Cationic complexes of palladium(II) and platinum(II) as Lewis acid catalysts for the Diels-Alder reaction

Article abstract of DOI:10.1021/om0003943

The synthesis of a series of Pd(II) and Pt(II) cationic complexes of the types [(P-P)M(solv)](Y)₂, [(P-P)M(η²-Y)](Y), and [(P-P)M(μ-Cl)]₂(Y)₂ (where P-P = various diphosphines including chiral diphosphines (M = Pd, Pt; Y = TfO^-, ClO₄^-, BF₄^-) is reported. All complexes are excellent catalysts for the Diels-Alder reaction between cyclopentadiene or cyclohexadiene and a number of simple dienophiles such as acrolein, methacrolein, and methylvinyl ketone. Some of the complexes modified with chiral diphosphines can act as enantioselective catalysts in the Diels-Alder reaction under mild condition with moderate ee's. A critical point in this respect is the presence of only one vacant coordination site in the catalyst.

Full text of DOI:10.1021/om0003943

5160
Organometallics 2000, 19, 5160-5167
Ca tion ic Com p lexes of P a lla d iu m (II) a n d P la tin u m (II) a s
Lew is Acid Ca ta lysts for th e Diels-Ald er Rea ction
Kristian Pignat, J acopo Vallotto, Francesco Pinna, and Giorgio Strukul*
Department of Chemistry, University of Venice, 30123 Venezia, Italy
Received May 9, 2000
The synthesis of a series of Pd(II) and Pt(II) cationic complexes of the types [(P-P)M-
(solv)](Y)2, [(P-P)M(η2-Y)](Y), and [(P-P)M(µ-Cl)]2(Y)2 (where P-P ) various diphosphines
including chiral diphosphines (M ) Pd, Pt; Y ) TfO-, ClO4-, BF4-) is reported. All complexes
are excellent catalysts for the Diels-Alder reaction between cyclopentadiene or cyclohexa-
diene and a number of simple dienophiles such as acrolein, methacrolein, and methylvinyl
ketone. Some of the complexes modified with chiral diphosphines can act as enantioselective
catalysts in the Diels-Alder reaction under mild condition with moderate ee’s. A critical
point in this respect is the presence of only one vacant coordination site in the catalyst.
In tr od u ction
observed, especially in the case of R,â-unsaturated
substrates.7
Square-planar complexes of Pd(II) and, to a minor
extent, Pt(II) have long been recognized to play a central
role in homogeneous catalysis in processes such as the
classical Wacker process1 or, more recently, the copo-
lymerization of carbon monoxide and olefins.2 In the
past two decades we have extensively studied the
properties of cationic Pd(II) and Pt(II) complexes in a
variety of oxidation reactions using hydrogen peroxide
as oxidant.3-5 A typical feature displayed by the metal
in the latter reactions is the ability to increase the
reactivity of the substrate by coordination, thereby
making it more prone to attack by the peroxidic oxidant.
An essential requirement to promote this kind of
reactivity is the presence of a positive charge on the
metal, which, therefore, possesses a significant Lewis
acidity.
The Diels-Alder reaction is a typical Lewis acid
catalyzed process, which consists of a 4+2 cycloaddition
between a diene and a dienophile. Being concerted and
stereoselective, it is widely used in synthetic organic
chemistry.8 In the past two decades several attempts
have been made to accomplish the reaction enantiose-
lectively, initially, by complexing ordinary Lewis acids
(AlX3, BX3, SnX4, etc.) with suitable chiral ligands9 and,
subsequently, by using transition metal complexes,
among which those of Co,9c Fe,9c,10 Mo,11 Ni,12 Cu,13 Ru,14
V,15 W,11 Yb,16 and Ti9c,17 have proved to be particularly
interesting, often showing ee’s in the >95% range.
The behavior of noble metal complexes in the Diels-
Alder reaction has not been satisfactorily exploited,
(7) Nieddu, E.; Cataldo, M.; Pinna, F. Strukul, G. Tetrahedron Lett.
1999, 40, 6987.
(8) As general references see for example: (a) Ciganek, E. Org.
React. 1984, 32, 1. (b) Craig, D. Chem. Soc. Rev. 1987, 16, 187. (c)
Carruthers, W. Cycloaddition Reactions in Organic Synthesis; Perga-
mon: Oxford, 1990.
(9) (a) Maruoka, K.; Sakurai, M.; Fujiwara, J .; Yamamoto, H. Chem.
Lett. 1986, 4895. (b) Rebiere, F.; Riant, O.; Kagan, H. B. Tetrahedron:
Asymm. 1990, 1, 199. (c) Kagan H. B.; Riant, O. Chem. Rev. 1992, 92,
1007.
