C-C activation in the solid state in an organometallic σ-complex

Article abstract of DOI:10.1021/ja2047599

Herein is reported the synthesis, by a solid-state reaction from [Ir(NBD)₂(PⁱPr₃)][BAr^F₄], of the first example of a C-C σ-complex with iridium, [Ir(BINOR-S)(P ⁱPr₃)][BAr^F

Full text of DOI:10.1021/ja2047599

ARTICLE
pubs.acs.org/JACS
CꢀC Activation in the Solid State in an Organometallic σ-Complex
Adrian B. Chaplin, Jennifer C. Green, and Andrew S. Weller*
Inorganic Chemistry Laboratory, Department of Chemistry, University of Oxford, South Parks Road, Oxford OX1 3QR, U.K.
S Supporting Information
b
ABSTRACT: Herein is reported the synthesis, by a solid-state reaction from [Ir(NBD)2-
(PiPr3)][BArF ], of the first example of a CꢀC σ-complex with iridium, [Ir(BINOR-S)-
(PiPr3)][BArF4]. This compound is unique in that in the solid state it undergoes reversible
4
activation of the CꢀC single bond that interacts with the metal center, establishing a
temperature-dependent equilibrium between Ir(III) CꢀC σ/Ir(V) bis-alkyl complexes. This
process has been interrogated by variable-temperature X-ray diffraction, NMR spectroscopy,
and DFT calculations.
Scheme 1. (a) Representation of the Transition-Metal-
’ INTRODUCTION
Mediated Activation of σ-Bonds and (b) Structure of
The coordination chemistry of transition-metal centers with
saturated bonds, such as dihydrogen (HꢀH), alkanes (CꢀH),
silanes (SiꢀH), and boranes (BꢀH), is of significant interest
with regard to their fundamental structures and bonding, activa-
tion, and subsequent utilization in chemical synthesis.1ꢀ3 These
so-called σ-complexes are prototypically defined by the coordi-
nation chemistry of the simplest member of the family, H2, as
[Rh(BINOR-S)(PiPr3)][BArF ] (1)
4
reported in 1984 by Kubas.4 Related intermolecular M HꢀC
3 3 3
bonds, i.e., alkane complexes, are scarce and metastable in
solution,5ꢀ7 while intramolecular agostic M HꢀC complexes
3 3 3
are much better represented.2 The analogous M CꢀC σ-
3 3 3
complexes (intra- or intermolecular) are of considerable inte-
rest with regard to the cleavage and formation of carbonꢀ
carbon bonds using transition metals,8ꢀ12 processes of significant
relevance to organic synthesis and petroleum research. Despite
this, examples of M CꢀC σ-complexes are rare compared to
3 3 3
M
HC, and limited to intramolecular systems,13ꢀ26 as kine-
3 3 3
tically such complexes are disfavored due to the relative inacces-
sibility and orbital directionality associated with the CꢀC bond
compared with CꢀH.8 We recently reported the synthesis of a
rhodium complex with a well-defined intramolecular σ-CꢀC
of the σ-bound ligand occurs or not can be controlled by the
relative position of the transition metal in a particular triad,
among other factors;29 heavier congeners with their energetically
more accessible higher oxidation states can favor the products of
oxidative cleavage. For example, [Rh(η5-C5Me5)(PMe3)H-
interaction, [Rh(BINOR-S)(PiPr3)][BArF ] (1, BINOR-S =
4
1,2,4,5,6,8-dimetheno-S-indacene and ArF = C6H3(CF3)2).27,28
This complex has a {Rh(PiPr3)}+ fragment coordinated to a
saturated organic ligand (BINOR-S) through a metallacyclobu-
tane ring and σ-CꢀC interaction from a cyclopropane unit
(H2)][BArF ] is formulated as a Rh(III) hydride/dihydrogen
4
complex,30 while [Ir(η5-C5Me5)(PMe3)H3][BF4] is an Ir(V)
trihydride.31 Recently, Brookhart, Goldberg, and co-workers
demonstrated similar behavior in methane complexes of the
group 9 {M(2,6-(tBu2PO)2C5H3N)}+ (M = Rh, Ir) fragments.
The rhodium complex is a σ-methane Rh(I) species,7 while the
iridium complex is an Ir(III) hydrido-methyl (albeit in fast self-
exchange via a proposed Ir(I) σ-methane complex).32 As σ-CꢀC
(Scheme 1). There is also a CꢀH Rh agostic interaction.
3 3 3
σ-Complexes are particularly interesting as they represent
intermediates in the activation of strong covalent bonds by
oxidative cleavage (Scheme 1).1,3 For example, in the now classic
study by Kubas, interconverting tautomers of W(PCy3)2-
(CO)3H2 were found to exist: a W(II)ꢀdihydride, the product
of oxidative cleavage, and a W(0)ꢀdihydrogen σ-complex.4
Equilibrium mixtures between σ-alkane and alkyl hydride com-
plexes have also been observed in alkane complexes of the {Re-
(η5-C5H5)(CO)(PF3)} fragment.5 Whether oxidative cleavage
Received: May 24, 2011
Published: July 07, 2011
r
2011 American Chemical Society
13162
dx.doi.org/10.1021/ja2047599 J. Am. Chem. Soc. 2011, 133, 13162–13168
|

Journal of the American Chemical Society
ARTICLE
Scheme 2. Formation of [Ir(BINOR-S)(PiPr3)][BArF ] (3)
4
complexes are rare, such an isomeric pair has yet to be observed
for analogous carbonꢀcarbon bonds.
