A new chiral metalladiphosphine for early-late chemistry. Crystal structure of a D-camphor-based zirconium enolato phosphine and rearrangement of heterobimetallic Zr/Pd complexes

Article abstract of DOI:10.1039/b008227i

The bis(phosphinoenolato) zirconocene 2, prepared from the phosphino camphor derivative 3-exo-PPh₂C₁₀H₁₅O 1, represents the first C₂-symmetric chiral diphosphine containing an early transition metal centre tethered to the phosphino groups via an enolato linkage. It allows the selective assembly of Lewis-acid early-late bimetallic combinations, as exemplified with the Zr/Cu and Zr/Ag complexes 4 and 5. The monophosphinoenolato titanocene 3 was also prepared. Reactions of 2 or 3 with [Pd(dmba)(μ-Cl)]₂ (dmba = o-C₆H₄CH ₂NMe₂) or [PdCl₂(SEt₂)₂] led to complete transfer of the functional ligand to the Pd(II) centre, with formation of [(dmba)Pd(PPh₂C₁₀H₁₄O)] 6 or cis-[Pd(PPh₂C₁₀H₁₄O)₂] 8 and [Cp₂ZrCl₂]/[Cp₂TiCl₂], respectively. A heterobimetallic Zr/Pd intermediate complex 10 of this rearrangement could be detected. Complexes 6 and 8 were prepared independently by reaction of 1 with [Pd(dmba)(μ-Cl)]₂ or [Pd(SEt₂)₂Cl ₂], respectively, which first afforded complexes 7 and 9 which were then treated with KH. The crystal structures of 2, 6 and 8·THF· H₂O have been determined by X-ray diffraction. The Royal Society of Chemistry 2001.

Full text of DOI:10.1039/b008227i

View Article Online / Journal Homepage / Table of Contents for this issue
A new chiral metalladiphosphine for early–late chemistry. Crystal
structure of a D-camphor-based zirconium enolato phosphine and
rearrangement of heterobimetallic Zr/Pd complexes†
Chris Mattheis,a Pierre Braunstein*a and Axel Fischerb
a Laboratoire de Chimie de Coordination, UMR 7513 CNRS, Université Louis Pasteur,
4 rue Blaise Pascal, 67070 Strasbourg Cedex, France. E-mail: braunst@chimie.u-strasbg.fr
b Chemisches Institut, Otto-von-Guericke-Universität Magdeburg, Universitätsplatz 2,
Gebäude 16, 39106 Magdeburg, Germany
Received 12th October 2000, Accepted 22nd January 2001
First published as an Advance Article on the web 26th February 2001
The bis(phosphinoenolato) zirconocene 2, prepared from the phosphino camphor derivative 3-exo-PPh2C10H15O 1,
represents the first C2-symmetric chiral diphosphine containing an early transition metal centre tethered to the
phosphino groups via an enolato linkage. It allows the selective assembly of Lewis-acid early–late bimetallic
combinations, as exemplified with the Zr/Cu and Zr/Ag complexes 4 and 5. The monophosphinoenolato titanocene
3 was also prepared. Reactions of 2 or 3 with [Pd(dmba)(µ-Cl)]2 (dmba = o-C6H4CH2NMe2) or [PdCl2(SEt2)2] led to
complete transfer of the functional ligand to the Pd() centre, with formation of [(dmba)Pd(PPh2C10H14O)] 6 or
cis-[Pd(PPh2C10H14O)2] 8 and [Cp2ZrCl2]/[Cp2TiCl2], respectively. A heterobimetallic Zr/Pd intermediate complex 10
of this rearrangement could be detected. Complexes 6 and 8 were prepared independently by reaction of 1 with
[Pd(dmba)(µ-Cl)]2 or [Pd(SEt2)2Cl2], respectively, which first afforded complexes 7 and 9 which were then treated
with KH. The crystal structures of 2, 6 and 8ؒTHFؒH2O have been determined by X-ray diffraction.
Introduction
Two particularly active research areas in inorganic and organo-
metallic chemistry concern chiral diphosphine ligands owing
to their applications in homogeneous catalysis,1–3 and the study
of metal complexes with phosphorus–oxygen hybrid ligands
whose hard/soft character allows hemilabile behaviour in solu-
tion4 or confers unique catalytic properties.5 With the desire of
combining these two facets of metal–phosphine chemistry, we
prepared a new chiral diphosphine consisting of two enolato
phosphine ligands oxygen-bonded to a zirconocene unit. This
unique early metal centred ligand system should subsequently
allow the synthesis of early–late heterobimetallic complexes.
