Fesulphos-palladium(II) complexes as well-defined catalysts for enantioselective ring opening of meso heterobicyclic alkenes with organozinc reagents

Article abstract of DOI:10.1021/ja055692b

The air-stable and readily available cationic methyl palladium(II) complexes of planar chiral Fesulphos ligands [(Fesulphos)Pd(Me)(PhCN)] ⁺ X^- are highly efficient catalysts for the alkylative ring opening of oxa- and azabicyclic alk

Full text of DOI:10.1021/ja055692b

Published on Web 11/23/2005
Fesulphos-Palladium(II) Complexes as Well-Defined Catalysts
for Enantioselective Ring Opening of Meso Heterobicyclic
Alkenes with Organozinc Reagents
Silvia Cabrera, Ramo´n Go´mez Arraya´s, Ine´s Alonso, and Juan C. Carretero*
Contribution from the Departamento de Qu´ımica Orga´nica, Facultad de Ciencias, UniVersidad
Auto´noma de Madrid, 28049 Madrid, Spain
Received August 31, 2005; E-mail: juancarlos.carretero@uam.es
Abstract: The air-stable and readily available cationic methyl palladium(II) complexes of planar chiral
Fesulphos ligands [(Fesulphos)Pd(Me)(PhCN)]+ X- are highly efficient catalysts for the alkylative ring opening
of oxa- and azabicyclic alkenes with dialkylzinc reagents, showing broad scope with regards to both the
bicyclic substrate and the dialkylzinc reagent. Catalyst loading as low as 0.5 mol % is sufficient to achieve
good yields and enantioselectivities ranging 94 f 99% ee in most cases. {Fesulphos ) (1-phosphino-2-
sulfenylferrocene); X- ) tetrakis[3,5-bis(trifluoromethyl)phenyl]borate or PF6-].} The origin of the high
asymmetric induction has been rationalized by mechanistic studies combining computational calculations
and X-ray structural analysis.
Introduction
structures: the sulfur atom becomes stereogenic upon coordina-
tion to the metal, which imposes a unique asymmetric environ-
Two main structural concepts have proved to be greatly
successful in the design of chiral ligands for asymmetric
catalysis.1 The first concept is the reduction of the number of
possible diastereomeric transition states by using bidentate C2-
symmetrical chiral ligands with P/P, N/N, or O/O coordination
(e.g., BINAP, bisoxazolines, salen, or BINOL based ligands),
some of which have reached the status of “privileged ligands”
because of their broad applicability.2 The second strategy relies
on mixed bidentate structures equipped with a strong and weak
donor heteroatom pair, taking advantage of the different
electronic properties associated with each heteroatom-metal
bond (e.g., the trans influence). This electronic differentiation,
together with appropriate steric effects around the metal
coordinating heteroatoms, may create an asymmetric environ-
ment capable of inducing high levels of enantiocontrol. Some
bidentate P/N chiral ligands such as phosphine-oxazoline
systems and QUINAP constitute excellent examples of the latter
strategy.3
ment very close to the reactive metal center. Despite this, P/S
ligands reported to date have been successfully applied only to
a limited range of asymmetric transformations, being their
asymmetric reaction scope far away from that of the best ligands
based on P/P, N/N, or P/N coordination modes.
Within this context, we have recently developed a highly
tunable family of P/S ligands having only planar chirality, the
1-phosphino-2-sulfenylferrocenes (Fesulphos ligands 1, Chart
1).5 These compounds are readily prepared following a modular
three-step approach from commercially available ferrocene:
sulfinylation of ferrocenyllithium, fully diastereoselective ortho-
(4) For a review on chiral sulfur ligands, see: (a) Masdeu-Bulto´, A. M.;
Die´guez, M.; Martin, E.; Go´mez, M. Coord. Chem. ReV. 2003, 242, 159.
For leading recent references on P/S ligands in enantioselective catalysis,
see: sulfenyl phosphinites, (b) Evans, D. A.; Campos, K. R.; Tedrow, J.
S.; Michael, F. E.; Gagne, M. R. J. Am. Chem. Soc. 2000, 122, 7905. (c)
Pa`mies, O.; Die´guez, M.; Net, G.; Ruiz, A.; Claver, C. Organometallics
2000, 19, 1488. (d) Evans, D. A.; Michael, F. E.; Tedrow, J. S.; Campos,
K. R. J. Am. Chem. Soc. 2003, 125, 3534. (e) Kanayama, T.; Yoshida, K.;
Miyabe, H.; Takemoto, Y. Angew. Chem., Int. Ed. 2003, 42, 2054. (f)
Guimet, E.; Die´guez, M.; Ruiz, A.; Claver, C. Tetrahedron: Asymmetry
2005, 16, 959. Sulfenyl phosphines: (g) Enders, D.; Peters, R.; Lochtman,
R.; Raabe, G.; Runsik, J.; Bats, J. W. Eur. J. Org. Chem. 2000, 3399. (h)
Verdaguer, X.; Moyano, A.; Perica`s, M. A.; Riera, A.; Maestro, M. A.;
Mah´ıa, J. J. Am. Chem. Soc. 2000, 122, 10242. (i) Yan, Y.-Y.; RajanBabu,
T. V. Org. Lett. 2000, 2, 199. (j) Nakano, H.; Suzuki, Y.; Kabuto, C.;
Jujita, R.; Hongo, H. J. Org. Chem. 2002, 67, 5011. (k) Verdaguer, X.;
Perica`s, M. A.; Riera, A.; Maestro, M. A.; Mah´ıa, J. Organometallics 2003,
22, 1868. (l) Tu, T.; Zhou, Y.-G.; Hou, X.-L.; Dai, L.-X.; Dong, X.-C.;
Yu, Y.-H.; Sun, J. Organometallics 2003, 22, 1255. (m) Verdaguer, X.;
Lledo´, A.; Lo´pez-Mosquera, C.; Maestro, M. A.; Perica`s, M. A.; Riera, A.
J. Org. Chem. 2004, 69, 8053. (n) Zhang, W.; Shi, M. Tetrahedron:
Asymmetry 2004, 15, 3467. (o) Molander, G. A.; Burke, J. P.; Carroll, P.
J. J. Org. Chem. 2004, 69, 8062. (p) Nakano, H.; Takahashi, K.; Suzuki,
Y.; Fujita, R. Tetrahedron: Asymmetry 2005, 16, 609. (q) Sola´, J.; Riera,
A.; Verdaguer, X.; Maestro, M. A. J. Am. Chem. Soc. 2005, 127, 13629.
