Activation volumes of substitution reactions on neutral and cationic organometallic platinum(IV) complexes: Definite proof of selective associative activation

Article abstract of DOI:10.1021/om0499409

The activation volumes for the substitution reactions of a series of Pt(IV) sulfide organometallic complexes have been determined as a function of the entering and leaving ligands, as well as the inert organometallic skeleton. The results agree with the existence of a tunable associatively/dissociatively activated mechanism operating on these complexes, depending on the electronegativity of the substitution-inert spectator ligands.

Full text of DOI:10.1021/om0499409

2434
Organometallics 2004, 23, 2434-2438
Activa tion Volu m es of Su bstitu tion Rea ction s on Neu tr a l
a n d Ca tion ic Or ga n om eta llic P la tin u m (IV) Com p lexes:
Defin ite P r oof of Selective Associa tive Activa tion
Carlos Gallego,† Gabriel Gonza´lez,† Manuel Mart´ınez,*,† and Andre´ E. Merbach‡
Departament de Qu´ımica Inorga`nica, Universitat de Barcelona, Mart´ı i Franque`s 1-11,
E-08028 Barcelona, Spain, and Institut de Chimie Mole´culaire et Biologique,
EÄ cole Polytechnique Fe´de´rale de Lausanne, EPFL-BCH, CH-1015 Lausanne, Switzerland
Received J anuary 22, 2004
The activation volumes for the substitution reactions of a series of Pt(IV) sulfide
organometallic complexes have been determined as a function of the entering and leaving
ligands, as well as the inert organometallic skeleton. The results agree with the existence
of a tunable associatively/dissociatively activated mechanism operating on these complexes,
depending on the electronegativity of the substitution-inert spectator ligands.
In tr od u ction
yl complexes has also been carried out for some known
cyclometalated complexes.15,16 All of these reactions
have been found to operate via a dissociation of one of
the ligands around the platinum center as an initial step
prior to the oxidative addition of the C-X bond.17-20 The
existence of these three-coordinate intermediates had
already been proposed in some instances where the
number of Pt-C bonds existing in the complex seemed
to play a crucial role.11,21-24
We have also studied the substitution reactions on
similar cyclometalated octahedral Pt(IV) complexes.25-28
The effect of two Pt-CH3 and a Pt-Caryl bond seems
clearly to produce complexes where Pt(IV) resembles Pt-
(II), and in most of the cases the reaction mechanism is
dissociatively activated, as expected,24 including the
formation of a pentacoordinated unsaturated inter-
mediate.29-31 Nevertheless, when the Pt(IV) center is
coordinated to highly electron withdrawing groups or
atoms, such as F and CC5F4CHdNCH2Ph, and with
Substitution on inert Pt(IV) octahedral t2g6 complexes
has been a difficult subject to deal with, given the
important amount of side reactions which appear in
most of the processes not involving organometallic
compounds.1-5 Furthermore, the importance of the
oxidative-addition-reductive-elimination reactions in
most of the catalytic cycles in which platinum com-
pounds are involved has hindered the effort dedicated
to the subject.6-10 Nevertheless, it is clear that any of
these reactions is usually preceded by a substitution
process involving complexes with Pt-C σ-bonds. These
bonds are expected to produce important inductive
effects on the platinum center, which results in a certain
tendency for dissociatively activated reactions.11
We have been involved in the study of the mecha-
nisms by which electrophilic C-H bond activation takes
place on Pd(II) and Rh(II) complexes.12-14 Furthermore,
the study of the mechanism for the oxidative-addition
reactions of these bonds on square-planar Pt(II) dimeth-
(15) Crespo, M.; Mart´ınez, M.; Sales, J . J . Chem. Soc., Chem.
Commun. 1992, 822.
(16) Crespo, M.; Mart´ınez, M.; Sales, J . Organometallics 1993, 12,
4297.
(17) Crespo, M.; Mart´ınez, M.; de Pablo, E. J . Chem. Soc., Dalton
Trans. 1997, 1321.
† Universitat de Barcelona.
‡ EÄ cole Polytechnique Fe´de´rale de Lausanne.
(1) Tobe, M. L.; Burgess, J . Inorganic Reaction Mechanisms; Long-
man: Essex, U.K., 1999.
(2) Dixon, N. E.; Lawrance, G. A.; Lay, P. A.; Sargeson, A. M. Inorg.
