Molecular modeling of salt (lithium chloride) effects on the enantioselectivity of diethylzinc addition to benzaldehyde in the presence of chiral β-amino alcohols

Article abstract of DOI:10.1021/jo026811y

β-Amino alcohols (S,S,S)-1 and (R,R,S)-1, derived from cyclohexene oxide and containing α-phenylethyl auxiliaries, were examined as chiral promoters in the addition of diethylzinc to benzaldehyde. In agreement with literature precedent, the N-α-phenylethy

Full text of DOI:10.1021/jo026811y

Molecu la r Mod elin g of Sa lt (Lith iu m Ch lor id e) Effects on th e
En a n tioselectivity of Dieth ylzin c Ad d ition to Ben za ld eh yd e in th e
P r esen ce of Ch ir a l â-Am in o Alcoh ols
Martha Sosa-Rivadeneyra,† Omar Mun˜oz-Mun˜iz,‡ Cecilia Anaya de Parrodi,§
Leticia Quintero,*,† and Eusebio J uaristi*,‡
Centro de Investigacio´n de la Facultad de Ciencias Qu´ımicas, Universidad Auto´noma de Puebla,
72570 Puebla, Me´xico, Centro de Investigacio´n y de Estudios Avanzados del Instituto Polite´cnico
Nacional, Apartado Postal 14-740, 07000 Me´xico, D.F., Me´xico, and Centro de Investigaciones Qu´ımico
Biolo´gicas, Universidad de las Ame´ricas-Puebla, Santa Catarina Ma´rtir, 72820 Cholula, Me´xico.
ejuarist@enigma.red.cinvestav.mx
Received December 5, 2002
â-Amino alcohols (S,S,S)-1 and (R,R,S)-1, derived from cyclohexene oxide and containing R-phen-
ylethyl auxiliaries, were examined as chiral promoters in the addition of diethylzinc to benzaldehyde.
In agreement with literature precedent, the N-R-phenylethyl chiral auxiliary had no significant
impact on enantioinduction, which is determined by the configuration of the framework’s C*(OH),
with unlike induction. Contrary to some literature reports, stereoinduction by lithium salt
derivatives of (S,S,S)-1 and (R,R,S)-1 was lower than that obtained with the free amino alcohol.
Remarkable lithium chloride salt effects were observed in the reaction. In particular, an opposite
chiral induction was found with (S,S,S)-1-Li2 as ligand and in the presence of “inert” salt.
N-Alkylated derivatives (S,S,S)-3-7 proved to be more efficient ligands, providing higher yields
and enantioselectivities in the formation of carbinols (R)- or (S)-2. BP86/DN**//PM3 theoretical
calculations proved remarkably successful in reproducing the experimental observations and
permitted expansion of Noyori’s catalytic cycle [J . Am. Chem. Soc. 1995, 117, 6327] to understand
the relevant N-substitution and medium salt effects that determine the enantioselection in this
catalytic asymmetric reaction.
In tr od u ction
of diastereomeric analogues of complex D, whose relative
stabilities determine the sense and extent of stereo-
selection.6-8
Dialkylzinc addition to prochiral aldehydes in the
presence of chiral â-amino alcohols constitutes a highly
efficient procedure for the asymmetric construction of
C-C bonds.1-4 (Scheme 1).
A pioneering ab initio molecular orbital study {MP2//
HF/3-21G (H,C,N,O) + [8s4p2d]/(14s9p5d) for Zn} of this
reaction by Yamakawa and Noyori5 afforded essential
information on the structures, stabilities, and reactivities
of the organozinc complexes that are involved in the
catalytic process. Figure 1 summarizes the main features
of the reaction course, modeled with dimethylzinc, 2-ami-
noethanol, and formaldehyde.5 Consideration of chiral
ligands instead of achiral 2-aminoethanol and prochiral
aldehydes instead of formaldehyde leads to the formation
(4) For original reports on some of the most successful chiral â-amino
alcohol ligands, see: (a) Kitamura, M.; Suga, S.; Kawai, K.; Noyori, R.
J . Am. Chem. Soc. 1986, 108, 6071. (b) Tanaka, K.; Ushio, H.; Suzuki,
H. J . Chem. Soc., Chem. Commun. 1989, 1700. (c) Nakano, H.;
Kumagai, N.; Kabuto, C.; Matsuzaki, H.; Hongo, H. Tetrahedron:
Asymmetry 1995, 6, 1233. (d) Soai, K.; Ookawa, A.; Ogawa, K.; Kaba,
K. J . Chem. Soc., Chem. Commun. 1987, 467 (e) Urabe, T.; Ushio, H.;
Sato, F. Tetrahedron: Asymmetry 1992, 3, 5, (f) Noyori, R.; Suga, S.;
Kawai, K.; Okada, S.; Oguni, N.; Hayashi, M.; Kaneko, T.; Matsuda,
Y. J . Organomet. Chem. 1990, 382, 19. (g) Corey, E. J .; Hannon, F. J .
Tetrahedron Lett. 1987, 28, 5233. (h) Chaloner, P. A.; Perea, S. A. R.