Binap(S): (r) Faller, J. W.; Wilt, J. C.; Parr, J. Org. Lett. 2004, 6, 1301.
(s) Faller, J. W.; Wilt, J. C. Org. Lett. 2005, 7, 633. (t) Faller, J. W.; Wilt,
J. C. Organometallics 2005, 24, 5076.
Although much less developed than the mixed P/N ligands,
in recent years considerable effort has been devoted to the
synthesis of chiral ligands based on a P/S coordination mode.4
In addition to the strong electronic differentiation imposed by
the phosphorus and sulfur bonding to the metal, a key structural
feature makes phosphine-thioether chiral ligands very appealing
(1) For excellent overviews on asymmetric catalysis, see: (a) ComprehensiVe
Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.;
Springer-Verlag: Berlin, 1999. (b) Williams, J. M. J. Catalysis in
Asymmetric Synthesis; Academic Press: Sheffield, 1999. (c) Catalytic
Asymmetric Synthesis, 2nd ed.; Ojima, I., Ed.; Wiley-VCH: New York,
2000. See also: (d) Trost, B. M. PNAS 2004, 101, 5348.
(2) Yoon, T. P.; Jacobsen, E. N. Science 2003, 299, 1691.
(3) (a) Pfatz, A.; Drury, W. J., III. PNAS 2004, 101, 5723. For a review on
P/N ligands in asymmetric catalysis, see: Guiry, P. J.; Saunders, P. C.
AdV. Synth. Catal. 2004, 346, 497.
(5) Garc´ıa Manchen˜o, O.; Priego, J.; Cabrera, S.; Go´mez Arraya´s, R.; Llamas,
T.; Carretero, J. C. J. Org. Chem. 2003, 68, 3679.
9
17938
J. AM. CHEM. SOC. 2005, 127, 17938-17947
10.1021/ja055692b CCC: $30.25 © 2005 American Chemical Society

Fesulphos-Pd(II) Complexes as Catalysts for Ring Opening
A R T I C L E S
Chart 1. Fesulphos Family of Ligands (ee >99%)
Table 1. Asymmetric Ring Opening of 7-Oxabenzonorbornadiene
with Me2Zn and Et2Zn Using Fesulphos Ligands 1
entry
ligand
R
t (h)
product
yielda (%)
eeb (%)
1
2
1a
1a
1b
1b
1c
1c
1d
1d
1e
1e
1f
Me
Et
Me
Et
Me
Et
Me
Et
Me
Et
Me
Et
48
24
48
24
96
96
96
96
24
96
96
96
24
18
96
96
96
3a
3b
3a
3b
3a
3b
3a
3b
3a
3b
3a
3b
3a
3b
3b
4
72
70
63
20c
15
15
15
27
73
68
68
33
65
16d
30
e
88
88
82
81
74
31
55
14
94
93
73
57
67
70
6
3
4
deprotonation/phosphonation of the resulting enantiopure sulfi-
nylferrocene, and subsequent sulfoxide to thioether reduction.
Interestingly, a very bulky substitution at sulfur (tert-butyl
group) leads to a single epimer at sulfur upon coordination to
a transition metal such as Pd(II),5,6 Pt(II),6 or Cu(I)6,7 that orients
the tert-butyl group on sulfur in anti arrangement with regard
to the ferrocene unit. By fine-tuning the steric and electronic
properties at phosphorus, these ligands have proved to afford
very high enantioselectivities in a variety of reactions such as
the palladium-catalyzed allylic substitution,5 the Diels-Alder
reaction of cyclopentadiene with N-acryloyl oxazolidinones,6
and the formal aza Diels-Alder reaction of sulfonyl aldimines
with activated dienes.7 As an extension of the applicability of
Fesulphos ligands in asymmetric catalysis, herein we describe
that these P,S-ligands, in combination with a palladium(II)
source, act as extremely efficient catalysts in the alkylative ring-
opening reaction of meso oxa- and azabicyclic alkenes with
organozinc reagents.8 In particular, their isolable, air-stable, and
structurally well-defined cationic methyl-palladium complexes
feature low catalyst loading, very high enantioselectivity, easy
fine-tuning, mild reaction conditions, and broad structural
scope.9
5
6
7
8
9
10
11
12
13
14
15
16
17
1f
1g
1g
1h
1i
Me
Et
Et
Et
1j
Et
3b
NR
a In pure product after chromatography. b Determined by chiral HPLC
(Chiralpak AD column). c Product 4 was also obtained in 37% yield.
d Product 4 was also obtained in 40% yield. e Alcohol 4 was formed as the
main product.
(CH3CN)2Cl2 (5 mol %), Fesulphos 1a (5 mol %) in CH2Cl2 at
room temperature. Pleasingly, the reaction was complete in 24
h, providing selectively the cis-dihydronaphthol (R,R)-3b10 in
86% yield with 80% ee. With this promising preliminar result
in hand, a brief study of solvents led us to find that the
enantioselectivity was somewhat higher in toluene (88% ee) and
1,2-dichloroethane (84% ee), while xylene (60% ee), chloroform
(42% ee), and diisopropyl ether (32% ee) produced a substantial
decrease of asymmetric induction.
Results and Discussion
Having established toluene as an optimal solvent, Table 1
shows the results obtained in the reaction of 2 with Et2Zn and
Me2Zn in the presence of Fesulphos ligands with different
substitutions at phosphorus and sulfur (ligands 1a-j). Some
relevant conclusions can be drawn from this study:
(a) By far, the best reactivity profile was observed with
ligands 1a and 1e, which led to complete conversions within
24 h of reaction (entries 1-2 and 9-10). In contrast, low yields
were obtained in the case of the bulky phosphines 1c and 1d
(entries 5-8), the difurylphosphine 1f (entries 11 and 12), and
the p-tolylsulfenyl ligand 1h (entry 15), due to incomplete
conversions after 24 h. The addition of Et2Zn to 2 in the presence
of the electron poor phosphines 1b and 1g resulted in the
formation of dihydronaphthol 4 as the main product, via a
reductive ring-opening process (entries 4 and 14).
Desymmetrization of Meso Oxabenzonorbornadienes in
the Presence of a Combination of PdCl2(CH3CN)2 and
Fesulphos Ligands. One of the most recent and interesting
enantioselective processes mediated by palladium catalysts is
the nucleophilic alkylative ring opening of meso oxabicyclic
and azabicyclic alkenes with dialkylzinc reagents reported by
Lautens et al.8 For this transformation, chiral phosphino-oxazo-
lines (especially t-Bu-PHOX),8a-d Tol-BINAP,8a-d MeOBiphep,8e
and the P-chiral diphosphine QuinoxP*8f proved to be highly
efficient ligands, affording the final alcohols with excellent
enantioselectivities (up to 97.6% ee).
