Experimental and computational evidence for the participation of nonclassical dihydrogen species in proton transfer processes on Ru-Arene complexes with uncoordinated N centers. Efficient catalytic deuterium labeling of H2 with CD3OD

Article abstract of DOI:10.1021/om300015j

The protonation with Bronsted acids HB (HBF₄ and CF ₃CO₂H) of [RuH₂(arene)(PPh₂Het)] derivatives (PPh₂Het = dpim, Het = 2-N-methylimidazolyl; PPh ₂Het = PPh₂py, Het = 2-pyridyl) that contain uncoordinated N atoms has been analyzed experimentally by NMR spectroscopy and through computational studies. Initially, at low temperature, the uncoordinated N atoms of the phosphine are protonated and a proton-hydride exchange is observed by ¹H NMR spectroscopy. The proton transfer leads to the corresponding cationic trihydride intermediates, which exhibit a dual character of classical and nonclassical hydrides, with the nonclassical species being more stable, especially when Het = 2-pyridyl. In fact, the release of H₂ and the formation of the respective monohydride complexes [RuH(arene) (κ²-N,P-PPh₂Het)]B was observed at room temperature. The participation of the uncoordinated N center in the proton transfer process in the stabilization of RuH(H₂)⁺ with respect to RuH ₃⁺ species and in the hydride-dihydrogen exchange (cis effect) are discussed. Calculations on the complex [RuH₃(p-cymene) (PPh₃)]⁺ have also been carried out for the sake of comparison. A dual character was also found, but in this case the classical species is more stable. H/D exchange of the hydride ligands of the dihydride complexes, using CD₃OD as the deuterium source, has been studied. The very rapid deuterium labeling of H₂, catalyzed by the aforementioned dihydrides, has been observed. The known compound [RuH₂(p-cymene) (PPh₃)] is also active in this labeling process, and the possible mechanism for both the H/D exchange and the deuterium labeling of H₂ is discussed in light of theoretical studies.

Full text of DOI:10.1021/om300015j

Article
pubs.acs.org/Organometallics
Experimental and Computational Evidence for the Participation of
Nonclassical Dihydrogen Species in Proton Transfer Processes on
Ru−Arene Complexes with Uncoordinated N Centers. Efficient
Catalytic Deuterium Labeling of H2 with CD3OD
Gustavo Espino,† Agustín Caballero,‡ Blanca R. Manzano,‡ Lucía Santos,§ Mercedes Perez-Manrique,†
́
Miquel Moreno,⊥ and Felix A. Jalon*,‡
́
́
†
́
́
Departamento de Quımica, Facultad de Ciencias, Universidad de Burgos, Plaza Misael Banuelos s/n, 09001 Burgos, Spain
̃
‡
́ ́
́ ́
ica, Organica y Bioquımica, Facultad de Quımicas, Universidad de Castilla-La Mancha,
Departamento de Quımica Inorgan
Avda. Camilo J. Cela 10, 13071 Ciudad Real, Spain
§
́
́
́
Departamento de Quımica Fısica, Facultad de Quımicas, Universidad de Castilla-La Mancha, Avda. Camilo J. Cela s/n,
13071 Ciudad Real, Spain
⊥
́
̀
Departament de Quımica, Edifici Cn, Universitat Autonoma de Barcelona, 08193 Bellaterra, Spain
S
* Supporting Information
ABSTRACT: The protonation with Brønsted acids HB
(HBF4 and CF3CO2H) of [RuH2(arene)(PPh2Het)] deriva-
tives (PPh2Het = dpim, Het = 2-N-methylimidazolyl; PPh2Het =
PPh2py, Het = 2-pyridyl) that contain uncoordinated N atoms
has been analyzed experimentally by NMR spectroscopy and
through computational studies. Initially, at low temperature,
1
the uncoordinated N atoms of the phosphine are protonated and a proton−hydride exchange is observed by H NMR
spectroscopy. The proton transfer leads to the corresponding cationic trihydride intermediates, which exhibit a dual character of
classical and nonclassical hydrides, with the nonclassical species being more stable, especially when Het = 2-pyridyl. In fact, the
release of H2 and the formation of the respective monohydride complexes [RuH(arene)(κ2-N,P-PPh2Het)]B was observed at
room temperature. The participation of the uncoordinated N center in the proton transfer process in the stabilization of
RuH(H2)+ with respect to RuH3+ species and in the hydride−dihydrogen exchange (cis effect) are discussed. Calculations on the
complex [RuH3(p-cymene)(PPh3)]+ have also been carried out for the sake of comparison. A dual character was also found, but
in this case the classical species is more stable. H/D exchange of the hydride ligands of the dihydride complexes, using CD3OD as
the deuterium source, has been studied. The very rapid deuterium labeling of H2, catalyzed by the aforementioned dihydrides, has
been observed. The known compound [RuH2(p-cymene)(PPh3)] is also active in this labeling process, and the possible
mechanism for both the H/D exchange and the deuterium labeling of H2 is discussed in light of theoretical studies.
The initial interaction of the electrophilic proton of an acid,
HB, with the metal hydride moiety (step i) is believed to be the
first step in this process. In comparison with the direct proton
transfer onto the metal center, prior attack on the hydride ligand
(or on the M−H bond) has been established as kinetically more
favorable in some systems.5,7 The product of this first step is a
dihydrogen-bonded complex8 that, when sufficiently stable, can
be characterized by crystallographic9 or spectroscopic tech-
niques such as IR and NMR.10 As in a conventional hydrogen
bond, the H···H contact involved in dihydrogen-bonded species
is mainly electrostatic in nature and its formation implies the
weakening of the M−H and B−H bonds. As a consequence of
this contact, a downfield or upfield shift of the proton resonances
for the B−H and M−H bonds, respectively, is normally observed
INTRODUCTION
■
Protonation of transition-metal hydride complexes, LnMHx, with acids
of different strengths is one of the most useful ways to form non-
classical dihydrogen compounds,1 and interesting examples of classical
polyhydride2 or hydride−dihydrogen complexes3 have been reported
as products. In the majority of examples, the resulting poly-
hydrides4 or hydride−dihydrogen derivatives5 are structurally labile
and exchange their positions in the metal coordination sphere.
It is widely accepted that a proton transfer process to a
transition metal hydride or polyhydride follows several steps, as
depicted in eq 1.6
(i)
MH + HB ⇌ MH···HB
1
in the H NMR spectra. Additionally, a faster hydride relaxation
(ii)
+
−
HooI [M(H )] ···B
2
(iii)
Received: January 6, 2012
Published: March 20, 2012
2
+
−
HoooI [M(η ‐H )] + B
(1)
2
© 2012 American Chemical Society
3087
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
rate is also a consequence of the dihydrogen bond formation, an
aspect that can be studied by NMR techniques. Step ii involves the
full proton transfer from the conjugate base B− to the MH bond,
with the anion B− maintaining an electrostatic contact with the
coordinated H2 molecule as an ion pair. Cases in which this
intermediate has been detected spectroscopically are rare, but
Shubina and others have studied several of these systems.11 Step iii
involves ion separation induced by solvation. From this
dihydrogen-coordinated compound several transformation path-
ways are possible, depending on the electron density of the metal
center. Electron-rich centers promote the oxidative addition of H2
with or without a subsequent cis−trans MH2 isomerization. On the
other hand, in complexes with electron-poor metal centers the loss
of the H2 molecule would be possible with or without coordination
of the counteranion or a solvent molecule.12
If the sequence of reactions in eq 1 is regarded as reversible,
they would be the basis of interesting reactivity behavioreven
of catalytic processes, such as hydrogen transfer, hydrogenation
by heterolytic cleavage of H2,13 ionic hydrogenation,14 hydro-
genation of CO2,15 H2 and CH labeling,5 etc. The feasibility of
designing systems with a reversible reaction sequence is not
certain. However, some easy tests can be carried out to demon-
strate the ability of transition-metal hydride or polyhydride
derivatives to behave in such a way. Specifically, the proton−
hydride exchange between the hydride coordinated to a metal
and the external acid species (alcohols for instance) is con-
sidered as evidence that steps i and ii are consecutive and
reversible. In addition, the labeling of H2 with an external acid
as a deuterium source can be considered as a sign that the
sequence step i−step iii is reversible and that the dihydrogen
coordination is labile and also reversible.
In the work described here we focus on the study of the
protonation process of [RuH2(arene)(PPh2Het)] derivatives
(Het = 2-pyridyl, 2-N-methylimidazolyl) with HBF4, both by ex-
periments and with the aid of theoretical calculations. The main
goal of the work was to analyze the effect of the Het moiety of the
phosphines on the protonation process, more specifically the effect
on (i) the reversibility of the different steps of the proton transfer
process, (ii) the participation of hydrogen-bonded and nonclassical
dihydrogen intermediates, and (iii) the stability of classical and
nonclassical species. The possible labeling of H2 within these
species was another aim of this work. In order to elucidate the role
of the acid in the overall process, we also analyzed the effect of
replacing HBF4 by CF3CO2H, a weaker acid that has a higher
ability to coordinate its conjugated base.
RESULTS AND DISCUSSION
■
Preparation and Characterization of the Dihydride
Derivatives. Three new η6-arene ruthenium dihydrides of
general formula [RuH2(η6-arene)(κ1-P-PPh2Het)] (PPh2Het =
dpim, Het = 2-N-methylimidazolyl, arene = p-cymene (2b),
benzene (2c); PPh2Het = PPh2py, Het = 2-pyridyl, arene =
p-cymene (2d)) (see Scheme 1) were prepared from the
Scheme 1. Reaction Scheme for the Formation of 2a−d
As far as the chemistry of Ru−arene polyhydride derivatives is
concerned, it has been reported that the protonation of [RuH2-
(arene)(PR3)] (R = alkyl, aryl) leads to trihydride Ru(IV) com-
plexes: [RuH3(arene)(PR3)]+.2a,16 These compounds are now con-
sidered as classical species in analogy with the similar Os(IV)16
and neutral cyclopentadienyl half-sandwich Ru(IV) trihydrides.17
However, if the higher tendency of Ru vs Os and of the cationic vs
neutral complexes to stabilize coordinated H2 molecules is taken
into account, some doubts emerge regarding the nature of this
family of complexes.
We recently designed some systems bearing ligands with
noncoordinated nitrogen centers in order to control and
modify the thermodynamics and the kinetics of the proton
transfer reaction.18 These basic centers are first protonated and
serve as a proton relay, transforming the proton transfer to the
metal by the base into an internal sphere process18,19 so that
the intermediate structures shown in eq 1 are stabilized and can
be easily studied using spectroscopic techniques.20
corresponding dichlorides, which had been described pre-
viously,22 by reaction with excess KBH4 in ethanol. For the sake
of comparison, the previously known2a [RuH2(η6-p-cymene)(κ1-P-
PPh3)] (2a) was also prepared. In the case of 2d it was confirmed
that the synthesis is also possible using an excess of LiAlH4 as
the hydrogen source in tetrahydrofuran (THF) (Scheme 1).
