Coordinatively Unsaturated Cp*Ru Alkoxo Complexes
Organometallics, Vol. 16, No. 15, 1997 3279
Below -60 °C the signals broaden again and at -80 °C
appear as two AB quartets. Chemical shifts at ambient
and low temperature are different from those of 3-3 in
toluene. Thus, most of the Ru complex is present as
the sulfide adduct. Simulating line broadening of the
o-protons and of the Me triplets yields activation ener-
gies around 30 kJ /mol. Though these figures are not
very precise in this case, they nevertheless prove an
association equilibrium faster than the inversion with-
out extra ligand present. It appears that, notwithstand-
ing a more mobile association-dissociation equilibrium,
the virtual inversion can still be slower than in the
dimer. The peculiar temperature-dependent behavior
of the benzylic protons points to the existence of two
processes with different activation energies (not resolved
explicitly in the experiment). One slows down inversion
and the second one, operative at still lower temperature,
causes splitting into diastereoisomers. The point is
further discussed below.
one another in the temperature interval of slow to
medium exchange.
Me3SiCtCH. In the case of this unsymmetrical
acetylene, adduct formation is less easy to prove. When
the acetylene is added in about 3-fold excess, the Me3Si
singlet at δ 0.28 starts to broaden from -20 °C but
exchange is slow only at -50 °C, at which point separate
signals for complexed and free acetylene have developed.
Since the chemical shift of the free acetylene Me3Si
group has undergone only a minor change between
ambient and low temperature (and in the wrong direc-
tion, i.e. downfield), it must be concluded that associa-
tion equilibrium at ambient temperature is not only fast
but also to a larger extent is shifted toward the free
complex: that is, 3-3. Consequently, line broadening
of the benzylic singlet starts at the same temperature
as in neat toluene but much lower temperature is
needed for individual lines to reappear. The spectrum
at -90 °C, though not yet sharp in the benzyl region,
shows a pattern similar to that of the diastereoisomeric
pair of dimers of 3 in toluene, but with additional peaks
for an acetylene adduct. Two methyl signals of unequal
intensity for complexed acetylene around δ 0.44 at this
temperature can be interpreted as originating from two
orientations of the acetylene with respect to the benzyl
group in the adduct. From line broadening of the ortho
protons the activation energy Ea ) 18 kJ /mol can be
estimated for the inversion of the acac ligand, which
is much lower than the value found for the dimer in
toluene. In this case the adduct seems to be rather
weak and to compete with the dimer.
p-Me-C6H4S(Me)O. The sulfoxide was added to the
toluene solution of 3 as the pure S enantiomer to obtain
a 3:p-Me-C6H4S(Me)O ratio of 1:1.4. Signal broadening
is already obvious for some signals at ambient temper-
ature. All signals broaden below 20 °C and are sharp
again at -45 °C. At this temperature the S(O)Me group
and, most noticeably, the ortho protons from the sulf-
oxide arene around δ 7.9 have separated into three sets,
assigned to the two diastereoisomeric adducts RRu and
SRu and the signal for the uncomplexed ligand. In the
same manner were Cp* and Me singlets from 3 split
into two signals, each of intensity ratio 1:0.45 (at -45
°C). Benzylic protons, at -45 °C, appear as two sharp
AB quartets of unequal intensity with chemical shifts
and intensity ratios distinctly different from those of 3
in toluene. Temperature-dependent line shapes of the
different groups were simulated, and the resulting k
values were found close enough to ascertain the same
process as being responsible for all line shape changes.
The value for Ea given in Table 2 has been obtained by
simulating the line shape of the Cp* signal. It also
became obvious during the simulation that intensity
ratios had to be changed for different temperatures,
being closer to 1 at higher temperature. The ratio
change in the temperature interval where separate
signals are seen for e.g. the Cp* group (-45 to -15 °C)
is, however, too small to allow thermodynamic param-
eters for the equilibrium of diastereoisomers to be
determined.
Discu ssion
Addition of two-electron-donor ligands L to the solu-
tions of the dimers 2-2 and 3-3 in all cases cleaves them
into monomeric adducts (2,3)‚L. This occurs with
π-acceptor as well as with pure σ-donor ligands. The
adducts are, however, of widely differing stability,
ranging from isolable compounds for phosphines, phos-
phites, and CO to solution species, whose formation is
evident from the influence of added ligand on the
inversion rate of the dimer.
Due to a number of shortcomings, such as chemical
shifts and equilibrium populations varying with tem-
perature, which cannot be evaluated above coalescence
temperature, some of the activation parameters Ea and
∆Gq in Table 2 pertaining to ligand association can
273
bear a greater uncertainty which is not reflected in the
standard deviation of the Arrhenius plots and are
believed to be realistic within 5-10%. Bearing in mind
these uncertainties, it still follows from Table 2 un-
ambiguously that for the ligand MeSEt not only the
association equilibrium but also the inversion, measured
in that case from line broadening of phenylic ortho
protons, is faster than for the pure dimer (Ea is below
the value of the dimer in pure toluene). The same is
true for the Me3SiCtCH adduct, where Ea has been
found from ortho proton line broadening as low as about
18 kJ /mol. The situation arises if, in spite of an
equilibrium shifted toward the adduct 3‚L, the latter is
kinetically still very fast. This fact arises because the
observed NMR lifetime is determined by the faster of
the two competing equilibria, adduct formation-dis-
sociation or dimerization. This will hold as long as the
inversion in the monomer is much faster than either
3-Cya n op yr id in e. The LT NMR of a solution con-
taining 3 and 3-cyanopyridine in the ratio 1:1.23 shows
two sets of pyridine signals in a corresponding ratio with
relatively large displacements, assigned to complexed
and free pyridine. Benzylic protons are split into a
single AB quartet; aromatic protons from the Phacac
ligand are not subject to any temperature-dependent
line broadening in this case. Moreover, Me and Cp*
signals remain singlets down to low temperature. This
means that the ligand gives rise to one single dia-
stereoisomer only, or more realistic, to two diasteromers
still rapidly interconverting at low temperature through
rotation around the Ru-N bond. The existence of a
sharp AB quartet, on the other hand, shows that the
process occurs without Ru-N bond cleavage. The first-
order pyridine exchange rate and the rate of acac
inversion have been found to be not much different from