Investigation of the Effects of the Structure and Chelate Size of Bis-oxazoline Ligands in the Asymmetric Copper-Catalyzed Cyclopropanation of Olefins: Design of a New Class of Ligands

Article abstract of DOI:10.1021/jo962021a

A set of novel, C₂-symmetric bis-oxazoline ligands has been synthesized by mounting two oxazoline rings onto an optically active 1,3-dioxolane backbone. This design allows for the control of both orientation as well as the proximity of the oxazolinyl R-groups around the reactive site. As a result of the twist imparted by the 1,3-dioxolane ring, the stereogenic oxazolinyl substituents can be brought either toward or away from the complexed metal in a controllable fashion. Starting from L-amino alcohols and either L- or D-tartaric acid, two sets of ligands (6b-e and 7a,b) were synthesized and evaluated in the copper-catalyzed cyclopropanation of olefins. The comparison of benzyl and isopropyl derivatives of these ligands with previously reported five- and six-membered bis-oxazolines clearly indicates the beneficiary effect of the larger chelate size and the chiral tether of the tartrate-derived ligand. The effect of the different oxazolinyl groups along with the different substituents on the dioxolane tethers was also investigated. The influence of the alkyl group of the diazoacetate was studied, and the diazoacetate derived from (-)-8-phenylmenthol was found to be superior to (-)-menthyl diazoacetate. The cyclopropanation of vinyl acetate, a relatively unexplored substrate for this reaction, furnished cyclopropanol derivatives in good optical purity.

Full text of DOI:10.1021/jo962021a

2518
J . Org. Chem. 1997, 62, 2518-2526
In vestiga tion of th e Effects of th e Str u ctu r e a n d Ch ela te Size of
Bis-oxa zolin e Liga n d s in th e Asym m etr ic Cop p er -Ca ta lyzed
Cyclop r op a n a tion of Olefin s: Design of a New Cla ss of Liga n d s
Ashutosh V. Bedekar, Elise B. Koroleva, and Pher G. Andersson*
Department of Organic Chemistry, Institute of Chemistry, Uppsala University, Box 531,
S-75121 Uppsala, Sweden
Received October 29, 1996X
A set of novel, C2-symmetric bis-oxazoline ligands has been synthesized by mounting two oxazoline
rings onto an optically active 1,3-dioxolane backbone. This design allows for the control of both
orientation as well as the proximity of the oxazolinyl R-groups around the reactive site. As a result
of the twist imparted by the 1,3-dioxolane ring, the stereogenic oxazolinyl substituents can be
brought either toward or away from the complexed metal in a controllable fashion. Starting from
L-amino alcohols and either L- or D-tartaric acid, two sets of ligands (6b-e and 7a ,b) were
synthesized and evaluated in the copper-catalyzed cyclopropanation of olefins. The comparison of
benzyl and isopropyl derivatives of these ligands with previously reported five- and six-membered
bis-oxazolines clearly indicates the beneficiary effect of the larger chelate size and the chiral tether
of the tartrate-derived ligand. The effect of the different oxazolinyl groups along with the different
substituents on the dioxolane tethers was also investigated. The influence of the alkyl group of
the diazoacetate was studied, and the diazoacetate derived from (-)-8-phenylmenthol was found
to be superior to (-)-menthyl diazoacetate. The cyclopropanation of vinyl acetate, a relatively
unexplored substrate for this reaction, furnished cyclopropanol derivatives in good optical purity.
In tr od u ction
tions, and recently, J ørgensen6f reported on the Cu-
catalyzed hetero Diels-Alder reactions of glyoxylates.
Helmchen et al.7 were the first to report on the Rh-
catalyzed hydrosilylation of ketones with these ligands.
Addition of alkyllithium reagents to imines8 and Cu-
catalyzed allylic oxidation9 are other recent applications
of bis-oxazolines. Addition of trimethylsilyl cyanide10 and
silylketene acetal11 to aldehydes are other examples
where excellent enantioselectivities have been achieved.
Porter and co-workers12 used bis-oxazolines as chiral
auxiliaries in their enantioselective free-radical-initiated
addition of allyltributylstannane. The schematic repre-
sentation of the wide applications of bis-oxazolines is
presented in Figure 1. However, the asymmetric cyclo-
propanation of olefins is by far the most successful
application of the Cu(I)/bis-oxazoline catalytic system.
The C2-symmetric chiral bis-oxazolines have emerged
as an efficient class of ligands in an increasing number
of asymmetric transformations over the last few years.1
Since the first use of the structurally similar semicorrins
in asymmetric cyclopropanation of olefins by Pfaltz,1a the
bis-oxazoline unit has been found to be an excellent chiral
ligand for numerous asymmetric reactions. Pfaltz showed
its utility in a number of applications,2a which in addition
to cyclopropanation also included Ir-catalyzed transhy-
drogenation of ketones and Pd-catalyzed nucleophilic
allylic substitution.3 Masamune2c,d and Evans2e,f further
developed the Cu-catalyzed asymmetric cyclopropanation
of olefins into a highly enantioselective process. J acob-
sen4 recently introduced Cu-catalyzed aziridination of
imines in the presence of bis-oxazolines with moderate
enantioselectivity. Evans5 demonstrated the asymmetric
version of Cu-catalyzed aziridination of olefins with [N-(p-
toluenesulfonyl)imino]phenyliodinane in the presence of
bis-oxazolines. Corey6a,b and Evans6c,d,h utilized these
ligands with remarkable success for the Mg-, Cu-, and
Fe-catalyzed asymmetric Diels-Alder cycloaddition reac-
The conformationally rigid framework of the metal
chelate with the presence of stereocenters close to the
donor nitrogen atoms imposes a well-ordered chiral
(6) Diels-Alder reactions: (a) Corey, E. J .; Imai, N.; Zhang, H.-Y.
J . Am. Chem. Soc. 1991, 113, 728. (b) Corey, E. J .; Ishihara, K.
Tetrahedron Lett. 1992, 33, 6807. (c) Evans, D. A.; Miller, S. J .; Leckta,
T. J . Am. Chem. Soc. 1993, 115, 6460. (d) Evans, D. A.; Murry, J . A.;
von Matt, P.; Norcross, R. D.; Miller, S. J . Angew. Chem., Int. Ed. Engl.
1995, 34, 798. (e) Kagan, H. B.; Riant, O. Chem. Rev. 1992, 92, 1007.
(f) J ohannsen, M.; J ørgensen, K. A. J . Org. Chem. 1995, 60, 5757. (g)
Desimoni, G.; Faita, G.; Righetti, P. P. Tetrahedron Lett. 1996, 37,
3027. (h) Evans, D. A.; Kozlowski, M. C.; Tedrow, J . S. Tetrahedron
Lett. 1996, 37, 7481.
(7) Hydrosilylation of ketones: Helmchen, G.; Krotz, A.; Ganz, K.-
T.; Hansen, D. Synlett 1991, 257.
(8) Addition of organolithium reagents to imines: (a) Denmark, S.
E.; Nakajima, N.; Nicaise, O. J .-C. J . Am. Chem. Soc. 1994, 116, 8797.
(b) Denmark, S. E.; Nakajima, N.; Nicaise, O. J .-C.; Faucher, A. M.;
Edwards, J . P. J . Org. Chem. 1995, 60, 4884.
(9) Allylic oxidation of olefins: (a) Gokhale, A. S.; Minidis, A. B. E.;
Pfaltz, A. Tetrahedron Lett. 1995, 36, 1831. (b) Andrus, M. B.; Argade,
A. B.; Chen, X.; Pamment, M. G. Tetrahedron Lett. 1995, 36, 2945.
(10) Addition of trimethylsilyl cyanide to aldehydes: Corey, E. J .;
Wang, Z. Tetrahedron Lett. 1993, 34, 4001.
(11) Aldol additions to aldehydes: Evans, D. A.; Murry, J . A.;
Kozlowski, M. C. J . Am. Chem. Soc. 1996, 118, 5814.
X Abstract published in Advance ACS Abstracts, March 1, 1997.
(1) (a) Pfaltz, A. Acc. Chem. Res. 1993, 26, 339. (b) Bolm, C. Angew.
Chem., Int. Ed. Engl. 1991, 30, 542. (c) Pfaltz, A. Acta. Chem. Scand.
1996, 50, 189.
(2) For various bis-oxazoline-mediated organic syntheses see the
following for cyclopropanation of olefins: (a) Mu¨ller, D.; Umbricht, G.;
Weber, B.; Pfaltz, A. Helv. Chim. Acta 1991, 74, 232. (b) Reiser, O.
Angew. Chem., Int. Ed. Engl. 1993, 32, 547. (c) Lowenthal, R. E.; Abiko,
A.; Masamune, S. Tetrahedron Lett. 1990, 31, 6005. (d) Lowenthal, R.
E.; Masamune, S. Tetrahedron Lett. 1991, 32, 7373. (e) Evans, D. A.;
Woerpel, K. A.; Hinman, M. M.; Faul, M. M. J . Am. Chem. Soc. 1991,
113, 726. (f) Evans, D. A.; Woerpel, K. A.; Scott, M. J . Angew. Chem.,
Int. Ed. Engl. 1992, 31, 430. (g) Tokunoh, R.; Tomiyama, H.; Sodeoka,
M.; Shibasaki, M. Tetrahedron Lett. 1996, 37, 2449.
