A New CopperII/Isopropylidene-2,2-bis(oxazoline) catalyst and Its stable reactive complex with acryloyloxazolidinone in enantioselective reactions

Article abstract of DOI:10.1002/chem.200900908

A real ghost: The [TIPS-box/Cu-(OTf)₂] complex (box = isopropylidene-2,2-bis(oxazoline)) is a catalyst of the Diels-Alder and the 1,3-dipolar cycloaddition between acryloyl-1,3-oxazolidin-2-one with cyclopentadiene and diphenylnitrone with ee v

Full text of DOI:10.1002/chem.200900908

DOI: 10.1002/chem.200900908
A New CopperII/Isopropylidene-2,2-bis(oxazoline) Catalyst and Its Stable
Reactive Complex with Acryloyloxazolidinone in Enantioselective Reactions
Giovanni Desimoni,*[a] Giuseppe Faita,[a] Marco Toscanini,[a] and Massimo Boiocchi[b]
À
The formation of carbon carbon bonds through asymmet-
ric catalytic processes is one of the great topics of modern
synthetic organic chemistry. In general, the catalysts consist
of a cation coordinating an optically active organic ligand;
the resulting complex must have at least one free binding
site to coordinate one reagent in order to promote the for-
mation of the reactive complex involved in the catalytic
cycle. The ideal ligand should offer some advantages: it
should be easily prepared, cheap, resistant under the reac-
tion conditions and very selective for a large number of pro-
cesses, hence very flexible. The ideal complex should be
stable enough to be kept at the air on the bench.
Among the chiral ligands, pyridine-2,6-bis(oxazoline)
Scheme 1. Synthesis of TIPS-box (4) and its CuACTHNURGTNEUNG(OTf)2 complex 5.
(pybox)[1] and isopropylidene-2,2-bis(oxazoline) (box)[2]
proved to have many of the above-mentioned qualities, and
gave catalysts for a variety of enantioselective reactions.
One of the best pybox is (4’S,5’S)-2,6-bis[4’-TIPSoxymethyl-
5’-phenyl-1’,3’-oxazolin-2’-yl]pyridine (TIPS=triisopropylsil-
yl) because it is cheap, easily synthesized within two steps,
and very flexible in use.[3]
wal;[4] 3 was converted with good yields into (4’S,5’S)-2,2-bis-
[4’-(triisopropylsilyl)-oxymethy-5’-phenyl-1’,3’-oxazolin-2’-
yl]propane (4) (TIPS-box). concentrated equimolar
CH2Cl2 solution of 4 and Cu(OTf)2 was diluted with hexane
AHCTUNGTRENNUNG
A
AHCTUNGTRENNUNG
and, after standing at 08C, a blue-greenish solid precipitated
that recrystallized from cyclohexane to give stable crystals
of copper complex 5, suitable for X-ray structure determina-
tion (Figure 1).
The coordination around Cu has a square-pyramidal ge-
ometry. The box nitrogen and two oxygen atoms, one be-
Starting from dimethylmalononitrile (1) and aminol (2),
several authors have synthesized (4’S,5’S)-2,2-bis[4’-hydroxy-
methyl-5’-phenyl-1’,3’-oxazolin-2’-yl]propane (3) using Cd-
(OAc)2,[4] or ZnCl2.[5–7] The hydroxy groups were functional-
ized with a variety of substituents,[4,8,9] providing ligands for
many catalysts, some very efficient, some less. Among these
derivatives, the TBDMS-box was mentioned but not de-
scribed,[10] and to the best of our knowledge the TIPS ana-
logue has not been prepared thus far (Scheme 1). Com-
pound 3 was synthesized following the method of Aggar-
À
longing to a water molecule [Cu(1) O(1w) 1.981 ꢀ] and
À
one to a triflate anion [Cu(1) O(5) 2.045 ꢀ], are the equa-
torial ligands; this molecular fragment, in spite of the bulky
equatorial triflate, is about planar: the dihedral angles
O(1w)-Cu(1)-N(2)-C(5) and O(5)-Cu(1)-N(1)-C(3) diverge
in the same sense from planarity by +3.6 and +17.48, re-
spectively, and the plane O(1w)-Cu(1)-O(5) is slightly bent
downwards from the plane of the box. The second triflate
[a] Prof. G. Desimoni, Prof. G. Faita, Dr. M. Toscanini
Dipartimento di Chimica Organica, Universitꢁ di Pavia
Viale Taramelli 10, 27100 Pavia (Italy)
Fax : (+39)382-987323
À
anion is less bound in the apical position [Cu(1) O(8)
2.386 ꢀ].
