An efficient and general one-pot method for the synthesis of chiral bis(oxazoline) and pyridine bis(oxazoline) ligands

Article abstract of DOI:10.1055/s-2005-872672

An expeditious method for the synthesis of chiral box and pybox ligands is reported. The approach is based on a one-pot condensation reaction of chiral β-amino alcohols with a dinitrile using stoichiometric or catalytic amounts of zinc triflate. Yields gr

Full text of DOI:10.1055/s-2005-872672

LETTER
2321
An Efficient and General One-Pot Method for the Synthesis of Chiral
Bis(oxazoline) and Pyridine Bis(oxazoline) Ligands
b
a
a
b
b
a
a
a
E
A
fficient
O
ne-Pot
.
S
ynthesis of
C
Box and Pybox ornejo, J. M. Fraile, J. I. García,* M. J. Gil, V. Martínez-Merino, J. A. Mayoral, E. Pires,* I. Villalba
a
Departamento de Química Orgánica, Facultad de Ciencias, Instituto de Ciencia de Materiales de Aragón and Instituto Universitario de
Catálisis Homogénea, Universidad de Zaragoza-CSIC, 50009 Zaragoza, Spain
Fax +34(976)762077; E-mail: jig@unizar.es
b
Departamento de Química Aplicada, Universidad Pública de Navarra, 31006 Pamplona, Spain
Received 19 May 2005
acid derivatives to form the bis(hydroxy)amide
derivative. The hydroxyl groups are then activated and the
resulting intermediate is cyclized to provide the bis(ox-
Abstract: An expeditious method for the synthesis of chiral box
and pybox ligands is reported. The approach is based on a one-pot
condensation reaction of chiral b-amino alcohols with a dinitrile
using stoichiometric or catalytic amounts of zinc triflate. Yields
azoline) ligand. Several activating agents have been de-
greater than 90% are obtained in many cases without the need for scribed to favor the cyclization,6 but most of the papers
–8
further purification of the product.
describing non-commercial bis(oxazolines) ligands use
Key words: bis(oxazolines), pybox, synthesis, zinc triflate, tolu- the Evans method.9
ene, chiral ligands
It is worth noting that this method consists of a two-step
process in which several purifications are needed, with
overall yields that can reach 80%, but are often lower.
Chiral bis(oxazolines) (box) are widely used in the asym-
metric catalysis of an increasing variety of organic reac-
tions including cyclopropanations, Diels–Alder, ene,
aziridination, allylic substitutions, Mukaiyama–Michael,
Mukaiyama aldol reactions, to mention just a few. Box
systems bearing a one-carbon spacer between the two
oxazoline rings are the most frequently used.
A convenient synthesis of bis(oxazoline) ligands was de-
scribed by Lehn10 and Pfalz. The treatment of malono-
nitrile with anhydrous HCl in ethanol afforded the
corresponding imidate salt. Condensation with an optical-
ly active b-amino alcohol in dichloromethane furnished
the bis(oxazoline) in good yields. Unfortunately, this
method does not work with other spacers, for instance
when the methylene spacer is alkyl-substituted.
6
1
,2
Since 1997 many attempts have also been made to support
box ligands in heterogeneous systems.3
Bolm and coworkers11 reported the condensation of
phthalonitrile and 1,2-ethanedinitrile with an excess (1:3)
of several b-amino alcohols using chlorobenzene as sol-
The introduction of a third coordinating atom into the link
between the two oxazoline rings has led to the develop-
ment of tridentate bis(oxazoline) ligands, among which
pyridine bis(oxazoline) (pybox) derivatives represent the
best known examples. These ligands have been success-
fully used in the same types of reactions as bis(oxazol-
vent and catalytic amounts of ZnCl (5%). The best results
2
were obtained for bis(oxazolines) bearing a phenylene
spacer between the oxazoline rings, but moderate yields
were obtained when the spacer was an ethylene group.
The need for an excess of the amino alcohol, the moderate
yields, the need for purification of the reaction mixture
and the frequent loss of the starting enantiomerically pure
amino alcohol are significant drawbacks of this method.
1
,4
ines), and our group has described for the first time the
5
support of this family of ligands.
Interest in box and pybox ligands has increased due to the
wide variety of new compounds that have been synthe-
sized. Modifications include different substituents on the
oxazoline rings or variation of the nature and size of the
spacer between oxazoline rings. Such changes open new
possibilities for catalytic applications of these materials.