The new complexes 2b−d are sufficiently stable to be fully
1
characterized by elemental analysis and IR and 31P{1H}, H,
Most recently, Jimen
́
ez-Tenorio et al.21 studied proton
and 13C{1H} NMR spectroscopy. 2D-NMR experiments were
also carried out in order to assign all the signals.
transfer processes starting from [Cp*RuH3(κ1-P-iPr2PCH2X)]
(X = pyridine, quinoline) derivatives. Experimental and computa-
tional information was compared. As previously studied by us,18
these types of ligands are initially protonated at the heterocycle
N-donor center and stabilized as conventional hydrogen-bonded
(N···HA) or unconventional dihydrogen-bonded (RuH···HA)
compounds when weak acids are used. With strong acids, e.g.
CF3SO3H, such adducts are only stable at low temperature. An
increase in temperature initiates a complete proton transfer
process from the NH groups to the hydride ligands, ending in
the release of H2 and the formation of thermally stable
[Cp*Ru(H2)(κ2P,N-iPr2PCH2X)]+ species.
All of the new dihydride derivatives exhibit singlets in the
31P{1H} NMR spectra (acetone-d6), and these are at chemical
shifts higher than those exhibited by their dichloride precursors:
δ 52.06, 51.14, and 71.04 ppm for 2b−d, respectively, and 7.58,
11.85, and 23.54 ppm for 1b−d, respectively, in chloroform-d.
1
The hydride atoms give rise to one sharp doublet in the H
NMR spectra at about −10 ppm, and these integrate for two
protons. The 1H{31P} NMR data show that the coupling is with
the phosphorus atom (2JPH = 40−44 Hz), and the data are also
consistent with literature values for similar compounds.2a,16
3088
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
Protonation Experiments in an Equimolar Ratio. The
reaction at room temperature of the p-cymene derivatives 2b,d,
dissolved in acetone-d6, with equimolar amounts of the strong
acids HB (HBF4 and CF3CO2H) led to the formation of a
monohydride species of general formula [RuH(η6-p-cymene)-
(κ2P,N-PPh2Het)][B] (see Scheme 2). In this reaction, molecular
In order to shed some light on the proton transfer mechanism,
we carried out the acid addition at −90 °C and monitored the
1
proton transfer reaction from −90 to 20 °C by H and 31P{1H}
NMR spectroscopy. The NMR data are discussed in the following
paragraphs, and the HBF4 and CF3CO2H additions are covered
separately.
Addition of HBF4 to 2b or 2d in an Equimolar Ratio.
The addition of HBF4 to 2b,d was performed, and a compar-
able behavior was observed in each case. Thus, the results and
conclusions of both experiments will be explained together.
Initially, just after the addition of the HBF4·Et2O solution
(equimolar amounts) at low temperature (−90 °C), a new
species (3b,d) with 1H and 31P NMR resonances different from
Scheme 2. Formation of the Monohydride Species
1
those of 4b,d or the starting materials was observed. In the H
NMR spectra broad bands were observed at very low field:
13.40 and 12.57 ppm, respectively. These signals are assigned
to N−H protons (see Figure 1). It can be concluded that the
acid proton has been transferred to the basic sp2 nitrogen in
both coordinated ligands (dpim and PPh2py), giving rise to the
protonated derivatives [RuH2(η6-p-cymene)(κ1P-PPh2HetH)]-
[BF4] (3b,d) (see eq 2). New singlets were recorded in the
31P{1H} NMR spectra at 59.01 and 81.50 ppm for 3b,d, re-
spectively. Concerning the hydride resonances, the protonated
compound 3b shows, at −90 °C, an unchanged signal, in both
chemical shift and resolution, with respect to the non-
protonated 2b. In contrast, 3d presents at this temperature a
broadened signal for the hydride resonance that is shifted
upfield 0.65 ppm in comparison to 2d. This broadening and
shift of the hydride resonance of 3d can be ascribed in a first
attempt to an interaction with the proton in the NH group,
through a dihydrogen bond, that is operative in 3d but not in
3b at this temperature.
hydrogen was also released, as detected in experiments with
Young valve NMR tubes. These results contrast with the
reported formation of trihydride derivatives [RuH3(η6-arene)-
(PR3)][B]2a,16 (PR3 = trialkyl, triarylphosphines) as final prod-
ucts after the protonation of the dihydride complexes con-
taining PR3 instead of the PN ligands. This difference suggests
that in our family of trihydride complexes an easy access to a
stable hydride−dihydrogen structure is possible. One novel
aspect of this example is that this hydride−dihydrogen species
would undergo substitution of the H2 molecule by the basic N
1
center of the phosphine. The H NMR hydride resonances,
which integrate for one H, are doublets that appear at −5.91
and −5.30 ppm for 4b,d, respectively. In addition, the positions
of all of the signals in the 1H NMR spectra are modified and the
resulting patterns fit best to metal-based chiral compounds with
diastereotopic isopropyl methyl groups and ABCDX (X = P)
spin systems for the aromatic p-cymene protons. Upfield-shifted
singlets were observed in the corresponding 31P{1H} NMR
spectra, at 22.32 ppm (4b) and 4.59 ppm (4d), in agreement with
a chelating bidentate coordination mode for the PN ligands.22,23
6
1
[RuH (η ‐ p ‐ cymene)(κ P‐PPh Het)]
2
2
PPh Het = dpim (2b), PPh py (2d)
2
2
HBF
4
6
1
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ [RuH (η ‐ p ‐ cymene)(κ P‐PPh HetH)][BF ]
2
2
2
4
−90 °C, acetone‐d
6
PPh Het = dpim (3b), PPh py (3d)
2
(2)
Figure 1. Evolution of high-field peaks (RuH2, right) and low-field peaks (N−H, left) on increasing the temperature in the protonation experiment
on 2b (a) and 2d (b) with HBF4 in acetone-d6.
3089
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
Figure 2. T1(ms) determination as a function of temperature (K) for the hydride resonances of 2b,d (blue sequences) and 3b,d (red sequences) in
acetone-d6. A 300 MHz spectrometer was used.
The broadness of the N−H bands, for 3d but also for 3b,
depending on the nature of the phosphine ligand, and so each case
will be discussed separately in the following paragraphs.
suggests that at this temperature the acidic proton is inter-
−
changing between the N atom and (probably) the BF4 anion,
For the ligand dpim (pair of species 2b and 3b), the
determination of T1 was carried out for the hydride doublet
peaks (Figure 2a). All values obtained for 3b from 183 to 203 K
are in the range of a classical hydride species and are very
similar to those of 2b, meaning that 3b is essentially a classical
dihydride in this temperature range. Nevertheless, from 203 to
243 K, a significant deviation to lower T1 values was obtained
for 3b in comparison to 2b. This observation, along with the
unchanged chemical shift, suggests the participation of an
incipient dihydrogen-bonded species. Importantly, from 243 K
on, the slope of the curve that represents the T1 values changes
from positive to negative, which could indicate a more signifi-
cant participation of nonclassical species at the highest tem-
peratures studied.
With regard to ligand PPh2py (species 2d and 3d), T1
determinations were carried out on the hydride doublet for
2d and on the N−H signal in the case of 3d, since the last
resonance shows fast exchange with the hydride resonance and
the latter signal is too broad to record the T1 measurement over
a wide temperature range (Figure 2b). The T1 value for N−H
and metal−H groups has been demonstrated to be equal when
both hydrogen atoms are in mutual exchange in a transition-
metal derivative.18c,24 In contrast to the results obtained for 3b,
T1 values for 3d over the whole temperature range fall within
the accepted range for dihydrogen species. The minimum value
is T1,min(3d) = 30 ms at 161 K (extrapolated value), which is
clearly lower than the value of 2d, T1,min(2d) = 240 ms at
193 K. In addition, it is possible to calculate the d(H−H) value for
the slow rotation regime, 1.20 Å, and the fast rotation regime,
0.95 Å.25 Both distances are in agreement with calculated values
for other nonclassical hydride−dihydrogen ruthenium com-
pounds in the literature.26,27 Therefore, it can be concluded that
3d can access a nonclassical RuH(H2) state even more easily
than 3b. As a comparative point with 3d, the T1,min value was
also determined for the N−H group of the chloride complex
[RuCl2(p-cymene)(PPh2pyH)]BF4 (300 MHz) in acetone-d6.
The value obtained of 298 ms is very different from that of 3d,
showing the effect of the hydride on relaxation.
maybe as a consequence of ion pairing. The solvent, acetone-d6,
could alternatively participate in this interchange. Analogous
“HBF4 adducts” of precursors 1b,c with the general formula
[RuCl2(η6-arene)(κ1-P-dpimH)][BF4] (arene = p-cymene,
benzene) have been reported by our group, thus showing the
ability of Ru-coordinated dpim to accept one proton.22 Like-
wise, the ligand PPh2py has been shown to favor intramolecular
proton-transfer reactions after the pyridine protonation.18a,c
When the temperature is increased, the N−H resonance and
the hydride doublet became broader, even for 3b, a fact that
points to a second dynamic interchange phenomenon between
the two atoms (Figure 1) that is opened at lower temperature
in 3d than in 3b. At 0 °C the broad band for the N−H protons
had disappeared from the baseline for both compounds. The
hydride resonance at 0 °C has also disappeared for 3d, but it
still appears at approximately −10.5 ppm as a broad doublet for
3b. At 20 °C this doublet had transformed into an unresolved
broad signal centered at −10.98 ppm. The position of this
resonance supposes an upfield shift of 0.36 ppm with respect to
the hydride signal of 2b at the same temperature and suggests
the presence of a hydrogen bond in 3b. The coalescence
temperatures for both compounds could not be determined,
due to the large separation between the two resonances, but a
magnetization transfer experiment confirmed unequivocally the
H+/H− exchange. An incipient doublet at −5.91 ppm for 4b
appeared at 20 °C, whereas at −5.30 ppm the analogous signal
appeared for 4d even at −60 °C. Both of these hydride resonances
integrated for one proton. These signals became more intense with
time at the expense of the hydride resonances for 3b,d.
The formation of the monohydride derivatives 4b,d
constitutes indirect evidence for the release of H2 from an
intermediate of the type [Ru(arene)H(H2)(PPh2Het)]BF4. This
coordinated dihydrogen molecule must be easily replaced by the
N-donor atom of the −Het fragment.
T1 Measurements and T1(min) Calculations. T1 mea-
surements were carried out in order to provide evidence about
the participation of intermediates, such as dihydrogen-bridged
or nonclassical species, contributing to the hydride relaxation in
the proton transfer mechanism. A 300 MHz spectrometer and
the inversion-recovery method were used in all these studies.
