Diruthenium(I,I) saccharinate complexes: Synthesis, molecular structure, and evaluation as catalysts for carbenoid reactions of diazoacetates

Article abstract of DOI:10.1016/j.jorganchem.2006.02.013

The dinuclear ruthenium complexes [Ru₂(μ-sac)₂(CO)₆] (1), [Ru₂(μ-sac)₂(CH₃CN)₂(CO)₄] (3), [Ru₂(μ-sac)₂(CO)₅(PPh₃)] (4) and [Ru₂(μ-sac)₂(CO)₄(PPh₃)₂] (5) as well as the tetranuclear ruthenium complex [Ru₂(μ-sac)₂(CO)₅]₂ (2) (sac = saccharinate, C₇H₄NO₃S^-) were synthesized starting from Ru₃(CO)₁₂ and saccharin. X-ray crystal structure analysis of 1, 3A × p-xylene, 4 × CH₂Cl₂ and 5 × 3CH₂Cl₂ showed that the Ru₂^{2 +} core is bridged through the amidate moieties of the two saccharinate ligands, with a head-tail arrangement in complexes 1, 3A and 5, and a head-head arrangement in 4. For complex 3, an equilibrium mixture of the head-head regioisomer 3A and a second species 3b exists in solution. Complexes 1 and 2 are suitable catalysts for the cyclopropanation of nucleophilic alkenes (styrene, cyclohexene and 2-methyl-2-butene) with methyl diazoacetate.

Full text of DOI:10.1016/j.jorganchem.2006.02.013

Journal of Organometallic Chemistry 691 (2006) 2774–2784
www.elsevier.com/locate/jorganchem
Diruthenium(I,I) saccharinate complexes:
Synthesis, molecular structure, and evaluation as catalysts for
carbenoid reactions of diazoacetates
*
Stefan Buck, Gerhard Maas
Abteilung Organische Chemie I, Universita¨t Ulm, Albert-Einstein-Allee 11, D-89081 Ulm, Germany
Received 31 January 2006; accepted 10 February 2006
Available online 6 March 2006
Abstract
The dinuclear ruthenium complexes [Ru2(l-sac)2(CO)6] (1), [Ru2(l-sac)2(CH3CN)2(CO)4] (3), [Ru2(l-sac)2(CO)5(PPh3)] (4) and
[Ru2(l-sac)2(CO)4(PPh3)2] (5) as well as the tetranuclear ruthenium complex [Ru2(l-sac)2(CO)5]2 (2) (sac = saccharinate, C7H4NO3Sꢀ)
were synthesized starting from Ru3(CO)12 and saccharin. X-ray crystal structure analysis of 1, 3A · p-xylene, 4 · CH2Cl2 and
5 · 3CH2Cl2 showed that the Ru22þ core is bridged through the amidate moieties of the two saccharinate ligands, with a head–tail
arrangement in complexes 1, 3A and 5, and a head–head arrangement in 4. For complex 3, an equilibrium mixture of the head–head
regioisomer 3A and a second species 3b exists in solution. Complexes 1 and 2 are suitable catalysts for the cyclopropanation of nucle-
ophilic alkenes (styrene, cyclohexene and 2-methyl-2-butene) with methyl diazoacetate.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Ruthenium(I) complexes; Saccharin; Catalysts; Cyclopropanation; Diazo compounds
1. Introduction
compounds constitute dinuclear rhodium(II) catalysts of
the type Rh2L4, where L represents a bridging anionic
ligand such as carboxylate, amidate, phosphate, or C6H4–
PPh2 [1]. Inspired by the success and versatility of these
rhodium complexes, we have investigated structurally
related dinuclear ruthenium complexes with the Ru22þ core
and the general composition ½Ru2ðl-L12ÞðCOÞ4L2L3ꢁ, where
L1 is a bidentate bridging ligand such as carboxylate
[10,11], triazenide [12] and pyridin-2-olate [11] while L2
and L3 represent axial ligands at one or both ruthenium
atoms. Up to present, the anion of saccharin (o-sulfobenzi-
mide, 1,2-benzisothiazol-3(2H)-one) has not been consid-
ered as a bidentate ligand in catalytically active dinuclear
Rh or Ru complexes, although saccharin features an ami-
date-like moiety that is reminiscent of several excellent
tetrakis(amidato)-dirhodium catalysts. In fact, the multi-
functional saccharinate anion (sac) can engage in several
different coordination modes [13], of which the simple N-
coordination is the most common one [14]. Notably, com-
plexes in which saccharinate acts as a bidentate ligand via
the amidate (N–C@O) fragment have been reported only
Carbenoid reactions of aliphatic diazo compounds
represent valuable tools in synthetic organic chemistry
[1,2]. These transformations are catalyzed effectively and
efficiently by a variety of catalysts based in particular on
copper and rhodium. Recently, ruthenium complexes, with
different oxidation states of the metal and a variety of
ligand systems, have emerged as novel catalysts for diazo
decomposition and carbene transfer reactions. While the
majority of studies so far have focused on alkene cyclo-
propanation reactions [3], increasing attention is now given
to other carbenoid reactions, e.g., insertion into C–H [4]
and X–H [5] bonds, formation of ylides (carbonyl ylides
[6], P-ylides with subsequent aldehyde olefination [7], and
N-ylides [8]), and formation of carbene dimers [9].
Almost all of the rhodium catalysts which have been
successfully applied to carbenoid transformations of diazo
*
Corresponding author.
E-mail address: gerhard.maas@uni-ulm.de (G. Maas).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.02.013

S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
2775
occasionally. This coordination motif has been found in
mononuclear lead(II) complexes [15] and dinuclear copper
complexes [13a,16]. Dinuclear chromium complexes of the
type [Cr2(sac)4] or [Cr2(sac)4L2] [17], where four sacchari-
nate ligands bridge a Cr(II)–Cr(II) core, represent the clos-
est structural analogy to the ruthenium complexes which
are described in this communication.
equilibrium mixture of two species as indicated by the
NMR spectra. Due to incomplete signal separation of the
1H signals, the ratio could be determined only approxi-
mately as A:B ꢂ 2.9:1 at 295 K, and it appeared that this
ratio changed only slightly between 243 and 328 K. Crys-
tallization of 3 from CH3CN/p-xylene for an XRD analysis
furnished the head–tail regioisomer 3A. It could not be
clarified so far whether the minor species 3B, which was
detected in solution and which has 13C and 1H NMR
chemical shifts quite similar to those of 3A (Dd less than
0.8 ppm), is the head–head isomer or a different species.
Compound 3 was also obtained by heating of hexacarbonyl
complex 1 in acetonitrile solution. Efforts to separate 3A
and 3B by chromatography on silica gel were fruitless:
Although two fractions with different Rf values could be
isolated, their NMR spectra were identical with those
before chromatography.
