Substitution reactivity of the hydrido sulfido bridged dirhenium complex [Re2(μ-H)(μ-Snaph)(CO)8] (naph = 2-naphthyl)

Article abstract of DOI:10.1039/b107859c

The substitution reactivity of the hydrido sulfido bridged dirhenium complex [Re₂(μ-H)(μ-Snaph)(CO)₈] 1 (naph = 2-naphthyl) has been investigated. The former complex can be easily substituted with one equivalent of tmno (trimethylamine-N-oxide) followed by addition of a ligand L giving complexes of the general formula [Re₂(μ-H)(μ-Snaph)(CO)₇L] [L = NCMe (2a), NC(Bu^t) (2b), pyridine (2c), PH₃ (2d), PMe₃ (2e), H₂P(p-C₆H₄OMe) (2f), PPh₃ (2g), P(OMe)₃ (2h)]. L is always perpendicularly co-ordinated to one of the Re atoms with respect to the Re-(μ-SR)-Re plane. In case of ligands with small cone angles two isomers are found. They differ in the relative orientation of L with respect to the substituent R attached to the sulfido bridge (syn and anti isomer). Temperature dependent 2D-EXSY NMR experiments with 2a dissolved in CDCl₃ proved that both isomers are in a dynamic equilibrium, which is based on the pyramidal inversion of the bridging sulfur. The thermodynamic data of this process at 298.15 K were determined to be: ΔG? = 68.6 ± 1.2 kJ mol^-1, ΔH? = 49.0 ± 4.2 kJ mol^-1 and ΔS? = -66 ± 21 J K^-1 mol^-1. Owing to intramolecular sterical hindrance ligands with large cone angles give only the anti isomer. Disubstituted complexes [Re₂(μ-H)(μ-Snaph)(CO)₆L₂] [L = NCMe (3a), pyridine (3b), PPh₃ (3c)] and [Re₂(μ-H)(μ-Snaph)(CO)₆(μ-L-L)] [L-L = dppm (4a), dppe (4b)] were prepared in a similar way. The reaction of [Re₂(μ-H)(μ-SH)(CO)₈] 5 with one equivalent of tmno and PPh₃ gave [Re₂(μ-H)(μ-SH)(CO)₆(PPh₃)₂] 6 in 38% yield. Compound 6 is rigid on the timescale of ³¹P NMR spectroscopy. The molecular structures of 2a, 2c, 2g, 2h, 3a, 4b and 6 were confirmed by X-ray structure analyses.

Full text of DOI:10.1039/b107859c

View Article Online / Journal Homepage / Table of Contents for this issue
Substitution reactivity of the hydrido sulfido bridged dirhenium
complex [Re2(ꢀ-H)(ꢀ-Snaph)(CO)8] (naph ؍ 2-naphthyl)
Hans Egold,* Detlef Schwarze and Ulrich Flörke
Anorganische und Analytische Chemie der Universität Paderborn, Fachbereich 13,
Chemie und Chemietechnik, Warburger Straße 100, Paderborn, D-33098, Germany.
E-mail: he@chemie.uni-paderborn.de
Received 31st August 2001, Accepted 11th December 2001
First published as an Advance Article on the web 8th February 2002
The substitution reactivity of the hydrido sulfido bridged dirhenium complex [Re2(µ-H)(µ-Snaph)(CO)8] 1
(naph = 2-naphthyl) has been investigated. The former complex can be easily substituted with one equivalent
of tmno (trimethylamine-N-oxide) followed by addition of a ligand L giving complexes of the general formula
[Re2(µ-H)(µ-Snaph)(CO)7L] [L = NCMe (2a), NC(But) (2b), pyridine (2c), PH3 (2d), PMe3 (2e), H2P(p-C6H4OMe)
(2f ), PPh3 (2g), P(OMe)3 (2h)]. L is always perpendicularly co-ordinated to one of the Re atoms with respect to the
Re–(µ-SR)–Re plane. In case of ligands with small cone angles two isomers are found. They differ in the relative
orientation of L with respect to the substituent R attached to the sulfido bridge (syn and anti isomer). Temperature
dependent 2D-EXSY NMR experiments with 2a dissolved in CDCl3 proved that both isomers are in a dynamic
equilibrium, which is based on the pyramidal inversion of the bridging sulfur. The thermodynamic data of
this process at 298.15 K were determined to be: ∆G‡ = 68.6 1.2 kJ molϪ1, ∆H‡ = 49.0 4.2 kJ molϪ1 and
∆S‡ = Ϫ66 21 J KϪ1 molϪ1. Owing to intramolecular sterical hindrance ligands with large cone angles give
only the anti isomer. Disubstituted complexes [Re2(µ-H)(µ-Snaph)(CO)6L2] [L = NCMe (3a), pyridine (3b),
PPh3 (3c)] and [Re2(µ-H)(µ-Snaph)(CO)6(µ-L–L)] [L–L = dppm (4a), dppe (4b)] were prepared in a similar
way. The reaction of [Re2(µ-H)(µ-SH)(CO)8] 5 with one equivalent of tmno and PPh3 gave [Re2(µ-H)-
(µ-SH)(CO)6(PPh3)2] 6 in 38% yield. Compound 6 is rigid on the timescale of 31P NMR spectroscopy.
The molecular structures of 2a, 2c, 2g, 2h, 3a, 4b and 6 were confirmed by X-ray structure analyses.
