Intramolecular C - H bond activation and redox isomerization across two-electron mixed valence diiridium cores

Article abstract of DOI:10.1021/om7007748

Metal-metal cooperativity enables the reaction of carbon-based substrates at diiridium two-electron mixed valence centers. Arylation of Ir ₂^0,II(tfepma)₃Cl₂ (1) (tfepma = bis[(bistrifluoroethoxy)phosphino]methyl

Full text of DOI:10.1021/om7007748

Organometallics 2008, 27, 1073–1083
1073
Intramolecular C-H Bond Activation and Redox Isomerization
across Two-Electron Mixed Valence Diiridium Cores
Arthur J. Esswein,§ Adam S. Veige,‡ Paula M. B. Piccoli,† Arthur J. Schultz,† and
Daniel G. Nocera*,§
Department of Chemistry, 6-335, Massachusetts Institute of Technology, 77 Massachusetts AVenue,
Cambridge, Massachusetts 02139-4307, and the Intense Pulsed Neutron Source, Argonne National
Laboratory, Argonne, Illinois 60439-4814
ReceiVed July 30, 2007
Metal-metal cooperativity enables the reaction of carbon-based substrates at diiridium two-electron
mixed valence centers. Arylation of Ir20,II(tfepma)3Cl2 (1) (tfepma ) bis[(bistrifluoroethoxy)phosphi-
no]methylamine) with RMgBr (R ) C6H5 and C6D5) is followed by C-H bond activation to furnish the
bridging benzyne complex Ir2II,II(tfepma)3(µ-C6H4)(C6H5)H (2), as the kinetic product. At ambient
temperature, 2 isomerizes to Ir2I,III(tfepma)3(µ-C6H4)(C6H5)H (3) (kobs ) 9.57 ( 0.10 × 10-5 s-1 at 31.8
°C, ∆Hq ) 21.7 ( 0.3 kcal/mol, ∆Sq ) -7.4 ( 0.9 eu), in which the benzyne moiety is conserved and
the IrIII center is ligated by terminal hydride and phenyl groups. The same reaction course is observed
for arylation of 1 with C6D5MgBr to produce 2-d10 and 3-d10 accompanied by an inverse isotope effect,
kh/kd) 0.44 (kobs ) 2.17 ( 0.10 × 10-4 s-1 in C6D6 solution at 31.8 °C, ∆Hq ) 24.9 ( 0.7 kcal/mol,
∆Sq ) -6.4 ( 2.4 eu). 2 reacts swiftly with hydrogen to provide Ir2II,II(tfepma)3H4 as both the syn and
anti isomers (4-syn and 4-anti, respectively). The hydrides of 4-syn were directly located by neutron
diffraction analysis. X-ray crystallographic examination of 2, 2-d10, 3, and 4-syn indicates that cooperative
reactivity at the bimetallic diiridium core is facilitated by the ability of the two-electron mixed valence
framework to accommodate the oxidation state changes and ligand rearrangements attendant to the reaction
of the substrate.
hydride into a position bridging the two metal centers produces
the preferred transition state from which elimination may
proceed smoothly. Owing to the bimolecular order of the above
transformation, reactions proceeding through this dinuclear
elimination-type transition state should be greatly enhanced if
supported intramolecularly within a bimetallic core. We have
shown this to be the case with two-electron mixed valence
complexes, Mn-Mn+2 (M ) Rh, Ir), that are coordinated by
bis(difluorophosphino)methylamine (dfpma, CH3N(PF2)2) and
bis[(bistrifluoroethoxy)phosphino]methylamine (tfepma, CH3-
N[P(OCH2CF3)2]2) ligands. The ability of the bimetallic core
to accommodate the electronic and steric asymmetries
associated with oxidation state changes and ligand rearrange-
ments is essential to the reactivity of two-electron mixed
valence Mn-Mn+2 complexes.10 For the case of H2 activation,
computational studies show that the internal asymmetry of
the two-electron mixed valence core is preserved as hydrogen
is shuttled from one metal to the other through a critical
bridging intermediate.11
Introduction
Cooperative reactivity at a metal-metal core is based on the
elementary premise that affixed metal centers working in concert
will effect chemistry otherwise unattainable to mononuclear
complexes.1–6 For instance, some metal acyls, alkyls, and
dihydrides are particularly sluggish toward intramolecular
elimination, but become extremely unstable in the presence of
a second metal complex. The enhanced reactivity has been
ascribed to the following transformation (R ) hydride, alkyl,
or acyl).7–9
(1)
Dissociation of a ligand opens a coordination site to permit
attack by a coordinatively saturated partner. Migration of the
* Corresponding author. E-mail: nocera@mit.edu.