(3) Nucleophilic allylic substitution: von Matt, P.; Pfaltz, A. Angew.
Chem., Int. Ed. Engl. 1993, 32, 566.
(4) Aziridination of imines: Hansen, K. B.; Finney, N. S.; J acobsen,
E. N. Angew. Chem., Int. Ed. Engl. 1995, 34, 676.
(5) Aziridination of olefins: Evans, D. A.; Faul, M. M.; Bilodeau,
M. T.; Anderson, B. A.; Barnes, D. M. J . Am. Chem. Soc. 1993, 115,
5328.
(12) Free-radical addition of alkyl iodides to olefins: Wu, J . H.;
Radinov, R.; Porter, N. D. J . Am. Chem. Soc. 1995, 117, 11029.
S0022-3263(96)02021-X CCC: $14.00 © 1997 American Chemical Society

Cu-Catalyzed Cyclopropanation of Olefins
J . Org. Chem., Vol. 62, No. 8, 1997 2519
F igu r e 3.
F igu r e 1.
F igu r e 4.
F igu r e 2.
The bis-oxazolines of type 1, where the oxazoline rings
are directly connected and thus form a five-membered
chelate with Cu, give trans product with poor enantio-
selectivity2e (3% ee for 1a ; Figure 3). On the other hand,
the ligands in which the oxazoline rings are separated
by one carbon atom, of the type 2 and 3, lead to a six-
membered chelate and have been found to be more
effective in Cu-catalyzed cyclopropanation. Masamune2c
used ligand 2a and obtained the trans product with 46%
ee in the cyclopropanation of styrene under similar
conditions. Evans et al.2e obtained the trans product in
the same reaction with 49% ee with the use of 3a and
could enhance the selectivity to 99% with 3b. It is
noteworthy that the use of t-Bu derivatives of 2 and 3
has been found to be crucial in order to achieve high
asymmetric induction in cyclopropanation as well as in
several other catalytic reactions.2a,c
environment at the catalytic site. The relatively easy
accessibility and the possibility to modify the chiral
centers also allow for fine-tuning of the ligand structure
for specific applications. The size of the chelate in the
reactive metal complex of bis-oxazolines is another
important feature of the catalyst since it will control the
orientation of the substituents on the two oxazolines
around the metal ion. As the bite angle of the bidentate
bis-oxazoline ligand increases from a five-membered
chelate in A to a six-membered chelate in B, the R-groups
will come in closer proximity to the metal ion, and this
effect will be even more pronounced in a seven-membered
chelate system, C, as shown in Figure 2. It should also
be noted that the chelated metal ion will be more deeply
embedded in complex C, as the larger chelate size will
push the R-groups upward toward the substrate.
We have recently reported15 a series of new bis-
oxazoline ligands16 capable of forming seven-membered
chelates and their use in the asymmetric Cu-catalyzed
cyclopropanation of olefins. In this paper, we present full
experimental details for their preparation, along with a
comparative study on the effect of the chelate size on the
selectivity/reactivity of this important class of C2-sym-
metric ligands (Figure 4).
Bis-oxazolines forming five- and six-membered chelates
are the most widely used ligands of this type. Recently,
Corey13 has reported a new catalyst for the intramolecu-
lar cyclopropanation in the asymmetric synthesis of
sirenin, where the two oxazoline rings are fixed on a
biaryl framework at the 2,2′ positions, increasing the size
of chelate to a nine-membered ring. However, few
investigations have been focused on the seven-membered
chelate systems, and we therefore decided to explore this
factor in order to better understand the influence of the
chelate size on the reactivity/selectivity of bis-oxazoline
ligands.
Syn th esis of Bis-oxa zolin es
In the design of our new ligands it was crucial to have
a rigid, cyclic backbone, and 2,2-dimethyl-1,3-dioxolane-
4,5-dicarboxylic acid dimethyl ester was selected for this
purpose since both enantiomeric forms are readily avail-
Methods for the preparations of bis-oxazolines 1-5 (R
) i-Pr) are available in the literature,2c,e,14 and to compare
the effect of the chelate size we chose the cyclopropana-
tion of styrene with ethyl diazoacetate catalyzed by
CuOTf as the model reaction.
(15) Bedekar, A. V.; Andersson, P. G. Tetrahedron Lett. 1996, 37,
4073.
(16) After the publication of our preliminary results on the asym-
metric cyclopropanation of olefins with this new class of ligands, two
other groups have reported on similar ligands: (a) Harm, A. M.;
Knight, J . G.; Stemp, G. Synlett 1996, 677. (b) Harm, A. M.; Knight,
J . G.; Stemp, G. Tetrahedron Lett. 1996, 37, 6189. (c) Imai, Y.; Zhang,
W.; Kida, T.; Nakatsuji, Y.; Ikeda, I. Tetrahedron: Asymmetry 1996,
7, 2453.
(13) Grant, T. G.; Noe, M. C.; Corey, E. J . Tetrahedron Lett. 1995,
36, 8745.
(14) Bolm, C.; Weickhardt, K.; Zehnder, M.; Ranff, T. Chem. Ber.
1991, 124, 1173.

2520 J . Org. Chem., Vol. 62, No. 8, 1997
Bedekar et al.
Sch em e 1
Sch em e 3a
Sch em e 2a
a
Key: (i) ZnCl2, paraformaldehyde, 150 °C (64%); (ii) trimethyl
orthoformate, p-toluenesulfonic acid, MeOH; (iii) (+)-dimethyl
tartrate, MeOH (80-90%); (iv) (+)-dimethyl tartrate, refluxing
benzene (64%); (v) (+)-dimethyl tartrate, HgO, BF3‚Et2O, Et2O
(66%); (vi) ZnCl2, (+)-dimethyl tartrate (65%).
a
Key: (i) L-amino alcohol, NaCN, MeOH, 45 °C; (ii) MsCl, Et3N,
CH2Cl2, 0 °C; (iii) KOH, MeOH-H2O (50-84% overall yield).
To study the effect of different substitutions on the tether
dioxolane ring, on which two oxazoline rings are mounted,
a series of another set of ligands, 13a -f, derived from
L-valinol was synthesized. For the synthesis of these
different i-Pr derivatives of bis-oxazolines by the same
reaction sequence as shown in Scheme 2, the correspond-
ing dioxolane diesters were required (Scheme 3). There
are several methods available for the preparation of
tartrate dioxolanes from ketones, but no general pathway
was found to be useful for all the required derivatives,
and after much investigation we had to rely on different
syntheses for each derivative of the dioxolane diesters.
For instance, the diester 12a with two hydrogens instead
of alkyl groups could not be prepared from paraformal-
dehyde and dimethyl tartrate by p-toluenesulfonic acid-
catalyzed acetalization but was instead obtained when
ZnCl2 was used as a catalyst, as described by Seebach.19
The dioxolane diesters 12b and 12e were prepared from
diethyl ketone and acetophenone, respectively, by treat-
ment with trimethyl orthoformate and a catalytic amount
of p-toluenesulfonic acid in methanol, followed by treat-
ment with (+)-dimethyl tartrate, as described in the
literature.20 However, this method could not be applied
to the preparation of the n-Bu analogue, 12c, which was
instead synthesized via acid-catalyzed reaction of
5-nonanone and (+)-dimethyl tartrate in refluxing ben-
zene with continuous removal of water over 4 days.19
2-Methyl-1,3-dioxolane-4,5-dicarboxylic acid dimethyl es-
ter (12d ) was prepared from (+)-dimethyl tartrate and
able.17 Although there are several synthetic methods
available for the preparation of oxazolines, many of these
methods utilize acidic reagents or require drastic reaction
conditions. The acid-sensitive nature of the dioxolane
ring of the tether precluded the use of acidic reagents
such as zinc chloride,14 dichlorodimethylstannane,2c meth-
anesulfonic acid,6b etc., otherwise used for the synthesis
of oxazolines. Instead, another approach using the
method recently reported by Denmark8b for the prepara-
tion of bis-oxazolines was investigated and found to be
compatible with the dioxolane ring (Scheme 1).
The dimethyl 2,3-O-isopropylidenetartrates were con-
verted into the corresponding dihydroxy diamides 8
under cyanide-catalyzed (10 mol %) neutral conditions18
in warm methanol. The crude samples of these dihy-
droxy diamides were treated with methanesulfonyl chlo-
ride (MsCl) (2.2 equiv) and Et3N (4.4 equiv) in dichlo-
romethane at 0 °C to afford the corresponding bis-
mesylates 9, which were treated with an aqueous
methanolic solution of NaOH (8 equiv) to afford the bis-
oxazolines in good to moderate overall yield (Scheme 2).
In all cases, the intermediate dihydroxy diamides and
bis-mesylates were too unstable to tolerate flash chro-
matographic purifications; thus, the crude material was
used immediately in the following step. The final ligands
were conveniently purified by flash chromatography on
silica gel in the presence of triethylamine.
The i-Pr derivative, 6c, was found to be the most
effective for the asymmetric cyclopropanation of styrene.
(17) Mash, E. A.; Nelson, K. A.; van Deusen, S.; Hemperly, S. B.
