Ruthenium titanocene and ruthenium titanium half-sandwich bimetallic complexes in catalytic cyclopropanation

Article abstract of DOI:10.1016/j.jorganchem.2004.09.021

The reaction of the phosphine functionalised titanium half-sandwich complexes 7, 9 and 10 with the binuclear complex [(p-cymene)RuCl ₂]₂ allowed the access to three new early-late bimetallic complexes (p-cymene)[(μ-η⁵:η

Full text of DOI:10.1016/j.jorganchem.2004.09.021

Journal of Organometallic Chemistry 690 (2005) 301–306
www.elsevier.com/locate/jorganchem
Ruthenium titanocene and ruthenium titanium
half-sandwich bimetallic complexes in catalytic cyclopropanation
*
*
´ ˆ
Jerome Goux, Pierre Le Gendre , Philippe Richard, Claude Mo¨ıse
` ` ´ ´ ´
Laboratoire de Synthese et Electrosynthese Organometalliques, LSEO-UMR 5188, Faculte des Sciences Gabriel, Universite de Bourgogne,
6 bd Gabriel, 21000 Dijon, France
Received 17 May 2004; accepted 13 September 2004
Available online 29 December 2004
Abstract
The reaction of the phosphine functionalised titanium half-sandwich complexes 7, 9 and 10 with the binuclear complex [(p-cym-
ene)RuCl2]2 allowed the access to three new early-late bimetallic complexes (p-cymene)[(l-g5:g1-C5H4(CH2)nPR2)TiX3]RuCl2 (11–
13). The structure of 11 (n = 0, X = Cl) has been confirmed by X-ray diffraction. The ruthenium titanium half-sandwich bimetallic
complexes so formed and the ruthenium titanocene analogues 4–6 catalyse the addition of ethyl diazoacetate to styrene with high
selectivity toward cyclopropanation versus metathesis contrary to the monometallic complexes (p-cymene)RuCl2PR3.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Titanium; Ruthenium; Heterobimetallic complexes; Cyclopropanation
1. Introduction
titanium atom with a stronger Lewis acid character
and a less crowded environment should be more able
to interact with both ruthenium and substrate. Continu-
ing our screening of catalytic reactions we focused then
our work on cyclopropanation and we compared the
activity of the bimetallic complexes with the monometal-
lic species (p-cymene)RuCl2PR3.
Early-late heterobimetallic complexes represent
promising catalysts owing to the potential cooperative
reactivity of widely different metals [1]. However, their
synthesis often constitute a challenge and therefore a
screening of their catalytic behaviour is not possible.
To start with the design of such systems, we prepared
early-late complexes which combined two well known
fragments in catalysis field, a titanocene and an
(arene)ruthenium moieties [2]. The bimetallic complexes
4–6 were obtained easily via the reaction of the titano-
cene phosphines 1–3 with the binuclear complex [(p-
cymene)RuCl2]2 and have already proved to be active
catalyst in ring closing metathesis [3] and in enol esters
synthesis (Scheme 1) [4].
2. Results and discussion
By using a strategy which has proved to be effective in
the ruthenium titanocene bimetallic series [2,3], we first
concentrate our efforts in the preparation of phosphine
functionalised titanium half-sandwich complexes. The
complex (g5-C5H4PPh2)TiCl3 (7) was prepared accord-
ing to an earlier method involving the synthesis of the
trimethylsilyl cyclopentadiene derivative C5H4(Si-
Me3)PPh2, followed by treatment with TiCl4 [5]. In or-
der to expand our choice of ligands we attempted to
replace the chloride atoms at the titanium atom in 7
Here, we report the synthesis and the characterisation
of a complementary series of ruthenium titanium
half-sandwich bimetallic complexes. We feel that the
*
Corresponding author.
0022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.09.021

302
J. Goux et al. / Journal of Organometallic Chemistry 690 (2005) 301–306
Ru
n
n
PR2
P
R2
Cl
Cl
1/2 [(p-cymene)RuCl2]2
benzene, 4 h, r.t.
Cl
Cl
Cl
Cl
Ti
Ti
1: n = 0, R = Ph
2: n = 2, R = Ph
3: n = 2, R = Cy
4
5
6
Scheme 1.
by alkoxy groups. We just began by refluxing 7 in meth-
anol [6], but it resulted in products of decomposition.