the 2-naphthyl residue facilitates crystallisation of the prepared
Introduction
compounds. Starting from 1, the mono- and di-substituted
products [Re2(µ-H)(µ-Snaph)(CO)7L] and [Re2(µ-H)(µ-Snaph)-
(CO)6L2] (L = two-electron donor ligand) were prepared and the
different substitution patterns with respect to the cone angle of
L were explored. In addition the observed pyramidal inversion
of the bridging sulfur in [Re2(µ-H)(µ-Snaph)(CO)7NCMe] 2a
was examined by 2D-EXSY NMR spectroscopy.9
The chemistry of polynuclear sulfido bridged (µ-SR) metal
complexes is very rich.1 Almost every transition metal forms
such complexes with one, two or three sulfido ligands bridging a
metal–metal vector. The sulfido bridge has to be considered as a
three-electron donor ligand having one lone pair of electrons
left. In many of these complexes the sulfido bridges are non-
rigid ligands due to pyramidal inversion of the bridging sulfur
atom. The low inversion energy is a consequence of metal–
sulfur interaction. The metal reduces the s-character of the lone
pair ground state allowing easier access to the transition state
in which the inverting lone pair of electrons is regarded as
having pure p-character. In addition, pπ–dπ overlap of sulfur
and metal orbitals reduces the energy barrier for pyramidal
inversion as well giving in many cases configurationally non-
rigid molecules.2 Typical values for the sulfur atomic inver-
sion energy ∆G‡ in sulfido bridged complexes range from
Results and discussion
Monosubstituted complexes [Re2(ꢀ-H)(ꢀ-Snaph)(CO)7L]
(naph ؍ 2-naphthyl)
On treating [Re2(µ-H)(µ-Snaph)(CO)8] 1 with one equivalent of
tmno (trimethylamine-N-oxide) at 0 ЊC in THF, one CO ligand
is removed. Subsequent addition of one equivalent of a ligand
L generates the complexes [Re2(µ-H)(µ-Snaph)(CO)7L] 2 [L =
NCMe (2a), NC(But) (2b), pyridine (2c), PH3 (2d), PMe3 (2e),
H2P(p-C6H4OMe) (2f ), PPh3 (2g), P(OMe)3 (2h)] in good yields
(Scheme 1). Compound 2d was obtained from the reaction of 1
with P(SiMe3)3 and subsequent methanolysis. Naturally, their
ν(CO) spectra (not shown) are very similar, showing seven
absorptions and suggesting that all complexes have an analo-
gous structure. On the basis of the X-ray structure analyses of
2a, 2c, 2g and 2h we conclude that L is always axially bound
[perpendicular with respect to the Re–(µ-SR)–Re plane]. The
chemical shift of the µ-H ligand in these complexes is strongly
dependent on the ligand L. In general, on substitution a low
field shift of the µ-H resonance relative to 1 is observed.
45 to 80 kJ molϪ1 2,3
There are many reports on the chemistry
.
of dinuclear sulfido bridged complexes of manganese and
rhenium,4 but only few refer to such stereodynamic processes.
For example, in the case of the complexes [M2(µ-SR)2(CO)8]
(M = Mn, Re; R = H, organic residue), only for R = H is the
inversion of the bridging sulfur mentioned.5 The similar mole-
cules [M2(µ-PR2)(µ-SRЈ)(CO)8] (M = Mn, Re; R, RЈ = organic
residue)6 and [Re2(µ-ER)(µ-EЈRЈ)(CO)8] (E, EЈ = S, Se, Te; R,
RЈ = organic residue)7 are known to show evidence of
pyramidal inversion at room temperature. Rate constants and
thermodynamic data were not determined in any of these cases.
We have recently reported the synthesis the sulfido bridged
complexes [Re2(µ-H)(µ-SR)(CO)8] (R = H, organic residue)8
and now wish to report on the facile synthesis and character-
isation of their substitution products. The investigations
focused on [Re2(µ-H)(µ-Snaph)(CO)8] 1 (naph = 2-naphthyl) as
Complexes substituted by ligands with small cone angles
according to Tolman10 (2a, 2b, 2d, 2f ) consist of two isomers as
evidenced by two high field shifted 1H NMR signals of the µ-H
ligands. These signals must be attributed to a syn and an anti
1078
J. Chem. Soc., Dalton Trans., 2002, 1078–1084
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b107859c

View Article Online
Scheme 1 Preparation of substitution products of 1.
Table 1 syn : anti Ratio of 2a in various solvents according to
1H NMR data
isomer. The isomers differ in the relative orientation of L with
respect to the 2-naphthyl residue attached to the bridging sulfur
atom. The ratio strongly depends on the solvent used. We have
investigated this dependence for 2a as this is the only pair
syn : anti
Solvent
Polarity of solvent
1
of isomers allowing assignment of syn and anti isomer by H
1 : 1.37
1 : 1.22
1 : 1.12
1 : 0.68
1 : 0.67
1 : 1.07
CD3CN
d6-Acetone
d8-THF
CD2Cl2
CDCl3
0.65
0.56
0.45
0.42
0.40
0.32
NMR spectroscopy. The syn isomer always exhibits the reson-
ance of the methyl group of the acetonitrile ligand at high field
(1.21 ppm in CDCl3) whereas the corresponding resonance of
the anti isomer is located at lower field (2.33 ppm in CDCl3).
The assignment of syn or anti is based on a comparison with
other acetonitrile complexes of rhenium11 like, for example,
d6-Benzene
1
᎐
[Re (µ-H)(µ-C᎐CPh)(CO) (NCMe)] [δ(Me) = 2.20 in CD -
temperature H NMR spectra of 2a in CDCl3 were recorded.
Upon heating, the signals of the µ-H ligands coalescenced at
310 K confirming a fluxional process. On further heating to
320 K the H1 signals of the 2-naphthyl group coalescenced, and
finally upon heating to 330 K only one sharp resonance for the
µ-H ligands and for the H1 protons remained. At this point the
signals of the acetonitrile ligand are very broad, but coales-
cence of their resonances was not yet reached. Since no free
acetonitrile was observed the detected dynamic process is
almost certainly based on the inversion of the sulfido bridge
(Scheme 2). In 2a the inversion of the sulfido bridge proceeds
᎐
2
7
2
Cl2].11b In all of these cases the signal of the methyl group of
acetonitrile is located well above 2 ppm. The unusual high field
shift of one of the methyl groups in the 2a experiments is most
likely caused by an electronic shielding effect of the neighbour-
ing 2-naphthyl residue. Therefore we assign this signal to the
syn isomer, and consequently the signal at δ = 2.33 to the anti
isomer. The syn : anti ratios of 2a in various solvents are sum-
marised in Table 1. The ratio of isomers was determined by
integration of the signals of the µ-H ligands or of the methyl
resonances of the acetonitrile ligands, respectively. Apparently,
the sterically favoured anti isomer predominates in all solvents
but dichloromethane and chloroform. From chloroform–
hexane, syn-2a crystallised as confirmed by a single crystal
X-ray analysis.
Interconversion of syn and anti isomers
The above-described results indicated that syn and anti isomers
are interconverting. In order to verify this hypothesis, variable-
Scheme 2 Interconversion of syn and anti isomers of 2a.
J. Chem. Soc., Dalton Trans., 2002, 1078–1084
1079

View Article Online
even at 298.4 K, as established by a 1H 2D-EXSY9 NMR spec-
trum recorded in CDCl3. From that spectrum the rate constant
of the labile NCMe. In 3c the PPh3 ligands are co-ordinated cis
to each other and anti with respect to the 2-naphthyl residue
due to their bulkiness. The two different types of substitution
pattern can be distinguished by ν(CO) IR spectroscopy. The
trans substituted products exhibit three absorption bands,
whereas in 3c an analogous pattern with an additional band at
1952 cmϪ1 is observed.
