Kinetics and mechanism of indene C-H bond activation by [(COD)Ir(μ2-OH)]2

Article abstract of DOI:10.1021/om400230j

The hydroxy-bridged dimer [(COD)Ir(μ₂-OH)]₂ (COD = 1,5-cyclooctadiene) cleanly cleaves C-H bonds in indene and cyclopentadiene to produce (COD)Ir(η³-indenyl) and (COD)Ir(η⁵-C ₅H₅), respectively. The kinetics of the formation of (COD)Ir(η³-indenyl) are consistent with a mechanism that involves coordination of indene to [(COD)Ir(μ₂-OH)]₂ followed by rate-determining C-H activation from the iridium dimer-indene unit. Transition-state analysis of the Ir and Rh hydroxy dimers indicates that the C-H activation proceeds through a direct deprotonation of indene by the M-OH unit rather than a stepwise oxidative addition/reductive elimination mechanism. The crystal structure of [(COD)Ir]₅(μ₄-O)(μ₃- O)(μ₂-OH), a dehydration product of [(COD)Ir(μ₂-OH)] ₂, is presented.

Full text of DOI:10.1021/om400230j

Article
Kinetics and Mechanism of Indene C−H Bond Activation by
[(COD)Ir(μ2‑OH)]2
Tonia S. Ahmed, Ian A. Tonks, Jay A. Labinger,* and John E. Bercaw*
Arnold and Mabel Beckman Laboratories of Chemical Synthesis, California Institute of Technology, Pasadena, California 91125,
United States
S
* Supporting Information
ABSTRACT: The hydroxy-bridged dimer [(COD)Ir(μ2-OH)]2 (COD = 1,5-cyclooctadiene) cleanly cleaves C−H bonds in
indene and cyclopentadiene to produce (COD)Ir(η3-indenyl) and (COD)Ir(η5-C5H5), respectively. The kinetics of the
formation of (COD)Ir(η3-indenyl) are consistent with a mechanism that involves coordination of indene to [(COD)Ir(μ2-
OH)]2 followed by rate-determining C−H activation from the iridium dimer−indene unit. Transition-state analysis of the Ir and
Rh hydroxy dimers indicates that the C−H activation proceeds through a direct deprotonation of indene by the M−OH unit
rather than a stepwise oxidative addition/reductive elimination mechanism. The crystal structure of [(COD)Ir]5(μ4-O)(μ3-
O)(μ2-OH), a dehydration product of [(COD)Ir(μ2-OH)]2, is presented.
reaction with indene as compared to the hydroxy-bridged
dimer. In both cases, the rate-limiting step was the displace-
ment of the coordinated solvent or water ligand by indene. We
also explored the related dimer [(COD)Rh(μ2-OH)]2,
anticipating that solvent displacement from a neutral (COD)-
Rh(OH)(solv) species would be faster than from the cationic
group 10 analogues.7 The hydroxy-bridged dimer [(COD)Rh-
(μ2-OH)]2 was found to react with indene directly via the
dimer, and the rate-limiting step was found to be C−H
activation rather than ligand displacement. Furthermore, the Rh
dimer reacts with indene more slowly than do the group 10
dimers. Thus, we have turned our attention to the Ir congener
[(COD)Ir(μ2-OH)]2, anticipating that it may undergo C−H
activation more rapidly than [(COD)Rh(μ2-OH)]2. Herein we
report activation of C−H bonds of indene and cyclopentadiene
by the iridium dimer, together with the results of kinetics
studies of the mechanisms of these reactions.
INTRODUCTION
■
Efficient and selective oxidative functionalization of the C−H
bonds of hydrocarbons and other substrates could have a major
impact on the development of catalysts for producing value-
added fuels, commodity chemicals, and pharmaceuticals.1 Many
transition metal complexes capable of C−H activation are
unstable or unreactive in the presence of water, which
represents a significant obstacle because water is a potential
byproduct of oxidative C−H functionalizations utilizing O2.
Additionally, many model systems for electrophilic C−H
activation require making a new C−H bond (e.g., with
formation of methane) as the target C−H bond is broken.2
In consideration of these issues, C−H bond activation using
metal alkoxy and hydroxy complexes in the presence of water
and alcohols has been the focus of recent studies.3−5
Our group has recently found that the air- and water-stable
hydroxy-bridged dimers [(diimine)M(μ2-OH)]2 and the
2+
bis(aquo) dications [(diimine)M(OH2)2]2+ (M = Pd, Pt) are
capable of activating C−H bonds of arenes and allylic positions
of olefins.6 Kinetic studies of these reactions revealed that Pd
dimers dissociate to [(diimine)PdOH(solv)]+ monomers in the
presence of weakly coordinating solvents prior to reaction with
indene, whereas the Pt dimers require acid assistance to
generate [(diimine)Pt(OH2)2]2+, which undergoes faster
RESULTS AND DISCUSSION
■
[(COD)Ir(μ2-OH)]2 (1; COD = 1,5-cyclooctadiene) reacts
with 2 equivalents of indene to quantitatively (by NMR)
Received: March 18, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om400230j | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
generate (COD)Ir(η3-indenyl), 2, with release of water (eq 1).