Org. Synth. 1990, 68, 92.
(18) Ho¨gberg, T.; Stro¨m, P.; Ebner, M.; Ra¨msby, S. J . Org. Chem.
1987, 52, 2033.
(19) Seebach, D.; Beck, A. K.; Imwinkelried, R.; Roggo, S.; Wonna-
cott, A. Helv. Chim. Acta 1987, 70, 954.
(20) McClelland, R. A.; Engell, K. M.; Larsen, T. S.; Sorensen, P. E.
J . Chem. Soc., Perkin Trans. 2 1994, 2199.

Cu-Catalyzed Cyclopropanation of Olefins
J . Org. Chem., Vol. 62, No. 8, 1997 2521
Sch em e 4a
Ta ble 1. Com p a r ison of En a n tioselectivities Usin g
i-P r -su bstitu ted Bis-oxa zolin es in th e Cu -Ca ta lyzed
Cyclop r op a n a tion of Styr en e w ith Eth yl Dia zoa ceta te
%
diastereoselectivity % eeb % eec
entry ligand yielda
trans:cis
trans
cis
ref
1
2
3
4
5
6
7
1a
2a
3a
4
5
6c
7a
d
76
d
49
12
76
77
66:34
71:29
69:31
64:36
64:36
70:30
67:33
3
36
49
59
8
8
2e
2c
2e
f
f
f
15
45
30
e
65
73
84
68
f
a
b
Isolated. Determined by HPLC analysis using a CHIRAL-
CEL OD-H column. c Determined by optical rotation. Not re-
d
ported. e Not determined. f This work.
a
Key: (i) L-valinol, NaCN, MeOH, 45 °C; (ii) MsCl, Et3N,
CH2Cl2, 0 °C; (iii) KOH, MeOH-H2O (32-78% overall yield).
vinyl acetate by the action of red HgO and BF3‚Et2O, as
described by Seebach.21 The derivative 12f, with phenyl
and hydrogen substituents at the 2,2 position of diox-
olane, was prepared via ZnCl2-catalyzed acetalization of
benzaldehyde and (+)-dimethyl tartrate, as reported
earlier22 (Scheme 3).
The series of i-Pr derivatives, 13a -f, was synthesized
via the dihydroxy diamides and the bis-mesylates pre-
pared from the corresponding diesters, 12a -f, and opti-
cally pure L-valinol (Scheme 4).
Resu lts a n d Discu ssion
The Cu-catalyzed cyclopropanation of olefins with
different diazoacetates in the presence of tartrate-
tethered bis-oxazolines was investigated in detail in order
to determine the efficiency of the new ligand system. To
compare various derivatives of ligands, as well as the
different reaction conditions, styrene was chosen as a
standard substrate. Different Cu salts and complexes
were employed, and the best results were obtained using
F igu r e 5. Positioning of the oxazoline R-groups as a function
of the chelate size and the configuration of the ligand back-
bone.
product. The steric interactions between the alkyl groups
of the ligand and the carboxy group of the carbenoid
intermediate will determine the approach of the olefin
and hence the stereochemical outcome of the reaction.
With an increase of the bite angle, the substituents on
the oxazoline rings will come closer toward the metal and
the carboxy group in the intermediate, resulting in a
more severe interaction.
23
CuOTf‚1/2C6H6 in accordance with literature obser-
vations.2e In the first part, different i-Pr derivatives of
bis-oxazolines, which will form different chelate sizes in
their complexes with copper, were investigated to deter-
mine the influence of the chelate size on the enantiose-
lectivity in this reaction. The cyclopropanation of styrene
was carried out with ethyl diazoacetate in the presence
of 1.00 mol % of CuOTf and 1.05 mol % of chiral ligand
in dry dichloromethane.
To investigate this effect further, we conducted the
same experiment with bis-oxazoline 4, tethered with an
ethylene unit, and capable of forming a seven-membered
chelate with Cu. The trans product was formed in 59%
ee, which is substantially higher than for 2a and 3a . This
observation clearly supports the assumption that in the
seven-membered chelate the substituents on the two
oxazoline rings will be brought closer to the metal as the
bite angle is increased (Figures 2 and 5), which in turns
results in a higher enantioselectivity for the reaction. The
other bis-oxazoline 5 tethered by a benzene ring, how-
ever, resulted in a considerable drop in both catalytic
activity (12% yield) as well as enantioselectivity (8% ee).
Encouraged by the results obtained using ligand 4, we
decided to concentrate on making a bis-oxazoline with a
more rigid two-carbon tether. One such way to restrict
the flexibility of this bidentate ligand would be to attach
the two oxazoline rings onto the tartrate backbone as
demonstrated by Seebach19,25 and Narasaka26 in their
TADDOL ligands. Other examples of the beneficiary
The beneficial effect of an increased chelate size is
evident from the data presented in Table 1. While the
five-membered chelate formed with ligand 1a resulted
in a nearly nonselective reaction, ligands 2a and 3a ,
which form six-membered chelates, gave rise to consider-
ably higher enantioselectivities. It has been proposed24
that the copper-ligand complex first reacts with the alkyl
diazoacetate to form a metal-carbene complex which
then is attacked by the olefin to form the cyclopropanated
(21) Seebach, D.; Plattner, D. A.; Beck, A. K.; Wang, Y. M.; Hunziker,
D.; Petter, W. Helv. Chim. Acta 1992, 75, 2171.
(22) Fischer, H. O. L.; Appel, H. Helv. Chim. Acta 1934, 1574.
(23) Solomon, R. G.; Kochi, J . K. J . Am. Chem. Soc. 1973, 95, 3300.
(24) (a) Fritschi, H.; Leutenegger, U.; Pfaltz, A. Helv. Chim. Acta
1988, 71, 1553. (b) Doyle, M. P. Chem. Rev. 1986, 86, 919. (c)
Brookhart, M.; Studabaker, W. B. Chem. Rev. 1987, 87, 411.
(25) Seebach, D.; Weidmann, B.; Widler, L. In Modern Synthetic
Methods; Scheffold, R., Ed.; J ohn Wiley and Sons, Inc.: New York,
1983; Vol. 3, p 217.
(26) Narasaka, K.; Iwasaka, N.; Inoue, M.; Yamada, T.; Nakashima,
M.; Sugimori, J . J . Am. Chem. Soc. 1989, 111, 5340.

2522 J . Org. Chem., Vol. 62, No. 8, 1997
Bedekar et al.
Ta ble 2. Effect of Differ en t Su bstitu en ts of th e
Dioxola n e Rin g of th e i-P r Der iva tive of Bis-oxa zolin es
in th e Cu -Ca ta lyzed Cyclop r op a n a tion of Styr en e w ith
Eth yl Dia zoa ceta te
%
diastereoselectivityb % eec,d
entry ligand
R′
R′′ yielda
trans:cis
trans
1
2
3
4
5
6
7
13a
6c
13b Et
13c
13d Me
H
Me
H
Me
Et
53
76
71
60
59
26
50
70:30
70:30
73:27
54:46
67:33
74:26
72:28
50
84
74
25
51
66
69
n-Bu n-Bu
H
H
Ph
13f
13e
Ph
Me
a
b
F igu r e 6. CS Chem3D presentation of L-tartrate ligand 6 and
D-tartrate ligand 7.
Isolated. Determined by 1H NMR analysis of crude sample.
c Determined by HPLC analysis using
column. The configuration was established to be 1R,2R in all the
a CHIRACEL OD-H
d
examples by the sign of optical rotation.
effect of placing the donor centers on a cyclic backbone
compared to a linear one is pointed out by J acobsen27 in
his work on Mn(salen)-catalyzed epoxidations. For this
purpose, the 1,3-dioxolane ring was selected due to its
easy preparation and readily availability in both antipo-
des starting from (+)- or (-)-tartaric acid. Another
benefit with this system lies in the additional chiral
centers in the ligand, which gives us a chance to study
the effect of different combination of the tartrate and
oxazolinyl chirality. In the rigid tartrate derived ligands,
6 and 7 (Figures 5 and 6), the two oxazoline rings are
slightly twisted (instead of being nearly coplanar as in
1-3), pushing the oxazolinyl R groups (represented in
Figure 6 as a carbon atom for simplicity) either toward
the metal ion as in 6c (prepared from L-valinol and
L-tartrate) or away from it in 7a (prepared from L-valinol
and D-tartrate). We were pleased to find that the rigid
bis-oxazoline 7a with a D-tartrate backbone having a wide
angle between the i-Pr groups led to an enantioselectivity
of 68% for the trans product. This could be improved
considerably to 84% by the use of 6c, derived from
L-tartrate, where the i-Pr groups are much closer to the
metal ion as depicted in Figures 5 and 6. The model
cyclopropanation was carried out in CH2Cl2 using 1.00
mol % of CuOTf and 1.05 mol % of ligand 6c. The alkyl
diazoacetate was slowly added with a syringe pump to a
solution of styrene, 1.00 mol % of CuOTf, and 1.05 mol
% of the ligand at 0 °C, and the reaction mixture was
stirred at room temperature for 24 h. Keeping the
reaction temperature constant at 0 °C throughout the
reaction time (28 h) resulted in similar enantioselectivity
but with considerable drop in chemical yield. It was also
observed that the enantioselectivity remained unaffected
when 1.00 mol % of CuOTf and 2.00 mol % of ligand was
used.
