Mechanistic Insight into H/D Exchange by a Pentanuclear Ni-H Cluster and Synthesis and Characterization of Structural Analogues of Potential Intermediates

Article abstract of DOI:10.1021/acs.organomet.7b00763

Experiments to gain mechanistic insight into catalytic H/D exchange of unactivated arenes by [(ⁱPr₃P)Ni]₅H₆ (1) are described. The reaction order with respect to 1, arene substrate, and added ⁱPr₃P were determined, as well as the temperature dependence of reaction rate. Site-selectivity data were obtained by monitoring the reaction of 1 with fluorobenzene and 2-methoxynaphthalene. H/D exchange competition reactions between arenes reacted with 1 were performed. The addition of an excess of Hg to a solution of 1 provided the new heterometallic cluster [(ⁱPr₃P)Ni]₅H₆(μ₄-Hg) (2); this species also undergoes H/D exchange with C₆D₆, albeit more slowly than 1. Reaction of cluster 1 with TlCp (Cp = C₅H₅) gave (η⁵-Cp)Ni[(ⁱPr₃P)Ni]₄(μ₂-H)₄(μ₄-Tl) (3) with the loss of H₂. A similar transfer of cyclopentadienyl to Ni occurred when 1 was reacted with MgCp₂, to give (η⁵-Cp)Ni[(ⁱPr₃P)Ni]₄H₆(μ₄-MgCp) (4), but not H₂ loss. Reaction of 1 with cyclopentadiene gave the 5-coordinate hydride complex (η⁵-Cp)Ni[(ⁱPr₃P)Ni]₄(μ₂-H)₄(μ₅-H) (5). The Cp? analogue, (η⁵-Cp)Ni[(ⁱPr₃P)Ni]₄(μ₂-H)₄(μ₅-H) (6) (where Cp? = C₅Me₅), was synthesized by reacting 1 with LiCp? in THF, or by reaction of 1 with pentamethylcyclopentadiene.

Full text of DOI:10.1021/acs.organomet.7b00763

Article
Cite This: Organometallics XXXX, XXX, XXX−XXX
Mechanistic Insight into H/D Exchange by a Pentanuclear Ni−H
Cluster and Synthesis and Characterization of Structural Analogues
of Potential Intermediates
Manar M. Shoshani, Junyang Liu, and Samuel A. Johnson*
Department of Chemistry and Biochemistry, University of Windsor, Sunset Avenue 401, Windsor, ON, N9B 3P4, Canada
S
* Supporting Information
ABSTRACT: Experiments to gain mechanistic insight into
catalytic H/D exchange of unactivated arenes by [(iPr3P)-
Ni]5H6 (1) are described. The reaction order with respect to 1,
i
arene substrate, and added Pr3P were determined, as well as
the temperature dependence of reaction rate. Site-selectivity
data were obtained by monitoring the reaction of 1 with
fluorobenzene and 2-methoxynaphthalene. H/D exchange
competition reactions between arenes reacted with 1 were
performed. The addition of an excess of Hg to a solution of 1
provided the new heterometallic cluster [(iPr3P)Ni]5H6(μ4-
Hg) (2); this species also undergoes H/D exchange with C6D6,
albeit more slowly than 1. Reaction of cluster 1 with TlCp (Cp
= C5H5) gave (η5-Cp)Ni[(iPr3P)Ni]4(μ2-H)4(μ4-Tl) (3) with
the loss of H2. A similar transfer of cyclopentadienyl to Ni
occurred when 1 was reacted with MgCp2, to give (η5-Cp)Ni[(iPr3P)Ni]4H6(μ4-MgCp) (4), but not H2 loss. Reaction of 1 with
cyclopentadiene gave the 5-coordinate hydride complex (η5-Cp)Ni[(iPr3P)Ni]4(μ2-H)4(μ5-H) (5). The Cp* analogue, (η5-
Cp*)Ni[(iPr3P)Ni]4(μ2-H)4(μ5-H) (6) (where Cp* = C5Me5), was synthesized by reacting 1 with LiCp* in THF, or by reaction
of 1 with pentamethylcyclopentadiene.
even water. Mechanistic studies of the CC bond cleavage
reaction show that all five Ni centers in 1 aid in the binding of
alkenes via initial triple C−H bond activation with minimal
reaction barriers. Reactivity studies with 1 have shown the
propensity of 1 to form stable tetranuclear clusters with
bridging N, O, B, S, and Cl heteroatoms.30 Although these
reactions give little direct insight into the mechanism of
unactivated arene H/D exchange catalyzed by 1, the studies of
CC bond cleavage suggest that it could occur with cooperativity
from all five metal centers.11
Hydrogen isotope exchange (HIE) facilitated by transition
metals is well-studied due to the application in selectively
deuterating or tritiating organic molecules that are used in
pharmaceuticals.31−37 Though several techniques have been
reported that are effective for HIE, in many cases, the
mechanisms are poorly understood and several require harsh
conditions.38 The ability of 1 to catalyze H/D exchange of
unactivated arenes such as benzene or toluene is of interest,
particularly given that room-temperature activation of arene
C−H bonds at Ni typically requires activating groups like ortho-
F substituents.39−49 Herein, we report mechanistic studies of
the catalytic H/D exchange of 1. Although no intermediates
could be observed or isolated in H/D exchange, reactivity
INTRODUCTION
■
The past decade has seen several reports of transition metal
clusters1,2 undergoing unprecedented bond transformations
with traditionally inert bonds,3−11 including the activation of
C−C bonds in hydrocarbons such as benzene4,9 and ethylene.11
This has spurred on the rational design of transition metal
clusters and polynuclear complexes,12−20 in an attempt to
access similar cooperative reactivity. Interactions among
proximal late transition metal centers21 could be envisioned
to facilitate metal surface-like transformations. Thousands of
transition metal clusters have been reported in the Cambridge
Structure Database22 with the vast majority of them composed
of either late transition metals with π-acceptor ligands like
CO23 or early transition metals with π-donor ligands.24 Few
examples of clusters are known that are composed of solely late
transition metals and electron-rich phosphines and hy-
drides.25−29 Our group has reported examples of inert-bond
activation by the electron-deficient pentanuclear Ni cluster,
[(iPr3P)Ni]5H6 (1), which includes ultra-deep hydrodesulfuri-
zation (HDS),10 CC bond cleavage of unactivated alkenes,11
and catalytic H/D exchange of unactivated aromatics such as
benzene (Scheme 1).9 Moreover, these transformations all
proceed at room temperature. Cluster 1 is highly reactive in the
presence of substrates such as unencumbered alkenes; however,
1 is surprisingly selective and functional group tolerant.
Complex 1 does not react rapidly with amines, alcohols, or
Received: October 12, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1. Inert-Bond Activation Facilitated by 1
Figure 1. (a) H/D exchange of cluster 1 with arenes to form a series of isotopologues. (b) 31P{1H} NMR of a statistical mixture of 1-d0−6
.
studies are described that provide complexes that give insight
into how these clusters could interact with arene π-ligands and
facilitate H/D exchange.
in solution at room temperature. The decrease in reaction rate
suggested that Pr3P generated by decomposition of 1 was
i
inhibiting H/D exchange. By monitoring the kinetics of H/D
exchange with 1 in the presence of different added
concentrations of phosphine, it was found that the rate of
reaction was inversely correlated to [iPr3P], suggestive of an H/
D exchange mechanism that involves phosphine dissociation at
or before the rate-determining step. Similar experiments were
conducted to determine the order of reaction of 1 and C6D6.
RESULTS AND DISCUSSION
■
Stoichiometric H/D Exchange with 1. The stoichio-
metric reaction of 1 with an excess of deuterated arene provides
a simple noncatalytic system to study H/D exchange (Figure
1). The room-temperature addition of 1 to deuterated arenes,
such as C6D6 and C7D8, provides the isotopologues 1-dn
(where n = 1−6), as shown in Scheme 2, along with
isotopologues of the arene. These reactions proceed at room
temperature, and the degree of H/D exchange into 1 is easily
monitored by 31P{1H} NMR, because the resonances for 1-d0−6
have large temperature-dependent isotope shifts that arise due
to a low-energy paramagnetic state.9 The extent of the isotope
shift is dependent upon the degree of deuteration, with a
separation between the 1-d0 and 1-d1 31P{1H} NMR shifts of
0.9 ppm, and a separation between the 1-d5 and 1-d6 shifts of
0.6 ppm, at 298 K. The incorporation of deuterium into 1 gives
a statistical mixture of 1-d0−6 that can be modeled to accurately
determine the percent deuteration, and thus monitor the rate of
H/D exchange.
i
These reactions were done with added Pr3P to avoid errors
associated with the thermal decomposition of 1 liberating iPr3P.
The rate of exchange was found to be first-order with respect to
both the concentration of Ni-H in 1-d0−6 and in [C6D6]. The
addition of partial pressures of H2 as low as 1.3 × 10−6 atm
caused a 10-fold decrease in rate; however, identical rates were
observed at higher H2 pressures of 1.3 × 10−5, and 6.6 × 10−5
atm. This observation can be rationalized by proposing two
mechanisms of H/D exchange. One mechanism, which
dominates in the absence of added H2 is postulated to involve
equilibrium loss of H2 prior to a rate determining C−D bond
activation. The second mechanism that dominates in the
presence of even minimal added H2 would have rate
determining C−D bond activation occur without prior H2
reductive elimination, either by an oxidative addition or σ-bond
metathesis mechanism.
