Synthesis of new C1-symmetric bis(oxazoline) ligands with a chelating sidearm for stereoselective Mukaiyama aldol condensations

Article abstract of DOI:10.1016/j.jorganchem.2007.02.009

Novel C₁-symmetric bis(oxazoline) ligands with a secondary binding sidearm were prepared in enantiomerically pure form in good yields, in only four steps starting from commercially available reagents. These new chiral ligands were tested in the

Full text of DOI:10.1016/j.jorganchem.2007.02.009

Journal of Organometallic Chemistry 692 (2007) 2120–2124
www.elsevier.com/locate/jorganchem
Communication
Synthesis of new C1-symmetric bis(oxazoline) ligands with
a chelating sidearm for stereoselective Mukaiyama aldol condensations
a
a,
a
b
Simonetta Orlandi , Maurizio Benaglia *, Gianmaria Dell’Anna , Giuseppe Celentano
a
Centro di Eccellenza CISI, Dipartimento di Chimica Organica e Industriale, via Golgi 19, I-20133 Milano, Italy
Istituto di Chimica Organica, Facolta’ di Farmacia, Universita’ degli Studi di Milano, via Golgi 19, I-20133 Milano, Italy
b
Received 11 January 2007; received in revised form 7 February 2007; accepted 7 February 2007
Available online 20 February 2007
Abstract
Novel C1-symmetric bis(oxazoline) ligands with a secondary binding sidearm were prepared in enantiomerically pure form in good
yields, in only four steps starting from commercially available reagents. These new chiral ligands were tested in the enantioselective
Mukaiyama aldol condensation between the trimethylsilyl keteneacetal of methyl isobutyrate and a non-chelating substrate such as benz-
aldehyde to afford the product in up to 55% ee.
Ó 2007 Elsevier B.V. All rights reserved.
Keywords: Chiral bis-oxazolines; Copper complexes; Enantioselective catalysis; Mukaiyama aldol condensation; Chelation
1. Introduction
concave pocket around the metal center and to synthesize
a more stable catalytic system. Alternatively tridentate bis
(oxazolinyl)-type ligands have been designed by introduc-
ing a donor atom into the link connecting two chiral oxaz-
oline rings [4]. Among the different structures it is worth
mentioning Kanemasa’s ligand 3 [5], Zhang’s bis(oxazoli-
nyl-methyl)amine 4 [6] and specially the ‘‘pybox’’ ligands
5 first developed by Nishiyama [7] (Scheme 1).
The versatility of chiral bis(oxazolines) to act as ligands
in a variety of catalytic enantioselective transformations is
well recognised [1]. For example, copper (II) complexes of
C2-symmetric ligand type 1 were shown to be efficient cat-
alysts for several stereoselective C–C bonds formation
reactions [2] (see Scheme 1).
However, with these chiral catalytic systems very high
enantioselectivities are obtained only by employing sub-
strates which can participate in catalyst chelation. The
strict requirement of using a bidentate substrate that con-
tributes in determining a well-defined geometry of the cat-
alyst/substrate complex represents an obvious limitation to
the methodology’s generality.
However it must be noted that also these tridentate
catalytic systems usually promote high enantioselective
C–C bond formation reactions with bidentate chelating
structures.
2. Results and discussion
The idea to introduce an extra chelating element into the
chiral ligand has been recently explored and has led to the
development of tris(oxazoline) [3] ligand types 2 (see
Scheme 1). The reasons beyond the preparation of this
class of molecules were basically to build a deeper chiral
In order to overcome this substrate limitation we decided
to explore a different approach. The idea was to synthesize a
Evan’s bis(oxazoline) type 1 ligand but with a chelating
sidearm, to afford a structure of type 6; the new ligand
should bind the copper (II) ion following the coordination
behaviour of the C2-symmetric analogs whose complexes
geometry has been well studied [8] (see Scheme 2). The
active catalytic species has now only one coordination site
*
Corresponding author. Tel.: +39 0 25031 4171; fax: +39 0 25031 4159.
E-mail address: maurizio.benaglia@unimi.it (M. Benaglia).