Short treatment of 2 with triphenylphosphane led to a
mixture of the yellow complex [Ru2(l-sac)2(CO)5(PPh3)]
(4) and the orange complex [Ru2(l-sac)2(CO)4(PPh3)2]
(5). Crystallization from dichloromethane/pentane yielded
a mixture of yellow and orange crystals which could be sep-
arated manually. XRD analysis established the different
constitution of both compounds and showed that the
head–head arrangement of the two sac ligands is present
in 4, while 5 constitutes the head–tail regioisomer. A 31P
NMR control experiment showed that 4 (dP = 22.0 ppm)
was slowly converted into 5 (dP = 14.1 ppm) in the pres-
ence of PPh3: starting from equimolar amounts of 4 and
PPh3 in CDCl3, the 4:5 ratio was 5:1 after 1 h and 4:1 after
18 h; combination of 4 with two equivalents of PPh3 fur-
nished a 0.4:1 mixture of 4 and 5 after 1 h. A mixture of
4 and 5 was also obtained when hexacarbonyl complex 1
was treated with PPh3 (Scheme 1).
For related 6-halogenopyridonate complexes, we have
recently observed that the interconversion of head–head
and head–tail regioisomers occurs quite easily and that
the result depends on the axial ligands at the two ruthe-
nium centers [22]. The isolation of 4, with one axial CO
ligand, indirectly confirms the proposed structure of the
dimeric complex 2 and excludes a polymeric catena struc-
ture of the type [Ru2(l-L)2(CO)4]n (i.e., Ru–O coordina-
tion at both ruthenium centers) which has been found
with bridging carboxylate as well as pyridonate ligands
[23].
Complexes 1–5 were also characterized spectroscopi-
cally (Table 1). In the IR spectra, the carbonyl absorptions
of 3 and 5 show the typical pattern of a sawhorse-shaped
M2(CO)4 unit (see, e.g., literature [19,20,24,25a]), while
the hexacarbonyl and pentacarbonyl complexes (1, 2, 4)
show a more complex absorption pattern. Absorptions at
high wavenumbers (2083–2107 cmꢀ1) are observed in the
complexes containing axial carbonyl ligands.
2. Results and discussion
2.1. Preparation of ruthenium complexes 1–5
Following the established method for the preparation of
other dinuclear ruthenium(I,I) complexes, the ruthenium
saccharinate complexes were prepared from Ru3(CO)12
and the protonated form of the briding ligand (Hsac).
When Ru3(CO)12 and saccharin (1:3 stoichiometry) were
heated in toluene at 90 ꢁC for 36 h, the complex [Ru2(l-
sac)2(CO)6] (1) was obtained as an air- and moisture-stable
pale yellow solid in 80% yield (Scheme 1). The compound is
fairly soluble in noncoordinating organic solvents such as
dichloromethane. The composition of 1 is in agreement
with the elemental analysis, and the head–tail arrangement
of the two sac ligands (1,1 regioisomer) as well as the pres-
ence of an axial CO ligand at each ruthenium atom is
uncovered by the results of an XRD structure determina-
tion (see Section 2.2).
It should be noted that the thermal stability of com-
plexes of the type [Ru2(l-L2)2(CO)6] varies considerably:
While 1 as well as related triazenide [12] and pyrazolate
[18] complexes can be isolated as stable complexes, the
pyridonate ([Ru2(l-Opy)2(CO)6] [19]) and carboxylate
complexes ([Ru2(l-OOCR)2(CO)6] [20]) readily lose their
axial CO ligands in the absence of a CO atmosphere. On
the other hand, complex 1 splits off carbon monoxide when
kept in boiling toluene (110 ꢁC) and is transformed into an
air- and moisture-stable yellow powder. The same com-
pound was obtained in 89% yield when Ru3(CO)12 and sac-
charin were heated in boiling toluene for 5 h. The product
was completely insoluble in pentane or cyclohexane and
only barely soluble in chloroalkanes. Although no final
structural proof could be obtained so far, we assume that
the product represents the tetranuclear ruthenium complex
[Ru2(l-sac)2(CO)5]2 (2), a coordination dimer of a dinu-
clear ruthenium complex. Recently, we have established
the structure of an analogous complex with fluoropyrido-
nate rather than saccharinate ligands [21]. Obviously, the
thermal conversion of 1 into 2 proceeded by expulsion of
one axial CO ligand, rearrangement from the head–tail
(0,2) to the head–head (1,1) regioisomer and dimerization
across a Ru–O bond. Steric interactions with the SO2
group of the second sac ligand are likely to prevent the
analogous dimerization of the head–tail regioisomer.
As expected, suitable Lewis bases were able to cleave the
coordination dimer 2 (Scheme 1). Thus, exposure of 2 to
acetonitrile furnished the yellow complex [Ru2(l-
sac)2(CH3CN)2(CO)4] (3) which in solution existed as an
For characterization by mass spectrometry, MALDI-
TOF spectra gave unsatisfactory results. In contrast,
clean ESI mass spectra were obtained, but the intact com-
plexes could not be observed. The peaks of highest intensity

2776
S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
O
+
Ru3(CO)12
3
NH
S
O2
toluene
90 ˚C, 36 h
toluene
110 ˚C, 5 h
CO
Ru
CO
Ru
N
SO2
O
N
CO
O2S
N
O2S
O
N
O
toluene
110 ˚C, 4 h
OC
CO
OC
SO2
CO
Ru
O
N
Ru
Ru
CO
OC
OC OC
Ru
O
OC
SO2
N
O
O2S
CO
CO
CO CO
1
2
CH3CN,
50 oC
CH3CN,
rfx
O2S
O
N
3B
SO2
CCH3
O
N
N
H3CC N
OC
Ru Ru
OC
CO
CO
3A
2 PPh3
CH2Cl2, r.t., 15 min
1
2
O2S
O2S
+
O
N
O
N
SO2
CO
N
PPh3
CO
PPh3
O
Ru
OC
Ru
Ph3P Ru
OC
Ru
N
O
OC
4 PPh3
CH2Cl2, r.t., 15 min
O2S
CO CO
CO CO
4
5
Scheme 1.
correspond to the fragment obtained after cleavage of the
axial CO ligand(s) in 1 and 4, while dimers of these frag-
ments can also be observed with low intensities (Table 1).
In the case of 2, the base peak at m/z = 679 corresponds
to the monomeric unit (M/2ꢀCO) of the postulated dimer;
a fragment corresponding to a loss of one CO ligand from
the dimeric unit appears with low intensity (MꢀCO+Na).
The thermogravimetric analysis (TGA) of 1 and 2 shows
a correlation with the ESI-MS behavior. For 1, the TGA
curve at 210–245 ꢁC shows a mass loss corresponding to
two CO molecules. For 2, a mass loss is observed in the
range 188–282 ꢁC which corresponds to a loss of two CO
molecules from the tetranuclear complex or of one CO mol-
ecule from one half of the coordination dimer.
