. The transition state involving a Felkin type
transition state, commonly proposed for the dibenzyl
amino group,24 and the chelated free hydroxyl group leads
to the observed stereochemistry (Scheme 9).
metric induction). We presume that the observed C4-
H,C6-H anti-relationship arises from an attack of the
Z- or E-crotyl reagent (in rapid equilibrium)25 on a rigid
cyclic substrate formed by π-chelation between the nitrile
and the magnesium alkoxide, as shown in Scheme 10. 26
Therefore, of the four possible transition states, two are
the result of re face attack via a pseudo-boat (A) or a
pseudo-chair (B) transition state. The others are the
result of a si face attack again via a pseudo-boat (C) or a
pseudo-chair (D) transition state. A, C, and D are
destabilized by steric interactions. The most favorable
situation is obtained in transition state B, which leads
to the anti C4-H,C6-H relationship. The anti,anti,syn,-
Concerning the formation of 16d , the configuration of
the C6-H allylic center was controlled during the addi-
tion of the prochiral organometallic reagent (1,3-asym-
(25) Yamamoto, T.; Asao, N. Chem. Rev. 1993, 93, 2207-2293.
(26) A free hydroxyl group seems to be required for the addition of
the Grignard reagent. Indeed, we have observed that such addition
on a OSit-BuMe2 protected derivative did not occur.
(24) (a) Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1991, 30, 1531-
1546. Reetz, M. T. Chem. Rev. 1999, 99, 1121-1162.
8932 J . Org. Chem., Vol. 67, No. 25, 2002

Diastereoselective Aldolization of R-Aminonitriles
SCHEME 10
SCHEME 11
â-hydroxy-R-aminonitriles. These compounds are easily
transformed into the syn,anti-protected â,γ-diamino al-
cohols by a two-step procedure involving an addition of
a Grignard reagent to the nitrile followed by reduction
of the resulting imine. The N-tert-butyl group is then
cleaved by a simple treatment with acid. The application
of the aldol reaction to chiral aldehydes occurs with poor
facial selectivity. Nevertheless, starting from D-isopro-
pylideneglyceraldehyde and using crotylmagnesium bro-
mide, the anti,anti,syn,anti-(2R,3S,4S,5R,6R)-diaminot-
riol 18 is prepared in acceptable yield and with a good
level of diastereoselectivity. Development of an asym-
metric aldol reaction and preparation of the other ste-
reotriads of â,γ-diamino alcohols are under investigation.
Exp er im en ta l Section
anti-isomer 16d was then obtained by diastereoselective
reduction as shown in Scheme 9.
Gen er a l Con sid er a tion s. Experiments involving organo-
metallics were carried out under dry nitrogen atmosphere. All
glassware was dried at 120 °C and assembled while hot under
a stream of nitrogen. All moisture-sensitive reactants were
handled under a nitrogen atmosphere. Low-temperature ex-
periments were carried out by cooling a three-necked round
botton flask with an acetone (-80 °C), acetonitrile (-40 °C)
bath, frozen with liquid nitrogen. The flask was equipped with
an internal thermometer, a nitrogen inlet, and a septum cap.
Diethyl ether and tetrahydrofuran were distilled from sodium-
benzophenone ketyl. Organomagnesium and organolithium
reagents were titrated with 2-butanol (1 M solution in toluene)
using, respectively, 2,2′-biquinolyl and 1,10-phenanthroline as
indicators. Column chromatographies were performed over
silica gel Si 60 (0.015-0.040 or 0.040-0.063 mesh). Thin layer
chromatography (TLC) was performed over silica gel (0.25 mm;
F-254) and visualized under a UV lamp (254 and 366 nm) and
by using a 10% phosphomolybdic acid solution in ethanol
(heating) or vanillin reagent (heating) or Dragendorf reagent.
Optical rotations were measured at 20 °C. 1H NMR spectra
were recorded at 200 or 400 MHz and 13C NMR spectra at 50
or 100 MHz, in CDCl3 as a solvent. Chemical shifts are
The final cleavage of the N-t-Bu bond, although easily
achieved in the amide series,27 is rarely described for a
tertiary amine.28 We were very pleased to observe that
simple treatment with a saturated methanolic solution
of HCl under reflux afforded, after neutralization, the free
amines 17a ,b and 18 in good yields and without isomer-
ization (Scheme 11).
In conclusion, the lithiated N-benzyl-N-tert-butylami-
noacetonitrile undergoes clean 1,2-addition to aldehydes,
providing the corresponding diastereomerically pure anti-
(27) Earle, M. J .; Fairhurst, R. A.; Heaney, H.; Papageorgiou, G.
Synlett 1990, 621-623.
(28) For the cleavage of an N-tert-butyl group on a secondary amine,
see: De Kimpe, N.; Sulmon, P.; Brunet, P. J . Org. Chem. 1990, 55,
5777-5784. Bundy, G. l.; Banitt, L. S.; Dobrowolski, P. J .; Palmer, J .
R.; Schwartz, T. M.; Zimmermann, D. C.; Lipton, M. F.; Mauragis, M.
A.; Veley, M. F.; Appell, R. B.; Clouse, R. C.; Daug, E. D. Org. Process
Dev. 2001, 5, 144-151. Sachs, M. WO Patent 9636602, 1996. Kuz-
netsov, A. I.; Romanova, K. I.; Basargin, E. B.; Moskovkin, A. S.;
Unkovski, B. V. Khim. Geterotsikl. Soedin. 1990, 4, 538-542.
J . Org. Chem, Vol. 67, No. 25, 2002 8933

Leclerc et al.
1
reported in ppm (reference ) TMS for 1H spectra and CDCl3
Found: C, 77.09; H, 9.04; N, 13.78. (9b) H NMR (400 MHz):
δ 3.6 (s, 2H), 2.44 (d, J ) 4.5 Hz, 2H), 1.75 (m, 5H), 1.18 (m +
s, 13H), 0.83 (m, 2H). 13C NMR (100 MHz): δ 118.6, 54.9,
54.11, 36.7, 36.1, 31.5, 27.2, 29.9, 26.2.
for 13C spectra).
