Towards a library of Early-late Ti-Ru bimetallic complexes

Article abstract of DOI:10.1002/ejic.200401028

A series of new titanocene phosphanes 3-6 have been prepared by replacing both chloride atoms at the titanium atom of the complexes [TiCl ₂(η⁵-C₅H₅){η⁵- C₅H₄(CH₂)₂PR₂}] (1: R = Phi 2: R = Cy) by sodium fluoride or sodium benzoate in two-phase systems. Treatment of these new metalloligands with the binuclear complex [(p-cymene)RuCl₂]₂ affords the targeted titanocene difluoride and titanocene dibenzoate bimetallic ruthenium complexes 8-11. The first chiral Ti-Ru bimetallic complex 12 bearing a binaphthyloxy ligand at the titanium centre has been synthesised in this way. In each series, an X-ray crystal structure has been determined. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2005.

Full text of DOI:10.1002/ejic.200401028

FULL PAPER
Towards a Library of “Early-Late” Ti–Ru Bimetallic Complexes
Laurianne Bareille,[a] Pierre Le Gendre,*[a] Philippe Richard,[a] and Claude Moïse*[a]
Keywords: Heterometallic complexes / Metallocenes / Ruthenium / Titanium
A series of new titanocene phosphanes 3–6 have been pre- tallic ruthenium complexes 8–11. The first chiral Ti–Ru bime-
pared by replacing both chloride atoms at the titanium atom
of the complexes [TiCl2(η5-C5H5){η5-C5H4(CH2)2PR2}] (1: R =
Ph; 2: R = Cy) by sodium fluoride or sodium benzoate in two-
phase systems. Treatment of these new metalloligands with
the binuclear complex [(p-cymene)RuCl2]2 affords the tar-
geted titanocene difluoride and titanocene dibenzoate bime-
tallic complex 12 bearing a binaphthyloxy ligand at the tita-
nium centre has been synthesised in this way. In each series,
an X-ray crystal structure has been determined.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
zoate in a biphasic system (CHCl3/H2O).[9] Therefore we
carried out the reaction with the titanocene dichloride phos-
phanes 1 and 2 under similar conditions (Scheme 1). The
colour of the reaction mixture, initially red, turned slowly
to orange. After this treatment, titanocene dibenzoate phos-
phanes 3 and 4, respectively, were isolated as orange pow-
ders in good yields. The NMR spectroscopic data of 3 and
4 are very similar to their dichloride counterparts 1 and 2,
with additional resonances for the benzoate hydrogens. The
IR spectra exhibit two (COO) stretching bands around 1630
and 1320 cm–1, consistent with the monodentate coordina-
tion of the benzoate ligands.[10] Thus, in one step, both the
steric and electronic features of the titanocene phosphanes
1 and 2 have been modified. In an extension of this pro-
cedure, we conducted a reaction between sodium fluoride
and the titanocene dichloride phosphanes 1 and 2 under
similar conditions. The halide exchange proceeded
smoothly and led to the titanocene difluoride phosphanes
5, and 6, respectively, in very good yields. The 19F{1H}
NMR spectrum of 5 shows singlets at δ = 64.3 ppm in the
range reported for [Cp2TiF2].[11] The 19F{1H} NMR spec-
trum of 6 also contains a singlet but with a chemical shift
shifted slightly downfield at δ = 86.8 ppm. All other NMR
spectroscopic data of 5 and 6 are in accordance with their
structural features. At this point we should mention that
[Cp2TiF2] has recently been converted, under mild condi-
tions, into [Cp2TiH], which has proved to be a very effective
catalyst for the hydrosilylation of lactones[12] and the dehy-
dropolymerisation of silanes.[13] The potential of the bime-
tallic complexes constructed with the fluorinated titanocene
phosphane is thus reinforced.
Introduction
Complexes containing early as well as late transition met-
als are particularly attractive materials owing to their po-
tential application in homogeneous catalysis.[1–3] Indeed,
the coordination of the substrate to one of the two metals
may increase its reactivity towards the other. Such com-
pounds, which often show a cooperative effect, may consti-
tute a new class of catalysts that are able to improve the
efficiency of known processes or to give new reactions. Nev-
ertheless, probably because of their laborious access, only
a few early–late bimetallic complexes have been tested in
catalysis.[4] Our strategy to construct such systems was to
first synthesise a titanocene dichloride phosphane complex
and then to treat these early metal ligands with the binu-
clear complex [(p-cymene)RuCl2]2.[5] We obtained a series
of Ti–Ru heterobimetallic complexes in sufficient amounts
to carry out a screening of their catalytic behaviour.[6–8] An-
other aspect of interest of this approach is the structural
flexibility of the titanocene phosphane as regards a possible
tuning of the bimetallic complexes towards the target cata-
lytic reaction. Indeed, by replacing the two chloride atoms
on the titanium atom by other ligands it should be possible
to design new phosphanes and therefore to expand substan-
tially this class of bimetallic complexes. Here we report the
synthesis and the characterisation of a series of new Ti–Ru
bimetallic complexes by using this strategy.