Chem. 1983, 22, 846.
(3) Drougge, L.; Elding, L. I. Inorg. Chim. Acta 1986, 121, 175.
(4) Dixon, N. E.; Lawrance, G. A.; Lay, P. A.; Sargeson, A. M. Inorg.
Chem. 1984, 23, 2940.
(18) Anderson, C. M.; Crespo, M.; J ennings, M. C.; Lough, A. L.;
Ferguson, G.; Puddephatt, R. J . Organometallics 1991, 10, 2672.
(19) Crespo, M.; Puddephatt, R. J . Organometallics 1987, 6, 2548.
(20) Crespo, M.; Mart´ınez, M.; Sales, J .; Solans, X.; Font-Bard´ıa,
M. Organometallics 1992, 11, 1288.
(5) Peloso, A. Coord. Chem. Rev. 1973, 10, 123.
(6) Parshall, G. W.; Ittel, S. D. Homogeneus Catalysis, 2nd ed.;
Wiley: New York, 1992.
(7) Holt, M. S.; Wilson, W. L.; Nelson, J . H. Chem. Rev. 1989, 89,
11.
(8) van der Veen, L. A.; Keeven, P. K.; Kamer, P. C. J .; van
Leeuwen, P. W. N. M. J . Chem. Soc., Dalton Trans. 2000, 2105.
(9) Deibele, C.; Permin, A. B.; Petrosyan, V. S.; Bargon, J . Eur. J .
Inorg. Chem. 1998, 1915.
(21) Schmu¨lling, M.; van Eldik, R. Chem. Ber./ Recl. 1997, 130, 1791.
(22) Nakayama, K.; Kondo, Y.; Ishihara, K. Can. J . Chem. 1998,
76, 62.
(23) Scott, J . D.; Puddephatt, R. J . Organometallics 1983, 2, 1643.
(24) Romeo, R. Comments Inorg. Chem. 1990, 11, 21.
(25) Bernhardt, P. V.; Gallego, C.; Mart´ınez, M.; Parella, T. Inorg.
Chem. 2002, 41, 1747.
(26) Bernhardt, P. V.; Gallego, C.; Mart´ınez, M. Organometallics
2000, 19, 4862.
(10) Gladiali, S.; Alberico, E.; Pulacchini, S.; Kolla`r, L. J . Mol. Catal.
1999, 143, 155.
(11) Frey, U.; Helm, L.; Merbach, A. E.; Romeo, R. J . Am. Chem.
Soc. 1989, 111, 8161.
(12) Estevan, F.; Gonza´lez, G.; Lahuerta, P.; Mart´ınez, M.; Peris,
E.; van Eldik, R. J . Chem. Soc., Dalton Trans. 1996, 1045.
(13) Go´mez, M.; Granell, J .; Mart´ınez, M. Organometallics 1997,
16, 2539.
(27) Font-Bard´ıa, M.; Gallego, C.; Gonza´lez, G.; Mart´ınez, M.;
Merbach, A. E.; Solans, X. Dalton 2003, 1106.
(28) Esteban, J .; Font-Bard´ıa, M.; Gallego, C.; Gonza´lez, G.;
Mart´ınez, M.; Solans, X. Inorg. Chim. Acta 2003, 351, 269.
(29) Bartlett, K. L.; Goldberg, K. I.; Thatcher Borden, W. Organo-
metallics 2001, 20, 2669.
(30) Williams, B. S.; Goldberg, K. I. J . Am. Chem. Soc. 2001, 123,
2576.
(14) Go´mez, M.; Granell, J .; Mart´ınez, M. Eur. J . Inorg. Chem. 2000,
217.
(31) Fekl, U.; Zahl, A.; van Eldik, R. Organometallics 1999, 18,
4156.
10.1021/om0499409 CCC: $27.50 © 2004 American Chemical Society
Publication on Web 04/10/2004

Selective Associative Activation in Pt(IV) Complexes
Organometallics, Vol. 23, No. 10, 2004 2435
Ch a r t 1
Sch em e 1
constant for the dissociatively activated mechanism, k1,
has been found to be the same as the value obtained
for the exchange of coordinated SMe2, kex, as measured
from magnetization transfer data.32 Furthermore, moni-
toring of the substitution reactions in the absence of
leaving ligand produced well-behaved first-order absor-
relatively small entering (PPh3) and leaving (SMe2)
ligands the presence of an associatively activated path
has been detected.25
We present in this paper the study of the activation
volumes for the substitution reactions of a series of
bance versus time traces, from which the value of k1 is
27,28
obtained as kobs
.