The existence of such isomers is suggested on the basis of
experimental and computational investigations on 1.27,28 Using
NMR spectroscopy, it was shown that this Rh(III) complex
undergoes rapid reversible CꢀC oxidative cleavage/reductive
coupling on the NMR time scale in solution. The calculated
mechanism invoked a putative Rh(V) alkyl intermediate: the
product of activation of the σ-CꢀC bond. Although this dynamic
process can be arrested at low temperature in solution,27,28
allowing the barrier to be determined, the key Rh(V) inter-
mediate remains elusive. In order to realize the synthesis of such
an intermediate, we targeted the synthesis of the Ir congener to
1, hoping to exploit the more accessible (V) oxidation state of
iridium compared to rhodium. To this end, we now report the
synthesis of [Ir(BINOR-S)(PiPr3)][BArF ] (3), in which both
4
the Ir(III) σ-complex and Ir(V) oxidative cleavage products
coexist in the solid state. We find that these two complexes
are in equilibrium with one another, and the position of the
equilibrium can be adjusted by changing the temperature.
Although reversible CꢀC activation has been reported
previously28,33ꢀ35 and σ-bound complexes have been impli-
cated in such processes,28,36,37 the observation of intimately
connected σ and oxidative cleavage product complexes is, we
believe, unknown.
Figure 1. Molecular structures of 2 (a) and 3 (b): ORTEP plots with
thermal ellipsoids drawn at the 50% probability level. Data were
collected at 150 K. The structure of 3 presented here is refined without
cation disorder modeling (see text). Anions and solvent molecules are
omitted for clarity. Selected bond distances (Å) for 2: Ir1ꢀP1,
2.4768(8); Ir1ꢀC1, 2.190(4); Ir1ꢀC6, 2.169(4); Ir1ꢀC3, 2.275(3);
Ir1ꢀC4, 2.250(4); Ir1ꢀC8, 2.254(3); Ir1ꢀC13, 2.237(4); Ir1ꢀC10,
2.225(3); Ir1ꢀC11, 2.192(3). Selected parameters for 3 are listed in
Table 1.
’ RESULTS AND DISCUSSION
The precursor complex [Ir(NBD)2(PiPr3)][BArF ] (2,
in CD2Cl2 solution it proceeds slowly to form [Ir(BINOR-S)-
4
(PiPr3)][BArF ] (3) alongside decomposition to unidentified
NBD = 2,5-norbornadiene) was prepared by reaction of excess
4
NBD and Na[BArF ] with [Ir(COE)(PiPr3)Cl]2 (COE = cyclo-
products (see Supporting Information). Complex 3 is itself un-
stable in solution at room temperature (half-life of 5.3 ( 0.2 h),
decomposing to give multiple products including free BINOR-S.
This stability profile is similar to that of the Rh congener 1, which
decomposes in CH2Cl2 over 24 h. Thus, although the formation of
3 can be observed by NMR spectroscopy, its low stability means
that its isolation as a pure material using solution techniques is
problematic. To circumvent this, we carried out the synthesis of
3 in the solid state by reaction of 2 at 40 °C to afford pure 3
after 20 h in quantitative yield. Monitoring this reaction over a
range of temperatures using NMR spectroscopy (samples ana-
lyzed in CD2Cl2 solution, see Supporting Information) showed
a reaction profile that fitted zero-order reaction kinetics39 with
activation parameters of ΔHq = 106.2 ( 1.1 kJ molꢀ1 and
4
octene) in CH2Cl2, as shown in Scheme 2. By comparison, the
Rh congener of 2 cannot be observed, as it presumably proceeds
immediately to form 1.27 Complex 2 is isolated as an air- and
temperature-sensitive crystalline material. The solid-state struc-
ture of the cation, which is best described as having a distorted
square-based pyramidal geometry, is shown in Figure 1. The
structure of a PMe3 analogue of 2, [Ir(NBD)2(PMe3)]+ (20), has
recently been calculated and shows very similar structural
metrics.38 The overall barrier to dimerization of the coordinated
NBD units to form BINOR-S in 20 was also calculated and was
shown to be 123.9 kJ molꢀ1, significantly higher than that for the
rhodium analogue (92.4 kJ molꢀ1) and consistent with our
experimental observations. Although complex 2 can be isolated,
13163
dx.doi.org/10.1021/ja2047599 |J. Am. Chem. Soc. 2011, 133, 13162–13168

Journal of the American Chemical Society
ARTICLE
Table 1. Selected Distances (Å) and Angles (°) for the Structures of 1 and 3 at Different Temperaturesa
1
3
100 K
250 K
100 K
250 K
150 K
cation structure
MꢀP1
ordered
2.2711(5)
2.368(2)
2.356(2)
1.610(3)
64.32(15)
2.043(2)
2.036(2)
91.80(16)
2.906(2)
ordered
2.2718(10)
2.369(4)
2.359(4)
1.609(7)
64.4(3)
disordered, not modeled
2.2884(8)
2.305(3)
disordered, not modeled
2.3545(15)
2.220(6)
disordered, two-component model (3a, 3b)
2.3191(9)
MꢀC8
2.26(2),b 2.27(4)c
2.284(9),b 2.18(2)c
1.645(7),b,d 2.069(8)c,d
66.6(6),b 82.7(14)c
2.057(4)
MꢀC10
2.294(3)
2.206(8)
C8ꢀC10
C8ꢀC9ꢀC10
MꢀC1
1.704(5)
1.833(13)
76.3(7)
68.4(2)
2.033(4)
2.028(4)
92.3(3)
2.061(3)
2.046(7)
MꢀC3
C1ꢀC2ꢀC3
MꢀC16
2.050(3)
2.043(7)
2.045(4)
92.4(3)
92.3(6)
92.3(3)
2.921(5)
2.955(4)
3.004(9)
2.970(5)
a See Supporting Information for complete listing of structural parameters. b Parameters associated with disordered component 3a only. c Parameters
associated with disordered component 3b only. d These parameters are restrained directly (see main text).