H/D exchange reactions between 1 and C6D6 are first-order
with respect to overall Ni-H concentration in solution for
several hours, but slowed as complex 1 gradually decomposed
B
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 2. (a) The Observed Ortho and Meta H/D Exchange between C6D6 and Fluorobenzene with Catalytic 1. (b) Deuterium
Incorporation into All sp2 C-H Sites of 2-Methoxy Naphthalene apart from the 1-Site. (c) General Scheme for Competition
a
Reactions Conducted with Various Substituents
a
Relative rates between substrate C-H sites are shown beside their respective sites.
In both arene solvents and pentane, the addition of higher
pressures of dihydrogen to solutions of 1 gives an equilibrium
amount of a new hydride containing species. Attempts to
crystallize this equilibrium complex gave only low yields of its
less soluble precursor 1, even under 4 atm H2 and at −80 °C,
so its characterization was limited to solution NMR. The
complex features a broad hydride resonance at δ −19.0 and a
broad resonance in the 31P{1H} NMR spectrum that overlaps
with 1. Increasing the pressure of dihydrogen gives an increase
in the concentration of the unknown product. It is likely that
this species is the 68 electron complex [(iPr3P)Ni]5H8; the Pt
analogue, [(tBu2PhP)Pt]5H8, is known, though no reactivity
studies have been reported.26 Although the reaction of 1 with
1 and C6D6 to monitor the regioselectivity of H/D exchange.
Site-selectivity studies with cluster 1 and fluorobenzene
revealed approximately equal meta and ortho H/D exchange
with no observed para activation, as shown in Scheme 2a. The
observed site selectivity is different from known mononuclear
Ni C−H activation mechanisms which are generally ortho
selective.50 The regioselectivity with fluorobenzene suggests
that 1 does not dissociate into mononuclear fragments that
then undergo reversible C−H bond oxidative addition. Cluster
1 and C6D6 were also exposed to 2-methoxynaphthalene, and
1H and 13C{1H} NMR spectra revealed deuterium incorpo-
ration in all aromatic C-H sites apart from the 1-site, which is
electronically similar to the 3-site. The lack of reactivity at the
1-site is presumably due to the steric influence of the closest
C−H bond on the adjacent ring. Peculiar regioselectivities were
observed with a few of the analyzed substrates. For example,
anisole showed a strong preference for meta H/D exchange,
which was not observed for the electronically similar
dimethylaniline. In addition, ortho H/D exchange was observed
in anisole but not dimethylaniline. This latter fact may be
rationalized by the larger steric influence of the NMe2 group. A
similar lack of regioselectivity was observed in α,α,α
trifluorotoluene, with no exchange at the sterically encumbered
ortho positions, but near equal rates of exchange at the meta and
para sites.
D2 gives rapid incorporation of deuterium to give 1-d1−6
,
catalytic incorporation of deuterium into C6H6 was exceedingly
slow using D2 gas, due to the inhibition of the faster H/D
exchange mechanism by added D2.
To provide further mechanistic insight, the effects of
temperature on the rate of H/D exchange were determined.
A solution composed of 0.021 M of 1, 1.623 M of C6D6, and
0.095 M of iPr3P in n-pentane was monitored at 15, 20, 25, and
30 °C by 31{P} NMR to determine the rate of H/D exchange.
The Eyring plot graphed showed a linear correlation as can be
seen in Figure S4. The enthalpy and entropy of activation were
41.5 kcal·mol−1 and 67.4 cal·K−1·mol−1. The large positive
entropy of activation is typical of a dissociative mechanism. The
influence of both iPr3P dissociation and C6H6 association steps
should give a net marginal entropy of activation. The large
positive entropy of activation is consistent with the dominant
mechanism involving further equilibrium loss of H2 under these
conditions, to give potentially the 68 electron species
(C6D6)Ni[(iPr3P)Ni]4(μ2-H)4, which could then undergo
rate-limiting C−D bond activation.
Competition reactions were conducted with a variety of
monosubstituted arenes in the presence of catalytic amounts of
1 and C6D6, to monitor the effect of changing electronic
parameters on the rate of H/D exchange, as shown in Scheme
2b. However, with the exception of α,α,α-trifluorotoluene and
fluorobenzene, the change in substituents induced a minimal
influence on the rate of H/D exchange. Although the most
electrophilic arenes react the fastest, the selectivities are not
consistent with a nucleophilic attack at the arene. The electron-
withdrawing nature of the CF3 and F substituents may
Catalytic H/D Exchange and Site Selectivity. Several
monosubstituted arenes were exposed to a catalytic amount of
C
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
influence the arene-binding step of the mechanism, which
would account for the increase in rate of exchange. The lack of
typical selectivity in these reactions may indicate a transition
state for C-H activation with little Ni−C bond formation.
When these competition reactions were run for extensive
periods of time, there was evidence of Ni metal particle
formation that also facilitated H/D exchange, but with different
selectivities.51 For example, when 1 was exposed to C6D6 and
toluene, no initial H/D exchange into the sp3 C−H bonds of
toluene was observed, but after 16 h at room temperature,
significant deuteration of the methyl site was detected,
consistent with H/D exchange mediated by a Ni surface.52,53
The regioselectivities presented in Scheme 2c were determined
from initial rates of exchange.
hydride region of the H NMR spectrum, which has two
singlets at δ −34.75 and δ −16.30 in a 4:2 ratio. An Arrhenius
activation energy of 8.1 kcal·mol−1 was calculated for the
exchange process. The 13C{1H} NMR of 2 displays a pair of
resonances for the methyl and methine environments of the
phosphine ligand, with the latter coupling to the phosphine to
give a doublet. Despite our best efforts, 199Hg{1H} NMR
spectra yielded no resolvable signal from a chemical shift range
of δ −5000 to 5000, even at −60 °C.
The crystallization of 2 is facile and yields seemingly perfect
cubic crystals that diffract well; however, the solid-state
structure from single-crystal X-ray diffraction could not be
determined accurately, likely due to twinning, and only
connectivity could be confirmed. Around 82% of the diffraction
data indexed to a tetragonal P unit cell with a volume of
approximately 5800 Å3, but weaker diffraction fit to an apparent
tetragonal cell with 4 times the volume. In multiple different
data sets, it proved difficult to assign both the unit cell and
space group, though, in many attempted space groups, the
Ni5Hg core could be observed, but with complete disorder of
the supporting ligands. Further details are provided in the
Supporting Information. Cluster 2 features a [Ni5Hg]6+ core in
an octahedral geometry and could be considered an octahedral
closo cluster. It bears an electron count of 76 electrons, which is
considered electron deficient compared to the anticipated 86
electrons. The Wade−Mingos electron counting rules are based
on the 18-electron rule for transition metal complexes,58 while
many 16 electron complexes are known for Ni centers.59
Accounting for the 2-electron difference per Ni center, cluster 2
is electron precise. Only a few Ni-Hg complexes have been
reported in the Cambridge Structural Database;60−67 Kubiak
has reported a Hg center bound to four Ni centers, albeit in a
distorted planar-Hg geometry.67 The closest structural
analogues with respect to the capping μ4-Hg environment in
2 have been reported in Au, Au/Pd, and Au/Pt clusters.68,69
Similarly to 1, samples of 2 in C6D6 were found to undergo H/
D exchange with the solvent, which could be monitored by a
decrease of the hydride resonance intensity in the 1H NMR and
an increase in the intensity of the C6D5H resonance. The
exchange with 2 in neat C6D6 is significantly slower than with 1,
with a t1/2 of 4.76 × 104 s versus the t1/2 of 1.8 × 103 s for 1.
Reactions of 1 with Cyclopentadienide Moieties and
Cyclopentadiene. Cyclopentadienylthallium was reacted with
cluster 1 at room temperature, with the intention that the CpTl
fragment would bind in a similar manner to Hg in complex 2.
The 31P{1H} and 1H NMR showed immediate conversion to a
new hydride bearing complex along with the loss of an
equivalent of iPr3P. Rather than having a TlCp fragment bound
to the square plane of the cluster, (η5-Cp)Ni[(iPr3P)Ni]4(μ2-
H)4(μ4-Tl) (3) was formed, which features Tl bound to the 4
Ni of the basal square plane, with the Cp ligand replacing the
phosphine on the capping Ni along with the loss of 1 equiv of
dihydrogen, as shown in Scheme 4. Cluster 3 was isolated
through recrystallization from n-pentane at −40 °C in a yield of
66%.
H/D Exchange in the Presence of Hg. The first-order
kinetics observed in our kinetics studies are not consistent with
Ni metal from the decomposition of 1 being the active species
for H/D exchange; however, as noted in the absence of added
iPr3P and over extended reaction times, a precipitate formed
from solutions of 1, and H/D exchange reactions are observed
that 1 is not capable of, such as the weaker sp3 C−H bonds in
the methyl group of toluene. In an attempt to amalgamate any
Ni nanoparticles competing with cluster 1, mercury metal was
added to solutions of 1 and C6D6. This reaction was monitored
1
by H and 31P{1H} NMR spectroscopy, and both showed the
immediate disappearance of 1 and the formation of [(iPr3P)-
Ni]5(H6)(μ4-Hg) (2) as shown in Scheme 3. This reaction
Scheme 3. Reaction of Cluster 1 with Hg To Give the
Heterometallic Cluster 2
proceeds rapidly at room temperature when a large excess of
Hg metal is used. Despite the quantitative conversion observed
by NMR spectroscopy, an isolated yield of approximately 58%
was obtained after recrystallization from n-pentane.