P r ep a r a tion of Am in oa ceton itr iles. N-Ben zh yd r yl-N-
ben zyla m in oa ceton itr ile (7). Benzaldehyde (2.22 g, 0.021
mol) is added to a stirred solution of aminodiphenylmethane
(3.66 g, 0.020 mol) in dichloromethane (25 mL) in the presence
of molecular sieves (4 Å, 10 g). The mixture is stirred for 1 h
at room temperature. Molecular sieves are filtered off. The
filtrate is dried over sodium carbonate and the solvant is
removed in vacuo to give the crude N-benzylidenebenzhydry-
lamine. The residue is then dissolved in methanol (15 mL).
Sodium cyanoborohydride (4.88 g, 0.030 mol) is added, and
then trifluoroacetic acid (3.10 mL, 0.040 mol) is added dropwise
until dissolution of the suspension. The mixture is stirred at
room temperature for 20 h and then neutralized by a 7:3
ammonium chloride/ammonium hydroxyde aqueous solution.
The aqueous layer is extracted with diethyl ether (2 × 25 mL).
The organic layer is dried over sodium sulfate and filtered,
and the solvents are removed under reduced pressure. The oily
residue is dissolved in diethyl ether and filtered over silica
gel. Evaporation of diethyl ether affords N-benzylbenzhydry-
lamine (5.30 g, 97%) as a white solid which is then used
without further purification. To an acetonitrile (50 mL) of
N-benzylbenzhydrylamine (5.30 g, 0.020 mol) are added bro-
moacetonitrile (1.46 mL, 0.021 mol), potassium carbonate (4.15
g, 0.030 mol), and sodium iodide (3.70 g, 0.022 mol). The
mixture is refluxed during 24 h, and then 100 mL of a
saturated aqueous sodium carbonate solution is added. The
solution is extracted with diethyl ether (4 × 25 mL), and the
combined organic layers are washed with water (6 × 50 mL),
dried over magnesium sulfate, filtered, and evaporated. The
crude product is recrystallized in pentane/diethyl ether (8:2),
yielding 2.58 g (41%) of 7 as white crystals. 1H NMR (400
MHz): δ 7.62-7.22 (m, 15H, m), 4.76 (s, 1H), 3.72 (s, 2H), 3.39
(s, 2H). 13C NMR (50 MHz): δ 141.5, 137.5, 129.1, 128.8, 127.8,
127.7, 114.8, 72.9, 55.8, 39.6. Anal. Calcd for C22H20N2: C,
84.58; H, 6.45; N, 8.97. Found: C, 84.59; H, 6.47; N, 8.95.
N-Ben zyl-N-tr ityla m in oa ceton itr ile (8). To an acetoni-
trile solution (50 mL) of benzylamine (5.30 g, 0.020 mol) are
added bromoacetonitrile (1.46 mL, 0.021 mol), potassium
carbonate (4.15 g, 0.030 mol), and sodium iodide (3.70 g, 0.022
mol). The mixture is stirred at room temperature for 20 h and
filtered. Trityl chloride (8.36 g, 0.030 mol) and triethylamine
(6.40 mL, 0.040 mol) are then added, and the mixture is stirred
overnight. A 250-mL portion of a saturated aqueous sodium
carbonate solution is added. The solution is extracted with
diethyl ether (4 × 75 mL), and the combined organic layers
are washed with water (6 × 50 mL), dried over magnesium
sulfate, filtered, and evaporated. The crude product is recrys-
tallized in cyclohexane/ethyl acetate (9:1), yielding 3.18 g (41%)
of 8 as a white solid. 1H NMR (400 MHz): δ 7.65-7.25 (m,
20H, m), 3.90 (s, 2H), 3.69 (s, 2H). 13C NMR (100 MHz): δ
147.0, 142.0, 137.4, 129.3, 128.9, 128.4, 128.0, 127.0, 115.0,
110.7, 77.6, 52.7, 38.5.
Gen er a l P r oced u r e for th e Ald oliza tion . To a THF
solution (15 mL) of aminonitrile (1.0 mmol) at -80 °C under
nitrogen atmosphere is added dropwise LDA (2 M in heptane,
1.1 mmol, 0.550 mL). After 15-30 mn at this temperature, a
THF solution (2 mL) of aldehyde (1.1 mmol) is added dropwise.
The solution is stirred (for the conditions, see the text) and
then quenched with a 7:3 ammonium chloride/ammonium
hydroxyde aqueous solution. The mixture is extracted with
diethyl ether (3 × 15 mL), and the combined organic layers
are dried over magnesium sulfate and evaporated. The crude
adduct is purified by column chromatography over silica gel.
a n ti- An d syn -N,N-Diben zyl-2-a m in o-3-h yd r oxy-3-p h e-
n ylp r op ion itr ile (5a ). Purification by column chromatogra-
phy over silica gel (eluant: pentane/diethyl ether 80:20) affords
1
pure anti- and syn-5a (0.255 g, 75%) as colorless crystals. H
NMR (400 MHz): δ 7.45-7.23 (m, 15H), 4.76 (d, J ) 9.7 Hz,
1H), 4.13 (d, J ) 13.3 Hz, 2H), 3.85 (s, 1H), 3.59 (d, J ) 9.7
Hz, 1H), 3.56 (d, J ) 13.3 Hz, 2H). (m, 15H), 3.85 (d, J ) 8.6
Hz, 1H), 3.78 (s, 1H), 3.43 (d, J ) 13.7 Hz, 2H). 13C NMR (100
MHz): δ 140.1, 138.4, 137.6, 136.9, 133.8, 129.5, 129.4, 129.1,
128.9, 127.3, 116.9, 115.17, 73.6, 71.2, 61.0, 60.5, 56.4, 56.3.