The substitution of 1 by bidentate ligands like dppe [1,2-bis-
(diphenylphoshino)ethane] or dppm [bis(diphenylphospino)-
methane] is best performed by reacting a THF solution of it
with one equivalent of tmno and subsequent addition of one
equivalent of these ligands followed by heating to reflux for
several hours. The expected substitution products [Re2(µ-H)-
(µ-Snaph)(CO)6(µ-L–L)] [L–L = dppm (4a), dppe (4b)] were
obtained in 90% yield. Reacting 1 with two equivalents of tmno
at 0 ЊC in THF and subsequent addition of dppm or dppe,
respectively, gives the same products, but the yields are
distinctly lower (about 30%). In both products the bidentate
ligands are bridging the two rhenium atoms. As expected they
are co-ordinated in an anti orientation with respect to the
2-naphthyl residue. The structure of 4b has been confirmed by
a single crystal X-ray analysis.
of the interconversion was determined to be 6.03 0.05 sϪ1
.
Its value was calculated from the integrals of the acetonitrile
resonances according to literature procedures.9 Further 2D-
EXSY spectra were recorded at 308.0, 314.9 and 322.3 K. At
322.3 K, however, the solution turned yellow within 3 hours
indicating slow decomposition of 2a. From the remaining data,
Eyring activation parameters at 298.15 K were calculated by
plotting ln(k/T ) vs. 1/T giving ∆G‡ = 68.6 1.2 kJ molϪ1, ∆H‡
= 49.0 4.2 kJ molϪ1 and ∆S‡ = Ϫ66 21 J KϪ1 molϪ1. As
these data are based on three temperatures only, their margin of
error is quite big. Nevertheless ∆S‡ is negative, strongly sug-
gesting an intramolecular rearrangement by pyramidal inver-
sion of the sulfido bridge. Moreover, the value of ∆G‡ is close
to those derived for other sulfido bridged transition metal com-
plexes like [HOs3(µ-SMe)(µ3-η2-C6H4)(CO)9] (∆G‡ = 61.5 kJ
3d
molϪ1
)
or [Mo2(CO)3(µ-SR)3(η7-C7H7)] [∆G‡ = 61.9, 58.8 kJ
molϪ1 (for R = Me, But)].3e In order to find out whether
the residue attached to the bridging sulfur has a significant
influence on the inversion process the 2-naphthyl residue in 2a
was replaced by the stronger electron withdrawing C6F5 residue.
It became immediately apparent that the inversion barrier is
lowered in this case. A 1H NMR spectrum of [Re2(µ-H)-
(µ-SC6F5)(CO)7NCMe] 2i in CDCl3 recorded at 300 K shows
only one µ-H resonance and two broad resonances for the
methyl groups of the acetonitrile ligand. Hence, the signals of
the µ-H ligands have already coalescenced due to faster inver-
sion of the sulfido bridge at this temperature. The C6F5 group is
probably stabilising the planar transition state by p(sulfur)
π*(C6F5) orbital overlap.2 From the big variety of mono-
substituted complexes prepared we can conclude that those
substituted by ligands L with cone angles larger than approx-
imately 100Њ (2c, 2e, 2g, 2h) exhibit only the anti isomer as
confirmed by a single 1H NMR resonance for the µ-H ligand. A
hypothetical syn isomer is unlikely to exist in these cases due to
steric repulsion between the 2-naphthyl residue and the bulky
ligand L. A comparison of these results with other hydrido
sulfido bridged metal complexes would be most desirable. How-
ever, for most of these complexes there is no indication whether
they are dynamic with respect to pyramidal inversion of sulfur
or not. For example, the analogous manganese complexes
[Mn2(µ-H)(µ-SR)(CO)8] are known to be stable compounds, but
so far only three examples have been obtained in very low yield
preventing a closer inspection of their chemical properties.12
The inversion of sulfur has been characterised for the com-
plexes [Ir2(µ-H)(µ-SR)Cp*2Cl2] as being not rapid on the NMR
timescale,13 but this statement is too vague for comparison with
the above-described complex. Since complexes like [Re2(µ-SPh)-
All attempts to prepare triply substituted derivatives of 1 by
tmno substitution, heating or irradiation in the presence of an
excess of ligand failed, giving disubstituted products only and
confirming the high thermodynamic stability of the Re(CO)3
units that has been observed for other sulfido bridged rhenium
carbonyl complexes.14
Substitution of [Re2(ꢀ-H)(ꢀ-SH)(CO)8] 5 with PPh3
The substitution of CO ligands in [Re2(µ-H)(µ-SH)(CO)8] 5 was
tested with tmno and PPh3. Although only one equivalent of
the latter compound was used [Re2(µ-H)(µ-SH)(CO)6(PPh3)2] 6
was the main product (40% yield). The reaction was accom-
panied by strong decomposition reactions. According to 1H
NMR and 31P NMR spectra 6 was contaminated with small
amounts of an unknown phosphorus-containing compound
[31P(CDCl3): δ = 43.9] which was lost on recrystallisation from
chloroform–hexane. A single crystal X-ray analysis revealed
that in 6 the two PPh3 ligands are co-ordinated in a trans
arrangement confirming that the cis arrangement in 3c is based
1
on steric factors only. The H NMR spectrum exhibits triplets
for the µ-H ligand (δ = Ϫ12.88, JPH = 7.7 Hz) and for the hydro-
gen attached to the sulfido bridge (δ = Ϫ1.08, JPH = 10.0 Hz) due
to coupling with the PPh3 ligands. The strong highfield shift of
the latter hydrogen resonance is typical for SH bridged mole-
cules {e.g. [Re2(µ-SH)2(CO)8] δ(H) = Ϫ0.89;15 6 δ(µ-SH) =
Ϫ0.47}.8 The 31P NMR spectrum shows two sharp resonances
at δ = 11.3 and 15.8 proving that the molecule is rigid on the
timescale of 31P NMR spectroscopy, since on rapid inversion
of the sulfido bridge only one resonance would have been
expected.
6a
(ER)(CO)8] (E = S, Se, Te)7 and [Re2(µ-PCy2)(µ-SPh)(CO)8]
also exhibit pyramidal inversion of sulfur at room temperature,
we conclude that the inversion in systems with a Re2(µ-Y)-
(µ-SR) (Y = bridging ligand) core is not strongly dependent
upon the bridging ligand Y.
Molecular structures of 2a, 2c, 2g, 3a, 4b and 6
These compounds exhibit equal molecular structures with
almost planar Re2(µ-H)(µ-S) cores. The basic structural ele-
ments of this type of molecular structure have been discussed
earlier.8 So the following will focus on the most interesting
differences only.