Figure 2. Disorder models for the molecular structure of 3: ball-and-stick representations of the individual disordered components 3a (a) and 3b
(b), the combined model (c), and an ORTEP plot (thermal ellipsoids drawn at the 50% probability level) of 3 refined without cation disorder modeling
(d). Anions and hydrogen atoms are omitted for clarity. Data were collected at 150 K.
ΔSq = +6 ( 4 J molꢀ1 Kꢀ1. ΔGq(298 K) = 104 ( 2 kJ molꢀ1
,
and calculations suggest a Cs-symmetric Rh(V) intermediate.28
For 3, the barrier to this process must be considerably lower. The
changes with temperature observed by NMR spectroscopy for 3
suggest a change in population between chemically distinct
species that are in rapid equilibrium with one another.
The solid-state structure of 3 was determined by X-ray
diffraction over a wide temperature range (100ꢀ250 K; see
Figure 1 for the 150 K structure), and the key structural metrics
that arise from these experiments are presented in Table 1.
Analysis of these data suggests an apparent lengthening of the σ-
CꢀC distance with temperature [C8ꢀC10, 1.704(5) Å at 100 K,
1.833(13) Å at 250 K], alongside a concomitant apparent
lengthening of the Ir1ꢀP1 bond [2.2884(8) Å at 100 K,
2.3545(15) Å at 250 K]. The metrics associated with the
comparing well with that calculated for the equivalent process
involving 20 (123.6 kJ molꢀ1).38 Crystals suitable for an X-ray
diffraction experiment were grown at low temperature (243 K,
CH2Cl2/pentane) inan argon-filled glovebox. Inthe solidstate 3 is
extremely sensitive and is unstable outside a purified Ar or vacuum
atmosphere, decomposing rapidly. The dimerization of NBD on
group 9 metals has been reported previously.40ꢀ42
Solution 31P{1H} NMR data for 3 show a single, sharp, signal
over the temperature range from 298 to 200 K (CD2Cl2) that
moves significantly to lower frequency on cooling (δ 47.2 at 298
K; δ 36.2 at 200 K). By contrast, 1 shifts by only 1.8 ppm over
the same temperature range.27 For 3, local C2v symmetry is
1
observed for the BINOR-S ligand by H and 13C{1H} NMR
spectroscopy, even on cooling to 200 K. As for the 31P{1H}
NMR spectrum, changes in chemical shift with temperature are
also observed. These NMR data suggest a rapid, time-averaged,
fluxional process that makes the two sides of the BINOR-S ligand
equivalent, by movement of the phosphine from one side of the
molecule to the other, as in the solid state the ligand has only Cs
symmetry. For 1, we have previously shown that this process can
be frozen out (200 K, ΔGq(298 K) = 44.8 ( 4 kJ molꢀ1) to reveal
solution NMR data fully consistent with the solid-state structure,
metallacyclobutane (Ir,C1,C2,C3) and CꢀH Ir agostic inter-
3 3 3
action (Ir1ꢀC16) do not change significantly. These data
suggest increasing activation of the CꢀC σ-bond with tempera-
ture and that these structural changes are localized primarily to
one part of the molecule. Repeating these measurements on a
different crystalline sample, using a different diffractometer, and
cycling the temperature (cooling from 250 to 100 K, followed by
warming to 250 K) on the same sample all gave the same results.
13164
dx.doi.org/10.1021/ja2047599 |J. Am. Chem. Soc. 2011, 133, 13162–13168

Journal of the American Chemical Society
ARTICLE
The behavior observed for 3 in the solid state is rationalized
using a crystallographic disorder model involving a mixture of
two cationic components in the crystal lattice (Figure 2), the
relative ratio of which changes with temperature, resulting in
apparent changes in the C8ꢀC10, Ir1ꢀC8, Ir1ꢀC10, and
Ir1ꢀP1 distances. The main disordered component is an Ir(III)
σ-CꢀC complex, 3a, that is isostructural with its Rh congener 1.
In this structure there is a metallacyclobutane (Ir,C1,C2,C3) and
a cyclopropane moiety (C8A,C9A,C10A) that adopts a σ-CꢀC
interaction with the metal center. The other component, 3b, has
two metallacyclobutane units (Ir1,C1,C2,C3 and Ir1,C8B,C9B,
C10B) and a formal Ir(V) oxidation state. The phosphine ligand
and supporting CꢀH Ir agostic interaction (Ir1ꢀC16) are
3 3 3
common to both components. These two isomers, 3a and 3b,
were modeled by restraining the C8ꢀC10 distance in 3a and 3b
to 1.64 and 2.07 Å, respectively, distances that were suggested by
DFT calculations (see below) and are crystallographically and
chemically sensible for an activated cyclopropane and a metalla-
cyclobutane respectively.27,28,37 Similar progressive changes in
bond length with temperature that are modeled using a tem-
perature-dependent population change between two disordered
isomers have been observed for Feꢀcyanide spin-crossover
complexes, where the FeꢀN bonds show a systematic increase
with increasing temperature due to a change in the weighted
average of two different FeꢀN environments, in this case on
moving from a low-spin to a high-spin electronic configuration.43
Data collections on 1, which notably crystallizes in the same
space group as 3, were repeated over the same temperature range.