Results and Discussion
Titanocene dibenzoate can be synthesised, in good yields,
from the reaction of titanocene dichloride with sodium ben-
These results prompted us to use the same strategy to
convert the titanocene dichloride phosphanes 1 and 2 into
optically active ligands. According to well-known pro-
cedures for the resolution of titanium metallocenes,[14–18] we
carried out the reaction of one equivalent of sodium bi-
[a] Laboratoire de Synthèse et Electrosynthèse Organométalliques,
LSEO-UMR 5188, Université de Bourgogne
Faculté des Sciences Gabriel, 6 bd Gabriel, 21000 Dijon, France
Fax: +33-3-8039-6880
E-mail: pierre.le-gendre@u-bourgogne.fr
Eur. J. Inorg. Chem. 2005, 2451–2456
DOI: 10.1002/ejic.200401028
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2451

L. Bareille, P. Le Gendre, P. Richard, C. Moïse
FULL PAPER
tra of 8 and 9 are similar to those reported for the free
titanocene phosphanes. The 19F{1H} NMR chemical shifts
of 10 and 11 show signals in the range of [Cp2TiF2] at δ
= 62.3 and 62.5 ppm, respectively. This suggests that the
dicyclohexylphosphane group is responsible for the de-
shielding of the signal in the 19F{1H}NMR spectrum of 6.
All other spectroscopic data of the bimetallic complexes 8–
11 are in accordance with their structural features.
Scheme 1.
naphtholate with 1 in THF at room temperature to afford
the desired titanocene binaphtholate phosphane 7. How-
ever, a careful study of the 1H NMR spectrum of the crude
product revealed that a partial loss of the Cp(CH2)2PPh2
moiety occurs in this reaction. At this step, purification by
chromatography is effective but led to a dramatic loss of
yield. Therefore we conducted this reaction under milder
conditions and used binaphthol directly along with imid-
azole as an HCl trap (Scheme 2). The reaction proceeded
cleanly without any side product formation and in very
Scheme 3.
We then carried out the reaction of the optically active
phosphane 7 with 0.5 molar equivalents of the binuclear
complex [(p-cymene)RuCl2]2 in CH2Cl2. Purification of the
crude product by flash chromatography on silica, in order
to remove an excess of [(p-cymene)RuCl2]2, allowed the iso-
lation of the first chiral Ti–Ru bimetallic complex 12. The
31P{1H} NMR spectrum of 12 shows a single peak at δ =
24.7 ppm, which is shifted downfield with respect to 7. The
1H NMR spectrum of 12 exhibits a marked anisochrony
not only for the substituted Cp, as for 7, but also for the p-
cymene fragment on the ruthenium centre − a chemical
shift gap of 0.25 ppm between both methyl protons of the
isopropyl group is observed. At this point, we should men-
tion that the synthesis and the characterisation of chiral
early–late bimetallic complexes are remarkably scarce in the
literature.[19,20]
1
good yield. The H NMR spectrum of 7 differs from that
of the precursor phosphane 1 in that it displays four signals
for the substituted Cp instead of two. This anisochrony is
indicative of a loss of symmetry due to the introduction of
binaphthol. The 31P{1H} NMR spectrum of 7 displays a
single peak at δ = –14.3 ppm, which is shifted slightly down-
field with respect to 1. Thus, starting from a readily com-
mercially available chiral diol, a pure optically active phos-
phane is available without the usual difficulties of chiral li-
gand synthesis (resolution, measurement of optical purity,
etc.). Finally, it is worth mentioning that all attempts to
convert 2 into an optically active phosphane by either the
sodium binaphtholate or imidazole routes were unsuccess-
ful.
Crystals of 8 suitable for X-ray measurements were ob-
tained by slow diffusion of pentane into a saturated dichlo-
romethane solution of the complex. The asymmetric unit
contains the bimetallic complex, one dichloromethane sol-
vate and one benzoic acid molecule (originally present as
an impurity in the sample). The structure of the complex
consists of two fragments: a titanocene moiety with a tetra-
hedral geometry and a ruthenium moiety with a usual
three-legged piano stool structure (Figure 1).[21,22] These
Scheme 2.
Once this series of new titanocene phosphanes had been two blocks are tethered together through a phosphanylethy-
prepared, we investigated their complexation with ruthe- lene bridge. Most of the structural features of the titano-
nium by following the procedure previously described for cene dibenzoate fragment within the bimetallic complex 8
the titanocene dichloride phosphanes 1 and 2.[5,6] Thus, the are similar to those reported for [Cp2Ti(O2CPh)2],[9,10] as
treatment of 3–6 with 0.5 molar equivalents of the binuclear the lengths of the C–O bonds to the coordinated and unco-
complex [(p-cymene)RuCl2]2 in benzene gave the targeted ordinated oxygen atoms are significantly different, the
early–late bimetallic complexes 8–11, respectively, in good angles at the coordinated oxygen atoms (Ti–O–C) are large
yields (Scheme 3). The 31P NMR chemical shifts of these and both benzoate ligands tilt with respect to the O(1)–Ti–
complexes, in the δ = 20 ppm range, correspond to a down- O(2) plane, with dihedral angles of 42.6(2)° and 29.9(2)°,
field shift of 30–40 ppm relative to the free ligands. As ex- respectively. Despite the fact that benzoate ligands are
pected, the two carboxylate stretching bands in the IR spec- rather bulkier than chloride atoms, the phosphorus atom
2452
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 2451–2456

Towards a Library of “Early-Late” Ti–Ru Bimetallic Complexes
FULL PAPER
remains located on the open side of the titanocene but not
exactly in the bisecting plane. This asymmetry is induced
by a π-stacking effect observed between one phenyl ring of
a benzoate ligand and one phenyl ring of a diphenylphos-
phanyl group. Finally, the cocrystallised benzoic acid mole-
cule forms a hydrogen bond with the oxygen atom O(2) of
one benzoate ligand.