Tetrahedron Lett. 1987, 28, 3013. (i) Hanyu, N.; Aoki, T.; Mino, T.;
Sakamoto, M.; Fujita, T. Tetrahedron: Asymmetry 2000, 11, 4127. (j)
Dangel, B. D.; Polt, R. Org. Lett. 2000, 2, 3003. (k) Bolm, C.; Zenhnder,
M.; Bur, D. Angew. Chem., Int. Ed. Engl. 1990, 29, 205. (l) Chrisman,
W.; Camara, J . N.; Marcellini, K.; Singaram, B. Goralski, C. T.; Hasha,
D. L.; Rudolf, P. R.; Nicholson, L. W.; Borodychuck, K. K. Tetrahe-
dron Lett. 2001, 42, 5805. (m) Pinho, P.; Anderson, P. G. Tetrahedron
2001, 57, 1615 (n) Coob, A. J . A.; Marson, C. M. Tetrahedron:
Asymmetry 2001, 12, 1547, (o) Vidal-Ferran, A.; Moyano, A.; Pericas,
M. A.; Riera, A. J . Org. Chem. 1997, 62, 4970. (p) J uanes, O.;
Rodr´ıguez-Ubis, J . C.; Brunet, E.; Pennemann, H.; Kossenjans, M.;
Martens, J . Eur. J . Org. Chem. 1999, 3323. (q) Palmieri, G. Tetrahe-
dron: Asymmetry 2000, 11, 3361. (r) Wolf, C.; Francis, C. J .; Hawes,
P. A.; Shah, M. Tetrahedron: Asymmetry 2002, 13, 1733. (s) Lu, J .;
Xu, X.; Wang, C.; He, J .; Hu, Y.; Hu, H. Tetrahedron Lett. 2002, 43,
8367.
† Universidad Auto´noma de Puebla.
‡ Centro de Investigacio´n y de Estudios Avanzados del Instituto
Polite´cnico Nacional.
§ Universidad de las Ame´ricas-Puebla.
(1) For the first studies, see: (a) Mukaiyama, T.; Soai, K.; Sato, T.;
Shimizu, H.; Suzuki, K. J . Am. Chem. Soc. 1979, 101, 1455. (b) Oguni,
N.; Omi, T. Tetrahedron Lett. 1984, 25, 2823.
(2) Leading references: (1) Noyori, R.; Kitamura, M. Angew. Chem.,
Int. Ed. Engl. 1991, 30, 49. (b) Soai, K.; Niwa, S. Chem. Rev. 1992, 92,
833. (c) Noyori, R. Asymmetric Catalysis in Organic Synthesis; Wiley:
New York, 1994.
(3) Recent reviews: (a) Pu, L.; Yu, H.-B. Chem. Rev. 2001, 101, 757.
(b) Soai, K.; Shibata, T. In Comprehensive Asymmetric Catalysis;
J acobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin, 1999;
pp 911-922.
(5) Yamakawa, M.; Noyori, R. J . Am. Chem. Soc. 1995, 117,
6327.
(6) (a) Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.; Riera, A.
Tetrahedron Lett. 1997, 38, 8773. (b) Rasmussen, T.; Norrby, P.-O. J .
Am. Chem. Soc. 2001, 123, 2464.
10.1021/jo026811y CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/15/2003
J . Org. Chem. 2003, 68, 2369-2375
2369

Sosa-Rivadeneyra et al.
TABLE 1. En a n tioselective Ad d ition of Et 2Zn to
Ben za ld eh yd e in th e P r esen ce of Ch ir a l Liga n d s 1
(Sch em e 2)
additive
(equiv)
yieldc eed
(%)
major
(%) enantiomere
entrya
ligandb
1
2
3
4
5
6
7
8
(S,S,S)-1
(S,S,S)-1
(S,S,S)-1-Li2
(S,S,S)-1-Li2 LiCl (0.6)
(R,R,S)-1
(R,R,S)-1
(R,R,S)-1-Li2
(R,R,S)-1-Li2 LiCl (0.6)
58
50
80
58
80
65
78
55
57
53
47
24
50
52
35
22
(R)
(R)
(R)
(S)
(S)
(S)
(S)
(S)
LiCl (0.6)
LiCl (0.6)
F IGURE 1. Catalytic cycle proposed by Yamakawa and
Noyori5 for the reaction of dimethylzinc and formaldehyde,
under 2-aminoethanol activation.
a
All reactions were carried out in toluene-hexane (2:1), at 25
°C for 20 h.14 Dilithiated ligand obtained by metalation of the
b
â-amino alcohol ligand with 2 equiv of BuLi. c Isolated yield after
purification by flash chromatography (petroleum ether-ethyl
SCHEME 1
d
acetate (15:1). Determined by HPLC using
a Chiracel OD
column. e The absolute configuration of the 1-phenyl-1-propanol
was assigned from the sign of the specific optical rotation17 and
from the elution order in HPLC analysis on
stationary phase column.18
a Chiracel OD
â-amino alcohol (S,S,S)-1 and according to the standard
procedure,14 carbinol (R)-2 was obtained in 58% isolated
yield and with 57% enantiomeric excess (entry 1 in Table
1). Similarly, benzaldehyde reacted with diethylzinc in
the presence of chiral ligand (R,R,S)-1 to give carbinol
(S)-2 in 80% yield and 50% ee (entry 5, Table 1). Thus, it
is apparent that the N-R-phenylethyl substituent does
not have a significant influence on the stereoselectivity
of the organometallic addition. This result is in line with
the empirical rule that the configuration of the stereo-
genic center supporting the hydroxy group, C*(O), is more
SCHEME 2
In this context, Noyori et al.7b and others9 have shown
that the profile of asymmetric catalysis can be affected
by molecular interaction of the chiral catalyst with
achiral molecules present in the reaction medium, so that
unusual phenomena may be seen. We were particularly
interested in the determination of potential “inert” salt
effects10 on the stereochemical outcome of the addition
of Et2Zn to benzaldehyde catalyzed by N-[(S)-R-phenyl-
ethyl]-â-aminocyclohexanols (R,R,S)-1 and (S,S,S)-1
(Scheme 2).11-13
relevant for the stereochemical outcome of the alkylation
15
reaction than the configuration of the â-carbon, C*(N)
.
Consequently, it is appreciated that stereoinduction by
C*(O) is unlike;16 i.e., the configuration of the predominant
product is opposite to the one present at C*(O) in the chiral
â-amino alcohol ligand.
Comparison of entries 1 and 2, and 5 and 6 in Table 1
shows that addition of LiCl to the reactions activated
with free â-amino alcohols does not have any signifi-
cant effect on chemical yield or enantiomeric ratio of
carbinol 2.