No statistically significant variation in any of the measured
distances was found (Table 1). The unit cell volume changed
by only 3.5% for both 1 and 3 over this temperature range. This
demonstrates that the structural changes observed are due to the
replacement of Rh (4d metal) with Ir (5d metal) on moving from
1 to 3.
This structural change between 3a and 3b is thus formulated
as a CꢀC oxidative cleavage of a σ-bound cyclopropane ring.
Such a change from a σ-interaction (3a) to two IrꢀC covalent
bonds (3b) would be expected to have a pronounced effect on
the trans-disposed phosphine group, owing to the large trans-
influence of a bis-alkyl compared to a σ-CꢀC unit. Although the
heavy phosphorus atom was not explicitly split in the disorder
model this change is manifested in the apparent lengthening of
the measured Ir1ꢀP1 distance with temperature associated with
the increase in the relative population of the Ir(V) isomer
(Figure 2, Table 1).
Figure 3. Solid-state dynamic behavior of 3: reaction scheme demon-
strating the structural change between the tautomers of 3 (a), popula-
tions of the tautomers at different temperatures determined using X-ray
diffraction (b), and Van’t Hoff plot for this process (c) (R2(fit) = 0.999).
deviations from rigid-bond behavior for the C8ꢀC10 and
Ir1ꢀP1 bonds when the cation of 3 is refined without using this
model, indicating the presence of disorder (Hirshfield rigid-bond
test, see Supporting Information).48 In contrast, the bonds not
significantly involved in the structural change (i.e., Ir1ꢀC1,
Ir1ꢀC3) do not show any similar signs of disorder. Importantly,
no evidence of disorder is found across the complete temperature
range for 1, with C8ꢀC10 and Rh1ꢀP1 bonds notably showing
good rigid-bond behavior. We are thus confident that this
variation in C8ꢀC10 and Ir1ꢀP1 bond distances with tempera-
ture in 3 is a consequence of a reversible intramolecular change in
bonding and that this does not happen to any significant degree
for the Rh analogue.
As X-ray diffraction gives only the average of all possible
atomic displacements in the entire crystal on the time scale of the
X-ray experiment (i.e., hours),44 to probe this dynamic change
further we have obtained the solid-state 31P{1H} MASꢀNMR
spectra of 1 and 3 over a variable-temperature range (298 to 213
K); which gives information on the NMR time scale (seconds)
regarding local atomic environments. For 3, a single peak is
observed that progressively shifts from δ 35.2 (298 K) to δ 29.9
(213 K), mirroring that observation in solution. The signal
broadens upon cooling (fwhm 260 vs 320 Hz). These observa-
tions (albeit not covering the same range as for the X-ray
diffraction experiments) are indicative of a weighted average
chemical shift of two chemically distant species that are in
dynamic equilibrium with one another and suggest that the
barrier to interconversion in the solid state between 3a and 3b is
low, as might be expected for such a simple process that does not
involve major atomic reorganization. These changes are rever-
sible upon warming to 298 K, as for the X-ray experiments.
Higher temperatures resulted in the onset of decomposition. For
1, a single peak is observed at δ 46.7 [J(RhP) = 208 Hz] that, as
in solution, shifts by a significantly lesser degree upon cooling
to 213 K compared to 3, to δ 44.5 [J(RhP) = 220 Hz]. At
room temperature, the signal becomes broader (fwhm 80 vs
140 Hz). These results are consistent with a higher barrier to
That the relative populations of the two isomers 3a and 3b
change reversibly with temperature shows that there is an
equilibrium between the two disordered components. This is
similar to the dynamic disorder44 reported in the solid state for
mixed M3(CO)12 clusters (M = Fe, Ru, Os).45,46 In the solid state
at 100 K, the 3a:3b ratio is 83(2):17(2), which changes at 250 K
to 55(2):45(2). A Van’t Hoff plot (Figure 3) clearly demon-
strates the change in the position of the equilibrium with
temperature and yields the following thermodynamic para-
meters: ΔH = 1.93 ( 0.03 kJ molꢀ1, ΔS = +6.2 ( 0.2 J molꢀ1
K
ꢀ1, and ΔG(298 K) = +0.09 ( 0.08 kJ molꢀ1. These data show
the Ir(III) and Ir(V) complexes to be energetically very similar to
one another. A similar analysis has been used to map the
rotational progression of nitrosyl ligands in Feꢀporphyrin com-
plexes with temperature in the solid state.47
Further support for the description of the solid-state structure
of 3 using a two-component model comes from significant
13165
dx.doi.org/10.1021/ja2047599 |J. Am. Chem. Soc. 2011, 133, 13162–13168

Journal of the American Chemical Society
ARTICLE
showed a point of inflection at long CꢀC distances in a steadily
rising curve, which is related to the Rh(V) transition state for the
fluxional process for the BINOR-S ligand in solution.28 There is
also a strong correlation of the C8ꢀC10 distance with the changes
in Ir1ꢀP1, Ir1ꢀC8, and Ir1ꢀC10 bondlengths for 3. As the CꢀC
distance increases, the phosphine is also found to tilt slightly to one
side. All these trends are fully consistent with the structural
variation observed by X-ray diffraction at different temperatures,
and the steeper potential energy curve associated with the CꢀC
distance in 1 underscores the observation that the Rh(III)
oxidation state is preferred in the solid state, reflecting the relative
inaccessibility of the Rh(V) oxidation state.51 Examination of the
frontier orbitals (Supporting Information) of 3a and 3b showed
that, while 3a has three occupied orbitals principally of d character
and two unoccupied orbitals, for 3b the reverse is true, with two d
orbitals being occupied in this case and three unoccupied.
