Reactivity of the heteronuclear cluster Cp*IrOs3(μ-H) 2(CO)10 with some group 16 substrates

Article abstract of DOI:10.1016/j.jorganchem.2005.10.037

The mixed metal cluster Cp*IrOs₃(μ-H) ₂(CO)₁₀ (1) reacted readily with a number of group 16 substrates under chemical activation with TMNO. It reacted with C ₆H₅SH to afford the novel cluster Cp*IrOs ₃(μ-H)₃(CO)₉(μ-SPh) (2). It also reacted readily with Ph₃PSe to afford five new clusters, viz., Cp*IrOs₃(μ-H)₂(CO)₉(μ₃- Se) (3) Os₃(μ-H)₂(CO)₇(μ₃-Se) (PPh₃)₂ (4), Cp*IrOs₃(μ-H) ₂(CO)₉(PPh₃) (5), Cp*IrOs ₃(μ-H)₂(μ₃-Se)(CO)₈(PPh ₃) (6) and Cp*IrOs₃(μ-H) ₂(μ₃-Se)₂(CO)₇(PPh₃) (7). The reaction pathway for this reaction has been studied carefully and suggests that Ph₃PSe functioned primarily as a selenium atom transfer agent to give initially the even more reactive 3. The reaction of 1 with di-p-tolyl ditelluride yielded three new clusters, 8-10, which were non-interconverting stereoisomers with the formulation Cp*IrOs ₃(μ-H)₂(μ-Te-p-C₆H₄CH ₃)₂(CO)₈.

Full text of DOI:10.1016/j.jorganchem.2005.10.037

Journal of Organometallic Chemistry 691 (2006) 941–951
www.elsevier.com/locate/jorganchem
Reactivity of the heteronuclear cluster Cp*IrOs3(l-H)2(CO)10
with some group 16 substrates
*
Padmamalini Srinivasan, Mei En Tai, Weng Kee Leong
Department of Chemistry, National University of Singapore, Kent Ridge, Singapore 119260, Singapore
Received 30 September 2005; received in revised form 24 October 2005; accepted 25 October 2005
Available online 1 December 2005
Abstract
The mixed metal cluster Cp*IrOs3(l-H)2(CO)10 (1) reacted readily with a number of group 16 substrates under chemical activation
with TMNO. It reacted with C6H5SH to afford the novel cluster Cp*IrOs3(l-H)3(CO)9(l-SPh) (2). It also reacted readily with Ph3PSe to
afford five new clusters, viz., Cp*IrOs3(l-H)2(CO)9(l3-Se) (3) Os3(l-H)2(CO)7(l3-Se)(PPh3)2 (4), Cp*IrOs3(l-H)2(CO)9(PPh3) (5), Cp*Ir-
Os3(l-H)2(l3-Se)(CO)8(PPh3) (6) and Cp*IrOs3(l-H)2(l3-Se)2(CO)7(PPh3) (7). The reaction pathway for this reaction has been studied
carefully and suggests that Ph3PSe functioned primarily as a selenium atom transfer agent to give initially the even more reactive 3. The
reaction of 1 with di-p-tolyl ditelluride yielded three new clusters, 8–10, which were non-interconverting stereoisomers with the formu-
lation Cp*IrOs3(l-H)2(l-Te-p-C6H4CH3)2(CO)8.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Osmium; Iridium; Cluster; Thiol; Phosphine selenide; Ditelluride; Stereoisomers
1. Introduction
some group 16 substrates. We have chosen representatives
from three different classes of group16 substrates to be
examined, viz., a thiol, a phosphine selenide and a
ditelluride.
Reasons for the continued interests in transition metal
carbonyl clusters containing chalcogens include (a) the
ability of the chalcogens to act as stabilizing ligands, thus
preventing cluster fragmentation even under forcing reac-
tion conditions [1], (b) the chalcogen, especially the hea-
vier members, appear to be a key factor in cluster
growth reactions [2], and (c) the unusual coordination
modes and geometries they exhibit [3]. Much has been
reported on the reactivity of homonuclear carbonyl clus-
ters with group 16 substrates, but relatively little is known
on the reactivity of mixed-metal clusters. We have recently
initiated investigations into some osmium–iridium mixed-
metal clusters containing cyclopentadienyl ligands [4]. In
this paper, we would like to report our investigations into
the reactivity of the cluster Cp*IrOs3(l-H)2(CO)10 (1) with
2. Results and discussion
2.1. Reaction with thiophenol
Sulphur-containing clusters of the Group 8 and 9 metals
are of interest as hydrodesulfurization catalysts [1b]. Thiols
are known to react with homonuclear clusters via oxidative
addition across the S–H bond [5]. Interestingly, the reac-
tion involving the cluster Os3(CO)11(NCCH3) has been
reported to proceed via an intermediate with an agostic
interaction [6]. In contrast, there appears to have been only
one report on the reaction of a heteronuclear carbonyl
cluster with thiols [7].
Reaction of 1 with thiophenol under chemical activation
with trimethylamine N-oxide (TMNO) afforded bright
orange crystalline Cp*IrOs3(l-H)3(CO)9(l-SPh) (2).
*
Corresponding author.
E-mail address: chmlwk@nus.edu.sg (W.K. Leong).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.10.037

942
P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
Cluster 2 has been characterized completely, including by a
single crystal X-ray crystallographic analysis; the ORTEP
plot together with selected bond parameters are shown in
Fig. 1. Thus, the reaction involved oxidative addition of
the thiol S–H bond via an Os–Os bond cleavage and a
decarbonylation.
The molecular structure of 2 is that of a wingtip-bridged
butterfly cluster; an osmium and the iridium atom occupy
the hinge of the butterfly, and the benzene thiolate group
bridges the wingtips. The Os(3)ꢀ ꢀ ꢀOs(4) vector at
˚
3.6970(8) A is clearly non-bonding. There is a hydride
bridging each of the three iridium–osmium edges. Bridging
hydrides tend to elongate metal–metal bonds [8], and this is
evident in the three iridium–osmium bond lengths [Ir(1)–
˚
˚
˚
Os(2) = 2.9037(3) A, Ir(1)–Os(3) = 2.9242(3) A, and Ir(1)–
Os(4) = 2.9327(3) A]. The thiolate bridge is unsymme-
˚
trical [Os(3)–S(5) = 2.4343(15) A, Os(4)–S(5) = 2.4257(14)
˚
A], but the bond parameters are similar to those in, for
example, HOs3(l-SMe)(l-g2-C6H4)(CO)9 [2.418(4) and
Fig. 1. ORTEP plot, with selected bond parameters, of 2. Thermal
ellipsoids are drawn at 50% probability level. Organic hydrogens are
˚
2.433(5) A] [9].