The 13C NMR chemical shifts of 1–5 are also given in
Table 1. The axial CO ligands can be clearly distinguished

Table 1
Selected IR, MS and NMR data of complexes 1–5
1
2
3
4
5
IRa: m(CO) [cmꢀ1
MS (ESI)b, m/z
]
2107 vs, 2083 vs, 2075 s,
2042 s, 2025 vs, 2007 vs
2099 vs, 2042 vs, 2021 vs,
2011 s, 1948 s
2046 vs, 1996 m, 1956 vs
2086 s, 2034 vs, 1998 s,
1963 m, 1935 w
2035 s, 1996 m, 1965 s
703 (100%,
679 (100%, M/2ꢀCO),
702 (M/2)ꢀCO+Na),
1432 (MꢀCO+2Na)
–
965 (100%,
–
Mꢀ2CO+Na), 1381
MꢀCO+Na), 1904
(2 · (Mꢀ2CO)+Na)
(2 · (MꢀCO)+Na)
13C NMRc
d(COeq
)
195.8, 197.3
197.3, 201.8
Ad: 200.13, 201.19
Bd: 199.36, 201.53
198.1 (d, JC,P = 2.9),
203.2 (d, JC,P = 5.1)
202.5 (vte, J = 3.3), 204.0
(vt, J = 4.8)
3
d(COax
d(sac)
)
179.05
175.5
181.1 (d, JC,P = 32.9)
121.1 (CH), 124.6 (CH),
128.5 (C), 133.8 (CH),
134.0 (CH), 139.9 (C),
178.95 (NCO)
122.0, 124.4, 129.6,
133.9, 134.7, 142.8, 176.7
(NCO)
A: 120.21, 124.10, 129.15, 133.06,
133.17, 142.30, 175.85 (NCO)
B: 120.53, 123.77, 129.09, 132.78,
133.10, 142.68, 176.30 (NCO)
A: 3.57 (CH3CN), 122.97 (CN)
B: 4.16 (CH3CN)
178.9 (d, JC,P = 8.8)
(NCO)
178.6 (vt, J = 6.2)
(NCO)
Other signals
31P NMRc
22.0
14.1
a
KBr pellets; vs = very strong, s = strong, m = medium, w = weak.
b
c
Only the highest peak of the multiplet of isotope peaks is given, assignment in parentheses (M = molecular mass).
Solvent: C2D2Cl4 for 1, [D8]-THF for 2, CDCl3 for 3–5.
A mixture of two isomers was obtained, 3A:3B ꢂ 2.5:1.
d
e
vt = virtual triplet.

2778
S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
from the equatorial ones. On the other hand, the 13C data
do not allow a systematic distinction between head–head
and head–tail isomers. The spectra of 3, where both iso-
mers are present, illustrate that the chemical shifts of all
corresponding signals are very close. Due to the low solu-
bility of 2 in non-donor solvents, the 13C spectra of this
complex was recorded in [D8]-THF, and therefore, the data
most likely represent those of [Ru2(l-sac)2(CO)5(thf)].
2.2. Solid-state structures of 1, 3A, 4 and 5
The structures of 1, 3A, 4, and 5 were established by X-
ray single crystal structure analysis (Figs. 1–4). In three
cases, solvent of crystallization was incorporated in the
crystal: 3A · p-xylene, 4 · CH2Cl2 and 5 · 3CH2Cl2. In
all cases, a ‘‘sawhorse’’ arrangement, with a cis relationship
of the two saccharinate ligands at each metal center, is
observed which is characteristic for dinuclear complexes
of the general composition [Ru2(CO)4(l-L)2] (for examples,
see literature [12,18,19,21,23,25]). The saccharinate units
act as bidentate ligands via their amidate (N–C@O) moi-
ety. In 1, 3A and 5 they bridge the two ruthenium atoms
in a head–tail (or 1,1) fashion. The axial positions at both
metal atoms are occupied by CO (in 1), acetonitrile (in 3A),
or PPh3 (in 5) ligands. Thus, these dinuclear complexes
have an overall C2 topology which in the case of 3A coin-
cides with a crystallographic C2-axis. Compound 3A crys-
tallizes with one p-xylene molecule per formula unit; the
Fig. 2. Structure of 3A in the solid state; ellipsoids of thermal vibration
are shown at 50% probability. The p-xylene solvate molecule is not shown.
0
˚
Selected bond lengths (A) and angles (ꢁ): Ru–Ru 2.7113(4), Ru–N1
2.154(2), Ru–O30 2.136(2), Ru–N2 2.149(2), Ru–C8 1.862(3), Ru–C9
1.841(3), C8–O4 1.137(4), C9–O5 1.143(3), N1–C7 1.347(3), C7–O3
1.248(3), S–O1 1.433(2), S–O2 1.429(2), N2–O10 1.130(3); N1–Ru–O30
84.46(8), Ru0–Ru–N1 81.24(5), Ru0–Ru–O30 86.50(4), Ru0–Ru–N2
167.51(6), Ru–N2–C10 179.5(2); N1–Ru–Ru0–O30 ꢀ11.6(2), C9–Ru–
Ru0–C80 ꢀ11.8(2). Atoms marked with a prime are generated by the
symmetry operation: ꢀx + 1, y, ꢀz + 0.5.
xylene molecule also has crystallographic C2 symmetry
with both methyl carbons positioned on the rotation axis
(i.e., the methyl hydrogen atoms are positionally disor-
dered over two symmetry-related sites).
Complex 4, in contrast to 1, 3A and 5, has the head–
head arrangement of the two sac ligands, and the sterically
less accessible axial coordination site (neighbored by two
SO2 groups) is occupied by the small CO ligand. The tor-
sion angles around the Ru–Ru axis show that the in-plane
rotation of the equatorial coordination plane containing
Ru2 relative to that containing Ru1 is larger than in the
head–tail complexes.
In three of the four complexes, remarkably long Ru–Ru
˚
˚
distances are found (1: 2.7713(9) A; 3A: 2.7113(4) A;
˚
˚
4: 2.7754(5) A; 5: 2.7725(5) and 2.7747(5) A) which are lar-
ger than in similar dinuclear Ru(I,I) complexes with ami-
date or pyridin-2-olate bidentate bridging ligands and
various axial ligands, such as in the following examples:
2.688(1) A in [Ru2(l-HNC(Ph)O)2(CH3CN)2(CO)2] [25a],
2.609–2.710 A in complexes with pyridin-2-olate ligands
˚
˚
[21,26,27]. A comparison of the Ru–Ru distance in 1 and
3 shows that the strongly coordinating axial CO ligands
cause a significant bond lengthening as compared to aceto-
nitrile ligands. It is also informative to compare the metal–
metal distance in 1 with that in other diruthenium(I,I)
hexacarbonyl complexes such as [Ru2(l-OOCPh)2(CO)6]
Fig. 1. Structure of 1 in the solid state; ellipsoids of thermal vibration are
shown at 30% probability. Selected bond lengths (A) and angles (ꢁ): Ru1–
˚
Ru2 2.7713(9), Ru1–N2 2.156(6), Ru1–O7 2.137(5), Ru1–C1 2.002(9),
Ru1–C2 1.889(9), Ru1–C3 1.891(9), C1–O1 1.117(10), C2–O2 1.119(10),
C3–O3 1.141(10), Ru2–N1 2.134(6), Ru2–O10 2.126(5), Ru2–C4 1.989(8),
Ru2–C5 1.868(9), Ru2–C6 1.850(8), C4–O4 1.117(9), C5–O5 1.170(9), C6–
O6 1.153(9), O7–C7 1.259(8), C7–N1 1.325(10), O10–C14 1.264(9), C14–
N2 1.319(9); N2–Ru1–O7 86.3(2), N1–Ru2–O10 86.9(2), Ru2–Ru1–C1
173.3(2), Ru1–C1–O1 169.6(8), Ru1–Ru2–C4 174.7(2), Ru2–C4–O4
171.8(7); N2–Ru1–Ru2–O10 ꢀ9.3(2), N1–Ru2–Ru1–O7 –8.6(2), C2–
Ru1–Ru2–C5 ꢀ7.4(3).