Figure 2. Molecular structure of complex 10. The CDCl3 solvate
molecules have been omitted for clarity. Selected bond lengths [Å]
and angles [°] [data in square brackets correspond to the second
indepedent molecule]: Ti–Ct1 2.050(5) [2.055(5)], Ti–Ct2 2.065(5)
[2.056(5)], Ti–F1 1.891(3) [1.905(4)], Ti–F2 1.937(3) [1.896(4)], Ru–
Ct3 1.707(4) [1.709(4)], Ru–P 2.347(1) [2.344(1)], Ru–Cl1
2.4180(12) [2.4093(13)], Ru–Cl2 2.4126(13) [2.4200(11)]; Ct1–Ti–
Ct2 132.6(2) [136.2(2)], F1–Ti–F2 93.70(14) [95.7(2)], Ct3–Ru–P
131.0(1) [130.6(1)], Ct3–Ru–Cl1 126.1(1) [126.1(1)], Ct3–Ru–Cl2
127.7(1) [127.3(1)], P–Ru–Cl1 84.35(3) [83.47(3)], P–Ru–Cl2
86.92(4) [86.74(4)], Cl1–Ru–Cl2 86.80(4) [86.84(4)].
Figure 1. Molecular structure of complex 8. The CH2Cl2 solvate
molecule and the benzoic acid have been omitted for clarity. Se-
lected bond lengths [Å] and angles [°]: Ti–Ct1 2.049(3), Ti–Ct2
methylene carbon atom attached to the Cp ring] of 67°.
2.054(4), Ti–O1 1.947(2), Ti–O3 1.956(2), Ru–Ct3 1.707(3), Ru–P Thus, the areneruthenium moiety lies in an open area and
2.3516(8), Ru–Cl1 2.4182(9), Ru–Cl2 2.4211(8); Ct1–Ti–Ct2
not above the bulky binaphtholate group. The structural
132.4(2), O1–Ti–O3 90.8(1), Ct3–Ru–P 130.6(1), Ct3–Ru–Cl1
parameters of the chelate ring formed are similar to those
127.3(1), Ct3–Ru–Cl2 124.7(1), P–Ru–Cl1 85.25(3), P–Ru–Cl2
described in the literature.[14–17] The bite angle O(1)–Ti–
85.89(3), Cl(1)–Ru–Cl(2), 89.16(3) Ti–O(1)–C(35) 152.9(2), Ti–O3–
O(2) is equal to 92.0(1)° and the dihedral angle between
the naphthalene planes [64.8(1)°] is contracted from that of
cocrystallised “free” binaphthol (90° is generally observed).
It is noteworthy that two different Ti–O bond lengths are
observed [Ti–O(1) = 1.915(2) Å and Ti–O(2) = 1.981(2) Å].
This can be explained by the presence of a hydrogen bond
pointing from a hydroxide group of the “free” binaphthol
molecule toward the O(2) atom [O(3)–H···O(2) = 1.75 Å].
In fact, this binaphthol molecule plays an active role in the
crystallisation process in connecting two adjacent mole-
cules: the second hydroxy group points between the two
chlorine atoms of the Ru moiety of an adjacent complex
[O(4)–H···Cl(1) = 2.35 Å and O(4)–H···Cl(2) = 2.70 Å].[25]
C42 142.4(2).
Crystals of 10 suitable for an X-ray study were also ob-
tained by layering techniques with CDCl3/hexane (the sam-
ple came from a NMR solution). The asymmetric unit con-
tains two independent complexes of 10 and six CDCl3 sol-
vate molecules. The two molecules exhibit an almost iden-
tical conformation with opposite orientations of the p-cy-
mene ligand. The molecular structure of complex 10, repre-
sented in Figure 2, is very similar to the one with two chlo-
rides.[5] The diphenylphosphanyl group is on the open side
of the titanocene, near a bisecting position, with both phen-
yls rings directed toward the titanium atom and the ruthe-
nium away from it in the least congested environment. As
expected, the Ti–F bond lengths [ϽTi–FϾ = 1.91(2) Å,
mean over four distances] are shorter than the Ti–Cl bond
lengths [2.35(3) Å] and the F–Ti–F angle [95(1)°, mean over
two angles] similar to the one with chlorides [Cl–Ti–Cl =
Conclusions
This study provides a straightforward access to a series
95.26(8)°]. These structural parameters are close to those of new early–late transition metal complexes. The substitu-
observed for other titanocene difluoride derivatives.[23,24]
tion of both chloride atoms of a titanocene phosphane by
Recrystallisation attempts of 12, starting from a racemic other ligands such as benzoate, fluoride or binaphtholate is
mixture, proved to be difficult, but finally succeeded by slow the key step of this new route. The phosphanes so formed
evaporation under air of a CH2Cl2/pentane solution of the react readily with the binuclear complex [(p-cymene)-
crude product. Its molecular structure is shown in Figure 3. RuCl2]2 to give the targeted bimetallic complexes. The
The asymmetric unit contains the bimetallic Ti–Ru com- structures of three of them have been confirmed by X-ray
pound, one additional binaphthol molecule and 2.5 dichlo- diffraction studies. Assessment of the catalytic behaviour of
romethane solvates. Contrary to the bimetallic complexes 8 these early–late bimetallic complexes is currently being
and 10, the PPh2 substituent is rotated from the bisecting studied in our laboratory. The design of new titanocene
position, leading to a dihedral angle between the Ct(1)–Ti– phosphanes inspired by the fruitful chemistry of [Cp2TiCl2]
Ct(2) plane and the Ct(1)–C(11)–Ct(2) one [C(11) being the is also in progress.