The 31P{1H} NMR spectrum of 2 features a broad peak
centered at approximately δ 37.5 with a W1/2 of 1250 Hz (W1/2
= peak width at half-height). At −40 °C, the slow exchange
limit 31P{1H} NMR spectrum was observed, which features a
doublet at δ 37.2 with a JPP = 12.9 Hz that corresponds to the 4
phosphines in the basal plane, and a pentet at δ 94.5 for the
capping Ni with the same JPP = 12.9 Hz, but also 199Hg satellites
(I = 1/2, 16.8%) with a JPHg of 1208 Hz. The observation of
coupling to the Hg center for the phosphine on the capping Ni
but not for the phosphines on the Ni centers in the square base
is unique to 2, and not observed in structurally related μ4-Hg
2
Au clusters, where the JHgP coupling to the phosphines in the
basal plane is large.54,55 Antipodal ligands in carborane and
The solid-state structure of 3 is shown in Figure 2. The
molecule has four Ni centers in a square plane with Ni−Ni
bond distances ranging from 2.5107(4) to 2.5228(4) Å and a
capping Ni bound to all four Ni centers with Ni−Ni bond
lengths ranging from 2.3696(4) to 2.3730(4) Å. The four
phosphine centers all lie in the plane of the Ni square base. The
four bridging hydride ligands were located in an electron
density difference map and are also in the square plane. The η5-
metallocarborane clusters have been reported to influence
chemical shifts, but not coupling constants.56,57 The H NMR
1
spectrum has the expected PCH and Me resonances for the five
iPr3P in rapid exchange, along with a resonance for the six
hydrides at δ −28.62. The hydride resonance is exchange
broadened, with a W1/2 of 450 Hz. Cooling the solution of
cluster 2 to −60 °C gave the slow exchange limit for the
D
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
process exchanges hydride environments; the same is not true
in related complexes with a square planar Ni4H4 core.10 The
205Tl{1H} NMR spectrum features a broad singlet at δ 821.1.
Similar to 2, cluster 3 is a closo cluster with an octahedral core
geometry. Overall, the four hydride ligands and the Cp ligand
give cluster 3 five anionic X ligands. The thallium center is best
considered as formally cationic, consistent with thallium’s most
common oxidation state, Tl(I). The electron count of cluster 2
of 68 electrons, and once again the apparent deficiency could
be explained by the propensity of Ni complexes to form stable
16 electron compounds. Transition metal thallium complexes
are reported in the literature; however, examples of Ni−Tl
complexes are quite rare,70−72 though an example of a
trinuclear Ni cluster with a Tl−I fragment was reported by
Kubiak.72 The capping μ4-Tl binding mode of Tl in 3 is
unprecedented in any Tl containing transition metal cluster.
Cyclopentadienyl complexes are well documented in the
organometallic chemistry of Ni73 and have also found use as a
supporting ligand in Ni clusters.29 Of interest in H/D exchange
Scheme 4. Reaction of Cluster 1 with TlCp To Synthesize
the Cp Bearing Cluster 3
with 1 is that the formally anionic (η5-Cp)(iPr3P)4Ni5(μ2-H)4
−
fragment in 3 is isoelectronic to the proposed intermediate for
H/D exchange via the H2 dissociation mechanism, (η6-
C6D6)Ni[(iPr3P)Ni]4(μ2-H)4. The formal anionic fragment in
3 is stabilized by strong covalent interactions afforded by the
soft cation Tl, so attempts were made to synthesize an
analogous cluster with a harder cation that would afford only
ionic interactions. The reaction of [(iPr3P)Ni]5H6 with 1 equiv
of Cp2Mg at room temperature gave dark brown (η5-
Cp)(iPr3P)4Ni5(μ2-H6)(μ4-MgCp) (4) and an equivalent of
iPr3P, as shown in Scheme 5. Contrary to the reaction forming
3, no loss of dihydrogen is observed, and 4 retains six hydride
Figure 2. ORTEP depiction of cluster 3. Phosphine isopropyl groups
and hydrogens on the Cp ligand are omitted for clarity. Selected bond
distances (Å): Ni(1)−Ni(2) 2.3697(4), Ni(1)−Ni(3) 2.3696(4),
Ni(1)−Ni(4) 2.3718(4), Ni(1)−Ni(5) 2.3730(4), Ni(1)−Cpcentroid
1.788, Ni(2)−Ni(3) 2.5228(4), Ni(3)−Ni(4) 2.5124(4), Ni(2)−
Ni(5) 2.5107(4), Ni(4)−Ni(5) 2.5132(4), Ni(2)−Tl(1) 2.7334(3),
Ni(3)−Tl(1) 2.6798(3), Ni(4)−Tl(1) 2.7302(3), Ni(5)−Tl(1)
2.7158(3). Selected bond angles (deg): Tl(1)−Ni(1)−Cpcentroid
178.15, Ni(3)−Tl(1)−Ni(5) 81.838(9), Ni(2)−Tl(1)−Ni(4)
81.825(9).
1
ligands. Although the 31P{1H} and H NMR spectra showed
near quantitative conversion to 4, an isolated yield of 68% was
obtained after recrystallization from n-pentane at −40 °C.
The solid-state structure of 4 was obtained by X-ray
crystallography and is shown in Figure 3. The structure
features two-fold disorder of the CpNi5MgCp core. The
proximal carbon atoms of the C(1)−C(5) and C(6)−C(10)
cyclopentadienyl rings were constrained to have identical
thermal parameters. The nickel hydrides, which should also
be two-fold disordered, could not be accurately located in the
electron density difference map, and thus were not included in
the model. Cluster 4 features a distorted square pyramid
geometry composed of 5 Ni centers, with the square base Ni−
Ni bond distances ranging from 2.417(10) to 2.502(10) Å and
the Ni−Ni bond distances from the capping Ni to the square
base Ni centers ranging from 2.485(10) to 2.650(10) Å. Four
terminal phosphine ligands are bound to each of the Ni centers
in the square base, while the capping Ni(1) has an η5-Cp. The
Cp ligand is bound to the capping Ni with a Ni(1)−Cp
centroid distance of 1.788 Å. The Tl center is located opposite
of the capping Ni of the cluster and is bound to the four Ni
centers in the square plane with Ni−Tl bond distances ranging
from 2.6798(3) to 2.7334(3) Å. The Cpcentroid−Ni(1)−Tl(1)
bond angle is 178.15°.
The 31P{1H} NMR spectrum of 3 features a singlet at δ 31.4.
The 13C{1H} NMR spectrum displays a singlet for the PiPr
methyl groups, whereas a second-order AXX′2X″ multiplet with
2JPC = 10.9 Hz is observed for the methine environment. A
1
singlet at δ 85.2 is observed from the Cp carbons. The H
NMR spectrum displays a pair of resonances corresponding to
i
the Pr substituents and a singlet at δ 5.20 for the Cp ligand.
The hydride resonance δ −24.20 is a pentet with JPH = 2.8 Hz
that integrates to 4 hydrides relative to the four phosphine
ligands. The observation of a binomial pentet suggests that this
is an average coupling to all four phosphines, and a fluxional
Scheme 5. Reaction of Cluster 1 with Cp2Mg To Give 4
E
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Cluster 4 has six hydride ligands in comparison to cluster 3
which only has four. The difference corresponds to a change in
core oxidation state with 4 bearing a formally [Ni5]8+ core in
comparison to 3 which bears a [Ni5]6+ core. This result has
implication on the potential mechanism for H/D exchange,
with the Ni5 core being able to support a capping π-ligand with
both six and four hydride ligands.
The pentanuclear Ni cluster [(iPr3P)Ni]5H6] also reacts
immediately with 1 equiv of freshly distilled cyclopentadiene at
room temperature to give dark brown (η5-Cp)Ni[(iPr3P)-
Ni]4(μ2-H)4(μ5-H) (5), as shown in Scheme 6. The reaction
produces 1 equiv of dihydrogen and releases an equivalent of
Figure 3. ORTEP depiction of cluster 4, with disorder not shown.
Carbons and hydrogens on phosphine ligands excluded for clarity.
Hydride ligands were not located. Selected bond distances (Å):
Ni(1)−Ni(2A) 2.616(9), Ni(1)−Ni(3A) 2.497(4), Ni(1)−Ni(2B)
2.651(10), Ni(1)−Ni(3B) 2.485(10), Ni(1)−C(1−5)centroid 1.763,
Ni(2A)−Ni(3A) 2.417(10), Ni(3A)−Ni(2B) 2.435(10), Ni(2B)−
Ni(3B) 2.502(10), Ni(3B)−Ni(2A) 2.485(10), Ni(2A)−Mg(1)
2.682(16), Ni(3A)−Mg(1) 2.562(10), Ni(2B)−Mg(1) 2.713(16),
Ni(3B)−Mg(1) 2.947(13). Selected bond angles (deg): Mg(1)−
Ni(1)−C(1−5)avg 176.24, Ni(2A)−Mg(1)−Ni(2B) 81.5(4), Ni(3A)−
Mg(1)−Ni(3B) 76.5(3).
1
iPr3P. The H and 31P{1H} NMR spectra shows quantitative
conversion to 5, and an isolated yield of 56% was obtained after
crystallization from n-pentane at −40 °C.