IR (film, cm-1): 3440, 3060, 2220. Anal. Calcd for C23H22N2O:
C, 80.67; H, 6.48; N, 8.18. Found: C, 80.70; H, 6.51; N, 8.15.
a n ti- An d syn -N,N-Diben zyl-2-a m in o-3-h yd r oxy-3-m es-
itylp r op ion itr ile (5b). Purification by column chromatogra-
phy over silica gel (eluant: pentane/diethyl ether 70:30) affords
pure anti-5b (0.140 g, 36%) as colorless crystals. 1H NMR (400
MHz): δ 7.30-7.07 (m, 10H), 6.79 (s, 2H), 5.44 (d, J ) 9.2 Hz,
1H), 4.28 (d, J ) 9.2 Hz, 1H), 4.00 (d, J ) 13.9 Hz, 2H), 3.45
(d, J ) 13.9 Hz, 2H), 2.35 (s, 3H), 2.02 (s, 6H). 13C NMR (100
MHz): δ 138.1, 137.6, 137.4, 132.3, 130.7, 128.9, 128.8, 127.9,
117.2, 69.5, 58.4, 56.7, 21.2, 20.5. Purification by column
chromatography over silica gel (eluant: pentane/diethyl ether
90:10) affords pure syn-5b (0.100 g, 26%). Recrystallization
from pentane, ether (9/1) gives colorless crystals (mp 137 °C).
1H NMR (400 MHz): δ 7.46-7.36 (m, 10H), 6.82 (s, 1H,), 6.72
(s, 1H, s), 5.30 (d, J ) 10.5 Hz, 1H), 4.16 (d, J ) 13.1 Hz, 2H),
4.09 (d, J ) 9.2 Hz, 1H), 3.57 (d, J ) 13.1 Hz, 2H,), 3.47 (s,
1H), 2.41 (s, 3H), 2.24 (s, 3H,), 1.73 (s, 3H). 13C NMR (100
MHz): δ 138.1, 137.6, 137.3, 136.6, 131.7, 129.8, 129.5, 129.4,
129.1, 128.3, 115.3, 66.6, 56.1, 56.0, 21.1, 20.9, 19.6.
a n ti- An d syn -N,N-Dib en zyl-2-a m in o-3-h yd r oxyl-4-
en en itr ile (5c). Purification by column chromatography over
silica gel (eluant: pentane/diethyl ether 75:25) affords pure
1
anti-5c (0.176 g, 57%) as a yellow oil. H NMR (400 MHz) δ
7.38-7.25 (m, 10H), 5.78 (m, J ) 15.3 Hz, 1H), 5.39 (ddd, J )
15.3 Hz, J ) 7.2 Hz, J ) 1.6 Hz, 1H), 4.38 (m, 1H), 3.97 (d, J
) 13.6 Hz, 2H), 3.56 (d, J ) 8.0 Hz, 1H), 3.43 (d, J ) 13.6 Hz,
2H), 2.02 (d, J ) 4.4 Hz, 1H), 1.78 (dd, J ) 7.2 Hz, J ) 1.6 Hz,
3H). 13C NMR (100 MHz): δ 137.6, 130.4, 130.0, 128.9, 128.7,
127.7, 116.5, 71.7, 58.8, 56.1, 17.8. IR (film, cm-1 ): 3440, 2210.
Anal. Calcd for C20H22N2O: C, 78.40; H, 7.24; N, 9.14. Found:
C, 78.24; H, 7.20; N, 8.99. Purification by column chromatog-
raphy over silica gel (eluant: pentane/diethyl ether 90:10)
affords pure syn-5c (0.075 g, 23%) as a yellow oil. 1H NMR
(400 MHz): δ 7.42-7.32 (m, 10H), 5.93 (m, 1H), 5.39 (ddd, J
) 15.2 Hz, J ) 7.1 Hz, J ) 1.1 Hz, 1H), 4.21 (dd, J ) 9.6 Hz,
J ) 9.6 Hz, 1H), 4.04 (d, J ) 13.4 Hz, 2H), 3.51 (d, J ) 13.4
Hz, 2H), 3.45 (d, J ) 9.6 Hz, 1 H), 3.32 (s, 1H), 1.74 (d, J )
6.6 Hz, 3H). 13C NMR (100 MHz): δ 136.7, 132.7, 129.1, 129.0,
128.1, 127.6, 115.1, 69.8, 59.0, 55.9, 17.9.
N-Ben zyl-N-ter t-bu tyla m in oa ceton itr ile (9a ) a n d N-
Cycloh exylm eth yl-N-ter t-bu tyla m in oa ceton itr ile (9b). To
an acetonitrile solution (30 mL) of N-benzyl-tert-butylamine
or N-cyclohexylmethyl-tert-butylamine (0.020 mol) are added
bromoacetonitrile (1.46 mL, 0.021 mol), potassium carbonate
(5.53 g, 0.040 mol), and sodium iodide (3.00 g, 0.020 mol). The
mixture is stirred overnight at room temperature, and then
100 mL of a saturated aqueous sodium carbonate solution is
added. The solution is extracted with diethyl ether (4 × 25
mL), and the combined organic layers are washed with water
(6 × 50 mL), dried over magnesium sulfate, filtered, and
evaporated. The crude product is chromatographied over silica
gel (eluant: pentane/diethyl ether 9:1), yielding 3.20 g (79%)
a n ti- An d syn -N,N-Diben zyl-2-a m in o-3-h yd r oxyh ep -
ta n en itr ile (5d ). Purification by column chromatography over
silica gel (eluant: pentane/diethyl ether 75:25) affords pure
1
of 9a or 3.3 g (79%) of 9b as colorless oils. (9a ) H NMR (400
1
MHz): δ 7.38-7.26 (m, 5H), 3.84 (s, 2H), 3.45 (s, 2H), 1.30 (s,
9H).13C NMR (100 MHz): δ 138.9, 128.7, 127.5, 118.2, 55.3,
51.5, 35.8, 27.5. IR (film, cm-1 ): 3080, 3060, 3030, 2930, 2920,
2880. Anal. Calcd for C13H18N2: C, 77.18; H, 8.97; N, 13.85.
anti-5d (0.170 g, 53%) as a colorless oil. H NMR (400 MHz):
δ 7.42-7.27 (m, 10H), 3.98 (d, J ) 13.5 Hz, 2H), 3.92 (m, 1H,
m), 3.51 (d, J ) 8.2 Hz, 1 H), 3.45 (d, J ) 13.5 Hz, 2H), 2.00
(s, 1 H), 1.85-1.80 (m, 1H), 1.33-1.14 (m, 5H), 0.88 (m, J )
8934 J . Org. Chem., Vol. 67, No. 25, 2002

Diastereoselective Aldolization of R-Aminonitriles
7.3 Hz, 3H). 13C NMR (100 MHz): δ 137.9, 129.3, 129.0, 128.1,
117.0, 70.9, 59.9, 56.9, 33.7, 27.0, 22.9, 14.4. IR (film, cm-1 ):
3060, 2230. Purification by column chromatography over silica
gel (eluant: pentane/diethyl ether 85:15) affords pure syn-5d
(0.050 g, 17%). 1H NMR (400 MHz): δ 7.40-7.28 (m, 10H),
4.03 (d, J ) 13.4 Hz, 2H), 3.85-3.78 (m, 1H), 3.48 (d, J ) 13.4
Hz, 2H), 3.41 (d, J ) 9.5 Hz, 1H), 3.23 (s, 1H), 1.72-1.25 (m,
6H), 0.90 (m, J ) 6.7 Hz, 3H).