The Lewis acidity of these complexes has been re-
cently exploited in the acetalyzation of a variety of
aldehydes and ketones,6 where some unusual selectivi-
ties with respect to ordinary Brønsted acids were
* Corresponding author at: Department of Chemistry, University
of Venice, Dorsoduro 2137, 30123 Venezia, Italy. Tel: +39 041 257
8551. Fax: +39 041 257 8517. E-mail: strukul@unive.it.
(1) As general references see for example: (a) Parshall, G. W.; Ittel,
S. D. Homogeneous Catalysis, 2nd ed.; Wiley-Interscience: New York,
1992; Chapter 6, p 137. (b) Sheldon, R. A.; Kochi, J . K. Metal Catalyzed
Oxidations of Organic Compounds; Academic Press: New York, 1981.
(2) See for example: (a) Drent, E.; van Broekhoven, J . A. M.;
Budzelaar, P. H. M. In Applied Homogeneous Catalysis with Organo-
metallic Compounds; Cornils, B., Herrmann, W. A., Eds.; VCH:
Weinheim, 1997; Vol. 1, Chapter 2.3.4, p 333. (b) Drent, E.; Budzelaar,
P. H. M. Chem. Rev. 1996, 96, 663. (c) Sen, A. Acc. Chem. Res. 1993,
26, 303. (d) Drent, E. Eur. Patent Appl. 251373, 1987, and references
therein.
(3) For reviews on this subject see: (a) Strukul, G. In Catalytic
Oxidations with Hydrogen Peroxide as Oxidant; Strukul, G., Ed.;
Kluwer: Dordrecht, 1992; Chapter 6, p 177. (b) Strukul, G. In Advances
in Catalyst Design, Vol. 2; Graziani, M., Rao, C. N. R., Eds.; World
Scientific: Singapore, 1993; p 53. (c) Strukul, G. Angew. Chem., Int.
Ed. 1998, 37, 1198.
(10) (a) Corey, E. J .; Imai, N.; Zhang, H. J . Am. Chem. Soc. 1991,
113, 728. (b) Ku¨ndig, E. P.; Bourding, B.; Bernardinelli, G. Angew.
Chem., Int. Ed. 1994, 33, 1856.
(11) (a) Bonnesen, P. V.; Puckett, C. L.; Honeychuck, R. V.; Hersh,
W. H. J . Am. Chem. Soc. 1989, 111, 6070. (b) Kuo, C. Y.; Fuh, Y. S.;
Chaen, M. C.; Yu, S. J . Tetrahedron Lett. 1999, 40, 6451.
(12) (a) Brunner, H.; Muschiol, M.; Prester, F. Angew. Chem., Int.
Ed. 1990, 29, 652. (b) Kanemasa, S.; Oderaotoshi, Y.; Sakaguchi, S.;
Yamamoto, H.; Tanaka, J .; Wada, E.; Curran, D. P. J . Am. Chem. Soc.
1998, 120, 3074. (c) Otto, S.; Engberts, J . B. F. N. J . Am. Chem. Soc.
1999, 121, 6798.
(13) (a) Evans, D. A.; Miller, S. J .; Letcka, T. J . Am. Chem. Soc.
1993, 115, 6469. (b) Evans, D. A.; Murry, J . A.; Matt, P.; Norcross, R.
D.; Miller S. J . Angew. Chem., Int. Ed. Engl. 1995, 34, 798. (c)
J ohannsen, M.; J ørgensen, K. A. J . Org. Chem. 1995, 60, 5757. (d)
J ørgensen, K. A.; J ohannsen, M.; Yao, S.; Audrain, H.; Thorhauge, J .
Acc. Chem. Res. 1999, 32, 605, and references therein. (e) Evans, D.
E.; Barnes, D. M.; J ohnson, J . S.; Lectka, T.; von Matt, P.; Miller S. J .;
Murry, J . A.; Norcross, R. D.; Shaughnessy, E. A.; Campos, K. R. J .
Am. Chem. Soc. 1999, 121, 7582. (f) Evans, D. A.; J ohnson, J . S.;
Olhava, E. J . J . Am. Chem. Soc. 2000, 122, 1635.
(4) Gavagnin, R.; Cataldo, M.; Pinna, F.; Strukul, G. Organometallics
1998, 17, 661.
(5) Paneghetti, C.; Gavagnin, R.; Pinna, F.; Strukul, G. Organome-
tallics 1999, 18, 5057.
(6) Cataldo M.; Nieddu, E.; Gavagnin, R.; Pinna, F.; Strukul, G. J .