Eur. J. Inorg. Chem. 2005, 2451–2456
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2453

L. Bareille, P. Le Gendre, P. Richard, C. Moïse
FULL PAPER
(s, PPh2) ppm. C38H33O4PTi (632.51): calcd. C 72.16, H 5.26; found
C 72.00, H 5.02.
[Ti(O2CPh)2(η5-C5H5){η5-C5H4(CH2)2PCy2}] (4): This compound
was obtained following the above procedure but with [TiCl2(η5-
C5H5){η5-C5H4(CH2)2PCy2}] (2) and toluene/H2O as the biphasic
system (0.38 g, 62% yield). IR (KBr): ν(COO) = 1635, 1326 cm–1.
1H NMR (C6D6): δ = 1.10–1.90 (m, 24 H, Cy + CH2), 2.85 (pseudo
3
2
3
4
q, JH,H = JH,P = 8 Hz, 2 H, CH2), 6.14 (pseudo t, JH,H = JH,H
3
= 3 Hz, 2 H, C5H4), 6.28 (s, 5 H, C5H5), 6.41 (pseudo t, JH,H
=
4JH,H = 3 Hz, 2 H, C5H4), 7.35 (t, JH,H = 8 Hz, 2 H, Ph), 7.39
3
3
3
(pseudo t, JH,H = 8 Hz, 4 H, Ph), 8.54 (d, JH,H = 8 Hz, 4 H, Ph)
ppm. 31P{1H} NMR (C6D6):
δ
=
–3.79 (s, PCy2) ppm.
C38H45O4PTi (644.60): calcd. C 70.80, H 7.04; found C 70.54, H
7.16.
[TiF2(η5-C5H5){η5-C5H4(CH2)2PPh2}] (5): A 100-mL flask was
charged with NaF (0.065 g, 1.54 mmol), [TiCl2(η5-C5H5){η5-
C5H4(CH2)2PPh2}] (1; 0.36 g, 0.78 mmol), 4 mL of degassed water
and 36 mL of CHCl3. The mixture was stirred for 1.5 h during
which time the colour turned to deep yellow. Then, the organic
phase was removed, washed with 5 mL of water, dried with MgSO4,
filtered and the solvents evaporated to dryness to yield a yellow
1
3
powder (0.16 g, 50% yield). H NMR (CDCl3): δ = 2.29 (t, JH,H
3
= 8 Hz, 2 H, CH2), 2.65 (pseudo q, JH,H
=
2JH,P = 8 Hz, 2 H,
CH2), 6.05 (pseudo s, 2 H, C5H4), 6.38 (m, 7 H, C5H5 + 2 H C5H4),
7.25–7.35 (m, 6 H, PPh2), 7.37–7.44 (m, 4 H, PPh2) ppm. 19F NMR
(CDCl3): δ = 64.28 (s, TiF2) ppm. 31P{1H} NMR (CDCl3): δ =
–14.99 (s, PPh2) ppm. C24H23F2PTi (428.28): calcd. C 67.31, H
5.41; found C 66.82, H 5.26.
Figure 3. Molecular structure of complex rac-12. Only one enanti-
omer is shown for clarity. The free binaphthol molecule and the
CH2Cl2 solvate molecules are also not shown. Selected bond
lengths [Å] and angles [°]: Ti–Ct1 2.099(3), Ti–Ct2 2.088(4), Ti–O1
1.915(2), Ti–O2 1.981(2), Ru–Ct3 1.703(3), Ru–P 2.3442(8), Ru–
Cl1 2.4210(7), Ru–Cl2 2.4157(8); Ct1–Ti–Ct2 130.48(13), O1–Ti–
O2 91.96(9), Ct3–Ru–P 130.95(9), Ct3–Ru–Cl1 126.04(9), Ct3–Ru–
Cl2 126.42(9), P–Ru–Cl1 84.97(3), P–Ru–Cl2 86.80(3), Cl1–Ru–Cl2
87.22(3).
[TiF2(η5-C5H5){η5-C5H4(CH2)2PCy2}] (6): This compound was ob-
tained following the above procedure but with [TiCl2(η5-C5H5){η5-
C5H4(CH2)2PCy2}] (2) and toluene/H2O as the biphasic system
(0.17 g, 50% yield). 1H NMR (C6D6): δ = 1.10–1.27 (m, 12 H, Cy),
3
1.55–1.84 (m, 12 H, Cy+CH2), 2.87 (pseudo q, JH,H
= =
2JH,P
8 Hz, 2 H, CH2), 5.84 (pseudo s, 4 H, C5H4), 5.99 (s, 5 H, C5H5)
ppm. 19F NMR (C6D6): δ = 86.76 (s, TiF2) ppm. 31P{1H} NMR
(C6D6): δ = –3.41 (s, PCy2) ppm. C24H35F2PTi (440.37): calcd. C
65.46, H 8.01; found C 65.15, H 8.22.