˚
(2.704(1) A) [23b], [Ru2(l-3,5-dimethylpyrazolate)2(CO)6]
˚
(2.705(2) A) [18], and [Ru2(l-ArNNNAr)2(CO)6] (Ar =
˚
4-chlorophenyl) (2.6701(6) A) [12]. It is evident that
the saccharinate ligands contribute considerably to the
stretching of the Ru–Ru bond. Cotton et al. [17] have made

S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
2779
˚
Fig. 3. Structure of 4 · CH2Cl2 in the solid state; ellipsoids of thermal vibration are shown at 30% probability. Selected bond lengths (A) and angles (ꢁ):
Ru1–Ru2 2.7754(5), Ru1–N1 2.149(2), Ru1–N2 2.119(2), Ru1–C1 2.003(3), Ru1–C2 1.878(3), Ru1–C3 1.872(3), C1–O1 1.115(4), C2–O2 1.141(4), C3–O3
1.135(4), Ru2–O6 2.128(2), Ru2–O9 2.162(2), Ru2–C4 1.842(3), Ru2–C5 1.842(3), Ru2–P1 2.431(1); Ru2–Ru1–C1 174.79(9), N1–Ru1–Ru2 79.67(6),
Ru1–Ru2–P1 173.12(2), O6–Ru2–O9 88.45(8); N1–Ru1–Ru2–O6 15.42(8), N2–Ru1–Ru2–O9 16.19(8), C2–Ru1–Ru2–C4 21.25(13). Closest intermolec-
˚
˚
ular contact of the dichloromethane molecule: C38–O7 3.284(5) A, H38A–O7 2.67 A, C38–H38a–O7 120.7ꢁ.
Fig. 4. Structure of 5 in the solid state; ellipsoids of thermal vibration are shown at 30% probability. Only one of the two molecules in the asymmetric unit
˚
is shown and the dichloromethane solvate molecules as well as the hydrogen atoms are omitted. Selected bond lengths (A) and angles (ꢁ): Ru1–Ru2
2.7725(5), Ru1–O1 2.148(3), Ru1–N2 2.166(4), Ru1–P1 2.462(1), Ru1–C15 1.861(5), Ru1–C16 1.834(6), C15–O7 1.148(6), C16–O8 1.153(6), N2–C8
1.339(5), N2–S2 1.656(4), C8–O4 1.246(6), Ru2–O4 2.146(3), Ru2–N1 2.163(4), Ru2–P2 2.470(1), Ru2–C17 1.852(6), Ru2–C18 1.840(5), C17–O9 1.152(6),
C18–O10 1.159(6), N1–C1 1.345(6), N1–S1 1.666(4), C1–O1 1.266(6); Ru2–Ru1–P1 169.96(3), Ru2–Ru1–N2 81.21(8), Ru2–Ru1–O1 82.40(7), O1–Ru1–
N2 88.45(2); O1–Ru1–Ru2–N1 19.2(2), Ru1–Ru2–P2 169.16(3), N2–Ru1–Ru2–O4 19.0(2), C15–Ru1–Ru2–C18 21.6(2).
similar observations on the dinuclear chromium complexes
[Cr2(sac)4L2] (L = thf, pyridine) and have suggested that
the effect could be related in part to the high acidity of
the parent saccharin ligand.
2.3. Ru-catalyzed carbene transfer reactions
Complexes 1 and 2 were tested for their ability to cata-
lyze carbene transfer from methyl diazoacetate (MDA) to

2780
S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
R3
R1
2. It was found that for the three alkenes, the yields of
R1
R3
R2
H
N2=CHCO2Me
catalyst
R1
R3
R2
H
H
H
H
cyclopropanation catalyzed by 1 or 2 are in general a little
lower than with [Ru2(l-OAc)2(CO)4]n (10, entry 4) under
the same conditions [10a], but they compare favorably with
other catalysts of this type bearing carboxylate [10,11], tria-
zenide [12] and 2-pyridonate [11] ligands. Initially, we had
expected that the presence of the SO2 group in the ligands
would enhance the electrophilicity of catalysts 1 and 2, with
a concomitant increased ability for the diazo decomposi-
tion reaction and increased efficiency and effectiveness as
cyclopropanation catalysts. These expectations are not
met by the results shown in Table 2 and disagree with
the considerably longer reaction times for the more highly
substituted alkenes, which due to their nucleophilicity
should be more susceptible to reaction with an electrophilic
metal–carbene intermediate (LnRu@CHCO2Me). The rela-
tive slowness of the reactions could be caused by catalyst
deactivation (competitive coordination of olefin or reaction
products instead of the diazo compound), but an increase
of the amount of catalyst 2 from 1 to 3 mol% (entries 2
and 3) did not lead to major improvement. On the other
hand, when the amount of 2-methyl-2-butene was reduced
by a factor of 10 (i.e., equimolar amounts of diazoacetate
and olefin were involved in the reaction), the reaction time
for complete consumption of MDA after its addition to the
alkene/catalyst phase decreased from 6 to 2 h (catalyst 2),
but the yield of cyclopropanes E/Z-6 went down to 20%
while the yield of E/Z-7 increased to 37%. This indicates
that the diazoester competes strongly with the olefin for
the ruthenium–carbene intermediate. The diastereoselectiv-
ities of the cyclopropanation reactions are in the expected
ranges [10–12], but the increased Z- (or syn-) selectivity
with catalyst 1, similar to related ruthenium triazenide
complexes [12], merits attention.
+
R2
-N2
CO2Me
CO2Me
R1 = Ph; R2, R3 = H
R1, R2 = -(CH2)4-, R3 = H
R1, R2, R3 = Me
(Z)-6
(E)-6
H
+
CO2Me
MeO2C
(E/Z)-7
CO2Me
CH2CO2Me
+
OMe
N2=CHCO2Me
catalyst
+
(E/Z)-7
-N2
MeO
MeO
8
9
cat. 2 (0.5 mol%): 13%
cat. 10 (1 mol%): 24%
3%
84%
Scheme 2.
olefins and arenes (Scheme 2). Styrene, cyclohexene, and 2-
methyl-2-butene were chosen as the olefinic substrates
because they represent different electronic and steric condi-
tions and the results can be compared with our preceding
investigations using related ruthenium catalysts. The reac-
tions were carried out by slow addition of MDA to a large
excess of the liquid olefin diluted with CH2Cl2 and contain-
ing the catalyst. While complex 1 was well soluble in the
alkene-CH2Cl2 phase, complex 2 dissolved completely only
after addition of a small amount of MDA, indicating the
cleavage of the coordination dimer of 2 by the diazoester.