Some years ago, Corey and Hannon19 reported the
improved enantioselectivity achieved by lithium salt
derivatives of chiral â-amino alcohol as ligands in the
addition of Et2Zn to aldehydes. Motivated by these obser-
vations,20 we carried out two experiments (entries 3 and
Resu lts a n d Discu ssion
When benzaldehyde was treated with diethylzinc (2.0
equiv), in the presence of 6.0 mol % (0.06 equiv) of
(7) (a) Yamakawa, M.; Noyori, R. Organometallics 1999, 18, 128.
(b) See, also: Kitamura, M.; Suga, S.; Oka, H.; Noyori, R. J . Am. Chem.
Soc. 1998, 120, 9800.
(8) (a) Goldfuss, B.; Houk, K. N. J . Org. Chem. 1998, 63, 8998. (b)
Goldfuss, B.; Steigelmann, M.; Khan, S. I.; Houk, K. N. J . Org. Chem.
2000, 65, 77.
(9) (a) Costa, A. M.; J imeno, C.; Gavenonis, J .; Carroll, P. J .; Walsh,
P. J . J . Am. Chem. Soc. 2002, 124, 6929. (b) Vogl, E. M.; Groger, H.;
Shibasaki, M. Angew. Chem., Int. Ed. 1999, 38, 1570. (c) Kobayashi,
S.; Ishitani, H. J . Am. Chem. Soc. 1994, 116, 4083.
(13) For reviews on the use of (R)- and (S)-R-phenylethylamine in
asymmetric synthesis, see: (a) J uaristi, E.; Escalante, J .; Leo´n-Romo,
J . L.; Reyes, A. Tetrahedron: Asymmetry 1998, 9, 715. (b) J uaristi, E.;
Leo´n-Romo, J . L.; Reyes, A.; Escalante, J . Tetrahedron: Asymmetry
1999, 10, 2441.
(10) (a) Loupy, A.; Tchoubar, B.; Astruc, D. Chem. Rev. 1992, 92,
1141. (b) Seebach, D.; Beck, A. K.; Studer, A. Modern Synthetic Methods
1995; VCH: Weinheim, 1995; pp 1-178. (c) See, also: J uaristi, E.;
Beck, A. K.; Hansen, J .; Matt, T.; Mukhopadhyay, T.; Simson, M.;
Seebach, D. Synthesis 1993, 1271.
(11) For the preparation of chiral â-amino alcohols (R,R,S)-1 and
(S,S,S)-1, see: (a) Anaya de Parrodi, C.; J uaristi, E.; Quintero-Corte´s,
L. An. Quı´m., Int. Ed. 1996, 92, 400. See, also: (b) Overman, L. E.;
Sugai, S. J . Org. Chem. 1985, 50, 4154. (c) Pracejus, H.; Pracejus, G.;
Costisella, B. J . Prakt. Chem. 1987, 329, 235. (d) Barbaro, P.; Bianchini,
C.; Sernau, V. Tetrahedron: Asymmetry 1996, 7, 843.
(12) For a preliminary account of this work, see: Anaya de Parrodi,
C.; J uaristi, E.; Quintero-Corte´s, L.; Amador, P. Tetrahedron: Asym-
metry 1996, 7, 1915.
(14) The standard procedure described by Soai and Niwa was
followed: Soai, K.; Niwa, S. J . Chem. Soc., Perkin Trans. 1 1991, 2717.
(15) Kitamura, M.; Okada, S.; Suga, S.; Noyori, R. J . Am. Chem.
Soc. 1989, 111, 4028.
(16) For a discussion on the like/unlike stereochemical descriptors,
see: (a) Seebach, D.; Prelog, V. Angew. Chem., Int. Ed. Engl. 1982,
21, 654. (b) See, also: J uaristi, E. Introduction to Stereochemistry and
Conformational Analysis; Wiley: New York, 1991; pp 52-54.
(17) Pickard, R. H.; Kenyon, J . J . Chem. Soc. 1914, 1115.
(18) Soai, K.; Ohno, Y.; Inoue, Y.; Tsuruoka, T.; Hirose, Y. Recl. Trav.
Chim. Pays-Bas 1995, 114, 145.
(19) (a) Corey, E. J .; Hannon, F. J . Tetrahedron Lett. 1987, 28, 5233.
(b) See, also: Soai, K.; Ookawa, A.; Kaba, T.; Ogawa, K. J . Am. Chem.
Soc. 1987, 109, 7111.
2370 J . Org. Chem., Vol. 68, No. 6, 2003

Salt Effects on Enantioselectivity
F IGURE 2. (a) BP86/DN** transition state structure for anti-(4). (b) BP86/DN** transition state structure for syn-(4) (ab initio
results5 are in parentheses).
TABLE 2. Im a gin a r y F r equ en cies (in cm -1) Ca lcu la ted
7, Table 1) where (S,S,S)-1 and (R,R,S)-1 were treated
for Tr a n sition Str u ctu r es F -H
transition structure ab initio value5 BP86/DN** (this work)
with 2 equiv of BuLi, so that lithium salts (S,S,S)-1-Li2
and (R,R,S)-1-Li2 are the active ligands. It is important
to emphasize that Corey and Soai et al.19 use the “mono”
lithium salt (alkoxide) of â-tert-amino alcohol, while the
structure of the catalyst in the present manuscript is the
“di” lithium salt of â-“sec”-amino alcohol. Unfortunately,
the ee of product 1-phenylpropanol (2) actually decreased
in these two cases (compare entries 1 vs 3 and 5 vs 7 in
Table 1).
anti-(4)
syn-(4)
anti-(6)
343 (i)
308 (i)
659 (i)
374 (i)
329 (i)
685 (i)
H. Agreement is even better between BP86/DN** and ab
initio calculations, as shown in Figure 2.
Furthermore, BP86/DN** calculations were particu-
larly successful in performing the analysis of frequencies
associated to transition structures F -H (Table 2). A
single imaginary frequency was revealed for each transi-
tion structure, corresponding to the stretching frequency
for the C-C bond that is formed upon methyl group
transfer from dimethylzinc to formaldehyde (Table 2).