This preliminary result and the weak stability of 7
prompted us to investigate a second route previously
described by Rausch and co-workers [7]. Thus, the reac-
tion of the lithium diphenylphosphino cyclopentadie-
nide with ClTi(OiPr)3 allowed us the access to the
triisopropoxy titanium half-sandwich complex (g5-
C5H4PPh2)Ti(OiPr)3 (8). In an extension of this proce-
dure, we conducted a reaction between the lithium
[(diphenylphosphino)ethyl] or [(dicyclohexylphosph-
ino)ethyl] cyclopentadienides with ClTi(OiPr)3 and
obtained the desired complexes [g5-C5H4(CH2)2-
PPh2]Ti(OiPr)3 (9) and [g5-C5H4(CH2)2PCy2]Ti(OiPr)3
(10) in good yields. Both 1H NMR spectra are very sim-
ilar to the spectrum of 8. All three show doublet and
heptet resonances for the three isopropoxy groups and
also two pseudotriplet resonances for the protons of
the cyclopentadienyl ring. The 31P NMR spectra of 9
and 10 appear as singlets with chemical shifts at ꢀ18.1
and ꢀ3.6 ppm, respectively.
The 31P NMR chemical shift of the complex 11 at
23.5 ppm corresponds to a downfield shift of 31.5 ppm
1
relatively to the free ligand. The H NMR spectrum of
11 exhibits all the signals relative to both metallic frag-
ments. Single crystal of 11, suitable for X-ray diffraction
was obtained by layering technique. ORTEP view of 11
is reported in Fig. 1 and selected bond lengths and an-
Cl(4)
Cl(5)
CT(2)
Ru
CT(1)
P
C(1)
Ti
Once a series of phosphine functionalised titanium
half-sandwich complexes with complementary steric
and electronic features has been prepared, we investi-
gated their complexation with ruthenium. Thus, the
reaction of 7 with 0.5 molar equivalent of the binuclear
complex [(p-cymene)RuCl2]2 gave the targeted early-late
bimetallic complex 11 (Scheme 2).
Cl(1)
Cl(3)
Cl(2)
Fig. 1. ORTEP view of the compound 11 with thermal ellipsoids
drawn at the 50% probability level. Hydrogen atoms and solvate
molecules are omitted for clarity.
PR2
Ru
n
n
P
1/2 [(p-cymene)RuCl2]2
Cl
R2
Ti
Ti
Cl
benzene or CH2Cl2
X
X
X
X
4 h, r.t.
X
X
7: n = 0, X = Cl, R = Ph
11
9: n = 2, X = OiPr, R = Ph
10: n = 2, X = OiPr, R = Cy
12
13
Scheme 2.

J. Goux et al. / Journal of Organometallic Chemistry 690 (2005) 301–306
303
Table 1
cies in similar conditions (Table 2). This procedure, if
successful, could provide good insight of the influence
of the titanocene and of the titanium half-sandwich frag-
ments on its late metal neighbour.
˚
Selected geometrical parameters [bond lengths (A) and bond angle (ꢁ)]
for 11
Ti–Ct(1)
Ti–Cl(2)
Ti–Cl(3)
Ti–Cl(1)
Ru–Ct(2)
Ru–P
2.020(5)
2.2249(14)
2.2264(14)
2.2329(14)
1.711(5)
2.345(1)
2.413(1)
2.406(1)
1.823(4)
All the complexes showed an acceptable level of
activity. The ratios cis/trans are quite comparable and
always in favor of the trans isomer. Looking at these re-
sults more in detail, it appears that the early metal frag-
ment plays a subtle role on the reactivity of the catalyst.
Indeed, while (p-cymene)RuCl2PPh3 gives a better yield
than (p-cymene)RuCl2PCy3, the bimetallic complexes
Ti–Cp(CH2)2PCy2/Ru 6 and 13 are more active than
Ti–Cp(CH2)2PPh2/Ru 5 and 12. The competitive forma-
tion of stilbene could explain the inconsistency between
these two results. Indeed a turnover number of 40 for
the metathesis of styrene is reached with (p-cym-
ene)RuCl2PCy3 while (p-cymene)RuCl2PPh3 gives only
10. This result confirms the fact that bulky and basic
phosphine are required for metathesis reaction to turno-
ver [11]. So the sluggish complex (p-cymene)RuCl2PPh3
leads to better yield in cyclopropane than (p-cym-
ene)RuCl2PCy3. A similar phenomenon with bimetallic
species is not observed. Indeed, in agreement with previ-
ous observations [3], it seems that the titanium fragment
inhibits metathesis reaction and so allows the Ti–
Cp(CH2)2PCy2/Ru complexes 6 and 13 to be more active
than Ti–Cp(CH2)2PPh2/Ru 5 and 12. It is also notewor-
thy that the overall best result was obtained with the
bimetallic complex Ti–CpPPh2/Ru 7 which belongs to
the titanocene series with the closest distance between
both metals. We should also mention that the adjunc-
tion of a catalytic amount of Cp2TiCl2 or CpTiCl3 to
a run performed with (p-cymene)RuCl2PCy3 has almost
no effect on the reactivity and the selectivity of the reac-
tion. At last this first comparison between the ruthenium
titanium half-sandwich bimetallic complexes and the
ruthenium titanocene bimetallic complexes is in favour
of the titanocene series.