Having established that the i-Pr derivative of 6 with
a backbone derived from the naturally occurring (+)-
tartaric acid was superior among the bis-oxazoline ligands
in Table 1 for the model reaction, we further investigated
the other structural features of these new ligands.
Seebach21,25 and Narasaka26 have demonstrated the
effect of the ketal substituents on the efficiency of their
TADDOL catalysts for asymmetric Diels-Alder and
carbonyl addition reactions. Narasaka observed that C1-
symmetric ligands having two different substituents on
the ketal were superior to the corresponding C2-sym-
metric analogues. This observation prompted us to
prepare a series of C2- and non-C2-symmetric ligands,
13a -f, having different substituents on the dioxolane
ring. These ligands were then evaluated in the Cu-
Ta ble 3. Effect of Differ en t Oxa zolin yl Gr ou p s of
Bis-oxa zolin es in th e Cu -Ca ta lyzed Cyclop r op a n a tion
of Styr en e w ith Eth yl Dia zoa ceta te
confgn of
tartrate
entry ligand backbone
diastereo-
selectivityb trans
% eec
% eed
%
cis
R
yielda trans:cis (confgnd) (confgnd)
1
2
3
4
5
6
7
6a
6b
6c
6d
6e
7a
7b
L
L
L
L
L
D
D
Me2
58
77
76
58:42
68:32
70:30
64:36
73:27
67:33
70:30
18
(1S,2S)
50
(1R,2R) (1R,2S)
84 65
(1R,2R) (1R,2S)
e
Bn
39
i-Pr
t-Bu 78
2
e
(1R,2R)
77
i-Bu
i-Pr
80
77
70
(1R,2R) (1R,2S)
68 73
(1R,2R) (1R,2S)
84 85
(1R,2R) (1R,2S)
t-Bu 85
a
b
Isolated. Determined by 1H NMR analysis of the crude
mixture. c Determined by chiral HPLC analysis using a CHIRAL-
CEL OD-H column. Determined by chiroptical comparison. e Not
d
determined.
catalyzed cyclopropanation of styrene with ethyl diaz-
oacetate, and the results are presented in Table 2.
As shown in Table 2, the presence of two geminal
methyl groups on the dioxolane ring resulted in optimum
selectivity for the reaction. Replacing the methyl groups
by hydrogen drastically decreased the selectivity. On the
other hand, increasing the size of the dioxolane substit-
uents from methyl to ethyl gave a slightly lower enan-
tioselectivity (74% ee for the trans product; Table 2, entry
3) compared to the dimethyl analogue (84% ee; Table 2,
entry 2). The ligand 13c, with the even larger n-Bu
substituents, was less efficient, which is in accordance
with the observation made by Narasaka for the TAD-
DOL-catalyzed asymmetric Diels-Alder reactions. How-
ever, introduction of two different substituents on the
dioxolane ring did not improve the selectivity (Table 2,
entries 5-7).
A number of bis-oxazolines prepared from different 1,2-
amino alcohols with the 2,2-dimethyldioxolane tether
prepared from (+)- and (-)-tartaric acid were next
compared in the reaction under investigation (Table 3).
The ligand 6a , with a gem-dimethyloxazolinyl substitu-
ent, gave very poor asymmetric induction. Its failure to
induce enantioselectivity in the reaction indicates that
the presence of the chiral backbone alone is not respon-
sible for enantioselective addition of the carbene to the
olefin. The benzyl derivative, 6b, gave the cyclopropa-
nation products of styrene with 50 and 39% ee for trans
and cis isomers, respectively. Masamune2c previously
reported that the benzyl derivative of bis-oxazoline 2 with
(27) J acobsen, E. N.; Zhang, W.; Muci, A. R.; Ecker, J . R.; Deng, L.
J . Am. Chem. Soc. 1991, 113, 7063.

Cu-Catalyzed Cyclopropanation of Olefins
J . Org. Chem., Vol. 62, No. 8, 1997 2523
Ta ble 4. Solven t Effect on th e Cu -Ca ta lyzed
Cyclop r op a n a tion of Styr en e w ith Eth yl Dia zoa ceta te
in th e P r esen ce of 6c
diastereoselectivityb
entry
solvent
% yielda
trans:cis
% eec trans
1
2
3
4
5
6
CH2Cl2
CHCl3
Styrene
THF
Toluene
Pentane
76
54
48
34
36
d
70:30
67:33
74:26
80:20
78:22
84
73
63
37
59
a
b
Isolated. Determined by 1H NMR analysis of crude mixture.
c Determined by chiral HPLC analysis using a CHIRALCEL OD-H
column. No reaction.
d
Ta ble 5. Cyclop r op a n a tion of Styr en e w ith Va r iou s
Dia zoa ceta tes in th e P r esen ce of 6c
diastereo-
%
%
%
selectivityb ee/% de ee/% de
entry
diazoacetate
ethyl
tert-butyl
(-)-menthyl
(-)-8-phenylmenthyl
yielda
trans:cis
trans
cis
1
2
3
4
76
84
86
80
70:30
82:18
85:15
80:20
84
88c
89e
96d
65
84d
89e
91d
a
b
Isolated. Determined by 1H NMR analysis of the crude
sample. c Determined by optical rotation of the acid. Determined
d
by chiral HPLC analysis using a CHIRALCEL OD-H column.
e Determined by GC analysis.
for i-Pr derivative 6c, but the cis product was also
obtained with equal optical purity, 85% ee (Table 3,
entry 7).
The conclusion that the ligand 6d failed to form a
uniform bidentate complex with CuOTf due to its large
1
oxazolinyl substituents was also supported by H NMR
study of separate experiments in which the ligands were
treated with various cuprous complexes. The ligands 6c
and 6d were treated with different concentrations of [Cu-
F igu r e 7. 1H NMR study of the complex formation between
the ligands 6c, 6d , and [Cu(CH3CN)4]PF6 in CDCl3.
28
1
(CH3CN)4]PF6 and analyzed by H NMR spectroscopy.
The comparison is presented in Figure 7. The i-Pr ligand
6c showed a single set of signals with downfield shift of
the oxazolinyl protons (Figure 7a) indicating rapid com-
plex formation. On the other hand, the t-Bu derivative
6d gave rise to at least two sets of signals upon addition
of the cuprous complex (Figure 7b), indicating the forma-
tion of a mixture of ligand-copper complexes. A similar
pattern of signals was observed when ligands 6c and 6d
were treated with CuOTf‚1/2C6H6 and [Cu(CH3CN)4]-
ClO4,28 but the signals were not as sharp and clear as
those observed for [Cu(CH3CN)4]PF6.
A systematic study of solvent effect on the cyclopro-
panation of styrene in the presence of 6c was carried out
to establish the best conditions for this reaction, and the
results are presented in Table 4. Dichloromethane and
chloroform were very suitable solvents, with the former
being marginally superior in all respects. Performing the
reaction in neat styrene resulted in a considerable drop
in both chemical and optical yields, but the level of
stereochemical induction was maintained. In tetrahy-
drofuran and toluene even lower yields and enantiose-
lectivities were observed. The reaction could not be
carried out in pentane due to the insolubility of the
catalyst in this medium.
methylene backbone gave the same products in 36 and
15% ee under similar reaction conditions. This again
shows the considerable beneficial effect of the twisted
oxazoline rings mounted on the tartrate backbone.
The bulkier tert-butyl derivative, 6d , with the tether
prepared from L-tartaric acid, was expected to give even
higher enantioselectivities compared to the sterically less
demanding i-Pr derivative, 6c (Table 3, entry 4). Con-
trary to expectations, this ligand resulted in the forma-
tion of almost racemic products, although the chemical
yield was high. The t-Bu derivative of the bis-oxazoline
with a dioxolane tether prepared from D-tartaric acid, 7b,
however, proved to be a much more effective ligand giving
the trans and cis cyclopropanated products with 84 and
85% ee, respectively. The t-Bu derivative of bis-oxazoline
3b was observed to be by far the most effective of the
ligands studied for the model reaction, because the large
substituents are capable of efficiently controlling the
olefin approach. In the case of 6d , the combination of
two bulky t-Bu oxazolinyl groups and the inward twist
of the L-tartrate backbone resulted in a severe steric
hindrance. This does not leave sufficient space for the
copper to simultaneously coordinate with both nitrogens
of the oxazoline rings. Hence, with 6d , the reaction is
probably catalyzed by free or monocoordinated Cu, lead-
ing to optically inactive products in high yield. In the
case of ligand 7b the oxazoline rings are twisted outward
by the D-tartrate-derived backbone, allowing enough
space for bis-oxazoline to complex with the copper. The
trans product of cyclopropanation of styrene catalyzed by
7b was found to be as enantiomerically enriched as that
The influence of the diazoacetate substituent on the
selectivity in the cyclopropanation of styrene was also
studied using different ester groups (Table 5). Not
unexpectedly, sterically demanding diazoacetates, such
as tert-butyl and menthyl diazoacetate, resulted in
increased diastereoselectivities and enantioselectivities.
(28) Kubas, G. J . Inorg. Synth. 1979, 19, 90.

