Phosphine-substituted and phosphido-bridged metal clusters in homogeneous catalysis I. Synthesis and reactivity of (η5-C5H5)NiM3(μ-H)3(CO)(9-n)Ln (M = Ru, Os, n = 1,2: L = PPh3, PPh2H, PCy3, PEt3) and M3(CO)(12-n)Ln (M = Ru, N = 1-3: L = PPh3,

Article abstract of DOI:10.1016/0022-328X(88)80047-0

The complexes (η⁵-C5H5)NiM3(μ-H)3(CO)(9-n)Ln (M = Ru, Os, n = 1.2: L = PPh3, PPh2H, PCy3, PEt3) and M3(CO)(12-n)Ln (M = Ru, n = 1-3: L = PPh3, PPh2H, PCy3, PEt3; M = Os, n = 1,2: L = PPh3) have been synthesized by new or known procedures.Reacti

Full text of DOI:10.1016/0022-328X(88)80047-0

100
filtration and TLC yielded about 75% of 7e and 10-15% of a yellow, unidentified
derivative.
Synthesis of the (Cp)NiM3H3(CO)9 _nLn derivatives
The new complexes 2m, 2n, 2nn were obtained by the method used previously for
the analogous osmium derivatives [11-14]. The new substituted derivatives of I were
obtained by two procedures, as follows.
(a) Reactions of I with PPh3, PPh2H, PCy3, PEt3 and Me3NO. Treatment of 1
with a two-fold molar excess of PPh3 in the presence of MeaNO in refluxing
heptane for 2 rain resulted in extensive decomposition. The main soluble product
was (Cp)NiRu3H3(CO)7(PPh3) 2 (laa) (20%), together with traces of Ru3(CO)I0-
(PPh3) 2. In 3 rain in refluxing hexane about 10% of (Cp)NiRuaHa(CO)s(PPh3) (la)
and 70% of laa are formed.
The reaction of PPh2H with 1 and Me3NO in hexane, with gentle warming for 5
rain below the reflux temperature, yielded a red green solution, containing some
unchanged 1, ca. 10% of Ru3(CO)10(PPh2H)2, and 30% of Ru3(CO)9(PPhEH)3.
Treatment of 1 (200 mg, 0.283 mmol) with 0.5 ml of liquid PEt3 and Me3NO (50
mg) in refluxing for 30 s yielded a green-red solution containing some Ru3(CO)a-
(PEt 3 )3, and a green product tentatively identified as (Cp)NiRu3n 3(CO)6(PEt 3)3.
Treatment of 1 (250 mg 0.366 mmol) with 0.5 g (1.78 mmol) of PCy3 and Me3NO
(50 mg) in refluxing hexane for 4 rain yielded about 40% of (Cp)NiRu3H3(CO)s-
(PCy3) (2m) and 40% of (tentatively identified) (Cp)NiRu3Ha(CO)7(PCy3)2 (2ram).
(b) Reactions of Ru3(CO)12_n(PPh3)n (n = 1-3) with (Cp)2Ni2(CO)2. (n = 1)
The conditions used were similar to those for the synthesis of 1 [4,17]. Treatment of
Ru3(CO)ll(PPh3) (100 mg, 0.114 mmol) with (Cp)ENi2(CO)2 (100 mg, 0.329
mmol) in refluxing diglyme for 20 min under a current of HE gave a dark brown
solution. This was reduced to small volume and subjected to TLC to give traces of
the unchanged nickel substrate, about 20% of la, 20% of laa, and several unidenti-
fied trace products.
(n = 2) Treatment of Ru3(CO)10(PPh3) 2 (100 mg, 0.090 mmol) with (Cp)2Ni2(CO)2
(100 mg, 0.329 mmol) in diglyme at 100°C for 55 min under a current of n 2
yielded only 15% of laa, a dark yellow unidentified derivative, and the products of
extensive decomposition.