Mol. Catal. A: Chem. 1999, 142, 305.
(14) Odenkirk, W.; Rheingold A. L.; Bosnich, B. J . Am. Chem. Soc.
1992, 114, 6392.
(15) (a) Togni, A. Organometallics 1990, 9, 3106. (b) Togni, A.;
Pastor, S. T. Chirality 1991, 3, 331.
10.1021/om0003943 CCC: $19.00 © 2000 American Chemical Society
Publication on Web 11/02/2000

Cationic Complexes of Pd(II) and Pt(II)
Ta ble 1. Sp ectr oscop ic F ea tu r es of th e Com p lexes Sh ow n in Ch a r t 1a
IR (cm-1 31P{1H} NMR (ppm, J in Hz)
33.0 (s); 37.2 (s)b
Organometallics, Vol. 19, No. 24, 2000 5161
complex
)
ΛM (Ω-1 mol-1 cm2)
1a x
2294, 2322
1096
259c
1a y
2a
2b
1097
37.6 (s)b
265c
124
123
100
1286, 1258, 1224, 1153, 1100, 1034
1282, 1256, 1227, 1167, 1098, 1032
1290, 1261, 1227, 1163, 1099, 1032
36.87 (s); 33.86 (s)
34.47 (s); 32.2 (s)
2
2c
24.27 (d), J P-P 16, 48.0 (d) 2J P-P 46;
2
2
28.9 (d), J P-P 16, 48.8 (d), J P-P 16
2d
2e
3a
3b
4a x
4bx
4by
4bz
5a z
5bx
5by
5bz
1289, 1261, 1225, 1167, 1100, 1027
1288, 1250, 1227, 1167, 1101, 1030
1291, 1259, 1225, 1153, 1104, 1030
1284, 1254, 1227, 1167, 1101, 1030
298, 278
303, 279
303, 280
303, 280
295, 282
303, 280
303, 280
303, 280
21.86 (s); 24.85 (s)
140
118
151
141
143d
2
2
42.03 (d), J P-P 21, 42.96 (d), J P-P 21
1
13.58 (s), J P-Pt 3656
1
2.31 (s), J P-Pt 4039
36.93 (s)
34.55 (s); 29.82 (s)b
34.32 (s); 29.76 (s)b
177
203
272
216
227
218
34.29 (s); 32.8 (d), J P-P 7.9, 31.0 (d), J P-P 7.9; 29.76 (s)b
2
2
1
13.49 (s), J P-Pt 3668
1
8.78 (s), J P-Pt 3816
1
8.53 (s), J P-Pt 3820
1
8.60 (s), J P-Pt 3819
a
Experimental conditions: IR in KBr pellet, NMR in CD2Cl2 at room temperature unless otherwise posted, conductivity 10-3M solutions
in acetone. bAt -70 °C. c 10-3 M solutions in MeCN. 10-3 M solutions in MeOH.
d
Ch a r t 1
Sch em e 1
behind the choice of these complexes is to have at least
one positive charge located on the potential catalytically
active fragment, to have a sufficient Lewis acid char-
acter. As will be clear in the foregoing, the different
types of complexes shown in Chart 1 reflect an evolution
of the catalyst that was found necessary depending on
the results observed in the catalytic reactions. Although
the synthetic procedures outlined in Scheme 1 are quite
well known, the formulation of the complexes as in
Chart 1 is in many cases merely conventional, as is
shown by their characterization data based on IR and
31P{1H} NMR spetroscopies and on molar conductivity
measurements (Table 1).
Complex 1a x shows the typical asymmetric and
symmetric IR bands of coordinated acetonitrile along
with a strong, broad band centered at 1096 cm-1 typical
of ClO4-. The homologous 1a y does not give evidence
for the coordination of THF; however an IR spectrum
in DCE where a stoichiometric amount of MeCN was
added showed again the presence of bands at 2291 and
2325 cm-1, indicating the presence in the original
complex of labile solvento ligands. Conductivity data in
MeOH for both 1a x and 1a y are consistent with the
presence of 2:1 electrolytes.20 The NMR spectra of these
complexes in CD2Cl2 are more complex than expected.
In the case of 1a x, two singlets are present at room
temperature (Figure 1 middle), which can be associated
with species having magnetically equivalent phosphorus
nuclei. Addition of an excess of MeCN increased sig-
nificantly the intensity of the signal at 32.8 ppm (Figure
probably because being electron-rich their choice is less
obvious, as they are generally considered more as bases
rather than acids. Some interesting examples in which
Pd(II) and Pt(II) species have been used to catalyze the
asymmetric aldol reaction or asymmetric 1,3-dipolar
cycloadditions have recently appeared in the litera-
ture.18,19 In this paper we wish to report our attempts
to apply the Lewis acidity of a wide class of homologous,
cationic complexes of Pd(II) and Pt(II) in the Diels-
Alder reaction, with the ultimate aim of performing this
transformation enantioselectively.