The hydroxy-bridged iridium dimer is only slightly soluble in
C. Alternatively, the rate dependence on [TFE-d3] could be a
result of a change in solvent polarity; TFE-d3 constitutes greater
than 10% of solvent by volume.
Compared to indene C−H activation with [(COD)Rh(μ2-
OH)]2, the reaction of indene with 1 proceeds at roughly the
same rate and through a similar mechanism. The enthalpy of
activation is larger for the Ir system (20.9(1) kcal·mol−1 for Ir
vs 16.6(1) kcal·mol−1 for Rh), but is partially offset by a less
negative entropy of activation (−12.7 eu for Ir vs −21.7 eu for
Rh). A possible explanation for the differences in ΔH⧧ could be
differences in M−OH bond strength; if the transition state of
C−H activation involves significant M−OH bond breakage, a
stronger Ir−OH bond would lead to a larger ΔH⧧ than the
corresponding Rh analogue. Since oxidation of IrI to IrIII should
be more facile than for RhI to RhIII, which would result in a
smaller ΔH⧧ in contradiction to the observations, we
tentatively favor direct attack on the coordinated indene C−
H bond by hydroxide rather than oxidative addition of the C−
H bond followed by reductive elimination of water (Scheme 2).
Hydroxy-bridged dimer 1 does not activate the C−H bonds
of cyclopentene, cyclohexene, or other substrates having
stronger C−H bonds. However, 1 does rapidly react with the
C−H bond of cyclopentadiene to produce (COD)Ir(η5-C5H5),
3 (eq 2).
neat DCE-d4 (DCE = 1,2-dichloroethane), so solvent mixtures
of TFE-d3 (TFE = 2,2,2-trifluoroethanol) and DCE-d4 were
utilized in the studies discussed below.
The reaction kinetics were followed under the conditions
shown in eq 1, with varying concentrations of indene and TFE-
d3. Dimer 1 and product 2, but no intermediates, were observed
1
by H NMR spectroscopy. Kinetic runs at varying [indene]
(see Supporting Information (SI), Figures S1 and S2) indicate
that the reaction is first order with respect to both indene and
1. Similar to C−H activation with [(COD)Rh(μ2-OH)]2, there
is a dependence on [TFE] that is consistent with a two-term
rate law, wherein one pathway has a first-order dependence on
[TFE] and the other is independent of [TFE] (see SI, Figure
S3). C−H activation is strongly implicated as rate-determining
by comparing the rates of reaction of 1 with indene and with
1,1,3-trideuteroindene, kH/kD = 3.1(1) (see SI, Figure S4). An
Eyring analysis of the temperature dependence of reaction rate
from 300 to 333 K indicates activation parameters ΔH⧧
=
20.9(1) kcal·mol−1 and ΔS⧧ = −12.7(1) eu, the latter
consistent with coordination of indene to 1 in the transition
state.
These kinetic data suggest a mechanism similar to that for
reaction of indene with [(COD)Rh(μ2-OH)]2 (Scheme 1).
The first-order dependence on [indene] indicates its pre-
equilibrium coordination to 1 affording adduct A, with the
equilibrium strongly favoring 1 because no intermediates are
1
detected by H NMR. In accord with the kinetics, the rate-
determining step, likely C−H bond activation based on the
KIE, occurs from a dimeric Ir species A generating 1 equivalent
each of 2 and [(COD)Ir(OH)(solv)], C. Monomer C then
reacts rapidly with another equivalent of indene to generate a
second equivalent of 2. The [TFE-d3] dependence may indicate
competitive precoordination of TFE to generate B, which can
undergo C−H activation in a similar manner to generate 2 and
Interestingly, 1 is stable in dichloromethane that has not
been dried, but it is unstable in dry dichloromethane. When
yellow 1 was dissolved in dichloromethane-d2 either dried over
CaH2 or passed through an activated alumina column, the
solution turned a deep orange color over the course of a few
hours. After standing overnight, the solution had precipitated
needle-shaped, dark red crystals. An X-ray diffraction analysis of
Scheme 1
B
dx.doi.org/10.1021/om400230j | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 2
the red crystalline material revealed it to be a product of 1
resulting from loss of water, [(COD)IrI]5(μ4-O)(μ3-O)(μ2-
OH), 4 (eq 3, Figure 1). Pentamer 4 contains a trinuclear Ir3
CONCLUSIONS
■
[(COD)Ir(μ2-OH)]2 activates both indene and cyclopenta-
diene C−H bonds to produce (COD)Ir(η3-indenyl) (2) and
(COD)Ir(η5-C5H5) (3), respectively. The kinetic isotope effect
and rate law reveal that 1 reacts directly from the substrate/
dimer adduct and that C−H bond activation is rate-limiting.
