p-Cymene ruthenium thioether complexes

Article abstract of DOI:10.1016/j.jorganchem.2006.11.046

Thioethers PhC₂H₄SMe, PhC₃H₆SⁱPr and MeSAllyl form substitutionally labile monomeric adducts (p-cymene)RuCl₂(SRR′) (2a-c) upon treatment with the {(p-cymene)RuCl₂}₂ dimer (p-cymene = η⁶-MeC₆H₄ⁱPr-1,4). Pure adducts were obtained by crystallization from CH₂Cl₂/Et₂O, and 2a,c as well as the bis(thioether) complex [(p -cymene) RuCl (SMe₂)²]⁺ SbF₆^- (3) were studied by X-ray crystallography. The trichloro bridged diruthenium complex [(p -cymene) Ru}₂ (μ -Cl)₃]⁺ SbF₆^- is formed as a byproduct in the preparation of 3 and was also crystallographically characterized. In solution, pure samples 2a-c equilibrate with free thioether and the dimeric starting complex 1. The amount of 1 present in these mixtures increases with increasing bulk of the thioether substituents. Attempts to thermally replace the cymene ligand by the dangling arene substituent of the thioether ligand of 2a,b failed. Complexes 2a-c as well as the dimethylsufide derivative 2d were studied by cyclic voltammetry and display a close to reversible (2a,c,d) or partially reversible (2b) oxidation near +0.85 V and an irreversible reduction at rather negative potential. New peaks observed after oxidation and reduction point to dissociation of the thioether ligand as the main decomposition pathway of the associated radical cations and anions.

Full text of DOI:10.1016/j.jorganchem.2006.11.046

Journal of Organometallic Chemistry 692 (2007) 1496–1504
www.elsevier.com/locate/jorganchem
p-Cymene ruthenium thioether complexes
b
b
b
Jadranka Cubrilo a,b, Ingo Hartenbach , Falk Lissner , Thomas Schleid ,
ˇ
b
a,
*
Mark Niemeyer , Rainer F. Winter
a
Institut fu¨r Anorganische, Chemie der Universita¨t Regensburg, Universita¨tsstraße 31, D-93040 Regensburg, Germany
b
Institut fu¨r Anorganische Chemie der Universita¨t Stuttgart, Pfaffenwaldring 55, D-70569 Stuttgart, Germany
Received 18 October 2006; received in revised form 29 November 2006; accepted 29 November 2006
Available online 6 December 2006
Abstract
Thioethers PhC2H4SMe, PhC3H6SiPr and MeSAllyl form substitutionally labile monomeric adducts (p-cymene)RuCl2(SRR0) (2a–c)
i
upon treatment with the {(p-cymene)RuCl2}2 dimer (p-cymene = g6-MeC6H4 Pr-1,4). Pure adducts were obtained by crystallization
2 þ
from CH2Cl2/Et2O, and 2a,c as well as the bis(thioether) complex ½ðp-cymeneÞRuClðSMe2Þ ꢀ SbFꢁ (3) were studied by X-ray crystallog-
6
raphy. The trichloro bridged diruthenium complex ½ðp-cymeneÞRug2ðl-ClÞ3ꢀþSbF6ꢁ is formed as a byproduct in the preparation of 3 and
was also crystallographically characterized. In solution, pure samples 2a–c equilibrate with free thioether and the dimeric starting com-
plex 1. The amount of 1 present in these mixtures increases with increasing bulk of the thioether substituents. Attempts to thermally
replace the cymene ligand by the dangling arene substituent of the thioether ligand of 2a,b failed. Complexes 2a–c as well as the dimethyl-
sufide derivative 2d were studied by cyclic voltammetry and display a close to reversible (2a,c,d) or partially reversible (2b) oxidation near
+0.85 V and an irreversible reduction at rather negative potential. New peaks observed after oxidation and reduction point to dissoci-
ation of the thioether ligand as the main decomposition pathway of the associated radical cations and anions.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Ruthenium; Arene complexes; X-ray; Electrochemistry; Thioether
1. Introduction
bifunctional bidentate ligands (L = O, NTos) are the hall-
mark of asymmetric hydrogen transfer hydrogenation of
ketones [10–13]. Cationic complexes with neutral bidentate
chelate ligands have been employed in arene hydrogenation
[14], asymmetric Diels–Alder reactions [15,16], hydrogen
transfer hydrogenations [17], and styrene polymerization
[18], to mention only a few important applications. In con-
trast to arene half-sandwich complexes bearing phosphine
and amine ligands, similar complexes of thioethers have
only been little explored, probably because of the inherently
weaker Ru–SR2 bond. Thus, studies by Dixneuf and
coworkers have revealed, that thioether adducts (g6-are-
ne)RuCl2(SR2) are labile and exist in equilibrium with their
dimeric halide bridged precursors. Dicationic bis(adducts)
[(g6-arene)RuCl2(SR2)2]+ are more stable but may still eas-
ily exchange one of the thioether moieties for a better donor
such as a phosphine [19]. Yamamoto and his coworkers
have recently reported on more stable ruthenium thioether
complexes with 1,3- and 1,4-dithiane ligands [20]. The use of
Half-sandwich arene complexes of ruthenium with het-
eroatom donor ligands such as phosphines and amines are
being intensively used as catalysts for many important pro-
cesses. Cationic phosphine substituted allenylidene com-
plexes [(g6-arene)Cl(PR3)Ru@C@C@CAr2]+ have recently
been identified as highly efficient catalysts for olefin metath-
esis, rivalling the Grubbs carbene systems [1–6]. Simple
phosphine adducts (g6-arene)RuCl2(PR3) catalyze the atom
transfer radical polymerization (ATRP) of olefins [7,8], the
isomerization of allylic alcohols and the hydration of alky-
nes to ketones or aldehydes [9]. Amine/amide or amine/alk-
oxy complexes (g6-arene)Ru(NH2CHRCHRL)H with
*
Corresponding author. Tel.: +49 941 943 4485; fax: +49 941 943 4488/
711 685 4165.