2524 J . Org. Chem., Vol. 62, No. 8, 1997
Bedekar et al.
Ta ble 6. Cyclop r op a n a tion of Differ en t Olefin s w ith
Va r iou s Dia zoa ceta tes
trometer. Mass spectra were recorded on a INCOS 50 GC/
MS system. Melting points were determined on a Leitz
apparatus and are uncorrected. GC analysis was performed
on Varian 3400 capillary gas chromatograph using a 25 m SE
54 column. HPLC analyses were performed on a Waters M
45 pump coupled with a Varian 9065 polychrom diode array
detector. Toluene, methylene chloride, and chloroform were
dried over calcium hydride and freshly distilled under nitrogen.
Styrene and R-methylstyrene were distilled prior to use. THF
and diethyl ether were distilled over sodium/benzophenone
% yielda
entry
1
olefin
diazoacetate ligand (trans:cis)b % ee/% de
1,1-diphenyl- tert-butyl
ethylene
R-methyl-
styrene
6c
6c
90
70
92c
2
(-)-menthyl
85 (trans)d
(57:43)
44
(75:25)
64
(72:28)
51
(72:28)
63 (cis)d
3
4
5
vinyl acetate ethyl
6c
6c
7b
64 (trans)e
60 (cis)e
under nitrogen. Preparative TLC plates, SIL G-100 UV254
,
vinyl acetate (-)-menthyl
vinyl acetate (-)-menthyl
74 (trans)d
88 (cis)d
were purchased from Macherey-Nagel. Ethyl diazoacetate was
purchased from Aldrich Co., and tert-butyl diazoacetate was
prepared according to a literature procedure.31 CuOTf‚1/2C6H6
was prepared according to a previously described procedure,23
and (-)-8-phenylmenthyl diazoacetate was prepared in a
similar procedure24a as described for (-)-menthyl diazoacetate.
(4R,5R)-2,2-Dimethyl-1,3-dioxolane-4,5-dicarboxylic acid dim-
ethyl ester (10) and (4S,5S)-2,2-dimethyl-1,3-dioxolane-4,5-
dicarboxylic acid dimethyl ester (11) were synthesized accord-
ing to a previously described procedure.17 It is the first step
in the preparation of the following ligands.
Gen er a l P r oced u r e for th e Syn th esis of Liga n d s.
Unless otherwise mentioned, the bis-oxazoline ligands were
prepared according to the three-step process outlined below.
(-)-(4R,5R)-Bis[(S)-4-isop r op yloxa zolin -2-yl]-2,2-d im -
eth yl-1,3-d ioxola n e (6c). To a solution of L-2-amino-3-
methyl-1-butanol (L-valinol) (0.84 g, 8.07 mmol) in dry metha-
nol (12 mL) was added (4R,5R)-2,2-dimethyl-1,3-dioxolane-4,5-
dicarboxylic acid dimethyl ester (10) (0.80 g, 3.67 mmol)
together with a catalytic amount of NaCN (18 mg, 0.03 mmol).
The reaction mixture was stirred under inert atmosphere at
45 °C for 24 h. After being cooled to room temperature, the
solution was concentrated to give an oil that was dissolved in
ethyl acetate (25 mL) and washed with water (5 mL). The
organic phase was dried (MgSO4), and evaporation of the
solvent yielded a yellow oily product pure enough to be used
in the next step without further purification: 1H NMR (300
MHz) δ 7.20 (bd, J ) 8.0 Hz, 2 H), 4.58 (s, 2 H), 3.75 (m, 4 H),
3.65 (m, 2 H), 1.92 (m, 2 H), 1.53 (s, 6 H), 0.98 (d, J ) 6.8 Hz,
6 H), 0.96 (d, J ) 6.1 Hz, 6 H).
62 (trans)d
80 (cis)d
a
b
Isolated. Determined by 1H NMR analysis of crude mixture.
c Determined by optical rotation of the acid. Determined by GC
d
analysis. e Established by 1H NMR analysis with chiral shift
reagent.
We also prepared (-)-8-phenylmenthyl diazoacetate,
a new diazoester, from (-)-8-phenylmenthol in analogy
with the method reported24a for the preparation of (-)-
menthyl diazoacetate. Cyclopropanation of styrene with
this (-)-8-phenylmenthyl diazoacetate in the presence of
bis-oxazoline 6c gave trans and cis products with 96 and
91% de, respectively.
The efficiency of the new ligand system for the cyclo-
propanation was not restricted to monosubstituted ole-
fins. The cyclopropanation of 1,1-diphenylethylene with
6c/CuOTf and tert-butyl diazoacetate gave the R enan-
tiomer in 92% ee and 90% yield (Table 6).
Another interesting class of substrates consists of enol
derivatives that will form optically active precursors to
cyclopropanols upon asymmetric cyclopropanation. De-
spite the synthetic utility of these in organic synthesis,29
there are only a few reports30 on the enantioselective
cyclopropanation of such substrates in the literature. We
therefore chose to study the cyclopropanation of vinyl
acetate with our new ligand system (Table 6, entries
3-5). Using ligand 6c, the cyclopropanated products
could be obtained in much higher enantioselectivities
than what has previously been reported30a with chiral
dirhodium complexes.
Su m m a r y. By introducing a tartrate backbone be-
tween the two oxazoline rings, we have synthesized a new
class of bis-oxazoline ligands that have shown high
enantioselectivity in cyclopropanation of olefins compared
to the corresponding conventional ligands having a
methylene bridge between the two oxazoline rings. The
valine-derived ligand 6c was found to be superior to the
analogous ligands 2a and 3a , but less selective than
ligand 3b, which in turn is derived from the more
expensive tert-leucine. This new ligand should find many
applications in asymmetric reactions catalyzed by C2-
symmetric ligands as recently reported by Ikeda16c for
hydrosilylation of ketones.
To an ice-cooled solution of the dihydroxy diamide (1.21 g,
3.37 mmol) and Et3N (1.53 g, 15.1 mmol) in CH2Cl2 (15 mL)
was added MsCl (0.96 g, 8.4 mmol) slowly. The mixture was
allowed to warm to room temperature, stirred for 1 h, and
washed with water (3 mL). The organic phase was dried (Na2-
SO4) and concentrated to dryness in vacuo to give a yellow
oil, which was used in the next reaction without purification.
The bis-mesylated compound was treated with NaOH (0.14
g, 3.4 mmol in 24 mL of a 5:1 MeOH/H2O mixture). The
reaction mixture was stirred at room temperature for 2 days,
at which point no trace of the starting material could be seen
on TLC (5% v/v MeOH in CHCl3). The solvent was removed
and the residual oil dissolved in CH2Cl2 (25 mL), washed with
water (5 mL), dried (Na2SO4), and concentrated in vacuo.
Purification by flash chromatography (1.5% triethylamine, 20%
chloroform in pentane) gave the ligand 6c as a pale yellow,
low-melting solid in 77% overall yield: [R]21 ) -137.7 (c )
D
1
1.00, CH2Cl2); H NMR (300 MHz) δ 4.94 (s, 2 H), 4.31 (dd, J
) 9.7, 8.3 Hz, 2 H), 4.05 (app t, J ) 8.3 Hz, 2 H), 3.97 (m, 2
H), 1.78 (m, 2 H), 1.51 (s, 6 H), 0.95 (d, J ) 6.8 Hz, 6 H), 0.87
(d, J ) 6.8 Hz, 6 H); 13C NMR (75.5 MHz) δ 163.1, 112.7, 74.0,
72.0, 71.7, 32.4, 26.4, 18.6, 17.9; IR (CCl4, cm-1) 2960, 1673;
MS (EI) m/ z (rel intensity) 324 (M+ - 15; 47), 281 (11), 267
(18), 223 (100), 195 (19), 154 (77), 114 (42), 85 (43), 79 (57).
Anal. Calc for C17H28N2O4: C, 62.94; H, 8.70; N, 8.63.
Found: C, 62.64; H, 8.56; N, 8.47.
Exp er im en ta l Section
Optical rotations were recorded on a thermostated Perkin-
Elmer 241 polarimeter using a 1.0 dm cell. 1H and 13C NMR
spectra were recorded for CDCl3 solutions at 400/300 and 100.4
/75.5 MHz (Varian Unity 400/Varian XL 300), respectively.
Chemical shifts for protons are reported using the residual
1H in CDCl3 as the internal reference (δ 7.26). Carbon shifts
are referenced to the 13C signal of CDCl3 at 77.0 ppm. Infrared
spectra were recorded on a Perkin-Elmer 1600 FT-IR spec-
(-)-(4R,5R)-Bis(4,4-dim eth yloxazolin -2-yl)-2,2-dim eth yl-
1,3-d ioxola n e (6a ). The ligand 6a was prepared from 10
(0.50 g, 2.29 mmol) and 2-amino-2-methyl propanol (0.45 g,
5.04 mmol) in 46% overall yield (colorless oil) following the
same procedure as reported for 6c: [R]21 ) -32.5 (c ) 1.17,
D
CH2Cl2); 1H NMR (300 MHz) δ 4.87 (s, 2 H), 4.00 (s, 4 H), 1.49
(s, 6 H), 1.28 (s, 6 H), 1.26 (s, 6 H); 13C NMR (75.5 MHz) δ
(29) Kulinkovich, O. G. Russ. Chem. Rev. 1993, 62, 839.