Experimental Section
General Remarks: All reactions were carried out under an atmo-
sphere of purified argon. The solvents and eluents were dried by
the appropriate procedures and distilled under argon immediately
before use. Standard Schlenk techniques and conventional glass
vessels were employed. Elemental analyses were carried out with a
EA 1108 CHNS-O FISONS Instruments. 1H (500 MHz), 19F
(282 MHz) and 31P{1H} (202 MHz) spectra were collected on a
Bruker 500 MHz Avance DRX spectrometer. Chemical shifts are
quoted relative to internal TMS (1H), external CFCl3 (19F) or ex-
ternal H3PO4 (31P). IR spectra were obtained with a Bruker IFS
66v spectrometer. The titanocene phosphanes 1 and 2 were synthe-
sised as reported previously.[5,6]
[Ti{(R)-Binaphtholate)}(η5-C5H5){η5-C5H4(CH2)2PPh2}] (7): Imid-
azole (0.14 g, 2.15 mmol) in 10 mL of dichloromethane and (R)-
binaphthol (0.32 g, 1.13 mmol) in 25 mL of dichloromethane were
successively added to a solution of [TiCl2(η5-C5H5){η5-C5H4-
(CH2)2PPh2}] (1; 0.52 g, 1.13 mmol) in 25 mL of dichloromethane
at room temperature. The mixture was stirred for 4 h. After fil-
tration, the solvent was removed from the filtrate under reduced
pressure and 40 mL of toluene was added to the resulting product.
Filtration through celite and evaporation of the solvent afforded 7
(0.5 g, 70% yield) as a red solid. 1H NMR (CDCl3, 500 MHz): δ =
1.65–1.75 (m, 2 H, CH2), 2.20–2.35 (m, 2 H, CH2), 5.28–5.35 (m,
1 H, C5H4), 5.85–5.95 (m, 1 H, C5H4), 5.98–6.04 (m, 1 H, C5H4),
6.09 (s, 5 H, C5H5), 6.60–6.66 (m, 1 H, C5H4), 7.15–8.15 (m, 22 H,
phenyl-H and binaphthyl-H) ppm. 31P{1H} NMR (CDCl3,
202 MHz): δ = –14.3 (s, PPh2) ppm. C44H35PTiO2 (674.59): calcd.
C 78.34, H 5.23; found C 78.48, H 5.29. [α]2D0 = –2163 (c = 0.1,
CH2Cl2).
[Ti(O2CPh)2(η5-C5H5){η5-C5H4(CH2)2PPh2}] (3): A 100-mL flask
was charged with PhCO2Na (0.3 g, 2 mmol), [TiCl2(η5-C5H5){η5-
C5H4(CH2)2PPh2}] (1; 0.4 g, 1 mmol), 5 mL of degassed water and
44 mL of CHCl3. The mixture was stirred for 1.5 h during which
time the colour turned to orange. Then, the organic phase was
removed, washed with 5 mL of water, dried with MgSO4, filtered
and the solvents evaporated to dryness to yield an orange powder
[(p-Cymene)RuCl2{(η5-C5H5)[μ-η5:η1-C5H4(CH2)2PPh2]Ti(O2CPh)2}]
(0.45 g, 75% yield). IR (KBr): ν(COO) = 1634, 1324 cm–1. 1H (8): A 25-mL Schlenk flask was charged under argon with 3 (0.45 g,
NMR (CDCl3): δ = 2.17 (t, 3JH,H = 8 Hz, 2 H, CH2), 2.61 (pseudo
0.71 mmol), [(p-cymene)RuCl2]2 (0.16 g, 0.27 mmol) and degassed
benzene. The mixture was stirred at room temperature for 4 h, dur-
ing which time a brick-red precipitate slowly formed. The solvent
3
2
3
4
q, JH,H = JH,P = 8 Hz, 2 H, CH2), 6.42 (pseudo t, JH,H = JH,H
3
= 3 Hz, 2 H, C5H4), 6.59 (s, 5 H, C5H5), 6.61 (pseudo t, JH,H
=
4JH,H = 3 Hz, 2 H, C5H4), 7.16–7.20 (m, 10 H, PPh2), 7.44 (pseudo was removed by filtration and the red residue was dried under vac-
uum (0.38 g, 75% yield). IR (KBr): ν(COO) = 1636, 1355 cm–1. 1H
3
3
t, JH,H = 7 Hz, 4 H, Ph), 7.54 (t, JH,H = 7 Hz, 2 H, Ph), 8.03 (d,
3
3JH,H = 7 Hz, 4 H, Ph) ppm. 31P{1H} NMR (CDCl3): δ = –14.87
NMR (500 MHz, CDCl3): δ = 0.74 [d, JH,H = 7 Hz, 6 H, CH-
2454
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 2451–2456

Towards a Library of “Early-Late” Ti–Ru Bimetallic Complexes
FULL PAPER
3
(CH3)2], 1.81 (s, 3 H, CH3), 2.17 (m, 2 H, CH2), 2.45 [sept, JH,H 6.15 (m, 1 H, C5H4), 6.20 (s, 5 H, C5H5), 6.49–6.57 (m, 1 H, C5H4),
3
= 7 Hz, 1 H, CH(CH3)2], 2.63 (m, 2 H, CH2), 5.01 (d, JH,H
=
6.80–8.00 (m, 22 H, Ph and binaphthyl) ppm. 31P{1H} NMR
24.66 (s) ppm. C54H49Cl2O2PRuTi
(980.78): calcd. C 66.13, H 5.04; found C 66.60, H 5.10. [α]2D0
3
6 Hz, 2 H, p-cymene), 5.18 (d, JH,H = 6 Hz, 2 H, p-cymene), 6.14 (CDCl3, 202 MHz):
δ =
(pseudo t, 3JH,H = 4JH,H = 2 Hz, 2 H, C5H4), 6.48 (pseudo t, 3JH,H
=
4
= JH,H = 2 Hz, 2 H, C5H4), 6.49 (s, 5 H, C5H5), 7.19 (pseudo t, –1920 (c = 0.1, CH2Cl2).