The solid-state structure of 5 was determined by X-ray
crystallography and is shown in Figure 4. Complex 5 has Ni
bond distance from the capping Ni(1) to the Cp centroid is
1.794 Å which is similar to the Ni−Cpcentroid distance featured
in 3. The bottom face of the cluster features a MgCp fragment
that is bound to the 4 Ni centers in the square base with Ni−
Mg distances ranging from 2.562(10) to 2.947(13) Å. The Cp
ligand bound to the Ni center lies almost directly above the Ni
center along with the MgCp fragment below with a Cpcentroid
−
Ni(1)−Mg(1) bond angle of 170.1°.
Figure 4. ORTEP depiction of cluster 5. Carbons and hydrogens on
and hydrogens on Cp ligand were excluded for clarity. Selected bond
distances (Å): Ni(1)−Ni(2) 2.3649(5), Ni(1)−Ni(3) 2.3752 (5),
Ni(1)−Ni(4) 2.3643(5), Ni(1)−Ni(5) 2.3719(5), Ni(1)−Cpcentroid
1.738, Ni(1)−H(5) 1.54(5), Ni(2)−Ni(3) 2.5273(5), Ni(3)−Ni(4)
2.5487(5), Ni(2)−Ni(5) 2.5414(5), Ni(4)−Ni(5) 2.5531(5), Ni(2)−
H(5) 1.77(5), Ni(3)−H(5) 1.82(5), Ni(4)−H(5) 1.81(5), Ni(5)−
H(5) 1.80(5). Selected bond angles (deg): Cpcentroid−Ni(1)−H(5)
174.76, N(3)−H(5)−Ni(5) 179.39(19), Ni(2)−H(5)−Ni(4)
178.46(15).
Although the solid-state structure of 4 did not reveal the
1
hydride locations, the room-temperature H NMR had a sharp
singlet hydride resonance at δ −22.23 that integrates to six. The
CpNi and CpMg resonances were observed at δ 5.91 and δ
6.78. Cooling a sample of 3 in n-pentane as low as −90 °C did
not cause decoalesence of the hydride resonance. The 31P{1H}
NMR spectrum had a singlet at δ 51.5, consistent with the C4v
symmetry of cluster 3. Selective decoupling of the nonhydride
resonances was attempted to further confirm the assignment of
6 hydride ligands; however, coupling to the hydride ligands
only caused broadening of the 31P NMR resonance. The
13C{1H} NMR spectrum features a pair of resonances
corresponding to the methyl and methine environments on
the phosphine ligands, at δ 20.6 and δ 24.8, respectively. The
methyl peak is a singlet, but the methine resonance is a second-
order multiplet consistent with the spin system AXX′2X″ and a
2JPC = 11.1 Hz. Two additional singlets are observed at
approximately δ 85.8 and δ 97.3, for the CpNi and CpMg
fragments.
centers in a distorted square pyramidal geometry. The bond
distances between the square-base Ni centers range from
2.5273(5) to 2.5531(5) Å. The Ni−Ni bond distances are
considerably shorter to the capping Ni center, and range from
2.3643(5) to 2.3752(5) Å. A phosphine donor is bound to each
Ni center in the square base, and the capping Ni(1)−Cpcentroid
distance is 1.7381 Å. The five hydride ligands were located in
the electron density difference map. Four are μ2-bound
between the basal Ni centers, in the Ni4 square plane. The
Scheme 6. Reaction of Cluster 1 with Cyclopentadiene To Form the 5-Coordinate Hydride Bearing Cluster 5
F
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 7. Reaction of Cluster 1 with Lithium Pentamethylcylopentadienide To Form the 5-Coordinate Hydride Bearing
Cluster 6
distinct hydride ligand is located in the Ni square plane, below
the capping Ni(1) with a short Ni(1)−H(5) distance of
1.54(5) Å. The Ni−H(5) distances for the Ni centers in the
square plane are longer, and range from 1.77(5) to 1.82(5) Å.
Only one other example of a 5-coordinate hydride complex has
been reported.74
The 31P{1H} NMR of 5 has a sharp singlet at about δ 48.5,
1
consistent with apparent C4v symmetry. The H NMR features
a pair of resonances corresponding to the phosphine ligands
along with a singlet at δ 5.56 for the protons on the Cp ligand.
The hydride region has a single broad resonance at δ −29.66
Figure 5. ORTEP depiction of Cluster 6. Carbons and hydrogens on
for all five hydride ligands. This is consistent with a fluctional
phosphine ligands and hydrogens on methyl groups of Cp* excluded
process that exchanges the μ5-H hydride ligand and the four μ2-
H. When 5 in n-pentane is cooled to −60 °C, the hydride
for clarity. Selected bond distances (Å): Ni(1)−Ni(2) 2.3991(6),
Ni(1)−Ni(3) 2.3852 (7), Ni(1)−Ni(4) 2.4063(7), Ni(1)−Ni(5)
resonance decoalesces to give two peaks in a 4:1 ratio at δ
2.3977(6), Ni(1)−Ccentroid 1.748, Ni(1)−H(5) 1.50(7), Ni(2)−Ni(3)
−30.52 and δ −22.28, respectively. The variable-temperarure
1H NMR spectra line widths were used to determine an
Arrhenius activation energy of 21.1 kcal·mol−1 for exchange.
The 13C{1H} NMR features singlets for the methyl group and a
broad resonance with a W1/2 = 25.3 Hz for the methine
environments, which is presumably due to an unresolved
second-order multiplet. A resonance is also observed at δ 85.4
for the Cp ligand.
2.5659(6), Ni(3)−Ni(4) 2.5528(6), Ni(2)−Ni(5) 2.5510(7), Ni(4)−
Ni(5) 2.5570(5), Ni(2)−H(5) 1.77(5), Ni(3)−H(5) 1.82(5), Ni(4)−
H(5) 1.81(5), Ni(5)−H(5) 1.80(5). Selected bond angles (deg):
Ni(1)−Cavg−H(5) 177.29, N(3)−H(5)−Ni(5) 176(5), Ni(2)−
H(5)−Ni(4) 174(5).
methine signal of the phosphine ligands along with a sharp
singlet at δ 2.18 for the methyl peak of the Cp* ligand. A broad
hydride resonance, which integrates to five hydrides with
respect to the phosphine environments, is observed at δ
−29.02. Cooling an isolated sample of 6 to −65 °C gave the
slow-exchange spectrum, where the hydride resonance split into
two peaks in a 4:1 ratio at δ −31.1 and δ −21.1, respectively.
Cluster 5 bears a [Ni5]6+ core in a square pyramidal
geometry and can be considered a nido cluster that is one vertex
deficient of being a closed polyhedron. The suggested stable
electron count for these clusters is 74 electrons; however, 1
bears 68 electrons. Once again, the deficiency may arise from
the propensity of Ni centers to disobey the 18-electron rule.
Cluster 1 is stable at room temperature in the solid-state and in
solution over a period of days.
1
The variable-temperature H NMR line widths were modeled
to give an Arrhenius activation energy of 15.9 kcal·mol−1, which
is 5.2 kcal·mol−1 lower than in 5. The 13C{1H} NMR of 2
features a pair of singlets, for the methyl and methine carbons
of the phosphine ligands, at δ 20.9 and δ 25.3. Two singlets are
also observed at δ 13.0 and δ 92.2 for the methyl and
cyclopentadienyl carbons on the Cp* ligand, respectively.
Cluster 6 shares many electronic similarities to 5; the core
oxidation state of [Ni5]6+ and electron count of 68 electrons are
identical. The bond distances and bond angles are also very
similar. One important conclusion from the structure of 6 is
that these clusters can bind even sterically encumbered π-
ligands to the capping Ni center without much of an effect on
the neighboring metal centers or supporting ligands. In fact, the
Ni(1)−Cp*centroid bond distance in 6 is practically identical to
the Ni(1)−Cpcentroid distance in 5. This has the implication that,
in the H/D exchange mechanism of 1, the arene-binding step
may not be greatly influenced by arene size, consistent with the
observed reactivity of larger substituted naphthalenes. Clusters
3−6 serve as structural analogues of potential intermediates for
H/D exchange by 1. Cluster 6 also facilitated H/D exchange in
neat C6D6, observed by the pseudo-first-order decrease of the
1H NMR hydride integral, with a half-life of 5.62 × 104 s.
The pentamethylcyclopentadienyl (Cp*) analogue (η5-
Cp*)Ni[(iPr3P)Ni]4(μ2-H)4(μ5-H) (6) was synthesized from
either the reaction of [(iPr3P)Ni]5H6 with pentamethylcyclo-
pentadiene or by reaction with an excess of Cp*Li in THF over
a period of 2 days (Scheme 7). It is believed that the latter
reaction occurs because Cp*Li slowly deprotonates THF to
generate pentamethylcyclopentadiene, and no initial direct
reaction of Cp*Li with 1 to generate an analogue of 4 was
observed. Both routes to 6 proceed quantitatively as observed
1
by 31P{1H} and H NMR, and 6 was isolated with a yield of
55% after crystallization at −40 °C from n-pentane.
The solid-state structure of 6 was determined by X-ray
crystallography and is shown in Figure 5. All five hydride
ligands in 6 were found in the electron density map and lie
approximately in the Ni(2) to Ni(5) square plane, with the
shortest Ni−H distance of 1.50(7) Å belonging to the capping
Ni(1) and the 5-coordinate H(5).