The measurements were made at different temperatures for
both the original dihydrides (2b,d), as reference systems, and the
corresponding protonated species (3b,d) in order to make suitable
comparisons possible (Figure 2). Different results were obtained
Considering the data presented above, there are several
pieces of evidence that support the easy access to nonconven-
tional trihydride species:
(i) the observed broadening of the hydride signals at inter-
mediate temperatures, which suggest chemical exchange
with the acidic proton through dihydrogen-bonded species
3090
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
may indicate a formula of the type [RuH(p-cymene)(S)(κ1P-
PPh2HetH)](BF4)2 (S = solvent) for these compounds, but a
definitive characterization was not carried out due to a lack of
stability. On the other hand, the 31P NMR spectra of these final
compounds, with resonances at 55.0 and 68.7 ppm for products
from 2b,d, respectively, suggest κ1P coordination of the
phosphine, and this leads to resonances very different from
those found for an κ2N,P coordination in 4b,d (22.32 (4b) and
4.59 ppm (4d)). Similarities in the T1,min values of the hydride
resonances (240 ms at 240 K) with those calculated for 2b,d for
these final hydride compounds also points to their classical
nature. On the basis of these observations, we can conclude
that, unfortunately, release of the coordinated dihydrogen is not
avoided by the protonation of the N-donor centers. This
finding is in contrast with the results obtained for the afore-
mentioned protonation of [Cp*RuH3(κ1P-iPr2PCH2X)] com-
pounds, which give [Cp*RuH2(H2)(κ1P-iPr2PCH2XH)] de-
rivatives.21 In our opinion, the stabilization of the H2
coordination in the cyclopentadienyl compounds may be due
to the existence of a more electron-rich Ru center due to the
more pronounced donor character of Cp* with respect to the
p-cymene and the presence of two hydrides instead of one.
Computational Studies on the Proton Transfer
Process. Considering the experimental evidence discussed
above on the mechanism for the proton transfer reaction, we
carried out a theoretical study at the DFT (B3LYP) level for
the protonation of 2b with a single H+. The goal was to
determine a reaction pathway for the proton transfer process
from 2b to 4b and to evaluate the feasibility of the existence of
nonclassical and classical polyhydride species.
The protonated compound 3b can undergo different
transformations, and these are shown in Chart 2 along with
the pure electronic + nuclear energies as directly obtained from
the quantum mechanical calculations. Evaluation of the zero
point vibrational energy, the enthalpy energy, and the entropy
term leads to the Gibbs function energies, which are also given
for all the structures shown in Chart 2. Solvation energies
derived from single-point calculations using a continuum model
have been added to all the calculated structures.
As a reasonable starting point we chose 3b, whichas deduced
from the NMR studiesat low temperature has undergone
protonation on the N-imidazolyl atom. Two dihydrogen-
bonded minima, BF and FS1, prior to the proton−hydride
exchange process were calculated that are similar in energy. BF
is a bifurcated dihydrogen-bonded complex with a H···H distance
(NH···HRu) around 1.94 Å, whereas in the dihydrogen-bonded
FS1 the NH···HRu bond is not bifurcated, a fact that allows a
smaller H···H distance of 1.78 Å. The nonbifurcated species, with
a structure closer to a molecule with coordinated H2, is better in
terms of explaining the excess hydride relaxation and the proton−
hydride exchange. The stated H···H distance for this interaction in
FS1 is relatively large and compares well with that evaluated for
other weak dihydrogen bonds such as Nb−H···HOMe29 or
ReH···HOH.30 The stabilization due to the formation of the
dihydrogen bond is small compared to that observed between
alcohols and Nb29 or Ru31 hydrides. Given that the transformation
between these three species essentially involves the rotation of the
P−imidazolium bond, especially high barriers are not expected for
the mutual interconversion.
(ii) the observed release of hydrogen at room temperature
(iii) the anomalous evolution of T1 with temperature for 3b
and the low T1 values for 3d, which fall in the range
expected for nonclassical hydride−dihydrogen compounds
The practically unaffected chemical shift of the hydride
resonance in 3b,d with respect to the corresponding dihydride
precursors can be due to the similarity in the chemical shift
between the involved species in the chemical exchange. In fact,
2a
the [Ru(arene)H3P]+
and [Ru(arene)H2P] species28 re-
ported until now exhibit very comparable chemical shifts.
The observations outlined above have several implications,
and a mechanism for the whole process is proposed in Chart 1.
Chart 1
(1) At low temperature, step i involves the transfer of the
acidic proton to the basic N atom. This proton is likely to
be interchanged between the N atom and the BF4− anion
(formation of a hydrogen bond, HB).
(2) The second transfer step, step ii, involves proton−
hydride interactions that allow chemical exchange between
the acidic proton and both hydride groups. This exchange
can be explained through the participation of dihydrogen-
bonded intermediates (DHB) in equilibrium with
nonclassical hydride−dihydrogen species (HDH) (step iii).
Both DHB and HDH are consistent with the observed
decrease in the T1 values. In the case of the DHB inter-
mediate, a bifurcated dihydrogen-bonded structure, with
the proton interacting with both hydride ligands, can also
be envisaged.
(3) Finally the release of dihydrogen (step iv) yields com-
pounds 4b,d.
This multistep mechanism has been previously exemplified
by us18c and Jimenez-Tenorio21 in systems with similar phos-
phines coordinated to Ru. This mechanism is in contrast with
the previously reported behavior of ruthenium arene trihydride
derivatives of general formula [RuH3(arene)(phosphine)]+,
which behave as classical hydride compounds and are stable at
room temperature (hydrogen loss was not observed).2a,16
Addition of HBF4 to 2b or 2d in a 2:1 Molar Ratio. We
added an excess of HBF4 to 2b or 2d (2:1 molar ratio). An
excess of acid could lead to the possibility of stabilization of
hydride−dihydrogen derivatives, since the shift of the coordi-
nated H2 molecule by the N-donor center could be hampered
by its protonation.
Protonation with HBF4 in a 2:1 molar ratio at very low
temperature (−80 to −70 °C) gave results similar to those
obtained with a 1:1 ratio (See Figure S10 in the Supporting
Information). However, for mid-range temperatures (−60 to
−40 °C), resonances due to the protonated species 3b,d
disappeared and several hydride doublet signals in the region
7−8.5 ppm were observed in an irreversible process. Finally, at
room temperature a doublet (different from that observed for
the monohydrides 4b,d) was obtained: −7.18 ppm (JHP = 54.0
Hz) for 2b and −7.28 ppm (JHP = 51.3 Hz) for 2d. The
difference in the chemical shift with respect to the hydride of
4b,d, in conjunction with similarities in the coupling constant,
Another minimum, FS2, was found for the proton transfer to
the hydride atom. FS2 is a hydride−dihydrogen compound
with a short H−H bond distance of 0.919 Å and with the
N-imidazolyl atom pointing toward the transferred proton at a
3091
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
a
Chart 2. Reaction Coordinate of the Proton Transfer Process from 3b to 4b
a
Electronic (EE) and Gibbs energies (GE) in the solvent, indicated in parentheses, are in kcal mol−1 (EE/GE). The EE scale is represented in the
chart. For a GE scale see Figure S2 in the Supporting Information.
a
Chart 3. (a) Fundamental and Transition States for the Proton Transfer Process in 3b−d . and (b) Comparison of the Proton
b
−
Transfer Process for Compound 3b with (Blue) and without (Black) the Counteranion BF4
a
b
Relative free energy values in the solvent (kcal mol−1) are in parentheses for 3b,d and in brackets for 3c. Relative free energy values in the solvent
(kcal mol−1) are given in parentheses.
distance of 2.027 Å. As a consequence, the H2 ligand could be
considered as polarized by this atom. Accordingly, the
calculated Mulliken charge of the transferred proton is 0.139:
i.e., higher than the value for the other H of the H2 molecule
(0.095) (Table S2, Supporting Information). We also located
the transition state for the dihydrogen-bonded FS1 to the
hydride−dihydrogen FS2 transformation (TS12), and the latter
is 12.2 kcal mol−1 higher than FS1 (ΔGacetone values). This value
is very low compared with those found for other proton
transfer reactions on transition-metal hydrides.19,20 In our
opinion this low barrier reflects the ease of this process when
the proton is previously located on a basic center close to the
coordination sphere of the complex.
To compare the experimental differences observed between
complexes 3b and 3d in the accessibility to nonclassical species,
the calculated energies in the proton transfer process are
compared for both complexes (see Tables S1 and S3,
Supporting Information). The structural parameters of the
species involved are essentially the same for the two derivatives,
but significant energy differences were found. A relatively lower
barrier for the transition state and a lower energy for the
hydride−dihydrogen FS2 minima were found for the PPh2py
derivative 3d (see Chart 3a). This finding is consistent with the
lower values of T1,min obtained for the derivative 3d compared
to those for 3b and thus supports an easier access and relatively
higher stability of the nonclassical species, FS2, in the case of
3092
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
the PPh2py derivative. Calculations were also carried out on 3c.
The results are similar to those obtained for 3b (see energy
values in brackets in Chart 3a).
It is noteworthy that the process shown in Chart 3a is the key
for the proton−hydride exchange and that the relative energies
of the species involved confirm the reversibility of the process
for these derivatives.
due to the more favorable back-donation from the Ru orbitals
in this orientation (coplanar with the hydride) and to a possible
interaction with the adjacent hydride (“cis effect”).
The aforementioned cis effect34 must lead to a weakening of
the H2 bond and to the consequent formation of a nascent
bond between the hydride and the central hydrogen of the H2
molecule. Covalent34 and an electrostatic35 mechanisms have
been proposed to explain this effect. Both of these views
support the proposal of Crabtree et al., in which an acid di-
hydrogen ligand transfers a proton to an adjacent hydride
base.36 On considering the reaction coordinate of this process
from the hydride-dihydrogen complexes FS3 to FS5, we found
FS4 as an intermediate and two transition states, TS34 and
TS45. The H−H distances in the intermediate FS4 are long
and are both very similar, 1.461 and 1.472 Å, whereas the
shorter H−H distances in the respective TSs are 1.269 Å
(TS34) and 1.279 Å (TS45). The H−H distances in this
reaction coordinate provide conclusive evidence that FS4 is a
classical trihydride structure that possibly arises as a consequence
of the elongation of the H2 bond due to the cis effect.
Chart 3b is useful for comparing the effect of the
−
counteranion BF4 on the stabilization of the different
structures in the proton transfer process (see Table S4 in the
Supporting Information for energies and structural parameters).