Having validated BP86/DN**//PM3 as a reliable com-
putational method for the molecular modeling of the
reaction of interest, we proceeded to include chiral
ligands (S,S,S)-1 and (R,R,S)-1, diethylzinc, and prochiral
benzaldehyde in the calculations. It is clear that the sense
and extent of stereoselection will be determined by the
relative stabilities of the rate-determining diastereomeric
transition states.
Scheme 3 summarizes the computed results from
hypothetical equilibrium between diastereomeric transi-
tion structures leading to (R)-2 and (S)-2 under standard
conditions; i.e., free â-amino alcohol ligand in the absence
of LiCl. It is appreciated that with (S,S,S)-1 ligand, alkyl
addition to the pro-(R) face of benzaldehyde is favored
by 1.12 kcal/mol. This calculated ∆E value corresponds
to an anticipated (R)-2:(S)-2 ratio ) 87:13 and ee ) 74%.
Experimentally, (R)-2 is indeed the major product, with
ee ) 57% (entry 1, Table 1).
More intriguing to us was the effect of LiCl addition21
on the reaction of Et2Zn with PhCHO, activated by
(S,S,S)-1-Li2 and (R,R,S)-1-Li2. Most interesting is the
opposite chiral induction observed in entries 3 and 4 in
Table 1. Thus, diethylzinc addition to benzaldehyde
under (S,S,S)-1-Li2 catalysis affords predominantly (R)-
2, with 47% ee, whereas the same reaction in the
presence of LiCl (10 equiv relative to the ligand) gives
carbinol (S)-2 as the main product (24% ee).
To better understand this dramatic salt effect, Noyori’s
catalytic cycle (Figure 1 and refs 5-8) was reexamined
computationally, including ligands (S,S,S)-1 and (R,R,S)-
1, diethylzinc, benzaldehyde, and of course LiCl. In view
of the large size of the system involved, fully ab initio
theoretical methods are unsuitable for the proposed
study. Instead, the semiempirical PM3 method22 was
used to optimize molecular structures of the organome-
tallic complexes, and the hybrid BP/DN** method22 was
employed to determine the corresponding energies.
We first confirmed, as already done by Goldfuss and
Houk,8a that PM3 correctly reproduces Noyori’s ab initio
geometries for the model transition structures F , G, and
(20) See, also: Kitamura, M.; Suga, S.; Kawai, K.; Noyori, R. J . Am.
Chem. Soc. 1986, 108, 6071.
Scheme 3b presents the results with (R,R,S)-1 ligand.
Ethyl addition to the pro-(S) face in benzaldehyde is now
predicted to be more favorable relative to addition to the
(21) For remarkable examples of the consequences of LiCl addition
to enolate alkylation reactions, see: (a) Seebach, D.; Bossler, H. G.;
Flowers, R.; Arnett, E. M. Helv. Chim. Acta 1994, 77, 291. (b) J uaristi,
E.; Leo´n-Romo, J . L.; Ram´ırez-Quiro´s, Y. J . Org. Chem. 1999, 64, 2914.
(c) Reyes, A.; J uaristi, E. Tetrahedron: Asymmetry 2000, 11, 1411. (d)
See, also: J uaristi, E.; D´ıaz, F.; Cue´llar, G.; J ime´nez-Va´zquez, H. A.
J . Org. Chem. 1997, 62, 4029.
(22) PC Spartan-Pro, version 1.0; Wavefunction, Inc.: Irvine, CA,
1999.
J . Org. Chem, Vol. 68, No. 6, 2003 2371

Sosa-Rivadeneyra et al.
SCHEME 4
F IGURE 3. Catalytic cycle proposed in this work for the
reaction of dimethylzinc and formaldehyde, under dilithiated
2-amino alcohol activation. All organometallic structures were
optimized by semiempirical PM3 calculations.
SCHEME 3
SCHEME 5
SCHEME 6
pro-(R) face, with ∆E ) 0.92 kcal/mol, calculated (R)-2:
(S)-2 ratio ) 17:83, and ee ) 66%. These results are in
good agreement with experiment, since the (R,R,S)-1
â-amino alcohol induces the predominant formation of
the (S)-2 carbinol, with 50% ee (Table 1, entry 5).
Also in agreement with the experimental results,
stereoinduction by the N-phenylethyl group is marginal.
Molecular modeling of the alkylation reactions acti-
vated by dilithiated ligands (S,S,S)-1-Li2 and (R,R,S)-1-
Li2 (entries 3 and 7, Table 1) required reexamination of
Noyori’s model catalytic cycle of 2-aminoethanol-promot-
ed reaction of dimethylzinc and formaldehyde5 (Figure
1). Calculations (BP86/DN**//PM3) support the modified
catalytic cycle presented in Figure 3.
Salient features in the catalytic cycle advanced in
Figure 3 are (1) the lowest-energy structures I-M,
presenting ion-triplet configuration23 of the dilithium
salts, and (2) in contrast with the mechanism presented
in Figure 1,5 only 1 equiv of dialkylzinc should be required
for the alkyl addition to take place.
mined by the configurations at the stereogenic centers
in the carbocyclic framework. Unlike16 stereoinduction is
favored, in agreement with the experimental results,
although the extent of both the calculated (Scheme 4) and
experimental (Table 1) enantioselectivity is lower in the
case of the dilithiated ligands.
To understand the effect produced by the presence of
LiCl in the reaction of diethylzinc and benzaldehyde, with
activation by dilithiated amino alcohol (entries 4 and 8,
Table 1), transition structure L (Figure 3) was optimized
following incorporation of one, two, and three molecules
of LiCl. The mixed aggregate of lowest energy at the
BP86/DN**//PM3 level corresponds to N, as shown in
Scheme 5.