3
3JH,H = 6 Hz, 4 H, PPh2), 7.25–7.35 (m, 6 H, PPh2), 7.49 (t, JH,H
X-ray Crystallographic Study of 8: C48H47Cl2O4PRuTi·
C7H6O2·0.5(CH2Cl2), MW = 1103.28, monoclinic, space group C2/
c, a = 25.0136(3), b = 18.9231(2), c = 21.2279(3) Å, β = 98.595(1)°,
V = 9935.1 (2) Å3, Z = 8, Dcalc = 1.475 gcm–3; F(000) = 4536. The
structure was solved by the heavy-atom method using
SHELXS97.[26] Refinement, based on F2, was carried out by full-
matrix least-squares with the SHELXL97 and WINGX pro-
3
= 8 Hz, 2 H, Ph), 7.56 (pseudo t, JH,H = 8 Hz, 4 H, Ph), 7.82 (d,
3JH,H = 8 Hz, 4 H, Ph) ppm. 31P{1H} NMR (202 MHz, CDCl3): δ
= 24.99 (s, PPh2) ppm. C48H47Cl2O4PRuTi (938.70): calcd. C 61.42,
H 5.05; found C 60.92, H 5.21.
[(p-Cymene)RuCl2{(η5-C5H5)[μ-η5:η1-C5H4(CH2)2PCy2]Ti(O2CPh)2}]
(9): This compound was obtained following the above procedure
but with 4 (0.41 g, 80% yield). IR (KBr): ν(COO) = 1635, grams.[26,27] Non-hydrogen atoms were refined anisotropically. H
1328 cm–1. 1H NMR (CDCl3): δ = 0.85–2.17 (m, 24 H, Cy + CH2),
atoms were included in calculated positions and included in the
δ = 1.24 [d, 3JH,H = 7 Hz, 6 H, CH(CH3)2], 2.02 (s, 3 H, CH3), 2.60 refinement with a riding-motion model with Uiso = 1.2Ueq of the
3
(m, 2 H, CH2), 2.75 [sept, JH,H = 7 Hz, 1 H, CH(CH3)2], 5.01 carrier atom (1.5 for methyl groups and OH). Convergence was
3
4
(pseudo s,, 4 H, p-cymene), 6.37 (pseudo t, JH,H = JH,H = 3 Hz,
reached at wR2 = 0.125 for all data (11386 intensities), R1 = 0.047
for 7904 intensities with I Ͼ 2σ(I) and S = 1.044 for 615 parame-
ters. The residual electron density in the final difference Fourier
map was 1.12 and –1.18 eÅ–3.
2 H, C5H4), 6.55 (pseudo t, 3JH,H = 4JH,H = 3 Hz, 2 H, C5H4), 6.60
3
(s, 5 H, C5H5), 7.46 (pseudo t, JH,H = 8 Hz, 4 H, Ph), 7.50 (t,
3
3JH,H = 8 Hz, 2 H, Ph), 8.07 (d, JH,H = 8 Hz, 4 H, Ph) ppm.
31P{1H} NMR (CDCl3): δ = 25.84 (s, PCy2) ppm. C48H59Cl2O4P-
X-ray Crystallographic Study of 10: C34H37Cl2F2PRuTi·3(CHCl3),
RuTi (950.80): calcd. C 60.23, H 6.25; found C 60.43, H 6.55.