After the addition of MDA was completed, it took several
hours before all the diazoester had been consumed. While
the reaction times increased from styrene to 2-methyl-2-
butene, a clear-cut and general difference between the per-
formance of catalysts 1 and 2 could not be observed. Yields
of cyclopropanes E/Z-6 and the formal carbene dimers
dimethyl fumarate and maleate (E/Z-7) are given in Table
Similar to complex 10 [10a], the saccharinato-ruthe-
nium complexes reported here are not suited for profi-
cient intermolecular carbene transfer to arenes. Thus,
decomposition of MDA in neat benzene catalyzed by 2
does not give a carbene addition product. When the
more electron-rich anisole was used as substrate and sol-
vent, the cycloheptatrienecarboxylate 8 and the isomeric
(methoxyphenyl)acetate 9 were obtained in a combined
yield of 16% (Scheme 2), somewhat less than with 10
as catalyst. The low yield may be due in part to the
Table 2
a
Cyclopropanation of alkenes with methyl diazoacetate (MDA) in CH2Cl2
Entry
Catalyst
Catalyst loading (mol%)
Yields of cyclopropanes 6 [%, relative to MDA] (E/Z or anti/syn ratio)b
From styrene
From cyclohexene
From 2-methyl-2-butene
1
2
3
4
a
[Ru2(l-sac)2(CO)6] (1)
[Ru2(l-sac)2(CO)5]2 (2)
[Ru2(l-sac)2(CO)5]2 (2)
[Ru2(l-OAc)2(CO)4]n (10)
1
1
3
1
62 (1.0)
76 (1.8)
69 (2.0)
95 (1.6)
45 (1.1)
60 (1.5)
61 (2.7)
68 (3.7)
43 (0.13)
46 (0.20)
61 (0.26)
61 (0.16)
Conditions: see Section 4.3.1.
b
Dimers E/Z-7 (ꢃ1:1) were also formed in all cases. Yield of E/Z-7 (%): entry 1: styrene 11, cyclohexene 28, 2-methyl-2-butene 14; entry 2: styrene 7,
cyclohexene 22, 2-methyl-2-butene 16; entry 3: styrene 11, cyclohexene 21, 2-methyl-2-butene 13; and entry 4: not determined.

S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
2781
insufficient amount of catalyst (0.5 mol%) which was
dictated, however, by the low solubility of 2 in anisole.
In contrast to the alkene cyclopropanation reactions
described above, the solubility of 2 did not improve even
after complete addition of the diazoester.
415 mg (0.57 mmol, 80% based on Ru3(CO)12). Decompo-
sition of the complex, indicated by a color change to red
and brown, started above 200 ꢁC. IR (KBr): 2107 vs,
2083 vs, 2075 s, 2042 s, 2025 vs, 2007 vs, 1616 m, 1585
m, 1380 m, 1328 m, 1177 m, 1167 m cmꢀ1 1H NMR
.
(CDCl3): d = 7.66–7.73 (m, 4H), 7.81 (dd, 2H), 7.87 (dd,
2H). Anal. Calc. for C20H8N2O12Ru2S2 (734.5): C, 32.70;
H, 1.10; N, 3.81. Found: C, 32.55; H, 1.14; N, 3.75%.
3. Conclusion
Dinuclear Ru22þ complexes of the type ½Ru2ðl-L1Þ2-
ðCOÞ4L22ꢁ where L1 is a bridging saccharinate ligand can
easily be synthesized from Ru3(CO)12 and saccharin and
are very stable towards air and moisture. Depending on
the axial ligands (CO, CH3CN, PPh3), they are obtained
with a head–head or a head–tail arrangement of the two
saccharinate ligands or as a mixture of isomers. Contrary
to expectation, the presence of electron-withdrawing SO2
groups in the saccharinate ligands does not enhance the
efficiency of these complexes for decomposition of diazo-
acetic esters compared to, e.g., the related acetate com-
plexes [Ru2(l-OAc)2(CO)4L2]. Nevertheless, [Ru2(l-sac)2-
(CO)6] (1) and [Ru2(l-sac)2(CO)5]2 (2) were found to be
suitable catalysts for the cyclopropanation of nucleophilic
olefins with methyl diazoacetate. The yields compare favor-
ably with those obtained with related Ru22þ complexes
bearing carboxylate, triazenide and 2-pyridonate bridging
ligands, and catalyst 1 induces a higher diastereoselectivity
for Z(syn)-cyclopropanes than [Ru2(l-OAc)2(CO)4]n.
4.2.2. [Ru2(l-sac)2(CO)5]2 (2)
A solution of Ru3(CO)12 (250 mg, 0.39 mmol) and sac-
charin (215 mg, 1.17 mmol) in toluene (40 ml) was heated
at reflux for 5 h. After 30 min a yellow solid began to pre-
cipitate. The suspension was concentrated to about 20 ml
and cooled to 7 ꢁC, then the solid was filtered off and
washed with diethylether and pentane (5 ml each). The yel-
low powder was dried at 160 ꢁC/0.02 mbar for 4 h, but
residual toluene was still present after this treatment. Yield
of 2: 367 mg (0.26 mmol, 89%). Decomposition of the com-
plex, indicated by a color change to brown and black,
started above 210 ꢁC. IR (KBr): 2099 vs, 2042 vs, 2021
vs, 2011 s, 1948 s, 1615 m, 1584 m, 1333 m, 1314 s, 1180
m, 1148 m cmꢀ1 1H NMR ([D8]-THF): d = 2.30 and
.
7.06–7.20 (residual toluene solvent, 0.04 mol%), 7.64–7.94
(m, all H-sac). Anal. Calc. for C38H16N4O22Ru4S4
(1413.1): C, 32.30; H, 1.14; N, 3.96. Found: C, 32.55; H,
1.30; N, 3.82%. C38H16N4O22Ru4S4 · 0.04 toluene
requires: C, 32.45; H, 1.16; N, 3.95.
4. Experimental
4.2.3. [Ru2(l-sac)2(CH3CN)2(CO)4] (3)
4.1. General remarks
Method 1. Complex 2 (33 mg) was dissolved in boiling
acetonitrile (4 ml). The solvent was evaporated and the yel-
low powder of 3 was dried for 1 h at 45 ꢁC/20 mbar.
Decomposition of the complex, indicated by a color change
to brown, started above 215 ꢁC. According to the NMR
spectra, a mixture of two species (3A,B) is present in solu-
tion, A:B ꢂ 2.9:1 at 295 K. IR (KBr): 2315 vw, 2288 vw,
2046 vs, 1996 m, 1956 vs, 1618 m, 1583 m, 1465 m, 1374
NMR spectra: Bruker DRX 400 (1H: 400.13 MHz; 13C:
100.62 MHz; 31P: 161.98 MHz); the solvent signal was used
as the internal standard for the 1H and 13C spectra (CDCl3:
dH = 7.26, dC = 77.0 ppm; THF: dH = 1.72 ppm; [1,2-D2]-
tetrachloroethane: dC = 73.7 ppm), while the dP values
are referenced to external H3PO4 (dp = 0 ppm). – IR spec-
tra: Bruker Vector 22. – Mass spectra, ESI: Waters micro-
mass ZMD. – Elemental analyses: Elementar Vario EL.
TGA: Mettler Toledo TGA/SDTA851, heating rate
10 ꢁC/min, N2 flow (50 ml/min). – GC: Varian CP-3800
with a FID. Ru3(CO)12 [28] and methyl diazoacetate [29]
were prepared by published procedures. All reactions were
performed under an inert atmosphere (argon) and in dry
solvents.
1
m, 1362 m, 1320 m, 1176 m, 1164 m, 1129 w cmꢀ1. H
NMR (CDCl3): d = 2.40 (s, CH3CN, major species A),
2.44 (s, CH3CN, minor species B), 7.57–7.87 (several m,
H-sac, A and B). Anal. Calc. for C22H18N4O10Ru2S2
(760.6): C, 34.74; H, 1.86; N, 7.37. Found: C, 34.68; H,
2.01; N, 7.33%. – Crystallizataion of 3 from CH3CN/
p-xylene afforded yellow crystals of bis(acetonitrile)-1jN:
2jN-tetracarbonyl-1j2C:2j2C-(l-saccharinato-1jN:2jO)-
(l-saccharinato-1jO:2jN)-diruthenium(Ru–Ru) (3A).