˚
˚
omitted for clarity. Ir(1)–Os(2) = 2.9037(3) A; Ir(1)–Os(3) = 2.9242(3) A;
˚
˚
Ir(1)–Os(4) = 2.9327(3) A; Os(2)–Os(3) = 2.8387(3) A; Os(2)–Os(4) =
2.2. Reaction with triphenylphosphine selenide
˚
˚
˚
2.8448(3) A; Os(3)–S(5) = 2.4343(15) A; Os(4)–S(5) = 2.4257(14) A; S(5)–
Os(3)–Os(2) = 78.00(3)ꢁ; S(5)–Os(4)–Os(2) = 78.01ꢁ; Os(3)–Os(2)–Ir(1) =
61.213(7)ꢁ; Os(4)–Os(2)–Ir(1) = 61.340(7)ꢁ.
The employment of tertiary phosphine selenides R3PSe
has been shown to be an effective method for synthesiz-
ing selenium-containing transition metal clusters. This
method takes advantage of the frailty of the P@Se bond,
which leads to its formal oxidative-addition onto the
cluster, resulting in the transfer of the selenium atom
[10].
The reaction of 1 with Ph3PSe under TMNO activation
proceeded at room temperature to afford up to four novel
selenium-bridged osmium–iridium clusters, as depicted in
Scheme 1. The cluster 5, a phosphine-substituted derivative
of 1, has been reported earlier [4]. It moved together as one
band on the TLC plate with cluster 4, and had to be
mechanically separated after crystallization; 4 and 5
appeared as orange-yellow and dark red crystals, respec-
tively. All the novel compounds 3, 4, 6 and 7, have been
characterized, including by single crystal X-ray crystallo-
graphic analyses.
The ORTEP plots of 3 and 6 are given in Fig. 2, and
common atomic numbering scheme and selected bond
parameters for them are tabulated in Table 1. Cluster 6 is
clearly a phosphine-substituted derivative of 3. Their
molecular structures comprise a butterfly tetrahedral core,
with the iridium and an osmium atom making up the wing-
tips. The Se atom is triply bridging across the wingtips and
one of the hinge Os atoms. The wing edges not spanned by
the selenium are bridged by a hydride each; although the
Se
Se
Ir
Ir
Os
H
O
Ph3P
O
s
s
Os
Os
Os
H
H
H
Os
H
Os
PPh3
H
Ir
PPh3
Os
5
3
4
2.6 eqTMNO
(22%)
(17%)
(<1%)
Ph3PSe
+
Os
Os
H
Se
Os
1
H
Ir
Ir
Os
H
PPh3
Se
Os
Se
H
H
Os
Os
H
7
6
Os
Ph3P
Os
(13%)
(22%)
Scheme 1.

P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
943
Fig. 2. ORTEP diagrams of 3 (left) and 6 (right). Thermal ellipsoids are drawn at 50% probability level. Organic hydrogens are omitted for clarity.
locations of these hydrides were placed by potential energy
calculations [11], their positions are also corroborated by
low angle electron density difference maps which show
residual electron densities suggesting hydrides bridging
those same edges. The 1H NMR spectra are also consistent
with these positions; that for 3 showing two doublets
hydride bridging an Os–Os bond. The interesting point
here is that, quite contrary to expectations, the hydride
bridging the Ir–Os bond in each structure does not bridge
the longer of the two such bonds! In comparing the
bond parameters between 3 and 6, although some of their
corresponding bond parameters are significantly different,
there does not appear to be any discernible trend. A final
point of interest is that the Os–P bond length in 6 is rather
short.
of equal intensities at
d
ꢁ16.90 and ꢁ15.03 ppm
(2JHH = 3.3 Hz), and that for 6 showing a doublet at d
ꢁ16.87 ppm (2JHH = 3.3 Hz) assignable to the hydride
bridging the osmium–iridium bond, and a doublet of dou-
blets at d ꢁ13.95 ppm (2JPH = 11.6 Hz) assignable to the
The molecular structure of 4 is shown in Fig. 3, together
with selected bond parameters. It is structurally similar to
Table 1
Common atomic numbering scheme and selected bond parameters for 3
and 6
Se
L
Ir
Os4
H
Os2
H
Os3
3
6
L
CO
PPh3
˚
Bond length (A)
Ir–Os2
Ir–Os3
Os4–Os2
Os4–Os3
Os2–Os3
Ir–Se
Os4–Se
Os2–Se
Os4–L
2.8927(5)
2.8163(5)
2.8123(5)
2.9484(5)
2.8477(5)
2.4057(9)
2.5242(9)
2.5196(9)
2.8512(7)
2.8395(7)
2.8135(7)
2.9812(7)
2.8489(7)
2.4038(13)
2.5100(13)
2.5379(12)
2.316(3)
Fig. 3. ORTEP diagram of 4. Thermal ellipsoids are drawn at 50%
–
probability level. Organic hydrogens are omitted for clarity. Os(1)–Os(2)=
Bond angle (ꢁ)
Os2–Ir–Os3
Os2–Os4–Os3
Dihedral angle between
Os2Os3Ir and Os2Os3Os4
˚
˚
˚
2.9858(4) A;
Os(1)–Os(3) = 2.7940(5) A;
Os(2)–Os(3) = 2.9419(4) A;
59.823(12)
59.195(13)
104.3
60.082(7)
58.811(17)
105.3
˚
˚
Os(1)–Se(4) = 2.4957(9) A; Os(2)–Se(4) = 2.5042(9) A; Os(3)–Se(4) =
˚
˚
˚
2.5020(9) A; Os(1)–P(1) = 2.334(2) A; Os(2)–P(2) = 2.356(2) A; Os(3)–
Os(1)–Os(2) = 61.089(11)ꢁ; Os(3)–Os(2)–Os(1) = 56.236(10)ꢁ; Os(1)–
Os(3)–Os(2) = 62.675(10)ꢁ.

944
P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
˚
bridge the longest edge [Os(2)–Os(4) = 3.0043(5) A]; the
lengthening of this bond is presumably in part due to
trans influence of the PPh3 ligand [P(7)–Os(4)–Os(2) =
173.39(6)ꢁ]. In fact, the Os(3)–Os(4) bond is remarkably
short, even for an Os–Os bond not bridged by a hydride;
˚
it should be contrasted with the mean value of 2.877 A in
˚
Os3(CO)12 [12], or the ꢂ2.84 A in Os3(CO)9(l-Se)2 [13],
although the reason for this is not clear. The four Os–Se
bond lengths vary over a fairly wide range, from
˚
2.4674(10) to 2.5613(10) A (for Os(3)–Se(5) and Os(4)–
Se(6), respectively).