E-mail: desimoni@unipv.it
The X-ray structures of eighteen [box/CuII] complexes
have been reported, eleven of them containing tert-butyl
box (A) or phenyl box (B).[2] Six complexes are known of
(S)-A and (S)-B with SbF6, Cl and Br as anions, but the li-
gands involved in the square-planar structures are H2O, Cl
[b] Dr. M. Boiocchi
Centro Grandi Strumenti, Universitꢁ di Pavia
Via Bassi 21, 27100 Pavia (Italy)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900908.
9674
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9674 – 9677

COMMUNICATION
tert-butyl substituents; the dihedral angles O-Cu-N-C are
+35.6 and À35.98, respectively, compared with +3.6 and
+17.48 for 5. One triflate is located in the axial position
with a weak interaction (Cu–OTf 2.624 ꢀ), the second tri-
flate is a non-coordinating anion (Cu–OTf 3.667 ꢀ).
Besides to the distorted or less distorted square planar
structures, a key factor of complex 5 is the presence of a
pentacoordinated copper(II) species with a geometry that,
as described above, has been found for some [box/CuII]
complexes.[11–13,17,18] This geometry has increasingly been rec-
ognized to play an important role in asymmetric CuII cataly-
sis[19] in alternative or together with tetracoordinated spe-
cies.
As mentioned previously, the angle a illustrated in
Figure 2 (representing the degree of distortion from planari-
ty of the structure) may be important for the square-planar
structure, and in the induction of stereoselectivity on a bi-
dentate substrate that may form only one possible coordina-
tion to copper by substitution of the equatorial ligands. On
the contrary, pentacoordination may induce more than one
possible coordination of the same substrate. In addition to a
coordination involving two equatorial positions, which gives
a result similar to a square planar complex, the reagent che-
lation may involve one equatorial (water or triflate) and the
axial position of 5, and this may induce opposite stereoselec-
tivity.[20]
With this scope, complex 5 was tested as catalyst for the
enantioselective Diels–Alder reaction (DA) between acrylo-
yl-1,3-oxazolidin-2-one (6), a bidentate substrate, and cyclo-
pentadiene (7a), a cycloaddition used as a benchmark to
test the efficiency of a chiral complex in the cluster of
copper-catalyzed DA reaction.[21]
Figure 1. An ORTEP view of the crystal structure of 5 (ellipsoid are
drawn at the 10% probability level) labelled with main crystallographic
atom names. CCDC 723861 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_-
request/cif.
and Br, respectively [(S)-A1–3 and (S)-B1–3]. The four
chiral complexes with halogens (A2,3 and B2,3) are strongly
distorted by about the same dihedral angle [X-Cu-N-C (a)
between 39 and 458].[11] The aquo complexes (A1 and B1) of
the six complexes in which the catalysts are employed in
enantioselective organic syntheses, are considerably differ-
ent in their structure. The square-planar structure of {Cu-
A
E
(SbF6)}
N
Experiments were run with or without 3 ꢀ molecular
sieves (MS) added to 5 or with the catalyst prepared in situ
torted with the water molecules located far away from the
tert-butyl groups (the dihedral angles O-Cu-N-C are +30.0
from 4 and CuACTHGUNETRNNU(G OTf)2. The results are very similar (Table 1,
and À36.08); contrarily {Cu
[(S,S)-Ph-box]
N
N
entries 1–3). The efficiency of 5 was excellent because 1%
ACHTUNGTRENNUNG
water molecules is reversed and the dihedral angles O-Cu-
N-C are À11.3 and +7.28, respectively (Figure 2). Similari-
ties can therefore be found between the crystal structure of
5 and B1.