Compound 2a (see Fig. 1) differs from the already reported
molecular structure of [Re2(µ-H)(µ-Snaph)(CO)8] 18 in the
substitution of an axially co-ordinated CO group with an
acetonitrile ligand which is in the sterically less favoured syn
position to the 2-naphthyl group. The crystallisation of this
isomer is in agreement with the above discussed dynamics since
the crystal was grown from a solution of 2a in chloroform in
which the syn isomer predominates. Unfortunately all attempts
to crystallise the anti isomer from acetone or acetonitrile,
respectively, in which the anti isomer predominates failed as no
Disubstituted complexes [Re2(ꢀ-H)(ꢀ-Snaph)(CO)6L2] and
[Re2(ꢀ-H)(ꢀ-Snaph)(CO)6(ꢀ-L–L)]
Disubstituted derivatives of 1 were prepared starting from the
monosubstituted precursors 2a, 2c, 2g. The latter were reacted
with one equivalent of tmno and subsequently one equivalent
of ligand L giving the disubstituted complexes [Re2(µ-H)-
(µ-Snaph)(CO)6L2] [L = NCMe (3a), pyridine (3b), PPh3 (3c)] in
70–75% yield. In 3a and 3b each rhenium is axially substituted
by one of the ligands L which are arranged trans to each other
1
(Scheme 1) as established by H NMR spectra. We have no
indication that the sulfido bridge in these complexes is fluxional
as in 2a. 1H NMR spectra of 3a in CDCl3 at 45 ЊC show a slight
broadening of the methyl resonances. At higher temperatures
the complex slowly decomposes, most likely due to dissociation
1080
J. Chem. Soc., Dalton Trans., 2002, 1078–1084

View Article Online
tution of one axial CO group by a pyridine ligand instead
of acetonitrile, but now the ligand is positioned anti to the
2-naphthyl group. In view of the fact that in solution only one
isomer is present we conclude that the latter is the anti isomer
too. Compared with 2a the Re–Re and Re–S bond parameters
remain unaffected. Concerning the µ-H position the same
considerations are valid as for 2a.
In 2g (see Fig. 3) the substituting ligand is triphenylphos-
Fig. 1 Molecular structure of 2a. Hydrogen atoms omitted. Selected
bond lengths (Å) and angles (Њ): Re1–Re2 3.1023(12), Re1–S1 2.480(2),
Re2–S1 2.478(2), Re2–N1 2.150(8), S1–C11 1.788(9); S1–Re1–Re2
51.24(6), S1–Re2–Re1 51.29(5), Re1–S1–Re2 77.47(7).
single crystals were obtained. The Re–Re single bond is slightly
elongated to 3.1023(12) Å [3.0909(8) Å in the unsubstituted
complex] but the two equal Re–S distances remain unchanged.
A rather large torsion angle of 9.5(4)Њ is observed for C1–
Re1–Re2–N1 and the acetonitrile ligand bends towards the
2-naphthyl group [S1–Re2–N 87.4(2)Њ] whereas the carbonyl
group 1 bends away from it [S1–Re1–C1 97.3(3)Њ]. This is
obviously due to the steric interactions of these ligands with
the 2-naphthyl group. The 2-naphthyl group is twisted around
the S1–C11 axis, the angle between the plane of C11–C20
and the Re–Re vector is 17.1(1)Њ and the Re1–S1–C11–C12
and Re1–S1–C11–C20 torsion angles are 32.9(9) and 65.5(8)Њ,
respectively (absolute values).
The µ-H position was not determined from Fourier maps,
but the orientation of the equatorial CO groups 2 and 7 with
Re–Re–C angles of 124.8(3) and 120.0(3)Њ, respectively, as well
as Orpen’s HYDEX program16 strongly indicate a position in
the Re2S plane midway between the carbonyl ligands. This is
in agreement with the crystallographically determined µ-H
positions in Re2(µ-H)(µ-SPh)(CO)8 8 and 4b and 6.
Fig. 3 Molecular structure of 2g. Hydrogen atoms omitted. Selected
bond lengths (Å) and angles (Њ): Re1–Re2 3.1238(13), Re1–P1 2.488(2),
Re1–S1 2.479(2), Re2–S1 2.465(2), S1–C40 1.790(5); P1–Re1–Re2
97.93(5), S1–Re1–Re2 50.60(4), S1–Re2–Re1 51.03(4), Re1–S1–Re2
78.37(5).
phine, again in a position anti to the 2-naphthyl group. A hypo-
thetical syn co-ordination is impossible as the van der Waals
radii of PPh3 and the 2-naphthyl residue would overlap. Hence,
in solution only the anti isomer is present. The Re–Re bond is
clearly elongated to 3.1238(13) Å but the Re–S distances are the
same as for 2a and 2c. The µ-H position is deduced the same
way as above.
Compound 3a (see Fig. 4) is a twofold substituted complex
Complex 2c (see Fig. 2) exhibits two crystallographically
independent but geometrically almost identical molecules per
asymmetric unit. The structure differs from 2a in the substi-
Fig. 4 Molecular structure of 3a. Hydrogen atoms omitted. Selected
bond lengths (Å) and angles (Њ): Re1–Re2 3.1049(11), Re1–N1
2.138(13), Re1–S1 2.485(4), Re2–N2 2.136(14), Re2–S1 2.486(4),
S1–C11 1.780(16); N1–Re1–S1 85.4(4), N1–Re1–Re2 84.7(4),
S1–Re1–Re2 51.36(10), N2–Re2–S1 85.9(4), N2–Re2–Re1 86.1(4),
S1–Re2–Re1 51.34(9), Re1–S1–Re2 77.30(12).
with acetonitrile ligands co-ordinated at both Re atoms. These
ligands are in the sterically favourable trans position to each
other, thus the one attached to Re1 is anti to the 2-naphthyl
group and the one attached to Re2 is in the syn position. This
constitution is in full agreement with the assignment from the
spectroscopic solution data. As observed for 2a the molecule
exhibits a large C3–Re1–Re2–N2 torsion angle of 9.2(6)Њ and
Fig. 2 Molecular structure of 2c. Hydrogen atoms omitted. Selected
bond lengths (Å) and angles (Њ): molecule 1: Re1–Re2 3.1093(15),
Re1–N1 2.27(2), Re1–S1 2.477(7), Re2–S1 2.465(6), S1–C21 1.77(2);
S1–Re1–Re2 50.83(13), S1–Re2–Re1 51.18(15), Re1–S1–Re2 77.99(18);
molecule 2: Re3–Re4 3.1054(15), Re3–S2 2.482(6), Re4–N2 2.225(17),
Re4–S2 2.469(7), S2–C51 1.80(2); S2–Re3–Re4 50.97(17), S2–Re4–Re3
51.34(15), Re3–S2–Re4 77.69(18).