(n = 3) Treatment of Ru3(CO)9(PPh3)3 (100 mg, 0.075 mmol) with (Cp)ENi2(CO)2
(100 mg, 0.329 mmol) in diglyme at 110 °C for 35 min under a current of HE gave
only traces of la, about 30% of laa, and trace amounts of a dark-green unstable
product, tentatively identified as (Cp)NiRu3H3(CO)6(PPh3) 3. Extensive decom-
position also occurred.
(c) Reactions of Ru3(CO)I2_n(PPh2H)n (n~-2,3) with (Cp2Ni2(CO)2. The
ruthenium derivatives are sparingiy soluble in aliphatic hydrocarbon solutions of the
nickel dimer. Suspensions were formed, which gave clear solutions at reflux temper-
atures.
Addition of
a 10 ml toluene solution of 50 mg (0.052 mmol) of
Ru3(CO)lo(PPh2H)2 to a hexane solution of 100 mg (0.329 mmol) of the nickel
complex under N2 followed by 30 min reflux gave a dark purple suspension. The
reaction yielded seven species identified as: small amounts of unchanged Ru3(CO)12
and nickel dimer along with H2Ru3(CO)s(/L-PPh2)2 (complex A) [15],
H2Ru3(CO)7(PPh2H)(/L-PPh2) 2 (complex B) [15], Ru2(CO)6(/~-PPh2)2 (complex

101
C) [15] and HRu3(CO)7(/.t-PPh2)3 (complex D) [15]; and with traces of an unidenti-
fied derivative. Yields of A and B were 5-10%, and those of C and D 10-20%. The
same result was obtained after only 2 min in refluxing heptane. Complexes A-D
were difficult to separate and isolate pure by TLC.
Addition of
a
10 ml toluene solution of 75 mg (0.067 mmol) of
Ru 3(CO)9(PPh2H)3 to a heptane solution of 100 mg of nickel dimer followed by 10
min reflux under N2 gave a dark purple solution containing the same products as
above.
Other reactions. Reactions of Ru3(CO)I2_nLn (L=PPh2H, n=1-3," L=PPh3,
n = 2) with 3,3-dimethyl-but-l-yne
A 10 ml toluene solution of Ru3(CO)ll(PPh2H) with a 100 ml heptane solution
containing 0.5 ml of the alkyne was refluxed under N2 for 5 min to yield a dark
yellow solution containing nine products, namely: HRu3(CO)9(C2Bu t) (5%) [19];
HRua(CO)a(PPh2H)(C2Bu t) (1%); HRu3(CO)7(PPh2H)2(C2But) (1%) *;
A
(5-10%); B (5-10%); C (10-20%); D (10-20%); an unidentified deep red complex
(E); and an unidentified green complex (F) * *
The reaction of Ru3(CO)10(PPh2H)2 under similar conditions with 6 rain reflux
gave the same products in comparable yields.
Ru3(CO)9(PPhEH)3 after 15 min in refluxing heptane yielded a brown solution
containing Rua(CO)lo(PPh2H)2 and products A-D. However, after 10-15 min
reflux in octane the product distribution above was observed.
In refluxing heptane, Rua(CO)10(PPh3)2 gave three products identified as
HRua(CO)9_n(L)n(C2Bu t) (n = 2,3) [20] and Ru3(CO)9(PPh3) 3.
Thermal modification a n d / o r disproportionation reactions of the phosphine-substituted
derivatives
Refluxing of octane suspensions of Rua(CO)9(PPh2H)3 (initially dissolved in a
few ml of toluene) for 40 min under N2 in the absence of other ligands yielded
solutions containing complexes A and B in ca. 5-10% and C and D in ca. 10-20%
yields.
Complexes of (Cp)NiOsEH3(CO)sL (L = PPh 3, PPhEH) slowly dispro-
portionated when their solutions were kept for 12-24 h at room temperature under
N 2. The reaction occurred in a few minutes in refluxing hexane, and 5% of
(Cp)NiOsaH3(CO)9 and of (Cp)NiOs3H3(CO)7L 2 were formed.