Combining two metals with contrasting properties in the same
molecule not only offers the potential of inducing novel
chemical transformations that would not be possible with either
metal alone but may also improve the yield or selectivity of
existing reactions.6
Results and discussion
Slight modifications of the literature method enabled us to
increase the yield of the -camphor-based β-keto phosphine
Scheme 1
(1R)-endo-(ϩ)-3-diphenylphosphino-1,7,7-trimethylnorborn-2-
one 1 from 37%7 to 65% (see Experimental section). Reactions
of 1 with [Cp2ZrCl2] and NEt3, or with [Cp2ZrMe2] were unsuc-
may enhance the sensitivity of the Zr–O bond to water, the
splitting of H2O to hydroxy ions and protons being triggered by
formation of phosphonium ions. The 31P{1H} NMR resonance
for compound 2 at δ Ϫ31.8 is shifted to higher field compared
to that of the free ligand 1 (δ 0.1 for 1), owing to the form-
ation of the enolate species. Similarly, the 13C NMR resonance
of C(2) (δ 184.44) is shifted to higher field compared to 1
cessful; however treatment of 1 with n-BuLi at Ϫ78 ЊC followed
by the addition of [Cp2ZrCl2] in a 2 : 1 stoichiometry led to the
formation of the colourless zirconocene complex 2 in 70% yield
(Scheme 1).
This highly moisture-sensitive complex was characterized by
NMR spectroscopy and X-ray diffraction (see Experimental
section). The presence of the basic P atoms within the molecule
2
(δ 218.2, d, JPC = 7.3 Hz in 17), whereas a downfield shift is
observed for C(3) (δ 100.24) and H(4) (δ 2.4) [for 1: δ C(3) =
1
54.2, d, JPC = 28.4 Hz;7 1H NMR: δ H(4) = 1.93]. Interest-
ingly, the carbon atoms of the Cp ligands show coupling to
† Dedicated to Professor H. Schnöckel on the occasion of his 60th
birthday, with our most sincere congratulations.
phosphorus.8
800
J. Chem. Soc., Dalton Trans., 2001, 800–805
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b008227i

View Article Online
Preliminary complexation studies with late transition metals
have shown that 2 is indeed able to behave as a chelating
metalladiphosphine. Rotation of the camphor moiety about
the Zr–O bond allows 2 to react with silver() triflate or
[Cu(NCMe)4]PF6 hexafluorophosphate to form the new chiral
heterobimetallic early–late transition metal complexes 4 and 5
with Lewis-acid properties. They were too sensitive to give
satisfactory elemental analysis but could be characterized by
NMR spectroscopy and mass spectrometry (see Experimental
section).
However, no heterobimetallic Zr/Pd or Ti/Pd complexes
could be isolated when 2 or 3 were reacted with [Pd(dmba)-
(µ-Cl)]2 (dmba = o-C6H4CH2NMe2). Instead, [Cp2ZrCl2] or
1
Fig. 1 An ORTEP view of the structure of 2. Thermal ellipsoids are
[Cp2TiCl2] were detected in the H NMR spectrum and the
singlet at δ 26.8 in the 31P NMR spectrum of the reaction
mixture was assigned to the palladium phosphino enolate
complex 6 [eqn. (1)].
drawn at the 30% probability level.
Single crystals of 2, suitable for X-ray analysis, were grown
from toluene–n-hexane. The ligand arrangement around the
metal is pseudo-tetrahedral (Fig. 1) and it is noteworthy that
the bulky enolato phosphine ligands are wrapped back towards
the Zr atom. The O–Zr–OЈ plane bisects the Cp–Zr–Cp (Cp =
midpoint of cyclopentadienyl ring) angle of 50.8Њ [Cp(C41–
C45)–(O–Zr–OЈ)] 25.5Њ; (O–Zr–OЈ)–Cp(C46–C50) 25.3Њ. The
ligand conformations within the two slightly different best
planes defined by P–C(3)–C(2)–O and PЈ–C(3Ј)–C(2Ј)–OЈ,
respectively, may result from packing effects. A related situation
(1)
has been observed with the enolate ligands in [Cp Ti(OCH᎐
᎐
2
CH2)2].9 The rather short Zr–O distances [2.002(4) Å av.] and
the large Zr–O–C bond angles [154.0(3) av.] indicate some
metal–oxygen π interactions.10,11 The C–O bond distances
2
[1.330(6) Å av.] are typical for a Csp –O single bond (statistical
average 1.34 Å) and together with the value of the C(2)–C(3)
bond length [1.361(6) Å av.], this indicates a reduced electron
delocalization within the enolato moiety.11,12
A careful choice of the metal is crucial for the formation
of a metal centred diphosphine ligand. Thus, reaction of
two equivalents of lithiated 1 with [Cp2TiCl2] led to a mixture
of titanium enolato phosphine species with partial loss of
the Cp moieties. This stands in contrast to the synthesis of
Similarly, reactions with the chloride-free, cationic complex
[Pd(dmba)(NCMe)2]PF6 only allowed identification of 6 among
the products.
Complex 6 was prepared independently by reaction of 7
(see below), obtained from 1 and [Pd(dmba)(µ-Cl)]2 (Scheme 2),
with KH. It is sensitive to hydrolysis.
the bisenolato complexes [Cp Ti(OCH᎐CH ) ]9 or [Cp Ti(OC-
᎐
2
2
2
2
Me᎐CH ) ]10a and may be due to increased steric hindrance
᎐
2
2
in our case. Accordingly, the use of only one equivalent of
lithiated 1 afforded the monoenolato phosphine titanium
complex 3 in 70% yield which was characterized by NMR
spectroscopy (see Experimental section).