These features are similar to those for the rhodium analogue,
[(COD)Rh(μ2-OH)]2. Comparisons of the activation enthal-
pies for reaction of indene with 1 vs [(COD)Rh(μ2-OH)]2
suggest that the mechanism of C−H activation by [(COD)-
M(μ2-OH)]2 (M = Rh, Ir) is likely intramolecular deprotona-
tion, rather than oxidative addition of the C−H bond to the
metal center to produce IrIII hydride followed by reductive
elimination of water. Hydroxy-bridged dimer 1 spontaneously
dehydrates in dry solvents to produce a pentameric IrI oxo-
bridged complex, 4, which may be of interest to water oxidation
schemes.
EXPERIMENTAL SECTION
General Considerations. All manipulations were carried out in a
glovebox under a nitrogen atmosphere. [(COD)Ir(μ2-OH)]2 and
■
9
1,1,3-trideuteroindene10 were prepared according to literature
procedures. Indene and dicyclopentadiene were reagent-grade
commercial samples purchased from Sigma-Aldrich. Dicyclopenta-
diene was freshly cracked prior to use. 1,2-Dichloroethane-d4 and
trifluoroethanol-d3 were purchased from Cambridge Isotopes and
degassed prior to use. For some experiments, methylene chloride was
dried via the method of Grubbs.11 1H NMR spectra were recorded
using a Varian INOVA 500 MHz spectrometer. X-ray crystallographic
data collection was carried out on a Bruker KAPPA APEX II
diffractometer with a 0.71073 Å Mo Kα source. Crystals were removed
from a scintillation vial to a microscope slide coated with Paratone N
oil. Samples were selected and mounted on a glass fiber with Paratone
N oil. The structures were solved by direct methods. All non-hydrogen
atoms were refined anisotropically. Details regarding refined data and
cell parameters are displayed in Table 1.
Figure 1. ORTEP drawings of 4 with 50% thermal ellipsoids (two
views: side-on, left, and offset, right). Selected bond distances (Å):
Ir1−O2 2.102(1); Ir2−O2 2.0885(1); Ir3−O2 2.1409(1); Ir4−O2
2.0827(1); Ir1−O3 2.0225(1); Ir2−O3 2.0467(1); Ir3−O3 2.0536(1);
Ir4−O1 2.0361(1); Ir5−O1 2.0906(1). Intermetallic distances (Å):
Ir1−Ir2 3.0443(1); Ir1−Ir3 2.7610(1); Ir1−Ir5 2.8632(2); Ir2−Ir3
2.7292(2); Ir2−Ir5 3.0890(2). H atoms and methylene groups of the
COD ligands have been removed for clarity.
core capped by two oxo ligands, similar to a trinuclear cluster
prepared by Cotton, [(COD)IrII]3(μ3-O)2(μ2-I).8 The reaction
is apparently driven by the expulsion of H2O from 1 into dry
methylene chloride and by the insolubility of 4. The reaction
does not appear to be reversible; addition of H2O to a solution
of 4 does not regenerate 1.
4 likely forms by initial dehydration of 1 to produce a
bridging Ir oxo complex, [(COD)IrOIr(COD)], which
proceeds to dimerize to form a tetranuclear Ir4O2 core,
followed by reaction with an additional half-equivalent of 1.
The net result of the reaction is the spontaneous intramolecular
deprotonation of a hydroxyl group to generate a bridging IrI
oxo species. This transformation is unprecedented on IrI, and
complex 4 appears to represent the first crystallographically
characterized example of a formal IrI species containing
bridging oxo ligands.