Substitution in N by chiral ligands, incorporation of
benzaldehyde, and diethylzinc provided the calculated
(BP86/DN**//PM3) data that is summarized in Scheme
6. Most relevant, the calculations predict that both
transition structures incorporating (S,S,S)-1-Li2 (Scheme
6a) and (R,R,S)-1-Li2 (Scheme 6b) favor alkyl group
Scheme 4 summarizes the computed data for the
diastereomeric transition structures that lead, irrevers-
ibly, to carbinols (R)-2 and (S)-2 under the activation of
dilithiated ligands (S,S,S)-1-Li2 and (R,R,S)-1-Li2. Again,
the sense of stereoinduction is little affected by the N-R-
phenylethyl chiral substituent and seems to be deter-
(23) (a) Streitwieser, A. Acc. Chem. Res. 1984, 17, 353. (b) Gutie´rrez-
Garc´ıa, V. M.; Reyes-Rangel, G.; Mun˜oz-Mun˜iz, O.; J uaristi, E. J . Braz.
Chem. Soc. 2001, 12, 652. (c) Gutie´rrez-Garc´ıa, V. M.; Reyes-Rangel,
G.; Mun˜oz-Mun˜iz, O.; J uaristi, E. Helv. Chim. Acta 2002, 85, 4189.
2372 J . Org. Chem., Vol. 68, No. 6, 2003

Salt Effects on Enantioselectivity
TABLE 4. Com p a r ison of Ca lcu la ted (BP 86/DN**//P M3)
a n d Exp er im en ta l (Ta bles 1 a n d 3) Resu lts
SCHEME 7
er (R:S)
additive
entry
ligand
(equiv)
exptl
calcd
1
2
3
4
5
6
7
8
(S,S,S)-1
78:22
73:27
38:62
25:75
33:67
39:61
82:18
87:13
66:34
28:72
17:83
40:60
24:76
98:2
(S,S,S)-1-Li2
(S,S,S)-1-Li2
(R,R,S)-1
(R,R,S)-1-Li2
(R,R,S)-1-Li2
(S,S,S)-3
LiCl (0.6)
LiCl (0.6)
(R,R,S)-3
4:96
3-7 relative to N-H ligand (S,S,S)-1. N-Ethylated ligand
provided the best results, 85% yield and er ) 87:13.
Con clu sion s
TABLE 3. En a n tioselectivity of Dieth ylzin c Ad d ition to
Ben za ld eh yd e in th e P r esen ce of N-Alk yla ted Liga n d s
(S,S,S)-3-7a
Although many reports in the literature describe
enantioselective additions of diethylzinc to benzaldehyde
using chiral â-amino alcohols, this paper involves inter-
esting new results. Most important, the use of the
dilithium salts of â-amino alcohols (S,S,S)-1 and (S,S,R)-1
is described, and the opposite sense of enantioselectivity
by the addition of lithium chloride (Table 4, entries 2 and
3) is described.
entry
ligand
R
yield (%)
enantiomer ratio (R:S)
BP86/DN**//PM3 theoretical calculations proved re-
markably successful in the reproduction of experimental
results from the reaction of diethylzinc with benzalde-
hyde with activation by chiral â-amino alcohols (S,S,S)-1
and (R,R,S)-1. Molecular modeling of the transition struc-
tures provided ∆E values for the diastereomeric orga-
nozinc complexes that are in good agreement with the
experimentally observed enantioselectivities (Table 4).
Most relevant, modification of Noyori’s catalytic cycle
permitted the rationalization of the empirical data
gathered from reactions carried out under activation by
the lithium salts of (S,S,S)-1 and (R,R,S)-1, with or
without the addition of lithium chloride “inert” salts.
Finally, the calculations were also able to reproduce the
beneficial effect of N-alkylation on enantioselectivity
(Table 4).
1
2
3
4
5
6
(S,S,S)-1
(S,S,S)-3
(S,S,S)-4
(S,S,S)-5
(S,S,S)-6
(S,S,S)-7
H
Me
Et
Pr
Bu
Pen
65
81
85
80
90
82
78:22
82:18
87:13
88:12
82:12
84:16
a
All reactions were carried out in toluene, at 25 °C for 20 h.
Molar ratio, Et2Zn/PhCHO/ligand ) 2.03:1.00:0.06.
addition on the pro-(S) face of benzaldehyde, although
enantioselection is anticipated to be low. Both of these
predictions are in line with the experimental results
(entries 4 and 8, Table 1).
Steric bulk around the nitrogen atom in chiral â-amino
alcohol ligands generally enhances the enantioselectivity
of Et2Zn addition to benzaldehyde.4a,24 As can be seen in
Scheme 7, this empirical observation is reproduced by
our BP86/DN**//PM3 calculations incorporating N-meth-
yl derivatives (S,S,S)-3 and (R,R,S)-3. The former ligand
induces preferential ethyl group addition on the pro-R
face of benzaldehyde, leading to the formation of (R)-2
with a calculated ee ) 96%. By comparison, N-H ligand
(S,S,S)-1 is predicted to give (R)-2 in a lower ee ) 74%
(Scheme 3). Similarly, according to the calculations,
N-CH3 ligand (R,R,S)-3 favors formation of carbinol (S)-2
by 1.97 kcal/mol; i.e., ee ) 92% (Scheme 7). By compari-
son, N-H ligand is anticipated to furnish (S)-2 in a lower
66% ee (Scheme 3).
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. Flasks, stirring bars,
and hypodermic needles used for the generation and reactions
of organometallic compounds were dried for ca. 12 h at 120
°C and allowed to cool in a desiccator over anhydrous CaSO4.