E-mail address: rainer.winter@chemie.uni-regensburg.de (R.F. Win-
ter).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.11.046

ˇ
J. Cubrilo et al. / Journal of Organometallic Chemistry 692 (2007) 1496–1504
1497
thiacrown or mixed thiol/thiolate chelate ligands also aids
to stabilize the ruthenium thioether bond [21–23]. We here
report on the synthesis and characterization of thioether
adducts (p-cymene)RuCl2(SR2) 2a–c, including structural
characterization of the MeSC2H4Ph (2a) and the MeSCH2-
CH@CH2 (2c) complexes, which constitute rare examples
of simple thioether adducts to be investigated by this
method. Also included are the X-ray structures of the
bis(thioether) complex ½ðp-cymeneÞRuClðSMe2Þ ꢀþSbFꢁ
copy. While the resonance signals of the dissociated and
the coordinated SRR0 ligand are usually coalesced, giving
just one set of signals, the relative amounts of complexes
2a–c and 1 in the equilibrium mixtures can be determined
by integrating the different sets of CH resonances of the
cymene ligands. Values of 12:1, 5:2 and 7.5:1 were obtained
for 2a–c in CD2Cl2 respectively. This order roughly corre-
sponds to the steric bulk of the substituents on the thioe-
ther ligand. Nearly identical values were obtained by
comparing the peak currents of 2a–c and 1 in cyclic vol-
tammetry experiments (see below). The thioether methy-
lene and methyl resonance signals always integrate as 1:1
and 3:2, respectively, with the sum of each set of CH-reso-
nance signals of 2a–c and 1, thus ruling out contamination
of the solid adducts by unreacted 1. In toluene-d8 2b ini-
tially forms a clear orange solution from which 1 slowly
crystallizes. The final solution equilibrium mixture contains
free PhC3H6SiPr and 2b in an approximate 1:1 ratio as
based on comparison of the integrals of the arene CH sig-
nals of 2b to those of the thioether methylene protons. A
similar 2b:1 ratio of 1:0.88 was observed in CD3OD, where
1 remains dissolved. For every thioether complex, the res-
onance signals of the alkyl substituents neighbouring the
S atom are broadened due to sulfur inversion [24].
2
6
(3) and of ½ðp-cymeneÞRug Cl3ꢀþSbFꢁ (4). The latter was
2
6
formed as a by-product in the preparation of compound
3. Redox potentials for the Ru(II/III) oxidations and of
the irreversible reductions of these complexes were deter-
mined by cyclic voltammetry. Some hints as to the degrada-
tion pathway of the associated radical cations and anions
were obtained from a comparison of their voltammograms
with those of 1.
2. Results and discussion
The monothioether complexes 2a–c were prepared by
reacting dimeric {(p-cymene)RuCl2}2 (1) in dichlorometh-
ane with a slight excess (ca. 1.5 equiv.) of the correspond-
ing thioether (see Chart 1). The new arylalkyl substituted
thioethers, isopropyl(3-phenylpropyl)sulfide, PhC3H6SiPr,
and methyl(2-phenylethyl)sulfide, PhC2H4SMe, were pre-
pared according to established procedures via nucleophilic
substitution of bromide or iodide by the respective thiolate.
Pure thioether complexes could be obtained as orange-red
crystals (2a,c) or microcrystals (2b) by recrystallization
from dichloromethane/ether mixtures. Simple thioether
adducts (g6-arene)RuCl2(SRR0) are reportedly labile in
solution and exist in equilibrium with the dichloro bridged
precursor 1 and free thioether [19]. Ligand dissociation
from complexes 2 is readily observed by NMR spectros-
Recent work on similar (g6-arene)dichloro complexes
bearing arylalkyl substituted phosphines with flexible
alkyl spacers has disclosed, that, upon thermal treatment,
the coordinated arene is readily replaced by a dangling
aryl substituent [3,25–27]. This reaction provides an easy
access to tethered complexes where the phosphine substi-
tuted arene serves as an eight electron donor chelate
ligand. Some of these complexes display reactivities that
markedly differ from those of their non-tethered analogs
[28,29]. In contrast, no such arene substitution was
observed for complexes 2a,b even in hot chlorobenzene.
Cl
+
Ru
+
Me
Ru
Ru
-
-
SbF6
Cl
SbF6
Cl
Cl
Cl
Cl
Cl
S
S
Me
Ru
Ru
Cl
Me
Me
3
4
1
Ru
S
Ru
Ru
Ru
S
Cl
Me
Cl
iPr
Cl
Cl
Me
Cl
Cl
Me
Cl
Cl
S
S
Me
Ph
2a
2b
2c
2d
Ph
Chart 1. Complexes under investigation.

ˇ
J. Cubrilo et al. / Journal of Organometallic Chemistry 692 (2007) 1496–1504
1498
At higher temperatures, e.g. in refluxing 1,2-dichloroben-
zene, 2a,b decompose.
C2
C9
C7
C1
C6
C5
Dixneuf had reported that the cationic bis(thioether)
complexes [(g6-arene)RuCl(SMe2)2]+ are considerably more
stable toward thioether dissociation, yet easily exchange
one SMe2 ligand for a better donor like a phosphine to give
chiral-at-metal complexes [(g6-arene)RuCl(SMe2)(L)]+
[19]. In the course of our work we also prepared the cat-
ionic p-cymene bis(dimethylthioether) complex 3 as its
SbFꢁ6 salt. In our hands, under the conditions reported
by Dixneuf, this complex was obtained along with small
quantities of the trichloþro bridged bis(arene) complex
½fðp-cymeneÞRug ðl-Cl3Þꢀ SbFꢁ (4). Crystallization from
C3
C8
C4
Ru
C10
Cl1
C12
Cl2
C16
C15
S
C17
C14
C19
C13
C18
C11
Fig. 1. Structure of the thioether complex 2a in the solid state with the
atom numbering. Thermal ellipsoids are drawn at a 50% probability level.
2
6
ethanol resulted in the crystallization of both 3 and 4. Crys-
tals of 3 and 4 could be separated based on their different
shapes – orange diamond-shaped crystals for 3, orange
rods for 4. The structures of both complexes were deter-
mined by X-ray diffraction as it will be discussed in the fol-
lowing section.
C1
Cl1
C11
C2
C3
Ru
C7
C6
The reaction of thioether and bis(thioether) chloro ruthe-
nium complexes with terminal alkynes in methanol was
reported to give methoxycarbenes [(g6-arene)Cl(SMe2)-
Ru@C(OMe)CH2Ph]+, most probably via vinylidene inter-
mediates [19]. In trying to access similar half-sandwich
ruthenium allenylidene or vinylidene derivatives with thioe-
ther ligands, 2a,c and [(p-cymene)RuCl(SMe2)2]+ (3) were
reacted with 1,1-diphenyl-prop-2-yn-1-ol, 2-phenylbutynol,
2-methylbutynol or AgSbF6/1,1-diphenyl-prop-2-yn-1-ol in
dichloromethane or methanol. These reactions either did
not proceed at all (3) or led to the formation of dark orange,
oily unidentified mixtures. The absence of any allenylidene
or vinylidene IR bands in these mixtures points to the fact
that such complexes, if formed at all, are too unstable to
allow for their isolation.
S
C13
C4
C14
C12
C5
Cl2
C481
C10
C9
Fig. 2. Structure of the thioether complex 2c in the solid state with the
atom numbering. Thermal ellipsoids are drawn at a 50% probability level.
C9
C8
3. X-ray structure determinations of 2a,c, 3 and 4
C5
C4
C3
Crystals of compounds 2a,c, 3 and 4 were grown by slow
diffusion of ether into a concentrated solution of the
respective complex in CH2Cl2 and investigated by X-ray
crystallography. Figs. 1–4 display the structures of the
complexes or complex cations in the solid state. The most
important interatomic distances and bond angles are col-
lected in Tables 1 and 2 while Table 3 provides information
pertinent to the data collection and refinement.