127.4, 126.9, 119.5, 72.0, 58.4, 56.9, 51.4, 33.8, 31.8, 29.1, 27.8,
25.7, 22.6, 14.1. IR (film, cm-1 ): 3400, 2220.
N-Ben zyl-N-ter t-bu tyl-2-a m in o-4,4-d im eth yl-3-h yd r ox-
yva ler on itr ile (10e). Recrystallization of the crude adduct
(pentane/diethyl ether 9:1) affords pure 10e (0.255 g, 88%) as
1
colorless crystals (mp 118 °C). H NMR (400 MHz): δ 7.44-
7.19 (m, 5H), 4.32 (s, 1H), 4.22 (d, J ) 17.6 Hz, 1H), 4.17 (d,
J ) 17.6 Hz, 1H), 3.44 (dd, J ) 6.6 Hz, J ) 0.8 Hz, 1H), 1.71
(d, J ) 6.6 Hz, 1H), 1.20 (s, 9H), 0.98 (s, 9H). 13C NMR (100
MHz): δ 143.0, 128.6, 126.6, 119.2, 82.5, 57.7, 52.3, 50.8, 35.9,
27.8, 26.1. IR (film, cm-1 ): 3430, 2230. Anal. Calcd for
a n ti- An d syn -N,N-Diben zyl-2-a m in o-3-cycloh exyl-3-
h yd r oxyp r op ion itr ile (5e). Purification by column chroma-
tography over silica gel (eluant: pentane/diethyl ether 75:25)
1
affords pure anti-5e (0.176 g, 50%) as a colorless oil. H NMR
C18H28N2O: C, 74.96; H, 9.78; N, 9.71. Found: C, 75.01; H,
(400 MHz): δ 7.40-7.27 (m, 10H), 3.97 (d, J ) 13.4 Hz, 2H),
3.74 (m, 1H), 3.67 (d, J ) 8.6 Hz, 1H), 3.45 (d, J ) 13.4 Hz,
2H), 2.04 (s, 1H), 1.78-0.81 (m, 11H). 13C NMR (100 MHz): δ
138.0, 129.4, 129.0, 128.1, 117.4, 74.7, 57.2, 56.9, 39.3, 30.6,
26.9, 26.6, 26.3, 24.3. IR (film, cm-1 ) 3460, 2230. Anal. Calcd
for C23H28N2O: C, 79.27; H, 8.10; N, 8.04. Found: C, 79.06;
H, 8.34; N, 7.61. Purification by column chromatography over
silica gel (eluant: pentane/diethyl ether 85:15) affords pure
syn-5e (0.050 g, 15%). 1H NMR (CDCl3): δ 7.40-7.27 (m, 10H),
4.03 (d, J ) 13.3 Hz, 2H), 3.70 (m, 1H), 3.61 (d, J ) 9.7 Hz, 1
H), 3.48 (d, J ) 13.3 Hz, 2 H), 1.77-0.85 (m, 11 H).
9.87; N, 9.66.
N-Ben zyl-N-ter t-bu tyl-2-am in o-3-h ydr oxy-3-ph en ylpr o-
p ion itr ile (10f). Recrystallization of the crude adduct (pen-
tane/diethyl ether 9:1) affords pure 10f (0.244 g, 79%) as
1
colorless crystals (mp 132 °C). H NMR (400 MHz): δ 7.38-
7.22 (m, 10H), 4.61 (d, J ) 6.0 Hz, 1H), 4.31 (d, J ) 6.0 Hz,
1H), 4.23 (d, J ) 17.0 Hz, 1H), 4.11 (d, J ) 17.0 Hz, 1H), 2.18
(s, 1H); 1.08 (s, 9H). 13C NMR (100 MHz): δ 142.0, 140.0, 128.7,
128.5, 127.5, 127.1, 126.9, 118.7, 75.3, 59.2, 57.3, 51.4, 27.6.
IR (film, cm-1 ): 3500, 2220. Anal. Calcd for C20H24N2O: C,
77.89; H, 7.84; N, 9.08. Found: C, 77.83; H, 7.98; N, 9.26.
a n ti-N,N-Dib en zyl-2-a m in o-4,4-d im et h yl-3-h yd r oxy-
va ler on itr ile (5f). Recrystallization (cyclohexane or pentane/
diethyl ether 2:1) affords pure anti-5f (0.209 g, 65%) as
N-Cych oh exylm eth yl-N-ter t-bu tyl-2-am in o-3-cycloh exyl-
3-h yd r oxyp r op ion itr ile (10g). Recrystallization of the crude
adduct (pentane/diethyl ether 9:1) affords pure 10g (0.233 g,
73%) as colorless crystals (mp 142 °C). 1H NMR (400 MHz): δ
3.95 (d, J ) 5.5 Hz, 1H), 3.51 (bd, J ) 5.5 Hz, 1H), 2.73 (dd,
J ) 7.5 Hz, J ) 5.5 Hz, 1H), 2.35 (dd, J ) 7.5 Hz, J ) 5.5 Hz,
1H), 2-0.9 and 1.36 (m + s, 31 H). 13C NMR (100 MHz): δ
120.8, 75.8, 57.15, 56.4, 55.2, 40.0, 38.1, 32.3, 31.9, 31.1, 28.2,
27.1, 26.8. Anal. Calcd for C20H36N2O: C, 74.95; H, 11.32; N,
8.74. Found: C, 74.99; H, 11.35; N, 8.69.
1
colorless crystals (mp 110 °C). H NMR (400 MHz): δ 7.47-
7.27 (m, 10H, m), 4.17 (d, J ) 13.8 Hz, 2H), 3.86 (d, J ) 1.8
Hz, 1H), 3.56 (d, J ) 1.8 Hz, 1H), 3.43 (d, J ) 13.8 Hz, 2H),
2.00 (s 1H), 0.78 (s, 9H). 13C NMR (100 MHz): δ 137.0, 132.7,
127.9, 127.6, 126.6, 120.4, 114.5, 79.9, 55.0, 54.3, 34.7, 24.3.