§ Massachusetts Institute of Technology.
‡ Current address: Department of Chemistry, University of Florida, P.O.
Box 117200, Gainesville, FL 32611-7200.
† Argonne National Laboratory.
(2)
(1) Bosnich, B. Inorg. Chem. 1999, 38, 2554–2562.
(2) McCollum, D. G.; Bosnich, B. Inorg. Chim. Acta 1998, 270, 13–
19.
(3) Gavrilova, A. L.; Bosnich, B. Chem. ReV. 2004, 104, 349–384.
(4) Fackler, J. P. Jr. Inorg. Chem. 2002, 41, 6959–6972.
(5) Adams, R. D.; Captain, B.; Zhu, L. J. Am. Chem. Soc. 2006, 128,
13672–13673.
Atom shuttling through the bridging position of the bimetallic
core has also been important to the activation, in two-electron
steps, of other small molecules such as HX (X ) Cl-, Br-),12
(6) Adams, R. D.; Captain, B.; Zhu, L. J. Am. Chem. Soc. 2007, 129,
7545–7556.
(7) Norton, J. R. Acc. Chem. Res. 1979, 12, 139–145.
(8) Martin, B. D.; Warner, K. E.; Norton, J. E. J. Am. Chem. Soc. 1986,
108, 33–39.
(10) Gray, T. G.; Veige, A. S.; Nocera, D. G. J. Am. Chem. Soc. 2004,
126, 9760–9768.
(11) Veige, A. S.; Gray, T. G.; Nocera, D. G. Inorg. Chem. 2005, 44,
17–26.
(9) Kristjánsdóttir, S. S.; Norton, J. A. In Transition Metal Hydrides;
Dedieu, A., Ed.; VCH: New York, 1990; Chapter 9.
(12) Heyduk, A. F.; Nocera, D. G. Science 2001, 293, 1639–1641.
10.1021/om7007748 CCC: $40.75
2008 American Chemical Society
Publication on Web 02/20/2008

1074 Organometallics, Vol. 27, No. 6, 2008
Esswein et al.
Table 1. Crystallographic Summary for Complexes 2-4-syn
2 · (CH2Cl2) 2-d10 · (CH2Cl2)
3
4-syn (X-ray)
4-syn (neutron)
formula
fw, g/mol
cryst syst
space group
color
cryst dimens, mm
a, Å
b, Å
C40H45F36Ir2N3O12P6Cl2 C40H35D10F36Ir2N3O12P6Cl2 C39H43F36Ir2N3O12P6 C27H37F36Ir2N3O12P6 C27H37F36Ir2N3O12P6
2084.91
triclinic
P1
2094.91
triclinic
P1
1999.98
monoclinic
P21/c
1849.82
triclinic
P1
1849.82
triclinic
P1
j
j
j
j
colorless
0.20 × 0.12 × 0.10
12.975(2)
13.951(2)
20.478(3)
95.609(3)
104.539(3)
108.063(3)
3348.6(10)
2
colorless
0.25 × 0.20 × 0.10
12.957(2)
13.860(2)
20.390(4)
95.552(3)
103.892(3)
107.641(3)
3317.6(10)
2
colorless
0.12 × 0.10 × 0.06
21.1423(13)
16.6501(10)
18.5674(12)
90
98.7750(10)
90
6459.6(7)
4
colorless
0.25 × 0.25 × 0.20
12.0419(6)
12.2511(6)
21.5922(11)
80.6240(10)
80.1890(10)
61.5970(10)
2748.7(2)
2
colorless
2 × 2 × 1
11.982(4)
12.206(5)
21.559(8)
80.42(3)
79.95(3)
61.61(3)
2718(2)
2
c, Å
R, deg
ꢀ, deg
γ, deg
V, Å3
Z
T, °C
-173(2)
-173(2)
-173(2)
-75(2)
-80(1)
no. ind reflns
no. data
restraints
params
16 261
15 974
15 975
13 584
3781
16 261
4
939
0.0341
0.0873
0.0383
0.0907
1.028
15 974
0
913
0.0334
0.0871
0.0362
0.0890
1.051
15 975
7
907
0.0297
0.0604
0.0412
0.0658
1.013
13 584
6
999
0.0269
0.0664
0.0327
0.0699
1.016
2258
35
581
a
R1 [I > 2σ(I)]
0.182c,d
0.184c,e
0.210
a
wR2 [I > 2σ(I)]
a
R1 (all data)
a
wR2 (all data)
0.196
GOFb
1.53
largest peak hole, e/Å3 1.692–2.174
2.228–1.753
1.513–1.528
1.378–0.874
0.172–0.163 fm/Å3
2
2
a R1 ) ∑|Fo| - |Fc|/∑|Fo|; wR2 ) [∑w(Fo - Fc ) /∑wFo
]