C7
C6
Cl
C10
C2
C1
Ru
C11
C14
The complexes (p-cymene)RuCl2(MeSC2H4Ph) (2a),
(p-cymene)RuCl2(MeSC3H5) (2c) and ½ðp-cymeneÞRuCl-
S1
S2
ðSMe2ÞꢀþSbFꢁ (3) represent rare examples of structurally
C13
6
characterized (arene)Ru complexes bearing simple thioe-
ther ligands. In 2a,c, two chloride and the thioether ligands
occupy cis-disposed coordination sites and form three
nearly equally long legs of the piano-stool structure which
is archetypical of [(g6-arene)RuL3] half-sandwich arene
C12
Fig. 3. Structure of the complex cation in complex 3 in the solid state with
the atom numbering. Thermal ellipsoids are drawn at a 50% probability
level.
˚
complexes. The Ru–S bond lengths are close to 2.40 A
˚
˚
(2.4032(8) A in 2a and 2.3978(8) A in 2c) and are slightly
Ru–S bonds are observed in arene half-sandwich com-
plexes where the thioether moiety is part of a macrocycle
[22] or a mixed thioether thiolate chelate ligand [21,22].
longer than the corresponding Ru–S bond length of
˚
2.357(4) A in (p-cymene)RuCl2(1,3-dithiane) [20]. Shorter

ˇ
J. Cubrilo et al. / Journal of Organometallic Chemistry 692 (2007) 1496–1504
1499
In such complexes Ru–S bond lengths usually range from
2.30 to 2.34 A. The longer Ru–S bonds reflect the inherent
C9
˚
C8
C4
weakness of the unsupported Ru–S(thioether) bond as it is
evident from the ready dissociation of the SR2 ligand and
partial formation of dichloro bridged {(g6-arene)RuCl2}2
dimers. Bond angles subtended by the ligands forming
the legs and the ruthenium atom are all close 90ꢁ in 2a
(Cl1–Ru–Cl2 = 89.52(3)ꢁ, Cl1–Ru–S = 88.14(3)ꢁ, Cl2–
Ru–S = 84.51(3)ꢁ) and attest to the overall octahedral
coordination of the metal atom. Larger deviations from
this geometry are observed for 2c, where the S–Ru–Cl
angles are somewhat acute at 80.37(2) and 81.50(2)ꢁ. The
C5
C6
C3
Cl2
C11
C17
C16
Ru1
C12
Ru2
C7
Cl1
F6
C2
C1
C15
C18
Sb2
Cl3
F3
C14
C19
F5
C20
˚
Sb1
F1
F2
Ru–Cl bond lengths of 2.4108(8) and 2.4389(8) A (2a) or
F4
2.4104(7) and 2.4121(11) (2c) are unexceptional for mono-
meric dichloro arene ruthenium complexes featuring neu-
tral two-electron donor ligands
C6Me3H3)RuCl2 (pyridine) (2.419(2), 2.415(2) A) [30],
L
like, e. g. (g6-
˚
(p-cymene)RuCl2(2-aminopyridine) (2.405(3) and 2.402(3)
Fig. 4. Structure of ½fðp-cymeneÞRug Cl3ꢀþSbFꢁ (3) in the solid state.
˚
2
6
A) [31], and (p-cymene)RuCl2(1,3-dithiane) (2.389(5) and
Note that the Sb atoms reside on special positions {Sb1/Sb1a: (0,0,1/2)
and (0,1,1/2), respectively and Sb2/Sb2a (1/2,1,0) bzw. (1/2,0,1),
respectively} and are thus related by an inversion centre. Thermal
ellipsoids are drawn at a 50% probability level.
˚
2.406(4) A) [20]. The Ru–C(cymene) bond lengths vary
˚
˚
from 2.149(3) A to 2.215(3) A. In 2a, the longest Ru–C
bond involves the methyl substituted carbon atom C2 as
it is observed for many cymene ruthenium complexes, while
this is not the case for 2c. The arene rings display a distinct
C–C bond length alternation, and the average values of the
˚
long and short bonds differ by 0.02 A. Essentially the same
Table 1
pattern is also observed for the bis(thioether) complex 3
and for one of the cymene ligands of the trichloro bridged
dimer 4 (vide infra).
Interatomic distances and bond angles for complexes 2a,c and 3
2a
2c
3
Ru–C2
Ru–C3
Ru–C4
Ru–C5
Ru–C6
Ru–C7
Ru–Cl
2.215(3)
2.177(3)
2.149(3)
2.194(3)
2.178(3)
2.191(3)
2.4108(8),
2.4389(8)
2.4032(8)
2.191(2)
2.165(2)
2.187(3)
2.207(2)
2.200(2)
2.190(2)
2.4104(7),
2.4121(11)
2.3978(8)
2.220(3)
2.200(3)
2.198(3)
2.239(3)
2.227(3)
2.183(3)
2.3944(7)
The structure of the complex cation of 3 (see Fig. 3)
resembles that of 2a,c in all its main structural characteris-
tics i.e. its three legged piano stool geometry with three
essentially equally long legs. When comparing the struc-
tures of 2a,c and of 3, a small yet significant decrease in
˚
Ru–S bond lengths to 2.3833(7) and 2.3881(7) A and of
˚
the RuCl bond length to 2.3944(7) A is observed. Similar
Ru–S
2.3833(7),
2.3881(7)
1.797(4),
1.797(4)
1.800(4),
1.792(4)
values werþe observed for ½ðp-cymeneÞRuClðg1 : g1-1;
4-dithianeÞꢀ PFꢁ6 [20]. This may be ascribed to the overall
positive charge of the complex resulting in a stronger
attraction between the metal atom and the electron-rich
donor ligands and slightly shorter bond lengths. In con-
trast, the Ru–C bonds become slightly longer and their
S–C11, S–C13a
S–C12, S–C14a
1.806(4)
1.848(4)
1.811(3)
1.824(4)
Cl–Ru–Cl
Cl–Ru–S
89.52(3)
88.14(3),
84.51(3)
–
88.31(4)
80.37(2),
81.50(2)
–
87.68(3),
81.86(3)
87.51(3)
˚
˚
average value increases from 2.185 A in 2a or 2.190 A in
S–Ru–S
˚
2c to 2.211 A in 3. In the crystalline state the two thioether
a
For 3.