¯
MW = 1092.6, triclinic, space group P1; a = 9.716(5), b =
[(p-Cymene)RuCl2{(η5-C5H5)[μ-η5:η1-C5H4(CH2)2PPh2]TiF2}] (10): 17.008(5), c = 28.927(5) Å, α = 75.609(5)°, β = 82.660(5)°, γ =
This compound was obtained following the procedure described
81.094(5)°, V = 4554 (3) Å3, Z = 4, Dcalc = 1.593 gcm–3; F(000) =
2192. The structure was solved by the heavy-atom method using
SHELXS97.[26] Refinement, based on F2, was carried out by full-
for the synthesis of 8 but with 5 (0.33 g, 83% yield). 1H NMR
3
(500 MHz, CDCl3): δ = 0.80 [d, JH,H = 7 Hz, 6 H, CH(CH3)2],
3
1.90 (s, 3 H, CH3), 2.25 (m, 2 H, CH2), 2.52 [sept, JH,H = 7 Hz, 1 matrix least-squares with the SHELXL97 and WINGX pro-
3
H, CH(CH3)2], 2.74 (m, 2 H, CH2), 5.09 (d, JH,H = 8 Hz, 2 H, p- grams.[26,27] Non-hydrogen atoms were refined anisotropically. H
3
cymene), 5.28 (d, JH,H = 8 Hz, 2 H, p-cymene), 5.93 (pseudo s, 2
atoms were included in calculated positions and included in the
H, C5H4), 6.25 (pseudo s, 2 H, C5H4), 6.36 (s, 5 H, C5H5), 7.48 (m, refinement with a riding-motion model with Uiso = 1.2Ueq of the
6 H, PPh2), 7.90 (m, 4 H, PPh2) ppm. 19F NMR (CDCl3): δ =
62.29 (s, TiF2) ppm. 31P{1H} NMR (202 MHz, CDCl3): δ = 24.57
carrier atom (1.5 for methyl groups). Two independent molecules
of 10 are present in the asymmetric unit and one of them has a
(s, PPh2) ppm. C34H37Cl2F2PRuTi (734.47): calcd. C 55.60, H 5.08; disordered cyclopentadienyl ring over two positions with occupanc-
found C 55.10, H 4.90.
ies refined to 0.53:0.47. Among the six chloroform solvates present
in the asymmetric unit, three are disordered over two positions with
occupancies refined to 0.59:0.41, 0.90:0.10 and 0.90:0.10. The chlo-
rine atoms with occupancies equal to 0.10 were isotropically refined
Convergence was reached at wR2 = 0.119 for all data (18928 inten-
sities), R1 = 0.051 for 15240 intensities with I Ͼ 2σ(I) and S =
1.039 for 1047 parameters. The residual electron density in the final
difference Fourier map was 1.25 and –1.03 eÅ–3.
[(p-Cymene)RuCl2{(η5-C5H5)[μ-η5:η1-C5H4(CH2)2PCy2]TiF2}] (11):
This compound was obtained following the procedure described
for the synthesis of 8 but with 6 (0.24 g, 60% yield). 1H NMR
3
(CDCl3): δ = 1.10–2.31 (m, 24 H, Cy + CH2), δ = 1.24 [d, JH,H
=
7 Hz, 6 H, CH(CH3)2], 2.11 (s, 3 H, CH3), 2.65 (m, 2 H, CH2),
3
2.85 [sept, JH,H = 7 Hz, 1 H, CH(CH3)2], 5.58 (pseudo s, 4 H, p-
3
4
cymene), 6.14 (pseudo t, JH,H = JH,H = 3 Hz, 2 H, C5H4), 6.31
3
4
(pseudo t, JH,H = JH,H = 3 Hz, 2 H, C5H4), 6.42 (s, 5 H, C5H5) X-ray
Crystallographic
Study
of
12:
C54H49Cl2O2-
ppm. 19F NMR (CDCl3): δ = 62.54 (s, TiF2) ppm. 31P{1H} NMR PRuTi·C20H14O2·2.5(CH2Cl2), MW = 1479.4, triclinic, space group
¯
(CDCl3): δ = 25.88 (s, PCy2) ppm. C34H49Cl2F2PRuTi (746.57):
P1; a = 12.8350(4), b = 15.5480(5), c = 19.4240(6) Å, α =
calcd. C 54.70, H 6.61; found C 54.55, H 6.41.
112.361(2)°, β = 97.889(2)°, γ = 105.773(2)°, V = 3321.2 (2) Å3, Z
= 2, Dcalc = 1.479 gcm–3; F(000) = 1518. The structure was solved
by the heavy-atom method using SHELXS97.[26] Refinement, based
on F2, was carried out by full-matrix least-squares with the
SHELXL97 and WINGX programs.[26,27] Non-hydrogen atoms
were refined anisotropically. H atoms were included in calculated
positions and included in the refinement with a riding-motion
model with Uiso = 1.3Ueq of the carrier atom. Two of the three
dichloromethane solvates are disordered: one, located close to an
inversion centre, was refined with an occupation factor of 0.5, the
other is disordered over two positions with occupancies refined to
0.77:0.23. Convergence was reached at wR2 = 0.129 for all data, R1
= 0.049 for 12707 intensities with I Ͼ 2σ(I) and S = 1.053 for
841 parameters. The residual electron density in the final difference
Fourier map was 1.8 and –1.7 eÅ–3 close to a disordered dichloro-
methane solvate.