We have carried out a series of reactions on 3, 5 and 6
with PPh3 and Ph3PSe, both in the presence and the
absence of TMNO, to determine how the various clusters
3–7 are related in the reaction pathway. Monitoring the
1
original reaction of 1 by H NMR spectroscopy showed
that 3–7 were formed directly from the reaction. Forma-
tion of 5 evidently resulted from the reaction of free
PPh3 with 1 under TMNO activation [4]. The liberation
of PPh3 during the reaction was confirmed by both 31P
NMR spectroscopy and FAB MS, and also suggests that
Ph3PSe functioned primarily as a selenium transfer agent.
This function was again evident when 5 in turn reacted
with Ph3PSe in the presence of TMNO to afford 6. It is
also in accord with our previous findings [10g]. We have
found that 3 reacted with PPh3 even in the absence of
TMNO, to afford 4 and 6, and forms 7 when reacted with
Ph3PSe under TMNO activation; these point to 3 as the
precursor to 4, 6 and 7. Interestingly, 6 did not give rise
to 7 either with PPh3 or Ph3PSe both in the presence or
absence of TMNO. It was also found that a 1:2.6:2.6 ratio
of 1:TMNO:Ph3PSe was required to ensure completion of
reaction of 1; reducing the ratio of TMNO or Ph3PSe
resulted in recovery of a large amount of unreacted 1. This
evidently points to the subsequent reactions of 3, or the
formation of 1, being more rapid than the formation of
3 itself. We have thus proposed a reaction sequence as
given in Scheme 2. The formation of 4 from 3 requires
the loss of a Cp*IrCO fragment, although we do not know
at this point what is the ultimate fate of this fragment. An
example of such a cluster fragmentation with Ph3PSe is
that of the reaction of RuCo3(l-H)(CO)12, which resulted
in the loss of a Co fragment to give trinuclear clusters
[10e].
Fig. 4. ORTEP diagram of 7. Thermal ellipsoids are drawn at 50%
probability level. Organic hydrogens are omitted for clarity. Ir(1)–Os(2) =
˚
˚
˚
2.9232(5) A;
Os(2)–Os(3) = 2.9154(6) A;
Os(2)–Os(4) = 3.0043(5) A;
˚
˚
Os(3)–Os(4) = 2.7423(5) A; Ir(1)–Se(5) = 2.4753(10) A; Ir(1)–Se(6) =
2.4967(10) A; Os(2)–Se(5) = 2.5530(10) A; Os(3)–Se(5) = 2.4674(10) A;
Os(3)–Se(6) = 2.5346(10) A;
˚
˚
˚
˚
˚
Os(4)–Se(6) = 2.5613(10) A;
Os(4)–P(7)
˚
= 2.337(2) A; Os(3)–Os(2)–Ir(1) = 77.214(14)ꢁ; Os(3)–Os(2)–Os(4) =
55.168(12)ꢁ;
60.770(13)ꢁ.
Os(4)–Os(3)–Os(2) = 64.062(14)ꢁ;
Os(3)–Os(4)–Os(2) =
that of the previously reported clusters Os3(CO)9(l3-
Se)(PPh3) (4a) and Os3(l-H)2(CO)8(l3-Se)(PPh3) (4b)
[10g]. Indeed, 4 is a further substituted analogue of 4b, in
which the additional phosphine (P(1)) occupies a pseudo-
equatorial position on an adjacent osmium. As may be
expected, the further substitution of a phosphine ligand
has led to further elongation of the Os(1)–Os(2) bond that
lies in between the two phosphines, although the two cor-
responding bonds in 4b are rather unsymmetrical
˚
˚
[2.9858(4) A compared to 2.9440(4) and 2.9961(4) A in 4
and 4b, respectively. In general, the bond lengths in 4 are
slightly elongated with respect to those in 4b, no doubt
the result of increased electron density on the cluster frame-
work from the additional phosphine.
The molecular structure of 7 is shown in Fig. 4, together
with selected bond parameters. Its structure suggests that it
may be regarded as derivable from 6 by the capping of the
IrOs2 wing with an additional l3-Se fragment. The
˚
Ir(1)ꢀ ꢀ ꢀOs(3) vector is >3.6 A, and hence is clearly non-
2.3. Reaction with di-p-tolyl ditelluride
bonding. The hydride positions as obtained from a poten-
tial energy calculation [11], were corroborated by a low
The chemistry of tellurium-containing clusters can be
quite different from those of the sulphur or selenium ana-
logues. For instance, the difference in reactivity of the trinu-
clear iron clusters, Fe3(CO)9(l3-E)2, (E = S, Se, Te),
towards Lewis bases has been attributed to the larger size
of the Te atom which results in a more strained Fe–Te–Fe
angle in Fe3(CO)9(l3-Te)2 than in its sulphur and selenium
analogues [14]. Nevertheless, reports on the reaction of clus-
ters with diorganoditellurides appear to be scant [15]. The
reaction of 1 with di-p-tolylditelluride at ambient tempera-
1
angle electron density difference map and the H NMR
spectrum which showed a doublet at d ꢁ12.10 ppm
(2JPH = 8.3 Hz) assignable to the hydride bridging the
Os(3)–Os(4) edge, and a singlet at ꢁ18.72 assignable to that
bridging the Ir(1)–Os(2) edge; no 1H–1H coupling was
2
observable. The relatively small JPH value is also consis-
tent with the hydride bridging the Os(3)–Os(4) edge which
is cis to the phosphine, rather than the Os(2)–Os(4) edge
which is trans to it. As in 3 and 6, neither of the hydrides

P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
945
Ir
Ir
Ph3P, TMNO
Os
Os
H
Os
Os
H
PPh3 substitution
Os
H
Os
H
5
PPh3
1
Ph3PSe, TMNO
Ph3PSe, TMNO
Se transfer
Se transfer
PPh3
PPh3
Se
Se
Ir
Os
H
Ir
Os
H
PPh3
PPh3
Os
Os
substitution
H
3
H
Os
Os
6
oxidative addition
Ph3PSe, TMNO
- "Cp*IrCO"
Se
PPh3
substitution
Ir
Se
H
H
Se
H
Os
PPh3
4
Os
H
Os
Ph3P
Os
7
Scheme 2.
ture under chemical activation with TMNO afforded an
orange-red solution which after chromatographic separa-
tion afforded three new compounds 8–10, which were
stereoisomers having the formulation, Cp*IrOs3(l-H)2(l-
Te-p-C6H4CH3)2(CO)8 (Scheme 3). All three have been char-
acterized, including by single crystal X-ray crystallography,
and their ORTEP plots are shown in Figs. 5–7, respectively.