The 31P{1H} NMR of 6 has a singlet at δ 46.4, which is
consistent with apparent C4v symmetry. The 1H NMR is similar
to 5, with a pair of resonances observed for the methyl and
G
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 8. (a) Catalytic Conversion of 1,3- and 1,4-Cyclohexadiene to Cyclohexene and Benzene by 1. (b) Reaction of 1 with
1,4-Pentadiene Forming the Known Carbide Cluster [(iPr3P)Ni]5(H)4 (μ5-C)
Incorporation of deuterium into the methyl sites of the Cp*
ligand was also observed, but at a slower rate and not with first-
order kinetics. A likely mechanism for H/D exchange into the
Cp* ligand is via a tuck-in complex, though the slow
decomposition of 6 under these conditions prevented a
complete analysis.
elimination prior to the rate-determining step. Similar rates of
catalytic ortho and meta H/D exchange in fluorobenzene
suggests a very different mechanism of H/D exchange than the
C−H bond oxidative addition or H-atom transfer known for
mononuclear Ni(0)-phosphine complexes.
The competition of monosubstituted arenes with a wide
array of electronic parameters demonstrated an unusual lack of
selectivity in H/D exchange for Ni complexes. Exposing 1 to
arenes in the presence of Hg led to the coordination of Hg to
the Ni square plane forming the heterometallic cluster 2.
Exposing 1 to the isolobal TlCp yielded cluster 3, which
featured a cationic Tl center bound to the 4 Ni centers of the
square base and the Cp ligand displacing the phosphine of the
capping Ni. Three additional Cp bearing clusters (4−6) were
synthesized by exposing 1 to MgCp2, cyclopentadiene, and
LiCp*, respectively, with 5 and 6 featuring a 5-coordinate
hydride ligand. The dissociation of the phosphine ligand to
bind an arene supports the kinetics data provided indicating
phosphine inhibition and a dissociative type of mechanism.
Moreover, clusters 3−6 range in the number of hydride ligands
they bear, with four hydrides on 3, six hydrides on 4, and five
hydrides in 5 and 6. The isolation of analogues of potential
intermediates of H/D exchange with a range of hydride ligands
has major mechanistic implications by showing that 1 can
potentially bind an arene and maintain structural integrity upon
loss of H2 or transfer of a hydride ligand. Cluster 1 has shown
remarkable reactivity at room temperature with unactivated
substrates including unactivated hydrocarbons such as benzene
and ethylene. To date, little remains known regarding the
chemistry of small clusters supported solely by electron-rich
donor ligands. The electron counts and accessible core
oxidation states and ligand bonding modes seen here all
provide some insight into the possible transformations
accessible via these unique species.
The observed regioselectivity of 1 in H/D exchange noted in
Scheme 2 is not fully consistent with a nucleophilic mechanism
for H/D exchange, but the failure of fluorobenzene to perform
exchange at the para site could be explained by such a
mechanism.50 A species isoelectronic to 5 and 6, (η5-
C6D6H)Ni[(iPr3P)Ni]4(μ2-H)4(μ5-H), is a potential intermedi-
ate for H/D exchange by a mechanism that does not include H2
dissociation. This intermediate could be formed from 1 by
iPr3P dissociation and by hydride transfer to benzene-d6 to give
the C6D6H anion. In an attempt to make the protio species (η5-
C6H7)Ni[(iPr3P)Ni]4(μ2-H)4(μ5-H) directly, 1 was reacted
with both 1,4- and 1,3-cyclohexadiene as shown in Scheme
8a. In both cases, 1 was not immediatly consumed, but acted as
a hydrogen-transfer catalyst, forming equivalent amounts of
benzene and cyclohexene immediately at room temperature.
Similar reactivity with nickel complexes has been reported.75
Over the course of an hour, the cyclohexene produced reacted
with 1, forming multiple products that included the known
carbide cluster, [(iPr3P)Ni]5H4(μ5-C), from CC bond cleavage.
In an attempt to avoid the thermodynamic driving force
provided by aromatization in these reactions, 1,4-pentadiene
was reacted with 1 in an attempt to make the acyclic analogue
(η5-C5H7)Ni[(iPr3P)Ni]4(μ2-H)4(μ5-H). Instead, the previ-
ously reported cluster [(iPr3P)Ni]5H4(μ5-C) was formed
rapidly at room temperature,11 along with a series of organic
products from isomerization, hydrogenation, and carbon atom
abstraction, including pentane, 1-pentene, 2-pentene, butane, 1-
butene, 2-butene, and propane, as shown in Scheme 8b.
CONCLUSIONS
■
EXPERIMENTAL SECTION
■
In summary, mechanistic insight was obtained from kinetics
experiments conducted with 1. It was shown that the rate of H/
D exchange was first-order with respect to the concentration of
Reaction of 1 with C6D6 and Characterization of 1-dn (n = 1−
6). Complex 1 (20 mg, 0.018 mmol) was dissolved in C6D6 (600 mg,
7.68 mmol). At 5 min, the 31P{1H} NMR spectrum featured resolved
peaks for the isotopologues 1-d1, 1-d2, and 1-d3. After 25 min, the
31P{1H} NMR spectrum had resolved peaks for all isotopologues (1-d0
to 1-d6). Through modeled spectra, it was determined at this point the
mixture was a 1:6:15:20:15:6:1 statistical mixture of 1-d0:1-d1:1-d2:1-
d3:1-d4:1-d5:1-d6 arising from a 50% deuterium incorporation into the
i
1 and arene, whereas Pr3P showed first-order inhibition.
Temperature effects on the rate of exchange determined a large
positive value for the entropy of activation, indicative of a
dissociative mechanism, which is in support of the dominant
mechanism in the absence of added H2 involving H2 reductive
H
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
hydride sites of 2. At 85 min, the solution of 1 was found to have only
13% Ni−H remaining, with 87% deuteration.
Catalytic Site-Selectivity Studies with 2-Methoxynaphthalene. A
C6D6 solution of 1 (25 mg, 0.023 mmol) and 2-methoxynaphthalene
(100 mg, 0.63 mmol) was left at room temperature for 16 h and then
filtered through a plug of Celite. The selectivity of D incorporation was
1H NMR (298 K, C6D6, 500 MHz): δ −26.57 (1−d1), −26.55 (1-
d2),−26.54 (1-d3), −26.53 (1-d4), −26.51(1-d5) (s, 6H, Ni−H).
31P{1H} NMR (298 K, C6D6, 121.5 MHz): δ 59.8 (1−d1), 59.1 (1-d2),
58.3 (1-d3), 57.6 (1-d4), 56.9 (1-d5), 56.1 (1-d6) (s, 5P, NiPiPr3).
1
determined by H, and 13C{1H} NMR.
Monosubstituted Arene Competition Reactions. Solutions of 1
(5−25 mg, 0.0045−0.023 mmol) in C6D6 were made, and designated
amounts of monosubstituted arenes were added. The following
proportions were used with each competition: (a) C6H5F (118 mg,
1.23 mmol) vs C6H5N(Me)2 (150 mg, 1.23 mmol); (b) C6H5F (164
mg, 1.71 mmol) vs C6H5CH3 (161 mg, 1.71 mmol); (c) C6H5F (150
mg, 1.56 mmol) vs C6H5CF3 (228 mg, 1.56 mmol); (d) C6H5CF3
(225 mg, 2.08 mmol) vs C6H5OCH3 (60 mg, 0.41 mmol). The
solutions were left to react at room temperature for 16 h and then
were filtered through a plug of Celite. Degree of deuteration was
determined through the use of 13C{1H} NMR, either by modeling the
relative intensity of the resulting C−D coupling multiplet or by the
isotope shift caused by adjacent deuterium incorporation. The NMR
spectra were modeled through TOPSPIN and used to determine the
relative rate of each site.
i
H/D Exchange Rate vs [1]. A stock solution of Pr3P (63 mg, 0.38
mmol) was dissolved in 5 mL of C6D6. 1.575 g of the solution was
used to dissolve 101 mg of complex 1 to make solution B.
Approximate masses of 74, 123, 202, 450, and 800 mg of solution B
were weighed directly into 5 NMR tubes and diluted to 0.800 mL by
stock solution A. The NMR tubes were kept in liquid nitrogen to avoid
H/D exchange until the reaction could be monitored to obtain initial
rates. The probe was preset to 298 K to avoid changes in chemical
shift, and the ratio of the metal hydride complex to the initial ratio was
recorded versus time by the use of 31P{1H} NMR spectroscopy and
modeled using the stochastic kinetics simulation program Tenua.
Using the determined rate law, initial rates were calculated to be 4.30
× 10−5, 9.18 × 10−5, 1.18 × 10−4, 2.68 × 10−4, and 5.64 × 10−4 mol·
L−1·s−1. The plot of concentration of complex 1 vs rate was linear,
consistent with a first-order dependence of [1] on rate of H/D
exchange.
H/D Exchange Rate vs [C6D6]. A stock solution was made by
dissolving 1 (95 mg, 0.086 mmol) and iPr3P (60 mg, 0.38 mmol) in 5
mL of pentane. Approximately 400 mg of stock solution was weighed
directly into 5 different NMR tubes, followed by the addition of 75, 40,
28, 10, and 7 mg of C6D6 into each NMR tube. The NMR tubes were
then diluted with n-pentane to a total volume of 0.900 mL. The NMR
tubes were kept in liquid nitrogen to avoid unmonitored H/D
exchange. The NMR probe was preset to 298 K to avoid changes in
chemical shift, and the ratio of the metal hydride complex to the initial
ratio was recorded versus time by the use of 31P{1H} NMR
spectroscopy modeled using Tenua. Using the rate law, kobs values
were determined to be 4.32 × 10−4, 2.68 × 10−4, 1.95 × 10−4, 6.17 ×
10−5, and 4.55 × 10−5 s−1. The plot of concentration of C6D6 vs kobs
was linear, consistent with a first-order dependence of [C6D6] on rate
of H/D exchange.