Synthesis of (COD)Ir(η3-indenyl) (2). A 5.0 mg amount of 1
(0.0079 mmol, 1 equiv) was dissolved in 650 μL of DCE-d4 and 100
μL of TFE-d3 in a J-Young NMR tube. Then 9.2 μL of indene (0.080
mmol, 10 equiv) was added to the solution, and the tube sealed and
heated to 50 °C. The progression of the reaction was followed by
NMR, with complete conversion occurring after 4 h. The identity of
1
the product was confirmed by comparing the H NMR spectrum to
that of an authentic sample of 2 purchased from Strem.
Synthesis of (COD)Ir(η5-C5H5) (3). A similar preparation to that
used to produce 2 was used, substituting freshly cracked cyclo-
1
pentadiene for indene. The reaction was followed by H NMR, and
complete conversion to product (by NMR) occurred within 30 min.
The identity of 3 was confirmed by comparing the 1H NMR spectrum
C
dx.doi.org/10.1021/om400230j | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
methylene chloride-d2. However, 1 remained intact and did not
produce [(COD)Ir]5(μ4-O)(μ3-O)(μ2-OH).
Table 1. Crystal and Refinement Data for Complex 4
CCDC number
empirical formula
fw
885854
Standard Reaction Protocol for Kinetics Experiments. A 750
μL sample of a 6.5:1 DCE-d4/TFE-d3 stock solution containing 0.010
M 1 (5 mg, 0.0079 mmol) and 5 μL of TMS2O (as a standard) was
syringed into a J-Young NMR tube inside a glovebox. The sample was
cooled to 0 °C, and the desired amount of a 1.575 M stock solution of
indene in DCE-d4 was added to the reaction (50, 100, or 200 μL) via
syringe. An additional amount of DCE-d4 was added to reactions that
contained less than 200 μL of the indene stock solution to give all
reactions a total volume of 0.95 mL. The reactions were then inserted
into the preheated NMR probe, and the temperature was allowed to
equilibrate for 5 min before locking and shimming the instrument. An
additional 5 min was utilized for locking and shimming before finally
starting data collection 10 min after initial sample insertion. Single
spectra were collected every 5 min until the starting material was
completely consumed. The spectra were processed using the
MestReNova software package, and the starting material and product
peaks were integrated against the internal TMS2O standard. The two
peaks used for all kinetic determinations were the vinylic COD peak in
1 (Figure 2, 3.68 ppm) and the 2-position of the η3-indenyl in 2
(Figure 2, 6.04 ppm). In all cases, disappearance of the starting
material was directly correlated to appearance of product. Several of
the reactions were run in duplicate or triplicate to confirm
reproducibility. The KIE experiments were carried out in a similar
manner, utilizing 1,1,3-trideuteroindene in place of normal indene.
C40H60O3Ir5·CH2Cl2
1635.93
T (K)
100(2)
a, Å
10.3282(4)
22.4057(9)
12239(7)
90
b, Å
c, Å
α, deg
β, deg
92.280(2)
90
γ, deg
volume, Å3
3982.6(3)
4
Z
cryst syst
space group
monoclinic
P21/n
d
calc, g/cm3
2.726
range, deg
μ, mm−1
2.17 to 32.63
16.819
abs correction
GOF
semiempirical
1.436
a
R1
R1 = 0.0236
wR2 = 0.0381
b
wR2 [I > 2σ(I)]
a
b
R1 = ∑∥Fo| − |Fc∥/∑|Fo|. wR2 = [∑[w(Fo2 − Fc2)2]/∑[w(Fo )2]1/2
.
2
to a sample prepared via salt metathesis of [(COD)Ir(μ2-Cl)]2 with
NaCp following a literature procedure.12
Synthesis of [(COD)Ir]5(μ4-O)(μ3-O)(μ2-OH) (4). A 10 mg (0.016
mmol) portion of 1 was suspended in 2 mL of dry, degassed
methylene chloride-d2 in a J-Young NMR tube. The tube was sealed,
and the suspension was refluxed in an oil bath for 1 h to give an orange
solution with a small amount of 1 undissolved at the bottom. The
solution was cooled to room temperature and left to stand overnight,
during which time small red crystals of 4 formed. After 12 h, near
quantitative formation of 4 occurred. 4 was identified via X-ray
crystallography since it was insoluble in all solvents tested. The
formation of 4 could also be achieved without heating the suspension,
but the general insolubility of 1 resulted in this method being slower
and lower yielding. For comparison purposes, 1 was dissolved in “wet”
ASSOCIATED CONTENT
■
S
* Supporting Information
Log plots and Eyring plots for all kinetic analyses. Tables of
bond lengths, angles, and anisotropic displacement parameters
for 4. X-ray crystallographic data file (CIF) for 4. The
crystallographic data for 4 have been deposited with the
Cambridge Crystallographic Data Center and can be obtained
by requesting the deposition number. This material is available
free of charge via the Internet at http://pubs.acs.org.