Table 1. Diels–Alder reaction between 6 and 7a in CH2Cl2 at À508C cat-
alyzed by 1 mol% catalyst.
n
Catalyst
t
[8a]/
8a ee [%]
(configuration) (configuration)
9a ee [%]
[h][a] [9a]
1
2
3
4
5
6
7
8
5
12
12
12
24
18
24
93:7
90:10
92:8
92:8
90:10
90:10
93:7
91:9
74 (S)
76 (S)
76 (S)
73 (S)
81 (S)
80 (S)
82 (S)
81 (S)
60 (R)
50 (R)
55 (R)
52 (R)
53 (R)
54 (R)
57 (R)
55 (R)
5/MS
4 + Cu
Figure 2. Schematic representation of the six X-ray crystal structures of
the complexes between CuII, box (S,S)-tBu-box (A) and (S,S)-Ph-box
(B).[11–16]
ACHTUNGERTN(NUNG OTf)2
4 + CuACTHNGURTNE(NUG ClO4)2/MS
5/MS + LiOTf[b]
5/MS + LiOTf[c]
4+Cu
ACHUTNGTREN(NUG ClO4)2ACHTUGNTRENNNUG
(H2O)6 24
5 + 5H2O[d]
24
For copper(II) triflate, only the X-ray crystal structure of
the complex with tBu-box [(S)-A] has been reported.[11,12,17]
The distorted square pyramidal geometry has two water
molecules in the equatorial positions, located away from the
[a] All reactions give quantitative yields. [b] 2 mol%
LiOTf. [c] 5 equiv LiOTf. [d] 2 mol% 5 and 1 equiv LiOTf. [e] 2 mol% 5
and 10 equiv water.
5 and 1 equiv
Chem. Eur. J. 2009, 15, 9674 – 9677
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9675

G. Desimoni et al.
mol catalyst is enough to complete the DA reaction at
À508C. The addition of MS does not influence the result
and the stereoselectivity is satisfactory (about 75% ee).
A further test of 5 as a catalyst was performed on the 1,3-
dipolar cycloaddition (1,3-DC) between 6 and diphenylni-
trone 7b, which represents a reaction seldom catalyzed by
[box/CuII] complexes.[22] The 1,3-DC was performed at
À208C in CH2Cl2 with addition of MS required to avoid the
decomposition of nitrone; the results are reported in Table 2
(entries 1,2). The catalyst is strongly exo-selective and
(3’R,4’R)-8b is obtained in a good 85% ee.[23]
determination (Figure 3). This is the first crystal structure of
a box complex containing a bound 3-alkenoyl-2-oxazolidi-
none as a reaction intermediate proposed in enantioselective
catalyses involving 6 and its homologues. The structure
shows the behavior of 6 as a bidentate reagent with its car-
bonyl groups occupying the equatorial positions, and con-
firms the cisoid conformation of the a,b-unsaturated carbon-
yl fragment, a conformation forceful assumed in the litera-
ture to rationalize the stereochemical outcome. The reacting
complex 10 keeps a nearly untwisted square pyramidal
structure [the dihedral angles O(7)-Cu(1)-N(1)-C(3) and
O(5)-Cu(1)-N(2)-C(5) are +12.8 and +7.88, respectively],
with the water molecule bound in the apical position
[Cu(1)–O(1w) 2.283 ꢀ].
Table 2. 1,3-Dipolar cycloaddition between 6 and 7b in CH2Cl2 at À208C
catalyzed by 1 mol% catalyst.[a]
n
Catalyst[b]
[8b]/[9b] 8b [%]
9b [%]
(configuration) (configuration)
1
2
3
4
5
5 and MS
98:2
98:2
84 (3’R,4’R)
86 (3’R,4’R)
86 (3’R,4’R)
>90 (3’S,4’R)
>90 (3’S,4’R)
>90 (3’S,4’R)
>90 (3’S,4’R)
>90 (3’S,4’R)
4 + Cu
4 + Cu
ACHTUNGTRENNUNG(OTf)2 + MS
A
5/MS + LiOTf[c]
(ClO4)2 + MS[e] 98:2
98.5:1.5 84 (3’R,4’R)
98.5:1.5 84 (3’R,4’R)
5/MS + LiOTf[d]
[a] Reactions in entries 1–4 give quantitative yields (reaction time, 8–
12 h). [b] In the absence of MS nitrone decomposes. [c] 2 mol% 5 and
1 equiv LiOTf. [d] 5 equiv LiOTf. [e] Quantitative yield, reaction time
24 h.