H/D Exchange of 1 and C6D6 at Variable Temperatures. A stock
solution of Ni5L5H6 (100 mg, 0.088 mmol), triisopropylphosphine (60
mg, 0.37 mmol), and C6D6 (600 mg, 7.14 mmol) was dissolved in 4.4
mL of pentane. The stock solution was transferred evenly into 4
different NMR tubes. The NMR tubes are kept in liquid nitrogen to
avoid unmonitored H/D exchange. Four different temperatures were
preset in the NMR probe, 20, 25, 30, and 35 °C, and the rate constants
for each temperature were recorded by the use of 31P{1H} NMR
spectroscopy and modeled with Tenua. Using the determined rate
constants, an Eyring plot was produced and an enthalpy of activation
of 41.5 kcal·mol−1 and an entropy of activation of 67.4 cal·K−1·mol−1.
Synthesis and Characterization of [(iPr3P)Ni]5H6(μ4-Hg) (2). A
solution of [(iPr3P)Ni]5H6 (550 mg, 0.50 mmol) in 15 mL of n-
pentane was stirred in the presence of an excess of elemental Hg (9.4
g, 47.1 mmol) for 2.5 h. All volatiles were then removed in vacuo, and
the residue was extracted with n-pentane (2 × 10 mL). The solution
was then left at −40 °C for 48 h, at which point dark brown block
crystals suitable for X-ray diffraction were obtained. The mother liquor
was then concentrated and multiple crops yielded 373 mg of 1
(57.4%). 1H NMR (298 K, C6D6, 500 MHz): δ −28.62 (broad s, W1/2
= 453.4 Hz, 6H, Ni−H), 1.40 (dd, 3JPH = 11.8 Hz, 3JHH = 7.0 Hz, 90H,
H/D Exchange Rate vs [PiPr3]. A solution of 1 (100 mg, 0.082
mmol) in 2.1 mL of n-pentane was made. 215 mg of stock solution
was weighed directly into four different NMR tubes. Varying amounts
of PiPr3 were added amounting to 16, 8, 4, and 2 mg, each in 0.5 mL of
n-pentane. Finally, 100 mg of C6D6 was added to each NMR tube. The
NMR tubes are kept in liquid nitrogen to avoid unmonitored H/D
exchange. The probe was preset to 298 K to avoid changes in chemical
shift, and the ratio of the metal hydride complex to the initial ratio was
recorded versus time by the use of 31P{1H} NMR spectroscopy and
modeled with Tenua. Using the rate law, kobs values were determined
to be 5.83 × 10−5, 9.33 × 10−5, 1.12 × 10−4, 1.43 × 10−4, and 4.57 ×
10−4 s−1. A plot of 1/[iPr3P] vs kobs is near linear.
H/D Exchange of 1 and C6D6 in the Presence of H2. A stock
solution of 1 (100 mg, 0.088 mmol), triisopropylphosphine (25 mg,
0.16 mmol), and C6D6 (720 mg, 8.57 mmol) was diluted with 3 mL of
n-pentane. The solution was quickly transferred to 4 J. Young NMR
tubes and frozen in liquid nitrogen. Three of the samples were placed
under vacuum to remove the headspace and subsequently exposed to
1.3 × 10−5, 6.6 × 10−5, and 1.3 × 10−6 atm of H2. The samples were
warmed up in the NMR probe which was preset to 298 K to avoid
changes in chemical shift. The overall % deuteration of 1 was recorded
versus time using 31P{1H} NMR spectroscopy and revealed kobs values
of 1.2 × 10−5, 1.3 × 10−5, and 1.6 × 10−5 s−1, respectively. The kobs
value of the control sample was 1.35 × 10−4 s−1.
3
2
PCHMe2), 1.96 (apparent octet, JHH = 7.0 Hz, JHP = 6.7 Hz, 15H,
PCH). 31P{1H} NMR (298 K, C6D6, 121.5 MHz): δ 37.51 (broad s,
W1/2 = 1235.4 Hz, 5P, NiPiPr3). 13C{1H} NMR (298 K, C6D6, 75.5
2
MHz) δ 20.21 (s, 30 C, PCHMe2) 24.80 (d, JPC = 11.9 Hz, 15C,
PCH). 1H NMR (233 K, n-pentane, 500 MHz) δ −34.75 (s, 4H, Ni−
H) −16.30 (s, 2H, Ni−H) 31P{1H} NMR (233 K, n-pentane, 121.5
MHz) 37.16 (d, 3JPP = 12.9 Hz, 4P, square base NiPiPr3) 94.20 (p with
3
3
Hg satellites, JPP = 12.9 Hz, JHgP = 1208 Hz, 1P, cap NiPiPr3). Anal.
Calcd for C45H111HgNi5P5 (1298.38): C, 41.53; H, 8.60; Found: C,
41.23; H, 8.74.
Synthesis and Characterization of (η5-Cp)Ni[(iPr3P)Ni]4(μ2-H)4(μ4-
Tl) (3). A solution of [(iPr3P)Ni]5H6 (600 mg, 0.55 mmol) in 15 mL of
toluene was stirred in the presence of an excess of thallium(I)
cyclopentadiene (2.4 g, 8.9 mmol) for 0.25 h. All volatiles were then
removed in vacuo, and the residue was extracted with n-pentane (2 ×
10 mL). The solution was cooled to −40 °C for 16 h, at which point
dark brown block crystals suitable for X-ray diffraction were obtained.
The mother liquor was then concentrated and multiple crops yielded
1
432 mg of 1 (66.1%). H NMR (298 K, C6D6, 500 MHz): δ −24.20
(p, 2JPH = 2.8 Hz, 4H, Ni−H), 1.39 (dd, 3JPH = 12.2 Hz, 3JHH = 7.1 Hz,
3
2
72H, PCHMe2), 2.01 (apparent octet, JHH = 7.2 Hz, JPH = 7.4 Hz,
12H, PCH), 5.25 (s, 5H, Cp). 31P{1H} NMR (298 K, C6D6, 121.5
MHz): δ 31.39 (s, 4P, NiPiPr3). 13C{1H} NMR (298 K, C6D6, 75.5
MHz) δ 20.41 (s, 24 C, PCHMe2) 25.28 (dd, 2JCP = 6.6 Hz, 3JCP = 4.6
Hz, 12C, PCH), 85.31(s, 5C, NiCp) 207Tl{1H} NMR (298 K, C6D6,
288.73 MHz). δ 821.1 (broad s, W1/2 = 796.2 Hz, 1 Tl, Ni4Tl). Anal.
Calcd for C41H93Ni5P4Tl (1204.27): C, 40.77; H, 7.78. Found: C,
40.40; H, 7.87.
Catalytic Site-Selectivity Studies with Fluorobenzene. A solution
of 1 (52 mg, 0.047 mmol), triisopropylphosphine (15 mg 0.093
mmol), C6D6 (450 mg, 4.69 mmol), and fluorobenzene (60 mg, 0.63
mmol) was diluted with 1.3 mL of n-pentane. Deuterium exchange
into fluorobenzene was monitored by 1H, 19F{1H}, and 13C{1H} NMR
early in the reaction. The relative rates of incorporation into the ortho,
meta, and para sites were obtained by modeling peak intensities by
line-fitting using TOPSPIN.
I
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Synthesis and Characterization of (η5-Cp)Ni[(iPr3P)Ni]4H6(μ4-
MgCp) (4). A solution of [(iPr3P)Ni]5H6 (550 mg, 0.50 mmol) in
10 mL of toluene was stirred with 1 equiv of Cp2Mg (77 mg, 0.50
mmol) for 0.25 h. All volatiles were then removed in vacuo, and the
residue was extracted with n-pentane (2 × 10 mL). The solution was
then left at −40 °C for 16 h, at which point dark brown block crystals
suitable for X-ray diffraction were obtained. The mother liquor was
then concentrated and multiple crops yielded 396 mg of 1 (68.2%). 1H
NMR (298 K, C6D6, 500 MHz): δ −22.23 (s, 6H, Ni−H), 1.38 (dd,
pentane, 1-pentene, 1-butene, n-butane, and propane. Both NMR
spectra also revealed quantitative conversion to the known carbide
cluster, [(iPr3P)Ni]5H4(μ5-C).
Variable-Temperature NMR Spectra of 1. A n-pentane solution of
6 (20 mg, 0.018 mmol) was transferred to a J. Young NMR tube.
1
31P{1H} and H NMR spectra were collected at 0, −30, −40, −50,
−60, and −70 °C in attempts to track decoalesence of the single
hydride resonance. WINDNMR was used to model the spectra, and
corresponding rate constants of 2500, 400, 145, 58, 33, and 15 s−1
were determined. Using the Arrhenius equation, an Ea of 8.13 0.70
kcal·mol−1 was calculated.
3
3JPH = 11.1 Hz, JHH = 7.4 Hz, 72H, PCHMe2), 1.78 (apparent octet,
3JHH = 7.4 Hz, 2JHP = 6.9 Hz 12H, PCH), 5.91 (s, 5H, CpNi) 6.78 (s,
5H, CpMg); 31P{1H} NMR (298 K, C6D6, 121.5 MHz): δ 51.5 (s, 4P,
Variable-Temperature NMR Spectra of 5. A n-pentane solution of
5 (50 mg, 0.050 mmol) was transferred to a J. Young NMR tube.