Thus, the reaction of 1 with the diorganoditelluride is
that of an oxidative addition reaction, upon decarbonyla-
over a period of time at ambient temperature did not show
any interconversion between the isomers; 10 was not suffi-
ciently stable in solution over any appreciable period of
time to allow a similar check on isomerization but its sep-
aration chromatographically alludes to a slow isomeriza-
tion process, if any.
The crystal quality for 9 was, unfortunately, rather poor
and the structure was modeled as exhibiting disorder of the
heavy atom positions. Nevertheless the gross structural fea-
tures are discernible. All three clusters have an open butter-
fly IrOs3 metal core, with the iridium at one of the wingtips,
and a telluride bridging each of the two Os–Os of the Os3
wing. Besides the Cp*, the iridium also carries a carbonyl
group. Clusters 8 and 9 are isomers differing in the relative
orientation of the tolyl groups; in 8, one of the tolyl is ori-
entated away (exo) from the cluster core and the other
inwards (endo), while in 9, both tolyl groups are exo. Clus-
ter 10 differs from 8 and 9 in that (a) one of the tellurium is
oriented inwards towards the butterfly rather than away as
in the others, (b) the wingtip osmium has only two instead
1
tion with TMNO. The H NMR spectrum of the crude
reaction mixture showed the presence of all three isomers.
Thus the products were not formed via decomposition on
silica-gel plates, as has been observed in a related system
[15a]. The 1H NMR spectra of 8, 9 and 10 showed two res-
onances each at high field, indicative of bridging metal
hydrides. Two methyl signals each were observed for 8
and 10, and one for 9, consistent with their solid-state
structures. The aromatic resonances for 8 were also par-
tially assigned through selective decoupling experiment.
1
Solutions of 8 and 9 (d6-benzene) monitored by H NMR

946
P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
Ir
8
Os
H
Os
H
(5.6%)
Os
Te
tolyl
Te
tolyl
1
+ (p-tolyl)2Te2 + TMNO
Ir
9
Os
H
Os
tolyl
H
(10.6%)
Os
Te
tolyl
Te
H
Ir
10
(3.4%)
Te
Os
Os
tolyl
H
Os
tolyl
Te
Scheme 3.
Fig. 6. ORTEP diagram of the major component of the disordered
molecule of 9. Thermal ellipsoids are drawn at 50% probability level.
Organic hydrogens are omitted for clarity.
Fig. 5. ORTEP diagram of 8. Thermal ellipsoids are drawn at 50%
probability level. Organic hydrogens are omitted for clarity.
energy calculations indicated the positions given, which are
also consistent with those in 8. A common atomic number-
ing scheme, together with selected bond parameters, for 8–
10 are tabulated in Table 2.
Strangely, as in 3, 6 and 7, the hydrides did not always
bridge the longer metal–metal bonds. For instance, one of
the hydrides bridges the shortest Os–Os bond in 8, and the
longest hinge bonds among the three clusters is that of 10
which is not bridged by an hydride. The longest metal–
metal bonds are, in fact, bridged by a telluride. The
bridging hydrides though, manifest themselves in the asym-
of three carbonyls, this third carbonyl being now located at
one of the hinge osmiums, and (c) the positions of the
hydrides. The positions of the hydrides were located by a
potential energy calculation [11]. In the case of 8, the posi-
tions are corroborated by a low angle electron density map.
In 10, the hydrides failed to show up in the electron density
map, but the carbonyls along that edge [CO(32) and
CO(42)] are bent away from each other, indicative of the
steric influence of a hydride there. The disorder in 9 pre-
cluded reliable location of the hydrides, although potential

P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
947
the Cp*Ir moiety. Indeed this appears to be the case, the
Os4–Te6 bond being significantly longer than the Os3–Te6
˚
bond (2.7223(7) and 2.6180(7) A, respectively) and quite
unlike that for the other (exo) tellurium moieties which
all appear to lean inwards. This steric effect is also evident
in the increase in the dihedral angle between the IrOs2Os4
and Os2Os3Os4 planes in 10 (127.7ꢁ as compared to
121.3ꢁ in 8).
3. Concluding remarks
In this work, we have presented our findings on the
reactivity of 1 with three classes of Group 16 substrates.
Cluster 1 proved to be very reactive with these substrates
under chemical activation, all reactions proceeding at
ambient temperatures. With thiophenol and diorganodi-
telluride, 1 undergoes oxidative addition reactions. In
the latter case, stereoisomers are obtained. With phos-
phine selenide, selenium atom transfer seems to be the ini-
tial step but the product so obtained is more reactive and
quickly reacts further. One unusual structural feature in
many of these products is the presence of very short
metal–metal bonds, including some which are bridged
by a hydride. It, therefore, seems certain that this class
of reactions will turn up more unusual structures and
reactivities to come.
Fig. 7. ORTEP diagram of 10. Thermal ellipsoids are drawn at 50%
probability level. Organic hydrogens are omitted for clarity.
metry of the Ir–Os bond lengths. It may be expected that
the unusual position of Te6 in 10, being on the endo face
of the butterfly, would experience steric repulsion from
Table 2
A common atomic numbering scheme and selected bond lengths (A) and angles (ꢁ) for 8–10
˚
H
Ir
Ir
Ir
Te6
Os2
Os2
H
Os3
Os2
H
Os3
Os3
H
H
H
Os4
Os4
Te5
Te6
Te5
Os4
10
Te5
9
8
Te6
Compound
8
9a
10
˚
Bond lengths (A)
Ir–Os2
Ir–Os4
Os2–Os3
Os3–Os4
Os2–Os4
Os2–Te5
Os3–Te5
Os3–Te6
Os4–Te6
2.7519(5)
2.9299(4)
2.9817(4)
3.0086(4)
2.7803(4)
2.6149(7)
2.7293(6)
2.6927(6)
2.6232(6)
2.835(3)
2.835(3)
2.983(2)
2.983(2)
2.810(4)
2.612(4)
2.699(3)
2.708(3)
2.618(3)
2.7842(5)
2.9602(5)
2.8274(5)
2.9776(5)
2.9188(5)
2.6351(6)
2.6385(7)
2.6180(7)
2.7223(7)
Bond angles (ꢁ)
Os2–Ir–Os4
Os2–Os3–Os4
Os2–Te5–Os3
Os2–Te6–Os3
58.494(11)
55.306(10)
67.786(17)
68.926(16)
59.41(7)
56.18(7)
68.32(8)
68.12(8)
60.984(12)
60.305(12)
64.843(17)
67.743(17)
Dihedral angles (ꢁ)
IrOs2Os4 and Os2Os3Os4
Os2Os3Os4 and Os2Os3Te5
Os2Os3Os4 and Os3Os4Te6
Os2Os3Te5 and Os3Os4Te6
121.3
128.9
123.3
91.5
121.6
121.3
121.4
97.8
127.7
122.8
129.5
131.8
a
Disordered.