Scheme 2
Crystals of 6 suitable for X-ray determination were obtained
from dichloromethane–n-hexane. The coordination geometry
around the metal is square planar (Fig. 2). The palladium is
chelated by the dmba moiety and coordination of the phos-
phorus occurs in cis position to the aryl group, in agreement
with the antisymbiotic effect13 and with other related struc-
tures.14 The coordination sphere of the palladium atom is com-
pleted by the oxygen atom of the phosphinoenolate ligand. The
Pd–O and Pd–P bond distances in 6 [2.122(6) and 2.254(3) Å,
J. Chem. Soc., Dalton Trans., 2001, 800–805
801

View Article Online
Fig. 2 An ORTEP view of the structure of 6. Thermal ellipsoids are
drawn at the 30% probability level.
respectively] are slightly longer than in the related com-
plexes with a negatively charged oxygen donor atom [(dmba)-
Fig. 3 An ORTEP view of the structure of 8 in 8ؒTHFؒH2O. Thermal
ellipsoids are drawn at the 30% probability level.
14b
Pd{Ph2PCH2C(O)O}],14a [(dmba)Pd{Ph2PCH᎐C(᎐O)OEt}]
ؒؒ ؒؒ
Crystals of 8 suitable for X-ray determination were obtained
from THF–n-hexane. The unit cell of 8 contains one molecule
of water and of THF, which do not interact with the metal
complex. We find the expected cis-arrangement of the chelating
phosphinoenolate ligands around the square planar coordin-
ated palladium (Fig. 3). The Pd–O bond distances [2.097(4)
and 2.093(4) Å] are relatively short compared to the examples
cited above (see discussion for 6)14a–c whereas the Pd–P bond
distances [2.260(2) and 2.254(2) Å] are somewhat longer.
For the rearrangement leading to the transfer of the phos-
phinoenolate ligand from Zr to Pd, we suggest the sequence
shown in Scheme 4.
and [(dmba)Pd[Ph2PC{C(O)NHPh}C(Ph)O],14c [2.105(3)/
2.218(1), 2.117(5)/2.242(2) and 2.096(3)/2.234(1) Å, respect-
ively]. The bond distances and angles around the dmba moiety
and around the phosphorus atom all are in the expected range.
The air-stable complex 7 was obtained in 77% yield accor-
ding to Scheme 2. Conductivity measurements proved it to be
an 1 : 1 electrolyte with a molar conductivity of Λm = 59 S cm2
molϪ1 in methanol, which is less than the expected value for a
fully dissociated ion pair such as [(dmba)Pd(Ph2PC10H16OH)]Cl
which contains a limonene-based phosphino-alcohol ligand
(Λm = 74 S cm2 molϪ1, dissociated to 100% in methanol).15 We
thus assume the existence of an equilibrium between a neutral
(7a) and a cationic species (7b). In polar solvents (methanol)
7b is stabilized by coordination of the keto group, a situation
similar to that observed in other cationic β-ketophosphine
palladium complexes with a five-membered ring chelate,
such as [(dmba)Pd{Ph2PCH2C(O)Ph}](CF3SO3)16a and cis-
[Pd{Ph2PCH2C(O)Ph}2](BF4)2.16b
The reaction of 2 with [PdCl2(SEt2)2] also led to a ligand
exchange and resulted in the formation of the bis(phosphino-
enolate) palladium complex cis-[Pd(Ph2PC10H14O)2] 8 and [Cp2-
ZrCl2] (Scheme 3). Similar results were observed with the
Scheme 4
Reaction of 2 with [Pd(dmba)(µ-Cl)]2 should occur by a
classical phosphine-induced chloride-bridge splitting reaction
and yield the heterobimetallic complex 10 which is present
in a neutral (10a) and a cationic complex (10b). In cationic
10b the oxygen of the enolato function is coordinated to
the palladium. This can be compared with the cationic struc-
tures of 7b, [(dmba)Pd{Ph2PCH2C(O)Ph}](CF3SO3) and cis-
Scheme 3
[Pd{Ph2PCH2C(O)Ph}2](BF4)2.16 Complex 10b contains already
the structural components of the chelate complex 6. Nucleo-
philic attack of the chlorides at the zirconium and switching the
Zr–O electron pair to the palladium would lead to 6 and
[Cp2ZrCl2]. If the equilibrium of the bimetallic species is shifted
platinum precursor [PtCl2(NCPh)2]. Complex 8 was prepared
independently by reaction of 1 with [PdCl2(SEt2)2] (formation
of 9) followed by treatment with potassium hydride. Complexes
8 and 9 have been previously prepared by different routes.7,17
802
J. Chem. Soc., Dalton Trans., 2001, 800–805

View Article Online
to 10b but the latter rearranges much faster to 6 than reacts
back to 10a, it will not be possible to observe any Zr–Pd com-
plex. Although this remains difficult to prove, we could provide
at least partial evidence for this mechanism. By shifting the
equilibrium with added salt, neutral 10a could be stabilized
enough to be analysed by NMR spectroscopy and mass
spectrometry. In the presence of added LiBr a broad singlet was
observed in the 31P NMR spectrum at δ 25.25 which is close to
that for 6 (δ 26.76) (2 was always present in amounts less than
5%) and this is consistent with the very similar environments of
the phosphorus atoms in 6 and 10a. At Ϫ40 ЊC, the signals
in the 31P and 1H NMR spectra become sharper. Two signals are
observed in the 31P NMR spectrum, the major one corresponds
to 10a whereas the smaller is assigned to the bromide analogue
resulting from halide exchange. Adventitious hydrolysis of
10a in an NMR tube led to the formation of 7a,b and their
bromide analogues (δ 36.5, 36.8) and [(Cp2ZrCl)2O]. In the
mass spectrum, the peaks at m/z 832.0 and 575.1 were assigned
to A and 6, respectively, and could result from fragmentation
of 10a.