To account for the selectivity towards cyclopropana-
tion versus metathesis, we presume that the vacant sites
on the ruthenium atom resulting from the disengage-
ment of the p-cymene ligand are probably par-
tially blocked by a chelate effect of the metalloligand.
Therefore, the coordination of styrene to the carbene–
ruthenium complexes to form a ruthenacyclobutane is
no more possible. Nevertheless, the bimetallic complexes
would be able to transfer the carbene moiety onto sty-
rene not coordinated to the ruthenium atom and so to
catalyse the synthesis of cyclopropane. This bimolecular
process has been recently proposed for olefin cycloprop-
anation catalysed by ruthenium–arene complexes con-
taining chelating phosphine–arene ligands [9b]. The
splitting of the heterobimetallic species during catalysis
into monometallic ruthenium complexes containing
phosphines with pendent cyclopentadienes could also
explain this selectivity. Nevertheless, both ruthenium
Ru–Cl(4)
Ru–Cl(5)
P–C(1)
Ct(1)–Ti–Cl(1)
Ct(1)–Ti–Cl(2)
Ct(1)–Ti–Cl(3)
Cl(1)–Ti–Cl(2)
Cl(1)–Ti–Cl(3)
Cl(2)–Ti–Cl(3)
Ct(2)–Ru–Cl(4)
Ct(2)–Ru–Cl(5)
Ct(2)–Ru–P
113.99(13)
115.96(13)
115.76(13)
104.64(6)
100.46(6)
104.22(6)
128.05(15)
124.71(15)
129.66(15)
87.04(4)
P–Ru–Cl(5)
P–Ru–Cl(4)
86.25(4)
87.16(4)
Cl(5)–Ru–Cl(4)
C(1)–P–Ru
C(12)–P–Ru
113.04(13)
116.92(14)
C(1)–P–Ru–Ct(2)
179.1
gles in Table 1. The structure consists in two fragments:
a half titanocene trichloride moiety and a (p-cym-
ene)RuCl2 moiety linked by a phosphorus atom. The
geometry of the CpTiCl3 fragment is typically tetrahe-
dral with structural parameters similar to the phosphine
4 [5]. The ruthenium moiety which presents a usual
three-legged piano stool structure [8] lies on the exo face
of the cyclopentadiene ring in an open area.
Then we carried out the coordination of the phos-
phine functionalised triisopropoxy titanium half-
sandwich complexes 8–10 in similar conditions. The
addition of 8 to the binuclear complex [(p-cym-
ene)RuCl2]2 did not lead to the expected bimetallic com-
plex but gave the monometallic complex (C5H5PPh2)
RuCl2(p-cymene). The disengagement of the cyclopend-
iene results probably from a too big steric hindrance at
the titanium atom. On the other hand, both other com-
plexes 9 and 10, with an ethylene spacer between the Cp
ring and the phosphorus atom, allowed us the access to
the bimetallic complexes 12 and 13 in quantitative
yields. The 31P NMR spectra of 12 and 13 appear as sin-
glets with chemical shifts at 24.0 and 25.7 ppm, respec-
1
tively, and the H NMR spectra show all the signals
relative to each metallic fragment.