4 1/2. Refinement on Fo for all reflections (having Fo g –3σ(Fo2). wR2 and GOF based on
2
2 2
Fo2, R1 based on Fo, with Fo set to zero for negative Fo2. The observed criterion of Fo > 2σ(Fo2) is used only for calculating R1 and is not relevant to
2
2
2 2
2
the choice of reflections for refinement. b GOF ) [∑w(Fo - Fc ) /(n - p)]1/2 (n ) number of data, p ) number of parameters varied; w ) [σ2Fo
+
2
2
2
2
2
2
(0.0638P)2]-1, where P ) [max(Fo2, 0) + 2Fc ]/3. c Outliers with |Fo2/Fc | > 2 and |Fc /Fo2| > 2 and |(Fo - Fc )|/σFo > 6 were rejected, [I > 3σ(I)].
2
2 2
2
2
d Rw(F2) ) {∑[w(Fo - Fc ) ]/∑[w(Fo2)2]}1/2 e R(F2) ) ∑|Fo - Fc |/∑|Fo2|, [I > 3σ(I)].
.
prior to its introduction into reaction vessels. C6D5MgBr15 and
H2O, and NH3.13However, we have yet to exploit the unique
cooperative reactivity for carbonaceous substrates. We now report
organometallic transformations derived from Mn-Mn+2 complexes
that feature bimetallic cooperation of the two-electron mixed
valence core of Ir20,II(tfepma)3Cl2. The complex possesses an
electron-rich Ir0 center adjacent to a coordinatively unsaturated IrII
center, which can be readily functionalized with organic substrates.
The synthetic, X-ray, and neutron diffraction, redox isomerization,
and intramolecular C-H bond activation studies reported herein
serve to further expand the scope of Mn-Mn+2 chemistry.
16,17
Ir20,II(tfepma)3Cl2
procedures.
were synthesized according to published
X-ray Crystallographic Details. Single crystals were immersed
in a drop of Paratone N oil on a clean microscope slide, affixed to
either a glass fiber or a human hair coated in epoxy resin, and then
cooled to either -75 (4-syn) or -173 °C (2, 2-d10, and 3). The
crystals were then mounted on a Siemens three-circle goniometer
platform equipped with an APEX detector. A graphite monochro-
mator was employed for wavelength selection of the Mo KR
radiation (λ ) 0.71073 Å). The data were processed and refined
using the program SAINT supplied by Siemens Industrial Automa-
tion Inc. Structures were solved by a Patterson heavy atom map
and refined by standard difference Fourier techniques in the
SHELXTL program suite (6.10 v., Sheldrick G. M., and Siemens
Industrial Automation Inc., 2000). Disordered atoms in the
-CH2CF3 groups were fixed at idealized bond lengths where
necessary, and site occupancies were refined and refined anisotro-
pically for those groups that were split evenly. Otherwise the minor
components of unevenly split -CH2CF3 groups were refined
isotropically. Hydrogen atoms were placed in calculated positions
using the standard riding model and refined isotropically; all other
atoms were refined anisotropically. Unit cell parameters, morphol-
ogy, and solution statistics for complexes 2-4-syn are summarized
in Table 1. All thermal ellipsoid plots are drawn at the 50%
probability level, with -CH2CF3 groups, -NMe groups, and
solvents of crystallization omitted for clarity.