10b rearranges by loss of [Cp2ZrCl2] to 6 faster than it reacts
back to 10a. The rearrangement is driven by the favourable
formation of a chelate complex and by the nucleophilic attack
of the chlorides at the zirconium leading to [Cp2ZrCl2]. If the
tendency to form chelate complexes is reduced, heterobimetallic
complexes can be obtained, as shown in the case of the Zr–Ag/
Zr–Cu complexes 4 and 5, respectively.
These results indicate that a careful choice of the second
metal is necessary to observe a bridging mode for the P,O-
ligand. Another promising approach to prepare stable
phosphinoenolato heterobimetallic complexes consists of the
tuning of the ligands, for instance by using aromatic ligands
such as phenolates on the early transition metal (which would
also lower the sensitivity of the complexes towards water).
Finally, a reduced number of Cp ligands at the zirconium in
2 should lead to increased Lewis acidity of the metal.
Experimental
General procedures
All reactions and manipulations were carried out under an inert
atmosphere of purified nitrogen using standard Schlenk-tube
techniques. Nitrogen (Air liquide, R-grade) was passed through
BASF R3-11 catalyst and 4 Å molecular sieve columns to
remove residual oxygen and water. Solvents were dried and dis-
tilled under nitrogen before use: pentane, hexane and toluene
over sodium, tetrahydrofuran and diethyl ether over sodium–
benzophenone, acetonitrile and dichloromethane over calcium
hydride. Elemental C, H and N analyses were performed by the
Service de microanalyses du CNRS. The 1H, 31P{1H} and
13C{1H} NMR spectra were recorded at 300.1, 121.5 and 75.5
MHz respectively, on a Bruker AM300 instrument. Phosphorus
chemical shifts were externally referenced to 85% H3PO4 in
H2O with downfield chemical shifts reported as positive. All
NMR spectra were run at room temperature in CDCl3 if not
stated otherwise. Conductivity measurements were carried out
in methanol at room temperature. Electrospray mass spectra
were run on a HP 1100 series LC/MSD spectrometer.
An alternative, stepwise mechanism (Scheme 5) may be
envisaged which cannot be distinguished from that of Scheme 4
with the present data.
Preparations
Synthesis of 1. Reaction of one equivalent of (1R)-endo-(ϩ)-
3-bromocamphor in THF with 1.1 equivalent of n-BuLi at
Ϫ78 ЊC was followed by the addition of 1 equivalent of PPh2Cl
at Ϫ78 ЊC. The mixture was stirred overnight and the solvent
was removed under reduced pressure. Column chromatography
on silica gel with Et2O as the eluent afforded 1 in 65% yield as a
white solid. Analytical data were in agreement with literature
values.7
Synthesis of 2. Reaction of two equivalents of 1 in THF with
1.1 equivalent of n-BuLi at Ϫ78 ЊC was followed by the
addition of 0.5 equivalent of [Cp2ZrCl2] at Ϫ78 ЊC. The mix-
ture was stirred overnight and the solvent was removed under
reduced pressure. Recrystallisation from toluene–n-hexane
afforded 2 in 70% yield as a white solid. Satisfactory elemental
analyses could not be obtained owing to the moisture sensitivity
1
of the complex. H NMR: δ 0.66 [m, 2H, H(5)], 0.74 (s, 6H,
Scheme 5
Me), 0.93 (s, 6H, Me), 1.11 (s, 6H, Me), 1.48 [m, 2H, H(6)], 1.20
[m, 2H, H(6Ј)], 1.62 [m, 2H, H(5Ј)], 2.4 [br, 2H, H(4)], 6.44
(s, 10H, Cp), 7.29–7.36/7.52–7.57 (2 m, 20H, Ph); 13C NMR:
δ 11.09 [s, C(10)], 20.09/20.84 [2 s, C(8/9)], 27.68 [s, C(5)], 32.55
Conclusion
2
By using the new bis(phosphinoenolato)zirconocene complex 2
we could access new chiral heterobimetallic Zr/Ag and Zr/
Cu complexes. With Pd() precursors, a ligand redistribution
reaction leading to [Cp2ZrCl2] and 6 was observed for which a
neutral (10a) and a cationic (10b) Zr/Pd species are suggested
intermediates. They exist in an equilibrium but only 10a may
be observed by shifting the equilibrium with LiBr, because
[s, C(6)], 51.97 [d, JPC = 6.0, C(4)], 53.94 [s, C(7)], 57.21 [d,
1
3JPC = 5.9, C(1)], 100.24 [d, JPC = 9.5, C(3)], 113.83 (t,
5JPC = 3.0, Cp), 127.20 (s, p-Ph), 128.00 (s, p-PhЈ), 128.02
3
3
(d, JPC = 3.7, m-Ph), 128.16 (d, JPC = 8.2, m-PhЈ), 132.55 (d,
2JPC = 17.6, o-Ph), 134.07 (d, JPC = 21.1, o-PhЈ), 139.47 (d,
2
1JPC = 9.8, ipso-Ph), 140.88 (d, JPC = 6.4, ipso-PhЈ), 184.44
1
[d, 2JPC = 29.3 Hz, C(2)]; 31P{1H} NMR: δ Ϫ31.8 (s).