Recently, it has been demonstrated that arene
ruthenium complexes catalyse the addition of ethyldia-
zoacetate to styrene resulting in the formation of cyclo-
propane [9,10]. For comparative purposes, we examined
this reaction with the aid of mono- and bimetallic spe-

304
J. Goux et al. / Journal of Organometallic Chemistry 690 (2005) 301–306
Table 2
Addition of ethyl diazoacetate to styrene catalysed by the heterobimetallic complexes 4–6 and 11–13 or by the monometallic complexes (p-
a
cymene)RuCl2PR3
H
CO2Et
Ti-Ru catalyst
-N2
+
N2CHCO2Et
+
+
Ph
Ph
Ph
Ph
CO2Et
Ph
H
Catalyst
Cyclopropanation
Yield (%)b
Metathesis
TONb
trans/cisc
(p-Cymene)RuCl2PPh3
(p-Cymene)RuCl2PCy3
57
46
60
38
49
36
26
47
62/38
55/45
62/38
65/35
57/43
65/35
62/38
64/36
10
40
1
d
CpTiCl2–CpPPh2/Ru (4)
CpTiCl2–Cp(CH2)2PPh2/Ru (5)
CpTiCl2–Cp(CH2)2PCy2/Ru (6)
TiCl3–CpPPh2/Ru (11)
Ti(OiPr)3–Cp(CH2)2PPh2/Ru (12)
Ti(OiPr)3–Cp(CH2)2PCy2/Ru (13)
a
0
4
1
0
1
Conditions: catalyst (0.01 mmol), olefin (20 mmol), ethyl diazoacetate (1 mmol) diluted with the substrate up to 1 ml, 60 ꢁC, addition time (4 h),
overall reaction time (4.25 h).
b
Determinated by GC using n-decane as internal standard.
Determinated by GC analysis.
Demonceau and Noels [9c] have recently obtained a closely similar result.
c
d
titanocene bimetallic complexes and ruthenium titanium
half-sandwich bimetallic complexes should give, after
splitting, the same ruthenium complexes and so far sim-
ilar catalytic results. Moreover in case of the bimetallic
complexes 4 and 11, the phosphorus atom being directly
tethered to the cyclopentadienyl ring, the splitting of the
bimetallic complexes could not lead to a chelating ruthe-
nium complex.
tal analyses were carried out with a EA 1108 CHNS-O
FISONS Instruments. 1H (500 MHz) and 31P{1H}
(202 MHz) spectra were collected on a Bruker 500
MHz Avance DRX spectrometer. Chemical shifts are
relative to internal TMS (1H) or external H3PO4 (31P).
The catalytic cyclopropane synthesis was monitored on
a GC-17A SHIMADZU with a 007 Methyl 5% Phenyl
Silicone Capillary column (30 m). The titanocene phos-
phines 1, 2 and 3 and the heterobimetallic complexes 4, 5
and 6 were synthesised as previously reported [2,3]. The
cyclopentadienyl phosphine ligands Li[C5H4(CH2)2PR2]
[12] and the complexes (g5-C5H4PPh2)TiCl3 7 [5], (g5-
C5H4PPh2)Ti(OiPr)3 [7], (p-cymene)RuCl2(PR3) [13],
were prepared according to the literature methods.
3. Conclusion
This study provides a straightforward access to three
new early-late bimetallic complexes. Preliminary assess-
ment in catalytic cyclopropanation of styrene with
ruthenium titanium half-sandwich bimetallic complexes
and ruthenium titanocene bimetallic complexes as cata-
lyst revealed a cooperative effect. In fact, the early metal
fragment prevents any metathesis reaction to turn over
and so focused the catalytic activity of the late metal
on cyclopropanation. Further studies to rationalised
the different catalytic behaviours of the mono- and
bimetallic complexes are under progress.
4.1. Complex 9
A solution of 0.5 g (1.7 mmol) of Li[C5H4-
(CH2)2PPh2] in 20 ml of THF was added slowly at
0 ꢁC to an equimolar solution of ClTi(OiPr)3 in 20 ml
of THF. After stirring for 3 h at room temperature,
the solvent was removed under vacuo. The residue was
extracted with 80 ml of pentane, filtered through celite,
and evaporated to dryness to afford a yellow oil (84%
yield). 31P{1H} NMR (202 MHz, CDCl3) d ꢀ14.0 (s,
1
4. Experimental
PPh2). H NMR (500 MHz, CDCl3) d 1.26 (d, J = 6.5
i
Hz, 18H, Pr), 2.52–2.54 (m, CH2, 2H), 2.96–2.98 (m,
All reactions were carried out under an atmosphere
of purified argon. The solvents and eluents were dried
by the appropriate procedure and distilled under argon
immediately before use. Standard Schlenk techniques
and conventional glass vessels were employed. Elemen-
CH2, 2H), 4.61 (hept, J = 6.5 Hz, 3H, iPr), 6.11 (pseudo-
triplet, J = 2.5 Hz, Cp, 2H), 6.16 (pseudotriplet, J = 2.5
Hz, Cp, 2H), 7.17–7.23 (m, 6H, Ph), 7.57–7.60 (m, 4H,
Ph). Anal. Calc. for C28H39O3PTi (502.45): C, 66.93;
H, 7.82. Found: C, 66.71; H, 7.77%.