0022-328X/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.02.009

S. Orlandi et al. / Journal of Organometallic Chemistry 692 (2007) 2120–2124
2121
R
N
O
O
O
O
O
O
O
O
N
N
N
N
R
R
N
N
R
R
R
R
1
3
2
O
X
X
O
O
O
N
H
N
N
N
N
N
R
R
R
R
5
4
Scheme 1. Oxazoline-type ligands in asymmetric catalysis.
for the substrate that should be attacked by the reagent
under the stereocontrol exerted by the substituents at the
stereocentres of the oxazoline moieties [9].
or 9, in 81% and 85% yield, respectively. In order to build
the oxazoline ring bringing the chelating sidearm we
decided to employ the amino alcohol 10, easily prepared
in only three steps from the commercially available 4-ben-
zyl-(S)-aspartic acid [10]. After hydrolysis of esters 8 and 9,
the corresponding carboxylic acid amides were condensed
to the amino-alcohol 10 to give the corresponding dissimet-
ric bis amides 12 and 11 basically in quantitative yields, as
pure compounds without the need of chromatographic
purification [11].
Several methodologies were attempted for the closure of
bis-amides to the corresponding bis(oxazolines) 13 and 14
[12]. Finally, best results were obtained by reacting 11
and 12 with methansulfonyl chloride and triethylamine to
convert them in the corresponding mesylate derivatives
which were treated in situ with triethylamine and a cata-
lytic amount of DMAP to afford bis(oxazolines) 13 and
Here, we wish to report the synthesis of new C1-symmet-
ric bis(oxazoline) ligands with a secondary binding element
to be employed in promoting reactions that do not require
the use of bidentate substrates. Preliminary studies on the
use of these novel chiral ligands in the Mukaiyama aldol
condensation of trimethylsilyl keteneacetal of methyl
isobutyrate with benzaldehyde will be also described.
The synthesis of the new ligands follows the classical
route that involves the reaction of a dimethyl malonic acid
derivative with amino-alcohols to give a bis-amide interme-
diate that is converted into bis-oxazoline (see Scheme 3). In
our synthesis the first step is the EDC- promoted conden-
sation of dimethyl malonic acid monomethylester 7 with
(S)-phenyl glycinol or (S)-t-leucinol to afford the amide 8
O
O
O
O
N
N
N
N
R
R
R
O
6
1
Y
Y= OR, NRR'
O
O
O
O
N
O
H
N
OBn
N
O
H
N
Cu
Cu
si face
Scheme 2. Novel bis(oxazoline) with a chelating sidearm.

2122
S. Orlandi et al. / Journal of Organometallic Chemistry 692 (2007) 2120–2124
R
HO
NH
HO
OMe
1) EDC, HOBT (1 eq.), CHCl3
R-CH-(NH2)-CH2OH (1.1 eq.)
O
O
O
O
OH
2 )NaOH (3 eq.),
THF/H2O 1/1
7
8 R=Ph
9 R=tBu
+
-
NH TFA
BnO
3
EDC, HOBT (1 eq.),
TEA (2.5 eq.), CHCl3
RT, 24 h
O
OH
10
O
O
MsCl (2.2 eq.)
TEA (4.4 eq.)
H
N
R
BnO
NH
N
N
R
DMAP (cat.),
1,2-DCE
O
O
O
O
OH
OH
BnO
13 R=tBu
14 R=Ph
11 R=tBu
12 R=Ph
Scheme 3. Synthesis of new bis(oxazoline) ligands 13 and 14.
Table 1
14 in 53% and 45% yields, respectively after chromato-
graphic purification.
Following our interest in studying enantioselective reac-
tions promoted by copper complexes [13] the new enantio-
merically pure ligands were tested in the Cu(II) catalysed
Mukaiyama aldol condensation [14].
Mukaiyama aldol condensation promoted by Cu(OTf)2 complexes
Entry
Ligand
Reaction cond.
Yielda (%)
ee (%)b
1
2
3
4
5
a
1
2h, ꢀ78 °C; 18 h, RT
2 h, ꢀ78 °C; 18 h, RT
2 h, ꢀ78 °C; 18 h, RT
20 h, ꢀ78 °C
43
83
73
68
58
19
11
5
24
17
13
14
1
The condensation of the trimethylsilyl keteneacetal of
methyl isobutyrate with benzaldehyde was chosen as test
reaction, in the presence of 10% of the catalyst [15] (Scheme
4 R = Ph). First, the catalytic ability of Cu(II) trifluorom-
ethane sulfonate complexes was studied (Table 1). After 3 h
complexation of Cu(OTf)2 with the bis(oxazoline) ligand,
benzaldehyde was added, followed after 15 min by the
trimethylsilyl keteneacetal of methyl isobutyrate.
For sake of comparison the reaction was run also in the
presence of Evans’ bis(oxazoline) type 1 (Scheme 1, R = t-
Bu) complexed to Cu(OTf)2.
Unfortunately, the results were extremely disappointing.