948
P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
4. Experimental
(5 ml) was added a solution of TMNO (4.9 mg,
0.044 mmol) in CH2Cl2 (10 ml), and the mixture stirred
for 1 h. Removal of the solvent by rotary evaporation fol-
lowed by chromatographic separation on silica-gel TLC
plates with hexane and dichloromethane (4:1, v/v) as eluant
afforded, in order of elution, Cp*IrOs3(l-H)2(CO)9(l3-Se),
3 (3.6 mg, 17%); unreacted 1 (2.3 mg); Cp*IrOs3(l-
H)2(CO)9(PPh3) (5) (5.2 mg, 22%, identified spectroscopi-
cally [4]); Os3(l-H)2(CO)7(l3-Se)(PPh3)2 (4) (0.1 mg,
<1%); Cp*IrOs3(l-H)2(l3-Se)(CO)8(PPh3) (6) (3.2 mg,
13%); and Cp*IrOs3(l-H)2(l3-Se)2(CO)7(PPh3) (7) (5.6
mg, 22%). Clusters 4 and 5 travelled as one band on the
TLC plate and had to be mechanically separated after
crystallization.
All reactions were carried out using standard Schlenk
techniques under an atmosphere of nitrogen. Solvents used
in reactions were of AR grade, and were dried, distilled and
kept under argon in flasks fitted with Teflon valves prior to
use. The products were generally separated by thin-layer
chromatography (TLC), using plates coated with silica
gel 60 F254 of 0.25 mm or 0.5 mm thickness and extracted
with hexane or dichloromethane. Infrared spectra were
recorded as CH2Cl2 solutions unless otherwise stated on
a Bio-Rad FTS 165 FTIR spectrometer at a resolution of
1 cmꢁ1 using a solution cell with NaCl windows of path
length 0.1 mm. NMR spectra were acquired on a Bruker
ACF 300 MHz as CDCl3 solutions unless otherwise stated.
Selective decoupling experiments were carried out on a
Bruker Avance DRX500 or Bruker AMX500 machine.
Chemical shifts reported are referenced to residual protons
of the solvent.
Mass spectra were collected using the fast atom bom-
bardment (FAB) technique and were carried out on a
Finnigan MAT95XL-T mass spectrometer normally with
3-nitrobenzyl alcohol matrix. Microanalyses were carried
out by the microanalytical laboratory at the National
University of Singapore. The cluster 1 [4], and di-p-tolyl
ditelluride [16], were prepared according to published pro-
cedures. All other reagents were from commercial sources
and used as supplied.
3: IR (hexane): mCO 2065vs, 2059vs, 2046m, 2024vs,
1
2008m, 2000m, 1989s, 1967w, 1950m cmꢁ1. H NMR: d
2
1.41 (s, 15H, Cp*), ꢁ15.03 (d, 1H, JH–H = 3.3 Hz,
OsHOs), ꢁ16.90 (d, 1H, IrHOs). FAB-MS (m/z): 1230.8
[M+]. Calc. for C19H17IrO9Os3Se: C, 18.54; H, 1.39.
Found: C, 18.70; H, 1.45%.
4: IR: mCO 2057s, 2038vs, 1991vs, 1976vs, 1956sh, 1923w
cmꢁ1. 1H NMR: d 7.51–7.32 (m, 30H, Ph) ꢁ19.21 (dd, 1H,
2
2JP–H = 7.4 Hz, 7.4 Hz, OsHOs), ꢁ19.30 (d, 1H, JP–H
=
12.4 Hz, OsHOs). 31P{1H} NMR: d 1.22 (s), ꢁ4.55 (s).
Hi-res MS (m/z): 1371.9658. Calc. for C43H32P2O7-
78Se188Os192Os2: 1377.9585.
6: IR: mCO 2063m, 2019vs, 1996m, 1974sh, 1948w
cmꢁ1 1H NMR: d 7.56–7.42 (m, 15H, Ph), 1.59 (s,
.
2
2
15H, Cp*), ꢁ13.95 (dd, 1H, JP–H = 11.6 Hz, JH–H
=
3.3 Hz, OsHOs), ꢁ16.87 (d, 1 H, IrHOs). 31P{1H}
NMR: d 5.05 (s). FAB-MS (m/z): 1465.35 [M+]. Calc.
for C36H32IrO8Os3PSe: C, 29.50; H, 2.20. Found: C,
29.70; H, 2.25%.
4.1. Reaction of 1 with thiophenol
To a 250 ml three necked flask containing 1 (30.0 mg,
0.0254 mmol) in dichloromethane (10 ml) was added excess
thiophenol (4–5 drops). A solution of trimethylamine N-
oxide (3.4 mg, 0.0304 mmol) in deoxygenated dichloro-
methane (20 ml) was then introduced dropwise via a
pressure equalizing dropping funnel over a period of 2 h.
The solution was stirred for a further 1 h. Removal of
the solvent by rotary evaporation followed by chromato-
graphic separation (6:4, v/v, hexane/dichloromethane) on
silica gel TLC plates yielded a broad orange-red band of
unreacted 1 (10 mg) and a second broad dark orange band
of Cp*IrOs3(l-H)3(CO)9(l-SPh), 2 (18.7 mg, 58%).
7: IR: mCO 2060vs, 2005sh, 1991vs, 1979sh, 1939w,
1
1910w cmꢁ1. H NMR: d 7.83–7.76 (m, 15H, Ph), 1.49
2
(s, 15H, Cp*), ꢁ12.10 (d, 1H, JP–H = 8.3 Hz, OsHOs),
ꢁ18.72 (s, 1H, IrHOs). 31P{1H} NMR: d 12.30 (s). Hi-res
MS (m/z): 1514.8425. Calc. for C35H29O7PIrSe2188Os190
-
Os192Os: 1514.8414.
Diffraction-quality crystals for 3 were obtained from
hexane solution by slow cooling, and by slow diffusion
of hexane into a dichloromethane solution for 4, 5 and
6.