J. Chem. Soc., Dalton Trans., 2001, 800–805
803

View Article Online
Table 1 Selected 1H and 31P{1H} NMR data for 6, 7, 8, 9 and 10 (δ in ppm) in CDCl3
Me
Me
Me
H4
NMe2
NCH2
H3 a/Cpb
31P{1H}
6
7
0.75
0.77
0.94
0.91
0.97
0.91
2.21 (t, 3.5 Hz)
2.96 (t, 3.9 Hz)
2.86 (d, 2.4 Hz),
2.88 (d, 2.1 Hz)
2.75 (d, 2.4 Hz),
2.87 (d, 2.7 Hz)
—
3.92 (dd, 2.1, 13.5 Hz),
4.00 (d, 13.8 Hz)
3.82 (dd, 3.0, 10.5 Hz),
4.44 (m)
—
—
n.f.
—
26.76
4.24 (d, 13.3 Hz)
35.80, 35.79d
8
9
0.67
0.80
0.84
1.03
0.94
1.10
1.13
0.99
1.38
4.44 (m)
3.01 (t, 4.5 Hz)
2.25 (br)
—
3.57 (m)
5.50
31.07
18.26
25.25
—
10c
2.62, 2.83 (br)
a H3 in 7 and 9. b Cp in 10. c At Ϫ40 ЊC. d In C6D6.
Synthesis of 3. Reaction of one equivalent of 1 in THF
with 1.1 equivalent of n-BuLi at Ϫ78 ЊC was followed by the
addition of 1 equivalent of [Cp2TiCl2] at Ϫ78 ЊC. The mixture
was stirred overnight and the solvent removed under reduced
pressure. Recrystallisation from toluene–n-hexane afforded 3
ClNOPPd: C, 60.79; H, 6.09. Found: C, 60.1; H, 6.70%. See
NMR data in Table 1.
Synthesis of 8. To a solution of 9 (0.31 g, 0.37 mmol) in THF
(20 mL) potassium hydride (0.04 g, 1.0 mmol) was added
at 0 ЊC. After the mixture was stirred overnight at room
temperature, it was filtered and concentrated to 5 mL and then
n-hexane (30 mL) added. At Ϫ30 ЊC, 0.15 g (0.19 mmol, 52%)
of the beige product crystallized. Anal. Calc. for C44H48O2P2Pd:
C, 68.00; H, 6.22. Found: C, 67.5; H, 6.8%. See NMR data in
Table 1.
1
in 70% yield as a red solid. H NMR: δ 0.70 (s, 3H, Me), 0.88
(s, 3H, Me), 1.08 (s, 3H, Me), 2.44 [m, 1H, H(4)], 6.46 (s, 10H,
Cp), 7.26–7.32/7.50–7.54 (2 m, 10H, Ph); 13C NMR: δ 11.17 [s,
C(10)], 19.91/20.76 [2 s, C(8/9)], 27.60 [s, C(5)], 32.54 [s, C(6)],
2
3
52.43 [d, JPC = 6.0, C(4)], 54.31 [s, C(7)], 60.45 [d, JPC = 5.8,
1
C(1)], 100.24 [d, JPC = 9.7, C(3)], 117.40/117.91 (2 s, Cp),
127.32 (s, p-Ph), 128.11 (s, p-PhЈ), 128.09 (d, 3JPC = 4.1, m-Ph),
3
2
Synthesis of 9. To a solution of 1 (0.89 g, 2.65 mmol) in
THF (15 mL) [PdCl2(SEt2)2] (0.47 g, 1.32 mmol) was added at
room temperature. After the solution was stirred for 30
min, it was filtered and concentrated to 7 mL and then n-hexane
(30 mL) added. At Ϫ30 ЊC, 0.93 g (1.09 mmol, 83%) of the
orange product precipitated. Anal. Calc. for C44H50Cl2O2P2Pd:
C, 62.17; H, 5.93. Found: C, 61.4; H, 6.6%. See NMR data in
Table 1.
128.23 (d, JPC = 4.3, m-PhЈ), 132.29 (d, JPC = 17.5, o-Ph),
2
1
134.01 (d, JPC = 20.1, o-PhЈ), 138.85 (d, JPC = 7.9, ipso-Ph),
140.31 (d, 1JPC = 5.6, ipso-PhЈ), 189.70 [d, 2JPC = 27.4 Hz, C(2)];
31P{1H} NMR: δ Ϫ31.5 (s).