To check the feasibility of Fesulphos ligands in this trans-
formation, we chose as a model reaction the ring-opening
addition of Et2Zn to 7-oxabenzonorbornadiene (2) under the
standard reported reaction conditions:8 Et2Zn (150 mol %), Pd-
(b) With ligands 1i (entry 16), homologous with regard to
1a, and 1j (entry 17) either no reaction or formation of the
reductive ring-opened product 4 was observed, suggesting that
the formation of six-membered palladacycle intermediates
instead of five-membered ones preclude the alkylative ring
opening process.
(6) Garc´ıa Manchen˜o, O.; Go´mez Arraya´s, R.; Carretero, J. C. Organometallics
2005, 24, 557.
(7) Garc´ıa Manchen˜o, O.; Go´mez Arraya´s, R.; Carretero, J. C. J. Am. Chem.
Soc. 2004, 126, 456.
(8) (a) Lautens, M.; Renaud, J.-L.; Hiebert, S. J. Am. Chem. Soc. 2000, 122,
1804. (b) Lautens, M.; Hiebert, S.; Renaud, J.-L. Org. Lett. 2000, 2, 1971.
(c) Lautens, M.; Hiebert, S.; Renaud, J.-L. J. Am. Chem. Soc. 2001, 123,
6834. (d) Lautens, M.; Hiebert, S. J. Am. Chem. Soc. 2004, 126, 1437. (e)
Dotta, P.; Kuwar, P. G. A.; Pregosin, P. S.; Albinati, A.; Rizzato, S.
Organometallics 2004, 23, 2295. (f) Imamoto, T.; Sugita, K.; Yoshida, K.
J. Am. Chem. Soc. 2005, 127, 11934. For a review on enantioselective
metal-catalyzed opening reactions of oxabicyclic alkenes, see: (g) Lautens,
M.; Fagnou, K.; Hiebert, S. Acc. Chem. Res. 2003, 36, 48.
(9) For a preliminary communication, see: Cabrera, S.; Go´mez Arraya´s, R.;
Carretero, J. C. Angew. Chem., Int. Ed. 2004, 43, 3944.
(c) As in previously studied reactions with Fesulphos
ligands,5-7 the presence of the bulky tert-butyl group at sulfur
(10) The absolute configuration of the alcohols 3a-b, 7a-b, 8a-b, and 16a
was established by comparison with the chiral HPLC data reported for both
enantiomers of these compounds (see refs 8a-d).
9
J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005 17939

A R T I C L E S
Cabrera et al.
Table 2. Asymmetric Ring Opening of Substituted
Oxabenzonorbornadienes with Me2Zn and Et2Zn Using Fesulphos
Ligands 1
Scheme 1. Unreactive Substrates in the Presence of Ligands 1
Scheme 2. Proposed Catalytic Cycle
entry
substrate
ligand
R
t (h)
product
yielda (%)
eeb (%)
1
2
3
4
5
6
7
8
9
5
5
5
5
5
5
6
6
6
6
6
6
1a
1a
1b
1b
1e
1e
1a
1a
1b
1b
1e
1e
Me
Et
Me
Et
Me
Et
Me
Et
Me
Et
Me
Et
96
24
96
96
48
96
96
24
96
48
96
96
7a
7b
7a
7b
7a
7b
8a
8b
8a
8b
8a
8b
43
72
30
<15
38
<10
47
70
35
53
42
92
80
88
88
88
82
>99
76
86
76
10
11
12
38
a In pure product after chromatography. b Determined by chiral HPLC.
Mechanistic studies described by Lautens et al. support the
enantioselective carbopalladation of the C-C double bond of
the bicyclic alkene by a presumed cationic alkyl palladium
intermediate [L2PdR]+ generated in situ from PdCl2(CH3CN)2
and R2Zn as the key step of the reaction pathway. The sub-
sequent fast â-oxygen elimination of the resulting σ-alkyl palla-
dium species would afford the observed cis-substituted product
(Scheme 2).8c In agreement with this proposal, it has been
recently described that the cationic µ-hydroxo complex [{Pd-
((R)-BINAP)-(µ-OH)}2]2+(OTf-)2 is a much more active cata-
lyst than BINAP in the reaction of oxabicycle 2 with Et2Zn.8d
Inspired by this mechanistic work, we envisaged the preparation
of cationic methylpalladium complexes of Fesulphos [(1)-
PdMe]+ with the aim of developing much more active catalysts.
In fact, if these cationic palladium complexes act as the actual
active catalytic species in this process, a great enhancement of
the reaction rate and enantioselectivity could derive from their
use. Additionally, a careful structural analysis of these cationic
methyl palladium complexes and their precursors, coupled with
a comparative study of their catalytic activity, would provide
important mechanistic insights on the origin of the enantiose-
lectivity during the presumed key carbopalladation step.
Synthesis of Cationic Methyl-Palladium Fesulphos Com-
plexes and Their Precursors. DFT Calculations of Neutral
Methylpalladium(II) Complex (1a)Pd(Cl)(Me). Treatment of
enantiopure ligands 1a-e with a stoichiometric amount of
PdCl2(CH3CN)2 in CH2Cl2 at room temperature afforded in very
high or nearly quantitative yields the dichloro complexes (1)-
PdCl2, which were readily isolated as air-stable orange solids.
As expected, these complexes were obtained in all cases as
single epimers at sulfur, thus orienting the tert-butyl group at
the sterogenic sulfur atom in anti arrangement with regard to
the ferrocene backbone. This fact had previously been confirmed
by X-ray analysis of several (1)PdCl2 complexes5,6 (see also
Figure 1a).
was essential for achieving high asymmetric inductions. Thus,
it was not surprising that nearly racemic product 3b was isolated
in the reaction catalyzed by the p-tolylsulfenyl ligand 1h (entry
15), whereas the corresponding tert-butylsulfenyl ligand 1a
provided 3b with 88% ee.
(d) The dicyclohexylphosphine 1e provided the highest
asymmetric inductions, affording the alcohols 3a and 3b with
excellent 94% and 93% ee, respectively (entries 9 and 10). The
rest of the ligands showed lower asymmetric inductions than
the parent diphenylphosphino ligand 1a.