This behavior allows an easy
6
neutral and cationic Pt(IV) octahedral t2g complexes
determination of k1, as well as of the corresponding
volume of activation, with a single set of experiments.
For the associatively activated process, k2′, a full range
of entering ligand concentrations had to be used at each
pressure for the measurements.
depicted in Chart 1. The values obtained agree with the
operation of a dissociatively activated mechanism for
all the systems studied, the volume of activation being
directly related to the size of the leaving ligand, as well
as its intramolecular interactions with the inert skeleton
on decoordination. Furthermore, for the neutral sys-
tems, the solvent effects detected for systems with
suitable proton centers, which allow solvent-assisted
interactions of the leaving ligand with the rest of the
molecule in the transition state, are made evident by
the activation volume data determined. For the cationic
complexes, the solvent effects agree very well with what
should be expected from both the polarity of solvent used
and the degree of increase of electrostriction in the
transition state once the leaving ligand has began
dissociation.
For the dissociatively activated substitution processes
occurring on complexes 2ClSMe, 2ClSEt, 2ClSBzl,
2ClP P h , and 2Clp y the volumes of activation shown
in Table 1 correlate with the bulk of the leaving ligand
dissociating in the transition state. That is, for systems
in which no solvent involvement in the transition state
has been established from the values of ∆Sq (2ClSMe,
2ClP P h , and 2Clp y in acetone solution and 2ClSEt
and 2ClSBzl in toluene solution),25 the values increase
monotonically on going from SMe2 to SEt2, SBzl2, and
PPh3. Even for the substitution processes studied in
acetone solution for complexes with solvent involvement
in the transition state (2ClSEt and 2ClSBzl), the trend
is the same, although the difference between the expan-
sion on dissociative activation in acetone and toluene
for the SEt2 is larger. If the solvent-assisted interaction
of the CH2 protons of the ligand35 with the complex
spectator ligands is responsible for the negative values
obtained for ∆Sq, the less bulky SEt2 ligand could more
Resu lts a n d Discu ssion
Substitution processes taking place on the neutral
complexes indicated in Chart 1 (1F SMe, 1ClSMe,
1Br SMe, 2ClSMe, 2ClSEt, 2ClSBzl, 2ClP P h , 2Clp y,
3F SMe) have been found to occur through the combina-
tion of dissociatively and associatively activated reaction
mechanisms shown in Scheme 1.25 Nevertheless, only
for the compound 3F SMe has the path described by the
rate constant k2′ been found to be operative when the
entering phosphine ligands have cone angles θ e 165°.
The rate laws indicated in eqs 1 and 2 have been applied
(32) Magnetization transfer experiments between free and coordi-
nated SMe2 have been carried out on 1ClSMe at different tempera-
tures. The proton signal of the dimethyl sulfide ligand was separated
enough from that of free SMe2 to run such experiments (Figure S1,
Supporting Information). The values obtained for the chemical transfer
constant, kex, using the CIFIT program33,34 (1.8 × 10-2, 8.3 × 10-2
,
2.9 × 10-1 and 86 × 10-1 s-1 at 10, 20, 30, and 40 °C), are in perfect
agreement with the substitution-measured decoordination k1 values
(Figure S2, Supporting Information). The same sort of experiments
could not be carried out with the other complexes studied, due to the
proximity of the bound and free SMe2 proton signals.
k-1k-2[L] + k1k2[E]
kobs
)
[E] . [L]8 kobs ) k1 (1)
k-1[L] + k2[E]
kobs ) k2′[E]
(33) Bain, A. D.; Cramer, J . A. J . Magn. Reson. A 1996, 118, 21.
(34) North, M. R.; Swaddle, T. W. Inorg. Chem. 2000, 39, 2661.
(35) Goodfellow, R. J .; Goggin, P. L.; Duddell, D. A. J . Chem. Soc.
A 1968, 504.
(2)
in all cases; the value of the limiting first-order rate

2436 Organometallics, Vol. 23, No. 10, 2004
Gallego et al.