Method 2. A suspension of 1 (63 mg, 86 lmol) in
CH3CN (2 ml) was stirred at 50 ꢁC until a clear yellow
solution had formed. After cooling at r.t., the solvent was
evaporated, the residue was dissolved in dichloromethane
(2 ml), and pentane was added until the solution became
turbid (ca. 6 ml). The mixture was kept at ꢀ30 ꢁC for
two days, and the precipitate was collected and tried
(45 ꢁC/20 mbar); yellow solid (60 mg, 92%). The NMR
spectra were identical to those of the product obtained by
method 1.
4.2. Preparation of ruthenium complexes
4.2.1. Hexacarbonyl-1j3C:2j3C-(l-saccharinato-
1jO:2jN)-(l-saccharinato-1jN:2jO)-diruthenium-
(Ru–Ru), [Ru2(l-sac)2(CO)6] (1)
A solution of Ru3(CO)12 (300 mg, 0.47 mmol) and sac-
charin (258 mg, 1.40 mmol) in toluene (30 ml) was heated
at 90 ꢁC (bath temperature) for 36 h. The pale yellow pre-
cipitate was filtered off, washed with ether (5 ml) and pen-
tane (5 ml), then dried at 0.001 mbar/20 ꢁC for 4 h. Yield:

2782
S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
4.2.4. Pentacarbonyl-1j3C:2j2C-bis(l-saccharinato-
1jN:2jO)-triphenylphosphane-2jP-diruthenium(Ru–Ru),
[Ru2(l-sac)2(CO)5(PPh3)] (4) and tetracarbonyl-
1j2C:2j2C-(l-saccharinato-1jO:2jN)-(l-saccharinato-
1jN:2jO)-bis(triphenylphosphane-1jP:2jP-
diruthenium(Ru–Ru), [Ru2(l-sac)2(CO)4(PPh3)2] (5)
Method 1. Dichloromethane (5 ml) was added to com-
plex 1 (82 mg, 112 lmol) and triphenylphosphane (59 mg,
224 lmol) at 20 ꢁC. An orange solution was obtained
within a few minutes and gas evolution was observed.
After 10 min, the solvent was removed at 40 ꢁC/15 mbar.
In order to remove excess PPh3, the residue was sus-
pended in cyclohexane (10 ml) and treated with ultra-
syringe pump, a solution of methyl diazoacetate (0.10 g,
1 mmol) in dry dichloromethane and alkene (1 mmol)
was added during 4 h (10 h in the case of 2-methyl-2-
butene). Then, the reaction mixture was stirred until com-
plete consumption of the diazo compound was indicated by
IR (m(C@N2) = 2115 cmꢀ1) (from 5 h to more than 20 h).
The solution was passed through a short silica gel column
to remove the catalyst, and a defined amount of naphtha-
lene (for experiments with styrene and cyclohexene) or
mesitylene (for 2-methyl-2-butene) was added to the eluate
as an internal standard. The yields and diastereomer ratios
of cyclopropanes were determined by GC, using a Varian
CP-WAX 52 column (30 m · 0.32 mm, film thickness
0.25 lm) fitted with a retention gap. The response factor
of each cyclopropane diastereomer was determined on
samples prepared separately.
sound during 10 min.
A
solid was isolated by
centrifugation and was dried at 110 ꢁC/0.001 mbar for
24 h; yield: 117 mg. It consisted of a yellow and an orange
component which could be separated manually and were
identified as 4 and 5, respectively. It was observed that
the orange solid deposited faster than the yellow one from
the suspension in cyclohexane. The original ratio of 4:5
was 1:5 (31P NMR integration).
Method 2. Dichloromethane (5 ml) was added to com-
plex 2 (40 mg, 28 lmol) and triphenylphosphane (31 mg,
118 lmol) at 20 ꢁC. An orange solution was gradually
formed and after 15 min, the solvent was removed at
40 ꢁC/15 mbar. Further workup as described above pro-
vided 52 mg of a mixture of 4 and 5 (1:28 by 31P NMR
integration).
4.3.2. Method B: reaction with arenes
A solution of methyl diazoacetate (2.00 g, 20 mmol) in
the liquid arene (20 mmol) was added at r.t. during 26 h,
by means of a syringe pump, to a solution of the same
arene (180 mmol) containing 0.5 mol% of catalyst. Com-
plete disappearance of the diazo compound was observed
after 44–70 h (IR control). Products were isolated by col-
umn chromatography on silica gel (100 g, Merck silica
gel 60, 0.063–0.200 mm). Successive elution with pentane,
ether/pentane mixtures, and diethyl ether furnished excess
arene, products 8/9, and a mixture of dimethyl fumarate
and dimethyl maleate (E/Z-7). All products are known;
they were identified by their 1H NMR spectra [30] and their
yields were determined by product isolation or from the
crude product mixture by 1H NMR integration, using
naphthalene as the internal standard.
Spectroscopic and analytical data of 4: IR (CH2Cl2 film
on NaCl plate): 3062 w, 2086 s, 2034 vs, 1998 s, 1963 m,
1935 w, 1616 s, 1584 s, 1436 m, 1387 m, 1331 s, 1178 s,
1
1126 m, 1094 m cmꢀ1. H NMR (CDCl3): d = 6.98 (d,
JH,H = 7.6 Hz, 2H), 7.45–7.50 (m, 8H), 7.52–7.56 (m,
3H), 7.59 (d, JH,H = 1.5 Hz, 2H), 7.60–7.64 (m, 6H), 7.79
(d, JH,H = 7.8 Hz, 2H). 13C{1H} NMR (CDCl3):
d = 121.4, 123.6, 128.7, 128.8, 129.5, 130.5 (d,
JC,P = 2.2 Hz), 131.5, 131.9, 133.0, 133.3, 133.6, 133.7,
4.4. X-ray crystal structure determination for 1, 3A, 4, and 5
Single crystals of 1 were obtained by slow evaporation
of a dichloromethane solution, but the crystal quality was
rather poor. Crystals of 3A · p-xylene were obtained by
slow evaporation of an acetonitrile/p-xylene solution at
r.t. Crystals of 4 · CH2Cl2 and 5 · 3CH2Cl2 were obtained
from dichloromethane/pentane by a diffusion method,
starting from a 4/5 mixture. The crystals of 4 (yellow)
and 5 (orange) were separated manually. In the latter case,
the crystals were isolated and immediately coated with an
oil and cooled to prevent loss of solvent of crystallization
which would cause degradation of the crystals within a
few minutes. Data collection was performed on an image-
plate diffractometer (Stoe IPDS) using monochromated
3
141.2, 178.9 (d, JC,P = 8.8 Hz), 181.1 (d, JC,P = 32.9 Hz),
198.1 (d, JC,P = 2.9 Hz), 203.2 (d, JC,P = 5.1 Hz). 31P
(CDCl3): d = 22.0.
Spectroscopic and analytical data of 5: IR (CH2Cl2
film on NaCl plate): 3060 w, 2035 s, 1996 s, 1965 s,
1614 s, 1579 s, 1435 m, 1380 m, 1330 s, 1176 s cmꢀ1
.