Refluxing of a solution (Cp)NiRu3H3(CO)8(PPh3) in hexane for 30 s yielded
15% of (Cp)NiRu 3H3 (CO)9 and 40% of (Cp)NiRu 3H 3(CO)7 (PPh 3)2.
Thermal disproportionation reactions are probably likely to occur during the
syntheses described above.
Ligand exchange reactions on (Cp)NiOs3H3(CO)sL complexes
All the reactions were performed under N2 in refluxing hexane in the presence of
* HRu3(CO)s(PPh2H)(C2Bu t) IR: 2077s, 2055vs, 2049sh, 2044sh, 2025s, 2016vs, 1990m, 1970w. 1H
NMR: 7.60-7.18 mm (Ph): 1.29
t (But); -21.14 s (hydride). 31p NMR: +11.16 s.
HRu3(CO)7(PPh2H)2(C2But)- IR: 2062s, 2024vs, 2014vs, 1996vs, 1982s, 1960m. 1H NMR: 7.53-7.38
m (Ph): 1.99 s(But): --20.50 d (hydride). 31p NMR: +1.14 s. Satisfactory C and H analyses were
obtained for both complexes.
** See Discussion.

102
Table 1
Ligand exchange reactions of (Cp)NiOs3 H 3(CO) 8L complexes
(Cp)NiOs3H 3(CO)8L + L' ~ (Cp)NiOs3H 3(CO)sL' + L
L
L'
Reaction time
(rain)
Yield of L'
(%)
CH3 CN
CH3 CN
CH 3CN
C H 2 C N
SbPh 3
CO
3
3
3
2
5
5
5
6
2
100
100
90 a
PPh 3
C2 (PPh
SbPh 3
PPh 3
C2 (PPh
PPh 3
PPh 3
PPh 3
2
2
)2
)2
80 a
99 ~
SbPh 3
no exchange (30% decomposition)
no exchange (3% decomposition)
90
PPh2C2Pr i
PPh 2 H
PEt 3
no exchange
a
Some decomposition and some (Cp)NiOs3H3(CO)9 were observed.
a slight excess of the free ligand (1.2 to 1 molar ratio), except when CO was used (as
a slow current). Details of these reactions are given in Table 1.
Results and discussion
Synthesis and characterization of the complexes
(a) Trimetallic derivatives. Phosphine-substituted carbonyl dusters are nowa-
days readily made by various methods: (a) thermal or photochemical substitution
of phosphines for CO: (ii) displacement of labile ligands (e.g. acetonitrile): and (iii)
labilization of coordinated CO's by various reactants.
These methods were employed for the synthesis of complexes M3(CO)12_nLn
(M = Ru [21]; M = Os [22]). Thermal substitution reactions require high tempera-
tures and generally give products only with n = 3 [23]. Greater selectivity is achieved
by displacing CH3CN [24]. A selective high yield synthesis of Ru derivatives with
n = 1,2,3 (depending on the phosphine/cluster ratio) is obtained with Na-benzo-
phenylketyl as a radical-ion initiator [25].
We have found that use of Me3NO as a labilizing agent leads to fair yields of the
products with n = 1-3, except when the weakly reactive PCy3 or the very reactive
PEt 3 are involved. The reaction stoichiometry and time also influence the yields;
disproportionations are likely to occur, since the n = 2 products are generally
favoured for most of the phosphines.
We have also confirmed that phosphine exchange between metals can be used in
synthesis; Ru3(CO)9(PPh3) 3 was previously made from Ru3(CO)I2 and
Pt(stilbene)(PPh3) 2 [26]. Use of commercially available (PPh3)2Ni(CO)2 also gives
the M3(CO)12_n(PPh3)3 (M = Ru, Os) derivatives. Phosphine exchange between
metals under relatively mild conditions is evidently significant in view of the
disproportionation reactions observed.
Several Ma(CO)12_~Ln derivatives discussed here had previously only been
partially characterized (see ref. 21-26) and, in particular 31p NMR data were
lacking. Full data for the derivatives are listed in Table 2.