Next, to evaluate the substrate compatibility of our catalyst
system, we performed the reaction of the meso electron-poor
substrate 511 and the electron-rich alkene 612 with Et2Zn and
Me2Zn in the presence of the three Fesulphos ligands (1a, 1b,
and 1e) that showed the best asymmetric inductions in the model
reaction with 2 (Table 2). From a stereoselectivity point of view,
the three ligands behaved rather homogeneously, leading to the
cis dihydronaphthols 7 and 8 with enantioselectivities in the
range 76-92% ee. The exception to this behavior was found
in the reaction of 6 with Me2Zn promoted by ligand 1b, in which
a single enantiomer of 8a was isolated (ee >99%, entry 9).
However, compared to the results obtained with the parent
substrate 2, an important drawback associated with the ring-
opening of 5 and 6 was the usually low to moderate reaction
conversions and/or the competitive formation of reductive ring-
opened byproducts. Even prolonged reaction times (4 days
instead of 24 h) or higher temperatures did not significantly
improve the chemical yields.
A more dramatic example of this insufficient reactivity was
observed from the bicyclic substrates 9, 10, and 11, which were
recovered unaltered after 48 h of reaction with Et2Zn or Me2-
Zn in the presence of PdCl2(CH3CN)2 and ligands 1a or 1e
(Scheme 1).
Interestingly, the transmetalation reaction of (1)PdCl2 with
either Me4Sn8d (3.0 equiv) in CH2Cl2 at room temperature or,
preferably, Me2Zn (1.5 equiv) in CH2Cl2 at room temperature
was completely stereoselective, presumably as a result of the
(11) Caster, K. C.; Keck, C. G.; Walls, R. D. J. Org. Chem. 2001, 66, 2932.
(12) Jung, M. E.; Lam, P. Y.-S.; Mansuri, M. M.; Spelt, L. M. J. Org. Chem.
1985, 50, 1087.
9
17940 J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005

Fesulphos-Pd(II) Complexes as Catalysts for Ring Opening
A R T I C L E S
Figure 1. Crystal structures of (1a)PdCl2, (1a)Pd(Cl)(Me), and [(1a)Pd(Me)(PhCN)]+[B(ArF)4]-. The hydrogen atoms and the [B(ArF)4]- counterion have
been omitted for clarity.
Table 3. Selected Bond Distances (Å) and Angles (deg)
(1a)Pd(Cl)(Me)
(DFT calculations)a
[(1a)Pd(Me)(PhCN)]+-
(1a)PdCl2
(1a)Pd(Cl)(Me)
[B(ArF)4]-
d (Pd-Cl1)
2.346
2.302
2.314
2.242
2.359
2.409
d (Pd-Cl2)
d (Pd-S)
2.444
2.218
2.049
90.2
2.587
2.294
2.065
87.4
2.464
2.217
2.039
90.1
d (Pd-P)
d (Pd-Me)
θ (S-Pd-P)
90.8
89.4
88.3
θ (S-Pd-Cl1)
θ (P-Pd-Cl2)
θ (P-Pd-Me)
τ (Pd-S-C1-C2)
τ (Pd-P-C1-C2)
τ (C1-S-C3-C4)
τ (C2-P-C5-C6)
91.6
91.6
90.5
1.8
-11.9
67.5
93.1
3.5
-19.5
73.4
90.8
8.6
-15.3
64.4
0.1
-12.6
79.0
172.5
170.6
-165.9
-165.6
a B3LYP with a combination of basis sets and polarization functions for atoms attached to the metal (see computational details within the Supporting
Information).
strong electronic differentiation exerted by the phosphane and
thioether moieties, affording a single complex (1)Pd(Cl)(Me).13
These air-stable highly crystalline yellow-orange complexes
were readily purified by standard silica gel chromatography and
proved to be moisture and air-stable at room temperature. The
cis stereochemistry between the methyl group and the phosphane
was effected by its treatment with stoichiometric amounts of
AgPF6 or NaB(ArF)4 [ArF) 3,5-bis(trifluoromethyl)phenyl,
Brookhart’s reagent]15 in a 6:1 CH2Cl2-benzonitrile mixture.
Filtration of the resulting AgCl or NaCl precipitate, followed
by evaporation of the solvent and trituration with hexanes-
Et2O, led to the cationic methyl-palladium benzonitrile-
coordinated complexes as stable yellow-orange solids (Scheme
3). The coupling constants 2JP,C(Me) (<1.0 Hz) and 3JP,H(Me)(1.3-
1.6 Hz) were very similar to those of the precursor (1a)Pd(Cl)-
(Me), reflecting that the methyl group on palladium is positioned
cis with respect to the phosphorus atom. Pleasingly, careful
recrystallization of the tetraarylborate complex [(1a)Pd(Me)-
(PhCN)][B(ArF)4] afforded suitable crystals for X-ray analysis,
allowing confirmation of the anti arrangement of the benzonitrile
ligand to the phosphorus atom16 (Figure 1c).
The following structural features are worthy of consider-
ation: (a) All compounds exhibit a perfect square planar
geometry of the ligands around the metal center (sum of bond
angles around Pd in the range of 360.0-360.2°), with very
similar bond angles θ S-Pd-P, θ S-Pd-Cl(1), and θ S-Pd-
Me. (b) The five-membered palladacycle slightly deviates from
planarity, having dihedral angles close to zero for Pd-S-C(1)-
C(2) (somewhat higher in the cationic complex, 7.98°) and 12°-
1
moiety was established by H and 13C NMR spectra, the low
2
3
coupling constants JP,C(Me) (<2.1 Hz) and JP,H(Me) (2.2-2.8
Hz) being of great diagnostic value and similar to the values
reported for achiral methyl P,S-palladium complexes.14 Single-
crystal X-ray analysis of (1a)Pd(Cl)(Me) confirmed this ster-
eochemistry (Figure 1b). DFT calculations on (1a)Pd(Cl)(Me)
complex provides structural parameters very close to their X-ray
data (Table 3), giving us confidence in the reliability of the
computational approach used (see computational details in the
Supporting Information). Moreover, the complex with trans
relative orientation between the methyl group and the phosphane
moiety was optimized by the same method, resulting to be much
less stable (5.8 kcal mol-1) than the cis isomer, which is the
only one observed.
The cationic methyl-palladium complex was readily prepared
and isolated from (1a)Pd(Cl)(Me). The chloride ion abstraction
(13) Dimethylation of (1)PdCl2 was not observed even with an excess of Me2-
Zn (over 5.0 equiv) under prolonged reaction times.