Selected data for 4. Satisfactory elemental analyses could
not be obtained owing to the moisture sensitivity of the com-
1
plex. H NMR: δ 6.13 (s, 10H, Cp), 0.78 (s, 3H, Me), 1.02
(s, 3H, Me), 1.13 (s, 3H, Me), 2.35 [m, H(4)]; 31P{1H} NMR:
δ Ϫ13.6 (d, 1JP Ag = 478.6, 1JP Ag = 551.4 Hz, with the expected
magnetogyric ratios of 107Ag and 109Ag [γ(107Ag) : γ(109Ag) =
0.87 : 1]); FAB-MS m/z 998.8 (Mϩ).
Synthesis of 10. To a solution of 2 (0.40 g, 0.45 mmol) and
LiBr (0.04 g, 0.45 mmol) in THF (30 mL) [Pd(dmba)(µ-Cl)]2
(0.24 g, 0.44 mmol) was added at room temperature. After the
solution was stirred for 1 h, it was filtered and concentrated to
8 mL and then n-pentane (20 mL) added. At Ϫ80 ЊC, 0.40 g of
the beige product mixture precipitated.
107
109
Selected data for 5. Satisfactory elemental analyses could
not be obtained owing to the moisture sensitivity of the com-
plex. 1H NMR: δ 6.14 (s, 5H, Cp), 0.80 (s, 3H, Me), 1.05 (s, 3H,
Me), 1.20 (s, 3H, Me), 2.02 (s, MeCN), 2.52 [m, 1H, H(4)];
X-Ray crystal structure determination of 2, 6 and 8ؒTHFؒH2O
1
31P{1H} NMR: δ Ϫ26.57 (s), Ϫ143 (sept, JPF = 714 Hz, PF6);
Crystal data for complex 2. A colourless crystal of 2 was
coated with mineral oil, mounted on a glass fibre and trans-
ferred to the cold nitrogen stream (Bruker AXS Smart CCD
System with Mo-Kα radiation (λ = 0.71073 Å) and graphite
monochromator; Siemens LT-2 attachment). Crystal data:
C54H58O2P2Zr, M = 892.16, monoclinic, space group P21,
a = 9.582(8), b = 12.547(1), c = 18.959(16) Å, β = 97.67(1),
U = 2259(3) Å3, T = 143 K, Z = 2, Dc = 1.312 g cmϪ3, µ = 0.36
mmϪ1. A total of 24146 reflections (θ = 2.0–28.2Њ) were col-
lected (11141 independent reflections (Rint = 0.066)) and were
used for structure solution (direct methods) and refinement
(full-matrix least squares on F2); R1[I > 2σ(I)] = 0.0368, wR2
(all data) = 0.0559. The hydrogen atoms were calculated and
fixed in idealized positions. The absolute structure was deter-
mined with x = Ϫ0.06(2). For all computations, the SHELXTL
package (PC)18 was used. Selected bond lengths and angles are
given in Table 2.
FAB-MS: m/z 1195.1; does not correspond to the calculated
value for Mϩ = 952.3 but shows the expected Cu : Zr = 1 : 1
isotopic pattern.
Synthesis of 6. To a solution of 7 (0.26 g, 0.42 mmol) in THF
(20 mL), potassium hydride (0.04 g, 1 mmol) was added at 0 ЊC.
After the mixture was stirred overnight at room temperature,
it was filtered and concentrated to 5 mL and then n-hexane
(30 mL) added. At Ϫ30 ЊC, 0.15 g (0.26 mmol, 62%) of the
beige product crystallized. Anal. Calc. for C31H36NOPPd: C,
64.64; H, 6.30. Found: C, 64.6; H, 6.1%. 13C NMR: δ 9.62
[s, C(10)], 20.56/20.68 [2 s, C(8/9)], 28.56 [s, C(5)], 32.69 [s,
C(6)], 49.16/49.54 (2 s, NMe2), 55.74/55.91/56.78 [3 s, C(1/4/7)],
1
71.32 (s, NCH2), 88.15 [d, JPC = 57.2, C(3)], 122.11/123.41/
2
125.39 (3 s, C6H4), 127.93 (d, JPC = 10.4, p-Ph), 128.05
2
(d, JPC = 9.3, p-PhЈ), 129.35/129.52 (2 s, o-Ph/o-PhЈ), 133.12
3
3
(d, JPC = 11.5, m-P), 133.82 (d, JPC = 11.5, m-PhЈ), 134.41/
135.10 (2 s, ipso-Ph/ipso-PhЈ), 138.43 (d, JPC = 9.8, C6H4),
147.53 (d, JPC = 6.3, C6H4), 149.23 (s, C6H4), 200.23 [d,
2JPC = 19.3 Hz, C(2)].