Except for Ru 3(CO)9(PEt3)3, each complex apparently gave only one isomer of
the many axial-equatorial species possible; this could be the result of the failure of

103
the TLC technique to effect their separation or to rapid equilibration in solution.
Bruce and coworkers were also able to identify only one isomer for each complex
[25a].
By contrast Os3(CO)12_nLn (L = PPhMe2) gives separable isomers [27]; for
n = 2, 1,1- and 1,2-species were separated on silica, and for n = 3, two Cs, non-inter-
convertible, 1,2,3- and 1,1,2-isomers were observed. It is noteworthy that intramo-
lecular phosphine transfer was not observed.
(b) Heterometallic derivatives. These syntheses were performed as previously
described [11-14], starting from clusters 1 or 2, and the appropriate ligand, in the
presence of Me3NO. Details of the new osmium and ruthenium derivatives are given
in Table 3.
The type and yields of the products depend on the nature (and hence on the
reactivity) of the phosphine, as well as on the cluster metals; thus (Cp)NiOs3 H3(CO)9
usually gives n = 1 derivatives [11], and relatively long reaction times are required to
obtain n = 2 complexes [13]. The behaviour of PEt 3 is unusual, as the n = 2
derivative is obtained in good yield in a very short time. Derivatives with n = 3
could not be obtained even after long reaction time or by use of a large excess of
ligand, and only decomposition products were formed.
(Cp)NiRu3H3(CO)9 is considerably more reactive and generally gives n = 2
products; with PPh2H and PEt3 in particular, fragmentation of the cluster occurs
under very mild conditions. The latter ligand, however, gives small amounts of an
unstable green product, which could not be crystallized, but which was tentatively
identified as (Cp)3NiRu3H3(CO)6(PEt3)3.
Attempts to increase the selectivity and obtain n = 3 derivatives were made by
using the Ru3(CO)12_n(PPh3) n (n = 1-3) complexes, which were treated with
(Cp)2Ni2(CO)2 under H2 under conditions similar to those adopted for the
synthesis of cluster 1 [4,17]; these conditions are, however, rather drastic, so that
some disproportionation occurs. Once again the n = 2 product is favoured, and only
traces of the (assumed) n = 3 derivatives are obtained. This behaviour confirms that
the trisubstituted derivatives are unstable, probably for steric as well as electronic
reasons.
Reactions of Ru3(CO)12_n(PPh2H)~ (n = 1-3) with (Cp)2Ni2(CO)2 were also
examined, under conditions more closely similar to those used in the hydrogenation
experiments described elsewhere [16]. Under these conditions no formation of
substituted derivatives of 1 was observed; furthermore we did not observe reductive
coupling a n d / o r elimination of the phosphine hydrogen atoms with nickel contain-
ing fragments. Under a N2 atmosphere, only gradual loss of phosphine hydrogen
atoms occurs, leading to complexes A-D, observed previously by Geoffroy and
coworkers [15] as stable products of the photochemical reaction of Ru3(CO)9-
(PPh2H)3; the same products are obtained, under N2 and in comparable conditions,
by simple thermal treatment in the absence of nickel dimer.
For the (Cp)NiRu3H3(CO)9_~L~ derivatives, reported here for the first time, we
suggest the same structure and isomerism as found for the homologous osmium
compounds. Indeed, the IR, 1H and 31p NMR of the two classes of complexes are
closely comparable (see Table 3).
Other phosphine-substituted tetrahedral heterometallic clusters have been re-
ported; HRuCo3(CO)11(PPh3), obtained by "metal selective" substitution of PPh 3
on one basal cobalt atom, has the ligand in an axial position, at least in the solid
(Continued on p. 106)

104
%
+
r~
¢.q
I
I
+
+
•+
I
+
©
8
¢..)
°
.~
~d
e~
r..)
r..)
~3

1 0 5
r~
-t-
I
.+
+
+
-t-
~ , , , q
~ . , a -=
t"q
e~
C..)
I
, , , 4 ~
,..,
, , , 4 ~
©
z
,n-~"
, n : , 4
"L
,.,~ r4
v
8
v
.~a
e~
e~
ol
en
r,.)