Anal. Calcd for C21H26N2O: C, 78.22; H, 8.13; N, 8.69. Found:
C, 78.17; H, 8.26; N, 8.51.
3-Cycloh exyl-2-(2′,5′-d im eth ylp yr r olyl)-3-h yd r oxyp r o-
p ion itr ile (10a -1). Major diastereomer 1H NMR (400 MHz):
δ 5.83 (s, 2H), 5.03 (d, J ) 7.1 Hz, 1H) 3.98 (m, 1H), 2.73 (s,
1H), 2.33 (s, 6H), 1.81-1.10 (m, 11H). 13C NMR (100 MHz): δ
128.5, 116.7, 108.1, 76.1, 49.6, 39.0, 30.1, 26.4, 26.1, 25.9, 25.6,
13.6. Minor diastereomer 1H NMR (400 MHz): δ 5.80 (s, 2H),
4.94 (d, J ) 9.7 Hz, 1H), 3.92 (d, J ) 9.7 Hz, 1H), 2.58-1.14
(m, 17H). 13C NMR (100 MHz): δ 129.0, 115.9, 107.8, 75.6,
48.7, 39.6, 30.6, 26.4, 26.3, 26.0, 24.5, 13.8. Anal. Calcd for
N-Ben zyl-N-ter t-bu tyl-2-a m in o-4-(ter t-bu tyld im eth yl)-
silyloxy-3-h yd r oxy-4-p h en ylbu tyr on itr ile (11a ). Purifica-
tion by column chromatography over silica gel (eluant: pentane/
diethyl ether 95:05) affords pure aldol 11a (0.366 g, 81%) as a
mixture of two diastereomers (dr ) 50/50). Anal. Calcd for
C
27H41N2O2Si: C, 71.63; H, 8.91; N, 6.19. Found: C, 71.72; H,
9.08; N, 6.17. HRMS (CI+, CH4): calcd for C27H41N2O2Si (M +
H+) 453.2937, found 453.2930.
N-Ben zyl-N-ter t-bu tyl-2-a m in o-4-(ter t-bu tyld im eth yl)-
silyloxy-3-h yd r oxyva ler on itr ile (12a ). Purification by col-
umn chromatography over silica gel (eluant: pentane/diethyl
ether 9:1) affords pure aldol (0.230 g, 59%) as a mixture of
two diastereomers (dr ) 62:38). Major diastereomer . 1H NMR
(200 MHz): δ 7.42-7.15 (m, 5H), 4.47 (d, J ) 4.4 Hz, 1H),
4.22 (d, J ) 17.7 Hz, 1H), 4.07 (d, J ) 17.7 Hz, 1H), 3.79 (m,
1H), 3.52 (m, 1H), 1.90 (d, J ) 4.4 Hz, 1H), 1.18 (s, 9H), 1.16
(d, J ) 7.4 Hz, 3H), 0.87 (s, 9H), 0.06 (s, 3H), 0.05 (s, 3H). 13C
NMR (50 MHz): δ 142.7, 128.8, 126.9, 126.6, 118.8, 78.7, 77.4,
69.0, 57.7, 53.4, 51.4, 27.7, 25.9, 19.4, -4.1, -4.7. Minor
diastereomer. 1H NMR (200 MHz): δ 7.42-7.15 (m, 5H), 4.24
(d, J ) 17.2 Hz, 1H), 4.14 (m, 1H), 4.06 (d, J ) 8.8 Hz, 1H),
3.91 (d, J ) 17.2 Hz, 1H), 3.13 (dd, J ) 8.8 Hz, J ) 9.3 Hz,
1H), 1.24 (s, 9H), 0.91 (s, 9H), 0.87 (d, J ) 6.4 Hz, 3H), 0.15
(s, 3H), 0.14 (s, 3H). Anal. Calcd for C22H38N2O2Si: C, 67.64;
H, 9.80; N, 7.17. Found: C, 67.80; H, 9.97; N, 6.97.
N-Ben zyl-N-ter t-b u t yl-4,5-O-isop r op ylid en e-2-a m in o-
3,4,5-tr ih yd r oxyva ler on itr ile (14). Recrystallization of the
crude adduct (pentane/diethyl ether 9:1) affords the pure major
diastereomer 14 (0.146 g, 44%) as colorless crystals (mp 140
°C). [R]20D ) -9.0 (c 1.2, CHCl3). 1H NMR (400 MHz): δ 7.43-
7.22 (m, 5H), 4.49 (d, J ) 3.1 Hz, 1H), 4.26 (d, J ) 17.4 Hz,
1H), 4.06 (m, 1H), 4.05 (m, J ) 17.4 Hz, 1H), 3.99 (dd, J )
8.6 Hz, J ) 6.4 Hz, 1H), 3.88 (dd, J ) 8.6 Hz, J ) 4.1 Hz, 1H),
3.62 (m, 1H), 1.76 (d, J ) 4.8 Hz, 1H), 1.42 (s, 3H), 1.33
(s, 3H), 1.21 (s, 9H). 13C NMR (100 MHz): δ 142.1, 129.0,
127.1, 126.7, 117.9, 109.8, 75.9, 75.2, 66.2, 57.6, 53.7, 51.2, 27.6,
27.1, 25.0. IR (film, cm-1 ): 3460, 2220. Anal. Calcd for
C
15H22N2O: C, 73.13; H, 9.00; N, 11.37. Found: C, 72.93; H,
9.14; N, 11.17.