Crystal data for complex 6. Colourless crystals from
dichloromethane–n-hexane. A needle of 6 was coated with
mineral oil, mounted on a glass fibre and transferred to the
cold nitrogen stream (Bruker AXS Smart CCD System with
Mo-Kα radiation (λ = 0.71073 Å) and graphite monochrom-
ator; Siemens LT-2 attachment). Crystal data: C31H36NOPPd,
M = 575.98, monoclinic, space group C2, a = 15.9544(15),
b = 31.8046(15), c = 11.0466(12) Å, β = 95.916(6), U = 5575 Å3,
T = 173 K, Z = 8, Dc = 1.372 g cmϪ3, µ = 0.75 mmϪ1. A total
of 18772 reflections (θ = 1.9–28.2Њ) were collected (10447
Synthesis of 7. To a solution of 1 (0.52 g, 1.55 mmol) in
toluene (15 mL) [Pd(dmba)(µ-Cl)]2 (0.43 g, 0.78 mmol) was
added at room temperature. After the solution was stirred for
30 min, it was filtered and concentrated to 10 mL and then
n-hexane (20 mL) added. At Ϫ30 ЊC 0.73 g (1.2 mmol, 77%)
of the yellow product precipitated. Anal. Calc. for C31H37-
804
J. Chem. Soc., Dalton Trans., 2001, 800–805

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) for 2 with estimated
CNRS for financial support and an associate research position
(to C. M.) and the COST D12/0016/98 action of the European
Commission (Brussels).
standard deviations in parentheses
Zr–O
Zr–OЈ
1.995(2)
2.009(2)
2.522(3)
2.525(3)
2.518(3)
2.535(3)
2.534(3)
2.497(3)
Zr–C(47)
Zr–C(48)
Zr–C(49)
O–C(2)
C(2)–C(3)
P–C(3)
OЈ–C(2)
C(2Ј)–C(3Ј)
P–C(3Ј)
2.520(3)
2.537(3)
2.552(3)
1.327(3)
1.366(3)
1.791(3)
1.333(3)
1.355(3)
1.798(3)
References
Zr–C(41)
Zr–C(42)
Zr–C(43)
Zr–C(44)
Zr–C(45)
Zr–C(46)
1 H. Brunner and W. Zettlmeier, Handbook of Enantioselective
Catalysis with Transition Metal Compounds, VCH, Weinheim, 1993,
vol. 2.
2 R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley,
New York, 1994.
3 W. A. Herrmann and B. Cornils, Angew. Chem., Int. Ed. Engl., 1997,
36, 1049.
4 A. Bader and E. Lindner, Coord. Chem. Rev., 1991, 108, 27;
C. S. Slone, D. A. Weinberger and C. A. Mirkin, Prog. Inorg. Chem.,
1999, 48, 233; P. Braunstein and F. Naud, Angew. Chem., Int. Ed.,
2001, 40, 680.
O–Zr–OЈ
99.83(9)
154.7(2)
130.7(2)
124.5(2)
Zr–OЈ–C(2Ј)
O–C(2Ј)–C(3Ј)
C(2Ј)–C(3Ј)–P
153.3(2)
130.3(2)
125.4(2)
Zr–O–C(2)
O–C(2)–C(3)
C(2)–C(3)–P
5 See, for example: W. Keim, Angew. Chem., Int. Ed. Engl.,
1990, 29, 235; W. Keim, New J. Chem., 1994, 18, 93; P. Braunstein,
Y. Chauvin, S. Mercier, L. Saussine, A. DeCian and J. Fischer,
J. Chem. Soc., Chem. Commun., 1994, 2203; P. Braunstein,
Y. Chauvin, J. Nähring, A. DeCian, J. Fischer, A. Tiripicchio and
F. Ugozzoli, Organometallics, 1996, 15, 5551.
6 R. M. Bullock and C. P. Casey, Acc. Chem. Res., 1987, 20, 167;
G. S. Ferguson, P. T. Wolczanski, L. Párkányi and M. C. Zonnevylle,
Organometallics, 1988, 7, 1997; D. W. Stephan, Coord. Chem. Rev.,
1989, 95, 41; A. M. Baranger and R. G. Bergman, J. Am. Chem.
Soc., 1994, 116, 3822; P. Veya, C. Floriani, A. Chiesi-Villa and
C. Guastini, Organometallics, 1994, 13, 208; T. W. Graham,
A. Llamazares, R. McDonald and M. Cowie, Organometallics,
1999, 18, 3490; N. Wheatley and P. Kalck, Chem. Rev., 1999, 99,
3379; P. Braunstein and J. Rosé, in Metal Clusters in Chemistry, ed.
P. Braunstein, L. A. Oro and P. R. Raithby, Wiley-VCH, Weinheim,
1999, vol. 2, p. 616, and references therein.
7 D. A. Knight, D. J. Cole-Hamilton and D. C. Cupertino, J. Chem.
Soc., Dalton Trans., 1990, 3051.
8 S. Berger, S. Braun and H.-O. Kalinowski, NMR-Spektro-
skopie von Nichtmetallen, Thieme Verlag, Stuttgart, 1993, vol. 3,
p. 145.
9 M. D. Curtis, S. Thanedar and W. M. Butler, Organometallics,
1984, 3, 1855.