J. Goux et al. / Journal of Organometallic Chemistry 690 (2005) 301–306
305
i
4.2. Complex 10
CH3), 2.60 (hept, J = 6.9 Hz, 1H, Pr), 5.05 (d, J = 6.1
Hz, 2H, p-cymene), 5.24 (d, J = 6.1 Hz, 2H, p-cymene),
6.82 (pseudotriplet, J = 3.7 Hz, Cp, 2H), 7.14 (pseudo-
triplet, J = 3.7 Hz, Cp, 2H), 7.48–7.56 (m, 6H, Ph),
8.10–8.15 (m, 4H, Ph). Anal. Calc. for C27H28Cl5PRuTi
(709.69): C, 45.69; H, 3.98. Found: C, 45.36; H, 4.49%.
This compound was obtained following the above
procedure but with Li[C5H4(CH2)2PCy2] (yellow oil,
98% yield). 31P{1H} NMR (202 MHz, C6D6) d ꢀ3.60
(s, PCy2). 1H NMR (500 MHz, C6D6) d 1.27 (d,
J = 6.1 Hz, 18H, Pr), 1.30–1.94 (m, 22H, Cy), 1.85 (m,
2H, CH2), 3.00 (m, 2H, CH2), 4.64 (hept, J = 6.1 Hz,
i
3H, Pr), 6.14 (pseudotriplet, J = 2.3 Hz, Cp, 2H), 6.17
(pseudotriplet, J = 2.3 Hz, Cp, 2H). Anal. Calc. for
C28H51O3PTi (514.54): C, 65.36; H, 9.99. Found: C,
65.12; H, 10.63%.
i
4.3.1. X-ray structure determination of the complex 11
Single-crystal X-ray diffraction study of 11. Intensi-
ties were collected on an Enraf-Nonius KappaCCD dif-
fractometer at 110 K using Mo Ka radiation. The
structure was solved via a Patterson search program
[14] and refined with full-matrix least-squares methods
[15] based on |F2| with the aid of the WINGX program
[16]. All non-hydrogen atoms were refined with aniso-
tropic thermal parameters (except for one CH2Cl2 disor-
dered solvate molecule). Hydrogen atoms were included
with a riding model with isotropic temperature factors
fixed to 1.2 times (1.5 for methyl groups) those of the
corresponding parent atoms. One CH2Cl2 molecule is
located close to a twofold axis and was isotropically re-
fined with an occupation factor of 0.5 with restrained
bond distances and angle. The essential crystallographic
parameters are listed in Table 3.
4.3. Complex 11
A 25 ml Schlenk flask was charged under argon with
7 (0.14 g, 0.35 mmol), [(p-cymene)RuCl2]2 (0.11 g, 0.17
mmol) and degassed benzene. The mixture was stirred
at room temperature for 4 h during which time a brick
precipitate slowly formed. The solvent was removed by
filtration and the red residue was dried under vacuum
(brick-red powder, 75% yield). 31P{1H} NMR (202
1
MHz, CDCl3) d 23.50 (s, PPh2). H NMR (500 MHz,
CDCl3) d 1.05 (d, J = 6.9 Hz, 6H, Pr), 1.71 (s, 3H,
i
Table 3
Crystal data and structure refinement for 11
Formula
Molecular weight
T (K)
C27H28PCl5RuTi Æ 1.5CH2Cl2
837.07
110(2)
Monoclinic
Crystal system
Space group
C2/c
31.3285(4)
˚
a (A)
˚
b (A)
13.5589(3)
19.3432(3)
˚
c (A)
b (ꢁ)
126.620(1)
6594.7(2)
3
˚
V (A )
Z
8
3352
F(000)
Dcalc (g/cm3)
Diffractometer
Scan type
1.686
Enraf-Nonius KappaCCD
Mixture of / rotations and x scans
0.71073
˚
k (A)
l (mmꢀ1
)
1.417
0.50 · 0.45 · 0.45
0.65
Crystal size (mm3)
ꢀ1
˚
sin(h)/kmax (A
)
Index ranges
Absorption correction
ꢀ36 6 h 6 40, ꢀ16 6 k 6 17, ꢀ25 6 l 6 24
SCALEPACK
19067
7494 (0.029)
Number of reflections collected (RC)
Number of independent reflections (IRC) (Rint
IRCGT = IRC and [I > 2r(I)]
Refinement method
)
6001
Full-matrix least-square on F2
7494/3/360
R1a = 0.0511, wR2b = 0.1247
R1b = 0.0683, wR2b = 0.1341
1.042
Data/restraints/parameters
R for IRCGT
R for IRC
Goodness-of-fit (GoF)c
ꢀ3
˚
Largest difference peak and hole (e A
)
2.38 and ꢀ1.73
P
P
a
b
c
R1 = (iFo|ꢀ|Fci)/ |Fo|.