(14) For achiral cationic methyl P,S-palladium complexes, see: (a) Suranna,
G. P.; Mastrorilli, P.; Nobile, C. F.; Keim, W. Inorg. Chim. Acta 2000,
305, 151. (b) Daugulis, O.; Brookhart, M.; White, P. S. Organometallics
2002, 21, 5935.
(15) (a) Brookhart, M.; Grant, B.; Volpe, A. F., Jr. Organometallics 1992, 11,
3920. (b) Brookhart, M.; Wagner, M. J. Am. Chem. Soc. 1994, 116, 3641.
(16) Two slightly different structures were detected in the crystal, differing
mainly in the orientation of the counterion.
9
J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005 17941

A R T I C L E S
Cabrera et al.
Scheme 3. Synthesis of Fesulphos Palladium Complexes
Table 4. Asymmetric Ring Opening of 7-Oxabenzonorbornadiene with Me2Zn Catalyzed by Palladium(II) Complexes of Ligands 1
entry
cat. (x mol %)
additive (x mol %)
T (°
C)
t (min)
yielda [%]
eeb [%]
1
2
3
4
5
6
7
8
9
10
11
12d
13d
14d
(1a)PdCl2 (5)
(1a)Pd(Cl)(Me) (5)
25
25
25
25
25
25
25
25
25
25
25
1440
1440
420
360
10
30
30
15
30
360
10
30
60
300
80c
74c
83
68
86
78
82
71
74
80
77
88
87
88
81
87
83
61
78
78
72
69
64
46
83
95
97
97
(1a)Pd(Cl)(Me) (5)
AgPF6 (5)
[(1a)PdMePhCN]+(PF6)- (5)
(1a)Pd(Cl)(Me) (5)
NaB(ArF)4 (5)
(1a)Pd(Cl)(Me) (0.5)
NaB(ArF)4 (0.5)
[(1a)PdMePhCN]+(BArF
(1b)Pd(Cl)(Me) (5)
(1c)Pd(Cl)(Me) (5)
(1d)Pd(Cl)(Me) (5)
(1e)Pd(Cl)(Me) (5)
(1e)Pd(Cl)(Me) (5)
(1e)Pd(Cl)(Me) (1)
(1e)Pd(Cl)(Me) (0.2)
)
4
- (0.5)
NaB(ArF)4 (5)
NaB(ArF)4 (5)
NaB(ArF)4 (5)
NaB(ArF)4 (5)
NaB(ArF)4 (5)
NaB(ArF)4 (1)
NaB(ArF)4 (0.2)
-25
-25
-25
a
c
1
In pure product after chromatography. b Determined by chiral HPLC (Chiralpak AD column). Conversion yield calculated by H NMR. d Reaction
performed in 1,2-dichloroethane.
13° for Pd-P-C(2)-C(1). This geometry determines that one
of the phenyl groups at phosphorus and the bulky tert-butyl
group on sulfur are oriented in an almost parallel diaxial
arrangement, generating a sterically very congested area in the
upper face of the palladacycle moiety. (c) Some subtle differ-
ences are observed in the bond lengths around palladium. The
most significant is the elongation of the Pd-S bond in the
methyl-palladim complexes (1a)Pd(Cl)(Me) (2.444 Å) and [(1a)-
Pd(Me)(PhCN)]+ (2.464 Å) compared to that of the starting
dichloro complex (1a)PdCl2 (2.314 Å), as result of the strong
trans influence of the highly σ-donating methyl group. On the
other hand, both methyl-palladium complexes are geometrically
very similar around the metal center, the cationic complex
possessing a Pd-Me bond somewhat shorter (2.039 vs 2.049
Å) and the Pd-S somewhat longer (2.464 vs 2.444 Å) than the
neutral species, likely as a result of the enhanced σ-donation of
the methyl group to the more electron-deficient cationic Pd atom.
The reaction in the presence of catalyst (1a)PdCl2 exhibited
a similar result to that of the combination 1a+PdCl2(CH3CN)2
(see Table 1), reaching 80% conversion after 24 h to give the
alcohol (R,R)-3a in 81% ee (entry 1). A very similar reactivity
was observed in the case of using the complex (1a)Pd(Cl)(Me),
which provided 74% conversion after 24 h, leading to 3a in
87% ee (entry 2). Interestingly, a significant increase in the
reaction rate occurred when a stoichiometric amount of AgPF6
(5 mol %) was added to dissociate the chloride ligand, with the
consequent formation of a cationic palladium complex17 (entry
3). The reaction was finished in 7 h, providing 3a in 83% yield
and similar enantioselectivity (83% ee). This acceleration effect
was confirmed in the absence of silver, by using a sample of
the isolated salt [(1a)Pd(Cl)(PhCN)]+PF6- as catalyst (entry 4),
instead of the in situ generated complex (entry 3).
However, the most dramatic enhancement in the reaction rate
was produced when NaB(ArF)4 (5 mol %) was used as chloride
scavenger (entry 5). Under these conditions a complete conver-
sion was observed within just 10 min, suggesting that the
reactivity of the presumed key cationic palladium catalyst greatly
depends on the nature of the anionic counterion, the bulky
noncoordinating B(ArF)4- anion being optimal. This outstanding
reactivity allowed a dramatic decrease of the catalyst loading:
0.5 mol % of catalyst was sufficient to observe complete
disappearance of the starting material within 30 min without
loss of enantioselectivity (entry 6). As expected, a similar result
in terms of reactivity, chemical yield, and enantioselectivity was
Reaction of Oxabicycle 2 with Me2Zn Catalyzed by
Fesulphos-Palladium Complexes. To test the efficiency of the
well-defined Fesulphos-palladium catalysts previously synthe-
sized, we studied the reaction of oxabicycle 2 with Me2Zn. As
the starting point we used similar experimental conditions to
those previously employed when combining PdCl2(CH3-
CN)2+Fesulphos ligand (5 mol % of catalyst in toluene at room
temperature). The reactions were monitored by TLC until
disappearance of the starting alkene 2, except for very slow
transformations which were stopped after 24 h. The results are
summarized Table 4.
(17) A similar reactivity was observed using AgSbF6 instead of AgPF6.
9
17942 J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005

Fesulphos-Pd(II) Complexes as Catalysts for Ring Opening
A R T I C L E S
Scheme 4. Ring-Opening Reaction of 2 with Several Dialkylzinc
Reagents
loading, maintaining very reasonable reaction times. For
instance, the reactions of 2 with Me2Zn in 1,2-dichloroethane
at -25 °C in the presence of 1 mol % of this optimal catalyst
afforded alcohol 3a in an excellent 97% ee within 60 min (entry
13), while a 0.2 mol % of catalyst led to the same enantiose-
lectivity after 5 h (entry 14).