Anhydrous toluene was obtained by distillation from ben-
zophenone ketyl. N,N′-Dimethylpropyleneurea (DMPU) and
acetonitrile were dried over CaH2 and then distilled at reduced
pressure. The BuLi employed was titrated according to the
method we developed.25 TLC: DC-F254 plates, with detection
by UV light. Flash column chromatography: silica gel (0.040-
0.063 mm). HPLC: instrument fitted with UV-vis detector
and a chiral stationary phase of Chiralcel OD for the deter-
mination of the enantiomeric ratios. Melting points are not
corrected. 1H NMR spectra: 400 MHz. 13C NMR spectra: 100
MHz. Chemical shifts (δ) in ppm downfield from internal TMS
reference; the coupling constants (J ) are given in Hz. Elemen-
tal analyses were obtained from Galbraith Laboratories, Inc.,
Knoxville, TN, and Service of Microanalyses, I.C.S.N.-C.N.R.S.
Intrigued by this remarkable prediction, we prepared
N-alkylated â-amino alcohols (S,S,S)-3-7, whose ef-
fectiveness as chiral activators in the enantioselective
addition of diethylzinc to benzaldehyde was then exam-
ined (Table 3). It is seen that both the catalytic efficiency
and the enantioselectivity are higher with ligands (S,S,S)-
(24) Soai, K.; Yokoyama, S.; Hayasaka, T. J . Org. Chem. 1991, 56,
4264.
(25) J uaristi, E.; Mart´ınez-Richa, A.; Garc´ıa-Rivera, A.; Cruz-
Sa´nchez, J . S. J . Org. Chem. 1983, 48, 2603.
J . Org. Chem, Vol. 68, No. 6, 2003 2373

Sosa-Rivadeneyra et al.
91198, Gif sur Yvette. HRMS were obtained from Instituto de
Qu´ımica, UNAM, Me´xico.
â-amino alcohol (1S,2S,1′S)-1 (0.22 g, 1.0 mmol) was treated
with 0.12 mL (1.5 mmol) of ethyl iodide to furnish 1.0 g (39%
yield) of the desired N-ethyl derivative, [R]28 ) + 87.2 (c )
(1R,2R,1′S)- a n d (1S,2S,1′S)-2-[N-r-P h en yleth yl)-a m i-
n o]cycloh exa n ols [(R,R,S)-1 a n d (S,S,S)-1]. A dry two-
necked flask provided with addition funnel, condenser, and
magnetic stirrer was loaded, with stirring and under argon,
with an equimolar mixture of cyclohexene oxide (0.98 g, 10
mmol) and anhydrous lithium perchlorate (10 mmol) in freshly
dried acetonitrile (10 mL). The reaction mixture was cooled
in an ice-water bath to 0 °C before the dropwise addition of
(S)-R-phenylethylamine (1.2 g, 10 mmol). The resulting solu-
tion was heated to reflux for 18 h. Water (25 mL) was added
to the reaction mixture, and the organic phase was extracted
with CH2Cl2 (3 × 25 mL). The combined organic extracts were
dried over anhydrous Na2SO4, filtered, and concentrated in
vacuo to give 2.1 g (98% yield) of the diastereomeric mixture
of (S,S,S)-1 and (R,R,S)-1, in a 55:45 ratio. This crude mixture
was redissolved in dry ethanol (40 mL) and treated with
anhydrous oxalic acid (0.22 g, 0.25 equiv), and the resulting
solution was heated to reflux for 1 h. Upon standing, the
oxalate salt derived from (S,S,S)-1 precipitated, whereas
(R,R,S)-1 remained in solution. The precipitate was washed
several times with hot ethanol until no more â-amino alcohol
(R,R,S)-1 was detected by TLC. The combined organic layers
were treated at room temperature with anhydrous oxalic acid
(0.22 g, 0.25 equiv) to precipitate the oxalate salt of diastere-
omer (R,R,S)-1. Each oxalate was hydrolyzed with aqueous
7% K2CO3 to give the free â-amino alcohol.
D
1
3.1, CHCl3). H NMR (400 MHz, CDCl3) δ 1.01 (t, J ) 7.0 Hz,
3H), 1.20 (m, 4H), 1.38 (d, J ) 7.0 Hz, 3H), 1.70 (m, 3H), 2.05
(m, 1H), 2.41 (m, 1H), 2.65 (m, 2H), 3.28 (m, 1H), 3.58 (bs,
1H), 4.06 (q, J ) 7.0 Hz, 1H), 7.31-7.53 (m, 5H). 13C NMR
(100 MHz, CDCl3) δ 15.7, 16.9, 24.3, 26.0, 26.5, 33.4, 39.5, 56.7,
63.6, 69.7, 126.9, 127.5, 128.4, 145.1. HMRS (FAB) m/z calcd
for C16H25NO 247.1936 [M + H]+, found 247.1929.
(1S,2S,1′S)-2-N-P r op yl-[N-(r-p h en yleth yl)a m in o]cyclo-
h exa n ol [(1S,2S,1′S)-5]. According to the general procedure,
â-amino alcohol (1S,2S,1′S)-1 (0.22 g, 1.0 mmol) was treated
with 0.15 mL (1.5 mmol) of propyl iodide to furnish 0.11 g (41%
yield) of the desired N-propyl derivative, [R]28 ) + 96.2 (c )
D
1
2.1, CHCl3). H NMR (400 MHz, CDCl3) δ 0.77 (t, J ) 7.3 Hz,
3H), 1.17 (m, 5H), 1.37 (d, J ) 6.9 Hz, 3H), 1.45 (m, 1H), 1.65
(m, 1H), 1.75 (m, 2H), 2.06 (m, 1H), 2.35 (m, 1H), 2.53 (m,
2H), 3.26 (m, 1H), 3.58 (bs, 1H), 4.04 (q, J ) 6.9 Hz, 1H), 7.30-
7.50 (m, 5H). 13C NMR (100 MHz, CDCl3) δ 11.8, 16.2, 23.1,
24.3, 26.0, 26.7, 33.4, 47.7, 56.9, 63.7, 69.7, 127.0, 127.7, 128.4,
144.9. HMRS (FAB) m/z calcd for C17H27NO 261.2093 [M +
H]+, found 261.2089.