ligands adopt different orientations with respect to the
Table 2
Interatomic distances and bond angles for ½fðp-cymeneÞRug Cl3ꢀþSbFꢁ (4)
2
6
Ru1–Cl1
Ru2–Cl1
Ru1–C2
Ru1–C5
Ru2–C12
Ru2–C15
Cl1–Ru1–Cl2
Cl1–Ru2–Cl2
Ru1–Cl1–Ru2
2.4238(12)
2.4341(11)
2.176(6)
2.184(5)
2.172(5)
2.204(4)
79.61(4)
79.42(4)
85.07(4)
Ru1–Cl2
Ru2–Cl2
Ru1–C3
Ru1–C6
Ru2–C13
Ru2–C16
Cl2–Ru1–Cl3
Cl2–Ru2–Cl3
Ru1–Cl2–Ru2
2.4316(13)
2.4307(12)
2.157(6)
2.147(4)
2.139(5)
2.165(5)
79.17(5)
79.14(5)
84.97(4)
Ru1–Cl3
Ru2–Cl3
Ru1–C4
Ru1–C7
Ru2–C14
Ru2–C17
Cl1–Ru1–Cl3
Cl1–Ru2–Cl3
Ru1–Cl3–Ru2
2.4299(13)
2.4324(13)
2.143(6)
2.174(5)
2.171(5)
2.163(5)
79.53(5)
79.28(5)
84.97(4)

ˇ
J. Cubrilo et al. / Journal of Organometallic Chemistry 692 (2007) 1496–1504
1500
Table 3
Data pertaining to the data collection and structure refinement for complexes 2a,c, 3 and 4
Compound
2a
2c
3
4
Empirical formula/f. wt. (g/
mol)
Temperature (K)
Crystal system
C19H26Cl2RuS/458.43
C14H22Cl2RuS · 1/2
CH2Cl2/436.81
173(2)
Monoclinic
C2/c
C14H26ClF6RuS2Sb/630.74
C20H28Cl3F6Ru2Sb/812.66
293(2)
Monoclinic
P21/n
293(2)
Monoclinic
P21/c
293(2)
Triclinic
ꢀ
Space group
P1
Unit cell dimensions
˚
a (A)
11.2219(2)
12.1458(2)
14.8970(2)
22.674(5)
15.857(3)
11.948(2)
8.4828(1)
17.2331(2)
15.2935(2)
8.2786(1)
10.6537(2)
16.6834(2)
99.094(1)
101.480(1)
108.578(1)
1327.45 (3)
2
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
90.455(1)
120.58(3)
93.369(1)
3
˚
V (A )
2030.38(6)
4
1.500
1.135
3698.3(13)
8
1.569
1.382
2231.82(5)
4
1.877
2.238
Z
Dcalc (Mg/m3)
Absorption coefficient
(mmꢁ1
2.033
2.485
)
h Range for data collection
2.82–27.48
2.13–27.99
2.71–27.50
2.57–27.40
(ꢁ)
Limiting Indices
ꢁ14 < h < 14,
ꢁ15 < k < 15,
ꢁ19 < l < 19
ꢁ29 < h < 25,
ꢁ20 < k < 7,
ꢁ11 < h < 11,
ꢁ22 < k < 22,
ꢁ17 < l < 19
ꢁ10 < h < 10,
ꢁ13 < k < 13,
ꢁ21 < l < 21
ꢁ15 < l < 15
Reflections collected/uniquea 41082/4639 (0.0499)
4660/4550 (0.0337)
46487/5111 (0.0512)
47238/6040 (0.0832)
(Rint
)
Data/restraints/parameter
R (I > 2r(I))
Rw (all data)
4639/0/313
4550/2/183
5111/0/246
6040/0/331
R1 = 0.0350, wR2 = 0.0801
R1 = 0.0444, wR2 = 0.0844
1.080
R1 = 0.0332, wR2 = 0.1044 R1 = 0.0313, wR2 = 0.0750
R1 = 0.0350, wR2 = 0.1054 R1 = 0.0342, wR2 = 0.0770
1.853
R1 = 0.0460, wR2 = 0.0853
R1 = 0.0694, wR2 = 0.0882
1.926
GooF (all Data)
1.146
Maximum and minimum res. 0.591 and ꢁ0.515
1.352 and ꢁ1.314
0.736 and ꢁ0.885
1.053 and ꢁ0.686
dens. (e Aꢁ3
)
a
IP3r(I).
arene ligand. While the methyl groups on S1 are in an exo-
position with the methyl groups pointing away from the
cymene ligand the opposite is true for the second thioether
ligand associated with S2.
ne)Ru}2(l-Cl)3]+ where arene is C6H6, toluene [33], C6Me6
[34], or ethoxybenzene [35], and with the bis(p-cymene)
complex ½fðp-cymeneÞRug2ðl-ClÞ3ꢀþBPh4ꢁ [36]. Owing to
restrictions imposed by face-sharing, the Cl–Ru–Cl angles
lie in a narrow range from 79.14(5) to 79.61(4)ꢁ. The Ru–
Cl–Ru angles are all close to 85ꢁ and this againþclosely
resembles the values of ½fðp-cymeneÞRug2ðl-ClÞ3ꢀ BPhꢁ4 ,
where these angles range from 84.1 to 86.1ꢁ. Ru–C bonds
The structure of the dinuclear complex 4 is shown in
Fig. 3. A view of the unit cell may be found as Fig. 1 of
the Supplementary material. The complex cation adopts
the familiar structure of two face-sharing octahedra with
the arene ligands and the three chlorine atoms as the oppo-
site faces. This motif is highly common of complexes of the
general composition [L3M(l-L0)3ML3]n+ and has ample
precedence in arene ruthenium chemistry [32]. Individual
arene and Cl3 planes are nearly parallel to each other with
angles between their normals of 2.69ꢁ (Arene1-Cl3), 0.28ꢁ
(Arene2-Cl3) and 2.87ꢁ (Arene1–Arene2), where Arene1
and Arene2 denote the arene ligands bonded to atoms
Ru1 and Ru2, respectively. The metal atoms are slightly
shifted off-centre toward the common Cl3 face of the bisoc-
˚
are in the range from 2.139(5) and 2.204(4) A. For each
cymene ligand, the longest Ru–C bond involves the C atom
i
that bears the Pr group. The cymene ligands at Ru(1) and
i
Ru(2) differ in the orientation of the Pr-substituents as it
is shown by the torsional angles of C(4)–C(5)–C(8)–
C(9) = 93.7(9)ꢁ for Ru(1), resembling an orthogonal
orientation with respect to the ring plane, and C(14)–
C(15)–C(18)–C(20) = 30.8(7)ꢁ for Ru(2).
The related complexes ½fðg6-C6H6ÞRug2Cl3ꢀþBF4ꢁ and
½fðg6-C6H5MeÞRug2Cl3ꢀþBFꢁ4 display interesting intermo-
lecular interactions based on short CHꢂ ꢂ ꢂCl and CHꢂ ꢂ ꢂF
contacts between arene CH and the bridging chloride
ligands or the F-atoms of the BFꢁ4 counterions [33]. In 4,
˚
tahedron. Thus, the Ru–Cl3 distances amount to 1.640 A
˚
(Ru1–Cl3) and 1.644 A (Ru2–Cl3) while the Ru atoms are
˚
˚
1.647 A (Ru1) and 1.651 A (Ru2) away from the corre-
sponding arene planes. Ru–Cl distances range from
i
the Pr and the methyl substituents on the arene induce
˚
2.4238(12) to 2.4341(11) A and are thus slightly longer than
much larger anion/cation separations and prevent the for-
mation of such a hydrogen bonding network (see Fig. 1 of
the Supplementary material). All CHꢂ ꢂ ꢂCl contacts are
typical bonds to terminal chloride ligands as in complexes 2
or 3. The Ru–Cl distances in 4 also match well with the lit-
erature data of other complexes of the type [{(g6-are-
˚
longer than 2.93 A and are thus, at best, only slightly

ˇ
J. Cubrilo et al. / Journal of Organometallic Chemistry 692 (2007) 1496–1504
1501
shorter than the sum of the van der Waals radii. The same
holds for possible interion hydrogen bonded contacts to
the SbFꢁ6 counterion. Two notable exceptions are the short
CHꢂ ꢂ ꢂF contacts between atoms F2 and the hydrogen atom
*
˚
attached to C17 (2.440 A) and atoms F4 and the hydrogen
˚
atom at C7 (2.469 A). All other CHꢂ ꢂ ꢂF distances are iden-
*
tical to or larger than the sum of the van der Waals radii
˚
(2.55 A).