−
The introduction of the BF4 counteranion in structure FS1
found for 3b involves rupture of the dihydrogen-bonding
interaction in favor of proton capture by the imidazolyl moiety
(FS1-BF4), which maintains in this structure a conventional
hydrogen bond with the counteranion. The second minimum
in this process is FS2-BF4. This is a hydride−dihydrogen
compound with a weak interaction of one of the hydrogen
atoms of the coordinated H2 molecule with a F atom of the
anion at 1.99 Å, which leads to the formation of a polarized
H−H···FBF3 fragment. The H−H bond distance in the
coordinated H2 molecule decreases in comparison with that
in the same structure without the counteranion (0.90 Å in FS2-
BF4 and 0.92 in FS2). Reinforcement of the bonding in
coordinated H2 by the effect of the counteranion has been
observed previously in other systems.24,32 A transition state
(TS12-BF4) was found between the two minima. In this struc-
ture the counteranion maintains a weak F3BF···H interaction at
2.50 Å with the central hydrogen, which forms part of a
coordinated H2 molecule that interacts in turn with the N-
donor center of the imidazolyl moiety at 2.01 Å. In TS12-BF4
the H−H bond distance is 0.90 Å: i.e., markedly shorter than in
TS12 (0.99 Å). As far as the relative energies of these structures
are concerned, and considering FS1 and FS1-BF4 as reference
points, the activation energy of the proton transfer process
Evolution of the bonding pattern along the whole process
can be easily tracked out by analyzing the electronic density
surfaces in the Ru−H3 plane and using the atoms in molecules
(AIM) formalism (see Figure S5 in the Supporting
Information). The presence of H−H bond critical points
indicates that FS3 and FS5 are hydride−dihydrogen structures,
whereas FS4 is a classical trihydride complex. The charge at the
bond critical points also shows that the Ru−H2 bond is weaker
than the Ru−H bond (as it is usually seen) and that the H−H
bond in FS3 is stronger (by ca. 0.03 au) than in FS5.
On the other hand, the three fundamental states in this
process, of classical to nonclassical trihydride interconversion,
fall within 0.2 kcal mol−1 of each other in the free energy scale
and the energy profile from the classical trihydride intermediate
to the hydride−dihydrogen forms is practically flat. In all five
calculated structures of this reaction coordinate (from FS3 to
FS5) the N-imidazolyl atoms point to the central hydrogen
atom, indicating that this atom is the most electrophilic center
in this path. This situation was further confirmed by analyzing
the Mulliken charges of the hydrogen atoms bonded to the Ru
center in the stated structures (see Table S2 in the Supporting
Information). The relatively positive charge of the central
hydrogen atom in these structures favors the mechanism driven
by the cis effect on the basis of an electrostatic mechanism.35 In
relation with this statement, a relevant fact is that the elec-
trophilic character of this central hydrogen atom is induced by
the uncoordinated N center of the imidazolyl fragment.
Evidence for this is provided by a comparison of the relative
Mulliken charges on the H2 molecule atoms between the
RuH(H2) structures FS2 and FS3. Although both structures are
very similar, the Mulliken charge is more positive for the
hydrogen that interacts with the N center (Table S2) and, as
stated previously, its position changes with the H2 rotation.
At this point it is interesting to discuss our results in light of
the phenomenon known as quantum mechanical exchange37,
because the cationic trihydrides [RuH3(arene)L]+ are thought
to exhibit this phenomenon. Two mechanisms have been pro-
posed, depending on the way in which the exchange coupling
occurs. The first mechanism is based on the accessibility to
vibrational states in the bending modes of the MH2 unit in
which the coupling exchange is observed.38 An alternative
mechanism, proposed by Limbach and Chaudret, involves the
easy thermal accessibility of a hydride−dihydrogen state, MH(H2),
−
increases 7.1 kcal mol−1 due to the effect of the BF4 counter-
anion, whereas FS2-BF4 is destabilized by 4.3 kcal mol−1 due to
this effect. Although the energy of other critical points in Chart 2
could also be affected by the counteranion, we have not
extended our studies to other structures, because the step
analyzed is the one with the highest energy barrier in the global
́
process of proton transfer. For instance, Lledos and Basallote
have recently demonstrated the influence of weak interactions
of the BF4 anion in the rotation energy of H2.33
−
The experimentally observed proton−hydride interchange
must involve rotation of the H2 molecule. This takes place in
Chart 2 with the transformation of FS2 into FS3, also a
hydride−dihydrogen complex. The corresponding transition
state (TS23) was found and has a free energy barrier of 2.7 kcal
mol−1 from FS2. It is noteworthy that the N-imidazolyl atom
participates in this rotation as a polarizing center of the H2
molecule. In fact, it can be seen that this basic center is always
interacting with the transferred proton, shown in red in Chart 2,
and this is located on the right in FS2 and at the center in TS23
and FS3. The calculated Mulliken charges of the H2 atoms
reflect the effect of this polarization, and the most positively
charged hydrogen is the one that interacts with the N center
(Table S2, Supporting Information).
The H−H bond distance in the H2 molecule is shorter in
TS23 (0.854 Å) than in both minima (0.919 Å). Considering
that the N···H distance in TS23 is shorter (1.847 Å) than that
in FS2 (2.027 Å) or FS3 (2.390 Å), it can be concluded that the
H2 molecule is elongated in the equilibrium structuresprobably
3093
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
from a trihydride fundamental state, MH3.39 The same authors
also proposed that when the energy barrier between these
tautomers is very low, a classical exchange process (incoherent
exchange) involving the scrambling of all three hydrides occurs.
For many complexes that exhibit this phenomenon, the
accessibility to nonclassical species has been theoretically
shown to be the main parameter that governs the magnitude
of the exchange coupling.40 In our case, the very low barrier
calculated supports the experimental impossibility of slowing
this classical exchange down sufficiently to differentiate
hydrides in an AB2 system. This prevents the experimental
determination of quantum exchange constants. Heinekey
et al.16 demonstrated that some complexes of the type
[OsH3(arene)(PR3)]+ are examples of classical trihydrides
that show quantum exchange and they proposed that the
analogous Ru trihydrides probably have the same property,
although the low barrier for the classical exchange makes it
impossible to prove this proposal. The results of our theoretical
studies support the existence of a low energy barrier and the
existence of the mechanism proposed by Limbach and
Chaudret as the basis for the quantum exchange. However, in
contrast to the findings for other similar systems, the
Ru(arene)trihydride systems described here exhibit a non-
classical nature in the most stable structure, although the
accessibility to classical species is easy.
It is significant that the shortest N···H distances in the
structures of minimum energy (FS2, FS3, FS4, and FS5) and
their TS intermediates are shorter than the sum of the van der
Waals radii (2.75 Å),41 which suggests the existence to some
extent of an interaction. This situation increases the positive
charge on the affected H atom and could favor the cis-effect
mechanism and also reinforce the H−H bond in the
coordinated dihydrogen molecule. In order to evaluate this
hypothesis, the analogous structures for the classical to
nonclassical trihydride interconversion (TS3 to TS5) were
calculated for [RuH3(p-cymene)(PPh3)]+ (3a). The relative
electronic energy values, considering the classical RuH3
structures as a reference point, are shown in Chart 4. In this
is that the [Ru(arene)H3(P)]+ species reported until now might
be reformulated as dual classical−nonclassical species in nature.
Although we do this with caution, due to the small energy
differences between all these structures, the following
conclusions can be drawn from Chart 4.
(1) The energy differences between classical and nonclassical
species are not large for either 3a or 3b. However, there
is a clear distinction between the two derivatives: the
nonclassical species (FS3 and FS5) are the most stable
ones for 3b, whereas for 3a the classical structure (FS4)
is the lowest in energy.
(2) The H−H bond distances in H2 (for nonclassical) and in
contiguous hydride atoms (for classical) in the different
species of 3b are shorter than those in 3a.
(3) The transition states, starting from the classical structure,
are more accessible for 3b than for 3a.
As expected, all of these conclusions point to the polariza-
tion of the Ru−H bonds by the N-donor centers in the 3b
structures, with a consequent increase in the positive charge
(see Mulliken charges in the Supporting Information). This fact
increases the H2 bond strength and, as a consequence, stabilizes
the nonclassical with respect to the classical intermediates.
The final step in Chart 2 involves the release of H2 from FS5,
and this leads, through an exothermic process in the free energy
scale of 4.7 kcal mol−1, to the final product, 4b.
In conclusion, the presence of an uncoordinated basic N-
donor center close to the coordination sphere is important in
the proton transfer process, as it stabilizes the nonclassical
species, thus facilitating the proton−hydride exchange. The
stabilization effect of the N-donor center on the nonclassical
species is a rarely documented phenomenon, but the effects of
weak bases as the counteranions on the kinetics of the
deprotonation of H2 in trans-[FeH(H2)(dppe)2]+ 42 and on the
relative stabilization of the nonclassical species in a cis−trans
43
isomerization of [W(CO)Cp*H2(PMe3)2]BF4 have been
reported. The effects on the H···H distances of stretched
dihydrogen derivatives by the formation of IrCl···HIr
interactions are also noteworthy.32
H/D Exchange and Deuterium Labeling of H2 with
2a,b,d. Considering the easy protonation of the dihydride deri-
vatives, the possible reversibility of the proton transfer process
and the participation of hydride−dihydrogen intermediates,
which could allow a hydride scrambling in the coordination
sphere of the metal, we initially envisaged the study of proton−
deuterium exchange in the dihydride derivatives using
methanol-d4 as the deuterium source. In addition, it has been
reported that the existence of uncoordinated N-donor centers
facilitates the H/D exchange.44
Chart 4. Relative Energies in the Solvent (kcal mol−1) for the
Classical to Nonclassical Trihydride Interconversion for
3a,b, Considering the Classical Trihydride Structure (FS4)
a
as a Reference Point
Solutions of 2b,d in deuterated methanol were prepared, and
1
their evolution with time was monitored by H and 31P{1H}
NMR spectroscopy. A similar study was carried out on 2a with
the aim of ascertaining whether the N-donor centers were
necessary for this process to occur. The following observations
were made:
a
See Chart 2 for the structures and the Supporting Information for
(i) In all cases a decrease in the intensity of the hydride
doublets was observed, a fact that reflects H/D exchange
(see for instance Figures S8 and S9 for 2a in the Supporting
Information). Complete deuteration of the hydride groups
was achieved for all three compounds after different
reaction times and under different conditions.
selected structural parameters.
case ΔE and not Gibbs energies (ΔG) have been represented,
since the necessary energy correction at the zero point (ZPE)
would be, for some of the structures, higher in value that the
differences in ΔG between the fundamental and transition
states. A noteworthy conclusion we can extract from this study
(ii) The H/D exchange was slow for 2b and faster for 2a,d,
but in all cases the data fit a pseudo-first-order equation
3094
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
best. The initial rate constants were determined at 20 °C
Article
nature has solved with hydrogenases,1b,46 is currently only
available for relatively few systems.47 Many of these applications
give very low yields even under extreme conditions of tem-
perature and pressure.
for 2a,b and at −20 °C in the case of 2d because the
process was too fast in the latter case at the higher
temperature. kobs(2a,20 °C) = 7 × 10−4 s−1, kobs(2b,20 °C) =
4 × 10−5 s−1, kobs(2d,−20 °C) = 1 × 10−4 s−1.