(30) (a) Mu¨ller, P.; Band, C.; Ene´, D.; Motallebi, S.; Doyle, M. P.,
Brandes, B. D.; Dyatkin, A. B.; See, M. M. Helv. Chim. Acta 1995, 78,
459. (b) Dammast, F.; Reissig, H.-U. Chem. Ber. 1993, 126, 2727.
(31) Regitz, M.; Hocker, J .; Liedhegener, A. Organic Syntheses;
Wiley: New York, 1973; Collect. Vol. V, p 179.

Cu-Catalyzed Cyclopropanation of Olefins
J . Org. Chem., Vol. 62, No. 8, 1997 2525
161.5, 112.6, 79.6, 74.0, 67.4, 18.1, 27.9, 26.3; IR (neat, cm-1
)
(-)-(4R,5R)-Bis[(S)-4-isop r op yloxa zolin -2-yl]-1,3-d iox-
ola n e (13a ). The ligand 13a was prepared from 12a (0.50 g,
2.47 mmol) and L-valinol (0.56 g, 5.43 mmol) in 78% overall
yield following the same procedure as reported for 6c. Cy-
clization of the intermediate bis-mesylate with KOH required
only 1 day. Purification by flash chromatography (2% CHCl3,
1.5% Et3N in pentane) gave the pure compound as a colorless
3125, 2975, 1675, 1225; MS (EI) m/ z (rel intensity) 316 (M+
- 36, <1), 275 (90), 233 (30), 216 (39), 191 (42), 178 (60), 163
(35), 137 (100), 112 (33), 98 (26), 82 (24), 70 (23). Anal. Calcd
for C15H24N2O4: C, 60.79; H, 8.16; N, 9.45. Found: C, 60.58;
H, 8.09; N, 9.28.
(-)-(4R,5R)-Bis[(S)-4-ben zyloxazolin -2-yl]-2,2-dim eth yl-
1,3-d ioxola n e (6b). The ligand 6b was prepared from 10
(1.00 g, 4.58 mmol) and L-phenylalaninol (1.52 g, 10.1 mmol)
in 84% overall yield (colorless oil) following the same procedure
as reported for 6c: [R]21D ) -82.5 (c ) 1.00, CH2Cl2); 1H NMR
(300 MHz) δ 7.29 (m, 5 H), 7.21 (m, 10 H), 4.91 (s, 2 H), 4.46
(m, 2 H), 4.28 (app t, J ) 9.0 Hz, 2 H), 4.08 (app t, J ) 9.0 Hz,
2 H), 3.14 (dd, J ) 13.8, 5.2 Hz, 2 H), 2.68 (dd, J ) 13.7, 8.5
Hz, 2 H), 1.52 (s, 6 H); 13C NMR (75.5 MHz) δ 163.8, 137.4,
129.1, 128.5, 126.5, 112.8, 73.9, 72.4, 67.3, 41.3, 26.3; IR (neat,
cm-1) 2987, 2935, 1670, 1382, 1248; MS (EI) m/ z (rel intensity)
420 (M+, 34), 405 (14), 329 (72), 271 (41), 260 (20), 243 (33),
202 (40), 117 (54), 91 (100). Anal. Calcd for C25H28N2O4: C,
71.41; H, 6.71; N, 6.66. Found: C, 71.25; H, 6.79; N, 6.58.
oil in 78% yield: [R]21 ) -136.4 (c ) 1.00, CH2Cl2); 1H NMR
D
(400 MHz) δ 5.23 (s, 2 H), 4.94 (d, J ) 1.5 Hz, 2 H), 4.33 (dd,
J ) 9.6, 8.3 Hz, 2 H), 4.05 (app t, J ) 8.2 Hz, 2 H), 3.97 (ddd,
J ) 9.7, 8.1, 6.1 Hz, 2 H), 1.78 (m, 2 H), 0.96 (d, J ) 6.7 Hz,
6 H), 0.89 (d, J ) 6.7 Hz, 6 H); 13C NMR (100.4 MHz) δ 163.1,
127.5, 96.9, 74.0, 72.2, 70.9, 32.4, 18.6, 18.0; IR (CCl4, cm-1
)
2961, 1674; MS (EI) m/ z (rel intensity) 296 (M+, 16), 251 (83),
223 (100), 195 (59), 184 (16), 154 (93), 140 (34), 112 (34), 84
(27), 69 (62). Anal. Calcd for C15H24N2O4: C, 60.79; H, 8.16;
N, 9.45. Found: C, 60.56; H, 7.99; N, 9.23.
(-)-(4R,5R)-Bis[(S)-4-isopr opyloxazolin -2-yl]-2,2-dieth yl-
1,3-d ioxola n e (13b). The ligand 13b was prepared from 12b
(0.28 g, 1.14 mmol) and L-valinol (0.26 g, 2.51 mmol) in 32%
overall yield (pale yellow oil) following the same procedure as
(-)-(4R,5R)-Bis[(S)-4-ter t-bu tyloxa zolin -2-yl]-2,2-d im -
eth yl-1,3-d ioxola n e (6d ). The ligand 6d was prepared from
10 (0.80 g, 3.67 mmol) and L-tert-leucinol (0.95 g, 8.07 mmol)
in 54% overall yield (white solid) following the same procedure
1
reported for 6c: [R]21 ) -106.1 (c ) 1.08, CH2Cl2); H NMR
D
(300 MHz) δ 4.88 (s, 2 H), 4.30 (dd, J ) 9.1, 8.7 Hz, 2 H), 4.04
(app t, J ) 8.4 Hz, 2 H), 3.96 (dd, J ) 9.4, 6.1 Hz, 2 H), 1.75
(m, 6 H), 0.95 (d, J ) 6.7 Hz, 12 H), 0.86 (t, J ) 6.7 Hz, 6 H);
13C NMR (75.5 MHz) δ 162.9, 116.5, 74.3, 72.0, 70.6, 32.4, 29.8,
18.7, 17.9, 7.8; IR (CCl4, cm-1) 2964, 2879, 1673, 1466; MS
(EI) m/ z (rel intensity) 352 (M+, 6), 285 (23), 172 (59), 142
(35), 111 (29), 98 (21), 86 (35), 69 (89). Anal. Calcd for
C19H32N2O4: C, 64.75; H, 9.15; N, 7.95. Found: C, 64.73; H,
8.92; N, 8.05.
as reported for 6c: mp 91-93 °C; [R]21 ) -117.5 (c ) 1.00,
D
CH2Cl2); 1H NMR (300 MHz) δ 4.97 (s, 2 H), 4.26 (dd, J )
10.2, 8.8 Hz, 2 H), 4.15 (app t, J ) 8.5 Hz, 2 H), 3.90 (dd, J )
7.9, 5.7 Hz, 2 H), 1.51 (s, 6 H), 0.88 (s, 18 H); 13C NMR (75.5
MHz) δ 163.2, 112.7, 75.6, 74.0, 69.3, 33.6, 29.4, 25.7; IR (CCl4,
cm-1) 2958, 1675, 1247; MS (EI) m/ z (rel intensity) 337 (M+
- 15, 17), 295 (10), 237 (100), 168 (30), 112 (36), 84 (16). Anal.
Calcd for C19H32N2O4: C, 64.75; H, 9.15; N, 7.95. Found: C,
64.78; H, 9.20; N, 7.87.
(-)-(4R,5R)-Bis[(S)-4-isop r op yloxa zolin -2-yl]-2,2-d ibu -
tyl-1,3-d ioxola n e (13c). The ligand 13c was prepared from
12c (1.00 g, 3.31 mmol) and L-valinol (0.75 g, 7.28 mmol) in
35% overall yield (colorless oil) following the same procedure
as reported for 6c. The methanolic KOH treatment of the bis-
(-)-(4R,5R)-Bis[(S)-4-(1-m et h ylp r op yl)oxa zolin -2-yl]-
2,2-d im eth yl-1,3-d ioxola n e (6e). The ligand 6e was pre-
pared from 10 (0.70 g, 3.21 mmol) and L-isoleucinol (0.83 g,
7.06 mmol) in 50% overall yield (colorless oil) following the
mesylate required 4 days to reach completion: [R]21 ) -40.5
D
1
(c ) 1.26, CH2Cl2); H NMR (400 MHz) δ 4.88 (s, 2 H), 4.32
same procedure as reported for 6c: [R]21 ) -120.9 (c ) 1.12,
D
(dd, J ) 9.7, 8.3 Hz, 2 H), 4.06 (dd, J ) 8.1, 6.7 Hz, 2 H), 3.96
(m, 2 H), 1.74 (m, 4 H), 1.34 (bm, 8 H), 0.97 (m, 6 H), 0.89
(app t, J ) 6.7 Hz, 12 H); 13C NMR (100.4 MHz) δ 116.2, 77.7,
74.2, 72.1, 70.6, 64.2, 57.4, 37.2, 36.8, 32.4, 29.0, 25.8, 25.6,
22.8, 19.5, 18.7, 18.5, 18.0, 14.0; IR (CCl4, cm-1) 2960, 1731,
1273; MS (EI) m/ z (rel intensity) 413 (M+ - 30, 78), 387 (31),
285 (28), 172 (65), 142 (38), 130 (13), 104 (35), 85 (58), 69 (87).