,..g
o
o
o
o
e~
en
e~
e~
z
z
r..)
z

106
state [28]. The first substitution of PPh3 on H2FeRu3(CO)I3 occurs in an axial
position on one of the basal ruthenium atoms, whereas the second substitution, on
another ruthenium, can be either axial or equatorial [29]. 1H and 31p NMR data are
available for the latter compounds; the monosubstituted species shows a hydridic
signal at - 1 7 . 9 8 d ( J ( P - H ) 8.6 Hz) and a 31p NMR signal at +43.23 s; the
disubstituted complex shows complex hydridic signals at - 17.12 m, - 18.05 m, and
31p NMR chemical shifts at +41.9 s, +36.7 s( all of the spectra being recorded in
CDCI 3 at - 3 0 o C). These values fall in the range found at room temperature for
the (Cp)NiRu3H3(CO)9_~L ~ derivatives; the main difference is that the n = 2
complexes show only one 31p signal owing to equivalence of the phosphines.
Thus, axial substitution is always observed for the first phosphine ligand attach-
ing to a tetrahedral cluster formed by a triangle of identical metals capped by a
heterometal. By contrast, when M2M'2 tetrahedral clusters are substituted, as for
H2Ru2Rh2(CO)11(PPh3) [30], the position of the phosphine becomes a matter of
choice of the basal triangle.
Thermal reactions
(a) Disproportionation reactions. These have been observed for nearly all the
derivatives; Ru3(CO)12_nLn undergo this process quite easily, and apparently the
statistically expected n = 2 complexes are the favoured products. Ready thermal
disproportionation of R u 3 ( C O ) I2 or Ru3(CO)ll(P(OMe)3) with R u 3 ( C O ) 9 ( P ( O -
Me)3)3 has been reported [25]; it is probable that these reactions occur via intra- or
inter-molecular phosphine exchange rather than by break-up into metal fragments
followed by recondensation; the phosphine exchange observed between nickel and
ruthenium favours this suggestion. However, kinetic studies on Ru3(CO)9(PPh3)3
and related compounds [31] have shown that fragmentation can also occur, and that
there are complex equilibria between the tri-, bi- and mono-metallic species *
Disproportionation reactions also occur during the synthesis of the (Cp)Ni-
RuaH3(CO)9_nL n derivatives; it is impossible to discriminate between dispro-
portionation of the starting materials during the reaction and rearrangement of the
products. The latter have shown a tendency to disproportionate, the process being
faster for the ruthenium derivatives.
In addition to disproportionation, the Rua(CO)12_,(PPh2H)n (n = 1-3) com-
plexes undergo thermal modification reactions; in the absence of ligands these
occurs slowly, but the process is faster in the presence of (Cp)2Ni2(CO)2 or of
t-butyl-acetylene (see section b, below). Clusters A-D are formed, the same prod-
ucts that are obtained upon photochemical treatment of Ru3(CO)9(PPh2H)3 [15].
The formation of phosphido-bridged derivatives, from PPh2H substituted clusters
involves oxidative addition of the "fragments" PPh2 and H together with some loss
of hydrogen; the higher reaction rates observed in the presence of ligands show that
these must assist the process, probably by accepting hydrogen. This is relevant to
the catalytic behaviour of the PPh2H substituted complexes.
* We examined the reactions of Os3(CO)12 and Ru3(CO)I2 or of Ru3(CO)I2 and Fe3(CO)I2 with
(Cp)ENi2(CO)2 under HE in diglyme. Preliminary results indicate that the clusters (Cp)NiM2M'
(CO)9H3 are formed, and these are under investigation.

107
(b) Reactions with 3,3-dimethyl-but-l-yne. This ligand has been chosen because
of its well known coordinating ability [32]. Furthermore, phosphine-substituted,
acetylide-bridged triruthenium clusters have already been reported [20], making
easier the identification of new complexes of comparable structure.