Recently, Sajiki used M/C heterogeneous catalysts (M = Pd,
Pt, Rh, Ir) that were efficient in the deuteration of H2 gas at
room temperature.48 Himeda49 also recently reported the
activity of an Ir(III) dicationic catalyst with bpy (2,2-bipyridine)
type ligands. This catalyst is active in the deuterogenation of C
C and CO bonds using D2O as the deuterium source and H2 or
HCO2H as hydrogenation vectors. The cationic monohydride of
formula [Cp*Ir(H)(bipy)]+ is considered to be the active species.
In an effort to gain more information about possible mech-
anisms for the H/D exchange and the deuterium labeling of H2,
new calculations (DFT, B3LYP) were carried out, and the
conclusions are graphically represented in Chart 5.
(iii) The other signals in the spectra remained unchanged.
Thus, the exchange process only affects the hydride atoms.
(iv) Concerning the deuterium source, only the signal of the
residual OH group of the solvent increased in intensity in
these experiments.
(v) RuHD isotopomers were detected in the 1H and 31P{1H}
NMR spectra, and the resonance for the RuD2 isotopomers
was also seen in the corresponding 31P{1H} NMR spectra
(see Figure 3 for 2b).
Chart 5. Reaction Coordinate for the H/D Exchange (Red
Dashed Lines) and the Deuterium Labeling of H2 (Blue
a
Dotted Lines) from 2b in Methanol-d4
1
Figure 3. (a) H NMR spectra of 2b in the hydride region at 20 °C:
(i) after 100 min; (ii) after 48 h. (b) 31P{1H} NMR spectrum of 2b at
20 °C after 48 h. The resonances were assigned to the different
isotopomers.
Although we initially thought that this H/D exchange
process was N-mediated, the fact that 2a shows the same
phenomenon as 2b, but even faster, suggests that other
alternative mechanisms should be taken into account, at least
for 2a. Mechanistic considerations will be discussed below in
light of theoretical studies.
After these experiments, it was decided to study the possible
labeling of H2 to give D2 using methanol-d4 as a deuterium
source. This proposal is supported by the fact that this process has
been demonstrated to be feasible for other dihydrogen-bonded
complexes that exhibit proton−hydride exchange.45,18a,c In
a
Gibbs energies (GE) in methanol, indicated in parentheses, are in
kcal mol−1.
1
order to obtain the desired information, we used H NMR
Complex 2b has been chosen as starting point for this
calculation. Initially 2b reacts with a molecule of methanol to
give intermediate HB, which has a conventional N···HOMe
hydrogen bond with the N-donor center of the Me-imidazolyl
group. Alternatively, 2b could give DHB1, which has a
dihydrogen bond (RuH···HOMe). Although HB is more stable
than DHB1, the latter is also thermally accessible at 4.4 kcal
mol−1 from the starting point. In the case of 2a (with PPh3) the
species HB could not be formed and 2a would be transformed
into the species analogous to DHB1.
If we concentrate on the H/D exchange (red dashed lines in
Chart 5), the species DHB2, analogous to DHB1 but with
interchange of the H and D positions, should be formed.
Between both species we have found the transition state TS(TT)
that exhibits an elongated and coordinated H2 molecule with a
H−H bond distance of 1.09 Å. The oxygen atom of the MeO−
spectroscopy under conditions similar to those for the previous
H/D experiments to monitor the evolution of the dissolved
hydrogen (after bubbling H2 through a methanol-d4 solution)
using 2a,d as catalysts. The transformation of H2 into D2 was
quantitative after 20 min on using 2d. In this experi-
ment the formation of HD as an intermediate species was
also observed. Compound 2a was even more active, since only
the first spectrum acquired (after approximately 5 min) showed
traces of HD and these disappeared into the baseline after
5 min more. It is noteworthy that this process takes place at a
much higher rate than the H/D exchange, and consequently a
different pathway is expected.
The generation of D2 from protic solvents such as CD3OD
and D2O is a smart and environmentally friendly process that
has been widely pursued in recent years. This process, which
3095
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
anion interacts with both hydrogen atoms of the coordinated
H2 molecule at 1.68 and 1.72 Å. An IRC study showed that
TS(TT) connects with DHB1 and also with DHB2, thus
supporting its role as transition state in the H/D exchange. In
this way a “tick-tock” (TT) mechanism can be proposed for the
experimentally observed proton−hydride exchange in such a
way that the methoxy group could trap either the previously
transferred proton to give DHB1 or the hydride to give DHB2.
This “tick-tock” mechanism has been previously considered for
the H/H exchange in Re hydride derivatives and theoretically
supported by EHT calculations.50 The relatively high barrier for
this transition state, which is 35.4 kcal mol−1 from DHB2,
points to a slow chemical transformation which fits with the
slow H/D exchange experimentally observed.
In order to support the experimentally observed deuterium
labeling of H2 (blue dotted line in Chart 5), the structure
HydOMe + H2 was calculated. This fundamental state, model-
ing the loss of H2 from DHB2 or from DHB1, is a monohydride−
methoxy derivative that is 10.5 kcal mol−1 more stable than DHB2
and 5.4 kcal mol−1 more stable than DHB1. Unfortunately, we
were not able to locate a transition state between HydOMe +
H2 and any of the two dihydrogen-bonded derivatives, but the
connection between dihydrogen-bonded complexes and the
corresponding product of the dihydrogen release can be
considered as a reasonable reaction pathway.
Figure 4. Evolution with temperature of the 1H NMR hydride
resonances of 3e and 4e. Resonances labeled with asterisks can be
considered as being due to possible intermediates.
Chart 7. Mechanism Proposed for the Protonation of 2d
with CF3CO2H To Give 3e and Subsequent Evolution to 4e,
upon Increasing the Temperature, with Concomitant H2
Loss
In accord with the reaction coordinate discussed in the pre-
vious paragraphs, two related catalytic cycles can be proposed
for the H/D exchange and the deuterium labeling of H2 (Chart 6,
Chart 6. Proposed Catalytic Cycles for the H/D Exchange
(Red, Left) or Deuterium Labeling of H2 (Blue, Right)
(Chart 7, step i). The 31P{1H} NMR spectrum at this tem-
perature showed a single singlet at 80.08 ppm, whereas the
most salient spectroscopic features in the 1H NMR spectrum of
this compound are the low-field broad resonance at 20.40 ppm,
which integrates for one proton and is assigned to the N−H
group, and the high-field broad signal at −11.08 ppm
(integration for 2H), which corresponds to the hydride atoms.
It is noteworthy that the position of the resonances of 3e is not
the same as in 3d and there is a marked difference in the N−H
signal (of around 8 ppm). This finding points to an interaction
between the anion and the N−H group. In fact, the high
chemical shift of the N−H group suggests the existence of a
bifurcated dihydrogen bond, where the second basic center, in
addition to the N atom, would be the CF3CO2− anion. A similar
dihydrogen-bonded compound has been previously identified
by our group.18a Considering that both the N−H and hydride
signals are broad even at −90 °C, it is proposed that the
hydride−proton interaction is dynamic and involves both
hydride groups (Chart 7, step ii).
red and blue cycles, respectively). In the first step of these catalytic
cycles, from [Ru]-H, the participation of an intermediate of type
HB (see Chart 5) is reasonable in the case of compounds
bearing nitrogenated phosphines.
Addition of CF3CO2H to 2d. Finally, in order to assess
whether the previous results are general or depend on the
properties of the acid employed, we analyzed the effect of the
addition of trifluoroacetic acid to compound 2d. CF3CO2H is a
weaker acid than HBF4, but its anion has considerably higher
coordination ability.
The proton transfer process that takes place after the addi-
tion, in an NMR tube, of an equimolar amount of CF3CO2H to
a solution of 2d in acetone-d6, precooled to −90 °C, was
1
monitored by H and 31P{1H} NMR spectroscopy. The tem-
perature was slowly increased from −90 to +20 °C (Figure 4).
Initially, at −90 °C the proton acceptor is the basic nitrogen
atom in the pyridine moiety, and this gives rise to the cationic
derivative 3e, [RuH2(η6-p-cymene)(κ1P-PPh2pyH)][CF3CO2]
When the temperature is increased, both signals undergo
significant broadening, a change that is consistent with the ex-
change process between the three hydrogen atoms. The inter-
change between these atoms was confirmed by a magnetization
3096
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
transfer observed between the two resonances. Even though it
has not been observed, a nonclassical dihydrogen intermediate
is proposed to take part in such a phenomenon (3e′ in Chart 7,
step iii). At −80 °C the loss of dihydrogen starts, a fact that
implies a stability difference between 3e and 3d, since this loss
was not observed until room temperature in the case of 3d.
the complexes with PPh2py or PPh3. It was found that these
derivatives were extremely active in H2 labeling processes to
give D2 at room temperature (quantitative transformation in
several minutes), with the OD of methanol-d4 as the deuterium
source. Mechanistic proposals for this transformation have been
presented, and they are theoretically supported by a reaction
coordinate that includes a “tick-tock” mechanism for the H/D
exchange and the reversible loss of H2. The presence of the
uncoordinated nitrogen atom is not necessary for these catalytic
processes to occur.
When an acid containing a counteranion with a higher
coordinating ability (such as CF3CO2H) is used, the initial
protonation of the N atom and the proton−hydride exchange
also take place. In this case, however, a lower stability for the
dihydrogen species was found, possibly because the H2
molecule is more easily replaced by this anion.
1
From this temperature on, new signals in both the H and the
31P{1H} NMR spectra appeared and these were assigned to the
new monohydride 4e, [RuH(η6-p-cymene)(κ2-P,N-PPh2py)]-
[CF3CO2], the spectroscopic data for which are nearly identical
with those of 4d.