(1S,2S,1′S)-2-N-Bu tyl-[N-(r-p h en yleth yl)a m in o]cyclo-
h exa n ol [(1S,2S,1′S)-6]. According to the general procedure,
â-amino alcohol (1S,2S,1′S)-1 (0.22 g, 1.0 mmol) was treated
with 0.16 mL (1.5 mmol) of butyl bromide to furnish 0.09 g
(32% yield) of the desired N-butyl derivative, [R]28 ) +87.7
D
(c ) 2.0, CHCl3). 1H NMR (400 MHz, CDCl3) δ 0.83 (t, J ) 7.0
Hz, 3H), 1.22 (m, 8H), 1.39 (d, J ) 7.0 Hz, 3H), 1.64 (m, 1H),
1.74 (m, 2H), 2.06 (m, 1H), 2.35 (m, 1H), 2.56 (m, 2H), 3.26
(m, 1H), 3.56 (bs, 1H), 4.05 (q, J ) 7.0 Hz, 1H), 7.31-7.47 (m,
5H). 13C NMR (100 MHz, CDCl3) δ 14.1, 16.3, 20.6, 24.3, 26.0,
26.7, 32.3, 33.4, 45.5, 56.9, 63.7, 69.7, 126.9, 127.6, 128.4, 144.9.
HMRS (FAB) m/z calcd for C18H29NO 275.2249 [M + H]+,
found 275.2250.
(1R,2R,1′S)-1. The oxalate salt, a white powder, mp 199-
201 °C, gave the free â-amino alcohol (1.2 g, 54% yield) as
white crystals, mp 65-66 °C (lit.11c mp 51-56 °C. The
chlorhydrate salt was prepared by addition of HCl in ether,
followed by recrystallization from CH2Cl2, mp 282-283 °C;
[R]28 ) -69.4 (c ) 1, EtOH). Lit.11c mp 284-286 °C; [R]28
)
D
-69.8 (c ) 1, EtOH). 1H NMR, see ref 11b. 13C NMR (50 MHz,
CDCl3) δ 24.7, 25.5, 26.2, 30.0, 33.5, 54.4, 60.4, 74.6, 127.2,
127.5, 129.0, 145.0.
(1S,2S,1′S)-2-N-P en tyl-[N-(r-p h en yleth yl)a m in o]cyclo-
h exa n ol [(1S,2S,1′S)-7]. According to the general procedure,
â-amino alcohol (1S,2S,1′S)-1 (0.22 g, 1.0 mmol) was treated
with 0.19 mL (1.5 mmol) of pentyl bromide to furnish 0.09 g
(1S,2S,1′S)-1. The oxalate salt, a white powder, mp 234-
235 °C, gave the free â-amino alcohol (0.92 g, 43% yield) as a
colorless oil (lit.11c mp 25 °C). The chlorhydrate salt was
prepared by treatment with HCl in ether, followed by recrys-
(30% yield) of the desired N-pentyl derivative, [R]28 ) +72.9
D
tallyzation from CH2Cl2, mp 206-207 °C; [R]28 ) -11.7 (c )
(c ) 1.9, CHCl3). 1H NMR (400 MHZ, CDCl3) δ 0.83 (t, J ) 7.0
Hz, 3H), 1.20 (m, 11H), 1.38 (d, J ) 7.0 Hz, 3H), 1.64 (m, 1H),
1.75 (m, 2H), 2.07 (m, 1H), 2.36 (m, 1H), 2.54 (m, 2H), 3.27
(m, 1H), 4.05 (q, J ) 7.0 Hz, 1H), 7.02-7.33 (m, 5H). 13C NMR
(100 MHz, CDCl3) δ 14.2, 16.3, 22.6, 24.3, 26.0, 26.6, 29.6, 29.8,
33.4, 45.7, 57.0, 63.7, 69.7, 126.9, 127.6, 128.4, 144.9. Anal.
Calcd for C19H31NO: C, 78.84; H, 10.78. Found: C, 78.86; H,
10.44.
D
1, EtOH). Lit.11c mp 207 °C; [R]28 ) -11.5 (c ) 1, EtOH). H
1
D
NMR, see ref 11b. 13C NMR (50 MHz, CDCl3) δ 24.0, 24.8, 25.9,
31.8, 33.5, 55.7, 62.0, 74.6, 126.9, 127.6, 129.0, 142.1.
Gen er a l P r oced u r e for N-Alk yla tion of (1S,2S,1′S)-1.26
The â-amino alcohol (1.0 mmol) was dissolved in 10 mL
of DMPU-EtOH (1:10) before the addition of 2.0 equiv of
Na2CO3. To the resulting suspension was then added 1.5 equiv
of the alkylating agent, and the reaction mixture was heated
to reflux for 24 h. The reaction flask was brought to ambient
temperature before dilution with 10 mL of water, and the
crude product was extracted with three 10-mL portions of
CH2Cl2. The combined organic layers were dried with anhy-
drous Na2SO4, filtered, and evaporated. Final purification was
achieved by flash chromatography (petroleum ether).
Gen er a l P r oced u r e for Rea ction of Dieth ylzin c w ith
Ben za ld eh yd e in th e P r esen ce of Liga n d s 1, 3-7. To an
ice-cooled solution of chiral ligand (0.184 mmol) in dry toluene
(3 mL) was added Et2Zn (6.2 mmol, 6.2 mL of 1 M hexane
solution) over a period of 5 min. The mixture was stirred at
room temperature for 30 min, and benzaldehyde (0.32 mL, 3.05
mmol) was added at 0 °C. The reaction mixture was stirred
for 20 h at room temperature, and 1 M HCl was added to
quench the reaction. The mixture was extracted with CH2Cl2,
the organic extract was dried with Na2SO4, and the solvent
was evaporated under reduced pressure. The crude product
was purified by silica gel column chromatography using
hexane and ethyl acetate (15:1) as eluent. The product was
characterized by 1H NMR, and the enantiomeric excess was
determined by HPLC (Chiralcel OD, hexane/2-propanol (97:
3) as eluent, 0.5 mL/min flow rate).