31P{1H} and 1H NMR spectra were collected at −10, −20, −30, −40,
and −60 °C in attempts to see decoalesence of the single resonance.
WINDNMR was used to model the spectra, and corresponding rate
constants of 10 000, 3800, 1200, 98, and 1 s−1 were determined. Using
the Arrhenius equation, an Ea of 21.1 0.48 kcal·mol−1 was calculated.
Variable-Temperature NMR Spectra of 6. A n-pentane solution of
6 (50 mg, 0.051 mmol) was transferred to a J. Young NMR tube.
31P{1H} and 1H NMR spectra were collected at 0, −25, −40, −55, and
−70 °C in attempts to see decoalesence of the single resonance.
WINDNMR was used to model the spectra, and corresponding rate
constants of 30 000, 2600, 280, 45, and 1 s−1 were determined. Using
the Arrhenius equation, an Ea of 15.9 0.49 kcal·mol−1 was calculated.
NiPiPr3). 13C{1H} NMR (298 K, C6H6, 75.5 MHz) δ 20.60 (s, 24 C,
2
PCHMe2) 24.82 (2nd-order multiplet, spin system AXX′2X″, JPC
=
11.1 Hz, 3JPP = 14.2 Hz, 12C, PCH) 85.81(s, 5C, NiCp) 107.33 (s, 5C,
MgCp) Anal. Calcd for C46H100MgNi5P4 (1090.34): C, 50.46; H, 9.24;
Found: C, 50.70; H, 9.23.
Synthesis and Characterization of (η5-Cp)Ni[(iPr3P)Ni]4(μ2-H)4(μ5-
H) (5). A solution of [(iPr3P)Ni]5H6 (550 mg, 0.50 mmol) in 5 mL of
n-pentane was stirred at room temperature in the presence of 1 equiv
of freshly distilled cyclopentadiene (33 mg, 0.50 mmol) for a total of
0.1 h. The dark brown solution was filtered through a plug of Celite
and concentrated by removing volatiles until precipitate was formed.
The solution was then left at −40 °C for 16 h. Dark brown block
crystals suitable for X-ray diffraction were collected. Solid collected
from the first crop and an additional crop from the mother liquor
1
yielded a total of 281 mg of product (56% yield). H NMR (298 K,
ASSOCIATED CONTENT
* Supporting Information
3
■
C6D6, 500 MHz): δ −29.49 (s, 5H, Ni−H), 1.35 (dd, JPH = 10.1 Hz,
S
3JHH = 7.5 Hz, 72H, PCHMe2), 2.04 (broad m, W1/2 = 29.6 Hz, 12H,
PCH), 5.56 (s, 5H, CpNi). 31P{1H} NMR (298 K, C6D6, 121.5 MHz):
δ 48.55 (s, 4P, NiPiPr3). 13C{1H} NMR (298 K, C6H6, 75.5 MHz) δ
20.96 (s, 24 C, PCHMe2) 26.45 (s, 12C, PCH) 84.80(s, 5C, NiCp)
Anal. Calcd for C41H94Ni5P4 (1004.54): C, 49.02; H, 9.47; Found: C,
49.05; H, 9.09.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00763.
Crystallographic data and NMR spectra (PDF)
Synthesis and Characterization of (η5-Cp*)Ni[(iPr3P)Ni]4(μ2-
H)4(μ5-H) (6). A solution of [(iPr3P)Ni]5H6 (600 mg, 0.545 mmol)
in 5 mL of tetrahydrofuran was stirred at room temperature with 5.2
equiv of lithium pentamethylcyclopentadienide (420 mg, 3.0 mmol)
for a total of 4 h. All volatiles were removed from the dark brown
solution, and the precipitate was extracted with 4 × 10 mL of n-
pentane and filtered through a plug of Celite. The solution was then
left at −40 °C for 16 h. Dark brown block crystals suitable for X-ray
diffraction were formed and collected. Precipitate collected from the
first crop along with precipitate from the mother liquor yielded a total
of 322 mg of product (55% yield). 1H NMR (298 K, C63D6, 500 MHz):
Accession Codes
CCDC 1579454−1579457 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-
ing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: + 44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
*Fax: (+1) 519-973-7098. E-mail: sjohnson@uwindsor.ca.
■
3
δ −29.02 (s, 5H, Ni−H), 1.37 (dd, JPH = 12.2 Hz, JHH = 7.4 Hz,
3
2
72H, PCHMe2), 2.06 (apparent octet, JHH = 7.4 Hz, JHP = 7.1 Hz,
12H, PCH), 2.18 (s, 15H, C5Me5); 31P{1H} NMR (298 K, C6D6,
121.5 MHz): δ 46.36 (s, 4P, NiPiPr3). 13C{1H} NMR (298 K, C6H6,
75.5 MHz) 13.02 (s, 5C, C5Me5) δ 20.91 (s, 24 C, PCHMe2) 25.31
(m, 12C, PCH) 92.19(s, 5C, C5Me5) Anal. Calc’d for C46H104Ni5P4
(1074.68): C, 51.41; H, 9.75; Found: C, 51.58; H, 10.09.
ORCID
Samuel A. Johnson: 0000-0001-6856-0416
Notes
The authors declare no competing financial interest.
Reaction of 6 with C6D6. Cluster 6 (5 mg, 0.00466 mmol) was
dissolved in 0.525 mL of C6D6 and transferred to an NMR tube. The
NMR tubes are kept in liquid nitrogen to avoid unmonitored H/D
ACKNOWLEDGMENTS
■
The authors acknowledge the Natural Sciences and Engineering
Research Council (NSERC) of Canada for funding. M.M.S.
acknowledges the Natural Sciences and Engineering Research
Council (NSERC) scholarship support.
1
exchange. The solution was monitored by 31P{1H} and H NMR to
model the hydride resonance intensity over the course of the reaction.
A half-life of 5.62 × 104 s was determined.
Transfer Hydrogenation with 1 and Cyclohexadienes. Cluster 1
(55 mg, 0.05 mmol) was dissolved in 0.6 mL of n-pentane, followed by
either 1,3-cyclohexadiene or 1,4-cyclohexadiene (4 mg, 0.057 mmol).
The solution was monitored by 31P{1H} and 1H NMR, and resonances
consistent with the formation of C6H6 and cyclohexene were observed
immediately, along with the disappearance of resonances of the
corresponding diene.
Reaction of 1 with 1,4-Pentadiene. Cluster 1 (55 mg, 0.05 mmol)
was dissolved in 0.6 mL of C6D6, followed by 1,4-pentadiene (4 mg,
0.059 mmol). The solution was then transferred to an NMR tube and
was studied by 31P{1H}, 13C{1H}, and 1H NMR, which showed
resonances consistent with several alkanes and alkenes including n-
REFERENCES
■
(1) Gates, B. C.; Guczi, L.; Knozinger, H. Metal Clusters in Catalysis;
̈
Elsevier: Amsterdam, 1986; Vol. 29.
(2) Adams, R. D.; Cotton, F. A. Catalysis by Di-and Polynuclear Metal
Cluster Complexes; Wiley-VCH: New York, 1998.
(3) Shima, T.; Hu, S.; Luo, G.; Kang, X.; Luo, Y.; Hou, Z. Science
2013, 340, 1549−1552.
(4) Hu, S.; Shima, T.; Hou, Z. Nature 2014, 512, 413−415.
(5) Nakajima, Y.; Kameo, H.; Suzuki, H. Angew. Chem., Int. Ed. 2006,
45, 950−952.
J
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(6) Inagaki, A.; Takemori, T.; Tanaka, M.; Suzuki, H. Angew. Chem.,
Int. Ed. 2000, 39, 404−406.
(41) Doster, M. E.; Johnson, S. A. Angew. Chem., Int. Ed. 2009, 48,
2185−2187.
(42) Baird, D. A.; Jamal, S.; Johnson, S. A. Organometallics 2017, 36,
1436−1446.
(7) Takemori, T.; Inagaki, A.; Suzuki, H. J. Am. Chem. Soc. 2001, 123,
1762−1763.
(43) Cronin, L.; Higgitt, C. L.; Karch, R.; Perutz, R. N.
Organometallics 1997, 16, 4920−4928.
(8) Araake, R.; Sakadani, K.; Tada, M.; Sakai, Y.; Ohki, Y. J. Am.
Chem. Soc. 2017, 139, 5596−5606.
(44) Braun, T.; Foxon, S. P.; Perutz, R. N.; Walton, P. H. Angew.
Chem., Int. Ed. 1999, 38, 3326−3329.
(9) Beck, R.; Shoshani, M.; Johnson, S. A. Angew. Chem. 2012, 124,
11923−11926.
(45) Schaub, T.; Fischer, P.; Steffen, A.; Braun, T.; Radius, U.; Mix,
A. J. Am. Chem. Soc. 2008, 130, 9304−9317.
(10) Shoshani, M. M.; Johnson, S. A. Inorg. Chem. 2015, 54, 11977−
11985.
(46) Reinhold, M.; McGrady, J. E.; Perutz, R. N. J. Am. Chem. Soc.
2004, 126, 5268−5276.
(11) Shoshani, M. M.; Johnson, S. A. Nat. Chem. 2017, 9, 1282−
1285.
(47) Clot, E.; Eisenstein, O.; Jasim, N.; Macgregor, S. A.; McGrady, J.
E.; Perutz, R. N. Acc. Chem. Res. 2011, 44, 333−348.