J. Chem. Soc., Dalton Trans., 2002, 1078–1084
1081

View Article Online
bending of the CO(3) and acetonitrile N(2) ligands [S1–Re1–C3
98.8(5)Њ, S1–Re2–N2 85.9(4)Њ], accompanied by the twist of the
2-naphthyl group along the S1–C11 axis with torsion angles of
Re1–S–C11–C20 36.7(15)Њ and Re2–S–C11–C12 62.9(13)Њ,
respectively (absolute values). The other pair of axial ligands,
CO(8) and acetonitrile N(1) show an undistorted eclipsed
arrangement along the Re–Re vector with N1–Re1–Re2–C8 of
0.1(6)Њ. The Re–Re bond length of 3.1049(11) Å and the µ-H
position is the same as for 2a and 2c.
A 1,2-bis(diphenylphosphino)ethane group (dppe) serves as
the bridging ligand in 4b (see Fig. 5). Each Re atom is attached
Fig. 6 Molecular structure of 6. Hydrogen atoms omitted except µ-H.
Selected bond lengths (Å) and angles (Њ): Re1–Re2 3.1783(4), Re1–S1
2.4891(16), Re1–P1 2.4946(14), Re2–S1 2.4929(16), Re2–P2 2.4897(15);
S1–Re1–P1 94.48(5), S1–Re1–Re2 50.41(4), P1–Re1–Re2 93.87(3),
S1–Re2–P2 97.42(5), P2–Re2–Re1 100.63(4), S1–Re2–Re1 50.31(4),
Re1–S1–Re2 79.28(4).
Fig. 5 Molecular structure of 4b. Hydrogen atoms omitted except
µ-H. Selected bond lengths (Å) and angles (Њ): Re1–Re2 3.1427(7),
Re1–P1 2.477(2), Re1–S1 2.505(2), Re2–P2 2.464(2), Re2–S1 2.490(2),
S1–C61 1.811(9), C31–C32 1.509(12); P1–Re1–S1 93.72(7), P1–Re1–
Re2 99.80(6), S1–Re1–Re2 50.81(5), P2–Re2–S1 85.79(8), P2–Re2–Re1
97.78(6), S1–Re2–Re1 51.22(5), Re1–S1–Re2 77.97(7).
Instrumentation
The reaction products were characterised by ν(CO) FTIR
spectroscopy (Nicolet P510; CaF2 optics); H and 31P NMR
1
spectroscopy (Bruker AMX 300).
2D EXSY NMR spectra of 2a
to one P atom in axial position and the dppe ligand is anti to
the 2-naphthyl group. A hypothetical syn isomer cannot exist
owing to the same reason pointed out for 2g. Consequently, in
solution only the anti isomer is present. The corresponding
Re2S and Re2P2 planes show a dihedral angle of 97.6(1)Њ. The
dppe chelate ligand leads to a prominent enlargement of the
Re–Re single bond to 3.1427(7) Å and the Re–S distances tend
to be slightly elongated compared with those of the preceding
compounds. The steric strain caused by dppe results also in
torsion angles of the carbonyl and the phosphine ligands along
the Re–Re vector from 5.7(5) to 8.7(5)Њ (absolute values). The
µ-H atom was derived from a Fourier map in the expected
position.
Compound 6 (see Fig. 6) is a disubstituted complex with a
PPh3 ligand attached to each of the Re atoms. These two
ligands show the sterically favourable trans arrangement for
their axial positions confirming the assignment based on
spectroscopic solution data. The distortion of the eclipsed
ligand arrangement is only small with C–Re–Re–C and
P–Re–Re–C torsion angles in the range from 0.2(2) to 4.9(2)Њ
(absolute values). Re–S distances are similar to those of
2a–3a but the Re–Re bond length of 3.1783(4) Å is unexpect-
edly the largest of this series of compounds. The µ-H atom was
located from a Fourier map but the position of the µ-S bonded
hydrogen atom could not be determined.
The experiments were performed and analysed according to the
pulse sequence given in the literature.9a The mixing time τm was
150 ms and the temperatures were determined by an internal
methanol thermometer (accuracy 0.1 ЊC). The rate constants
of the syn : anti interconversion were determined from the
cross-peak to diagonal-peak intensity ratio of the methyl
groups of the MeCN ligands at δ = 1.21 and 2.33.
Starting materials
The compounds tmnoؒ2H2O, PPh3, dppm, dppe and P(OMe)3
were received from Fluka. PMe3 was purchased from Strem.
H2P(p-C6H4OMe),17 P(SiMe3)3
and Re2(µ-H)(µ-SR)(CO)8
18
(R = 2-naphthyl, C6F5 and H)8 were prepared according to
literature procedures. The tmno used in the reactions described
below was liberated from water by sublimation.
Preparations
Monosubstituted complexes [Re2(ꢀ-H)(ꢀ-SR)(CO)7L)] [R ؍
2-naphthyl, L ؍ NCMe (2a), NC(But) (2b), pyridine (2c), PH3
(2d), PMe3 (2e), H2P( p-C6H4OMe) (2f ), PPh3 (2g), P(OMe)3
(2h); R ؍ C6F5, L ؍ NCMe (2i)]. 200 mg (0.264 mmol) of 1
were dissolved in 10 ml THF and cooled to 0 ЊC. Subsequently
20 mg (0.266 mmol) of tmno were added. The solution turned
pale yellow and was stirred for 30 min, thereupon the solvent
was removed. To the pale yellow residue was added one equiv-
alent of ligand L and the resulting mixture was redissolved in
dichloromethane. After 1 h of stirring the solution was taken to
dryness and the residue subjected to TLC (eluent dichloro-
methane–hexane = 1 : 1 except for L = PPh3: 1 : 5). In all cases
two fractions were obtained. The one with the higher Rf-value
contained small amounts (<5%) of educt 1. The second fraction
was pure [Re2(µ-H)(µ-Snaph)(CO)7L] 2. For the preparation
Experimental
General conditions
All reactions were performed in oxygen-free solvents which
were dried according to literature methods, distilled and stored
under an argon atmosphere. TLC was carried out on glass
plates (20 × 20 cm) coated with a mixture of gypsum and silica
gel (Merck 60 PF254; 1 mm thick).
1082
J. Chem. Soc., Dalton Trans., 2002, 1078–1084

View Article Online
of 2d the ligand P(SiMe3)3 was used. In this case the resulting
dichloromethane solution was stirred for 1 h as described
above, but then 10 ml of methanol were added and the reaction
mixture was stirred for another 18 h. During this period the
co-ordinated ligand was hydrolysed to give co-ordinated PH3.
The final working up sequence was performed as described
above. The synthesis of 2i is completely analogous to the syn-
thesis of 2a. Elemental analysis (yield): 2a (50%) Found:
C, 29.42; H, 1.42; N, 1.61. C19H11O7NRe2S requires: C, 29.65;
H, 1.44; N 1.82%. 2b (68%) Found: C, 32.36; H, 1.98; N, 1.50.