We propose that the higher coordinating ability of the
−
CF3CO2− anion, compared to that of BF4 , is the reason for the
lower stability of the dihydrogen species in this system com-
pared to that of 3d. The anion could displace more easily the
coordinated H2 molecule.51 As a matter of fact, additional peaks
were found during the evolution with temperature in both the
31P{1H} NMR spectrum (at 20 °C: 56.90 (s) and 62.17 (s) ppm)
EXPERIMENTAL SECTION
■
1
General Comments. All reactions were carried out under nitrogen
using standard Schlenk techniques. Solvents were freshly distilled from
convenient drying agents and degassed under nitrogen prior to use.
dpim,52 PPh2py,53 [RuCl2(η6-arene)(κ1P-dpim)]22 (arene = p-cymene
(1b), benzene (1c)), and [RuCl2(η6-p-cymene)(κ1P-PPh2py)] (1d)54
were prepared according to previous procedures. [RuH2(η6-p-cymene)-
(κ1-P-PPh3)] (2a) was prepared by a procedure similar to that described
for 2b−d. RuCl3·xH2O, PPh3, HBF4, and CF3CO2H were purchased
from Aldrich.
and in the high-field region of the H NMR spectrum (at 20 °C:
−6.54 ppm (d, 2JHP = 52.4 Hz) and −6.88 ppm (d, 2JHP = 52.4 Hz)).
These resonances could correspond to two additional mono-
hydrides. One proposal is 3e″ (see Chart 7, step iv), although
the solvent (acetone-d6) could play a similar role in the stabi-
lization of a monohydride species. These signals finally dis-
appear when the sample is kept at room temperature for 16 h,
leaving 4e as the only detectable species.
NMR spectra were recorded on a Varian Unity Inova-400 (400
MHz for 1H; 161.9 MHz for 31P; 100.6 MHz for 13C), a Varian
UNITY-300 (300 MHz for 1H; 121.4 MHz for 31P; 75.4 MHz for 13C)
CONCLUSIONS
■
1
or a Varian Unity Inova-500 spectrometer (500 MHz for H; 202,4
In this work the protonation processes of arene Ru derivatives
of the type [RuH2(arene)(PPh2Het)] (arene = p-cymene,
benzene; Het = 2-pyridyl (PPh2py), 2-N-methylimidazolyl
(dpim)) containing an uncoordinated N atom were studied both
by experiments and with the help of theoretical calculations. The
behavior of these compounds was compared with that of the
derivatives [RuH2(arene)(PR3)].
MHz for 31P; 125.8 MHz for 13C). Chemical shift values are reported
1
in ppm. H chemical shifts were recorded using the residual proton
of the solvent as internal standard. Coupling constants (J) are in
hertz. See Chart 8 for atom numbering. All 31P chemical shifts were
Chart 8. Atom Numbering for the p-Cymene Ring and
Imidazolyl and Pyridyl Heterocycles
In the case of the protonation with HBF4, at low temperature
the N atoms are protonated first and a proton−hydride
exchange is present. At room temperature the formation of
monohydride species is observed. It was concluded that the
species [RuH3(arene)(PPh2Het)]BF4, which are formed as
intermediates after the proton transfer, have a dual character as
classical and nonclassical hydrides, with the nonclassical species
being more stable, especially when Het = 2-pyridyl. This
conclusion was based on the following facts: (i) the easy release
of H2 at room temperature with the concomitant coordination
of the nitrogen atom; (ii) the low T1 values measured; (iii) the
results of calculations, which show that the nonclassical hydrides
can access the slightly less stable classical species through ener-
getically accessible transition states. Calculations performed for
the species containing PPh3 also reflect the dual character, and
in our opinion the nature of this trihydride species might be
reformulated as a consequence.
referenced internally. For the variable-temperature spectra the
temperature of the probe ( 0.1 K) was controlled by a standard
unit calibrated with a methanol reference. Longitudinal relaxation time
(T1) measurements were made by the inversion-recovery method. In
the NMR data, s, d, t, b, spt, o, m, and p refer to singlet, doublet,
triplet, broad, septet, ortho, meta, and para, respectively. IR spectra
were recorded on a Nicolet Impact 410 spectrophotometer as KBr
pellets or on a Perkin-Elmer 883 (4000−200 cm−1 range) as Nujol
mulls deposited on a polyethylene film.
Ab initio calculations were performed using the Gaussian 03
revision C.02 series of programs.55 Density functional theory (DFT)
was applied with the three-parameter hybrid functional of Becke and
the Lee, Yang, and Parr correlation functional, more widely known as
B3LYP.56 The bulk effect of the solvent has been introduced through
the PCM57,58 continuum model by single-point calculations on gas-
phase optimized geometries with default parameters for acetone as the
solvent (ε = 20.7) to define the cavity in this model. The Gsolv values
The role played by the uncoordinated N center is particularly
relevant, as it interacts in all of the calculated structures with the
transferred hydrogen atom. More importantly, in the RuH-
(H2)+ and the RuH3 intermediates, this interaction makes the
+
central hydrogen atom more electrophilic and thus favors the
stabilization of the nonclassical species and the hydride−
dihydrogen exchange.
The derivatives [RuH2(p-cymene)(L)] (L = PPh2py, dpim,
PPh3) were active catalysts for the H/D exchange of the
hydride ligands with methanol-d4. The process was faster for
3097
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
account for the total free energy in solution with all non-electrostatic
terms and the thermal correction to the Gibbs free energy in gas phase.
For more computational details see the Supporting Information.
Synthesis of Complexes. Preparation of [RuH2(p-cymene)-
(κ1P-dpim)] (2b). [RuCl2(p-cymene)(dpim)] (1b; 109.5 mg,
0.19 mmol) was dissolved in ethanol (35 mL) to give an orange
solution. KBH4 (103 mg, 1.9 mmol) (molar ratio 1:10) was added,
and a vacuum was applied as hydrogen was released during the
reaction. The mixture became darker in a few minutes and was
stirred at room temperature for 7 h. The resulting suspension was
filtered to remove the additional KBH4. A dark brown solution was
obtained, and the solvent was evaporated under vacuum to afford
a solid, which was extracted with n-pentane (3 × 15 mL). The
liquid phase was filtered. The solvent of the resulting brown
solution was evaporated under vacuum to dryness to give
compound 2b. The product is soluble in n-pentane, acetone,
and ethanol. Yield: 28.7 mg (0.06 mmol), 35%. Anal. Calcd for
C26H31N2PRu: C, 61.99; H, 6.16; N, 5.56. Found: C, 62.23; H,
6.13; N, 5.7. 1H NMR (400 MHz, acetone-d6, room
temperature): 7.86−7.76 (m, 4H, Ho-Ph); 7.36−7.24 (m, 6H,
Hm,p-Ph); 7.15 (m, 1H, H5′-Im); 6.95 (m, 1H, H4′-Im); 5.28 (d,
JHH = 5.8 Hz, 2H, H3,5-cym); 5.14 (d, JHH = 5.2 Hz, 2H, H2,6-
4C, Cm-Ph); 122.73 (s, 1C, py); 109.98 (s, 1C, C4-cym); 101.5 (s, 1C, C1-
cym); 87.24 (d, JPC = 3.5 Hz, 2C, C3,5-cym); 82.79 (d, JPC = 4.0 Hz, 2C,
C2,6-cym); 31.05 (s, 1C, C7-cym); 22.00 (s, 2C, C8,9-cym); 19.50
(s, 1C, C10-cym) ppm. 31P{1H} NMR (121.4 MHz, acetone-d6, room
temperature): 71.0 (s) ppm.
Protonation Experiments: Formation of Trihydrides and
Monohydrides. Protonation experiments were carried out in 5 mm
NMR tubes. Samples were prepared under nitrogen. In a typical
experiment, 15−20 mg of dihydride 2b was dissolved in 0.5 mL of
CD3COCD3 and the tube was cooled to 183 K. HBF4·OEt2 or
CF3CO2H (1 equiv) was added using a microsyringe. The tube was
shaken to mix all the reactants and then was immediately introduced
1
into the probe at 183 K. H and 31P NMR spectra as well as T1
measurements were carried out at this temperature and then at
intervals of 20 K up to 298 K.
Deuteration Experiments. Deuteration experiments were carried
out in 5 mm NMR tubes. Samples were prepared under nitrogen. In a
typical experiment, 15−20 mg of dihydride 2b was dissolved in 0.5 mL
1
of CD3OD, and the evolution of the reaction was monitored by H
NMR spectroscopy over a range of times. For the experiments with
H2, the gas was bubbled through the deuterated solution for 2 min and
the NMR tube was then sealed with a Young valve under an H2
atmosphere. The NMR spectra were recorded.
3
cym); 3.51 (s, 3H, N-Me); 2.37 (spt, JHH = 6.9 Hz, 1H, H7-
3
cym); 1.65 (s, 3H, H10-cym); 1.18 (d, JHH = 6.9 Hz, 6H, H8,9-
2
cym); −10.62 (d, JPH = 43.5 Hz, 2H, RuH2) ppm. 13C{1H}
NMR (100.577 MHz, acetone-d6, room temperature): 147.9 (d,
ASSOCIATED CONTENT
* Supporting Information
■
S
1JPC = 68.4 Hz, 1C, C2′-Im); 140.1 (d, 1JPC = 49.2 Hz, 2C, Cipso
-
Text, figures, and tables giving energies and selection of
structural parameters for structures deduced from calculations,
Mulliken charges calculated for some of these structures, and
information on labeling processes. This material is available free
of charge via the Internet at http://pubs.acs.org.
2
4
Ph); 133.98 (d, JPC = 12.7 Hz, 4C, Co-Ph); 129.13 (d, JPC
=
3
2.2 Hz, 2C, Cp-Ph); 127.6 (d, JPC = 10.5 Hz, 1C, C4′-Im);
127.3 (d, 3JPC = 9.8 Hz, 4C, Cm-Ph); 124.4 (s, 1C, C5′-Im); 112
(d, JPC = 5.3 Hz, 1C, C4-cym); 101.5 (s, 1C, C1-cym); 87.5 (d,
JPC = 1.6 Hz, 2C, C3,5-cym); 82.9 (d, JPC = 3.5 Hz, 2C, C2,6-
cym); 34.8 (d, JPC = 2.3 Hz, 1C, N-Me); 32.3 (s, 1C, C7-cym);
3
31
24.1 (s, 2C, C8,9-cym); 5.1 (s, 1C, C10-cym) ppm. P{1H} NMR
AUTHOR INFORMATION
Corresponding Author
*E-mail: Felix.Jalon@uclm.es. Fax: (+34) 926295318. Tel:
(+34) 926295300.
■
(161.923 MHz, acetone-d6, room temperature): 52.1 (s) ppm.
Preparation of [RuH2(benzene)(κ1P-dpim)] (2c). This compound
was prepared as described for 2b using 82.5 mg (0.16 mmol) of
[RuCl2(benzene)dpim] (1c) and 86.2 mg (1.6 mmol) of KBH4 (molar
ratio 1:10). Yield: 10.7 mg (0.024 mmol), 15%. Pm (C22H23N2PRu):
447.26 g/mol. Analytical data for C22H23N2PRu are not available. The
compound is very air-sensitive. 1H NMR (400 MHz, acetone-d6, room
temperature): 8−7.1 (m, 12H, Ph, H5′,4′-Im); 5.21 (d, JPH = 0.9 Hz,
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
2
■
6H, C6H6); 3.5 (s, 3H, N−Me); −10.35 (d, JPH = 40.6 Hz, 2H,
RuH2) ppm. 13C{1H} NMR: not available, compound hardly soluble.