IR (CH2Cl2): mCO 2072m, 2049vs, 2016s, 1989ms,
4.3. Reaction of 3 with PPh3
1
1968m, 1936mw,br cmꢁ1. H NMR: d 7.34–6.90 (m, 5H,
Ph), 2.34 (s, 15H, Cp*), ꢁ15.58 (s, 2H, OsHIr), ꢁ16.60
(s, 1H, OsHIr). FAB-MS (m/z): 1261 [M+]. Calc. for
C25H23IrO9Os3S: C, 23.79; H, 1.84; S, 2.54. Found: C,
23.73; H, 2.17; S, 2.38%.
Diffraction quality crystals were grown from hexane and
dichloromethane by slow diffusion.
To a solution of 3 (5 mg, 0.004 mmol) in dichloro-
methane (5 ml) was added PPh3 (1.1 mg, 0.004 mmol)
and the mixture stirred at room temperature for 3 h.
Formation of 4 and 6 were verified by 1H NMR
spectroscopy.
4.4. Reaction of 3 with Ph3PSe
4.2. Reaction of 1 with Ph3PSe
To a solution of 3 (5 mg, 0.004 mmol) and Ph3PSe
(2.8 mg, 0.008 mmol) in dichloromethane was added
dropwise a solution of TMNO (1.0 mg, 0.008 mmol) in
In a typical reaction, to a solution of 1 (20.0 mg,
0.017 mmol) and Ph3PSe (15.3 mg, 0.045 mmol) in CH2Cl2

P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
949
dichloromethane (10 ml) at room temperature. The mix-
ture was stirred for 1 h. Formation of 7 was verified by
1H NMR spectroscopy.
orange solid of Cp*IrOs3(l-H)2(l-Te-p-C6H4CH3)2(CO)8
(8) (Rf = 0.57, 3.0 mg, 5.6%), a dark orange solid of 9
(Rf = 0.53, 5.6 mg, 10.6%), and pink crystals of 10
(Rf = 0.15, 1.8 mg, 3.4%).
4.5. Reaction of 5 with Ph3PSe
8: IR: mCO 2057s, 2009s, 1997vs, 1980m, 1960m, 1936ms,
1
3
1904m cmꢁ1. H NMR (C6D6): d 7.41 (d, 2H, JH–H
=
3
0
To a solution of 5 (5 mg, 0.003 mmol) and Ph3PSe
(1.2 mg, 0.003 mmol) in dichloromethane (10 ml) was
added TMNO (0.4 mg, 0.003 mmol) and the reaction mix-
ture stirred for 3 h. Formation of 6 was verified by H
NMR spectroscopy.
7.85 Hz, C6H2bH2b CH3), 7.30 (d, 2H, JH–H = 7.85 Hz,
0
0
C6H2aH2a CH3), 6.59 (d, 2H, C6H2bH2b CH3), 6.58 (d, 2H,
0
C6H2aH2a CH3), 2.04 (s, 15H, Cp*), 1.93 (s, 3H, CH3),
1
1.85 (s, 3H, CH3), ꢁ12.83 (s, 1H, MHM); ꢁ13.38 (s, 1H,
MHM). FAB-MS (m/z): 1561.1 [M+]. Calc. for C32H31Ir-
O8Os3Te2. 1/4C6H14: C, 25.41; H, 2.19. Found: C, 25.37;
H, 2.00%. Presence of hexane was confirmed by 1H
NMR spectroscopy.
4.6. Reaction of 1 with di-p-tolyl ditelluride
To a 250 ml three-necked flask containing 1 (40.2 mg,
0.0356 mmol) in dichloromethane (10 ml) was added di-p-
tolylditelluride (31.2 mg, 0.0711 mmol). A solution of
trimethylamine N-oxide (8 mg, 0.07 mmol) in dichloro-
methane (20 ml) was deoxygenated and then introduced
dropwise into the solution of 1 via a pressure equalizing
dropping funnel over a period of 2 h. The solution was stir-
red for a further 1 h. Removal of the solvent by rotary
evaporation followed by chromatographic separation
(9:1, v/v, hexane/dichloromethane) on silica gel TLC plates
yielded unreacted di-p-tolylditelluride (Rf = 0.7, trace
amounts) identified from its 1H NMR spectrum, unreacted
1 (Rf = 0.65, 3.5 mg) identified from its IR spectrum, an
9: IR: mCO 2058s, 2010s, 1997s, 1979m, 1960m, 1937ms,
1906m cmꢁ1 1H NMR (C6D6): d 7.57 (d, 4H, JH–H
. =
7.4 Hz, C6H4CH3), 6.73 (d, 4H, JH–H = 8.3 Hz,
C6H4CH3), 2.01 (s, 15H, Cp*), 1.93 (s, 6H, 2 · Me),
ꢁ13.16 (s, 1H, MHM); ꢁ15.11 (s, 1H, MHM). FAB-MS
(m/z): 1564.1 [M+]. Calc. for C32H31IrO8Os3Te2: C,
24.61; H, 2.00. Found: C, 24.51; H, 2.48%.