Ta ble 1. Kin etic a n d Activa tion P a r a m eter s for th e Su bstitu tion Rea ction s Stu d ied Accor d in g to Ch a r t 1
a
b
d
Activation volume data in acetone solution, unless differences are stated. From refs 25 and 26. c From ref 28. From the second-
order rate constant.25 e Interpolated from Eyring plots.
easily establish solvent-assisted interactions with the
inert skeleton of the complex, producing a larger com-
pressive contribution to the activation volume.
same reasoning applies to the difference in the values
on going from 2ClSMe (16.3 cm3 mol-1) to 3F SMe (10.3
cm3 mol-1), where perfluorination increases, even more,
the electron-withdrawing effect on the Pt(IV) center. In
this case, the enthalpic demands for the transition state
are even smaller for the 3F SMe complex, and this
should also produce an earlier position of the transition
state for the process; the limit would be the changeover
to an associatively activated substitution mechanism
detected for some of the reactions studied for this
complex, which are discussed below.
With respect to the trends observed for the 1F SMe,
1ClSMe, and 1Br SMe series of compounds, it is inter-
esting to note that the 1ClSMe and 1Br SMe pair
probably also are in agreement with the notion of a
longer Pt-S bond for the bromo derivative in the ground
state, as shown in the X-ray crystal structures of similar
Cl/Br pairs (0.03 C).27 For the 1F SMe fluoro complex,
neither the size nor the electronic characteristics of the
In the same line, for the dissociatively activated
systems, the differences found for the expansion on
dissociation of the dimethyl sulfide ligand from com-
plexes 1ClSMe (20.0 cm3 mol-1) and 2ClSMe (16.3 cm3
mol-1) has to be related with the early/late position of
the transition state in the reaction coordinate for the
same value of activation enthalpy (Table 1). The greater
electron-withdrawing characteristics of the perchlori-
nated ligand in 2ClSMe does not allow the same
advance of dissociation/expansion to go to a transition
state with the same enthalpic expense. Although these
differences are already visible from the ∆Sq data, the
values of ∆Vq are much more reliable and involve a
smaller intrinsic error. ∆Vq is calculated from the slope
of the ln k versus P plot (Figure 1, Table 1), not from
the extrapolation at T ) ∞ in classical Eyring plots. The

Selective Associative Activation in Pt(IV) Complexes
Organometallics, Vol. 23, No. 10, 2004 2437
F igu r e 2. Pressure dependence in acetone solution of the
value of k1 (PCy3 as entering ligand, 49.8 °C, O) and k2′
(PPh3 as entering ligand, 5.5 °C, b) for 3F SMe.
F igu r e 1. Pressure dependence of the value of the dis-
sociation constant, k1, in acetone solution: (O) 1F SMe (39.0
°C); (0) 1ClSMe (30.1 °C); (]) 1Br SMe (30.1 °C); (4)
2ClP P h (40.0 °C); (3) 2Clp y (13.0 °C).
substitution mechanism on 3F SMe with phosphines
with θ < 165°.25 As seen in Figure 2, the pressure
dependence of the first-order (k1) and second-order (k2′)
rate constants (Scheme 1) is the opposite, producing ∆Vq
values of 10.3 cm3 mol-1 for dissociative activation and
of -18.3 cm3 mol-1 for associative activation. This
change is not surprising, as the aforementioned rela-
tively low values of ∆Vq and ∆Hq for the dissociative
path (k1) point to a definitively lesser degree of dissocia-
tion in the transition state. This is especially true by
taking into account the inherently stronger Pt-S bond
in the complex, due to the increase of the electron-
withdrawing character of the spectator ligands. Never-
theless, these differences have not produced the same
effect on the 1F SMe complex or on the same perfluori-
nated complex with the SBzl2 ligand instead; in these
cases the mechanism is fully dissociative, as found for
the other systems.25 We have prepared the equivalent
SEtMe and SEt2 derivatives of the perfluorinated com-
plex38 in order to tune as much as possible the associa-
tiveness/dissociativeness of the substitution processes
studied. For both complexes the reactions carried out
in acetone solution at 20 °C have been found to occur
solely through the dissociatively activated path char-
acterized by k1, even with pyridine as entering ligand.