1H NMR (CDCl3): d = 7.06 (d, JH,H = 7.6 Hz, 2H),
7.37–7.41 (m, 17H), 7.44–7.48 (m, 3H), 7.56–7.62 (m,
4H), 7.64–7.69 (m, 12H). 13C{1H} NMR (CDCl3):
d = 120.8, 123.7, 128.0 (vt, J = 4.8 Hz), 128.4, 128.5,
128.7, 129.7, 129.7, 132.2, 132.9, 133.1, 134.2 (vt,
J = 5.9 Hz), 134.4, 134.5, 142.4, 178.6 (vt, J = 6.2 Hz),
202.5 (vt, J = 3.3 Hz), 204.0 (vt, J = 4.8 Hz) (vt = virtual
triplet). 31P (CDCl3): d = 14.1.
˚
Mo Ka radiation (k = 0.71073 A). No absorption correc-
tion was applied. Structure solution was achieved by direct
methods, and the structures were refined against F 2o values
using a full-matrix least-squares method. Hydrogen atom
positions were calculated geometrically and treated as rid-
ing on their bond neighbors in the refinement procedure.
Software for structure solution and refinement: SHELX-97
[31]; molecule plots: ORTEP-3 [32]. Crystallographic data
4.3. Catalytic carbene transfer reactions
4.3.1. Method A: cyclopropanation of alkenes
The catalyst was suspended at r.t. in a mixture of alkene
(9 mmol) and dry dichloromethane (4 ml). By means of a

S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
2783
Table 3
Summary of crystallographic data and structure refinement for compounds 1, 3A, 4, and 5
Compound
1
3A
4
5
a
b
b
Empirical formula
Formula weight
T (K)
C20H8N2O12Ru2S2
734.54
295(2)
C22H14N4O10Ru2S2 · C8H10
760.64 + 106.16
173(2)
C37H23N2O11Ru2S2 · CH2Cl2 C54H38N2O10Ru2S2 · 3CH2Cl2
937.86 + 84.95
1203.12 + 254.80
193(2)
220(2)
Crystal size (mm)
Crystal system
Space group
0.38 · 0.31 · 0.38
Monoclinic
P21/c
7.886(1)
14.760(1)
21.367(3)
90
97.57(2)
90
2465.3(5)
4
0.54 · 0.23 · 0.08
Monoclinic
C2/c
13.468(1)
16.610(2)
15.401(1)
90
106.32(2)
90
3306.5(6)
4
0.54 · 0.38 · 0.27
Triclinic
0.43 · 0.43 · 0.39
Monoclinic
P21/n
15.974(1)
34.863(3)
21.379(2)
90
93.25(1)
90
11887.1(17)
8
1.629
ꢀ
P1
˚
a (A)
12.876(2)
13.378(2)
13.535(2)
98.95(2)
112.11(2)
103.97(2)
2016.0(5)
2
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
3
˚
V (A )
Z
qber (g cmꢀ3
)
1.979
14.62
2.37–23.99
1.741
11.02
2.16–25.95
1.736
10.87
l (Mo Ka) (cmꢀ1
h Range (ꢁ)
)
0.96
2.39–26.00
ꢀ15/16, ꢀ16/16, ꢀ16/16
23873
1.94–24.13
ꢀ18/18, ꢀ39/40, ꢀ24/24
76619
Index ranges of h, k, l
Reflections collected
ꢀ8/8, ꢀ16/16, ꢀ24/24 ꢀ15/16, ꢀ20/20, ꢀ18/17
15159
12862
Independent reflections (Rint
Completeness to hmax (%)
Data/restraints/parametersb
Goodness-of-fit on F2
Final R indices R1,
)
3818 (0.1354)
99.1
3818/0/343
0.911
3027 (0.0618)
93.6
3027/0/222
0.963
7335 (0.0372)
92.6
7335/0/523
1.069
18615 (0.0609)
98.1
18165/6e/1478
0.802
0.0527, 0.1082
0.0275, 0.0683
0.0257, 0.0658
0.0350, 0.0776
c,d
wR2 [I > 2r(I)]
c,d
R indices R1, wR2
0.0867, 0.1162
0.0337, 0.0703
0.0342, 0.0745
0.0729, 0.0814
(all data)
Largest difference in
peak and hole (e A
0.76 and ꢀ0.75
0.63 and ꢀ1.345
1.19 and ꢀ0.90
0.80 and ꢀ0.90
ꢀ3
˚
)
a
p-Xylene solvate.
b
c
Dichloromethane solvate.
Refinement based on F2 values.
P
P
P
P
2
2 1=2
d
e
R1 = iFo| ꢀ |Fci/ |Fo|; wR2 ¼ ½ ðwðF 2o ꢀ F c2Þ Þ= wðF o2Þ ꢁ
.
One CH2Cl2 molecule is disordered over two positions.
and details of the refinement for the four structures are
given in Table 3.
References
[1] M.P. Doyle, M.A. McKervey, T. Ye, Modern Catalytic Methods for
Organic Synthesis with Diazo Compounds: From Cyclopropanes to
Ylides, Wiley, New York, 1998.
Acknowledgments
[2] (a) M.P. Doyle, D.C. Forbes, Chem. Rev. 98 (1998) 911–936;
(b) H.M.L. Davies, R.E.J. Beckwith, Chem. Rev. 103 (2003) 2861–
2903;
(c) C.A. Merlic, A.L. Zechman, Synthesis (2003) 1137–1156.
[3] G. Maas, Chem. Soc. Rev. 33 (2004) 183–190.
[4] (a) S.-L. Zheng, W.-Y. Yu, C.-M. Che, Org. Lett. 4 (2002) 889–
892;
This work has received financial support by the Deut-
sche Forschungsgemeinschaft. The support and sponsor-
ship conceded by COST action D24 ‘Sustainable
Chemical Processes: Stereoselective Transition Metal-
Catalyzed Reactions’ is kindly acknowledged.
(b) W.-H. Cheung, S.-L. Zheng, W.-Y. Yu, G.-C. Zhou, C.-M.
Che, Org. Lett. 5 (2003) 2535–2538;
Appendix A. Supplementary material
(c) M.K.-W. Choi, W.-Y. Yu, C.-M. Che, Org. Lett. 7 (2005)
1081–1084.
[5] (a) S. Sengupta, D. Das, D.S. Sarma, Tetrahedron Lett. 37 (1996)
8815–8818;
CCDC-285843 (1), CCDC-285844 (3A), CCDC-295914
(4) and CCDC-295915 (5) contains the supplementary crys-
tallographic data for this paper. These data can be
obtained free of charge at www.ccdc.cam.ac.uk/conts/
retrieving.html [or from the Cambridge Crystallographic
Data Centre, 12, Union Road, Cambridge CB2 1EZ,
UK; fax: (internat.) +44 1223 336 033; e-mail: depos-
it@ccdc.cam.ac.uk]. Supplementary data associated with
this article can be found, in the online version, at
doi:10.1016/j.jorganchem.2006.02.013.