Experimental Section
General Considerations. All manipulations were carried out in
an N2-filled glovebox or under an inert atmosphere provided by a
Schlenk line unless otherwise noted. All solvents were reagent grade
(Aldrich) or better and were dried and degassed by standard
methods.14
Methods. NMR data were collected at the MIT Department of
Chemistry Instrument Facility (DCIF) on a Varian Inova Unity 500
spectrometer. NMR solvents (C6D6, CD3CN, THF-d8) were pur-
chased from Cambridge Isotope Laboratories and purified by
standard procedures prior to use. 1H NMR spectra (500 MHz) were
referenced to the residual protio impurities of the given solvent;
31P{1H} NMR (202.5 MHz) spectra were referenced to an external
85% H3PO4 standard. All chemical shifts are reported in the
standard δ notation in parts per million; positive chemical shifts
are to higher frequency from the given reference. Infrared spectra
were recorded on a Bio-Rad XPS 150 FT-IR spectrometer in a
Perkin-Elmer liquid cell equipped with KBr windows. Elemental
analyses were performed by Robertson Microlit Laboratories.
[IrI(COD)Cl]2 (Strem) and C6H5MgBr (1.0 M in THF, Aldrich)
were used without further purification. Hydrogen gas (Grade 5.0,
Airgas) was passed through a U-tube immersed in liquid nitrogen
Neutron Diffraction. Neutron diffraction data were obtained at
the Intense Pulsed Neutron Source (IPNS) at Argonne National
Laboratory using the time-of-flight Laue single-crystal diffracto-
meter (SCD).18–20 At the IPNS, pulses of protons are accelerated
(15) Bianco, V. D.; Doronzo, S. Inorg. Synth. 1976, 12, 164–166.
(16) Heyduk, A. F.; Nocera, D. G. J. Am. Chem. Soc. 2000, 122, 9415–
9426.
(17) Heyduk, A. F.; Nocera, D. G. Chem. Commun. 1999,
1519–1520.
(18) Schultz, A. J.; Teller, R. G.; Williams, J. M.; Lukehart, C. M. J. Am.
Chem. Soc. 1984, 106, 999–1003.
(13) Veige, A. S.; Nocera, D. G. Chem. Commun. 2004, 1958–1959.
(14) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory
Chemicals, 4th ed.; Butterworth-Heinmann: Oxford, 1996.

Intramolecular C-H Bond ActiVation
Organometallics, Vol. 27, No. 6, 2008 1075
into a heavy-element target 30 times a second to produce pulses of
neutrons by the spallation process. With two position-sensitive area
detectors and a range of neutron wavelengths, a solid volume of
reciprocal space is sampled with each stationary orientation of the
sample and the detectors. The SCD has two 6Li-glass scintillation
position-sensitive area detectors, each with active areas of 15 ×
15 cm2 and a spatial resolution of <1.5 mm. One of the detectors
is centered at a scattering angle of 75° and a crystal-to-detector
distance of 23 cm, and the second detector is at 120° and 18 cm.
A crystal of syn-Ir2II,II(tfepma)3(H)4 (4-syn), with approximate
dimensions of 2 × 2 × 1 mm3 and weighing 6.8 mg, was coated
with fluorocarbon grease, wrapped in aluminum foil, and glued to
an aluminum pin that was mounted on the cold stage of a closed-
cycle helium refrigerator and cooled to -80(1) °C. Details of the
data collection and analysis procedures have been published
previously.18–23 The GSAS software package was used for structural
analysis.24 Initial atomic positions were taken from the X-ray
structure of 4; structural refinement was then followed using the
neutron diffraction data. Hydride ligands bound to iridium were
clearly located by difference Fourier maps. However elevated
temperatures were required to maintain crystallinity, the structure
showed complexity from disordered -CH2CF3 groups, and the
limited data restricted the solution to isotropic refinement of all
atoms in the final refinement. Full experimental details can be found
in the Supporting Information.