The values of the first-order dissociative rate constant
with PPh3 and py entering ligands (0.027 and 0.060 s-1
for the SEtMe and SEt2 derivatives, respectively) are
probably too high to enable the operation, under the
experimental conditions, of the associatively activated
path. The extra bulkiness of the sulfide ligand makes
difficult the partial association of a seventh ligand to
the platinum center; in fact, an increase of 5° (165 to
170°) in the cone angle of the entering phosphine ligands
has already been proved enough to produce a complete
changeover in the reaction mechanism from an as-
sociatively to a dissociatively activated one, as observed
F ligand explain in a simple way the small positive value
for the activation volume. For this compound, the
negative values of ∆Sq have been tentatively associated
with the formation of intramolecular interactions be-
tween the dissociating sulfide and the halogen ligand,
interactions that are not assisted by the solvent.25 In
fact, the values determined for ∆Sq are the same in
acetone and toluene solutions. Therefore, although the
dissociation of the SMe2 ligand should produce an
expansion effect, the existence of the above-mentioned
interactions should also create a compression effect,
even for this dissociatively activated path.36,37 Both
contributions would be reflected in a less positive ∆Vq
value than expected, as indicated in Table 1.
These general trends are evident also in the studied
ionic derivatives 4MeSMe and 4MeP P h in Chart 1,
where the increase in leaving ligand size from SMe2 to
PPh3 produces a dramatic increase of ∆Vq from 6.1 to
15.5 cm3 mol-1, the same increment of ca. 9 cm3 mol-1
as that observed for the above-described neutral 2ClS-
Me and 2ClP P h systems. Furthermore, the enthalpic
demand for the process agrees very well with such a
dissociative activation; the stronger PtIV-P bond needs
a much larger contribution of enthalpy to get to such
an expanded transition state. Equilibrium constant
determinations carried out on these types of complexes28
agree with the preference of this Pt(IV) compound for
phosphines versus sulfides (SMe2, SBzl2) and amines
(MeNH2, Me2NH, Me3N). The ionic nature of the com-
plex explains the generally smaller expansion to go to
the transition state; electrostriction of the solvent
(acetone) should occur on the increase of the charge
density produced by partial dissociation of the leaving
ligand. When the reaction is carried out in toluene, the
important difference in the value of ∆Vq is associated
with the very small compressive electrostriction contri-
bution of the solvent to the expansion on dissociation.
The negative ∆Vq entry for the 3F SMe/PPh3 system
in Table 1 corroborates the previously published obser-
vation of the operation of an associatively activated
(38) The equivalent SEtMe- and SEt2-derived complexes have been
prepared from substitution of the SMe2 ligand in 1F SMe with the
corresponding sulfides. The complexes were characterized by their
significant 1H NMR spectra (250 MHz, CDCl3, 298 K, ppm, TMS):
SEtMe derivative, 1.02 (d, J (FH) ) 7.7 Hz, J (PtH) ) 66 Hz, Meax),
1.63 (dd, J (FH) ) 9.3 Hz, J (FH) ) 7.3 Hz, J (PtH) ) 62 Hz, Meeq), 2.00
(s, J (PtH) ) 12 Hz, MeSEtMe), 1.10 (t, J (HH) ) 7.5 Hz, EtSEtMe, (CH3)),
2.51 (m, EtSEtMe, (CH2)), 8.65 (s, J (PtH) ) 47 Hz, CH); SEt2 derivative,
0.89 (d, J (FH) ) 7.4 Hz, J (PtH) ) 68 Hz, Meax), 1.58 (dd, J (FH) ) 9.3
Hz, J (FH) ) 7.2 Hz, J (PtH) ) 63 Hz, Meeq), 1.05 (t, J (HH) ) 7.4 Hz,
SEt2(CH3)), 2.53 (m, SEt2(CH2)), 9.17 (s, J (PtH) ) 48 Hz, CH).
(36) Benzo, F.; Bernhardt, P. V.; Gonza´lez, G.; Mart´ınez, M.; Sienra,
B. J . Chem. Soc., Dalton Trans. 1999, 3973.
(37) Mart´ınez, M.; Pitarque, M. A.; van Eldik, R. Inorg. Chim. Acta
1997, 256, 51.