C22H17O7NRe2S requires: C, 32.55; H, 2.11; N, 1.73%. 2c (77%)
Found: C, 32.88; H, 1.87; N, 1.44. C22H13O7NRe2S requires:
C, 32.71; H, 1.62; N, 1.73%. 2d (38%) Found: C, 26.77; H, 1.45.
C17H11O7PRe2S requires: C, 26.77; H, 1.44%. 2e (51%) Found:
C, 30.27; H, 2.07. C20H17O7PRe2S requires: C, 29.85; H, 2.13%.
2f (65%) Found: C, 32.45; H, 1.63. C23H15O7PRe2S requires:
C, 32.93; H, 1.80%. 2g (78%) Found: C, 42.20; H, 2.27.
C35H23O7PRe2S requires: C, 42.42; H, 2.34%. 2h (65%) Found:
C, 28.36; H, 2.16. C20H17O10PRe2S requires: C, 28.17; 2.01%. 2i
(39%) Found: C, 21.90; H, 0.41; N, 1.65. C15H4F5NO7Re2S
requires: C, 22.25; H, 0.50; N, 1.73%. Spectroscopic data (a = e =
2010 vs, 1915 vs. 3c 2033 m, 2011 s, 1952 m, 1911 vs. 1H NMR
(CDCl3): 3a δ Ϫ10.15 (s, 1H, µ-H), 1.14 (s, 3H, Mesyn), 2.30 (s,
3H, Meanti), 7.38–7.80 (m, 6H, naph), 8.06 [s, 1H, H1(naph)]. 3b
δ Ϫ7.70 (s, 1H, µ-H), 6.89 [s(broad), 4H, py], 7.29–7.96 (m, 9H,
naph, py), 8.46 [s(broad), 4H, py]. 3c δ 11.70 (t, 2JPH = 8.5 Hz,
1H, µ-H), 7.22–7.85 (m, 36H, Ph, naph), 7.95 [s, 1H, H1(naph)].
31P NMR (CDCl3): 3c δ 8.3 (s, PPh3).
Disubstituted complexes [Re2(ꢀ-H)(ꢀ-Snaph)(CO)6(ꢀ-L–L)]
[L–L ؍ dppm (4a), dppe (4b)]. 200 mg (0.264 mmol) of 1 were
reacted with one equivalent of tmno in THF at 0 ЊC as
described above. Now one equivalent of L–L was added and the
mixture was refluxed for 8 h. Removal of the solvent and TLC
(eluent dichloromethane–hexane = 1 : 1) gave pure 4a and 4b.
Elemental analysis (yield): 4a (90%) Found: C, 45.18; H, 2.47.
C41H30O6P2Re2S requires: C, 45.38; H, 2.79%. 4b (95%) Found:
C, 45.76; H, 2.62. C42H32O6P2Re2S requires: C, 45.90; H, 2.93%.
ν(CO) IR/cmϪ1 (THF): 4a 2038 vs, 2013 s, 1952 m, 1932 m, 1915
s. 4b 2037 vs, 2013 s, 1952 m, 1929 m, 1915 s. 1H NMR (CDCl3):
4a δ Ϫ12.53 (t, 2JPH = 9.6 Hz, 1H, µ-H), 2.65 (m, 1H, CH2), 3.42
(m, 1H, CH2), 7.26–7.87 (m, 26H, Ph, naph), 8.20 [s, 1H,
b
f
H1(naph)]. 4b δ Ϫ12.62 (t, JPH = 10.5 Hz, 1H, µ-H), 2.28
signal of major isomer,
=
= signal of minor isomer): ν(CO)
2
IR/cmϪ1 (THF): 2a 2102 m, 2029 vs, 2006 s, 1950 m, 1925 s. 2b
2102 m, 2030 vs, 2006 s, 1954 m, 1927 s. 2c 2102 m, 2023 vs,
2006 s, 1991 m, 1952 m, 1917 s. 2d 2102 m, 2033 vs, 2009 s, 1927
s. 2e 2102 m, 2023 vs, 2006 vs, 1991 s, 1948 m, 1915 s, 1907 m. 2f
2102 m, 2023 vs, 2008 s, 1944 m, 1913 s. 2g 2102 m, 2027 vs,
2004 s, 1992 m, 1952 m, 1938 m, 1923 s. 2h 2104 m, 2035 vs,
2006 s, 1995 sh, 1950 m, 1925 m, 1902 s. 2i 2106 m, 2035 vs,
2010 s, 1996 s, 1961 m, 1940 m, 1928 m. 1H NMR (CDCl3): 2a
δ Ϫ11.95 (s, µ-Hsyn), Ϫ11.90 (s, µ-Hanti), 1.21 (s, Mesyn), 2.33
(s, Meanti), 7.50–7.43 (m, naph), 7.80–7.70 (m, naph), 7.92 (s,
H1syn, naph), 8.05 (s, H1anti, naph). 2b δ Ϫ11.88 (s, µ-H)a, Ϫ11.87
(s, µ-H)b, 0.64 (s, But)a, 1.40 (s, But)b, 7.38–7.49 (m, naph),
7.51–7.83 (m, naph), 7.95 [s, H1(naph)]a, 8.07 [s, H1(naph)]b. 2c
δ Ϫ10.77 (s, 1H, µ-H), 7.27–7.32 (m, 2H, py), 7.43–7.54 (m, 2H,
naph), 7.76–7.85 (m, 5H, naph, py), 8.15 [s, 1H, H1(naph)], 8.78
[s(broad), 2H, CH2], 2.53 [s(broad), 2H, CH2]; 7.39–7.95 (m,
26H, Ph, naph), 8.28 [s, 1H, H1(naph)]. 31P NMR (CDCl3): 4a δ
2.6 (s, dppm). 4b δ Ϫ1.64 (s, dppe).