The reactions of this alkyne with (Cp)NiM3H3(CO)9 clusters have been discussed
previously [4,5]. Displacement of the capping (Cp)Ni generally occurs. Cluster
opening to give butterfly structures is dis-favoured.
Clusters Ru3(CO)12_nLn (L = PPh3, P P h 2 H) react with the alkyne in a complex
way; when n = 2, L = PPh 3, HRu3(CO)7(PPh3)2(C2Bu t) [20] and disproportiona-
tion products containing three phosphines, with or without the acetylide, are found.
This shows that the alkyne or the derived acetylide displace the coordinated CO
ligands rather than PPh 3.
When L = P P h 2 H and n = 1-3, some selectivity is observed; indeed, the same
products are obtained in increasingly drastic conditions (for n = 1,2,3 solvents were
refluxing heptane (short time), heptane (longer reaction time) and octane, respec-
tively). The nature of the products deserves some comments. Small amounts of
HRu3(CO)9(C2But) [19] are found, and this indicates that the alkyne can displace
the phosphine ligands, although with considerable difficulty; HRu3(CO)9_n(PPh z-
H)n(C2But) (n = 1,2) derivatives are also found, and were identified by comparison
of the spectroscopic data with those for known derivatives [20]. The main products
of these reactions are, however, the phosphido-bridged complexes A-D, the thermal
decomposition products. Finally, two unidentified products are also formed, but
only under rather drastic conditions; they may possibly be derived from the reaction
of complexes C or D with the alkyne *
(c) Ligand exchange on (Cp)NiOs3H3(CO)sL. The results of these reactions are
in Table 1; apparently the sequence in which ligands displace each other depends on
their ~r-donor abilities; the following series of ease of displacement have been
observed [33]:
(a) C H 3 C N >> SbPh 3 = dppm - C2(PPh2)2 > PPh3 > CO
(b) P P h 2 H = PCy3 = PEt 3 > PPh3 > CO
The positions of CO in the series may be deceptive, since the reactions were
performed under a slow current of the gas, and hence with a considerable excess of
it. Moreover the reactions described earlier indicate that CO is displaced more easily
than the phosphine ligands by organic ligands.
Concluding r e m a r k s
An investigation of the reactions of (Cp)NiM3H3(CO)9_nL n (M = Ru, Os)
derivatives has shown that warming in the presence of ligands or of "metal
* Reactions of complexes C or D with t-butyl-alkyne, by refluxing in octane for about 20 rain were
examined. Complex C gave a turquoise-green derivative (complex F) showing only terminal carbonyls
and signals in the 31p NMR typical of phosphido bridges; the ~H NMR data indicate the presence of
an organic ligand. Complex D gives a ruby-red derivative (complex E) with only terminal carbonyls
and giving a 31p NMR spectrum typical of phosphido-bridged derivatives; the 1H NMR indicates the
presence of an organic ligand. Hydridic signals were found for both compounds.
The green and red complexes are identical to those obtained by treating Ru3(CO)12_nLn with
t-butyl-alkyne. Unfortunately, attempts to obtain X-ray grade crystals gave only crystalline powders.
The nature of the complexes is still under investigation.

108
fragments" similar to those used in catalytic experiments gives rise to several
processes w h i c h may be parallel or competitive.
The alkyne displaces (Cp)Ni and CO from the tetrahedral clusters and m a i n l y
CO from the trimetallic derivatives; displacement of phosphines from the latter is
h a r d e r than that of CO. This behaviour may have implications for the catalytic
behaviour of the clusters. With the tetrahedral complexes exchange of the p h o s p h i n e
occurs readily, the ease being determined by the ~r-donor abilities of the ligands.
However, all the clusters examined u n d e r g o disproportionation, with preferential
formation of the n = 2 derivatives. W h e t h e r this behaviour involves the formation
and recombination of m e t a l fragments is still uncertain.
The behaviour of the P P h 2 H derivatives is noteworthy. T h e r m a l treatment of the
Ru3(CO)12_~Ln(PPh2H)~ complexes with or without o t h e r ligands gives p h o s -
phido-bridged derivatives. Thus, phosphine ligands tend to loose hydrogen and to
give the more stable phosphido-bridging units; the phosphido-bridged derivatives
still react with t-butyl-acetylene, p r o b a b l y by CO displacement a n d / o r insertion
into M - P or M - H bonds.