Figure 3. An ORTEP view of the crystal structure of the reactive inter-
mediate 10 formed from 5 and 6 involved in the DA and 1,3-DC with cy-
clopentadiene and diphenylnitrone (ellipsoid are drawn at the 10% prob-
ability level) labelled with crystallographic atom names. CCDC 723862
contains the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallograph-
ic Data Centre via www.ccdc.cam.ac.uk/data_request/cif
To test the importance of the triflate anion in the reacting
intermediates, the above reactions have been performed by
using a catalyst obtained from 4, [CuACTHNURGTENN(UG ClO4)2·ACHTUGNTRNE(NUGN H2O)6 and MS
(Tables 1, 2, entries 4 and 3, respectively), the reaction time
is somewhat longer, but the results are very similar to those
obtained with 5.
The efficiency of the [box/CuII] catalyst in terms of enan-
tioselectivity was proposed by Evans to be due to the pla-
narity of the ligation of the reagent in the reacting inter-
mediate and this was supported by solving the X-ray struc-
tures of the CuII complexes between benzylidene dimethyl-
malonate (BDM) and (A) or (B) box.[24] The dicarbonyl
fragment is coordinated at the position formerly occupied
by water in complexes A1 and B1; the configuration of [A/
The above crystal structure leaves no doubt about the site
of substitution of 6 into 5 during the reaction from the cata-
lyst 5 to 10, but the stereoselectivity induced into the DA
and the 1,3-DC is only the result of the diequatorial coordi-
nation of 6 or pentacoordination may be a cofactor?
Some intriguing results have been reported, in which lithi-
um triflate seems to stabilize a pentacoordinated CuII com-
plex involved in the enantioselective asymmetric Michael
addition of indole to benzylidene malonates.[19] Therefore,
both DA and 1,3-DC have been tested under the above cat-
alytic conditions, with the addition of 1 or 5 equivalents of
LiOTf. The enantiomeric excess of the DA product 8a is
slightly increased (Table 1, entries 5,6), that of the 1,3-DC
adduct 8b remains constant (Table 2, entries 4,5).
Cu
ACHTUNGTRENNUNG
Cu
ACHTUNGTRENNUNG
lectivity of the Michael reaction between enolsilane and
BDM catalyzed by A1 is excellent (99% ee), while the
result with B1 is poor (52% ee). Even if the moderate ste-
reoinduction obtained by using 5 could be rationalized in an
analogous manner, the possibility of competing coordination
modes in a pyramidal-square structure involving equatorial
and axial positions of 5, may also influence the reaction
enantioselectivity.
If the effect of LiOTf as additive stabilizing a square-pyr-
amidal CuII complex is to annihilate the dissociation of one
of its five ligands, the triflate should be a ligand participat-
ing to the pentacoordination. But 10, contrary to 5, has a
water molecule as the axial ligand. Therefore two further
Fortuitously, from 5 and 6, crystals of complex 10 were
isolated, which were suitable for an X-ray crystal structure
9676
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9674 – 9677

A New CopperII/box Catalyst
COMMUNICATION
experiments were performed, one with the catalyst formed
by 4 and CuACHTUNGTRENNUNG(ClO4)2·ACHTUNGTRENNU(G H2O)6, without MS, one with 5 equiva-
Acknowledgements
This work was supported by the Ministero dell’Universitꢁ e della Ricerca
(MIUR) through a PRIN fund, and by the University of Pavia.
lents water added to 5 (Table 1, entries 7 and 8). In general,
water does not have a beneficial effect on organometallic
catalysts since its Lewis-basic site tends to preclude binding
to the organic reactants, modifying the supramolecular ar-
chitecture of the complex involved in the catalysis.[25] In the
above experiments the enantioselectivity remains 81–82%
ee and, in our opinion, this supports 10 as the reacting inter-
mediate.
The catalytic cycle of the enantioselective DA and 1,3-DC
reactions between 6 and 7a or 7b can be therefore illustrat-
ed in Figure 4 in which not only the catalyst 5, but also the
reaction intermediate 10 have unambiguous structures. This
allows to understand why 10 is suitable to favor the Re-face
approach to coordinated 6 to give (S)-8a and (3’R,4’R)-8b.
Keywords: asymmetric catalysis · bis(oxazoline) · copper ·
reactive intermediate · structure elucidation
[1] G. Desimoni, G. Faita, P. Quadrelli, Chem. Rev. 2003, 103, 3119–
3154.
[2] G. Desimoni, G. Faita, K. A. Jørgensen, Chem. Rev. 2006, 106,
3561–3651.