The same method has also been used for the synthesis of
12
pybox ligands but the yields were no higher than 65%.
Pybox ligands have traditionally been prepared in multi-
13
step processes by the method reported by Nishiyama.
The need for several purifications of the intermediate
products results in yields no higher than 70%.
Recently, we have focused our attention in the synthesis
of these ligands in order not only to improve yields but
also to have a general method to design tailored box and
pybox ligands.
Other synthetic methods have been reported but the yields
14
always range from moderate to low.
If we review synthetic methods for box ligands, most of
them follow a general route in which the first step is the
condensation of the chiral amino alcohol with nitriles or
The aim of this paper is to describe a general method for
the one-pot synthesis of these types of ligands with high
or quantitative yields, without loss of chiral amino alcohol
and avoiding the need for product purification.
Optimization of the synthetic conditions was started by
SYNLETT 2005, No. 15, pp 2321–2324
Advanced online publication: 07.09.2005
DOI: 10.1055/s-2005-872672; Art ID: G13005ST
1
9
.0
9
.2
0
0
5
11
applying the Bolm method for 2,2-bis[4-(S)-phenyl-1,3-
oxazolinyl)propane] (1). Condensation of 2,2-dimethyl-
malononitrile (1 mmol) with (S)-phenylglycinol (3 mmol)
©
Georg Thieme Verlag Stuttgart · New York

2
322
A. Cornejo et al.
LETTER
using ZnCl (5%) in chlorobenzene furnished a mixture of Table 1 Results Using Zinc Triflate in the Condensation of 2,2-
2
Dimethylmalononitrile and (S)-Phenylglycinol in Toluene
the desired product 1a and oxazoline 2a (Scheme 1).
Chromatographic purification of the reaction mixture was
needed and a poor yield (28%) of bis(oxazoline) was
obtained. The reaction does not progress due to the forma-
Amount of Zn(OTf) (%)
Ratio of 2a:1a
Box yield (%)a
2
5
–b
0
12
65
95
tion of a strong bis(oxazoline) 1a–ZnCl complex. Bis(ox-
2
3
0
7:1
1:2.5
–c
azoline) competes favorably for the catalyst with the
nitrile group of oxazoline 2a in the reaction medium and
60
00
1
5
the reaction stops. The amount of catalyst should there-
fore be increased to obtain good yields of the bis(ox-
azoline).
1
a
b
c
1
Determined by H NMR.
Only oxazoline 2 was observed.
Only bis(oxazoline) 1 was observed.
Another drawback of Bolm’s conditions is the use of an
excess of amino alcohol, which is the most expensive and
difficult reagent to obtain. We therefore aimed to find new
conditions in which good yields were obtained, if possible
without the need for excess enantiomerically pure amino
alcohol and in a greener medium.
OH
O
O
O
NC
catalyst
solvent
NH2
+
+
N
NC
1
CN
mmol
N
N
Ph
H
H Ph
Ph
H
H Ph
3 mmol
1a
2a
Scheme 1
We first decided to change the solvent and use a non-
chlorinated and less toxic one. The long reaction times led
us to work with reaction temperatures over 100 °C and we
chose toluene as the solvent because of its appropriate
boiling point and the good solubility of the reactants. We
fixed a 1:2 ratio of 2,2-dimethylmalononitrile and amino
Figure 1 1H NMR signals showing 1a:2a ratio in synthesis reac-
tions with several amounts of Zn(OTf)2
alcohol, and tested different amounts of ZnCl in toluene.
2
Under these conditions only oxazoline 2a was obtained.
The reduced activity of ZnCl in toluene is certainly due
2
to solubility problems, so we considered other Zn salts
that would be more soluble in toluene.
OH
O
O
1
mmol Zn(OTf)2
NH2
+
NC
CN
toluene
N
N
The use of Zn(OAc) gave a complex reaction mixture
2
R
2
R'
mmol
R'
R'
R
R
containing oxazoline 2a, bis(oxazoline) 1a and a new
product. Chromatographic purification enabled us to iden-
tify that product as 2-methyl-4-phenyloxazoline. Finally,
1 mmol
1
i
t
R' = H and R = Ph (a), CH2Ph (b), Pr (c), Bu (d), Me (e), Indanyl (f)
R' = Me and R = Ph (g),
Zn(OTf) turned out to be the best catalyst for this reac-
tion. The results obtained with several proportions of the
catalyst are gathered in Table 1. When substoichiometric
2
Scheme 2
amounts of catalyst were used, oxazoline 2a appeared and latter leading to a non-previously described box
purification was required to obtain bis(oxazoline) 1a. (Scheme 2).19
Only near stoichiometric amounts of zinc triflate gave
It can be seen from the results in Table 2 that excellent
pure bis(oxazoline) 1a in very good yields (Figure 1).
yields were obtained in all cases, even when a quaternary
We therefore achieved the synthesis of the desired product amino alcohol was used. The only change required in this
in just one step, with an easy work up and almost quanti- case being an increase in reaction time.