Figure 2. Example spectral array for a kinetic run of 1 with 10 equivalents of indene at 50 °C. The peaks that were integrated were the vinylic COD
protons on the starting material 1 and the 2-H position on the product 2.
D
dx.doi.org/10.1021/om400230j | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
AUTHOR INFORMATION
Corresponding Author
*E-mail: jal@caltech.edu (J. A. Labinger), bercaw@caltech.edu
(J. E. Bercaw).
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Lawrence Henling and Dr. Michael Day for
assistance with the X-ray studies. The Bruker KAPPA APEXII
X-ray diffractometer was purchased via an NSF CRIF: MU
award to the California Institute of Technology, CHE-0639094.
This work has been supported by USDOE Office of Basic
Energy Sciences (Grant No. DE-FG03-85ER13431) and by the
Caltech Summer Undergraduate Research Fellowship program
(T.S.A.).
REFERENCES
■
(1) (a) Crabtree, R. H. Dalton Trans. 2001, 17, 2437. (b) Labinger, J.
A.; Bercaw, J. E. Nature 2002, 417, 507. (c) Fekl, U.; Goldberg, K. I.
Adv. Inorg. Chem. 2003, 54, 259. (d) Vedernikov, A. N. Curr. Org.
Chem. 2007, 11, 1401.
(2) (a) Driver, T. G.; Williams, T. J.; Labinger, J. A.; Bercaw, J. E.
Organometallics 2007, 26, 294−301. (b) Ackerman, L. J.; Sadighi, J. P.;
Kurtz, D. M.; Labinger, J. A.; Bercaw, J. E. Organometallics 2003, 22,
3884−3890.
(3) (a) Tenn, W. J.; Young, K. J. H.; Bhalla, G.; Oxgaard, J.; Goddard,
W. A.; Periana, R. A. J. Am. Chem. Soc. 2005, 127, 14172. (b) Tenn, W.
J.; Young, G. J.; Oxgaard, J.; Nielson, R. J.; Goddard, W. A.; Periana, R.
A. Organometallics 2006, 25, 5173. (c) Oxgaard, J.; Tenn, W. J.;
Nielson, R. J.; Periana, R. A.; Goddard, W. A. Organometallics 2007,
26, 1565.
(4) (a) Feng, Y.; Lail, M.; Barakat, K. A.; Cundari, T. R.; Gunnoe, T.
B.; Petersen, J. L. J. Am. Chem. Soc. 2005, 127, 14174. (b) Feng, Y.;
Lail, M.; Foley, N. A.; Gunnoe, T. B.; Barakat, K. A.; Cundari, T. R.;
Petersen, J. L. J. Am. Chem. Soc. 2006, 128, 7982. (c) Cundari, T. R.;
Grimes, T. V.; Gunnoe, T. B. J. Am. Chem. Soc. 2007, 129, 13172.
(5) (a) Kloek, S. M.; Heinekey, D. M.; Goldberg, K. I. Angew. Chem.,
Int. Ed. 2007, 46. (b) Hanson, S. K.; Heinekey, D. M.; Goldberg, K. I.
Organometallics 2008, 27, 1454.
(6) (a) Williams, T. J.; Caffyn, A. J. M.; Hazari, N.; Oblad, P. F.;
Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2008, 130, 2418.
(b) Bercaw, J. E.; Hazari, N.; Labinger, J. A.; Oblad, P. F. Angew.
Chem., Int. Ed. 2008, 47, 9941. (c) Bercaw, J. E.; Hazari, N.; Labinger,
J. A. J. Org. Chem. 2008, 73, 8654. (d) Oblad, P. F.; Bercaw, J. E.;
Hazari, N.; Labinger, J. A. Organometallics 2010, 29, 789.
(7) Bercaw, J. E.; Hazari, N.; Labinger, J. A. Organometallics 2009, 28,
5489.
(8) Cotton, F. A.; Lahuerta, P.; Sanau, M.; Schotzer, W. J. Am. Chem.
Soc. 1985, 107, 8284.
(9) Green, L. M.; Meek, D. W. Organometallics 1989, 8, 659.
(10) Yasuda, M.; Pak, C.; Sakurai, H. Bull. Chem. Soc. Jpn. 1980, 53,
502.
(11) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518.
(12) Sowa, J. R., Jr.; Angelici, R. J. J. Am. Chem. Soc. 1991, 113, 2537.
E
dx.doi.org/10.1021/om400230j | Organometallics XXXX, XXX, XXX−XXX