*
4. Electrochemistry
*
The thioether complexes 2a–c, known (p-cymene)RuCl2-
(SMe2) (2d), and the bis(thioether) adduct 3 (see Chart 1)
were studied by cyclic voltammetry. Relevant data are
compiled in Table 4. Figs. 5 and 6 display voltammograms
recorded for complexes 2b and 2c as a representative
examples.
All complexes undergo one partially to nearly reversible
oxidation and one chemically completely irreversible
reduction within the potential window of the CH2Cl2/
NBu4PF6 supporting electrolyte. The one-electron nature
1.0 0.5 0.0 -0.5
-2.0
-2.5 V
-1.0 -1.5
Fig. 6. Cyclic voltammograms of complex 2c in CH2Cl2/NBu4PF6 (0.1 M)
at RT and v = 0.1 V/s. Upper curve: oxidation scanned first, lower curve:
reduction scanned first. The peaks due to the reduction of 1 and to the
oxidation of the electroactive product formed from the radical anion of 1
are indicated by a star symbol.
of these waves was ascertained from peak potential separa-
tions and peak currents that resemble those of the internal
ferrocene standard present at equimolar concentration
(cyclic voltammetry), as well as peak half-widths (square
wave voltammetry) that correspond to the values expected
for this stoichiometry and from nearly identical peak
currents associated with the oxidation and the reduction pro-
cesses. At sweep rates of about 1 V/s or larger, peak-to-peak
separations of the ruthenium complexes are notably larger
than those of the internal ferrocene standard, pointing to
slightly lower electron transfer kinetics for the ruthenium
complexes. Chemical processes following oxidation could
be fully suppressed by applying higher sweep rates or low-
ering the temperature, and full chemical reversibility was
attained at 195 K in each case. Reduction, however,
remained a completely irreversible process. The mono(thi-
oether) complexes undergo oxidation at potentials near
0.85 V whereas the cationic bis(thioether) derivative 3 is
much harder to oxidize and gives an E1/2 of +1.44 V. The
oxidation potentials of 2a–c and of the related SMe2
derived complex 2d are by about 100 mV higher than those
of similar phosphine derivatives [26,37–39], and this signals
that SR2 ligands are inferior electron donors compared to
phosphines. A similar anodic shift is seen for the reduction
peak potential of 3 compared to 2a–c. The strong influence
of the complex charge on the oxidation and reduction
potentials indicates that both processes are centred on the
metal rather than at a ligand as it is common for half-sand-
wich ruthenium complexes. Irreversibility of the reduction
step usually arises from ligand dissociation from a reactive
Ru(I) species [40]. When the scan is reversed following
reduction a new anodic peak appears. This feature is com-
mon to every thioether complex 2a–d and has a peak
potential of ꢁ0.58 V (see Figs. 5 and 6). A similar feature
with exactly the same peak potential is observed on the
Table 4
Voltammetry data for complexes 1, 2a–d, 3 (CH2Cl2/0.1 M NBu4PF6)
a
b
1=2
Compound
E
(V)
Ep,c (V)
Ep;a follow (V)
ox
2a
2b
2c
2d
3
0.85
0.85
0.845
0.83
1.44
1.04
ꢁ2.09
ꢁ0.58
ꢁ0.58
ꢁ0.58
ꢁ0.58
ꢁ0.50
ꢁ0.57
c
ꢁ2.02
ꢁ2.19
ꢁ1.57
1
ꢁ1.36, ꢁ1.98
Potentials are reported vs. the ferrocene/ferrocenium couple.
a
Peak potential of an irreversible process.
Potential of the irreversible anodic peak following reduction.
Reduction peak overlapped with the second irreversible multielectron
b
c
reduction wave of 1.
*
a
*
*
b
*
0.5 1.0
-1.5
-2.0 -2.5 V
1.0
-0.5 -1.0
Fig. 5. Cyclic voltammograms of complex 2b in CH2Cl2/NBu4PF6
(0.1 M) at RT and v = 0.1 V/s. (a) Oxidation scanned first; (b) reduction
scanned first. The peaks due to the reduction of 1 and to the oxidation of
the electroactive product formed from the radical anion of 1 are indicated
by a star symbol.

ˇ
J. Cubrilo et al. / Journal of Organometallic Chemistry 692 (2007) 1496–1504
1502
anodic reverse scan following the reduction of {(p-cym-
ene)RuCl2}2 (1). This latter feature has been assigned as
arising from {(p-cymene)RuCl}2, which in turn results
from 1 by a sequence of reduction and chloride dissociation
steps [40]. Such a coincidence may be fortuitous, but the
formation of a common product from the reduction of thi-
oether complexes 2a–d still suggests that decomposition of
the reduced forms involves dissociation of the SRR0 ligand.
Voltammograms of 2a–c (see Fig. 5) also displayed a
distinct reduction peak at ꢁ1.37 V. This peak was identi-
fied as arising from the reduction of {(p-cymene)RuCl2}2,
(1), by comparison with an authentic sample (see Fig. 2
of the Supplementary material) and by comparing voltam-
mograms before and after addition of small amounts of 1
to the solutions of 2a–d. Relative amounts of 2a–d and 1
present in the equilibrated electrolyte solution were esti-
mated as 11:1, 5:2, 7:1, and 12:1 by comparing the peak
currents of the partially reversible oxidation of 2a–d and
of the reduction peak of 1 in good agreement with the
ratios determined by NMR spectroscopy. As it is shown
in Figs. 5 and 6, the peak at ꢁ1.37 V of 1 present in the
equilibrated solutions is enhanced when the partially
reversible oxidation is scanned first. Thioether dissociation
thus also constitutes a likely degradation pathway for the
[(p-cymene)RuC2(SRR0)]+ radical cations formed during
the oxidation step. This is also in line with the observation
that complex 2b with PhC3H6SiPr as the sterically most
demanding and most weakly coordinated thioether gives
the least stable radical cation, i.e. the one with the smallest
ip,c/ip,a peak current ratio of the thioether complexes under
study here.
was independently synthesized [19]. We also verified clean
formation of 2d when 1 and 3 were combined in an
NMR tube and charged with CD2Cl2. This reaction pro-
vides another instance of the ready exchange of one
SMe2 ligand from bis(thioether) complex 3.
5. Experimental
NMR spectra were recorded on a Bruker AC 250 instru-
ment at 303 K. The spectra were referenced to the residual
protonated solvent (1H) or the solvent signal itself (13C).