Resu lts a n d Discu ssion
Th e Ca ta lysts. A summary of the complexes used
in the present study is shown in Chart 1, and their
synthetic route is reported in Scheme 1. The basic idea
(16) (a) Kobayashi, S.; Hachiya, I.; Ishitani, H.; Araki, M. Tetrahe-
dron Lett. 1993, 34, 4535. (b) Kobayashi, S.; Ishitani, H. J . Am. Chem.
Soc. 1994, 116, 4083.
(17) (a) Mikami, K.; Terada, M.; Motoyama, Y.; Nakai, Y. Tetrahe-
dron: Asymm. 1991, 2, 643. (b) Corey, E. J .; Matsumara, Y. Tetrahe-
dron Lett. 1991, 32, 6289. (c) Narasaka, K.; Iwasawa, N.; Inoue, M.;
Yamada, T.; Nakashima, M.; Sugimori, J . J . Am. Chem. Soc. 1989,
111, 5340. (d) Hollis, T. K.; Robinson, N. P.; Bosnich, B. J . Am. Chem.
Soc. 1992, 114, 5464.
(18) Fujimura, O. J . Am. Chem. Soc. 1998, 120, 10032.
(19) Hori, K.; Kodama, H.; Ohta, T.; Furukawa, I. J . Org. Chem.
1999, 64, 5017.
(20) Geary, W. J . Coord. Chem. Rev. 1971, 7, 81.

5162 Organometallics, Vol. 19, No. 24, 2000
Pignat et al.
F igu r e 1. 31P{1H} NMR spectrum of complex 1a x in CD2Cl2: middle, the spectrum at room temperature; top, the same
+ excess MeCN added; bottom, at -70 °C.
1, top), indicating that this was most likely the expected
bis-solvento complex. We suggest that the species at
37.2 ppm is probably a η2 perchlorate derivative in
equilibrium with the expected complex (eq 1), probably
all complexes in MeOH fall in the range of 1/1 electro-
lytes. These observations are a strong argument in favor
of a formulation like the one reported in Chart 1.
However, the 31P{1H} NMR spectrum of 2a at room
temperature shows two moderately broad singlets, one
at 36.87 ppm and a much weaker (∼10% in intensity)
one at higher fields (33.86 ppm). At -60 °C the former
signal becomes the only one present in the spectrum,
indicating that this is the most stable species at low
temperature. Consistently with the IR and the molar
conductivity data, we suggest that the major species
observed in the NMR spectrum is a cationic complex
containing one OTf moiety coordinated η2 to the metal
and one TfO- moiety as external anion (eq 2), as was
via the Pd(NCMe)(OClO3) mixed ligand complex, which
would account for the broad minor signals present in
the spectrum. The formation of η2 perchlorate complexes
has already been suggested in the literature during the
synthesis of bis-solvento complexes of the type here
reported.21 The η2 perchlorate derivative appears to be
more favored at -70 °C as shown by the corresponding
NMR spectrum (Figure 1, bottom). The 31P{1H} NMR
spectrum of complex 1a y is quite similar. At room
temperature singlets at 33.9 and 37.0 ppm are observed
together with new singlets at 31.3 and 36.4 ppm. At -70
°C the singlet at about 37 ppm is the only detectable
signal, indicating that the η2 perchlorate derivative most
likely becomes the dominant species, in agreement with
the lower coordination properties of THF with respect
to MeCN.
already suggested in the literature for other triflate
complexes,23 while the minor component is most likely
the bis-triflate neutral complex. The situation is similar
for complexes 2b and 2d , where a formulation like the
one reported in Chart 1 applies to the major component
(∼90%) of the mixture observed by NMR. This is
associated with the signals at 34.37 ppm (2b) and 21.86
ppm (2d ). In the case of 2e only two doublets are
observed with evidence for P-P coupling, in agreement
with the magnetic inequivalence of the P donors in
norphos. No evidence for other signals to be associated
with the neutral species is seen in the 31P{1H} NMR
spectrum. The case of 2c is apparently more complex.
Two sets of signals of similar intensity are evident in
the 31P{1H} NMR spectrum, each corresponding to a
species having two inequivalent phosphorus nuclei
mutually coupled. As has been previously observed in
the case of some Pt complexes5 and in the case of the
The apparent discrepancy between the NMR data and
the molar conductivity data can be explained by simply
considering the different solvents (CD2Cl2 and MeOH,
respectively).