2438 Organometallics, Vol. 23, No. 10, 2004
Gallego et al.
for other known systems.39,40 Consequently, the ex-
pected stronger Pt-S bond for these complexes with
greater σ-donor capabilities35 cannot compensate for the
difficulties in the transient entry of a seventh ligand in
the coordination sphere.
wavelengths where a maximum increase and/or decrease of
absorbance was observed. For runs at variable pressure, a
homemade stopped flow device connected to a J &M TIDAS
spectrophotometer was used.42 No dependence of the observed
rate constant values on the selected wavelengths was detected,
as expected for reactions where a good retention of isosbestic
points is observed. The general kinetic technique is that
previously described;43 in all cases the platinum concentration
was kept at (2-6) × 10-4 M to avoid undesired decomposition
reactions.26 For the dissociatively activated processes, given
the kinetic approach taken to measure the values of the
limiting rate constants indicated in eq 1, no leaving ligand was
added to the reaction mixtures monitored;27,28 in all cases the
observed absorbance versus time traces behaved as first order,
producing reproducible rate constants within a 5% error. Table
S1 (Supporting Information) collects all the obtained kobs values
for all the complexes studied as a function of the starting
complex, entering ligand, solvent, pressure, and temperature.
The fitting of the results to the equations leading to the
relevant kinetic and activation parameters was done by
commercial least-squares packages; the errors quoted in Table
1 correspond to the standard deviation of the fitting.
Exp er im en ta l Section
Com pou n ds. 1FSMe, 1ClSMe, 1Br SMe, 2ClSMe, 2ClSEt,
2ClSBzl, 2ClP P h , 2Clp y, 3F SMe, and 4MeP P h have been
prepared according to established literature methods.20,25,26,28,41
Their characterization was confirmed via 1H NMR. 4MeSMe
has been prepared by treatment of a solution of [Pt(Me)3I(2,2′-
bpy)] (3.8 × 10-4 mol in 25 cm3 of acetone) with AgCF3SO3
(3.8 × 10-4 mol); after it was stirred for 15 min, the solution
was filtered through Celite, 6.8 × 10-3 mol of SMe2 was added,
and the mixture was stirred for another ca. 30 min. The solvent
was evaporated at reduced pressure and the solid washed
several times with hexane. Yield: 72%. Anal. Calcd (found)
for C16H23F3N2O3S2Pt‚H2O: C, 30.7 (30.7); H, 4.0 (3.7); N, 4.5
(4.5); S, 10.3 (10.9). 1H NMR (250 MHz, acetone-d6, 298 K,
ppm, TMS): 0.64 (s, 2J (PtH) ) 70 Hz, Meax), 1.13 (s,
3
2J (PtH) ) 68 Hz, 2 × Meeq), 1.98 (s, J (PtH) ) 12 Hz, SMe2).
Ack n ow led gm en t. We acknowledge financial sup-
port for the BQU2001-3205 project from the Ministerio
de Ciencia y Tecnolog´ıa.
195Pt NMR (53.5 mHz, acetone-d6, 298 K, ppm, H2[PtCl6]):
-2827.
Kin etics. The reactions were followed by UV-vis spectros-
copy in the 500-330 nm range, where none of the solvents
absorb. At atmospheric pressure runs with t1/2 > 170 s were
recorded on an HP8452A instrument equipped with a ther-
mostated multicell transport; for runs with t1/2 < 7 s an Applied
Photophysics stopped-flow instrument connected to a J &M
TIDAS spectrophotometer was used. Observed rate constants
were derived from absorbance versus time traces at the
Su p p or tin g In for m a tion Ava ila ble: kobs values for all
the systems studied (Table S1), a magnetization transfer
experiment for the 1ClSMe system (Figure S1), and an Eyring
plot for k1 in acetone solution for the 1ClSMe system (Figure
S2). This material is available free of charge via the Internet
at http://pubs.acs.org.
OM0499409
(39) Cossy, C.; Helm, L.; Merbach, A. E. Helv. Chim. Acta 1987,
70, 1516.
(40) Sisley, M. J .; Yano, Y.; Swaddle, T. W. Inorg. Chem. 1982, 21,
1141.
(42) Bugnon, P.; Ducommun, Y.; Sauvageat, P.; Merbach, A. E.;
Ith, R.; Tschanz, R.; Doludda, M.; Bergbauer, R.; Grell, E. Anal. Chem.
1996, 68, 3045.
(41) Bayler, A.; Canty, A. J .; Skelton, B. W.; White, A. H. J .
Organomet. Chem. 2000, 595, 296.
(43) Mart´ınez, M.; Pitarque, M. A.; van Eldik, R. J . Chem. Soc.,
Dalton Trans. 1994, 3159.