For complexes 2a–d and 3 the assignment of 13C NMR
spectra was aided by a DEPT-135 measurement. UV–Vis
experiments were performed on an Omega 10 spectrometer
by Bruins Instruments in HELMA quartz cuvettes whith
1 cm optical path lengths. Elemental analysis (C, H, N)
was performed at in-house facilities. Voltammograms were
recorded on an EG&G 273 or a BAS CV50 potentiostat in
a home-built vacuum tight one-compartment cell using Pt
or glassy carbon disk electrodes from BAS. {(p-cym-
ene)RuCl2}2 (1) was prepared according to reference [41].
The new thioethers PhC2H4SMe and PhC3H6SiPr were pre-
pared from phenylethanthiol and methyl iodide or from
ipropylthiol and 3-phenylpropylbromide following an
established literature method [42] and were obtained in
74% and 76% yield following vacuum distillation. Spectro-
1
scopic data: (2-phenylpropyl)isopropyl sulfide: H NMR
3
(CDCl3): d (ppm) = 0.86 (6H, d, JHH = 6.69 Hz, CH3-
3
(iPr)), 1.92 (2H, q, JHH = 7.35 Hz, CH2CH2CH2), 2.92
3
3
(1H, hept, JHH = 6.80 Hz, CH(iPr)), 2.55 (2H, t, JHH
=
3
7.31 Hz, CH2Ph), 2.73 (2H, t, JHH = 7.58 CH2S), 7.24
(5H, m, CH(Ph)); 13C NMR (CDCl3): d(ppm) = 23.48
(CH3(iPr)), 29.95 (s, CH2), 32.42 (s, CH2Ph), 34.81 (s,
CH(iPr)), 35.01 (s, CH2S), 125.90 (s,p-CH), 128.38 (s, m-
CH), 128.52 (s, o-CH), 141.55 (s, Cquart); methyl(2-phenyl-
ethyl)sulfide: 1H NMR (CDCl3): d(ppm) = 2.13 (3H, s,
When increasing quantities of 1 were added to a solution
of bis(thioether) complex 3 the disappearance of the origi-
nal waves of 3 and the appearance of a new, partially
reversible couple at considerably lower oxidation potential
of +0.83 V was observed (Fig. 7). After addition of about
half an equivalent of 1 this new couple constituted the
prominent feature in the anodic regime. The product
formed under these conditions was readily identified as
the known mono(thioether) complex (p-cymene)RuCl2-
(SMe2) (2d) by comparison with authentic material which
3
CH3S), 2.76 (2H, t, JHH = 8.0 Hz, CH2Ph), 2.89 (2H, t,
3JHH = 8.0 Hz, CH2S), 7.26 (5H, m, CH(Ph)); 13C NMR
(CDCl3): d(ppm) = 15.63 (s, CH3S), 35.75 (s, CH2Ph),
35.85 (s, CH2S), 127.2 (s, p-CH), 128.15 (s, m-CH),
128.57 (s, o-CHPh), 140.13 (s, Cquart).
5.1. Synthesis of (p-cymene)RuCl2(PhC2H4SMe) (2a)
a
b
A Schlenk tube was charged with 0.300 g (0.49 mmol) of
{(p-cymene)RuCl2}2 and a 223 ll of PhC2H4SMe and the
mixture stirred at room temperature overnight in 4 ml of
CH2Cl2. The solvent was evaporated and the crude product
was washed with 2 · 5ml of Et2O. The orange powder
obtained after drying in vacuo was dissolved in 3 ml
CH2Cl2, cautiously layered with the same quantity of
Et2O and left to cool in a refrigerator. Compound 2a crys-
tallized as long orange needles. Yield 65.4% (0.147 g,
0.32 mmol). 1H NMR (CDCl3): d(ppm) = 1.26 (6H, d,
3JHH = 6.86 Hz, CH3(iPr)), 2.18, 2.24 (each s, 3H,
CH3(cym), CH3S), 2.90 (2H, m CH2Ph), 2.92 (2H, m,
1.6 1.4 1.2
1.0 0.8 0.6
0.4
Fig. 7. Cyclic voltammograms of complex 3 in CH2Cl2/NBu4PF6 (0.1 M)
at RT and v = 0.2 V/s before (a) and after (b) addition of half an
equivalent of 1.
3
CH2S), 3.03 (1H, hept, JHH = 6.86 Hz, CH(iPr)), 5.18,

ˇ
J. Cubrilo et al. / Journal of Organometallic Chemistry 692 (2007) 1496–1504
1503
5.77 (each 2H, d, 3JHH = 6.04 Hz, CH (cym)), 7.26 (5H, m,
CH(Ph)); 13C NMR (CDCl3): d(ppm) = 18.29, 19.70
(CH3(cym), CH3S), 22.20 (CH3(iPr)), 30.62 (CH(iPr)),
33.97 (CH2Ph), 41.09 (CH2S), 82.89 (CH(cym)), 83.93 (s,
CH(cym)), 99.30 (s, Cquart(cym)), 104.58 (s, Cquart(cym)),
126.78 (s, p-CH), 128.67 (s, m-CH), 128.72 (s, o-CH),
139.15 (s, Cquart). Anal. Calc. for C19H26Cl2RuS: C,
49.78; H, 5.72. Found: C, 49.73; H, 5.79.
130.87 (s,@CH2). Anal. Calc. for C14H22Cl2RuS: C,
46.41; H, 6.12. Found: C, 46.76; H, 6.14.
5.4. ½ðp-CymeneÞRuClðSMe2Þ ꢀþSbF 6ꢁ (3) and
2
½fðp-cymeneÞRug2ðl-ClÞ ꢀþSbF 6ꢁ (4)
3
Following the procedure of Dixneuf [19], 0.200 g (0.326
mmol) of {(p-cymene)RuCl2}2, 0.176 g (0.68 mmol) of
NaSbF6 and 190 ll (2.61 mmol) of dimethyl sulfide were
stirred with 8 ml of MeOH for two days at room tempera-
ture. The resulting yellow mixture was evaporated to dry-
ness and the residue was dissolved in 4 ml of CH2Cl2.