1
6
tative yield.
To test this method we decided to try the synthesis of py-
This method was subsequently applied to the synthesis of box systems (Scheme 3).
several box systems using other enantiomerically pure b-
We first tried to optimize synthetic conditions using
amino alcohols, such as (S)-valinol, (S)-phenylglycinol,
smaller amounts of Zn(OTf) . As the need of using
2
(
S)-tert-leucinol, (S)-phenylalaninol, (1R,2S)-(+)-cis-1-
stoichiometric amounts in box synthesis was due to the
formation of strong catalyst–box complex we thought that
it could not be the case with pybox ligands.
17,18
amino-2-indanol and (S)-2-methylphenylglycinol,
the
Synlett 2005, No. 15, 2321–2324 © Thieme Stuttgart · New York

LETTER
Efficient One-Pot Synthesis of Box and Pybox
2323
Table 2 Synthesis of Box Ligands with Stoichiometric Amounts of
2
0
Zinc Triflate in Toluene
Entry
Box ligand
Reaction time (d) Box (yield, %)
1
2
3
4
5
6
7
1a
1b
1c
1d
1e
1f
2
2
2
3
2
2
3
95
100
90
90
75
100
85
1g
O
O
N
N
N
OH
Zn(OTf)2
H
3
H
R
R
NH2
+
NC
N
CN toluene
R
2
H
O
mmol
1 mmol
N
CN
i
t
R = Pr (a); Ph (b); Bu (c), indanyl (d)
N
H
4
R
Figure 2 1H NMR signals showing 3:4 ratio in synthesis reactions
of 3a with several amounts of Zn(OTf)2
Scheme 3
We started with the synthesis of i-Pr-pybox (3a) and
results showed that 5% of catalyst was enough to obtain ferent chiral amino alcohols in stoichiometric amounts.
excellent yields of pybox product 3a and no pyridine Only catalytic amounts of Zn(OTf) are needed in the case
2
oxazoline 4a is observed (Table 3 and Figure 2)
of pybox syntheses, which makes the process economical-
ly interesting. In the case box ligands, even if a stoichio-
The same study was carried out with the rest of pybox
ligands. Table 4 gathers the best conditions for each one,
best conditions meaning those in which no pyridine
metric amount of Zn(OTf) is necessary, the fact that
2
excess enantiomerically pure b-amino alcohol is not re-
quired may be economically advantageous in many cases,
given the relative prices of catalyst and some chiral amino
alcohols.
oxazoline (4) was observed, the minimum amount of
catalyst was used and good yields were obtained.21
In all cases good yields were achieved in one-pot reac-
tions and pybox ligands were isolated with no need of fur-
ther purification.
Furthermore, the need for product purification is avoided
and the method is environmentally friendly (improved
atom economy, use of a non-chlorinated solvent, reduced
amount of reaction solvent because there is only one syn-
thetic step, and solvents not used in purification steps).
We have described here an easy, efficient, general method
for the synthesis of box and pybox ligands in just one step.
This method gives high yields using a wide variety of dif-
Table 4 Optimized Conditions for the Synthesis of Pybox Ligands
with Zinc Triflate in Toluene
a,20
Table 3 Synthesis of Pybox Ligand 3a with Zinc Triflate in Toluene
Amount of Zn(OTf) (%)
Ratio of 3a:4aa
Pybox yield (%)a
Entry
R
Amount of
Yield (%)
2
Zn(OTf) (%)
2
0
0
1
5
.1
.5
1:5
3.8:1
–b
6
1
2
3
3a
3b
3c
3d
5
85
75
85
85
63
70
85
10
10
5
–b
4
a
1
Determined by H NMR.
Only pybox 3 was observed.
b
a
Reaction time 2 d.
Synlett 2005, No. 15, 2321–2324 © Thieme Stuttgart · New York

2
324
A. Cornejo et al.
LETTER
(16) General Procedure for the Preparation of Box.