N-Ben zyl-N-ter t-bu tyl-2-a m in o-3-cycloh exyl-3-h yd r ox-
yp r op ion itr ile (10b). Purification by column chromatography
over silica gel (eluant: pentane/diethyl ether 85:15) affords
1
pure 10b (0.264 g, 84%) as white solid. H NMR (400 MHz):
δ 7.36-7.18 (m, 5H), 4.22 (d, J ) 17.1 Hz, 1H), 4.13 (d, J )
7.8 Hz, 1H), 3.93 (d, J ) 17.1 Hz, 1H), 3.26-3.21 (m, 1H), 1.85
(d, J ) 5.4 Hz, 1H), 1.78-1.04 (m, 11H), 1.21 (s, 9H). 13C NMR
(100 MHz): δ 143.2, 128.6, 127.2, 126.9, 120.0, 75.9, 57.3, 55.4,
51.5, 38.4, 30.6, 27.9, 26.6, 26.4, 26.2, 25.4. IR (film, cm-1 ):
3420, 2220. Anal. Calcd for C20H30N2O: C, 76.39; H, 9.62; N,
8.91. Found: C, 76.42; H, 9.58; N, 8.86.
N-Ben zyl-N-ter t-b u t yl-2-a m in o-3-h yd r oxyh ep t a n en i-
tr ile (10c). Recrystallization of the crude adduct (pentane)
affords pure 10c (0.248 g, 86%) as colorless crystals (mp 67
°C). 1H NMR (400 MHz): δ 7.36-7.19 (m, 5H), 4.22 (d, J )
16.7 Hz, 1H), 3.91-3.86 (m, 2H), 3.28 (m, 1H), 2.01 (s, 1H),
1.86 (m, 1H), 1.44-0.88 (m, 5H), 1.20 (s, 9H), 0.86 (t, J ) 7.3
Hz, 3H). 13C NMR (100 MHz): δ 141.8, 128.5, 127.4, 126.8,
119.6, 72.0, 58.5, 57.0, 51.5, 33.5, 28.0, 27.8, 22.6, 14.1. IR (film,
cm-1 ): 3420, 2220. Anal. Calcd for C18H28N2O: C, 74.96; H,
9.78; N, 9.71. Found: C, 74.81; H, 9.94; N, 9.71.
N-Ben zyl-N-ter t-b u t yl-2-a m in o-3-h yd r oxyn on a n en i-
tr ile (10d ). Purification by column chromatography over silica
gel (eluant: pentane/diethyl ether 9:1) affords pure 10d (0.252
1
g, 80%) as white solid. H NMR (400 MHz): δ 7.36-7.14 (m,
5H), 4.21 (d, J ) 16.7 Hz, 1H), 3.92-3.21 (m, 2H), 3.24 (m,
1H), 2.06 (s, 1H), 1.85 (m, 1H), 1.45-0.92 (m, 9H), 1.19 (s, 9H),
0.87 (t, J ) 6.9 Hz, 3H). 13C NMR (100 MHz): δ 141.8, 128.4,
C
19H28N2O3: C, 68.65; H, 8.49; N, 8.43. Found: C, 68.42; H,
8.70; N, 8.38.
J . Org. Chem, Vol. 67, No. 25, 2002 8935

Leclerc et al.
Gen er a l P r oced u r e for th e Alk ylm a gn esia tion -Re-
d u ction Sequ en ce. To a THF solution (5 mL) of â-hydroxy-
R-Rminonitrile (1.0 mmol) at 0 °C under nitrogen atmosphere
is added dropwise the Grignard reagent (2.3 mmol). The
solution is stirred until complete consumption of the starting
material (TLC, eluant: cyclohexane/ethyl acetate 8:2), which
usually requires 2 h at 0 °C and 1 h at room temperature.
The reaction mixture is quenched with methanol (15 mL), and
sodium borohydride (0.080 g, 2.0 mmol) is added at 0 °C. The
solution is allowed to warm slowly to room temperature and
stirred overnight. A 7:3 ammonium chloride/ammonium hy-
droxyde aqueous solution is added and the mixture is extracted
with ethyl acetate (3 × 15 mL). The combined organic layers
are dried over magnesium sulfate, filtered, and evaporated
under reduced pressure. The crude â,γ-diamino alcohol is
purified by column chromatography over silica gel.
2-N-Ben zyl-2-N-ter t-b u t yl-2,3-d ia m in o-1-cycloh exyl-
h ep ta n -1-ol (15a ). 2-N-Ben zyl-2-N-ter t-bu tyl-1,2-d ia m in o-
1-p h en yln on a n -3-ol (15b). Purification by column chroma-
tography over silica gel (eluant: cyclohexane/ethyl acetate 7:3
to 0:1) affords pure 15b (0.275 g, 69%) as a yellow solid. 1H
NMR (200 MHz): δ 7.52-7.15 (m, 10H), 4.66 (d, J ) 18.2 Hz,
1H), 4.24 (d, J ) 3.0 Hz, 1H), 4.10 (d, J ) 18.2 Hz, 1H), 3.98
(m, 1H), 3.10 (d, J ) 3.0 Hz, 1H), 1.68-1.18 (m, 10H), 0.91-
0.84 (m, 3H), 0.71 (s, 9H). 13C NMR (50 MHz): δ 145.3, 128.3,
127, 126.9, 126.0, 75.3, 65.7, 56.8, 56.5, 49.6, 38.6, 32.1, 29.6,
28.2, 26.7, 22.9, 14.3. IR (film, cm-1): 3360, 3090, 3050, 3020.
Anal. Calcd for C26H40N2O: C, 78.74; H, 10.17; N, 7.06.
Found: C, 78.72; H, 10.17; N, 6.81.
1H), 4.09-3.93 (m, 3H), 3.41 (d, J ) 2.5 Hz, 1H), 3.01 (m, 1H),
2.67-2.35 (m, 2H), 1.41 (s, 3H), 1.37 (s, 3H), 1.12 (s, 9H). 13C
NMR (50 MHz): δ 144.9, 135.5, 128.4, 126.7, 125.9, 118.4,
109.5, 79.0, 77.6, 68.8, 57.6, 57.3, 51.9, 50.1, 42.1, 28.5, 26.5,
25.5. IR (film, cm-1 ): 3350, 3300, 3050, 2960, 1630. Anal.
Calcd for C22H36N2O3: C, 70.18; H, 9.64; N, 7.44. Found: C,
70.01; H, 9.76; N, 7.29.