Substitution of Re2(ꢀ-H)(ꢀ-SH)(CO)8 5 with PPh3
A mixture of 25 mg (0.095 mmol) of PPh3 and 60 mg (0.095
mmol) of 5 was dissolved at 0 ЊC in 10 ml of THF. Upon
addition of one equivalent of tmno the colour of the solution
turned pale yellow. After 20 min the colour turned to brown
indicating decomposition reactions. The reaction was brought
to an end after 1 h of stirring at room temperature by remov-
ing the solvent. The brown residue was worked up by TLC
(eluent dichloromethane–hexane = 1 : 2) giving one fraction
that contained Re2(µ-H)(µ-SH)(CO)6(PPh3)2 6 and an organic
phosphorus compound. The latter was removed by recrystal-
lisation from CHCl3–n-pentane affording pure 6. Elemental
analysis (yield): (38%) Found: C, 45.53; H, 2.65. C42H32O6-
P2SRe2 requires: C, 45.90; H, 2.93%. ν(CO) IR/cmϪ1 (THF):
3
2
(d, JHH = 5.2 Hz, 2H, py). 2d δ Ϫ13.83 (d, JPH = 8.3, µ-H)a,
Ϫ13.70 (d, 2JPH = 8.6 Hz, µ-H)b, 3.34 (d, 1JPH = 349 Hz, PH3)b,
3.99 (d, 1JPH = 348 Hz, PH3)a, 7.29–8.03 (m, naph). 2e δ Ϫ13.35
2
2
(d, JPH = 9.2 Hz, 1H, µ-H), 1.68 (d, JPH = 8.8 Hz), 7.46–7.58
(m, 2H, naph), 7.60–7.93 (m, 4H, naph), 8.13 [s, 1H, H1 (naph)].
2f δ Ϫ13.64 (d, 2JPH = 8.6 Hz, µ-H)e, Ϫ13.54 (d, 2JPH = 9.1 Hz,
1
2100 w, 2015 vs, 1934 s, 1917 s. H NMR (CDCl3): δ 12.88
2
3
(t, JPH = 7.7 Hz, 1H, µ-H), Ϫ1.05 (t, JPH = 9.9 Hz, 1H, SH),
7.45–7.58 (m, 30H, Ph). 31P NMR (CDCl3): δ 11.3 (s, PPh3),
15.8 (s, PPh3).
1
2
µ-H)f, 3.97 (dd, JPH = 350 Hz, JHH = 6.4 Hz, PH)b, 4.22 (dd,
1JPH = 350 Hz, JHH = 6.2 Hz, PH)b, 3.79 (s, OMe)a, 3.86
2
(s, OMe)b, 6.78–8.06 (m, 11H, naph, Ph). 2g δ Ϫ12.91 (d, 2JPH
=
8.6 Hz, 1H, µ-H), 7.44–7.83 (m, 21H, naph, Ph), 8.09 [s, 1H, H1
(naph)]. 2h δ Ϫ13.75 (d, 2JPH = 12.1 Hz, 1H, µ-H), 3.73 (d, 3JPH
Crystal structure determinations
=
11.4 Hz, 9H, Me), 7.41–7.53 (m, 2H, naph), 7.71–7.82 (m, 4H,
naph), 8.05 [s, 1H, H1(naph)]. 2i δ Ϫ11.88 (s, µ-H), 2.05
(s, broad, Me), 2.37 (s, broad, Me). 31P NMR (CDCl3): 2d
Pertinent crystallographic data for compounds 2a, 2c, 2g, 3a, 4b
and 6 are summarised in Table 2. All data sets were collected on
a Bruker AXS P4 diffractometer with graphite monochromated
Mo-Kα radiation. Standard reflections monitored after every
400 reflections showed only random deviations for 2a, 2c, 2g.
For 3a a decrease of 16% was monitored, for 4b 6% and for
6 7%. The intensities of these three data sets were corrected
accordingly. Intensities of all data sets were corrected for
Lorentz-polarisation effects and absorption corrections via
ψ-scans were applied. The structures were solved by direct and
conventional Fourier methods. Full-matrix, least-squares struc-
ture refinement based on F 2. All apart from hydrogen atoms
were refined anisotropically; geometrically placed hydrogen
atoms were refined with a ‘riding model’ and U(H) = 1.2U(Ciso).
The µ-H atom positions of structures 4b and 6 were determined
from ∆F maps. The µ-H atoms for 2a, 2c, 2g, 3a and the H atom
attached to µ-S in 6 were not located and not included in the
refinement. Programs used for calculations: SHELX-97.19
CCDC reference numbers 169418 2a, 139434 2c, 169419 2g,
169420 3a, 169421 4b, and 169422 6.
1
1
δ Ϫ160.4 (q, JPH = 349 Hz, PH3)a, Ϫ159.7 (q, JPH = 347 Hz,
PH3)b. 2e δ Ϫ40.9 (s, PMe3). 2f δ Ϫ73.2 [t, 1JPH = 351 Hz, H2P-
(p-C6HOMe)]a, Ϫ71.6 [t, 1JPH = 350 Hz, H2P(p-C6HOMe)]b. 2g
δ 10.9 (s, PPh3). 2h δ 118.9 [s, P(OMe)3].
Disubstituted complexes [Re2(ꢀ-H)(ꢀ-Snaph)(CO)6L2] [L ؍
NCMe (3a), pyridine (3b), PPh3 (3c)]. The preparation was
started with 200 mg of the monosubstituted complexes [Re2(µ-
H)(µ-Snaph)(CO)7L] [L = NCMe (0.269 mmol), pyridine (0.248
mmol), PPh3 (0.202 mmol)]. The complexes were reacted with
one equivalent of tmno as described above for the monosubsti-
tuted complexes. Addition of another equivalent of ligand L
and TLC [eluent dichloromethane–hexane = 1 : 1 (3a, 3b), 1 : 2
(3c)] gave pure [Re2(µ-H)(µ-Snaph)(CO)6L2]. Elemental analysis
(yield): 3a (73%) Found: C, 30.99; H, 1.80. C20H14O6N2Re2S
requires: C, 30.69; H, 1.80%. 3b (70%) Found: C, 35.99; 2.03.
C26H18O6Re2S requires: C, 36.36; H, 2.11%. 3c (75%) Found: C,
50.71; H, 2.54. C52H38O6P2Re2S requires: C, 50.97; H, 3.13%.
ν(CO) IR/cmϪ1 (THF): 3a 2031 m, 2013 vs, 1917 vs. 3b 2027 m,
See http://www.rsc.org/suppdata/dt/b1/b107859c/ for crystal-
lographic data in CIF or other electronic format.