10: IR: mCO 2063ms, 2040mw, 2026vs, 2012ms, 1982m,
1
3
1958m cmꢁ1. H NMR (C6D6): d 7.71 (d, 2H, JH–H
=
0
0
8.2 Hz, C6H2aH2a CH3), 7.57 (d, 2H, C6H2aH2a CH3), 6.83
3
0
(d, 2H, JH–H = 7.4 Hz, C6H2bH2b CH3), 6.77 (d, 2H,
0
C6H2bH2b CH3), 1.68 (s, Cp*), 1.99 (s, 3H, CH3), 1.96 (s,
3H, CH3); ꢁ18.45 (s, 1H, MHM); ꢁ20.42 (s, 1H,
Table 3
Crystal and structure refinement data for compounds 3, 4, 6 and 7
Compound
3
4
6
7
Empirical formula
Formula weight
Crystal system
C
19H17IrO9Os3Se
C43H32O7Os3P2Se
1372.19
Monoclinic
P21
C36H32IrO8Os3PSe
1465.35
Triclinic
C35H32IrO7Os3PSe2
1516.30
Triclinic
1231.09
Monoclinic
P21/n
ꢀ
ꢀ
Space group
P1
P1
Unit cell dimensions
˚
a (A)
8.9022(2)
19.5490(4)
14.5565(3)
90
93.829(1)
90
2527.60(9)
4
3.235
9.0916(3)
20.8054(8)
11.5487(4)
90
110.120(2)
90
2051.18(13)
2
2.222
11.4938(7)
11.8348(7)
16.9070(9)
105.050(1)
94.426(1)
118.879(1)
1888.10(19)
2
2.577
14.630
1332
0.20 · 0.04 · 0.04
2.04–26.37
27586
7720 [0.0620]
0.592 and 0.158
10.0835(6)
10.3971(6)
18.6602(11)
78.9320(10)
84.8280(10)
89.6690(10)
1911.98(19)
2
2.634
15.396
1372
0.22 · 0.12 · 0.07
2.03–26.37
24491
7814 [0.0476]
0.412 and 0.133
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
3
˚
Volume (A )
Z
Density (calculated) (Mg/m3)
Absorption coefficient (mmꢁ1
F(000)
)
21.765
2168
0.34 · 0.14 · 0.14
10.284
1276
Crystal size (mm3)
0.36 · 0.36 · 0.34
2.12–30.01
18676
10265 [0.0356]
0.128 and 0.119
Theta range for data collection (ꢁ) 2.08–26.37
Reflections collected
Independent reflections [Rint
Maximum and
24978
5161 [0.0350]
0.151 and 0.051
]
minimum transmission
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2r(I)]
R indices (all data)
5161/0/303
1.207
10265/1/505
1.036
7720/0/456
1.234
7814/0/447
1.044
R1 = 0.0358, wR2 = 0.0723 R1 = 0.0383, wR2 = 0.0867 R1 = 0.0551, wR2 = 0.1021 R1 = 0.0390, wR2 = 0.0841
R1 = 0.0381, wR2 = 0.0733 R1 = 0.0422, wR2 = 0.0889 R1 = 0.0650, wR2 = 0.1062 R1 = 0.0520, wR2 = 0.0971
Absolute structure
–
0.037(9)
–
–
Largest difference in
peak and hole (e A
1.529 and ꢁ0.995
3.466 and ꢁ1.354
1.484 and ꢁ1.769
1.712 and ꢁ1.379
ꢁ3
˚
)

950
P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
Table 4
Crystal and structure refinement data for compounds 2, 8–10
Compound
2
8
9
10
Empirical formula
Formula weight
Crystal system
C25H23IrO9Os3S
1262.29
Monoclinic
P21/n
C33.50H34.50 IrO8Os3Te2
1583.11
Monoclinic
P21/n
C32H31IrO8Os3Te2
1561.57
Trigonal
C32H31IrO8Os3Te2
1561.57
Monoclinic
P21/c
Space group
P31c
Unit cell dimensions
˚
a (A)
9.8007(2)
16.8974(4)
17.8438(4)
90
94.8210(10)
90
2944.59(11)
4
2.847
14.3405(5)
17.5501(7)
17.4210(6)
90
112.1920(10)
90
4059.7(3)
4
2.590
17.5141(6)
17.5141(6)
24.8205(16)
90
90
120
6593.5(5)
6
2.360
13.001
4176
0.24 · 0.08 · 0.06
2.12–26.36
89766
8967 [0.1450]
0.509 and 0.146
8.5464(4)
17.8810(8)
23.6636(10)
90
90.533(1)
90
3616.1(3)
4
2.868
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
3
˚
Volume (A )
Z
Density (calculated) (Mg/m3)
Absorption coefficient
F(000)
17.527
2264
0.34 · 0.26 · 0.22
14.079
2834
15.804
2784
Crystal size (mm3)
0.30 · 0.18 · 0.16
2.32–30.03
41835
11552 [0.0444]
0.212 and 0.101
0.20 · 0.14 · 0.10
2.06–29.61
30870
9344 [0.0606]
0.301 and 0.144
Theta range for data collection (ꢁ) 2.29–30.02
Reflections collected
Independent reflections [Rint
Maximum and
14495
8403 [0.0345]
0.113 and 0.066
]
minimum transmission
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2r(I)]
R indices (all data)
8403/0/357
1.071
11552/0/453
1.084
8967/57/233
1.114
9344/0/422
1.024
R1 = 0.0300, wR2 = 0.0641 R1 = 0.0460, wR2 = 0.0982 R1 = 0.0816, wR2 = 0.1629 R1 = 0.0439, wR2 = 0.0867
R1 = 0.0394, wR2 = 0.0668 R1 = 0.0630, wR2 = 0.1058 R1 = 0.1377, wR2 = 0.1869 R1 = 0.0669, wR2 = 0.0941
Largest difference in
peak and hole (e A
1.612 and ꢁ0.960
2.569 and ꢁ1.066
2.452 and ꢁ1.143
2.186 and ꢁ1.130
ꢁ3
˚
)
MHM). Hi-res MS (m/z): 1561.8530. Calc. for
was modeled with another alternative orientation for an
OsTe2 fragment. Appropriate restraints were placed to
keep the refinement stable.
C32H31O8191Ir128TeTe188Os190OsOs: 1561.8526.
4.7. X-ray crystal structure determinations
Acknowledgments
Crystals were mounted on quartz fibres. X-ray data were
collected on a Bruker AXS APEX system, using Mo Ka
radiation, at 223 K with the SMART suite of programs
[17]. Data were processed and corrected for Lorentz and
polarization effects with SAINT [18], and for absorption
effects with SADABS [19]. Structural solution and refinement
were carried out with the SHELXTL suite of programs [20].
Crystal and refinement data are summarized in Tables 3
and 4.
The structures were solved by direct methods to locate
the heavy atoms, followed by difference maps for the light,
non-hydrogen atoms. With the exception of one hydride in
9 which was located in a low angle difference map, the
hydrides were placed by potential energy calculations with
the program XHYDEX [11], given fixed isotropic thermal
parameters, and refined riding on one of the osmium
atoms they are attached to. All non-hydrogen atoms were
generally given anisotropic displacement parameters in the
final model, except for 9. Organic hydrogen atoms were
placed in calculated positions and refined with a riding
model.
This work was supported by the National University of
Singapore (Research Grant No. R143-000-190-112) and
one of us (P.S.) thanks the University for a Research
Scholarship.
Appendix A. Supplementary data
Experimental details of additional reactions involving 1
and Ph3PSe. Crystallographic data (excluding structure
factors) for the structures in this paper have been deposited
with the Cambridge Crystallographic Data Centre as sup-
plementary publication numbers CCDC 285088–285095.
Copies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge CB2
1EZ, UK, fax: +44 1223 336 033 or e-mail: deposit@ccdc.
cam.ac.uk. Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
j.jorganchem.2005.10.037.
References
Cluster 9 was refined as a racemic twin, and the struc-
ture also exhibited disorder of the heavy atoms, which
[1] (a) J. Mizutani, K. Matsumoto, Chem. Lett. (2000) 72;
(b) B.C. Wiegand, C.M. Friend, Chem. Rev. 92 (1992) 491.