[(p-Cymene)RuCl2{(η5-C5H5)[μ-η5:η1-C5H4(CH2)2PPh2]Ti[(R)-bi-
naphtholate)]}] (12): [(p-cymene)RuCl2]2 (0.041 g, 0.07 mmol) in
5 mL of dichloromethane was added to a solution of 7 (0.1 g,
0.15 mmol) in 10 mL of dichloromethane at 0 °C. The mixture was
stirred at room temperature for 4 h. The solvent was then removed
in vacuo. Purification by flash chromatography on silica (toluene/
THF, 50:50) under argon and at 10 °C afforded 12 as an orange
solid (0.070 g, 50% yield). Complex 12 in its racemic form, which
was used to obtain suitable crystal for X-ray diffraction, was pre-
1
pared in a similar way. H NMR (CDCl3, 500 MHz): δ = 0.70 [d,
3
3JH,H = 7 Hz, 3 H, CH(CH3)2], 0.95 [d, JH,H = 7 Hz, 3 H,
CH(CH3)2], 1.67–1.80 (m, 1 H, CH2), 1.87 (s, 3 H, CH3), 2.00–2.19
(m, 1 H, CH2), 2.45–2.65 (m, 1 H, CH2), 2.54 [sept, JH,H = 7 Hz,
3
1 H, CH(CH3)2], 2.74–2.96 (m, 1 H, CH2), 4.76–4.81 (m, 1 H,
3
3
C5H4), 4.88 (d, JH,H = 6 Hz, 1 H, p-cymene), 5.19 (d, JH,H
=
3
6 Hz, 1 H, p-cymene), 5.28 (d, JH,H = 6 Hz, 1 H, p-cymene), 5.39 CCDC-256921 (for 8), -256922 (for 10) and -225303 (for 12) con-
3
(d, JH,H = 6 Hz, 1 H, p-cymene), 5.97–6.04 (m, 1 H, C5H4), 6.09–
www.eurjic.org
tain the supplementary crystallographic data for this paper. These
Eur. J. Inorg. Chem. 2005, 2451–2456
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2455

L. Bareille, P. Le Gendre, P. Richard, C. Moïse
FULL PAPER
[14] F. R. W. P. Wild, L. Zsolnai, G. Huttner, H. H. Brintzinger, J.
Organomet. Chem. 1982, 232, 233–247.
[15] M. E. Huttenloch, B. Dorer, U. Rief, M. H. Prosenc, K.
Schmidt, H. H. Brintzinger, J. Organomet. Chem. 1997, 541,
219–232.
[16] M. S. Erickson, F. R. Fronczek, M. L. McLaughlin, J. Or-
ganomet. Chem. 1991, 415, 75–85.
[17] B. A. Kuntz, R. Ramachandran, N. J. Taylor, J. Guan, S. Col-
lins, J. Organomet. Chem. 1995, 497, 133–142.
[18] S. Collins, B. A. Kuntz, Y. Hong, J. Org. Chem. 1989, 54, 4154–
4158.
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[1]
[2]
[3]
R. Choukroun, D. Gervais, P. Kalck, F. Senocq, J. Organomet.
Chem. 1987, 335, C9–C12.
A. M. Trzeciak, J. J. Ziolkowski, R. Choukroun, J. Mol. Catal.
1996, 110, 135–139.
B. E. Bosch, I. Brümmer, K. Kunz, G. Erker, R. Fröhlich, S.
Kotila, Organometallics 2000, 19, 1255–1261.
N. Wheatley, P. Kalck, Chem. Rev. 1999, 99, 3379–3419.
P. Le Gendre, P. Richard, C. Moïse, J. Organomet. Chem. 2000,
605, 151–156.
[4]
[5]
[19] A. Kless, C. Lefeber, A. Spannenberg, R. Kempe, W. Baum-
ann, J. Holz, A. Börner, Tetrahedron 1996, 52, 14599–14606.
[20] M. Quirmbach, A. Kless, J. Holz, V. Tararov, A. Börner, Tetra-
hedron: Asymmetry 1999, 10, 1803–1811.
[6]
P. Le Gendre, M. Picquet, P. Richard, C. Moïse, J. Organomet.
Chem. 2002, 643–644, 231–236.
[21] J. F. Mai, Y. Yamamoto, J. Organomet. Chem. 1998, 560, 223–
[7] P. Le Gendre, V. Comte, A. Michelot, C. Moïse, Inorg. Chim.
Acta 2003, 350, 289–292.
[8] J. Goux, P. Le Gendre, P. Richard, C. Moïse, J. Organomet.
Chem. 2004, in press.
[9] K. Döppert, H. P. Klein, U. Thewalt, J. Organomet. Chem.
1986, 303, 205–211.
[10] D. M. Hoffman, N. D. Chester, R. C. Fay, Organometallics
1983, 2, 48–52.
[11] A. Herzog, F. Q. Liu, H. W. Roesky, A. Demsar, K. Keller, M.
Noltemeyer, F. Pauer, Organometallics 1994, 13, 1251–1256.
[12] X. Verdaguer, M. C. Hansen, S. C. Berk, S. L. Buchwald, J.
Org. Chem. 1997, 62, 8522–8528.
[13] F. Lunzer, C. Marschner, S. Landgraf, J. Organomet. Chem.
1998, 568, 253–255.
232.
[22] J. F. Mai, Y. Yamamoto, J. Organomet. Chem. 1997, 545, 577–
579.
[23] C. H. Winter, X. X. Zhou, M. J. Heeg, Inorg. Chem. 1992, 31,
1808–1815.
[24] A. J. Edwards, N. J. Burke, C. M. Dobson, K. Prout, S. J.
Heyes, J. Am. Chem. Soc. 1995, 117, 4637–4653.
[25] G. Aullon, D. Bellamy, L. Brammer, E. A. Bruton, A. G.
Orpen, Chem. Commun. 1998, 653–654.
[26] G. M.
Sheldrick,
SHELX97
(includes
SHELXS97,
SHELXL97); Programs for Crystal Structure Analysis (Release
97-2); University of Göttingen, Göttingen, Germany, 1998.
[27] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
Received: December 14, 2004
2456
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 2451–2456