J. Chem. Soc., Dalton Trans., 2002, 1078–1084
1083

View Article Online
Table 2 Crystallographic data
Compound
2a
2c
2g
3a
4b
6
Formula
M
C19H11NO7Re2S C22H13NO7Re2S C35H23O7PRe2Sؒtoluene C20H14N2O6Re2S C42H32O6P2Re2Sؒacetone C42H32O6P2Re2S
769.7
806.8
1083.1
782.8
1157.2
1099.1
Crystal system
Space group
Monoclinic
P21/n
Orthorhombic Triclinic
Monoclinic
P21/n
Monoclinic
P21/c
Triclinic
P1
¯
¯
Pca21
P1
T /K
a/Å
b/Å
c/Å
α/Њ
β/Њ
208(2)
298(2)
208(2)
298(2)
293(2)
293(2)
9.845(4)
13.269(4)
16.632(6)
15.464(5)
15.262(4)
20.333(3)
11.030(6)
12.673(6)
15.972(6)
76.24(2)
71.27(2)
67.31(2)
1934(2)
2
13.619(3)
13.340(3)
14.345(2)
16.715(4)
18.225(3)
15.022(3)
10.615(1)
10.683(1)
19.747(2)
75.86(1)
78.98(1)
68.87(1)
2012.3(3)
2
102.11(2)
116.48(2)
108.41(1)
γ/Њ
U/Å3
2124.3(13)
4
11.52
4799(2)
8
10.21
2332.7(8)
4
10.49
4342.0(15)
4
5.74
Z
µ(Mo-Kα)/mmϪ1
Reflections
measured/unique
6.39
8955/8652
6.19
10631/9164
4796/4657
6828/5936
6189/5144
11764/9962
R1 a/wR2
0.043/0.115
0.061/0.127
0.035/0.093
0.071/0.194
0.056/0.111
0.036/0.084
b
2
a R1[F > 4σ(F)] = Σ||Fo| Ϫ |Fc||/Σ|Fo|. b wR2(F 2, all data) = [Σw(Fo2 Ϫ Fc2)2/Σw(Fo )2]1/2
.
7 H. Egold and U. Flörke, Z. Anorg. Allg. Chem., 2001, 627,
2295.
8 H. Egold, D. Schwarze and U. Flörke, J. Chem. Soc., Dalton Trans.,
1999, 3203.
References
1 Recent examples: F. Takagi, H. Seino, Y. Mizobe and M. Hidai,
Can. J. Chem., 2001, 79, 632; A. R. Dias, M. H. Garcia, M. J. Villa
de Brito and A. Galvao, J. Organomet. Chem., 2001, 632, 75;
Y. Nishibayashi, I. Wakiji, K. Hirata, M. Rakowski DuBois and
M. Hidai, Inorg. Chem., 2001, 40, 578; S. M. T. Abedin, K. A. Azam,
M. B. Hursthouse, S. E. Kabir, K. M. A. Malik, M. A. Mottalib and
E. Rosenberg, J. Cluster Sci., 2001, 12, 5; M. M. Hossain, H.-M. Lin
and S.-G. Shyu, Eur. J. Inorg. Chem., 2001, 2655; M. Herberhold,
J. Peukert, M. Kruger, D. Daschner and W. Milius, Z. Anorg. Allg.
Chem., 2000, 626, 1289; R. S. Dickson, G. D. Fallon, W. R. Jackson
and A. Polas, J. Organomet. Chem., 2000, 607, 156.
9 (a) C. P. Perrin and T. J. Dwyer, Chem. Rev., 1990, 90, 935; (b) E. W.
Abel, T. P. J. Coston, K. G. Orrell, V. Sik and D. Stephenson,
J. Magn. Reson., 1986, 70, 34; (c) J. Jeener, B. H. Meier, P. Bachmann
and R. R. Ernst, J. Chem. Phys., 1979, 71, 4546.
10 C. A. Tolman, Chem. Rev., 1977, 77, 313.
11 (a) M. I. Bruce, P. J. Low, B. W. Skelton and A. H. White,
J. Organomet. Chem., 1996, 515, 65; (b) S. Top, M. Gunn, G. Jaouen,
J. Vaissermann, J. C. Daran and M. J. McGlinchey, Organometallics,
1992, 11, 1201; (c) K W. Lee and T. L. Brown, Organometallics, 1985,
4, 1025; (d ) P. O. Nubel, S. R. Wilson and T. L. Brown,
Organometallics, 1983, 2, 515.
2 E. W. Abel, S. K. Suresh and K. G. Orrell, Prog. Inorg. Chem., 1984,
32, 1.
3 (a) H. Torrens, Coord. Chem. Rev., 2000, 196, 331; (b) M. L. H.
Green and D. K. P. Ng, J. Chem. Soc., Dalton Trans., 1993, 11; (c)
E. W. Abel, N. A. Cooley, K. Kite, K. G. Orrell, V. Sik, M. B.
Hursthorse and H. M. Dawes, Polyhedron, 1987, 6, 1261; (d ) R. D.
Adams, D. A. Katahira and L. W. Yang, Organometallics, 1982, 1,
235; (e) I. B. Benson, S. A. R. Knox, P. J. Naish and A. J. Welch,
J. Chem. Soc., Dalton Trans., 1981, 2235; ( f ) S. D. Killops and S. A.
R. Knox, J. Chem. Soc., Dalton Trans., 1978, 1260; (g) G. Natile,
L. Maresca and G. Bor, Inorg. Chim. Acta, 1977, 23, 37.
4 P. M. Treichel and M. H. Tegen, J. Organomet. Chem., 1988, 358,
339; P. M. Treichel and P. C. Nakagaki, Organometallics, 1986, 5,
711; J. Grobe and R. Rau, J. Fluorine Chem., 1978, 11, 265.
5 V. Küllmer and H. Vahrenkamp, Chem. Ber., 1976, 109, 1560.
6 (a) H. Egold, S. Klose and U. Flörke, Z. Anorg. Allg. Chem., 2001,
627, 164; (b) J. Grobe, J. Vetter, B. Krebs and M. Pascaly, Z. Anorg.
Allg. Chem., 2000, 626, 430.
12 X. Zhang, C. A. Dullaghan, E. J. Watson, G. B. Carpenter and D. A.
Sweigart, Organometallics, 1998, 17, 2067; C. A. Dullaghan, G. B.
Carpenter, D. A. Sweigart, D. S. Choi, S. S. Lee and Y. K. Chung,
Organometallics, 1997, 16, 5688.
13 D. A. Vicic and W. D. Jones, Organometallics, 1997, 16, 1912.
14 V. Küllmer and H. Vahrenkamp, Chem. Ber., 1977, 110, 3799;
V. K üllmer and H. Vahrenkamp, Chem. Ber., 1976, 109, 1569; E. W.
Abel, A. M. Atkins, B. C. Crosse and G. V. Hutson, J. Chem. Soc. A,
1968, 687.
15 V. Küllmer and H. Vahrenkamp, Chem. Ber., 1977, 110, 3810.
16 A. G. Orpen, HYDEX, J. Chem. Soc., Dalton Trans., 1980, 2509.
17 U. Jüptner, PhD Thesis, Paderborn, Germany, 1993; E. Magnusson,
Phosphorus Sulfur, 1986, 28, 379.
18 E. Niecke and H. Westermann, Synthesis, 1988, 330.
19 G. M. Sheldrick, SHELX-97, A Program for Crystal Structure
Solution and Refinement, University of Göttingen, Germany, 1998.
1084
J. Chem. Soc., Dalton Trans., 2002, 1078–1084