P. Srinivasan et al. / Journal of Organometallic Chemistry 691 (2006) 941–951
951
[2] For examples, see (a) M. Shieh, S.-F. Lin, Y.-W. Guo, M.-H. Hsu,
Y.-W. Lai, Organometallics 23 (2004) 5182;
(b) R. Bau, R.G. Teller, S.W. Kirtley, T.F. Koetzle, Acc. Chem. Res.
12 (1979) 176.
(b) M. Brandl, H. Brunner, H. Cattey, Y. Mugnier, J. Wachter, M.
Zabel, J. Organomet. Chem. 659 (2002) 22;
[9] R.D. Adams, D.A. Katahira, L.W. Yang, Organometallics 1 (1982)
235.
(c) H. Brunner, D. Lucas, T. Monzon, Y. Mugnier, B. Nuber, B.
Stubenhofer, A.C. Stuckl, J. Wachter, R. Wanninger, M. Zabel,
Chem. Eur. J. 6 (2000) 493;
(d) S.W.A. Fong, T.S.A. Hor, J. Chem. Soc., Dalton Trans. (1999) 639;
(e) P. Mathur, R. Trivedi, P. Sekar, D. Chakrabarty, M.M. Hossain,
V.G. Puranik, J. Cluster Sci. 7 (1996) 191;
(f) P. Mathur, P. Sekar, J. Chem. Soc., Chem. Commun. (1996) 727;
(g) R.E. Bachman, K.H. Whitmire, J. van Hal, Organometallics 14
(1995) 1792;
(h)M.Shieh,T.-F.Tang,S.-M.Peng,G.-H.Lee,Inorg.Chem.34(1995)
2797;
[10] (a) D. Cauzzi, C. Graiff, R. Pattacini, G. Predieri, A. Tiripicchio, S.
Kahlal, J.-Y. Saillard, Eur. J. Inorg. Chem (2004) 1063;
(b) C. Graiff, G. Predieri, A. Tiripicchio, Eur. J. Inorg. Chem. (2003)
1659;
(c) D. Belletti, D. Cauzzi, C. Graiff, A. Minarelli, R. Pattacini, G.
Predieri, A. Tiripicchio, J. Chem. Soc., Dalton Trans. (2002)
3160;
(d) C. Graiff, A. Ienco, C. Massera, C. Mealli, G. Predieri, A.
Tiripicchio, F. Ugozzoli, Inorg. Chim. Acta 330 (2002) 95;
(e) P. Braunstein, C. Graiff, C. Massera, G. Predieri, J. Rose, A.
Tiripicchio, Inorg. Chem. 41 (2002) 1372;
(i) I.L. Eremenko, S. Nefedov, H. Berke, B.I. Kolobkov, V.M.
Novotortsev, Organometallics 14 (1995) 1132;
(f) D. Cauzzi, C. Graiff, C. Massera, G. Predieri, A. Tiripichio, J.
Cluster Sci 12 (2001) 259;
(j)V.W.Day,D.A.Lesch,T.B.Rauchfuss,J.Am.Chem.Soc.104(1982)
1290.
(g) W.K. Leong, W.L.J. Leong, J. Zhang, J. Chem. Soc., Dalton
Trans. (2001) 1087;
[3] (a) P. Braunstein, L.A. Oro, P.R. Raithby, Metal Clusters in
Chemistry, Wiley–VCH, Weinheim, New York, 1999;
(b) M.L. Steigerwald, T. Siegrist, E.M. Gyorgy, B. Hessen, Y.U.
Kwon, S.M. Tanzler, Inorg. Chem. 33 (1994) 3389.
[4] P. Srinivasan, W.K. Leong, J. Organomet. Chem. (accepted)..
[5] For example (a) S.M.T. Abedin, K.A. Azam, M.B. Hursthouse, S.E.
Kabir, K.M.A. Malik, M.A. Mottalib, E. Rosenberg, J. Cluster Sci.
12 (2001) 5;
(h) D. Cauzzi, C. Graiff, M. Lanfranchi, G. Predieri, A. Tiripicchio,
J. Chem. Soc., Dalton Trans. (1995) 2321.
[11] A.G. Orpen, XHYDEX, School of Chemistry, University of Bristol, UK,
1997.
[12] M.R. Churchill, B.G. DeBoer, Inorg. Chem. 16 (1977) 878.
[13] B.F.G. Johnson, J. Lewis, P.G. Lodge, P.R. Raithby, Acta Crystal-
logr., Sect. B B37 (1981) 1731.
[14] D.A. Lesch, T.B. Rauchfuss, Organometallics 1 (1982) 499.
[15] (a) For examples, see J. Zhang, W.K. Leong, J. Chem. Soc., Dalton
Trans. (2000) 1249;
(b) H. Egold, D. Schwarze, U. Florke, J. Chem. Soc., Dalton Trans.
(1999) 3203;
(c) P. Braunstein, J.R. Galsworthy, B.J. Hendan, H.C. Marsmann, J.
Organomet. Chem. 551 (1998) 125;
(d) M.C. Jennings, R. Puddephatt, J. Inorg. Chem. 27 (1988) 4280;
(e) E.G. Bryan, B.F.G. Johnson, J. Lewis, J. Chem. Soc., Dalton
Trans. (1977) 1328.
(b) A.A. Pasynskii, Yu.V. Torubaev, K.A. Lysenko, A.Yu. Lyakina,
Zh.V. Dobrokhotova, V.M. Novotortsev, Zh. Neorg. Khim. 43
(1998) 939.
[16] N. Petragnani, Tellurium in Organic Synthesis, Academic, London,
1994.
[6] K. Kiriakidou, M.R. Plutino, F. Prestopino, M. Johansson, L.I.
Elding, E. Nordlander, F. Prestopino, M. Monari, E. Valls, R.
Gobetto, S. Aime, Chem. Commun. (1998) 2721.
[7] S. Jaeaeskelaeinen, J. Pursiainen, T.A. Pakkanen, J. Organomet.
Chem. 487 (1995) 197.
[17] SMART version 5.628, Bruker AXS Inc., Madison, Wisconsin, USA,
2001.
[18] SAINT+ version 6.22a, Bruker AXS Inc., Madison, Wisconsin, USA,
2001.
[19] G.M. Sheldrick, SADABS, 1996.
[8] (a) M.R. Churchill, B.G. DeBoer, F. Rotella, J. Inorg. Chem. 15
(1976) 1843;
[20] SHELXTL version 5.1, Bruker AXS Inc., Madison, Wisconsin, USA,
1997.