The solution was then filtered and the orange filtrate lay-
ered with 15 ml of Et2O. Orange crystals of 3 along with
a small quantity of 4 formed upon slow diffusion of ether
and were isolated by decantation of the solvent, dried in
vacuo and then manually separated (3: orange diamonds,
4 orange needles). Yields were 230 mg (55.9 %) of 3 and
21 mg (7.9 %) of 4. Compound 3: 1H NMR (CDCl3):
5.2. (p-Cymene)RuCl2(PhC3H6SiPr) (2b)
Ph(CH2)3SiPr (193 ll, 0.98 mmol) was added to 0.200 g
(0.326 mmol) of {(p-cymene)RuCl2}2 dissolved in 5 ml of
CH2Cl2. The reaction mixture was stirred for 14 h at room
temperature, then layered with 4 ml of EtOH and left 5
days at room temperature. A small amount of a microcrys-
talline solid of 1 was formed and removed by canula filtra-
tion. The filtered solution was dried in vacuo. The dry
residue was washed with 3 · 4 ml of Et2O and the Et2O
removed after washing by canula filtration. The residue
was then recrystallized two times from CH2Cl2/Et2O to
give 2b as orange microcrystals. Yield: 0.296 g (0.59 mmol,
90.5%). 1H NMR (CDCl3): d(ppm) = 1.27 (12H, m,
3
d(ppm) = 1.27 (6H, d, JHH = 7.00 Hz, CH3(iPr)), 2.21
(3H, s, CH3(cym)), 2.27 (6H, s, CH3S), 2.96 (2H, hept,
3
3JHH = 7.00 Hz, CH(iPr)), 5.20, 5.40 (each 2H, d, JHH
=
6.01 Hz, CH(cym)); 13C NMR (CDCl3): d(ppm) = 18.03,
3
CH3(iPr)), 1.88 (2H, q, JHH = 7.57 Hz, CH2CH2CH2),
18.90 (CH3S, CH3(cym)), 21.90 (CH3(iPr)), 31.44 (s,
2.18 (3H, s, CH3(cym)), 2.64 (4H, m, CH2), 2.93 (2H, m,
CH(iPr)), 78.81 (s, CH), 88.57 (s, CH), 101.83 (Cquart
-
3
CH(iPr)), 5.21, 5.38 (each 2H, d, JHH = 5.85 Hz,
(cym)), 111.62 (Cquart(cym)). Anal. Calc. for C14H16Cl-
F6RuS2Sb: C, 26.66; H, 4.15. Found: C, 26.85; H, 4.15.
CH(cym)), 7.21 (5H, m, CH(Ph)); 13C NMR (CDCl3):
d(ppm) = 18.90 (CH3(cym)), 22.14 (CH3(iPr)), 30.36
(CH2), 30.61 (CH2Ph), 31.20 (CH(iPr)), 34.89 (CH(iPr)),
37.46 (CH2S), 80.53, 81.29 (CH(cym)), 96.74, 101.23
(Cquart(cym)), 125.90 (p-CH(Ph)), 128.38 (s, m-CH(Ph)),
128.50 (s, o-CH(Ph)), 141.55 (s, Cquart(Ph)). Anal. Calc.
for C22H32Cl2RuS: C, 52.79; H, 6.44. Found: C, 53.01;
H, 6.55.
5.5. Compound 4
3
1H NMR (CDCl3): d(ppm) = 1.30 (6H, d, JHH
=
6.9 Hz, CH3(iPr)), 2.22 (3H, s, CH3(cym)), 2.78 (1H, hept,
3JHH = 6.9 Hz, CH(iPr)), 5.46, 5.64 (each 2H, d,
3JHH = 6.20 Hz, CH). 13C NMR (CDCl3): d(ppm) =
18.96 (CH3(cym)), 22.17 (CH3(iPr)), 30.63 (CH(iPr)),
80.55, 81.32 (CH(cym)), 96.77 (Cquart(cym)), 101.24
(Cquart(cym)). Anal. Calc for C20H28Cl3F6Ru2Sb: C,
29.55; H, 3.47. Found: C, 30.02; H, 3.55.
5.3. [(p-Cymene)RuCl2(SMeC3H5)RuCl2 (2c)
[{(p-Cymene)RuCl2}2] (0.120 g, 0.196 mmol) was dis-
solved in 6 ml of CH2Cl2 and 52 ll (0.591 mmol) of
SMeC3H5 was added dropwise. The reaction mixture was
stirred for 12 h at room temperature. The solvent was evap-
orated from the dark orange solution. The dry residue was
washed with 3 · 4 ml of Et2O and the Et2O removed in
vacuo. The resulting orange powder was dissolved in 3 ml
of CH2Cl2 and layered with 3 ml of ether. After 3 days
orange crystals were obtained. Yield 74.4% (0.115 g,
0.294 mmol). 1H NMR (CDCl3): d(ppm) = 1.24 (6H, d,
3JHH = 6.95 Hz, CH3(iPr)), 2.15, 217 (each 3H, s,
6. X-ray crystallography
X-ray-quality crystals were obtained as described in Sec-
tion 5. Crystals were removed from Schlenk tubes and
immediately covered with a layer of viscous hydrocarbon
oil (Paratone N, Exxon). A suitable crystal was selected,
attached to a glass fiber, and, in the case of 2c, instantly
placed in a low-temperature N2-stream [43]. All data were
collected at 293 K (2a, 3, 4) or 173 K (2c) using either a
Bruker-Nonius Kappa CCD (2a, 3, 4) or a rebuilt Syntex
P21/Siemens P3 (2c) diffractometer. Crystal data are given
in Table 3. Calculations were carried out with the SHELXTL
PC 5.03 [44] and SHELXL-97 [45] program system installed on
a local PC. The structures were solved by direct methods
and refined on F 2o by full-matrix least-squares refinement.
An absorption correction was applied by using semiempir-
ical w-scans (2c) or by a numerical absorption correction
(2a, 3, 4) [46]. In 2c the disorder in the allyl group of the
3
CH3(cym), CH3S), 2.92 (1H, hept, JHH = 6.95 Hz,
3
CH(iPr)), 3.36 (2H, d, JHH = 7.3 Hz, CH2S), 5.17 (2H, d
3
(br), JHH = 16 Hz, CH@CHH), 5.19 (d, br, CH@CHH,
3
3JHH = 10.8 Hz), 5.25, 5.41 (each 2H, d, JHH = 5.66 Hz,
3
CH(cym)), 5.71 (1H, ddt, CH@CHH, JHH = 16.0, 10.8,
7.3 Hz); 13C NMR (CDCl3): d(ppm) = 18.12, 18.90
(CH3S, CH3(cym)), 22.14 (CH3(iPr)), 30.42 (CH(iPr)),
40.69 (CH2S), 82.87 (CH(cym)), 83.46 (CH(cym)), 98.76
(Cquart(cym)), 104.42 (Cquart(cym)), 120.91 (s, @CH),

ˇ
J. Cubrilo et al. / Journal of Organometallic Chemistry 692 (2007) 1496–1504
1504
[14] C.J. Boxwell, P.J. Dyson, D.J. Ellis, T. Welton, J. Am. Chem. Soc.
124 (2002) 9334.
[15] J.F. Faller, B.J. Grimmond, Organometallics 20 (2001) 2454.
[16] D.L. Davies, J. Fawcett, S.A. Garratt, D.R. Russell, Organometallics
20 (2001) 3029.
[17] S. Ogo, T. Abura, Y. Watanabe, Organometallics 21 (2002) 2964.
[18] S. Ogo, K. Uehara, T. Abura, Y. Watanabe, S. Fukuzumi, Organo-
metallics 23 (2004) 3047.
[19] M. Gaye, B. Demerseman, P.H. Dixneuf, J. Organomet. Chem. 411
(1991) 263.
thioether ligand was modeled with split positions
(s.o.f. = 0.667/0.333 for C14/C14a) and DFIX constraints
for the C13–C14, and C13–C14a distances. Anisotropic
thermal parameters were included for all non-hydrogen
atoms with the exception of the two disordered carbon
atoms. H atoms were placed geometrically and refined
using a riding model, including free rotation of methyl
groups and variable isotropic displacement parameters.