Table 3 Selected bond lengths (Å) and angles (Њ) for 6 with estimated
standard deviations in parentheses
Pd–P
Pd–O
Pd–N
2.254(3)
2.122(6)
2.118(8)
Pd–C(11)
P–C(1)
O–C(2)
1.961(10)
1.744(9)
1.286(11)
P–Pd–O
P–Pd–C(11)
O–Pd–N
86.06(18)
99.6(3)
92.1(3)
N–Pd–C(11)
O–Pd–C(11)
N–Pd–P
82.5(4)
173.8(3)
177.36(18)
Table 4 Selected bond lengths (Å) and angles (Њ) for 8ؒTHFؒH2O with
estimated standard deviations in parentheses
Pd–P(1)
Pd–P(2)
Pd–O(1)
Pd–O(2)
2.260(2)
2.254(2)
2.097(4)
2.093(4)
P(1)–C(13)
C(13)–C(17)
C(17)–O(1)
1.757(7)
1.355(9)
1.316(8)
P(1)–Pd–O(1)
O(1)–Pd–O(2)
86.1(1)
89.1(2)
O(2)–Pd–P(2)
P(2)–Pd–P(1)
85.9(1)
98.84(6)
10 (a) W. Spaether, K. Klass, G. Erker, F. Zippel and R. Froehlich,
Chem. Eur. J., 1998, 4, 1411; (b) W. A. Howard, T. M. Trnka,
M. Waters and G. Parkin, J. Organomet. Chem., 1997, 528, 95;
(c) A. R. Siedle, R. A. Newmark, W. M. Lamanna and J. C.
Huffman, Organometallics, 1993, 12, 1491; (d) S. Gambarotta, S.
Strologo, C. Floriani, A. Chiesi-Villa and C. Guastini, Inorg. Chem.,
1985, 24, 654, and references therein.
11 F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen
and R. Taylor, J. Chem. Soc., Perkin Trans. 2, 1987, S1 and
references therein.
independent reflections (Rint = 0.038)) and were used for struc-
ture solution (direct methods) and refinement (full-matrix least
squares on F 2); R1[I > 2σ(I)] = 0.0383, wR2 (all data) = 0.0775.
The hydrogen atoms were calculated and fixed in idealized
positions. The absolute structure was determined with
x = Ϫ0.06(3). For all computations, the SHELXTL package
(PC)18 was used. Selected bond lengths and angles are given in
Table 3.
12 S. Y. Lee and R. G. Bergman, J. Am. Chem. Soc., 1996, 118,
6396.
13 R. G. Pearson, Inorg. Chem., 1973, 12, 712; J. A. Davies and
F. R. Hartley, Chem. Rev., 1981, 81, 79; J. Vicente, A. Arcas,
D. Bautista and P. G. Jones, Organometallics, 1997, 16, 2127.
14 (a) P. Braunstein, D. Matt, D. Nobel, S.-E. Bouaoud and D.
Grandjean, J. Organomet. Chem., 1986, 301, 401; (b) P. Braunstein,
D. Matt, Y. Dusausoy, J. Fischer, A. Mitschler and L. Ricard, J. Am.
Chem. Soc., 1981, 103, 5115; (c) S.-E. Bouaoud, P. Braunstein,
D. Grandjean, D. Matt and D. Nobel, J. Chem. Soc., Chem.
Commun., 1987, 488; (d) F. Balegroune, P. Braunstein, T. M. Gomes
Carneiro, D. Grandjean and D. Matt, J. Chem. Soc., Chem.
Commun., 1989, 582.
15 C. Mattheis and P. Braunstein, J. Organomet. Chem., 2001, 621, 218.
16 (a) S.-E. Bouaoud, P. Braunstein, D. Grandjean, D. Matt and
D. Nobel, Inorg. Chem., 1986, 25, 3765; (b) P. Braunstein, D. Matt,
D. Nobel, F. Balegroune, S.-E. Bouaoud, D. Grandjean and
J. Fischer, J. Chem. Soc., Dalton Trans., 1988, 353; (c) P. Braunstein,
L. Douce, J. Fischer, N. C. Craig, G. Goetz-Grandmont and
D. Matt, Inorg. Chim. Acta, 1992, 194, 151.
Crystal data for complex 8ؒTHFؒH2O. Colourless crystals
from THF–n-hexane. Kappa CCD diffractometer. Crystal data:
C44H48O2P2PdؒC4H8OؒH2O, M = 867.34, monoclinic, space
group P21, a = 10.4840(2), b = 34.0610(7), c = 12.9620(2) Å,
β = 112.670(3), U = 4271.1(3) Å3, T = 173 K, Z = 4, Dc = 1.35 g
cmϪ3, µ = 0.552 mmϪ1, 4817 data with I > 3σ(I), R = 0.029,
Rw = 0.035. The absolute structure was determined with
x = Ϫ0.01(3). For all computations, the Nonius MoLEN
package was used.19 Selected bond lengths and angles are given
in Table 4.
CCDC reference numbers 158617–158619.
See http://www.rsc.org/suppdata/dt/b0/b008227i/ for crystal-
lographic data in CIF or other electronic format.
Acknowledgements
17 S. D. Perera and B. L. Shaw, J. Organomet. Chem., 1991, 402, 133.
18 G. M. Sheldrick, SHELXTL-PC, release 4.4, Siemens Analytical
X-Ray Instruments, Madison, WI, 1992.
19 C. K. Fair, MoLEN, An Interactive Intelligent System for Crystal
Structure Analysis, Nonius, Delft, Netherlands, 1990.
We are grateful to a referee for valuable comments, the Minis-
tère de l’Enseignement Supérieur, de la Recherche et de la
Technologie for financial support, the A. von Humboldt
Foundation for a Feodor Lynen fellowship (to C. M.) and the
J. Chem. Soc., Dalton Trans., 2001, 800–805
805