(48) Nakao, Y.; Kashihara, N.; Kanyiva, K. S.; Hiyama, T. J. Am.
Chem. Soc. 2008, 130, 16170−16171.
(49) Nova, A.; Reinhold, M.; Perutz, R. N.; Macgregor, S. A.;
McGrady, J. E. Organometallics 2010, 29, 1824−1831.
(50) Johnson, S. A.; Hatnean, J. A.; Doster, M. E. Prog. Inorg. Chem.
2011, 57, 255.
(51) Crawford, E.; Kemball, C. Trans. Faraday Soc. 1962, 58, 2452−
2467.
(52) Heys, J. R. J. Labelled Compd. Radiopharm. 2010, 53, 716−721.
(53) Hirota, K.; Ueda, T. Bull. Chem. Soc. Jpn. 1962, 35, 228−232.
(54) Gould, R. A.; Craighead, K. L.; Wiley, J. S.; Pignolet, L. H. Inorg.
Chem. 1995, 34, 2902−2913.
(55) Bour, J.; Van den Berg, W.; Schlebos, P.; Kanters, R.;
Schoondergang, M.; Bosman, W.; Smits, J.; Beurskens, P.; Steggerda,
J.; Van der Sluis, P. Inorg. Chem. 1990, 29, 2971−2975.
(56) Siedle, A.; Bodner, G.; Garber, A.; Beer, D.; Todd, L. Inorg.
Chem. 1974, 13, 2321−2324.
(57) Fehlner, T. P.; Czech, P. T.; Fenske, R. F. Inorg. Chem. 1990, 29,
3103−3109.
(12) Zhao, Q.; Harris, T. D.; Betley, T. A. J. Am. Chem. Soc. 2011,
133, 8293−8306.
(13) Kanady, J. S.; Tsui, E. Y.; Day, M. W.; Agapie, T. Science 2011,
333, 733−736.
(14) Fout, A. R.; Zhao, Q.; Xiao, D. J.; Betley, T. A. J. Am. Chem. Soc.
2011, 133, 16750−16753.
(15) Murray, L. J.; Weare, W. W.; Shearer, J.; Mitchell, A. D.;
Abboud, K. A. J. Am. Chem. Soc. 2014, 136, 13502−13505.
(16) Di Francesco, G. N.; Gaillard, A.; Ghiviriga, I.; Abboud, K. A.;
Murray, L. J. Inorg. Chem. 2014, 53, 4647−4654.
(17) Murahashi, T.; Inoue, R.; Usui, K.; Ogoshi, S. J. Am. Chem. Soc.
2009, 131, 9888−9889.
(18) Murahashi, T.; Kato, N.; Uemura, T.; Kurosawa, H. Angew.
Chem. 2007, 119, 3579−3582.
(19) Jandl, C.; Pankhurst, J. R.; Love, J. B.; Pothig, A. Organometallics
̈
2017, 36, 2772−2783.
(20) Ishikawa, Y.; Kimura, S.; Yamamoto, K.; Murahashi, T. Chem.
Eur. J. 2017, 23, 14149−14152.
(21) van der Vlugt, J. I. Eur. J. Inorg. Chem. 2012, 2012, 363−375.
(22) Groom, C. R.; Bruno, I. J.; Lightfoot, M. P.; Ward, S. C. Acta
Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 2016, 72, 171−179.
(23) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry;
Wiley: New York, 1988; Vol. 594.
(24) Saito, T. In Early Transition Metal Clusters with π-Donor Ligands;
Chisholm, M. H., Ed.; VCH: New York, 1995; pp 63−164.
(25) Ingleson, M. J.; Mahon, M. F.; Raithby, P. R.; Weller, A. S. J. Am.
Chem. Soc. 2004, 126, 4784−4785.
(26) Gregson, D.; Howard, J. A.; Murray, M.; Spencer, J. L. J. Chem.
Soc., Chem. Commun. 1981, 716−717.
(58) Tolman, C. Chem. Soc. Rev. 1972, 1, 337−353.
(59) Pearson, R. G. J. Am. Chem. Soc. 1969, 91, 4947−4955.
(60) Titova, S.; Bychkov, V.; Domrachev, G.; Razuvaev, G.;
Zakharov, L.; Alexandrov, G.; Struchkov, Y. T. Inorg. Chim. Acta
1981, 50, 71−77.
(61) Yamamoto, Y.; Ehara, K.; Takahashi, K.; Yamazaki, H. Bull.
Chem. Soc. Jpn. 1991, 64, 3376−3383.
(62) Bennett, M. A.; Contel, M.; Hockless, D. C.; Welling, L. L.;
Willis, A. C. Inorg. Chem. 2002, 41, 844−855.
(27) Frost, P. W.; Howard, J. A.; Spencer, J. L.; Turner, D. G.;
Gregson, D. J. Chem. Soc., Chem. Commun. 1981, 1104−1106.
(28) Goodfellow, R. J.; Hamon, E. M.; Howard, J. A.; Spencer, J. L.;
Turner, D. G. J. Chem. Soc., Chem. Commun. 1984, 1604−1606.
(63) Schneider, J. J.; Denninger, U.; Hagen, J.; Kruger, C.; Blaser, D.;
̈
̈
Boese, R. Chem. Ber. 1997, 130, 1433−1440.
(64) Tsunoda, M.; Fleischmann, M.; Jones, J. S.; Bhuvanesh, N.;
Scheer, M.; Gabbaï, F. P. Dalton Trans. 2016, 45, 5045−5051.
(65) Kreye, M.; Freytag, M.; Jones, P. G.; Williard, P. G.;
Bernskoetter, W. H.; Walter, M. D. Chem. Commun. 2015, 51,
2946−2949.
́
(29) Buchowicz, W.; Herbaczynska, B.; Jerzykiewicz, L. B.; Lis, T.;
Pasynkiewicz, S.; Pietrzykowski, A. Inorg. Chem. 2012, 51, 8292−8297.
(30) Shoshani, M. M.; Beck, R.; Wang, X.; McLaughlin, M. J.;
Johnson, S. A. Inorg. Chem. 2017, DOI: 10.1021/acs.inorg-
chem.7b02546.
(66) Fenske, D.; Magull, J. Z. Naturforsch., B: J. Chem. Sci. 1990, 45,
121−126.
(31) Yu, R. P.; Hesk, D.; Rivera, N.; Pelczer, I.; Chirik, P. J. Nature
2016, 529, 195−199.
(67) Gong, J.; Huang, J.; Fanwick, P. E.; Kubiak, C. P. Angew. Chem.,
Int. Ed. Engl. 1990, 29, 396−397.
(32) Palmer, W. N.; Chirik, P. J. ACS Catalysis 2017, 7, 5674−5678.
(33) Burhop, A.; Prohaska, R.; Weck, R.; Atzrodt, J.; Derdau, V. J.
Labelled Compd. Radiopharm. 2017, 60, 343−348.
(34) Jones, W. D. Acc. Chem. Res. 2003, 36, 140−146.
(35) Brown, J. A.; Cochrane, A. R.; Irvine, S.; Kerr, W. J.; Mondal, B.;
Parkinson, J. A.; Paterson, L. C.; Reid, M.; Tuttle, T.; Andersson, S.;
Nilsson, G. N. Adv. Synth. Catal. 2014, 356, 3551−3562.
(36) Chatterjee, B.; Gunanathan, C. Chem. Commun. 2016, 52,
4509−4512.
(68) Gould, R. A.; Craighead, K. L.; Wiley, J. S.; Pignolet, L. H. Inorg.
Chem. 1995, 34, 2902−2913.
(69) Bour, J.; Van den Berg, W.; Schlebos, P.; Kanters, R.;
Schoondergang, M.; Bosman, W.; Smits, J.; Beurskens, P.; Steggerda,
J.; Van der Sluis, P. Inorg. Chem. 1990, 29, 2971−2975.
(70) Emerich, B. M.; Moore, C. E.; Fox, B. J.; Rheingold, A. L.;
Figueroa, J. S. Organometallics 2011, 30, 2598−2608.
(71) Fox, B. J.; Millard, M. D.; DiPasquale, A. G.; Rheingold, A. L.;
Figueroa, J. S. Angew. Chem., Int. Ed. 2009, 48, 3473−3477.
(72) Johnson, M. J.; Gantzel, P. K.; Kubiak, C. P. Organometallics
2002, 21, 3831−3832.
(37) Hatano, M.; Nishimura, T.; Yorimitsu, H. Org. Lett. 2016, 18,
3674−3677.
(38) Atzrodt, J.; Derdau, V.; Fey, T.; Zimmermann, J. Angew. Chem.,
Int. Ed. 2007, 46, 7744−7765.
(73) O’Connor, J. M.; Casey, C. P. Chem. Rev. 1987, 87, 307−318.
(74) Bau, R.; Drabnis, M. H.; Garlaschelli, L.; Klooster, W. T.; Xie,
Z.; Koetzle, T. F.; Martinengo, S. Science 1997, 275, 1099−1102.
(75) Wallick, J.; Riordan, C. G.; Yap, G. P. J. Am. Chem. Soc. 2013,
135, 14972−14974.
(39) Hatnean, J. A.; Johnson, S. A. Organometallics 2012, 31, 1361−
1373.
(40) Johnson, S. A.; Mroz, N. M.; Valdizon, R.; Murray, S.
Organometallics 2011, 30, 441−457.
K
DOI: 10.1021/acs.organomet.7b00763
Organometallics XXXX, XXX, XXX−XXX