By running the reaction in different temperature conditions
the product (S)-methyl 2,2-dimethyl-3-hydroxy-3-phenyl-
propanoate 15 was obtained in good yields but with very
low enantioselectivities, comparable to those afforded by
bis(oxazoline) 1.
13
20 h, ꢀ78 °C
Isolated yields after flash chromatography.
As determined by HPLC on a chiral stationary phase.
b
Even running the reaction at lower temperature did not
produce any appreciable improvement as the enantioselec-
tivity was not significantly increased [16]. It is interesting to
observe that both new ligands 13 and 14 promoted the
reaction at room temperature with better yields than ligand
1 (entries 1–3, Table 1), suggesting maybe an enhanced sta-
bilization of the catalytically active complex at room
temperature.
The catalytic behaviour of Cu(SbF6)2 as copper (II)
source [17] was then investigated (Scheme 4, R = Ph,
L = SbF6).
At low temperature the new ligands 13 and 14 promoted
the reaction once again in better yields than ligand 1
(entries 1 and 2 vs. entry 3). When the reaction was run
for 2 h at ꢀ78 °C and allowed to rinse to room temperature
for 18 h the enantioselectivity increased up to 41% ee,
always maintaining a very good chemical efficiency (entries
4 and 5, Table 2).
Since the trend seemed to suggest better performances of
the catalysts at room temperature the aldol condensation
was run at 25 °C for 20 h; in these conditions the ligand
OH
O
OTMS
OMe
10 mol%
COOMe
Ph
+
Ph
H
L* / Cu(L)2
15
Scheme 4. Mukaiyama aldol condensation promoted by Cu(II)
complexes.

S. Orlandi et al. / Journal of Organometallic Chemistry 692 (2007) 2120–2124
2123
Table 2
were referenced to tetramethylsilane (TMS) at 0.00 ppm.
13C NMR spectra were recorded at 75 MHz and were
referenced to 77.0 ppm in CDCl3. Optical rotations were
measured at the Na-D line in a 1 dm cell at 22 °C. IR
spectra were recorded on thin film or as a solution in
CH2Cl2.
Mukaiyama aldol condensation promoted by Cu(SbF6)2 complexes
Entry
Ligand
Reaction cond.
Yielda (%)
ee (%)b
1
2
3
4
1
20 h, ꢀ78 °C
20 h, ꢀ78 °C
20 h, ꢀ78 °C
2 h, ꢀ78 °C;
18 h, RT
63
64
81
85
5
7
15
21
13
14
13
Synthesis of bis(oxazoline) 14: To a stirred solution of
bis-amide 12 (0.44 g, 1 mmol) and triethylamine
(0.63 mL, 4.4 mmol) in 1,2-dichloroethane (5 cm3) kept
under nitrogen and cooled at 0 °C, mesyl chloride
(0.18 mL, 2.2 mmol) was added dropwise. The mixture
was stirred at RT for 2 h. A solution of triethylamine
(0.63 mL, 4.4 mmol) and a catalytic amount of DMAP in
1,2-dichloroethane (2 mL) was added dropwise and the
reaction mixture was allowed to stir at 50 °C for 40 h.
The organic phase was concentrated under vacuum to give
the crude product, that was purified by chromatography
with a 6:4 CH2Cl2:AcOEt mixture as eluant. The product
(0.18 g, 0.45 mmol, 45% yield) was a pale yellow thick oil
that solidified on standing in the freezer. It had ½aꢁD22
ꢀ27.4 (c 0.4 in CH2Cl2). Anal. Calc. for C24H26N2O4
requires: C, 70.92; H, 6.45; N, 6.89. Found: C, 71.05; H,
6.51; N, 6.95%. IR(DCM): m 1712, 1655, 1265 cmꢀ1 1H
NMR (300 MHz, CDCl3): d 7.30–7.20 (m, 10 H), 5.16
(m, 1H), 5.10 (s, 2H), 4.55 (t, J = 6.5 Hz, 1H), 4.45 (m,
2H), 4.15 (m, 2H), 2.92 (dd, J = 3.5 Hz, J = 12.5 Hz,
1H), 2.48 (dd, J = 5.5 Hz, J = 12.5 Hz, 1 H), 1.55 (s, 3H),
1.52 (s, 3H). 13C NMR (75 MHz, CDCl3): d 170.8, 170.1,
169.8, 142.3, 140.0, 131.1, 130.0, 128.6, 128.0, 127.7,
127.5, 127.0, 126.5, 75.4, 73.0, 69.4, 66.5, 62.5, 40.1, 39.0,
24.4, 24.2.