Acknowledgment
A 100-mL two-necked round-bottomed flask fitted with a
reflux condenser was charged with 2,2-dimethyl
malononitrile (564 mg, 6 mmol) and zinc triflate (2, 177.4
mg, 6 mmol). The system was purged with argon and anhyd
toluene (40 mL) was added. The solution was stirred during
This work was made possible by the financial support of the Mini-
sterio de Educacion y Ciencia (Project PPQ2002-04012). A.C. and
I.V. wish to thank the MEC for a grant. We thank DSM-Deretil for
furnishing us pure (S)-2-methylphenylglycine.
5
min and a solution of the b-amino alcohol (12 mmol) in
anhyd toluene (20 mL) was added. The solution was heated
under reflux for 48 h [72 h when tert-leucinol and (S)-2-
methylphenylglycinol were used]. The system was allowed
to cool. The reaction was then washed with brine (3 × 75
References
(
(
1) McManus, H. A.; Guiry, P. J. Chem. Rev. 2004, 104, 4151.
2) Ghosh, A. K.; Mathivanan, P.; Cappiello, J. Tetrahedron:
Asymmetry 1998, 9, 1.
mL) and NaHCO (3 × 70 mL), dried with MgSO and the
3
4
solvent evaporated to give pure product.
17) (S)-2-Methylphenylglycinol was obtained from (S)-2-
(3) (a) Rechavi, D.; Lemaire, M. Chem. Rev. 2002, 104, 3467.
(
(
(
b) Fraile, J. M.; García, J. I.; Herrerías, C. I.; Mayoral, J. A.;
methylphenylglycine by the Meyers method: McKennon, M.
Harmer, M. J. Catal. 2004, 221, 532. (c) Fraile, J. M.;
García, J. I.; Herrerías, C. I.; Mayoral, J. A.; Gmough, S.;
Vaultier, M. Green Chem. 2004, 6, 93.
J.; Meyers, A. I. J. Org. Chem. 1993, 58, 3568.
1
18) (S)-2-Methylphenylglycinol: H NMR (400 MHz, CDCl ):
3
d = 1.42 (s, 3 H), 2.06 (br s, 3 H), 3.59 (dd, 2 H, J = 10.68
(
(
4) Desimoni, G.; Faita, G.; Quadrelli, P. Chem. Rev. 2003, 108,
1
Hz, J = 13.11 Hz), 7.19 (t, 1 H, J = 6.57 Hz), 7.35 (dd, 2 H,
3119.
2
1
3
J = 6.57 Hz, J = 8.08 Hz), 7.45 (d, 2 H, J = 8.08 Hz).
C
5) (a) Cornejo, A.; Fraile, J. M.; García, J. I.; Gil, M. J.;
Legarreta, G.; Luis, S. V.; Martínez-Merino, V.; Mayoral, J.
A. Org. Lett. 2002, 4, 3927. (b) Cornejo, A.; Fraile, J. M.;
García, J. I.; Gil, M. J.; Luis, S. V.; Martínez-Merino, V.;
Mayoral, J. A. J. Org. Chem. 2005, 24, 3448.
NMR (100 MHz, CDCl ): d = 27.09, 56.38, 71.71, 125.29,
3
2
5
126.85, 128.49, 146.31. [a]D 10.97 (c 0.75, HCl 1 N). Anal.
Calcd for C H NO: C, 71.49; H, 8.67; N, 9.26. Found: C,
9
13
7
1.40; H, 8.65; N, 9.31.
1
(
19) 2,2-Bis[4-(S)-methyl-4-phenyl-1,3-oxazolinyl]propane: H
(
(
(
(
6) Müller, D.; Umbricht, G.; Weber, B.; Pfalz, A. Helv. Chim.
Acta 1991, 74, 232.
7) Corey, E. J.; Imai, N.; Zhang, H. J. Am. Chem. Soc. 1991,
NMR (400 MHz, CDCl ): d = 1.61 (s, 3 H), 1.65 (s, 3 H),
3
4
.25 (dd, 2 H, J = 8.08 Hz, J = 25.00 Hz), 7.19 (m, 1 H),
1
2
1
3
7.30 (m, 2 H), 7.35 (m, 2 H). C NMR (100 MHz, CDCl3):
1
13, 728.
8) Lowental, R. E.; Abiko, A.; Masamune, S. Tetrahedron Lett.