Complexes 2 and 3 show in all cases a series of bands
in the 1000-1300 cm-1 region (Table 1) that can be
attributed to the different stretching modes of the
-
coordinated CF3SO3 anion, in agreement with that
reported in the literature22 for Pt(II) complexes, together
with other minor bands due to the stretching of unco-
ordinated triflate. In agreement with this assignment
is the observation that the molar conductivity data of
(21) Davis, J . A.; Hartley, F. R.; Murray, S. G. J . Chem. Soc., Dalton
Trans. 1980, 2246.
(22) Beck, W.; Olgemo¨ller B.; Olgemo¨ller L. Inorg. Synth. 1990, 28,
(23) Lawrance, G. A. Chem. Rev. 1986, 86, 17, and references
28.
therein.

Cationic Complexes of Pd(II) and Pt(II)
Organometallics, Vol. 19, No. 24, 2000 5163
F igu r e 2. 31P{1H} NMR spectra of different complexes in CD2Cl2 at -70 °C: top, complex 4bx; middle, complex 4by;
bottom, complex 4bz.
corresponding Pd dichlorides,5 these signals are at-
tributed to an equilibrium of the type shown in eq 3,
ppm for complex 4bx is the same observed in complex
2b and is attributed to [(binap)Pd(η2-OTf)]+, where the
Cl ligand has been replaced by TfO- and is present only
as the external counterion. By analogy, the signal at
34.32 ppm in the spectrum of the perchlorate derivative
(4by) is attributed to a monomeric species having the
same features (Figure 2), similarly to what was already
observed for 1a x and 1a y. The case of 4bz may be
different. Indeed, examples of η2 coordination in the case
where the N donor in the diphosphine is able to bind
the metal from the axial position giving rise to a square
pyramidal isomeric species. Other minor signals (<10%)
present in the NMR spectrum are attributed to neutral
species as in the previous cases, on the basis of the
extent of the molar conductivity of complex 2c.
The corresponding complexes of platinum (3a , 3b)
show simpler 31P{1H} NMR spectra, where only one
singlet is present at room temperature flanked by 195Pt
satellites. Molar conductivity data correspond to 1/1
electrolytes,20 supporting the formulation of complexes
as in Chart 1.
-
of BF4 have never been reported in the literature. A
formulation like in 4bx and 4by implies the presence
of two F atoms coordinated to the metal mutually cis,
and this would require in the 31P{1H} NMR spectrum
a coupling of each P atom with both the F atom in cis
and the F atom in trans. The spectrum in Figure 2 gives
no evidence for that, and the species at 34.32 ppm is
probably better described as a bis-solvento species.
Additionally, the third species present in 4bz is consis-
tent with a complex having both Cl and a solvent
molecule coordinated to Pd.
All complexes of type 4 and 5 show M-Cl stretching
frequencies in the range 270-310 cm-1 as well as molar
conductivities that fall in the range of 2/1 electrolytes.
The corresponding 31P{1H} NMR spectra at room tem-
perature are regular and correspond to the formulation
of Chart 1 in the cases of complex 4a x and complexes
of type 5. In all cases a sharp singlet (with P-Pt
coupling in the case of platinum species) is observed,
and lowering the aquisition temperature causes no
appreciable changes. Conversely, binap-containing pal-
ladium complexes 4bx, 4by, and 4bz all give a very
broad signal centered at 31-33 ppm at room tempera-
ture. At -70 °C the spectra shown in Figure 2 are
observed. As can be seen, they all show a singlet at 29.8
ppm and another stronger singlet in the 34.3-34.6
In summary, complexes 4b display significant differ-
ences from what was initially expected. The proposal
for the existence of η2-coordinated ClO4 ligands although
not common is not unprecedented,21,24 while the case of
-
the BF4 derivative is to some extent surprising, as it
seems to evidence that Cl- when coordinated to Pd is
so labile that it can be replaced by a weak donor such
as CH2Cl2. For the understanding of the catalytic results
that will be presented below, it has to be pointed out
that although at -70 °C the spectra of 4bx and 2b are
practically identical, at the temperature at which
catalytic runs have been carried out the composition of
the catalyst mixture is most likely different.
Ca ta lytic Exp er im en ts. Complexes 1a , 2a , and 3a
were tested to check their suitability as catalysts for
the Diels-Alder reaction. A selection of the results
-
range. In the case of the complex containing BF4 also
a third species appears, showing inequivalent phospho-
rus nuclei with evidence of P-P coupling. The singlet
at 29.8 that is present in all spectra can be most likely
attributed to the expected cationic bridging chloro
complex with different counterions. The species at ∼34.5
(24) (a) Favier, F.; Bargues, S.; Pascal, J . L.; Belin, C.; Tillard-
Charbonnel, M. J . Chem. Soc., Dalton Trans. 1994, 3119. (b) Honey-
chuck, R. W.; Hersh, W. H. Inorg. Chem. 1989, 28, 2869, and references
therein.