This behaviour must be t a k e n into account when considering the homogeneous
catalytic properties of the complexes [16].
Acknowledgement
Financial support for this work was provided by the Ministero della Pubblica
Istruzione and by C. N. R. (Rome). J o h n s o n M a t t h e y is t h a n k e d for a loan of RuC13
(to E.S.).
References
1 M. Castiglioni, E. Sappa, M. Valle, M. Lanfranchi and A. Tiripicchio, J. Organomet. Chem., 241
(1983) 99, and references therein.
2 S.G. Shore, W.S. Hsu, C.R. Weisenberger, M.L. Castle, M.R. Churchill and C. Bueno, Inorg. Chem.,
22 (1983) 510.
3 G. Lavigne, F. Papageorgiou, C. Bergounhou and J.J. Bonnet, Inorg. Chem., 22 (1983) 2485.
4 M. Castiglioni, R. Giordano and E. Sappa, J. Organomet. Chem., 319 (1987) 167.
5 M. Castiglioni, R. Giordano, E. Sappa, A. Tiripicchio and M. Tiripicchio CameUini, J. Chem. Soc.,
Dalton Trans., (1986) 23.
6 P. Moggi, G. Albanesi, G. Predieri and E. Sappa, J. Organomet. Chem., 252 (1983) C89.
7 G. Albanesi, R. Bernardi, P. Moggi, G. Predieri and E. Sappa, Gazz. Chim. Ital., 116 (1986) 385.
8 M. Castiglioni, R. Giordano, E. Sappa, G. Predieri and A. Tiripicchio, J. Organomet. Chem., 270
(1984) C7.
9 M. Castiglioni, R. Giordano, E. Sappa and P. Volpe, J. Chromatogr., 349 (1985) 173.
10 M. Castiglioni, R. Giordano and E. Sappa, J. Mol. Catal., 39 (1986) 287; idem, ibid., 40 (1987) 65.
11 E. Sappa, M. Valle, G. Predieri and A. Tiripicchio, Inorg. Chim. Acta, 88 (1984) L23.
12 G. Predieri, A. Tiripicchio, C. Vignali, E. Sappa and P. Braunstein, J. Chem. Soc., Dalton Trans.,
(1986) 1135.
13 E. Sappa, M.L. Nanni Marchino, G. Predieri, A. Tiripicchio and M. Tiripiochio Camellini, J.
Organomet. Chem., 307 (1986) 97.
14 E. Sappa, G. Predieri, A. Tiripicchio and M. Tiripicchio Camellini, J. Organomet. Chem., 297 (1985)
103.
15 R.P. Rosen, G.L. Geoffroy, C. Bueno, M.R. Churchill and R.B. Ortega, J. Organomet. Chem., 254
(1983) 89.
16 M. Castiglioni, R. Giordano and E. Sappa, J. Organomet. Chem., the next, 342 (1988) 111.
17 G. Lavigne, F. Papageorgiou, C. Bergounhou and J.J. Bonnet, Inorg. Synth., XXV, in press.

109
18 E. Sappa, O. Gambino, L. Milone and G. Cetini, J. Organomet. Chem., 39 (1972) 169.
19 M. Catti, G. Gervasio and S.A. Mason, J. Chem. Soc., Dalton Trans., (1977) 2260.
20 C. Jangala, E. Rosenberg, D. Skinner, S. Aline, L. Milone and E. Sappa, Inorg. Chem., 19 (1980)
1571.
21 (a) M.A. Bennett, M.I. Bruce and T.W. Matheson, in E.W. Abel, F.G.A. Stone and G. Wilkinson
(Eds.), Comprehensive Organometallic Chemistry, Pergamon Press, Oxford, 1982; Ch. 32.5, p. 872 et
seq; (b) S.C. Tripathi, S.C. Srivastava, R.P. Mani and A.K. Shrimal, Inorg. Chim. Acta, 15 (1975)
249.