Interestingly, Et2Zn and Bu2Zn did also efficiently participate
in the ring opening reaction of 2 using 0.5 mol % of catalyst
(Scheme 4). Particular attention deserves the addition of Bu2-
Zn, wich led to alcohol 3c with complete enantiocontrol (>99%
ee).
obtained using 0.5 mol % of the corresponding isolated cationic
complex [(1a)Pd(Me)(PhCN)]+[B(ArF)4]- (entry 7), instead of
the in situ generated catalyst (entry 6).
To fine-tune the enantioselectivity of the process, we next
studied the effect of the substitution at phosphorus in the in
situ generated cationic Fesulphos-palladium complexes [(1b-
e)Pd(Me)]+. For comparison purposes all reactions were con-
ducted in the presence of 5 mol % of (1)Pd(Cl)(Me) and
NaB(ArF)4 (entries 8-11). With the exception of the complex
containing the bulky naphthylphosphane ligand 1d (entry 10),
very high reactivity was observed again in all cases, the reaction
reaching completion within just 10-30 min.
The observed enantioselectivities nicely parallel those found
with the couple PdCl2(CH3CN)2+ligand 1 as the catalyst system
(Table 1). Thus, complexes derived from ligands 1b, 1c, and
1d provided a lower enantioselectivity (46-69% ee) than that
achieved from the complex of 1a (78% ee, entry 6), while the
palladium complex containing the dicyclohexyl phosphine 1e
afforded the highest asymmetric induction (entry 11, 83% ee).
The combination (1e)Pd(Cl)(Me)+NaB(ArF)4 proved to be so
effective that the reaction could be conducted at a much lower
temperature (-25 °C), which resulted in a remarkable enhance-
ment in the enantioselectivity (95% ee, entry 12). Outstandingly,
this catalytic system also admits a severe reduction in the catalyst
Scope of the Alkylative Ring-Opening Catalyzed by (1)-
Pd(Me)+. Once developed a very active and enantioselective
catalyst for the alkylative ring opening of compound 2 with
several dialkylzinc reagents, we turned our efforts toward
studying the scope and generality of the process with regard to
the alkene substrate. We focused our attention in the reaction
of Me2Zn, Et2Zn and Bu2Zn with the meso oxabicyclic
compounds 5, 6 and 9, which showed modest or very poor
reactivity with the standard combination of PdCl2(CH3CN)2 and
Fesulphos ligand (Table 2). To confirm the feasibility of using
a low catalyst loading and the influence of the substitution at
phosphorus, we systematically explored the addition of the
dialkylzinc reagent in the presence of 0.5 mol % of the in situ
formed cationic methyl-Pd complex of the three ligands that
provided the best results in the opening of the model substrate
2 [(1a)PdMe+, (1b)PdMe+, and (1e)PdMe+]. All reactions were
carried out in DCE at room temperature by addition of
dialkylzinc reagent to a mixture of the corresponding complex
(1)Pd(Cl)(Me) (0.5 mol %), NaB(ArF)4 (0.5 mol %), and the
oxabicyclic alkene. The results are collected in Table 5.
Table 5. Asymmetric Ring Opening of Oxabenzonorbornadienes with R2Zn Catalyzed by Cationic Pd-Complexes of 1a, 1b, and 1e
ent.
substrate
R1
R2
R3
catalyst
t (min)
product
yielda (%)
eeb (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
5
5
5
5
5
5
5
6
6
6
6
6
6
6
9
9
9
9
9
9
9
F
F
F
F
F
F
F
H
H
H
H
H
H
H
H
H
H
H
H
H
H
OMe
OMe
OMe
OMe
OMe
OMe
OMe
Me
Et
Me
Et
Me
Et
Bu
Me
Et
Me
Et
Me
Et
Bu
Me
Et
Me
Et
1a‚PdMe+
1a‚PdMe+
1b‚PdMe+
1b‚PdMe+
1e‚PdMe+
1e‚PdMe+
1e‚PdMe+
1a‚PdMe+
1a‚PdMe+
1b‚PdMe+
1b‚PdMe+
1e‚PdMe+
1e‚PdMe+
1e‚PdMe+
1a‚PdMe+
1a‚PdMe+
1b‚PdMe+
1b‚PdMe+
1e‚PdMe+
1e‚PdMe+
1e‚PdMe+
60
60
600
600
30
30
60
10
30
120
1440
10
30
d
7a
7b
7a
7b
7a
7b
7c
8a
8b
8a
8b
8a
8b
8c
12a
12b
12a
12b
12a
12b
12c
87
80
43c
74
95
79
75
62
77
58
45
71
61
d
80
90
89
54
98
85
91
83
81
77
87
95
96
96
90
88
88
81
>99
94
OCH2O
OCH2O
OCH2O
OCH2O
OCH2O
OCH2O
OCH2O
H
H
H
H
H
30
91
89
95
66
97
94
98
180
1200
1440
15
60
30
Me
Et
Bu
H
H
a In pure product after chromatography. b Determined by chiral HPLC (Chiralpak AS, Chiralcel AS and Chiralpak AD columns). c Conversion yield.
d Rapid aromatization of the addition product was observed.
9
J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005 17943

A R T I C L E S
Cabrera et al.
Table 6. Enantioselective Ring Opening of Nonaromatic
Scheme 5. Attempt of Kinetic Resolution of Trisubstituted Alkene
13
[2.2.1]Oxabicyclic Alkenes with R2Zn Catalyzed by Cationic
Methyl-Pd Complexes of 1a and 1e
To our delight, all ring-opening processes occurred with good
chemical yields and high enantioselectivities, leading to the
known cis-dihydronaphthols 7a-b8a and 8a-b,8a and the new
compounds 7c and 12a-c, without formation of significant
amounts of side products. However, the most outstanding result
was the astonishing great stereochemical fidelity displayed by
this family of Fesulphos ligands. Again, as for substrate 2, in
all cases the catalyst containing the electron-rich phosphine 1e
provided the best asymmetric induction, allowing a very high
enantiocontrol of the process (94f99% ee, entries 5-7, 12,
13, and 19-21). The second most efficient catalyst proved to
be the diphenylphosphino complex (1a)Pd(Me)+ (81-91% ee,
entries 1, 2, 8, 9, 15, and 16), while the electron-poor bis(p-
fluorophenyl)phosphino complex (1b)Pd(Me)+ was both less
stereoselective and sharply less reactive (66-88% ee, entries
3, 4, 10, 11, 17, and 18).
entry
R1
R2
x
ligand
time
product yielda (%)
eeb (%)
1
2
3
4
5
6
Bn
Bn
Me
Me
5
5
1
1
1
1
1a
1e
1a
1e
1e
1e
20 h
22 h
2 h
15a
15a
16a
16a
68
79
70
61
>99
92
96
97
OTBDPS Me
OTBDPS Me
OTBDPS Et
OTBDPS Bu
2 h
25 min 16b
1 h 16c
70(71)c
67
83(90)c
90
a In pure product after chromatography. b Determined by chiral HPLC.
c Reaction performed at -20 °C.