5164 Organometallics, Vol. 19, No. 24, 2000
Pignat et al.
Sch em e 2
Ta ble 2. Ca ta lytic Activity in th e Diels-Ald er
Rea ction of Ach ir a l Com p lexes 1-3
Ta ble 3. En a n tioselective Diels-Ald er
Tr a n sfor m a tion s Usin g Com p lexes 2 a n d 3 a s
Ca ta lystsa
a
Reaction conditions: catalyst 10 µmol, diene 1 mmol, dieno-
phile 1 mmol, DCE 3 mL, T 25 °C, N2 1 atm. a. Catalyst 1 µmol,
T 0 °C.
a
Reaction conditions: catalyst 10 µmol, diene 1 mmol, dieno-
phile 1 mmol, DCE 3 mL, N2 1 atm.
obtained is presented in Table 2. As can be seen, in the
reaction between cyclopentadiene and acrolein the sol-
vento complexes of type 1 show excellent activity (even
at low temperature) and a moderate selectivity toward
endo compounds. In the case of less reactive substrates
such as cyclohexadiene again a good catalytic activity
is observed even with dienophiles such as methacrolein
and methylvinyl ketone that are known to be moder-
ately reactive. Some interesting features are also ob-
served, namely, the generally higher selectivity toward
endo compounds and the preferential formation of the
exo isomer in the reaction with methacrolein with 1a y
as catalyst. In these experiments, no particular differ-
ences were observed between Pd and Pt complexes.
proceeds through the displacement of the labile ligands
(solvent molecules in the case of 1 and TfO- in the case
of 2 and 3), leading to intermediates in which the
dienophile binds the catalyst through the oxygen donor.
Under these circumstances the use of complex 1 and
that of complexes 2 and 3 as catalysts should lead to
the formation of closely related intermediates probably
still containing a very labile ligand (a solvent molecule
or η1-TfO-). The possibility that after the 4+2 cycload-
dition the presence of a vacant coordination site on the
metal may easily lead to the formation of an enolate
(Scheme 2, charges omitted for simplicity), washing out
possible ee’s, was checked using methacrolein that does
not contain enolizable hydrogens. As can be seen in
Table 2, negligible ee’s are observed even in this case.
The idea that two adjacent vacant coordination sites
on the catalyst may lead to a negative effect on the
asymmetric induction seems to be confirmed by the
work of Inoue and co-workers.25 While the present work
was in progress, these authors reported that using Pd
complexes very similar to 1 and 2, good ee values could
be observed only using N-acryloyloxazolidinone as the
dienophile. This dienophile is rather common in asym-
metric transition metal catalyzed Diels-Alder reactions
as, being able to chelate the transition metal as in
Figure 3A, it creates a rigid template and forces the
system to present preferentially one side of the CdC
double bond to the incoming diene. With ordinary
dienophiles (see in Figure 3B,C the case of acrolein) the
availability of two adjacent coordination sites will lead
to two alternative situations, pro-R and pro-S, with
On the basis of the good results reported above,
complexes 2 and 3 were tested in the enantioselective
version of some of the reactions reported in Table 3.
Cyclopentadiene and cyclohexadiene were chosen as the
dienes, while acrolein was chosen as the dienophile, at
least for the initial screening. The results are sum-
marized in Table 3. As can be seen, the reactivity in
the conversion of cyclopentadiene is good for most
complexes even at low temperature. The Pt complex 3b
shows a conversion similar to the Pd homologue 2b at
0 °C, while at -20 °C its reactivity is lower. With
cyclohexadiene, reactions were carried out at 25 °C
because at lower temperatures the reaction times
become exceedingly long. In the case of Pd complexes
the ee’s observed are virtually negligible, whereas in the
case of the Pt complex 3b the results are only slightly
better. It has to be pointed out that, in the case of exo
isomers, ee apparent values around 10% may fall within
the experimental error associated with analysis (GC).
(25) Oi, S.; Kashiwagi, K.; Inoue, Y. Tetrahedron Lett. 1998, 39,
It is conceivable that the Diels-Alder transformation
6253.