Final R values are listed in Table 3. Important bond
parameters are given in Tables 1 and 2.
[20] Y. Yamamoto, S. Sakamoto, Y. Ohki, A. Usuzawa, M. Fujita, T.
Mochida, Dalton Trans. (2003) 3534.
[21] M.A. Bennett, L.Y. Goh, A.C. Willis, J. Am. Chem. Soc. 118 (1996)
4984.
[22] R.Y.C. Shin, M.A. Bennett, L.Y. Goh, W. Chen, D.C.R.
Hockless, W.K. Leong, K. Mashima, A.C. Willis, Inorg. Chem.
42 (2003) 96.
[23] R.Y.C. Shin, S.Y. Ng, G.K. Tan, L.L. Koh, S.B. Khoo, L.Y. Goh,
R.D. Webster, Organometallics 23 (2004) 547.
[24] G. Tresoldi, L. Baradello, S. Lanza, P. Cardiano, Eur. J. Inorg.
Chem. (2005) 2423.
Acknowledgements
Financial support of this work by Deutsche Forschungs-
gemeinschaft (Grant Wi1262/4-1) is gratefully acknowl-
edged. RFW also wishes to thank Dr. Reiner Ruf for his
help with the structure refinement of complex 3.
[25] P.D. Smith, A.H. Wright, J. Orgnanomet. Chem. 559 (1998) 141.
[26] M.A. Bennett, A.J. Edwards, J.R. Harper, T. Khimyak, A.C. Willis,
J. Organomet. Chem. 629 (2001) 7.
Appendix A. Supplementary material
[27] K. Umezawa-Vizzini, I.Y. Guzman-Jimenez, K.H. Whitmire, T.R.
Lee, Organometallics 22 (2003) 3059.
[28] K. Umezawa-Vizzini, T.R. Lee, Organometallics 22 (2003) 3066.
[29] K. Umezawa-Vizzini, T.R. Lee, Organometallics 23 (2004) 1448.
[30] L. Carter, D.L. Davies, J. Fawcett, D.R. Russell, Polyhedron 12
(1993) 1123.
[31] R. Aronson, M.R. Elsegood, J.W. Steed, D.A. Tocher, Polyhedron
10 (1991) 1727.
[32] G.A. Heath, G. Hefter, D.R. Robertson, W.J. Sime, T.A. Stephen-
son, J. Organomet. Chem. 152 (1978) C1.
[33] F. Greponi, D. Braga, P.J. Dyson, B.F.G. Johnson, F.M. Sanderson,
M.J. Calhorda, L.F. Veiros, Organometallics 14 (1995) 121.
[34] D.S. Pandey, A.N. Sahay, O.S. Sisoida, N.K. Jha, P. Sharma, H.E.
Klaus, A. Caberra, J. Organomet. Chem. 592 (1999) 278.
[35] J. Soleimannejad, C. White, Organometallics 24 (2005) 2538.
[36] D.A. Tocher, M.D. Walkinshaw, Acta Crystallogr., Sect. B 38 (1982)
3083.
CCDC 611182, 611183, 611184 and 611185 contain the
supplementary crystallographic data for 2a, 2c, 3 and 4.
These data can be obtained free of charge via htpp://
www.ccdc.cam.ac.uk/conts/retrieving.html, or from the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or
e-mail: deposit@ccdc.cam.ac.uk. Supplementary data asso-
ciated with this article can be found, in the online version,
at doi:10.1016/j.jorganchem.2006.11.046.
References
[1] A. Furstner, M. Picquet, C. Bruneau, P.H. Dixneuf, Chem. Commun.
¨
(1998) 1315.
[37] B. Therrien, L. Vieille-Petit, J. Jeanette-Gris, P. Stepnicka, G. Suss-
¨
[2] A. Fu¨rstner, L. Ackermann, Chem. Commun. (1999) 95.
Fink, J. Organomet. Chem. 689 (2004) 2456.
[3] A. Furstner, M. Liebl, C.W. Lehmann, M. Picquet, R. Kunz, C.
¨
ˇ
[38] J. Cubrilo, I. Hartenbach, T. Schleid, R.F. Winter, Z. Anorg. Allg.
Bruneau, D. Touchard, P.H. Dixneuf, Chem. Eur. J. 6 (2000) 1847.
Chem. 632 (2006) 400.
´
[4] R. Castarlenas, D. Semeril, A.F. Noels, A. Demonceau, P.H.
[39] G. Bhalla, J. Oxgaard, W.A. Goddard III, R. Periana, Organomet-
allics 24 (2005) 5499.
Dixneuf, J. Organomet. Chem. 663 (2002) 235.
[5] R. Castarlenas, P.H. Dixneuf, Angew. Chem. 115 (2003) 4662.
[40] P. Lahuerta, J. Latorre, J. Soto, Y. Mugnier, Polyhedron 8 (1989)
2803.
´
[6] M. Bassetti, F. Centola, D. Semeril, C. Bruneau, P.K. Dixneuf,
Organometallics 22 (2003) 4459.
[41] M.A. Bennett, T.-N. Huang, T.W. Matheson, A.K. Smith, S. Ittel,
W. Nickerson, in: J.F. Fackler (Ed.), Inorg. Synth. 21 (1974) 75.
[42] P. Cagniant, G. Jecko, D. Cagniant, Bull. Soc. Chim. Fr. (1960)
1798.
[7] F. Simal, A. Demonceau, A.F. Noels, Angew. Chem., Int. Ed. Engl.
38 (1999) 538.
[8] F. Simal, S. Sebille, L. Hallet, A. Demonceau, A.F. Noels, Macormol.
Symp. 161 (2000) 73.
[43] H. Hope, Prog. Inorg. Chem. 41 (1995) 1.
[44] SHELXTL PC 5.03, 5.03, Siemens Analytical X-Ray Instruments Inc.,
Madison, WI, 1994.
´
[9] V. Cadierno, P. Crochet, S.E. Garcıa-Carrido, J. Gimeno, Dalton
Trans. (2004) 3635.
[10] K.-J. Haack, S. Hashiguchi, A. Fujii, T. Ikariya, R. Noyori, Angew.
Chem., Int. Ed. Engl. 36 (1996) 285.
[45] G.M. Sheldrick, SHELX-97, Program for Crystal Structure Solution
and Refinement, Institut fur Anorganische Chemie der Universita¨t,
¨
[11] R. Noyori, S. Hashiguchi, Acc. Chem. Res. 31 (1997) 97.
[12] M. Yamakawa, I. Yamada, R. Noyori, Angew. Chem., Int. Ed. Engl.
40 (2001) 2818.
Tammanstraße 4, Go¨ttingen, 1997.
[46] W. Herrendorf, H. Ba¨rnighausen, Program HABITUS, Karlsruhe,
Germany, 1993.
[13] R. Noyori, Angew. Chem., Int. Ed. Engl. 41 (2002) 2008.