Aldol condensation: To a stirred solution of Box 14
(0.020 g, 0.05 mmol) in dry CH2Cl2 (2.0 cm3) CuCl2
(0.007 g , 0.05 mmol) was added and the mixture stirred
at 23 °C for 3 h. Ag(SbF6) (0.034 g, 0.1 mmol) was added
and the mixture stirred at 23 °C for 12 h. To the complex
thus obtained benzaldehyde (0.05 mL, 0.5 mmol) was
added first and after 15 min the silyl keteneacetal
(0.16 mL, 0.80 mmol) was added too. After the mixture
was stirred for 20 h at 23 °C, the organic phase was concen-
trated under vacuum and the residue was purified by flash
chromatography with a hexanes:AcOEt 8:2 mixture as elu-
ant to afford the product (S)-15 as a white solid; m.p. 70–
71 °C (lit.: [19] 71–72 °C). The ee was established by HPLC
analysis on a Chiralpak AD column, flow rate 0.8 cm3/min,
k = 210 nm; hexane:ethanol 95:5; tR: 14.3 min (minor) and
20.4 min (major). The product had 1H NMR data identical
to those reported [23].
5
6
7
8c
9d
a
14
13
14
14
14
2 h, ꢀ78 °C; 18 h, RT
87
69
71
72
75
41
23
55
53
55
20 h, RT
20 h, RT
20 h, RT
20 h, RT
Isolated yields after flash chromatography.
As determined by HPLC on a chiral stationary phase.
20% of catalyst was used.
b
c
d
50% of catalyst was used.
14 promoted the reaction in slightly lower yields but with
enantioselectivity up to 55% ee (entry 7, Table 2) [18].
Increased catalyst loading did not allow to improve the
enantioselectivity (entries 8–9). The reaction of a non-
aromatic aldehyde was also performed; the silyl keteneac-
etal addition to hydrocinnamaldehyde (Scheme 4, R =
PhCH2CH2) in the same conditions of entry 5 lead to the
aldol product in similar yield but lower enantioselectivity
(65% yield, 39% ee) [19].
It must be noted that the copper (II) counterion plays an
important role; it has been already demonstrated that the
stereochemical outcome may be heavily influenced by the
presence or the absence of the counterion in the ligand/
copper complex [20]. A different geometry of the two com-
plexes 14/Cu(OTf)2 and 14/Cu(SbF6)2 might be responsible
for the different stereochemical efficiency [21].
3. Conclusions
In conclusion the synthesis of new enantiomerically pure
C1-symmetric bis(oxazolines) with a secondary binding ele-
ment has been successfully developed [22]. The new chiral
ligands should allow the promotion of stereoselective reac-
tions with substrates that do not require the presence of a
chelating element.
In these preliminary studies on the Mukaiyama aldol
condensation of trimethylsilyl keteneacetal of methyl
isobutyrate with benzaldehyde the product has been
obtained in very good chemical yields and with an ee up
to 55%. Although the level of enantioselectivity is quite
modest, the new C1-symmetric bis(oxazolines) have shown
a certain ability in stereodirecting the aldol reaction and
represent an interesting starting point for further develop-
ments in order to improve the stereoselectivity of such cat-
alytic reactions.
Acknowledgements
This work was supported by MIUR (Rome) within the
National Project ‘‘Nuovi metodi catalitici stereoselettivi e
sintesi stereoselettiva di molecole funzionali’’ and C.I.N.-
M.P.I.S. (Consorzio Interuniversitario Nazionale Metodol-
ogie e Processi Innovativi di Sintesi). We thank Prof.
Franco Cozzi for valuable discussion.
4. Experimental
1
General: H NMR spectra were recorded at 300 MHz
in chloroform-d (CDCl3) unless otherwise stated, and

2124
S. Orlandi et al. / Journal of Organometallic Chemistry 692 (2007) 2120–2124
References
S. Orlandi, M. Benaglia, F. Cozzi, Tetrahedron Lett. 45 (2004) 1747–
1749;
[1] A.K. Ghosh, P. Mathivanan, J. Cappiello, Tetrahedron: Asymmetry
9 (1998) 1–45.
M. Benaglia, D. Negri, G. Dell’Anna, Tetrahedron Lett. 45 (2004)
8705–8708;
[2] (a) D.A. Evans, T. Rovis, J.S. Johnson, Pure Appl. Chem 71 (1998)
1407–1421;
S. Orlandi, F. Colombo, M. Benaglia, Synthesis (2005) 1689–1692;
F. Colombo, M. Benaglia, S. Orlandi, F. Usuelli, J. Org. Chem. 71
(2006) 2064–2070.