P
P
2
2 1=2
2
wR2 ¼ ½ wðF o2 ꢀ F c2Þ = ½wðF o2Þ ꢁ where w ¼ 1=½r2ðF o2Þ þ 47:12P þ ð0:061PÞ ꢁ where P ¼ ðmaxðF o2; 0Þ þ 2ꢂF 2c Þ=3.
P
2
1=2
GoF ¼ ½ wðF 2o ꢀ F c2Þ =ðNo ꢀ NvÞꢁ
.

306
J. Goux et al. / Journal of Organometallic Chemistry 690 (2005) 301–306
4.4. Complex 12
of charge, on application to CCDC, 12 Union Road,
Cambridge, CB2 1EZ, UK (fax: +44 1223336033 or
e-mail: deposit@ccdc.cam.ac.uk).
A 25 ml Schlenk flask was charged under argon with
9 (0.5 g, 1 mmol), [(p-cymene)RuCl2]2 (0.3 g, 0.5 mmol)
and dichloromethane. After stirring for 4 h at room tem-
perature, the solvent was removed in vacuo and the
crude residue washed by 2 · 5 ml of pentane to afford
a red powder (100% yield). 31P{1H} NMR (202 MHz,
Acknowledgment
The authors thank Pr. A. Demonceau for helpful
discussions.
1
CDCl3) d 24.00 (s, PPh2). H NMR (500 MHz, CDCl3)
d 0.90 (d, J = 6.9 Hz, 6H, iPr), 1.20 (d, J = 6.1 Hz, 18H,
iPr), 1.84 (s, 3H, CH3), 2.70 (m, 2H, CH2), 2.76 (hept,
i
J = 6.9 Hz, 1H, Pr), 3.55 (m, 2H, CH2), 4.56 (hept,
References
J = 6.1 Hz, 3H, iPr), 5.04 (d, J = 6.0 Hz, 2H, p-cymene),
5.11 (d, J = 6.0 Hz, 2H, p-cymene), 6.00 (pseudotriplet,
J = 2.4 Hz, Cp, 2H), 6.02 (pseudotriplet, J = 2.4 Hz,
Cp, 2H), 7.23–7.27 (m, Ph, 6H), 8.07–8.10 (m, Ph,
4H). Anal. Calc. for C38H53Cl2O3PRuTi (808.64): C,
56.44; H, 6.61. Found: C, 56.40; H, 6.29%.
[1] (a) N. Wheatley, P. Kalck, Chem. Rev. (1999) 3379;
(b) L. Gelmini, D.W. Stephan, Organometallics 5 (1986) 67;
(c) R. Choukroun, D. Gervais, P. Kalck, F. Senocq, J. Organ-
omet. Chem. 335 (1987) C9;
(d) M.J. Hostetler, M.D. Butts, R.G. Bergman, J. Am. Chem.
Soc. 115 (1993) 2743;
(e) A.M. Trzeciak, J.J. Ziolkowski, R. Choukroun, J. Mol. Catal.
110 (1996) 135;
4.5. Complex 13
(f) Y. Yamaguchi, N. Suzuki, T. Mise, Y. Wakatsuki, Organo-
metallics 18 (1999) 996;
(g) B.E. Bosch, I. Brummer, K. Kunz, G. Erker, R. Fro¨hlich, S.
¨
This compound was obtained following the above
procedure but with 10 (red powder, 100% yield).
31P{1H} NMR (202 MHz, CDCl3) d 25.70 (s, PPh2).