Cationic Complexes of Pd(II) and Pt(II)
Organometallics, Vol. 19, No. 24, 2000 5165
Ta ble 4. Ca ta lytic Activity in th e Diels-Ald er
Rea ction of Ach ir a l Com p lexes 4 a n d 5b
a
T 73 °C. b Reaction conditions: catalyst 10 µmol, diene 1 mmol,
F igu r e 3. Intermediates involved in the interaction
between (P-P)Pd fragments and the dienophile: A, with
N-acryloyloxazolidinone according to ref 25; B and C,
diastereoisomers formed with acrolein.
dienophile 1 mmol, DCE 3 mL, T 25 °C, N2 1 atm.
Ta ble 5. En a n tioselective Diels-Ald er
Tr a n sfor m a tion s Usin g Ch ir a l Com p lexes 4 a n d 5
a s Ca ta lystsa
Sch em e 3
virtually equivalent probability. It seems therefore that
the use of transition metal systems having only one
available coordination site may probably be beneficial
in maximizing the energy difference between the transi-
tion states corresponding to the attack of the diene from
either side of the CdC double bond.
Evolu tion of th e Ca ta lyst. In view of the above
considerations, our attention shifted toward complexes
4 and 5, which in principle, generating fragments that
possess only one coordination site upon bridge splitting
(Scheme 3), represent an evolution of catalysts 1-3
considered so far. Indeed, the characterization data
reported above indicate that, at least for complexes 4b,
the formulation as in Chart 1 is not exhaustive, as
species generated as indicated in Scheme 3 are also
present. The latter, except for the external counterion,
bear close similarities (or are identical) to complex 2b
(the case of 4bz may be different).
a
Reaction conditions: catalyst 10 µmol, diene 1 mmol, dieno-
phile 1 mmol, DCE 3 mL, N2 1 atm.
to Pd at -20 °C. The other reactions reported in Table
5 yield poor ee’s.
The data of Table 5 for the Pd complexes at 0 °C in
the reaction between cyclopentadiene and acrolein
indicate the existence of an effect of the counterion
(TfO-, ClO4-, BF4-) on the enantioselectivity and on the
As can be seen in Table 4, the catalytic performance
of the achiral complexes 4 and 5 is in general rather
good, with a moderate selectivity toward endo com-
pounds. Asymmetric transformations are reported in
Table 5 using the chiral complexes 4 and 5, all modified
with binap, and confirm the idea of the single vacant
site. The first reaction studied was the one between
cyclopentadiene and acrolein at 0 and -20 °C. Pal-
ladium complexes display a moderate activity and fair
enantioselectivities that increase upon lowering the
temperature. The homologous Pt complexes appear to
be more effective catalysts at both 0 and -20 °C, as the
reaction times are about 1 order of magnitude shorter.
Ee’s are higher for Pt at 0 °C and practically equivalent
activity (see also Figure 4). As can be seen, the effect of
-
the couterion on the activity follows the order BF4
>
ClO4 ∼ TfO-. A significant contribution by the non-
catalyzed reaction may also be present (Figure 4). The
opposite effect is observed on the enatioselectivity. While
the effect on the activity could be easily related to the
-
coordinating ability of the counterion (TfO- and ClO4
-
-
bind more strongly, BF4 is probably noncoordinated),
the effect on the enantioselectivity is less obvious.
A comparison of the catalytic behavior between Pd-
Cl complexes and the homologous Pt complexes indi-
cates that the latter have a number of advantages,
probably related to their more uniform composition and

5166 Organometallics, Vol. 19, No. 24, 2000
Pignat et al.
involved is always [(binap)Pd(µ-Cl)]+. This view is
supported by the observation that Pt complexes 5b
where the catalytically active species is only [(binap)-
Pt(µ-Cl)]+ show no effect of the counterion on both the
activity and the enantioselectivity. Under these circum-
stances, the overall effect of the counterion on the
enantioselectivity can be rationalized as follows: for all
complexes 4b there will be a constant positive contribu-
tion due to pathway B; in addition where pathway A is
more favorable in terms of reaction rate (i.e., with 4bz)
its contribution to the overall process will be relatively
more significant and hence the overall ee lower. Sum-
marizing, it can be said that a lower association of the
counterion will result in a higher weight of pathway A
and hence lower ee’s. Platinum complexes for which only
pathway B is viable yield higher ee’s and no effect of
the counterion.
F igu r e 4. Reaction profiles for the reaction between
cyclopentadiene and acrolein with different Pd complexes
as catalysts: circles, complex 4bx; diamonds, complex 4by;
squares, complex 4bz; triangles, no catalyst.