990, 31, 6005.
d = 24.47, 28.69, 38.78, 72.47, 80.84, 125.38, 126.85,
2
5
1
28.42, 146.75, 168.50. [a]D –5.98 (c 1, CH Cl ). Anal.
1
2 2
Calcd for C H N O : C, 76.21; H, 7.23; N, 7.73. Found: C,
9) Evans, D. A.; Peterson, G. S.; Johnson, J. S.; Barnes, D. M.;
23 26
2
2
7
6.06; H, 7.76; N, 7.56..
Campos, K. R.; Woerpel, K. A. J. Org. Chem. 1998, 63,
1
(
(
20) H NMR and [a] data for box and pybox ligands bearing
4541.
D
phenyl, benzyl, isopropyl, tert-butyl or indanyl groups were
(
(
(
(
10) Hall, J.; Lehn, J. M.; DeCian, A.; Fischer, J. Helv. Chim.
Acta 1991, 74, 1.
11) Bolm, C.; Weickhardt, K.; Zehner, M.; Ranff, T. Chem. Ber.
7
,9,13a,22–24
compared with those described in the literature.
Purity of those products was verified by elemental analysis.
21) General Procedure for the Preparation of Pybox.
A 100-mL two-necked round-bottomed flask fitted with a
reflux condenser was charged pyridine-2,6-dicarbonitrile
1991, 124, 1173.
12) Chelucci, G.; Deriu, S.; Pinna, G. A.; Saba, A.; Valenti, R.
Tetrahedron: Asymmetry 1999, 10, 3803.
(
774 mg, 6 mmol) and the adequate amount of zinc triflate
13) (a) Nishiyama, H.; Kondo, M.; Nakamura, T.; Itoh, K.
Organometallics 1991, 10, 500. (b) Iwasa, S.; Nakamura,
H.; Nishiyama, H. Heterocycles 2000, 52, 939. (c) Iwasa,
S.; Takezawa, F.; Tuchiya, Y.; Nishiyama, H. Chem.
Commun. 2001, 59. (d) Nishiyama, H.; Itoh, Y.; Sugawara,
Y.; Matsumoto, H.; Aoki, K.; Itoh, K. Bull. Chem. Soc. Jpn.
(
5% mol for 3a and 3d and 10% mol for 3b and 3c). The
system was purged with argon and anhyd toluene (40 mL)
was added. The solution was stirred during 5 min and a
solution of the b-amino alcohol (12 mmol) in anhyd toluene
(
4
20 mL) was added. The solution was heated under reflux for
8 h. The system was allowed to cool, and reaction was
diluted with 50 mL of EtOAc. The solution was then washed
1995, 68, 1247. (e) Iwasa, S.; Tsushima, S.; Shimada, T.;
Nishiyama, H. Tetrahedron 2002, 58, 227.
with brine (3 × 75 mL) and NaHCO (3 × 70 mL), dried with
(
14) (a) Desimoni, G.; Faita, G.; Guala, M.; Pratelli, C.
Tetrahedron: Asymmetry 2002, 13, 1651. (b) Desimoni, G.;
Faita, G.; Filippone, S.; Mella, M.; Zampori, M.; Zema, M.
Tetrahedron 2001, 57, 10203. (c) Totleben, M. J.; Prasad, J.
S.; Simpsons, J. H.; Chan, S. H.; Vanyo, D. L.; Kuehner, D.
E.; Deshpande, R.; Kodersa, G. H. J. Org. Chem. 2001, 66,
3
MgSO and the solvent evaporated to give pure product.
4
(
22) Evans, D. A.; Woerpel, K. A.; Hinman, M. M.; Faul, M. M.
J. Am. Chem. Soc. 1991, 113, 726.
23) Von Matt, P.; Lloyd Jones, G. C.; Minidis, A. B. E.; Pfaltz,
A.; Macko, L.; Neuburger, M.; Zehnder, M.; Ruegger, H.;
Pregosin, P. S. Helv. Chim. Acta 1995, 78, 265.
(
1057. (d) Davies, I. W.; Gerena, L.; Lu, N.; Larsen, R. D.;
(
24) Davies, I. W.; Senanayake, C. H.; Larsen, R. D.; Verhoeven,
Reider, P. J. J. Org. Chem. 1996, 61, 9629.
1
T. R.; Reider, P. J. Tetrahedron Lett. 1996, 37, 813.
(
15) The H NMR signals of the box-ZnCl complex were
2
observed.
Synlett 2005, No. 15, 2321–2324 © Thieme Stuttgart · New York