’ CONCLUSIONS
We report here experimental and computational evidence that
an Ir(III) σ-complex and an Ir(V) oxidative cleavage product
coexist in the solid state. These two complexes are in equilibrium
with one another, with the position of the equilibrium can be
adjusted by changing the temperature. Related temperature-
dependent behavior of isomeric (tautomeric) forms of bound
σ complexes has previously been observed in solution for
dihydrogen complexes. Pons and Heinekey reported a marked
temperature dependence of J(HD) in isotopomers of [Ir(dmpm)-
(η5-C5Me5)H2][B(C6F5)4]2 (dmpm = bis(dimethylphosphino)-
methane), attributed to a rapidly established equilibrium be-
tween an Ir(III) dihydrogen complex and an Ir(V) cis-dihydride
tautomer. That higher temperatures resulted in increased J(HD)
values was interpreted as corresponding to an increased popula-
tion of the dihydrogen isomer,52,53 which was supported
by computational work.54ꢀ56 Saliently, the Rh congener [Rh-
(dmpm)(η5-C5Me5)H2][B(C6F5)4]2 is formulated as a conven-
tional Rh(III) dihydrogen analogue.54 The similarity with the
CꢀC σ-complexes 1 and 3 is thus striking and further strength-
ens our interpretation that both an Ir(III) σ-CꢀC complex and
the corresponding Ir(V) CꢀC oxidative cleavage product are
present in the solid state for 3 but not for 1. If we take this view, 3
completes the series for σ coordination and reversible activation
of the most important elementꢀelement σ-bonds with transition
metal centers: HꢀH, CꢀH, and CꢀC.
Figure 4. Calculated potential energy surface for CꢀC bond activation
in 1 and 3: SCF energy with changes in C8ꢀC10 bond distance for 1 and
3 and correlations between the C8ꢀC10, Ir1ꢀP1, Ir1ꢀC8, and
Ir1ꢀC10 bond distances during this process.
interconversion in 1 than in 3. Unlike in solution, where the
phosphine ligand would be free to move during the fluxional
process, making each side of the BINOR-S fragment equivalent on
the NMR time scale (i.e., time-averaged C2v symmetry), in the
solid state thisis not possible due to the constraints imposed bythe
lattice, and the molecule would retain approximate Cs symmetry.
DFT calculations49 were able to identify two minima for 3, one
with a C8ꢀC10 bond length of 1.64 Å (3a) and the other 2.07 Å
(3b), corresponding to, and fully consistent with, the inferred
structures of 3a (Ir(III)) and 3b (Ir(V)). Similarly, the calculated
structure of 1 shows excellent agreement with the observed
(static) structure of 1 in the solid state [e.g., C8ꢀC10, 1.61 Å
(calc) versus 1.610(3) Å (100 K)]. That the calculated CꢀC
sigma bond in 3a is longer than in 1 (1.64 versus 1.61 Å,
respectively) suggests a greater degree of CꢀC activation in 3a
and underscores that these structural changes are due to the
move from a 4d to a 5d metal. It also parallels the activation, but
not scission, of dihydrogen when bound to different metal
centers in stretched dihydrogen complexes.1 The SCF energy
of 3a was found to be 4 kJ molꢀ1 higher than that of 3b with an
enthalpy change ΔH0(298 K) = ꢀ10 kJ molꢀ1, in satisfactory
agreement with the small energy difference found experimen-
tally, though of the wrong sign. A transition state was identified
between the two species, with activation free energies of 4 kJ
molꢀ1 from 3a and 11 kJ molꢀ1 from 3b.50 This low barrier is
consistent with the results of the solid-state NMR experiments.
The C8ꢀC10 distance in the transition state was 1.82 Å. A scan
of the SCF energy with changing C8ꢀC10 bond length is shown
in Figure 4; two minima are clearly evident. For 1, a similar scan
’ EXPERIMENTAL SECTION
[Ir(NBD)2(PiPr3)][BArF ] (2). To a Schlenk flask charged with
4
[Ir(COE)(PiPr3)Cl]257 (0.200 g, 0.201 mmol) and Na[BArF ] (0.365 g,
4
0.412 mmol) was added a solution of NBD (0.2 mL, 2 mmol) in CH2Cl2
(5 mL). The resulting suspension was stirred at room temperature for 30
min and then filtered. The crude product was precipitated by addition of
excess hexane and isolated by filtration. Two successive recrystallizations
from CH2Cl2/hexane gave the pure product as a cream-white micro-
crystalline solid. Yield = 0.350 g (62%). The solid product was stored
under argon at ꢀ32 °C.