31P{1H} NMR (161.923 MHz, acetone-d6, room temperature):
51.1 (s) ppm.
This work was supported by the MICINN of Spain (projects
CTQ2011-24434 and CTQ2008-02403/BQU), the Junta de
Comunidades de Castilla-La Mancha-FEDER Funds (PEII11-
0214), and “Generalitat de Catalunya” (project 2009SGR409).
Preparation of [RuH2(p-cymene)(κ1-P-PPh2py)] (2d). Method 1.
[RuCl2(p-cymene)(PPh2py)] (1d; 125 mg, 0.22 mmol) was
dissolved in THF (20 mL) and LiAlH4 (84 mg, 2.2 mmol)
(molar ratio 1:10). The solution was stirred for 4 h at room
temperature. The resulting yellow solution was purified by
column chromatography (using Al2O3, 50−200 μm, and
toluene/diethyl ether (1/1) in a 14 cm column). The solvent
of the resulting brown solution was evaporated under vacuum
until dryness and compound 2d was obtained. Yield: 48.4 mg
(0.10 mmol), 45%.
DEDICATION
Dedicated to Francisco Urbanos in memoriam.
■
REFERENCES
(1) (a) Besora, M.; Lledos
■
́
, A.; Maseras, F. Chem. Soc. Rev. 2009, 38,
957−966. (b) Kubas, G. J. Chem. Rev. 2007, 107, 4152−4205.
(c) Papish, E. T.; Magee, M. P.; Norton, J. R. In Recent Advances in
Hydride Chemistry; Peruzzini, M., Poli, R., Eds.; Wiley: Amsterdam,
2001; pp 39−74.
Method 2. This compound was also prepared as described for 2b
using 109.5 mg (0.19 mmol) of [RuCl2(p-cymene)(PPh2py)] (1c) and
103 mg (1.9 mmol) of KBH4 (molar ratio 1:10). Yield: 60.3 mg (0.12
mmol), 65%. Anal. Calcd for C27H30NPRu: C, 64,78; H, 6.04; N, 2.79.
(2) (a) Werner, H.; Kletzin, H. J. Organomet. Chem. 1983, 243, C59−
C62. (b) Heinekey, D. M. J. Am. Chem. Soc. 1991, 113, 6074−6077.
(c) Bergman, R. G.; Gilbert, T. M. J. Am. Chem. Soc. 1985, 107, 3502−
3507. (d) Heinekey, D. M.; Millar, J. M.; Koetzle, T. F.; Payne, N. G.;
Zilm, K. W. J. Am. Chem. Soc. 1990, 112, 909−919.
(3) (a) Crabtree, R. H.; Lavin, M. J. Chem. Soc., Chem. Commun.
1985, 794−795. (b) Morris, R. H.; Sawyer, J. F.; Shiralian, M.;
Zubkowski, J. D. J. Am. Chem. Soc. 1985, 107, 5581−5582.
1
Found: C, 64.60; H, 5.86; N, 2.69. H NMR (300 MHz, acetone-d6,
room temperature): 8.12 (d, JHH = 4.7 Hz, 1H, CH-py); 7.74(t, JHH
=
=
8.0 Hz, 4H, Ho-Ph); 6.81 (t, JHH = 7.3 Hz, 4H, Hm-Ph); 6.73 (t, JHH
9.0 Hz, 2H, Hp-Ph); 6.62 (m, 1H, CH-py); 6.16 (m, 1H, CH-py); 5.24
3
(4H, CH-cym); 2.33 (spt, JHH = 7.0 Hz, 1H, H7-cym); 1.84 (s, 3H,
3
2
H10-cym); 1.21 (d, JHH = 7.0 Hz, 6H, H8,9-cym); −10.61 (d, JPH
=
3.6 Hz, 2H, RuH2) ppm. 13C{1H} NMR (75.4 MHz, acetone-d6, room
temperature): 148.70 (d, J = 12.5 Hz, 1C, py); 136.73 (d, J = 8.5 Hz,
1C, py); 134.05 (d, 2JPC = 11.5 Hz, 4C, Co-Ph); 131.93 (d, J = 4.5 Hz,
1C, py); 128.74 (d, 4JPC = 2.5 Hz, 2C, Cp-Ph); 127.20 (d, 3JPC = 9.5 Hz,
́
, A.; Clot, E.; Eisenstein, O. Chem. Rev. 2000,
(4) Maseras, F.; Lledos
100, 601−636.
(5) Jessop, Ph. G.; Morris, R. H. Coord. Chem. Rev. 1992, 121,
55−284.
3098
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
́
(6) (a) Epstein, L. M.; Belkova, N. V.; Shubina, E. S. In Recent
Advances in Hydride Chemistry; Peruzzini, M., Poli, R., Eds.; Wiley:
Amsterdam, 2001; pp 391−418. (b) Belkova, N. V.; Epstein, L. M.;
Shubina, E. S. Eur. J. Inorg. Chem. 2010, 3555−3565.
(7) Chinn, M. S.; Heinekey, D. M. J. Am. Chem. Soc. 1987, 109,
5865−5867.
Rodrıguez-Fernan
́
dez, A.; Jalon
́
, F.; Trofimenko, S. J. Am. Chem. Soc.
1995, 117, 7441−7451. (c) Chen, Y. Z.; Chan, W. C.; Lau, C. P.; Chu,
H. S.; Lee, H. L.; Jia, G. Organometallics 1997, 16, 1241−1246.
(d) Morris, R. H. Coord. Chem. Rev. 2008, 252, 2381−2394.
(28) Werner, H.; Kletzin, H. J. Organomet. Chem. 1982, 228,
298−300.
(8) Bakhmutov, V. I. Dihydrogen Bonds: Principles, Experiments, And
Applications; Wiley: Amsterdam, 2008.
(29) Bakhmutova, E. V.; Bakhmutov, V. I.; Belkova, N. V.; Besora,
́
M.; Epstein, L. M.; Lledos, A.; Nikonov, G. I.; Shubina, E. S.; Tomas,
(9) Custelcean, R.; Jackson, J. E. Chem. Rev. 2001, 101, 1963−1980
and references therein.
J.; Vorontsov, E. V. Chem. Eur. J. 2004, 10, 661−671.
(30) Orlova, G.; Scheiner, S. J. Phys. Chem. A 1998, 102, 4813−4818.
(10) Epstein, L. M.; Shubina, E. S. Coord. Chem. Rev. 2002, 231,
165−181.
́
(31) Belkova, N. V.; Besora, M.; Epstein, L. M.; Lledos, A.; Maseras,
F.; Shubina, E. S. J. Am. Chem. Soc. 2003, 125, 7715−7725.
(32) Gusev, D. G. J. Am. Chem. Soc. 2004, 126, 14249−14257.
́
(11) (a) Belkova, N. V.; Besora, M.; Epstein, L.; Lledos, A.; Maseras,
F.; Shubina, E. S. J. Am. Chem. Soc. 2003, 125, 7715−7725.
́ ́
(33) Algarra, A. G.; Fernandez-Trujillo, M. J.; Lledos, A.; Basallote,
M. G. Chem. Commun. 2009, 30, 4563−4565.
(b) Belkova, N. V.; Collange, E.; Dub, P.; Epstein, L. M.;
(34) (a) Van Der Sluys, L. S.; Eckert, J.; Eisenstein, O.; Hall, J. H.;
Huffman, J. C.; Jackson, S. A.; Koetzle, T. F.; Kubas, G. J.; Vergamini,
Ph. J.; Caulton, K. J. J. Am. Chem. Soc. 1990, 112, 4831−4841.
(b) Chaudret, B.; Chug, G.; Eisenstein, O.; Jackson, S. A.; Lahoz, F. J.;
Lopez, J. A. J. Am. Chem. Soc. 1991, 113, 2314−2316. (c) Riehl, J.-F.;
Lemenovskii, D. A.; Lledos
Revin, P. O.; Shubina, E. S.; Vorontsov, E. V. Chem. Eur. J. 2005, 11,
873−888. (c) Basallote, M. G.; Besora, M.; Castillo, C. E.; Fernandez-
Trujillo, M. J.; Lledos, A.; Maseras, F.; Manez, M. A. J. Am. Chem. Soc.
́
, A.; Maresca, O.; Maseras, F.; Poli, R.;
́
́
́
̃
2007, 129, 6608−6618. (d) Dub, P. A.; Filippov, O. A.; Belkova, N. V.;
Daran, J.-C.; Epstein, L. M.; Poli, R; Shubina, E. S. Dalton Trans. 2010,
39, 2008−2015.
Pel
́
issier, M.; Eisenstein, O. Inorg. Chem. 1992, 31, 3344−3345.
(35) (a) Maseras, F.; Duran
́
, M.; Lledos, A.; Bertran, J. J. Am. Chem.
́
́
Soc. 1991, 113, 2879−2884. (b) Bianchini, C.; Masi, D.; Peruzzini, M.;
Casarin, M.; Maccato, C.; Rizzi, G. A. Inorg. Chem. 1997, 36, 1061−
1069. (c) Trovitch, R. J.; Lobkovsky, E.; Chirik, P. J. Inorg. Chem.
2006, 45, 7252−7260.
(12) Cf. the seminal work of Kubas: Kubas, G. J. J. Chem Soc., Chem.
Commun. 1980, 61−62 and many other examples in ref 1b.
(13) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.
2004, 204, 2201−2237.
(36) Crabtree, R. H.; Lavin, M.; Bonneviot, L. J. Am. Chem. Soc.
1986, 108, 4032−4037.
(14) Bullock, R. M. Angew. Chem., Int. Ed. 2007, 46, 7360−7363.
(15) Laua, Ch. P.; Nga, S. M.; Jia, G.; Lin, Z. Coord. Chem. Rev. 2007,
251, 2223−2237.
(37) (a) Sabo-Etienne, S.; Chaudret, B. Chem. Rev. 1998, 98, 2077−
2091. (b) McGrady, G. S.; Guilera, G. Chem. Soc. Rev. 2003, 32, 383−
(16) Heinekey, D. M.; Harper, T. G. P. Organometallics 1991, 10,
2891−2895.
392. (c) Maseras, F.; Lledos
2000, 100, 601−636.
́
, A.; Clot, E.; Eisenstein, O. Chem. Rev.