[3] G. Desimoni, G. Faita, M. Guala, A. Laurenti, M. Mella, Chem. Eur.
J. 2005, 11, 3816–3824.
[4] V. K. Aggarwal, L. Bell, M. P. Coogan, P. J. Jubault, J. Chem. Soc.
Perkin Trans. 1 1998, 2037–2042.
[5] M. Schinnerl, M. Seitz, A. Kaiser, O. Reiser, Org. Lett. 2001, 3,
4259–4262.
[6] G. Klein, S. Pandiaraju, O. Reiser, Tetrahedron Lett. 2002, 43, 7503–
7506.
[7] M. Schinnerl, C. Bçhm, M. Seitz, O. Reiser, Tetrahedron: Asymme-
try 2003, 14, 765–771.
[8] M. Malkoch, K. Hallman, S. Lutsenko, A. Hult, E. Malmstrçm, C.
Molberg, J. Org. Chem. 2002, 67, 8197–8202.
[9] K. Hallman, A. Frçlander, T. Wondimagegn, M. Svenson, C. Mol-
berg, Proc. Natl. Acad. Sci. USA 2004, 101, 5400–5404.
[10] G. Koch, A. Pfaltz, unpublished results mentioned in ref. [41] of A.
Pfaltz, Acc. Chem. Res. 1993, 26, 339–345.
[11] J. Thorhauge, M. Roberson, R. G. Hazell, K. A. Jørgensen, Chem.
Eur. J. 2002, 8, 1888–1898.
[12] D. A. Evans, S. J. Miller, T. Lectka, P. von Matt, J. Am. Chem. Soc.
1999, 121, 7559–7573.
[13] D. A. Evans, T. Rovis, J. S. Johnson, Pure Appl. Chem. 1999, 71,
1407–1415.
[14] D. A. Evans, J. S. Johnson, C. S. Burgey, K. R. Campos, Tetrahedron
Lett. 1999, 40, 2879–2882.
[15] D. A. Evans, J. S. Johnson, E. J. Olhava, J. Am. Chem. Soc. 2000,
122, 1635–1649.
[16] D. A. Evans, K. A. Scheidt, J. N. Johnston, M. C. Willis, J. Am.
Chem. Soc. 2001, 123, 4480–4491.
Figure 4. Catalytic cycle of the DA and 1,3-DC reactions.
[17] D. A. Evans, E. J. Olhava, J. S. Johnson, J. M. Janey, Angew. Chem.
1998, 110, 3553–3557; Angew. Chem. Int. Ed. 1998, 37, 3372–3375.
[18] R. Rasappan, M. Hager, A. Gissibl, O. Reiser, Org. Lett. 2006, 8,
6099–6102.
[19] A. Schꢃtz, R. Rasappan, M. Hager, A. Gissibl, O. Reiser, Chem.
Eur. J. 2008, 14, 7259–7265.
In this research a new [box/CuII] complex 5 was easily
prepared; this complex is soluble in aprotic organic solvents
and stable on the bench. The enantioselectivity induced in
DA and 1,3-DC reaction is in the range 73–86% ee. The
complex has an undistorted square pyramidal structure and
the equatorial ligands are substituted by the reagent 6 to
give the reacting complex 10. Further research will clarify if
the pentacoordination can induce the optimal face selectivi-
ty that the untwisted square structure of a [box/CuII] com-
plex seems unable to induce.[2]
[20] R. Rasappan, D. Laventine, O. Reiser, Coord. Chem. Rev. 2008, 252,
702–714.
[21] S. Reymond, J. Cossy, Chem. Rev. 2008, 108, 5359–5406.
[22] a) L. M. Stanley, M. P. Sibi, Chem. Rev. 2008, 108, 2887–2902;
b) K. V. Gothelf, K. A. Jørgensen, Chem. Rev. 1998, 98, 863–910.
[23] G. Desimoni, G. Faita, M. Mella, M. Boiocchi, Eur. J. Org. Chem.
2005, 1020–1027.
[24] D. A. Evans, T. Rovis, M. C. Kozlowski, C. W. Downey, J. S. Tedrow,
J. Am. Chem. Soc. 2000, 122, 9134–9142.
[25] U. M. Lindstrom, Chem. Rev. 2006, 106, 2751–2772.
Received: April 7, 2009
Published online: August 15, 2009
Chem. Eur. J. 2009, 15, 9674 – 9677
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9677