1H NMR (500 MHz, CDCl3) d 1.10 (d, J = 6.1 Hz,
Kotila, Organometallics 19 (2000) 1255.
[2] P. Le Gendre, P. Richard, C. Mo¨ıse, J. Organomet. Chem. 605
(2000) 151.
i
i
[3] P. Le Gendre, M. Picquet, P. Richard, C. Mo¨ıse, J. Organomet.
Chem. 643–644 (2002) 231.
18H, Pr), 1.25 (d, J = 6.9 Hz, 6H, Pr), 1.39–2.16 (m,
22H, Cy), 2.07 (s, 3H, CH3), 2.30 (m, 2H, CH2), 2.71
[4] P. Le Gendre, V. Comte, A. Michelot, C. Mo¨ıse, Inorg. Chim.
Acta 350 (2003) 289.
i
(m, 2H, CH2), 2.82 (hept, J = 6.9 Hz, 1H, Pr), 4.49
(hept, J = 6.1 Hz, 3H, iPr), 5.52–5.56 (m, 4H, p-cymene),
6.00 (pseudotriplet, J = 2.7 Hz, Cp, 2H), 6.06 (pseudo-
triplet, J = 2.7 Hz, Cp, 2H). Anal. Calc. for
C38H65Cl2O3PRuTi (820.74): C, 55.61; H, 7.98. Found:
C, 55.59; H, 7.83%.
[5] J.C. Flores, R. Hernandez, P. Royo, A. Butt, T.P. Spaniol, J.
Okuda, J. Organomet. Chem. 593 (2000) 202.
[6] R. Gorsich, J. Am. Chem. Soc. 82 (1960) 4211.
[7] S. Barry, A. Kucht, H. Kucht, M.D. Rausch, J. Organomet.
Chem. 489 (1995) 195.
[8] (a) Structures including both ferrocene and (g6-arene)RuCl2
linked by a diphenyl phosphido bridge: J.F. Mai, Y. Yamamoto,
J. Organomet. Chem. 560 (1998) 223;
4.6. Cyclopropanations
(b) J.F. Mai, Y. Yamamoto, J. Organomet. Chem. 545 (1997)
577.
Cyclopropanation reactions were performed accord-
ing to the procedure described by Demonceau and
co-workers [9]. Ethyl diazoacetate (1 mmol diluted by
styrene up to 1ml) was added over the period of 4 h with
the aid of a syringe pump to a solution of styrene (20
mmol) containing the catalyst (0.01 mmol) and 1 mmol
of decane (internal standard). The reaction was stirred
at 60 ꢁC during an overall time of 4.25 h. Product yields
were determinated by GC using the experimentally
measured cyclopropane–decane response ratio.
[9] (a) F. Simal, A. Demonceau, A.F. Noels, Tetrahedron Lett. 39
(1998) 3493;
(b) F. Simal, D. Jan, A. Demonceau, A.F. Noels, Tetrahedron
Lett. 40 (1999) 1653;
(c) A. Demonceau, A.F. Noels, unpublished results.
[10] (a) H. Nishiyama, K. Aoki, H. Itoh, T. Iwamura, N. Sakata, O.
Kurihara, Y. Motoyama, Chem. Lett. (1996) 1071;
(b) I.W. Davies, L. Gerena, D. Cai, R.D. Larsen, T.R. Verhoeven,
P.J. Reider, Tetrahedron Lett. 38 (1997) 1145.
[11] S.T. Nguyen, L.K. Johnson, R.H. Grubbs, J.W. Ziller, J. Am.
Chem. Soc. 114 (1992) 3974.
[12] (a) T. Kauffmann, M. Bisling, J.H. Teuben, R. Konig, Angew.
Chem., Int. Ed. Engl. 19 (1980) 328;
(b) T.W. Graham, A. Llamazares, R. McDonald, M. Cowie,
Organometallics 18 (1999) 3490.
5. Supplementary material
[13] S. Serron, S.P. Nohan, Organometallics 14 (1995) 4611.
[14] G.M. Sheldrick, SHELXS-97: Program for Crystal Structure
Solution, University of Go¨ttingen, Go¨ttingen, Germany, 1997.
[15] G.M. Sheldrick, SHELXL-97: Program for Crystal Structure
Refinement, University of Go¨ttingen, Go¨ttingen, Germany, 1997.
[16] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837–838.
Crystallographic data (excluding structure factors)
have been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication No.
CCDC 237979. Copies of the data can be obtained, free