This conclusion could be better supported by the
knowledge of the catalyst mixture composition at 0 and
-20 °C. Unfortunately, the 31P{1H} NMR spectra at
these temperatures are not helpful, as the composition
can be determined only at -70 °C. Given the existence
of two nonselective contributions to the overall ee, i.e.,
pathway A and the likely involvement of the noncata-
lyzed reaction (Figure 4), it can be concluded that the
intrinsic enantiodifferentiating ability of [(binap)Pd(µ-
Sch em e 4
2+
Cl)]2 is probably much better than it appears from
Table 5.
Con clu sion s
This work has led to the synthesis of a variety of
cationic complexes of Pd(II) and Pt(II) that proved active
as catalysts in the Diels-Alder reaction. Spectroscopic
studies of these complexes have evidenced in some cases
the unexpected existence of an equilibrium in which the
external counterion (TfO-, ClO4-, BF4-) is able to
displace Cl- from the coordination sphere of the metal
and act as an η2 donor. All complexes have a high
catalytic activity (some are among the most active
catalysts reported in the literature), in many cases even
at low temperature, and some of them have an interest-
ing (albeit moderate) enantioselective discrimination
capacity. We have also demonstrated that in order to
observe enantiodiscrimination with simple reagents
such as acrolein only one coordination site must prefer-
ably be available in the catalytically active intermediate.
This is probably due to the formation of a more rigid
template interacting with the incoming diene in the
subsequent 4+2 cycloaddition.
the absence of equilibria, i.e., a higher activity in the
temperature range explored, a higher enantioselectivity,
and no dependence on the specific counterion used.
P ossible Or igin of th e En a n tioselectivity. It
should be reminded that at the temperature of catalytic
operations complexes 4b are indeed a mixture of two
different species, both catalytically active, i.e., [(binap)-
Pd(η2-Y)](Cl) (Y ) ClO4, TfO) or [(binap)Pd(solv)2](BF4)-
(Cl) and [(binap)Pd(µ-Cl)]2(Y)2 (Y ) BF4, ClO4, TfO).
Therefore, the enantioselective transformation, as well
as the catalytic transformation more generally, will be
the result of three different pathways (Scheme 4): the
spontaneous reaction (at least in the case of acrolein
and cyclopentadiene), i.e., pathway C, and the two
metal-mediated pathways A and B. In both cases the
interaction of the metal complex with the dienophile
leads to the formation of two diastereoisomers from
which the enantiomers are produced. Pathway A is
practically nonselective, as is demonstrated by the
results of Table 3, but on the other hand, it is sensitive
to the effect of the couterion on the activity. Therefore
it must be concluded that the enantiodifferentiating
ability of the overall system is due almost completely
to pathway B, which on the other hand is insensitive to
the counterion effect on the activity, as the species
Exp er im en ta l Section
Syn th esis of New Com p lexes. Full details on the spec-
troscopic methods, the starting materials, the complete ex-
perimental synthetic procedures, and the elemental analyses
of the new complexes are given separately in the Supporting
Information.
Ca ta lytic Rea ction s. These were carried out in a 25 mL
round-bottomed flask equipped with a stopcock for vacuum/
N2 operations and a sidearm fitted with a screw-capped silicone
septum to allow sampling. Constant temperature ((0.1 °C)
was maintained by water circulation through an external
jacket connected with a thermostat. Stirring was performed
by a Teflon-coated bar driven externally by a magnetic stirrer.

Cationic Complexes of Pd(II) and Pt(II)
Organometallics, Vol. 19, No. 24, 2000 5167
The following general procedure was followed. The required
amount of catalyst was placed solid in the reactor, which was
evacuated and filled with N2. Purified, N2-saturated solvent
was added in the required amount. After thermostating at the
required temperature for a few minutes, the dienophile and
the diene in the appropriate amount were sequentially injected
through the septum, and time was started.
All reactions were monitored with GC by direct injection of
samples taken periodically from the reaction mixtures with a
microsyringe. Separation of the products was performed on a
25 m HP-5 capillary column using a thermal conductivity
detector. Conversion percent was determined through the use
of calibration curves between the individual substrates and
toluene as internal standard.
The analytical conditions for the separation of the different
enantiomers are reported in the Supporting Information.
Ack n ow led gm en t. This work was supported by the
Ministry of University and Scientific Research. G.S.
wishes to thank Professor C. Bolm (RWTH, Aachen,
Germany) for first suggesting the possible use of Pd and
Pt complexes in the reaction here reported.
Su p p or tin g In for m a tion Ava ila ble: Complete experi-
mental procedures, elemental analyses of the new compounds,
and analytical conditions for the separation of the enantiomers
(Table 6). This material is available free of charge via the
Internet at http://pubs.acs.org.
The extent of the asymmetric induction (ee %) was deter-
mined with GC with the use of a Lipodex E chiral column.
OM0003943