22 R.D. Adams and J.P. Selegue, in E.W. Abel, F.G.A. Stone and G. Wilkinson (Eds.), Comprehensive
Organometallic Chemistry, Pergamon Press, Oxford, 1982; Ch. 33, p. 977 et seq.
23 (a) C.W. Bradford, Platinum Met. Rev., 11 (1967) 104; (b) F. Piacenti, M. Bianchi, E. Benedetti and
G. Sbrana, J. Inorg. Nucl. Chem., 29 (1967) 1389; (c) B.F.G. Johnson, R.D. Johnston, P.L. Josty, J.
Lewis and I.G. Williams, Nature, 213 (1967) 901; (d) F. Piacenti, M. Bianchi, E. Benedetti and C.
Braca, Inorg. Chem., 7 (1968) 1815; (e) A.J. Deeming, B.F.G. Johnson and J. Lewis, J. Organomet.
Chem., 17 (1969) 40; idem, J. Chem. Soc. (A), (1970) 897; (f) J.P. Candlin and A.C. Shortland, J.
Organomet. Chem., 16 (1969) 289; (g) R.J.H. Clark, C.W. Bradford, R.S. Nyhohn and W. van
Bronswijk, J. Chem. Soc. (A), (1970) 2889; (h) C.W. Bradford, R.S. Nyholm, G.J. Gainsford, J.M.
Guss, P.R. Ireland and R. Mason, J. Chem. Soc., Chem. Commun., (1972) 87; (i) G.J. Gamsford,
J.M. Guss, P.R. Ireland, R. Mason, C.W. Bradford and R.S. Nyholm, J. Organomet. Chem., 40 (1972)
C70; (1) C.W. Bradford and R.S. Nyholm, J. Chem. Soc., Dalton Trans., (1973) 529; (m) A.J.
Deeming, R.E. Kimber and M. Underhill, J. Chem. Soc., Dalton Trans., (1973) 2589; (n) M.I. Bruce,
J.G. Matisons, R.C. Wallis, J.M. Patrick, B.W. Skelton and A.H. White, J. Chem. Soc., Dalton Trans.,
(1983) 2365; (o) B.F.G. Johnson, J. Lewis, B.E. Reichert and K.T. Schorpp, J. Chem. Soc., Dalton
Trans., (1976) 1403.
24 G.A. Foulds, B.F.G. Johnson and J. Lewis, J. Organomet. Chem., 296 (1985) 147.
25 (a) M.I. Bruce, Coord. Chem. Rev., 76 (1987) 1, and references therein; (b) M.I. Bruce, J.G.
Matisons and B.K. Nicholson, J. Organomet. Chem., 247 (1983) 321.
26 M.I. Bruce, G. Shaw and F.G.A. Stone, J. Chem. Soc., Dalton Trans., (1972) 2094.
27 A.J. Deeming, S. Donovan-Mtunzi, S.E. Kabir and P.J. Manning, J. Chem. Soc., Dalton Trans.,
(1985) 1037.
28 M. Hidai, H. Matsuzaka, Y. Koyasu and Y. Uchida, J. Chem. Soc., Chem. Commun., (1986) 1451.
29 J.R. Fox, W.L. Gladfelter, T.G. Wood, J.A. Smegal, T.K. Foreman, G.L. Geoffroy, I. Tavanaiepour,
V.M. Day and C.S. Day, Inorg. Chem., 20 (1981) 3214.
30 J. Pursiainen and T.A. Pakkanen, J. Organomet. Chem., 315 (1986) 353.
31 A. Poe, Chemistry in Britain (1983) 997, and references therein.
32 E. Sappa, A. Tiripicchio and P. Braunstein, Chem. Rev., 83 (1983) 203.
33 Preliminary results have been reported: E. Sappa, G. Predieri and A. Tiripicchio, XII International
Conference on Organometallic Chemistry, Vienna Sept. 8-13 (1985) paper 98.