Data for 2: 1H NMR (CD2Cl2, 500 MHz, 298 K, using X-ray labeling
scheme) δ 7.70ꢀ7.74 (m, 8H, ArF), 7.56 (br, 4H, ArF), 4.29 (br, 2H,
H
2/9), 4.14 (app q, J = 4 Hz, 4H, H1/3/8/10), 3.55 (br, 2H, H5/12), 3.48
3
(dt, JPH = 7.2 Hz, 3JHH = 3.8 Hz, 4H, H4/6/11/13), 3.00 (app oct, J =
7 Hz, PCH), 1.37 (dd, 3JPH = 13.5 Hz, 3JHH = 7.2 Hz, Me), 0.96 (br m,
4H, H7/14); 31P{1H} NMR (CD2Cl2, 202 MHz, 298 K) δ ꢀ0.7 (s);
13C{1H} NMR (CD2Cl2, 126 MHz, 298 K) δ 162.3 (q, 1JBC = 50 Hz,
13166
dx.doi.org/10.1021/ja2047599 |J. Am. Chem. Soc. 2011, 133, 13162–13168

Journal of the American Chemical Society
ARTICLE
ArF), 135.4 (s, ArF), 129.4 (qq, 2JFC = 32 Hz, 3JBC = 3 Hz, ArF), 125.2 (q,
1JFC = 272 Hz, ArF), 118.0 (sept, 3JFC = 4 Hz, ArF), 66.8 (d, 4JPC = 1 Hz,
C7/14), 47.9 (d, 3JPC = 2 Hz, C5/12), 45.9 (d, 2JPC = 6 Hz, C1/3/8/10), 45.4
(3) Kubas, G. J. Chem. Rev. 2007, 107, 4152.
(4) Kubas, G. J.; Unksefer, C. J.; Swanson, B. I.; Fukushima, E. J. Am.
Chem. Soc. 1986, 108, 7000.
(5) Ball, G. E.; Brookes, C. M.; Cowan, A. J.; Darwish, T. A.; George,
M. W.; Kawanami, H. K.; Portius, P.; Rourke, J. P. Proc. Natl. Acad. Sci. U.
S.A. 2007, 104, 6927.
(6) Cowan, A. J.; Portius, P.; Kawanami, H. K.; Jina, O. S.; Grills,
D. C.; Sun, X. Z.; McMaster, J.; George, M. W. Proc. Natl. Acad. Sci. U.S.
A. 2007, 104, 6933.
(7) Bernskoetter, W. H.; Schauer, C. K.; Goldberg, K. I.; Brookhart,
M. Science 2009, 326, 553.
(8) Rybtchinski, B.; Milstein, D. Angew. Chem., Int. Ed. 1999, 38, 870.
(9) Jun, C. H. Chem. Soc. Rev. 2004, 33, 610.
(10) Jazzar, R. F. R.; Macgregor, S. A.; Mahon, M. F.; Richards, S. P.;
Whittlesey, M. K. J. Am. Chem. Soc. 2002, 124, 4944.
(11) Sattler, A.; Parkin, G. Nature 2010, 463, 523.
(12) Gandelman, M.; Vigalok, A.; Konstantinovski, L.; Milstein, D.
J. Am. Chem. Soc. 2000, 122, 9848.
(13) Goldfuss, B.; Schleyer, P. V.; Hampel, F. J. Am. Chem. Soc. 1996,
118, 12183.
(14) Tomaszewski, R.; Hyla-Kryspin, I.; Mayne, C. L.; Arif, A. M.;
Gleiter, R.; Ernst, R. D. J. Am. Chem. Soc. 1998, 120, 2959.
(15) Vigalok, A.; Rybtchinski, B.; Shimon, L. J. W.; Ben-David, Y.;
Milstein, D. Organometallics 1999, 18, 895.
(16) Gandelman, M.; Konstantinovski, L.; Rozenberg, H.; Milstein,
D. Chem.—Eur. J. 2003, 9, 2595.
(17) Jaffart, J.; Cole, M. L.; Etienne, M.; Reinhold, M.; McGrady,
J. E.; Maseras, F. Dalton Trans. 2003, 4057.
(18) Summerscales, O. T.; Cloke, F. G. N.; Hitchcock, P. B.; Green,
J. C.; Hazari, N. Science 2006, 311, 829.
(19) Harvey, B. G.; Mayne, C. L.; Arif, A. M.; Ernst, R. D. J. Am.
Chem. Soc. 2005, 127, 16426.
(20) Madison, B. L.; Thyme, S. B.; Keene, S.; Williams, B. S. J. Am.
Chem. Soc. 2007, 129, 9538.
(d, 3JPC = 2 Hz, C2/9), 42.4 (d, 2JPC = 4 Hz, C4/6/11/13), 29.3 (d, 1JPC
=
20 Hz, PCH), 20.3 (s, Me).
[Ir(BINOR-S)(PiPr3)][BArF ] (3). A solid sample of 2 (0.080 g,
4
0.057 mmol) was heated at 40 °C under argon for 20 h, resulting in
quantitative formation of the product as an orange solid.