Anal. Calcd for C23H40N2O4: C, 67.61; H, 9.87; N, 6.86.
Found: C, 67.48; H, 10.05; N, 6.65.
CH2Cl2); 1H NMR (300 MHz) δ 4.92 (s, 2 H), 4.28 (m, 2 H),
1.72 (m, 2 H), 1.49 (m, 10 H), 1.15 (dddq, J ) 14.3, 13.9, 7.4,
7.4 Hz, 2 H), 0.89 (t, J ) 7.4 Hz, 6 H), 0.80 (d, J ) 6.8 Hz, 6
H); 13C NMR (75.5 MHz) δ 162.9, 112.6, 74.0, 70.6, 70.2, 38.6,
26.3, 25.9, 14.1, 11.5; IR (neat, cm-1) 2963, 1673, 1382; MS
(EI) m/ z (rel intensity) 352 (M+, 1), 337 (42), 295 (32), 237
(100), 209 (17), 168 (50), 153 (18), 128 (19), 110 (21). Anal.
Calcd for C19H32N2O4: C, 64.75; H, 9.15; N, 7.95. Found: C,
64.66; H, 9.03; N, 7.87.
(-)-(4R,5R)-Bis[(S)-4-isop r op yloxa zolin -2-yl]-2-m eth yl-
1,3-d ioxola n e (13d ). The ligand 13d was prepared from 12d
(1.40 g, 6.86 mmol) and L-valinol (1.56 g, 15.09 mmol) in 34%
overall yield (colorless oil) following the same procedure as
reported for 6c. The methanolic KOH treatment of the bis-
mesylate required 1 day to reach completion: [R]21D ) -147.4
(+)-(4S,5S)-Bis[(S)-4-isop r op yloxa zolin -2-yl]-2,2-d im -
eth yl-1,3-d ioxola n e (7a ). The method described for the
preparation of 6c was also used to prepare the ligands with
D-tartrate backbone. The ligand 7a was prepared from 11
(0.60 g, 2.75 mmol) and L-valinol (0.62 g, 6.05 mmol) in 73%
1
overall yield (colorless oil) following the same procedure as
(c ) 1.25, CH2Cl2); H NMR (400 MHz) δ 5.39 (q, J ) 4.8 Hz,
1
reported for 6c: [R]21 ) -12.7 (c ) 0.99, CH2Cl2); H NMR
D
1 H), 4.90 (m, 2 H), 4.31 (m, 2 H), 4.04 (m, 2 H), 3.96 (m, 2 H),
1.78 (m, 2 H), 1 (m, 3 H), 0.95 (m, 6 H), 0.88 (m, 6 H); 13C
NMR (100.4 MHz) δ 163.6, 163.1, 104.0, 74.7, 74.2, 72.1, 72.0,
(300 MHz) δ 4.96 (s, 2 H), 4.32 (app. t, J ) 9.1 Hz, 2 H), 4.06
(app t, J ) 9.1 Hz, 2 H), 3.98 (ddd, J ) 9.1, 7.6, 6.1 Hz, 2 H),
1.78 (dqq, J ) 14.0, 6.7, 6.5 Hz, 2 H), 1.51 (s, 6 H), 0.95 (d, J
) 6.6 Hz, 6 H), 0.88 (d, J ) 6.7 Hz, 6 H); 13C NMR (75.5 MHz)
δ 163.1, 112.6, 74.1, 72.1, 70.6, 32.3, 26.3, 18.6, 17.8; IR (CCl4,
cm-1) 2960, 1673, 1371, 1242; MS (EI) m/ z (rel intensity) 324
(M+, 11), 309 (57), 281 (19), 267 (24), 223 (100), 195 (23), 154
(80), 114 (55), 85 (49), 69 (39). Anal. Calcd for C17H28N2O4:
C, 62.94; H, 8.70; N, 8.63. Found: C, 63.08; H, 8.66; N, 8.53.
70.9, 70.7, 46.3, 32.4, 29.4, 19.4, 18.7, 18.6, 18.0; IR (CCl4, cm-1
)
2960, 1729, 1675; MS (EI) m/ z (rel intensity) 310 (M+, <1%),
291 (41), 246 (41), 194 (37), 178 (19), 151 (100), 110 (31), 96
(13), 68 (30). Anal. Calcd for C16H26N2O4: C, 61.91; H, 8.44;
N, 9.03. Found: C, 61.95; H, 8.52; N, 9.15.
(-)-(4R,5R)-Bis[(S)-4-isop r op yloxa zolin -2-yl]-2-m eth yl-
2-p h en yl-1,3-d ioxola n e (13e). The ligand 13d was prepared
from 12e (0.60 g, 2.14 mmol) and L-valinol (0.49 g, 4.71 mmol)
in 25% overall yield (colorless oil) following the same procedure
as reported for 6c: [R]21D ) -99.7 (c ) 1.01, CH2Cl2); 1H NMR
(300 MHz) δ 7.54 (m, 2 H), 7.27 (m, 3 H), 5.00 (m, 2 H), 4.34
(m, 1H), 4.05 (m, 2 H), 3.99 (m, 1 H), 3.89 (app t, J ) 8.0 Hz,
1 H), 3.67 (m, 1 H), 1.78 (s, 3 H), 1.60 (m, 2 H), 0.95 (d, J )
6.8 Hz, 3 H), 0.87 (d, J ) 6.8 Hz, 3 H), 0.84 (d, J ) 6.8 Hz, 3
H), 0.76 (d, J ) 6.7 Hz, 3 H); 13C NMR (75.5 MHz) δ 162.6,
142.5, 128.0, 127.8, 125.5, 112.4, 74.4, 73.8, 72.0, 71.8, 70.7,
70.4, 32.4, 32.2, 28.3, 18.6, 18.5, 17., 17.8; IR (CCl4, cm-1) 2873,
1673, 1371; MS (EI) m/ z (rel intensity) 386 (M+, 9%), 371 (23),
327 (100), 223 (15), 105 (22). Anal. Calcd for C22H30N2O4: C,
68.37; H, 7.82; N, 7.25. Found: C, 68.22; H, 7.61; N, 7.09.
(+)-(4S,5S)-Bis[(S)-4-ter t-b u t yloxa zolin -2-yl]-2,2-d im -
eth yl-1,3-d ioxola n e (7b). The ligand 7b was prepared from
11 (0.60 g, 2.75 mmol) and L-tert-leucinol (0.71 g, 6.05 mmol)
in 52% overall yield (white solid) following the same procedure
as reported for 6c: mp 117-119 °C; [R]21 ) -40.8 (c ) 0.98,
D
1
CH2Cl2); H NMR (300 MHz) δ 4.96 (s, 2 H), 4.27 (app t, J )
8.0 Hz, 2 H), 4.14 (app t, J ) 8.0 Hz, 2 H), 3.91 (app t, J ) 8.0
Hz, 2 H), 1.51 (s, 6 H), 0.89 (s, 18 H); 13C NMR (75.5 MHz) δ
162.9, 112.6, 75.8, 74.2, 69.3, 33.6, 26.2, 25.7; IR (CCl4, cm-1
)
2958, 1675, 1365, 1246; MS (EI) m/ z (rel intensity) 337 (M+
- 15, 17), 295 (11), 237 (100), 168 (29), 112 (34), 84 (15). Anal.
Calcd for C19H32N2O4: C, 64.75; H, 9.15; N, 7.95. Found: C,
64.60; H, 9.25; N, 8.03.

2526 J . Org. Chem., Vol. 62, No. 8, 1997
Bedekar et al.
1
(-)-(4R,5R)-Bis[(S)-4-isop r op yloxa zolin -2-yl]-2-p h en yl-
1,3-d ioxola n e (13f). The ligand 13f was prepared from 12f
(1.00 g, 3.76 mmol) and L-valinol (0.85 g, 8.27 mmol) following
the same procedure as reported for 6c. Purification was
performed on preparative TLC on silica gel first treated with
triethylamine and then eluted with a solution of 2% MeOH
and 2% Et3N in chloroform to yield the ligand 13f in 45%
product was determined by H NMR spectroscopy to be 70:30
in the crude mixture. The cyclopropanes were isolated by flash
chromatography (2% ethyl acetate in pentane; 76% combined
yield). Optical purity could be determined by HPLC analysis
for the trans isomer (CHIRALCEL OD-H column, 0.5 mL/min
flow rate, 10% i-PrOH in hexane as eluent, 84% ee), while
comparison of optical rotation2f was used for the cis compound
(65% ee), the absolute configuration being confirmed by the
sign of the optical rotation [trans: (1R,2R), cis: (1R,2S)].