Gen er a l P r oced u r e for Rea ction of Dieth ylzin c w ith
Ben za ld eh yd e in th e P r esen ce of Dilith ia ted Liga n d s.
To an ice-cooled solution of chiral ligand (0.184 mmol) in dry
toluene (3 mL) was added butyllithium (0.368 mmol, 0.245 mL
of 1.5 M hexane solution). After 10 min of stirring, Et2Zn (6.2
mmol, 6.2 mL of 1 M hexane solution) was added over a period
of 5 min. The mixture was stirred at room temperature for 30
(1S,2S,1′S)-2-N-Meth yl-[N-(r-p h en yleth yl)a m in o]cyclo-
h exa n ol [(1S,2S,1′S)-3]. According to the general procedure,
â-amino alcohol (1S,2S,1′S)-1 (0.22 g, 1.0 mmol) was treated
with 0.09 mL (1.5 mmol) of methyl iodide to furnish 0.09 g
(40% yield) of the desired N-methyl derivative, [R]28D ) + 19.0
1
(c ) 1.37, CHCl3). H NMR (400 MHz, CDCl3) δ 1.20 (m, 5H),
1.36 (d, J ) 6.6 Hz, 3H), 1.59 (m, 1H), 1.72 (m, 2H), 2.04 (s,
3H), 2.16 (m, 1H), 2.66 (m, 1H), 3.40 (m, 1H), 3.68 (q, J ) 6.6
Hz, 1H), 7.23-7.31 (m, 5H). 13C NMR (100 MHz, CDCl3) δ 21.1,
22.7, 24.3, 25.6, 32.7, 33.4, 62.1, 64.8, 69.1, 127.1, 127.3, 128.5,
145.4. Anal. Calcd for C15H23NO: C, 77.25; H, 9.87. Found:
C, 77.48; H, 9.96.
(1S,2S,1′S)-2-N-E t h yl-[N-(r-p h en ylet h yl)a m in o]cyclo-
h exa n ol [(1S,2S,1′S)-4]. According to the general procedure,
(26) J uaristi, E.; Murer, P.; Seebach, D. Synthesis 1993, 1243.
2374 J . Org. Chem., Vol. 68, No. 6, 2003

Salt Effects on Enantioselectivity
min, and benzaldehyde (0.32 mL, 3.05 mmol) was added at 0
°C. The reaction mixture was stirred for 20 h at room
temperature, and then 1 M HCl was added to quench the
reaction. The mixture was extracted with CH2Cl2, and the
organic extract was dried with Na2SO4 before the solvent was
evaporated under reduced pressure. The crude product was
purified by silica gel column chromatography using hexane
and ethyl acetate (15:1) as eluent. The product was character-
ized by 1H NMR, and the enantiomeric excess was determined
by HPLC (Chiralcel OD, hexane/2-propanol (97:3) as eluent,
flow rate 0.5 mL/min.).
Gen er a l P r oced u r e for Rea ction of Dieth ylzin c w ith
Ben za ld eh yd e in th e P r esen ce of F r ee Liga n d s a n d LiCl.
To an ice-cooled solution of chiral ligand (0.184 mmol) and LiCl
(1.84 mmol, 0.077 g) in dry toluene (3 mL) was added Et2Zn
(6.2 mmol, 6.2 mL of 1 M hexane solution) over a period of 5
min. The mixture was stirred at room temperature for 30 min,
and benzaldehyde (0.32 mL, 3.05 mmol) was added at 0 °C.
The reaction mixture was stirred for 20 h at room temperature,
and then 1 M HCl was added to quench the reaction. The
mixture was extracted with CH2Cl2, and the organic extract
was dried with Na2SO4 before the solvent was evaporated
under reduced pressure. The crude product was purified by
silica gel column chromatography using hexane and ethyl
Gen er a l P r oced u r e for Rea ction of Dieth ylzin c w ith
Ben za ld eh yd e in t h e P r esen ce of Dilit h ia t ed Liga n d s
a n d LiCl. To an ice-cooled solution of chiral ligand (0.184
mmol) and LiCl (1.84 mmol, 0.08 g) in dry toluene (3 mL) was
added n-butyllithium (0.368 mmol, 0.245 mL of 1.5 M hexane
solution). After 10 min, Et2Zn (6.2 mmol, 6.2 mL of 1 M hexane
solution) was added over a period of 5 min. The mixture was
stirred at room temperature for 30 min, and benzaldehyde
(0.32 mL, 3.05 mmol) was added at 0 °C. The reaction mixture
was stirred for 20 h at room temperature, and then 1 M HCl
was added to quench the reaction. The mixture was extracted
with CH2Cl2, and the organic extract was dried with Na2SO4,
before the solvent was evaporated under reduced pressure.
The crude product was purified by silica gel column chroma-
tography using hexane and ethyl acetate (15:1) as eluent.
1
acetate (15:1) as eluent. The product was characterized by H
NMR, and the enantiomeric excess was determined by HPLC
(Chiralcel OD, hexane/2-propanol (97:3) as eluent, flow rate
0.5 mL/min.).
Ack n ow led gm en t. We are indebted to Conacyt for
financial support via grants 32200-E, 32202-E, and
33023-E. We are grateful to the reviewers for their
useful suggestions.
Su p p or tin g In for m a tion Ava ila ble: Computational re-
sults (Cartesian coordinates for intermediates and transition
states) at BP/DN** level of theory. This material is available
free of charge via the Internet at http://pubs.acs.org.
1
The product was characterized by H NMR, and the enantio-
meric excess was determined by HPLC (Chiralcel OD, hexane/
2-propanol (97:3) as eluent, flow rate 0.5 mL/min.).
J O026811Y
J . Org. Chem, Vol. 68, No. 6, 2003 2375