Data for 3 (higher temperature): 1H NMR (CD2Cl2, 500 MHz, 298
K, using X-ray labeling scheme) δ 7.70ꢀ7.74 (m, 8H, ArF), 7.56 (br, 4H,
ArF), 3.97 (br app t, J = 3 Hz, 4H, H1/3/8/10), 3.53 (br app t, J = 5 Hz, 2H,
H
H
2/9), 2.70 (second-order app oct, J ≈ 7 Hz, 3H, PCH), 2.30 (br, 2H,
5/12), 1.62 (br, 4H, H4/6/11/13), 1.33 (obscured app t, J = 2 Hz, 4H, H7/
14), 1.32 (dd, JPH = 14.5 Hz, JHH = 7.2 Hz, Me); 31P{1H} NMR
(CD2Cl2, 500 MHz, 298 K) 47.2 (s); 13C{1H} NMR (CD2Cl2, 126
MHz, 298 K) δ 162.3 (q, 1JBC = 50 Hz, ArF), 135.4 (s, ArF), 129.4 (qq,
2JFC = 32 Hz, 3JBC = 3 Hz, ArF), 125.2 (q, 1JFC = 272 Hz, ArF), 118.0
(sept, 3JFC = 4 Hz, ArF), 51.9 (s, C2/9), 46.9 (s, C4/6/11/13), 35.3 (d, 4JPC
= 3 Hz, C7/14), 34.1 (s, C5/12), 29.1 (d, 2JPC = 8 Hz, C1/3/8/10), 24.3 (d,
1JPC = 22 Hz, PCH), 20.4 (s, Me).
3
3
Data for 3 (lower temperature): 1H NMR (CD2Cl2, 500 MHz, 200
K) δ 7.69ꢀ7.74 (m, 8H, ArF), 7.53 (br, 4H, ArF), 3.74 (br app t, J = 4 Hz,
4H, H1/3/8/10), 3.24 (br app t, J = 5 Hz, 2H, H2/9), 2.53 (app oct, J = 8
Hz, 3H, PCH), 2.14 (br, 2H, H5/12), 1.55 (br, 4H, H4/6/11/13), 1.24 (br,
4H, H7/14), 1.20 (dd, JPH = 14.4 Hz, JHH = 7.2 Hz, Me); 31P{1H}
NMR (CD2Cl2, 202 MHz, 200 K) δ 36.2 (s); 13C NMR (CD2Cl2, 126
MHz, 200 K, selected data from HSQC) δ 49.7 (1JCH = 160 Hz, C2/9),
44.1 (1JCH = 136 Hz, C4/6/11/13), 34.7 (1JCH = 134 Hz, C7/14), 33.4
3
3
(1JCH = 148 Hz, C5/12), 26.0 (1JCH = 163 Hz, C1/3/8/10), 23.0 (1JCH
128 Hz, PCH), 19.2 (1JCH = 126 Hz, Me).
=
Anal. Calcd for C55H49BF24IrP (1399.96 g molꢀ1): C, 47.19; H, 3.53.
Found: C, 47.57; H, 3.63.
(21) Boulho, C.; Keys, T.; Coppel, Y.; Vendier, L.; Etienne, M.;
Locati, A.; Bessac, F.; Maseras, F.; Pantazis, D. A.; McGrady, J. E.
Organometallics 2009, 28, 940.
’ ASSOCIATED CONTENT
(22) Scheins, S.; Messerschmidt, M.; Gembicky, M.; Pitak, M.;
Volkov, A.; Coppens, P.; Harvey, B. G.; Turpin, G. C.; Arif, A. M.;
Ernst, R. D. J. Am. Chem. Soc. 2009, 131, 6154.
(23) Harvey, B. G.; Arif, A. M.; Ernst, R. D. J. Organomet. Chem.
2006, 691, 5211.
(24) Boulho, C. d.; Keys, T.; Coppel, Y.; Vendier, L.; Etienne, M.;
Locati, A.; Bessac, F.; Maseras, F.; Pantazis, D. A.; McGrady, J. E.
Organometallics 2009, 28, 940.
(25) An Ir(I) complex initially described as having a CꢀC σ
interaction has since been re-evaluted as simply having relatively close
nonbonding contacts but no lengthening of the CꢀC bond: Maseras, F;
Crabtree, R. H. Inorg. Chim. Acta 2004, 357, 345.
(26) An example of a SiꢀSi σ-complex has also been reported:
Chen, W.; Shimada, S.; Tanaka, M. Science 2002, 295, 308. Nikonov, G. I.
Angew. Chem., Int. Ed. 2003, 42, 1335.
S
Supporting Information. Full synthetic, crystallographic
b
characterization, and computational details (1H, 13C, 31P solu-
tion and solid-state NMR spectroscopy and details of the crystal-
lography on 1 and 3). This material is available free of charge via
the Internet at http://pubs.acs.org. Crystallographic data have
also been deposited with the Cambridge Crystallographic Data
Centre under CCDC 802171ꢀ802186. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
’ AUTHOR INFORMATION
Corresponding Author
andrew.weller@chem.ox.ac.uk
(27) Brayshaw, S. K.; Green, J. C.; Kociok-Kohn, G.; Sceats, E. L.;
Weller, A. S. Angew. Chem., Int. Ed. 2006, 45, 452.
(28) Brayshaw, S. K.; Sceats, E. L.; Green, J. C.; Weller, A. S. Proc.
Natl. Acad. Sci. U.S.A. 2007, 104, 6921.
’ ACKNOWLEDGMENT
We thank Prof. Judith Howard and Dr. Hazel Sparkes
(University of Durham, UK), Dr. David Watkin, and Dr. Amber
Thompson for discussions regarding the crystallography, the
EPSRC solid-state NMR service (Durham, UK), and in parti-
cular Dr. David Apperley.