Table 7. Enantioselective Ring-Opening of
Azabenzonorbornadiene with R2Zn
As a last oxa-benzonorbornadiene substrate, we attempted
the kinetic resolution of racemic 13, having a methyl group at
the alkene moiety. Unexpectedly, in this case the reaction with
Et2Zn in the presence of (1e)Pd(Me)+ promoted the endo to
exo isomerization of the C-C double bond (product 14) instead
of the alkylative ring-opening reaction (Scheme 5).
t
yielda
(%)
eeb
(%)
entry
R
catalyst
x
(h) product
1
2
3
4
5
Me (1a)Pd(Cl)(Me) + NaB(ArF)4
Me [(1a)Pd(Me)(PhCN)]+[B(ArF)4]-
Me [(1a)Pd(Me)(PhCN)]+(PF6)-
Me [(1a)Pd(Me)(PhCN)]+(PF6)-
Et [(1a)Pd(Me)(PhCN)]+(PF6)-
5
5
5
1
5
25 18a
24 18a
0.5 18a
48 18a
24 18b
79
98
97
70c
73 >99
60c
65
99
33
Alkylative Ring-Opening of Less Reactive Substrates.
From a synthetic point of view, meso nonaromatic [2.2.1]-
oxabicyclic alkenes are particularly interesting substrates for
enantioselective alkylative ring-opening reactions, since they
lead to substituted cyclohexenols having four contiguous
stereogenic centers. However, the use of these compounds is
much more restricted than that of their oxabenzonorbornadiene
counterparts due to their low reactivity. For instance, this
transformation promoted by Pd(Tol-BINAP)Cl2 requires either
high reaction temperatures and/or the addition of stoichiometric
amounts of additives such as Lewis acid [e.g., Zn(OTf)2] or
EtMgBr.8d Pleasingly, in our case, 5 mol % of catalysts (1a)-
PdMe+ and (1e)PdMe+ induced the ring opening of 10 with
Me2Zn in toluene at room temperature in the absence of any
additive, to give the cis-cyclohexenol 15a with reasonable yields
and excellent enantioselectivities (92f99% ee, Table 6, entries
1 and 2). Interestingly, the silyl ether derivative 11 was even
more reactive, the reaction reaching completion within 2 h at
room temperature in the presence of only 1 mol % of catalyst,
affording 16a with 96-97% ee (entries 3 and 4).10 The ring-
opening of alkene 11 with other dialkylzinc reagents such as
Et2Zn and Bu2Zn in the presence of 1 mol % of catalyst (1e)-
PdMe+was next studied. The reaction with Et2Zn was even
faster, leading to alcohol 16b with 70% yield and 83% ee after
25 min at room temperature (90% ee at -20 °C, entry 5). On
the other hand, the butyl derivative 16c was similarly obtained
with 67% yield and 90% ee by addition of Bu2Zn to 10 (1 h, rt,
entry 6).
a In pure product after chromatography. b Determined by chiral HPLC.
c Conversion yield determined by 1H NMR from the crude reaction mixture.
DCE due to their decreased reactivity.8b In this transformation,
unlike the behavior of the oxa-analogue 2, complex (1a)PdMe+
with the hexafluorophosphate anion was more reactive than the
corresponding tetraarylborate complex (Table 7). In addition,
the complex [(1a)Pd(Me)(PhCN)](PF6) (5 mol %) afforded the
corresponding ring-opened product 18a in good isolated yield
(73%) and virtually complete enantioselectivty (>99% ee, entry
4).18 The use of 1 mol % of [(1a)PdMe](PF6) resulted in a much
slower reaction (60% conversion after 48 h), although without
loss of enantioselectivity (99% ee, entry 4). The reaction with
Et2Zn was surprisingly much less enantioselective, giving 18b
in 65% yield with only 33% ee (entry 5).
Mechanistic Hypothesis. The X-ray structures of palladium
complexes of 1 (Figure 1), together with computational studies
gave us strong insights on the mechanistic forces behind the
high performance displayed by these P,S-palladium catalysts.
Considering that, according to the general mechanism proposed
by Lautens et al., the stereogenic carbon centers are created in
the syn-carbopalladation step on the key cationic palladium-
π-alkene complex; we studied theoretically this reaction by DFT
(B3LYP) method. As model structures in our theoretical study
we chose the parent ligand 1a and oxanorbornadiene as alkene.
(18) The cis stereochemistry of (+)-17a was confirmed by comparison of its
physical and spectroscopical data with those reported in the literature for
racemic cis-17a and trans-17a (see, for instance: Go´mez Arraya´s, R.;
Cabrera, S.; Carretero, J. C. Org. Lett. 2005, 7, 219). The absolute
stereochemistry of (+)-17a was assumed as (1R,2R) according to the
reaction mechanism.
This protocol was finally extended to the desymmetrization
of azabenzonorbornadiene derivatives, a kind of alkene whose
opening reaction with Me2Zn had been described in refluxing
9
17944 J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005

Fesulphos-Pd(II) Complexes as Catalysts for Ring Opening
A R T I C L E S
Figure 2. Reaction coordinate of carbopalladation to give the minor enantiomer (S,S)-3a. Hydrogen atoms have been omitted for clarity.
The reaction coordinates of the carbopalladation step to afford
the minor enantiomer (S,S)-3a and the major enantiomer (R,R)-
3a are presented in Figures 2 and 3, respectively, along with
the relative energies of the optimized geometries for each
intermediate and transition state. After considering several
geometries for the coordination of oxanorbornadiene with the
complex (1a)Pd(Me)+, structures I-IV were found as the
minimum of energy according to their harmonic frequencies.