´
(b) E. Galardon, S. Roue, P. Le Maux, G. Simonneaux, Tetrahedron
Lett. 39 (1998) 2333–2334;
(c) F. Simal, A. Demonceau, A.F. Noels, Tetrahedron Lett. 40 (1999)
63–66;
(d) E. Galardon, P. Le Maux, G. Simonneaux, Tetrahedron 56
(2000) 615–622;
(e) R.P. Wurz, A.B. Charette, Org. Lett. 4 (2002) 4531–4533;
(f) A. Del Zotto, W. Baratta, F. Miani, G. Verardo, P. Rigo, Inorg.
Chim. Acta 349 (2003) 249–252.

2784
S. Buck, G. Maas / Journal of Organometallic Chemistry 691 (2006) 2774–2784
[6] (a) M. Alt, G. Maas, Tetrahedron 50 (1994) 7435–7444;
(b) C.-Y. Zhou, W.-Y. Yu, C.-M. Che, Org. Lett. 4 (2002) 3235–
3238;
[17] (a) F.A. Cotton, L.R. Falvello, W. Schwotzer, C.A. Murillo, G.
Valle-Bourrouet, Inorg. Chim. Acta 190 (1991) 89–96;
(b) N.M. Alfaro, F.A. Cotton, L.M. Daniels, C.A. Murillo, Inorg.
Chem. 31 (1992) 2718–2723.
(c) C.-Y. Zhou, P.W.H. Chan, W.-Y. Yu, C.-M. Che, Synthesis
(2003) 1403–1412.
´
[18] (a) J.A. Cabeza, C. Landazuri, L.A. Oro, D. Belletti, A. Tiripicchio,
[7] (a) O. Fujimura, T. Honma, Tetrahedron Lett. 39 (1998) 625–626;
(b) Y. Chen, L. Huang, M.A. Ranade, X.P. Zhang, J. Org. Chem. 68
(2003) 3714–3717;
M. Tiripicchio-Camellini, J. Chem. Soc., Dalton Trans. (1989) 1093–
1100;
(b) J.A. Cabeza, C. Landa´zuri, L.A. Oro, A. Tiripicchio, M.
Tiripicchio-Camellini, J. Organomet. Chem. 322 (1987) C16–
C20.
(c) F.E. Kuhn, A.M. Santos, A.A. Jogalekar, F.M. Pedro, P. Rigo,
¨
W. Baratta, J. Catal. 227 (2004) 253–256.
[8] A. Del Zotto, W. Baratta, F. Miani, G. Verardo, P. Rigo, Eur. J.
Org. Chem. (2000) 3731–3735.
[19] P.L. Andreu, J.A. Cabeza, V. Riera, Y. Jeannin, D. Miguel, J. Chem.
Soc., Dalton Trans. (1990) 2201–2206.
[9] (a) J.P. Collman, E. Rose, G.D. Venburg, J. Chem. Soc., Chem.
Commun. (1993) 934–935;
[20] G.R. Crooks, B.F.G. Johnson, J. Lewis, I.G. Williams, J. Chem. Soc.
(A) (1969) 2761–2766.
(b) W. Baratta, A. Del Zotto, P. Rigo, Chem. Commun. (1997) 2163–
2164.
[21] L. Scha¨ffler, B. Mu¨ller, G. Maas, Inorg. Chim. Acta 359 (2006) 970–
977.
[10] (a) G. Maas, T. Werle, M. Alt, D. Mayer, Tetrahedron 49 (1993)
881–888;
(b) G. Maas, J. Seitz, Tetrahedron Lett. 42 (2001) 6137–6140;
(c) T. Werle, G. Maas, Adv. Synth. Catal. 343 (2001) 37–40.
[11] T. Werle, L. Scha¨ffler, G. Maas, J. Organomet. Chem. 690 (2005)
5562–5569.
[22] L. Scha¨ffler, G. Maas, unpublished results.
[23] (a) H. Schumann, J. Opitz, J. Pickart, J. Organomet. Chem. 128
(1977) 253–264;
(b) M. Spohn, T. Vogt, J. Stra¨hle, Z. Naturforsch. B 41 (1986) 1373–
1380.
[24] G. Suss-Fink, G. Herrmann, P. Morys, J. Ellermann, A. Veit, J.
¨
[12] C.-D. Leger, G. Maas, Z. Naturforsch. B 59 (2004) 573–578.
Organomet. Chem. 284 (1985) 263–273.
´
[13] See, for example: (a) L.R. Falvello, J. Gomez, I. Pascual, M. Tomas,
[25] (a) F. Neumann, H. Stoeckli-Evans, G. Suss-Fink, J. Organomet.
¨
E.P. Urriolabeitia, A.J. Schultz, Inorg. Chem. 40 (2001) 4455–4463;
(b) E.J. Baran, Quim. Nova 28 (2005) 326–328;
(c) V.T. Yilmaz, S. Guney, W.T.A. Harrison, Z. Naturforsch. B 60
(2005) 403–407;
Chem. 379 (1989) 139–150;
(b) M. Rotem, I. Goldberg, U. Shmueli, Y. Shvo, J. Organomet.
Chem. 314 (1986) 185–212.
´
[26] P.L. Andreu, J.A. Cabeza, G.A. Carriedo, V. Riera, S. Garcıa-
(d) S. Hamamci, V.T. Yilmaz, W.T.A. Harrison, Z. Naturforsch. B
60 (2005) 978–983.
Granda, J.F. Van der Maelen, G. Mori, J. Organomet. Chem. 421
(1991) 305–314.
[14] See, for example: (a) S.Z. Haider, K.M.A. Malik, K.J. Ahmed, H.
Hess, H. Riffel, M.B. Hursthouse, Inorg. Chim. Acta 72 (1983) 21–27;
(b) F.A. Cotton, G.E. Lewis, C.A. Murillo, W. Schwotzer, G. Valle,
Inorg. Chem. 23 (1984) 4038–4041.
[27] S.J. Sherlock, M. Cowie, E. Singleton, M.M. de V. Steyn, J.
Organomet. Chem. 361 (1989) 353–367.
[28] A. Mantovani, S. Cenini, Inorg. Synth. 16 (1976) 47–48.
[29] (a) N.E. Searle, Org. Synth. Coll. IV (1963) 424–426;
(b) H.-U. Reißig, J. Reichelt, T. Kunz, Org. Synth. 71 (1992) 189–
199.
´
[15] (a) G. Jovanovski, A. Hergold-Brundic, B. Kamenar, Acta Crystal-
logr., Sect. C 44 (1988) 63–66;
(b) E.J. Baran, C.C. Wagner, M. Rossi, F. Caruso, Z. Anorg. Allg.
Chem. 626 (2000) 701–705;
[30] (a) A.J. Anciaux, A. Demonceau, A.F. Noels, A.J. Hubert, R.
´
Warin, P. Teyssie, J. Org. Chem. 46 (1981) 873–876;
(c) V.T. Yilmaz, S. Guney, C. Thoene, Z. Anorg. Allg. Chem. 628
(2002) 1406–1410.
[16] P. Naumov, G. Jovanovski, S.-Z. Hu, I.-H. Suh, I.A. Razak, S.
Chantrapromma, H.-K. Fun, S.W. Ng, Acta Crystallogr., Sect. C 57
(2001) 1016–1019.
(b) N. Mandava, H. Finegold, Spectrosc. Lett. 13 (1980) 59–68.
[31] G.M. Sheldrick, SHELX-97: Program for the Solution and Refinement
of Crystal Structures from Diffraction Data, University of Go¨ttingen,
Go¨ttingen, Germany, 1997.
[32] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.