(b) K.A. Jorgensen, M. Johannsen, S. Yao, H. Audrian, J. Thorha-
uge, Acc. Chem. Res. 32 (1999) 605–619;
[14] For a PEG-supported bis(oxazoline)-copper(II) promoted reaction in
water see M. Benaglia, M. Cinquini, F. Cozzi, G. Celentano, Org.
Biomol. Chem. 2 (2004) 3401–3407.
[15] For the same reaction promoted by bis(oxazoline)-copper (II)
complexes in aqueous solvents see S. Kobayashi, S. Nagayama, T.
Busujiama, Tetrahedron 55 (1999) 8739–8746.
(c) D.A. Evans, J.S. Johnson, Acc. Chem. Res 33 (2000) 325–345;
(d) H.A. McManus, P.J. Guiry, Chem. Rev. 104 (2004) 4151.
[3] (a) J. Zhou, Y. Tang, J. Am Chem. Soc 124 (2002) 9030–9031;
(b) S. Bellemin-Laponnaz, L.H. Gade, Angew. Chem., Int. Ed. 41
(2002) 3473–3475.
[4] H.A. McManus, P.J. Guiry, J. Org. Chem 67 (2002) 8566–8573.
[5] V. Kanemasa, Y. Oderatoshi, S. Sakaguchi, H. Yamamoto, J. Tanaka,
E. Wada, D.P. Curran, J. Am. Chem. Soc. 120 (1998) 3074–3083.
[6] Y. Jiang, Q. Jiang, X. Zhang, J. Am. Chem. Soc. 120 (1998) 3817–
3818.
[16] At ꢀ78 °C ligand 14/ Cu(OTf)2 complex catalysed the reaction in 41%
yield basically without stereoselection.
[17] D.A. Evans, G.S. Peterson, J.S. Johnson, D.M. Barnes, K.R.
Campos, K.A. Woerpel, J. Org. Chem. 63 (1998) 4541–4544.
[18] In our hands the Cu(SbF6)/ligand 1 complex at RT promoted the
reaction with 7% ee. The result seems to suggest that the addition of a
sidearm in the ligand has some beneficial effect on the enantioselec-
tivity of the process. In this sense the figure of Scheme 2 represents a
very preliminary working model that accounts for the formation of
the product (S)-15.
[19] At the present based on these preliminary results it would be
premature to invoke the presence of p–p interactions.
[20] For a discussion on the stereochemical models for bis(oxazolines)
copper(II) complexes see Refs. [8,2c].
[7] (a) H. Nishiyama, Y. Itoh, H. Matsumoto, S.-B. Park, K. Itoh, J.
Am. Chem. Soc. 116 (1994) 2223–2224;
(b) D.A. Evans, Z.K. Sweeney, T. Rovis, J.S. Tedrow, J. Am. Chem.
Soc. 123 (2001) 12095–12096, and references cited.
[8] D.A. Evans, M.C. Kozlowski, J.A. Murry, C.S. Burgey, K.R.
Campos, B.T. Connell, R. Staples, J. Am Chem. Soc. 121 (1999)
669–685;
D.A. Evans, J.A. Murry, M.C. Kozlowski, J. Am Chem. Soc. 118
(1996) 5814.
[9] For a recent report about the use of secondary binding sites for the
asymmetric cyclopropanation of furans see M. Schinnerl, C. Bohm,
M. Seitz, O. Reiser, Tetrahedron: Asymmetry 14 (2003) 765–771.
[10] A.B. Maude, A.P. Mehrotra, D. Gani, J. Chem. Soc. Perkin Trans. I
(1997) 2513–2522.
[11] All the new products were fully characterized and show spectral data
in agreement with the proposed structure.
[12] Other combinations such as tosyl chloride/triethylamine or the use of
Burgess reagent gave the product in lower yields or did not work at
all.
[21] Several issues have to be addressed: for example the copper
coordination of a C1-symmetric bis(oxazoline) might bring about
the formation of two different complexes; our working hypothesis
is that for steric reasons one should be favoured over the other;
of course also the control of the approach of the aldehyde is
another point that deserves further studies. Preliminary attempted
NMR studies showed the presence of more than one species in
solution.
[22] For other catalysts prepared following the same design principles see:
J. C-D. Le, B.L. Pagenkopf, Org. Lett. 6 (2004) 4097.
[23] F. Fringuelli, O. Piermatti, F. Pizzo, J. Org. Chem. 60 (1995) 7006–
7009.
[13] A. Puglisi, M. Benaglia, R. Annunziata, A. Bologna, Tetrahedron
Lett. 44 (2003) 2947–2951;