Cyclop r op a n a tion of 1,1-Dip h en yleth ylen e w ith ter t-
Bu tyl Dia zoa ceta te. A mixture of CuOTf‚1/2C6H6 (3.0 mg,
0.012 mmol) and ligand 6c (4.1 mg, 0.013 mmol) was stirred
in dry CH2Cl2 (1 mL) under argon, for 30 min. This solution
was cooled to 0 °C, and 1,1-diphenyethylene (0.22 g, 1.2 mmol)
was added, followed by slow addition of a CH2Cl2 solution (2.5
mL) of tert-butyl diazoacetate (200 mg, 1.4 mmol) over 5 h via
syringe pump. The reaction mixture was allowed to warm to
room temperature, stirred for an additional 16 h, and then
quenched with a 10% aqueous solution of NH4Cl (5 mL). The
solution was diluted with Et2O (25 mL), washed with water
(5 mL) and brine (5 mL), and dried (MgSO4). Evaporation of
the solvent in vacuo afforded the crude product. The product
was isolated by column chromatography (2% ethyl acetate in
pentane; 0.32 g, 90% yield). The absolute configuration of the
pure tert-butyl ester was established as described by Evans2e
to be R [92% ee; [R]D -194 (c 0.29; CHCl3) for the acid obtained
upon hydrolysis of the tert-butyl ester].
overall yield as a colorless oil: [R]21 ) -72.5 (c ) 1.00, CH2-
D
Cl2); 1H NMR (400 MHz) δ 7.58 (d, J ) 6.2 Hz, 2 H), 7.37 (m,
3 H), 6.10 (s, 1 H), 5.10 (m, 2 H), 4.34 (dd, J ) 9.2, 12.3 Hz, 2
H), 4.09 (dd, J ) 6.1, 13.6 Hz, 2 H), 3.99 (m, 2 H), 1.79 (m, 2
H), 0.98 (d, J ) 6.7 Hz, 6 H), 0.90 (d, J ) 6.7 Hz, 6 H); 13C
NMR (100.4 MHz) δ 163.4, 162.9, 135.8, 129.7, 128.2, 127.3,
106.1, 74.9, 74.4, 72.2, 72.1, 70.0, 70.8, 32.5, 32.4, 18.7, 18.6,
18.2; IR (CCl4, cm-1) 2960, 1731; MS (EI) m/ z (rel intensity)
372 (M+, 3), 327 (41), 267 (16), 251 (100), 223 (21), 154 (34),
132 (15), 105 (35), 91 (17), 77 (24), 69 (19). Anal. Calcd for
C21H28N2O4: C, 67.72; H, 7.58; N, 7.52. Found: C, 67.66; H,
7.51; N, 7.41.
(-)-8-P h en ylm en th yl Dia zoa ceta te. A solution of chlo-
roacetyl chloride (2.67 g, 23.7 mmol) in 10 mL of dry ether
was added via a syringe pump during 2 h to a mixture of
8-phenylmenthol (5.50 g, 23.7 mmol) and pyridine (1.87 g, 23.7
mmol) in 50 mL of dry ether at 0 °C. The reaction was stirred
for 2 h at 22 °C, and the precipitate was removed by filtration.
The solution was washed with 2 M HCl (20 mL) followed by
saturated NaHCO3 (10 mL) and brine (10 mL) and dried
(MgSO4).The 8-phenylmenthyl chloroacetate obtained in quan-
titative yield on concentration of this solution was used without
purification in next step.
Crude 8-phenylmenthyl chloroacetate (7.30 g, 23.7 mmol)
in a mixture of DMF (80 mL) and 25% aqueous ammonia
solution (40 mL) was stirred for 96 h at 22 °C. The mixture
was then extracted with ether (3 × 100 mL), and the combined
organic phases was washed with saturated NaHCO3 (50 mL),
dried (Na2SO4), and concentrated to give 5.80 g crude 8-phe-
nylmenthyl glycinate (85%).
Cyclop r op a n a tion of r-m eth ylstyr en e w ith (-)-Men -
th yl Dia zoa ceta te. The mixture of CuOTf‚1/2C6H6 (12.2 mg,
0.048 mmol) and 6c (16.11 mg, 0.050 mmol) was stirred in
CH2Cl2 (2.5 mL) under argon for 30 min. This solution was
cooled to 0 °C, and R-methylstyrene (0.57 g, 4.8 mmol) was
added, followed by slow addition of a CH2Cl2 solution (3.5 mL)
of (-)-menthyl diazoacetate (1.3 g, 5.8 mmol) over 5 h via
syringe pump. The reaction mixture was allowed to warm to
room temperature, stirred for an additional 16 h, and then
quenched with a 10% aqueous solution of NH4Cl (5 mL). The
solution was diluted with Et2O (25 mL), washed with water
(5 mL) and brine (5 mL), and dried (MgSO4). Evaporation of
the solvent in vacuo afforded the crude product as a mixture
of trans and cis cyclopropanated products. The optical purity
of the crude products was determined by GC analyses.
Cyclop r op a n a tion of Vin yl Aceta te w ith (-)-Men th yl
Dia zoa ceta te. The mixture of CuOTf‚1/2C6H6 (8.4 mg, 0.033
mmol) and 6c (11.4 mg, 0.035 mmol) was stirred in CH2Cl2
(1.5 mL) under argon for 30 min. This solution was cooled to
0 °C, and vinyl acetate (280 mg, 3.3 mmol) was added, followed
by slow addition of a CH2Cl2 solution (3.0 mL) of (-)-menthyl
diazoacetate (920 mg, 4.1 mmol) over 5 h. The reaction
mixture was allowed to warm to room temperature, stirred
for an additionnal 16 h, and then quenched with a 10%
aqueous solution of NH4Cl (5 mL). The solution was diluted
with Et2O (35 mL), washed with water (5 mL) and brine (5
mL), and dried (MgSO4). Evaporation of the solvent in vacuo
afforded the crude product as a mixture of trans and cis
cyclopropanated products. The optical purity of the crude
products was determined by GC analyses.The crude product
was purified by chromatography (2% ethyl acetate in pentane)
to yield 0.60 g (64% combined yield) of cis- and trans-
cyclopropanes.
A mixture of crude 8-phenylmenthyl glycinate (5.8 g, 20.1
mmol), isopentyl nitrite (2.82 g, 24.1 mmol), and AcOH (0.34
mL, 6.00 mmol) in CHCl3 (40 mL) was heated to reflux for 4
h until the ninhydrin test was no longer positive. The mixture
was diluted with CH2Cl2 (100 mL) and washed with 1 N H2-
SO4 (25 mL), water (25 mL), saturated NaHCO3 (25 mL), and
again with water (25 mL). The organic phase was filtered
through cotton, evaporated in vacuo, and purified by chroma-
tography to afford (-)-8-phenylmenthyl diazoacetate (4.4 g,
73%) as yellow oil: [R]25 ) -13.7 (c ) 1.25, CHCl3); 1H NMR
D
(400 MHz) δ 7.27 (m, 4 H), 7.15 (m, 1 H), 4.00 (td, J ) 10.7,
4.4 Hz, 1 H), 4.21 (s, 1 H), 2.00 (ddd, J ) 12.3, 10.6, 3.5 Hz, 1
H), 1.93 (m, 1 H), 1.69 (dq, J )13.3, 3.4 Hz, 1 H), 1.64 (m, 1
H), 1.48 (m, 1 H), 1.33 (s, 3 H), 1.23 (s, 3 H), 1.10 (dd, J )12.8,
3.2 Hz, 1 H), 0.99 (app d, J ) 10.8 Hz, 1 H), 0.92 (m, 1 H),
0.87 (d, J ) 6.4 Hz, 3 H); 13C NMR (100.4 MHz) δ 151.4, 127.8,
125.3, 125.0, 74.7, 50.8, 46.2, 42.1, 39.7, 34.5, 31.3, 27.7, 26.6,
24.9, 21.7; IR (CDCl3, cm-1) 2112, 1681; MS (EI) m/ z (rel
intensity) 272 (M+ - 28, 10), 119 (100), 105 (13), 91 (29), 69
(10). Anal. Calcd for C18H24N2O2: C, 71.97; H, 8.05; N, 9.33.
Found: C, 71.65; H, 8.06; N, 9.10.
Gen er a l P r oced u r e for th e Asym m etr ic Cyclop r op a -
n a tion . A mixture of CuOTf‚1/2C6H6 (3.5 mg, 0.014 mmol)
and ligand 6c (4.7 mg, 0.015 mmol) was stirred in dry CH2Cl2
(1 mL) under argon for 30 min. This solution was cooled to 0
°C, and styrene (1.56 g, 14 mmol) was added, followed by slow
addition of a CH2Cl2 solution (2.5 mL) of ethyl diazoacetate
(159 mg, 1.4 mmol) over 5 h via syringe pump. The reaction
mixture was allowed to warm to room temperature and was
stirred for an additional 16 h, and then quenched with a 10%
aqueous solution of NH4Cl (5 mL). The solution was diluted
with Et2O (25 mL), washed with water (5 mL) and brine (5
mL), and dried (MgSO4). The solvents were evaporated in
vacuo, which afforded the crude cyclopropanated products as
a mixture of cis and trans isomers. The ratio of trans to cis
Ackn owledgm en t. Financial support from the Swed-
ish Natural Science Research Council and Tryggers
foundation are gratefully acknowledged. We are grate-
ful to Christel Carbonell, undergraduate research stu-
dent from the University of Aix-Marseille III, St J e´roˆme
(France), for her contribution to this work, and to Prof.
David Tanner for helpful suggestions during the prepa-
ration of the manuscript.
J O962021A