4-N-Ben zyl-4-N-ter t-bu tyl-1,2-O-isop r op ylid en e-4,5-d i-
a m in o-2,3-d ih yd r oxyn on a n -1-ol (16c). Recrystallization of
the crude adduct (cyclohexane/ethyl acetate 9:1) affords 16c
(0.240 g, 61%) as colorless crystals (mp 115 °C). [R]20D ) -13.2
(c 1.0, CHCl3). 1H NMR (200 MHz): δ 7.33-7.09 (m, 5H,), 4.69
(d, J ) 18.7 Hz, 1H); 4.19 (m, 1H), 4.18 (d, J ) 18.7 Hz, 1H),
4.05-3.92 (m, 3H), 3.35 (d, J ) 2.0 Hz, 1H), 2.94 (m, 1H),
1.85-1.58 (m, 2H), 1.49-1.01 (m, 4H), 1.42 (s, 3H), 1.38 (s,
3H), 1.11 (m, 9H), 0.96 (t, J ) 6.9 Hz, 3H). 13C NMR (50
MHz): δ 145.2, 128.4, 126.6, 125.8, 109.4, 79.0, 77.6, 68.7, 57.6,
57.1, 53.0, 50.3, 37.1, 28.5, 28.4, 26.5, 25.5, 22.9, 14.3. IR (film,
cm-1 ): 3350, 3300, 3050 Anal. Calcd for C23H40N2O3: C, 70.37;
H, 10.27; N, 7.14. Found: C, 70.29; H, 9.81; N, 7.19.
4-N-Ben zyl-4-N-ter t-bu tyl-1,2-O-isop r op ylid en e-4,5-d i-
a m in o-2,3-d ih yd r oxy-6-m eth ylp en t-7-en -1-ol (16d ). Puri-
fication by column chromatography over silica gel (eluant:
cyclohexane/ethyl acetate 7:3 to 0:1) affords pure 16d (0.310
g, 79%) as colorless crystals (mp 142 °C). [R]20 ) +23 (c 0.8,
D
1
CHCl3). H NMR (200 MHz): δ 7.48-7.15 (m, 5H), 5.60-5.52
(m, 1H), 5.20-5.09 (m, 2H), 4.68 (d, J ) 18.6 Hz, 1H), 4.23
(m, 1H), 4.20 (d, J ) 18.6 Hz, 1H), 4.06-3.93 (m, 3H, m), 3.60
(d, J ) 1.5 Hz, 1H), 2.86 (m,1H), 2.60 (dd, J ) 9.8 Hz, J ) 1.5
Hz, 1H), 1.43 (s, 3H, s), 1.37 (s, 3H, s), 1.15 (s, 9H), 1.10 (d, J
) 6.0 Hz, 3H). 13C NMR (50 MHz): δ 144.9, 141.5, 128.4, 126.6,
125.9, 116.9, 109.4, 78.4, 77.4, 68.9, 57.8, 57.0, 54.2, 50.4, 41.7,
28.6, 26.5, 25.5, 18.8. IR (film, cm-1 ): 3360, 3050, 2970, 2900,
1630. Anal. Calcd for C23H38N2O3: C, 70.73; H, 9.81; N, 7.17.
Found: C, 70.53; H, 9.87; N, 7.05.
2-N-Ben zyl-2-N-ter t-bu tyl-1,2-d ia m in o-1-p h en yln on a n -
3-ol (15b). Purification by column chromatography over silica
gel (eluant: cyclohexane/ethyl acetate 7:3 to 0:1) affords pure
46a (0.275 g, 69%) as a yellow solid. 1H NMR (200 MHz) δ
7.52-7.15 (m, 10H), 4.66 (d, J ) 18.2 Hz, 1H), 4.24 (d, J ) 3.0
Hz, 1H), 4.10 (d, J ) 18.2 Hz, 1 H), 3.98 (m, 1 H), 3.10 (d, J )
3.0 Hz, 1 H), 1.68-1.18 (m, 10 H), 0.91-0.84 (m, 3 H), 0.71 (s,
9 H). 13C NMR (50 MHz) δ 145.3, 128.3, 127.2, 126.9, 126.0,
75.3, 65.7, 56.8, 56.5, 49.6, 38.6, 32.1, 29.6, 28.2, 26.7, 22.9,
14.3. IR (film, cm-1): 3090, 3050, 3020 Anal. Calcd for
Dep r otection of th e N-ter t-Bu tyl-â,γ-Dia m in o Alco-
h ols. Gen er a l P r oced u r e. The â,γ-diamino alcohol (1.0
mmol) is dissolved in a saturated methanolic solution of
hydrogen chloride (5 mL). The mixture is refluxed for 1 h and
neutralized by a saturated aqueous solution of sodium carbon-
ate (10 mL). Extraction with ethyl acetate (3 × 15 mL) and
drying of the combined organic layers with magnesium sulfate,
filtration, and evaporation under reduced pressure afford the
crude free diamino alcohols.
C
26H40N2O: C, 78.74; H, 10.17; N, 7.06. Found: C, 78.72; H,
10.17; N, 6.81.
5-N-Ben zyl-5-N-ter t-bu tyl-4,5-d ia m in od od ec-1-en -6-ol
(15c). Purification by column chromatography over silica gel
(eluant: cyclohexane/ethyl acetate 7:3 to 0:1) affords pure 15c
1
(0.245 g, 68%) as a white solid. H NMR (200 MHz): δ 7.46-
2-N-Ben zyl-2,3-d ia m in o-1-cycloh exylh ep ta n -1-ol (17a ).
Purification by column chromatography over neutral alumi-
num oxyde (eluant: methanol) affords pure 17a (0.203 g, 64%)
7.08 (m, 5H), 5.73 (m, 1H), 5.16-5.02 (m, 2H), 4.64 (d, J )
18.7 Hz, 1H), 4.12 (d, J ) 18.7 Hz, 1H), 4.00 (m, 1H), 2.99 (m,
1H), 2.84 (s, 1H), 2.58-2.32 (m, 2H), 1.74-1.12 (m, 10H), 1.04
(s, 9H), 0.94-0.82 (m, 3H). 13C NMR (50 MHz): δ 144.9, 135.6,
128.3, 126.6, 125.8, 118.1, 75.4, 61.6, 57.1, 52.4, 49.8, 41.9, 38.6,
32.0, 29.5, 28.6, 26.5, 22.8, 14.2. Anal. Calcd for C23H40N2O:
C, 76.61; H, 11.18; N, 7.77. Found: C, 77.03; H, 11.34; N, 7.40.