(29) Kubas, G. J. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 6901.
(30) Taw, F. L.; Mellows, H.; White, P. S.; Hollander, F. J.; Bergman,
R. G.; Brookhart, M.; Heinekey, D. M. J. Am. Chem. Soc. 2002, 124,
5100.
(31) Gilbert, T. M.; Bergman, R. G. J. Am. Chem. Soc. 1985,
107, 3502.
(32) Bernskoetter, W. H.; Hanson, S. K.; Buzak, S. K.; Davis, Z.;
White, P. S.; Swartz, R.; Goldberg, K. I.; Brookhart, M. J. Am. Chem. Soc.
2009, 131, 8603.
(33) Gunay, A.; Jones, W. D. J. Am. Chem. Soc. 2007, 129, 8729.
(34) Escudero, D.; Weisheit, T.; Weigand, W.; Gonzalez, L. Dalton
Trans. 2010, 39, 9505.
’ REFERENCES
(1) Kubas, G. J. Metal Dihydrogen and σ-Bond Complexes; Kluwer
Academic/Plenum Publishers: New York, 2001.
(2) Brookhart, M.; Green, M. L. H.; Parkin, G. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 6908.
13167
dx.doi.org/10.1021/ja2047599 |J. Am. Chem. Soc. 2011, 133, 13162–13168

Journal of the American Chemical Society
ARTICLE
(35) Monreal, M. J.; Diaconescu, P. L. J. Am. Chem. Soc. 2010, 132,
7676.
(36) Evans, M. E.; Li, T.; Jones, W. D. J. Am. Chem. Soc. 2010, 132,
16278.
(37) Periana, R. A.; Bergman, R. G. J. Am. Chem. Soc. 1986, 108,
7346.
(38) Wu, Y.; Jin, L.; Xue, Y.; Lee, I. M.; Kim, C. K. J. Comput. Chem.
2010, 31, 2248.
(39) Khawam, A.; Flanagan, D. R. J. Phys. Chem. B. 2006, 110, 17315.
(40) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1971, 93, 3089.
(41) Acton, N.; Roth, R. J.; Maier, C. A.; Katz, T. J.; Frank, J. K.; Paul,
I. C. J. Am. Chem. Soc. 1972, 94, 5446.
(42) Bezman, S. A.; Bird, P. H.; Fraser, A. R.; Osborn, J. A. Inorg.
Chem. 1980, 19, 3755.
(43) Shatruk, M.; Dragulescu-Andrasi, A.; Chambers, K. E.; Stoian,
S. A.; Bominaar, E. L.; Achim, C.; Dunbar, K. R. J. Am. Chem. Soc. 2007,
129, 6104.
(44) Braga, D. Chem. Rev. 1992, 92, 633.
(45) Farrugia, L. J.; Senior, A. R. M.; Braga, D.; Grepioni, F.; Orpen,
A. G.; Crossley, J. G. J. Chem. Soc., Dalton Trans. 1996, 631.
(46) Farrugia, L. J.; Gillon, A. L.; Braga, D.; Grepioni, F. Organome-
tallics 1999, 18, 5022.
(47) Silvernail, N. J.; Barabanschikov, A.; Sage, J. T.; Noll, B. C.;
Scheidt, W. R. J. Am. Chem. Soc. 2009, 131, 2131.
(48) Rosenfield, R.; Tureblood, K.; Dunitz, J. Acta Crystallogr. 1978,
A34, 828.
(49) DFT calculations were carried out using the Gaussian suite with
the B3LYP functional. A GEN basis set employed with 6-31G(d,p) for
H, C, and P and SDD for Ir. Full details are given in the Supporting
Information.
(50) This very low energy barrier connecting 3a and 3b has analogy
with the situation found in “compressed dihydride” complexes of iridium
(see ref 52), which have extremely flat potential energy surfaces
connecting dihydrogen and dihydride structures.
(51) Rh(V) complexes see: McBee, J. L.; Escalada, J.; Tilley, T. D.
J. Am. Chem. Soc. 2009, 131, 12703 and references therein.
(52) Pons, V.; Heinekey, D. M. J. Am. Chem. Soc. 2003, 125, 8428.
(53) A related rapid equilibration of Irꢀdihydride and Irꢀelogo-
nated dihydrogen isomers in IrH2(2,6-(tBu2PO)2C6H3) has also been
reported: G€ottker-Schnetmann, I.; Heinekey, D. M.; Brookhart, M.
J. Am. Chem. Soc. 2006, 128, 17114.
(54) Gelabert, R.; Moreno, M.; Lluch, J. M.; Lledꢀos, A.; Pons, V.;
Heinekey, D. M. J. Am. Chem. Soc. 2004, 126, 8813.
(55) Mort, B. C.; Autschbach, J. J. Am. Chem. Soc. 2006, 128, 10060.
(56) Gelabert, R.; Moreno, M.; Lluch, J. M.; Lledꢀos, A.; Heinekey,
D. M. J. Am. Chem. Soc. 2005, 127, 5632.
(57) Dziallasa, M.; H€ohna, A.; Werner, H. J. Organomet. Chem. 1987,
330, 207.
13168
dx.doi.org/10.1021/ja2047599 |J. Am. Chem. Soc. 2011, 133, 13162–13168