The four structures exhibit a perfect square planar geometry of
the ligand around palladium (∑θ[Pd] ) 359.8°-360.6°). Since
I (in which only the oxygen atom is coordinated to the metal
[d(PdsO) ) 2.214 Å]) resulted the most stable structure, the
formation of this complex was considered the initial stage of
the carbopalladation step. A η2-coordination of the CdC double
bond to palladium was found in complexes II-IV, as deduced
from the PdsC distances (2.39-2.53 Å) and the slight elonga-
tion of the CdC bond (1.360-1.365 Å) compared to the same
bond length in intermediate I (1.334 Å). In complex II the
alkene moiety is in perpendicular arrangement with regard to
the coordination plane [(CdC)sPdsMe dihedral angle )
-70.4°], the oxygen atom pointing opposite to the methyl group.
Almost the opposite orientation of the alkene was found in
structure III, though a deviation from perpendicularity was
observed [(CdC)sPdsMe dihedral angle ) -56.3°], likely
to avoid steric interactions between the oxygen and the methyl
group. However, complex IV, formed by rotation of the alkene,
showed a quasi coplanar arrangement of the alkene, both carbon
atoms of the CdC double bond being almost within the
coordination plane of palladium [(CdC)sPdsMe dihedral
angle ) -10.4°].19 Considering the distances between the
methyl group and the closest alkene carbon, complex II can be
envisaged as an intermediate in the formation of the minor
enantiomer of the carbopalladation product (V), whereas
complex IV would be involved in the formation of the major
enantiomer VI.
In addition, carbopalladation products V and VI, as well as
the transition states leading to them (TS1 and TS2, respectively),
were optimized. Complex V [d(PdsC) ) 2.109 Å; d(PdsO)
) 2.226 Å] was found to be somewhat more stable than VI
[d(PdsC) ) 2.105 Å; d(PdsO) ) 2.251 Å], the latter having
the methyl group oriented to the more sterically hindered up
face of the ligand. Both transition states TS1 and TS2 have the
PdsC σ-bond almost completely formed (2.111 Å). However,
while the MesC bond (2.145 and 2.163 Å, respectively) is being
formed, the PdsMe bond (2.295 and 2.303 Å, respectively)
and CdC bond (1.435 and 1.431 Å, respectively) are being
broken. The transition state TS1, in which the oxygen atom is
directed to the more sterically hindered up face of the ligand,
was found to be 3.3 kcal mol-1 less stable than TS2, being this
large difference in energy consistent with enantioselectivities
greater than 90% observed experimentally.
These studies suggest that there are two key structural aspects
that determine the enantioselectivity in the carbopalladation
step: (a) coordination of the bicyclic alkene trans to phosphorus
(19) Perpendicular and coplanar arrangements of alkene ligands with respect to
the coordination plane have been reported for alkene-platinum complexes.
The actual orientation of the alkene has been proposed to be governed by
steric factors. See, for instance: Collman, J. P.; Norton, J. R.; Finke, R. G.
In Principles and Applications of Organotransition Metal Chemistry;
University Science Books, Mill Valley, CA, 1987; pp 41-42.
9
J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005 17945

A R T I C L E S
Cabrera et al.
Figure 3. Reaction coordinate of carbopalladation to give the major enantiomer (R,R)-3a. Hydrogen atoms have been omitted for clarity.
Scheme 6. Mechanistic Proposal
of the substitution at phosphorus, the optimal catalyst (1e)Pd-
(Me)+ combines a broad scope, low catalyst loadings (0.5 mol
% in most cases), and smooth reaction conditions (-25 °C or
rt), affording the corresponding ring-opened products in good
yields (65-95%) and with homogeneously high enantioselec-
tivities (90f99% ee). A combination of computational studies
at DFT (B3LYP) level and a complete X-ray structural analysis
of palladium complexes, including the presumed catalytically
active cationic methyl-Pd complex, shows that Fesulphos
ligands fulfill unique structural and electronic requirements to
provide high asymmetric induction in the key carbopalladation
step. Specifically, the high asymmetric performance of these
ligands in this transformation could rely on the strong trans effect
of the phosphine moiety, acting in combination with the great
sterically demanding environment imposed by the stereogenic
sulfur atom directly bonded to the palladium atom.
on the palladium atom and (b) coordination of the alkene from
its less hindered exo face, orienting the oxygen (or nitrogen)
bridge to the opposite side of the very bulky tert-butyl group
(intermediate IV). Therefore, we can conclude that the syner-
gistic effect derived from the strong electronic trans effect of
the phosphorus moiety, together with the great steric control
exerted by the close bulky stereogenic sulfur substitution, would
be responsible for the high asymmetric induction displayed by
Fesulphos catalysts in this transformation (Scheme 6).
Acknowledgment. This work was supported by the Ministerio
de Educacio´n y Ciencia (MEC, Project BQU2003-0508) y
Consejer´ıa de Educacio´n de la Comunidad de Madrid (Project
GR/MAT/0016/2004). S.C. thanks the UniVersidad Auto´noma
de Madrid (UAM) for a predoctoral fellowship. R.G.A. thanks
the MEC for a Ramo´n y Cajal Contract. We acknowledge the
skilled assistance of Dr. Jaime Rodr´ıguez Gonza´lez in solving
the X-ray crystal structure of the cationic palladium complex.
Prof. Jose´ L. Garc´ıa-Ruano is gratefully acknowledged for
unrestricted access to the HPLC equipment of his group. We
also thank the Centro de Computacio´n Cient´ıfica (UAM) for
computation time, and Dr. M. Alcam´ı (UAM) and Dr. Feliu
Maseras (Institute of Chemical Research of Catalonia) for
Conclusions
In summary, the highly tunable family of Fesulphos ligands
(1), owing P,S-coordination mode and exclusively planar
chirality, act as very efficient ligands in the enantioselective
alkylative ring opening of oxa- and azabicyclic alkenes with
dialkylzinc reagents. In particular, their readily available cationic
methylpalladium complexes display an unprecedented high
reactivity with regard to a variety of substrates. After fine-tuning
9
17946 J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005

Fesulphos-Pd(II) Complexes as Catalysts for Ring Opening
A R T I C L E S
helpful discussions. Johnson Matthey PLC is gratefully ac-
knowledged for a generous loan of PdCl2.
coordinates for all optimized structures and their absolute
energies. This material is available free of charge via the Internet
at http://pubs.acs.org.
Supporting Information Available: Detailed experimental
1
procedures and characterization data. Copies of H NMR and
13C NMR of new compounds. Computational details. Cartesian
JA055692B
9
J. AM. CHEM. SOC. VOL. 127, NO. 50, 2005 17947