1
as a colorless oil. H NMR (200 MHz): δ 7.32-7.17 (m, 5H),
3.94 (d, J ) 13.3 Hz, 1H), 3.68 (d, J ) 13.3 Hz, 1H), 3.43 (dd,
J ) 8.9 Hz, J ) 3.0 Hz, 1H), 2.96 (m, 1H), 2.51 (s, 1H), 2.23
(m, 1H), 1.85-0.75 (m, 19H). 13C NMR (50 MHz): δ 140.8,
128.7, 128.3, 127.0, 77.8, 56.4, 52.9, 51.5, 41.8, 36.1, 30.4, 28.9,
28.3, 26.6, 26.0, 22.8, 19.6, 14.1. IR (film, cm-1 ): 3300, 3060,
3040, 2920, 2850, 1950. HRMS (CI+, CH4): calcd for C20H35N2O
(M + H+) 319.2749, found 319.2743.
4-N-Ben zyl-4-N-ter t-bu tyl-1,2-O-isop r op ylid en e-4,5-d i-
a m in o-2,3-d ih yd r oxy-5-p h en ylp en ta n -1-ol (16a ). Purifica-
tion by column chromatography over silica gel (eluant: cyclo-
hexane/ethyl acetate 7:3 to 0:1) affords pure 16a (0.235 g, 57%)
5-N-Ben zyl-4,5-d ia m in o-6-h yd r oxyd od ec-1-en e (17b).
Purification by column chromatography over neutral alumi-
num oxide (eluant: methanol) affords pure 17b (0.212 g, 70%)
as a white solid. [R]20 ) -88 (c 1.0, CHCl3).1H NMR (200
D
MHz): δ 7.55-7.19 (m, 10H), 4.68 (d, J ) 18.2 Hz, 1H), 4.30-
4.24 (m, 2H), 4.16 (d, J ) 18.2 Hz, 1H), 4.13 (m, 1H), 4.05-
3.95 (m, 2H), 3.71 (d, J ) 2.1 Hz, 1H), 1.42 (s, 6H), 0.71 (s,
9H). 13C NMR (50 MHz): δ 144.8, 128.3, 128.2, 126.9, 126.8,
126.0, 109.5, 78.4, 77.5, 68.7, 60.7, 57.2, 56.1, 49.8, 27.8, 26.5,
25.4. IR (film, cm-1 ): 3360, 3300, 3085, 3040, 3010. HRMS
(CI+, CH4): calcd for C25H37N2O3 (M + H+) 413.2804, found
413.2803.
1
as a colorless oil. H NMR (200 MHz): δ 7.38-7.21 (m, 5H),
5.63 (m, 1H), 5.08-4.91 (m, 2H), 3.94 (d, J ) 12.8 Hz, 1H),
3.77 (m, 1H), 3.72 (d, J ) 12.8 Hz, 1H), 3.08 (m, 1H), 2.34 (m,
1H), 2.28-2.20 (m, 2H), 1.61-0.78 (m, 13H). 13C NMR (50
MHz): δ 140.6, 135.2, 128.7, 128.5, 127.2, 118.1, 73.2, 59.9,
53.1, 51.0, 40.5, 35.8, 32.0, 29.9, 26.4, 22.7, 14.3. IR (film, cm-1
): 3300, 3060, 3020, 2930, 2860, 1940, 1870, 1810. HRMS (CI+,
CH4): calcd for C19H33N2O (M + H+) 305.2593, found 305.2591.
4-N-Ben zyl-4-N-ter t-bu tyl-1,2-O-isop r op ylid en e-4,5-d i-
a m in o-2,3-d ih yd r oxyp en t-7-en -1-ol (16b). Purification by
column chromatography over silica gel (eluant: cyclohexane/
4-N-Ben zyl-1,2-O-isop r op ylid en e-4,5-d ia m in o-2,3-d ih y-
d r oxy-6-m eth ylp en t-7-en -1-ol (18). Purification by column
chromatography over neutral aluminum oxyde (eluant: metha-
ethyl acetate 7:3 to 0:1) affords pure 16b (0.255 g, 68%) as a
1
white solid. [R]20 ) -12 (c 1.0, CHCl3). H NMR (200 MHz):
nol) affords pure 18 (0.220 g, 75%) as a colorless oil. [R]20
)
D
D
δ 7.46-7.12 (m, 5H), 5.88-5.66 (m, 1H), 5.18-5.05 (m, 2H),
4.69 (d, J ) 18.2 Hz, 1H), 4.23 (m, 1H), 4.17 (d, J ) 18.2 Hz,
+21 (c 1.0, CHCl3). 1H NMR (200 MHz): δ 7.38-7.22 (m, 5H),
5.56 (m, 1H), 5.11-5.00 (m, 2H), 3.94 (d, J ) 12.3 Hz, 1H),
8936 J . Org. Chem., Vol. 67, No. 25, 2002

Diastereoselective Aldolization of R-Aminonitriles
3.90-3.62 (m, 5H), 3.00 (d, J ) 2.5 Hz, 1H), 2.72 (m, 1H); 2.29
(m, 1H), 0.88 (d, J ) 6.9 Hz, 3H). 13C NMR (50 MHz): δ 141.8,
140.3, 128.6, 128.5, 127.3, 116.5, 73.4, 73.0, 64.7, 56.3, 55.5,
53.2, 43.3, 17.7. IR (film, cm-1 ): 3340, 3060, 2960, 2920, 2860,
1930, 1880. HRMS (CI+, CH4): calcd for C16H27N2O3 (M + H+)
295.2022, found 295.2017.
Normant for many interesting discussions and Patrick
Herson for the crystal stucture determinations.
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra of 5a -f (anti and syn), 6, 7, 8, 9a -b, 10a -g, 11a , 12a ,
14, 15a -c, 16a -d , 17a -b, 18 and ORTEP drawings for anti-
5f, 10c, 14, 16c, and 16d . This material is available free of
charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. We thank Aventis for financial
support through a grant to E.L. We thank also Dr.
Vivien Henryon, Prof. Alex Alexakis, and Prof. J ean
J O025872T
J . Org. Chem, Vol. 67, No. 25, 2002 8937