(17) (a) Arliguie, T.; Border, C.; Chaudret, B.; Devillers, J.; Poilblanc,
R. Organometallics 1989, 8, 1308−1314. (b) Limbach, H. H.; Ulrich,
S.; Buntkowsky, G.; Sabo-Etienne, S.; Chaudret, B.; Kubas, G. J.;
Eckert, J. J. Am. Chem. Soc. 1998, 120, 7929−7943. (c) Heinekey,
D. M.; Payne, N. G.; Sofield, C. D. Organometallics 1990, 9, 2643−2645.
(38) (a) Zilm, K. W.; Heinekey, D. M.; Millar, J. M.; Payne, N. G.;
Demou, P. J. Am. Chem. Soc. 1989, 111, 3088−3089. (b) Zilm, K. W.;
Heinekey, D. M.; Millar, J. M.; Payne, N. G.; Neshyba, S. P.;
Duchamps, J. C.; Szczyrba, J. J. Am. Chem. Soc. 1990, 112, 920−929.
(c) Jones, D. H.; Labinger, J. A.; Weitekamp, D. P. J. Am. Chem. Soc.
1989, 111, 3087−3088.
(d) Grundemann, S.; Limbach, H. H.; Rodríguez, V.; Donnadieu, B.;
̈
Sabo-Etienne, S.; Chaudret, B. Ber. Bunsen-Ges. Phys. Chem. 1998, 102,
344−353.
(39) Limbach, H.-H.; Scherer, G.; Maurer, M.; Chaudret, B. Angew.
Chem., Int. Ed. 1992, 31, 1369−1372.
(18) (a) Caballero, A.; Jalon
1998, 1879−1880. (b) Caballero, A.; Carrion
F. A.; Manzano, B. R. Polyhedron 2003, 23, 361−371. (c) Jalon
Manzano, B. R.; Caballero, A.; Carrion, M. C.; Santos, L.; Espino, G.;
Moreno, M. J. Am. Chem. Soc. 2005, 127, 15364−15365.
(19) (a) Rossin, A.; Gonsalvi, L.; Phillips, A. D.; Maresca, O.; Lledos
́
, F. A.; Manzano, B. R. Chem. Commun.
, M. C.; Espino, G.; Jalon
, F. A.;
(40) (a) Barthelat, J. C.; Chaudret, B.; Daudey, J. P.; De Loth, Ph.;
Poilblanc, R. J. Am. Chem. Soc. 1991, 113, 9896−9898. (b) Jarid, A.;
́
́
,
́
Moreno, M.; Lledos
́
, A.; Lluch, J. D.; Bertran, J. J. Am. Chem. Soc.
, A.; Lluch,
J. D.; Bertran, J. J. Am. Chem. Soc. 1995, 117, 1069−1075. (d) Camanyes,
S.; Maseras, F.; Moreno, M.; Lledos, A.; Lluch, J. M.; Bertran, J. J. Am.
Chem. Soc. 1996, 118, 4617−4621.
(41) Bondi, A. J. Phys. Chem. 1964, 68, 441−51.
(42) Basallote, M. G.; Besora, M.; Castillo, C. E.; Fernan
M. J.; Lledos, A.; Maseras, F.; Manez, M. A. J. Am. Chem. Soc. 2007,
129, 6608−6618.
(43) Dub, P. A.; Belkova, N. V.; Filippov, O. A.; Daran, J.-C.; Epstein,
L. M.; Lledos, A.; Shubina, E. S.; Poli, R. Chem. Eur. J. 2010, 16,
189−201.
́
1993, 115, 5861−5862. (c) Jarid, A.; Moreno, M.; Lledos
́
́
,
A.; Peruzzini, M. Organometallics 2007, 26, 3289−3296. (b) Chu,
H. S.; Lau, C. P.; Wong, K. Y.; Wong, W. T. Organometallics 1998, 17,
2768−2777.
́
́
dez-Trujillo,
(20) (a) Epstein, L. M.; Belkova, N. V.; Gutsul, E. I.; Shubina, E. S.
Pol. J. Chem. 2003, 77, 1371−1383. (b) Belkova, N. V.; Epstein, L. M.;
Shubina, E. S. Eur. J. Inorg. Chem. 2010, 3555−3565.
́
́
̃
(21) Jimen
Ujaque, G.; Lledos
(22) Caballero, A.; Jalon
́
ez-Tenorio, M.; Puerta, M. C.; Valerga, P.; Moncho, S.;
, A. Inorg. Chem. 2010, 49, 6035−6057.
́
, F. A.; Manzano, B. R.; Espino, G.; Perez-
́
́
́
(44) Baur, J.; Jacobsen, H.; Burger, P.; Artus, G.; Berke, H.;
Dahlenburg, L. Eur. J. Inorg. Chem. 2000, 1411−1422.
(45) Lough, A. J.; Park, S.; Ramachandran, R.; Morris, R. H. J. Am.
Chem. Soc. 1994, 116, 8356−8357.
(46) (a) Caniguier, S.; Artero, V.; Fontecave, M. Dalton Trans. 2008,
315−325. (b) Mealli, C.; Rauchfuss, T. B. Angew. Chem., Int. Ed. 2007,
46, 8942−8944.
(47) (a) Collman, J. P.; Wagenknecht, P. S.; Hembre, R. T.; Lewis, N. S.
J. Am. Chem. Soc. 1990, 112, 1294−1295. (b) Heinen, A. V.;
Papadogianakis, G.; Sheldon, R. A.; Peters, J. A.; van Bekkum, H. J.
Mol. Catal. A: Chem. 1999, 142, 17−26. (c) Baba, A.; Nishikawa, M.;
Eguchi, T. J. Nucl. Mater. 1997, 250, 29−35. (d) von Hahn, H. E. A.;
Peters, E. J. Phys. Chem. 1971, 75, 571−579. (e) Grundler, P. V.; Yazyev,
O. V.; Aebischer, N.; Helm, L.; Laurenczy, G.; Merbach, A. E. Inorg.
Chim. Acta 2006, 359, 1795−1806. (f) Palibroda, N.; Grecu, E.;
Manrique, M.; Mucientes, A.; Poblete, F. J.; Maestro, M. Organo-
metallics 2004, 23, 5694−5706.
(23) Garrow, P. E. Chem. Rev. 1981, 81, 229−266.
(24) (a) Park, S. H.; Lough, A. J.; Yap, G. P. A.; Morris, R. H.
J. Organomet. Chem. 2000, 609, 110−122. (b) Ayllon, J. A.; Sayers, S.
F.; Sabo-Etienne, S.; Donnadieu, B.; Chaudret, B.; Clot, E.
Organometallics 1999, 18, 3981−3990.
(25) Desrosiers, P. J.; Cai, L.; Richards, R.; Halpern, J. J. Am. Chem.
Soc. 1991, 113, 4173−4184.
(26) Almeida, K.; Kranenburg, M.; Guari, Y.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M.; Sabo-Etienne, S.; Chaudret, B. Inorg. Chem.
2003, 42, 2859−2866.
́
(27) (a) Moreno, B.; Sabo-Etienne, S.; Chaudret, B.; Rodrıguez-
Fernan
́ ́
dez, A.; Jalon, F.; Trofimenko, S. J. Am. Chem. Soc. 1994, 116,
2635−2636. (b) Moreno, B.; Sabo-Etienne, S.; Chaudret, B.;
3099
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100

Organometallics
Article
Marginean, P. Isotopenpraxis 1986, 22, 435−438. (g) El-Nour, F. A.;
Belacy, N.; Abdel-Khalik, M.; Khalil, T.; Aly, H. F. Isotopenpraxis 1990,
26, 529−531.
(48) Kurita, T.; Aoki, F.; Mizumoto, T.; Maejima, T.; Esaki, H.;
Maegawa, T.; Monguchi, Y.; Sajiki, H. Chem. Eur. J. 2008, 14, 3371−
3379.
(49) Himeda, Y.; Miyazawa, S.; Onozawa-Komatsizaki, N.; Hirose,
T.; Kasuga, K. Dalton Trans. 2009, 6286−6288.
(50) Feracin, S.; Buergi, T.; Bakhmutov, V. I.; Eremenko, I.;
Vorontsov, E. V.; Vimenits, A. B.; Berke, H. Organometallics 1994, 13,
4194−4202.
(51) (a) Shubina, E. S.; Belkova, N. V.; Bakhmutova, E. V.;
Vorontsov, E. V.; Bakhmutov, V. I.; Ionidis, A. V.; Bianchini, C.;
Marvelli, L.; Peruzzini, M.; Epstein, L. M. Inorg. Chim. Acta 1998, 280,
302−307. (b) Belkova, N. V.; Ionidis, A. V.; Epstein, L. M.; Shubina, E.
S.; Gruendemann, S.; Golubev, N. S.; Limbach, H.-H. Eur. J. Inorg.
Chem. 2001, 1753−1761. (c) Belkova, N. V.; Shubina, E. S.; Gutsul, E.
I.; Epstein, L. M.; Eremenko, I. L.; Nefedov, S. E. J. Organomet. Chem.
2000, 610, 58−70.
(52) (a) Tolmachev, A. A.; Yurchenko, A. A.; Semenova, M. G.;
Feshchenko, N. G. Russ. J. Chem. 1993, 63, 504−506. (b) Jalil, M. A.;
Yamada, T.; Fujinami, S.; Honjo, T.; Nishikawa, H. Polyhedron 2001,
20, 627−633.
(53) Hintermann, L.; Dang, T.-Th.; Labonne, A.; Kribber, Th.; Xiao,
L.; Naumov, P. Chem. Eur. J. 2009, 15, 7167−7179.
́
(54) Moldes, I.; Encarnacion, E.; Ros, J.; Alvarez-Larena, A.; Piniella,
J. F. J. Organomet. Chem. 1998, 566, 165−174.
(55) See ref 1 in the Supporting Information.
(56) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (b) Becke,
A. D. J. Chem. Phys. 1992, 96, 2155−2160. (c) Lee., C.; Yang, W.; Parr,
R. G. Phys. Rev. 1988, B37, 785−789. (d) Ditchfield, R.; Hehre, W. J.;
Pople, J. A. J. Chem. Phys. 1971, 54, 724−728. (e) Ditchfield, R.;
Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257−2260.
(f) Hariharan, P. C.; Pople, J. A. Mol. Phys. 1974, 27, 209−214.
(g) Gordon, M. S. Chem. Phys. Lett. 1980, 76, 163−168. (h) Clark, T.;
Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P. v. R. J. Comput.
Chem. 1983, 4, 294−301.
(57) Foresman, J. B.; Keinth, T. A.; Wiberg, K. B.; Snoonian, J.;
Frisch, M. J. J. Phys. Chem. 1996, 100, 16098−16104.
(58) Cossi, M.; Barone, V. J. Chem. Phys. 2001, 115, 4708−4718.
3100
dx.doi.org/10.1021/om300015j | Organometallics 2012, 31, 3087−3100