NMR Kinetics. Variable-temperature NMR kinetics studies were
performed in C6D6, wherein a solution of complex 2 (or 2-d10) in
C6D6 was prepared. The solution was then filtered and sealed in a
J. Young NMR tube. Kinetic runs were carried out over the range
20–40 °C (calibrated temperatures 22.2, 31.8, and 41.5 °C) in 10
degree increments. Temperature calibration was carried out using
100% ethylene glycol by standard procedures (see Supporting
Information). Spectra were recorded every 10 min for 200 min for
each run. Relative concentrations of 2 and 3 (2-d10 and 3-d10) were
assessed by integration over one of the -NMe resonances
characteristic for each complex in C6D6 (2.60 and 2.48 ppm,
respectively). The integral over both peaks was then set to 6H, and
then each peak was integrated separately. In all cases, both the
appearance of 3 and the decay of 2 (3-d10 and 2-d10) were fit to a
single exponential with rate constants that agree within error.
Preparation of Ir2II,II(tfepma)3(µ-C6H4)(C6H5)H (2). A 500
mg amount of Ir20,II(tfepma)3Cl2 (0.261 mmol, 1 equiv) was
dissolved in 5 mL of THF, and the solution was frozen. In a separate
vial, 2.1 equiv of C6H5MgBr (548 µL, 1.0 M THF solution) was
diluted with 1 mL of THF, and the solution was frozen. Immediately
upon thawing, the Grignard was added to the iridium complex
dropwise, affecting a color change to light yellow. The solvent was
then immediately removed, and the residue triterated with pentane
(3 × 2 mL), then taken up in Et2O (5 mL), and filtered through a
plug of Celite. The light yellow solution was concentrated to about
3 mL and placed in a freezer (-35 °C) overnight. The supernatant
was decanted, and the resulting white solid was washed with 2 mL
of pentane to give 359 mg (69%) of Ir2II,II(tfepma)3(µ-C6H4)(C6H5)H
(2) as a colorless powder. 1H NMR (THF-d8) δ/ppm: –13.72 (ddt,
244.8 Hz, 20.4 Hz, 10.2 Hz, 1 H, Ir-H), 2.13 (m, 1H, -OCH2CF3),
2.74 (t, 8.9 Hz, 3H, -NMe), 2.82 (t, 10.2 Hz, 3H, -NMe), 2.83 (t,
6.4 Hz, 3H, -NMe), 3.24 (m, 1H, -OCH2CF3), 3.85 – 5.21 (m,
22H, -OCH2CF3), 6.62 (t, 6.8 Hz, 1H, C-H), 6.67 (t, 6.4 Hz, 2H,
C-H), 6.78 (t, 10.2 Hz, 1H, C-H), 6.83 (t, 10.2 Hz, 1H, C-H), 7.01
(t, 5.9 Hz, 1H, C-H), 7.64 (t, 6.8 Hz, 1H, C-H), 8.00 (t, 8.1 Hz,
2H, C-H). 31P{1H} NMR (THF-d8) δ/ppm: 24.09 (ddd, 370.4 Hz,
74.1, 26.7 Hz), 34.90 (m), 39.56 (dddd, 68.1 Hz, 35.6 Hz, 20.7
Hz, 8.9 Hz, 1P), 62.24 (dddd, 367.4 Hz, 91.8 Hz, 20.7 Hz, 8.9 Hz,
1P), 104.14 (dddd, 269.6 Hz, 30.0 Hz, 17.8 Hz, 8.9 Hz), 120.48
(ddd, 269.6 Hz, 26.7 Hz, 20.7 Hz). In C6D6 a multiplet is observed
at 2.35 ppm for 2. These resonances arise from the methylene
protons of one of the tfepma ligands. The multiplet arises from the
3JH-F splitting from the three fluorines of the CF3 group and the
3JH-P from the phosphorus, and they are observed in pure samples
of 2. IR(CD3CN) νIr-H/cm-1: 2069. Anal. Calc for C39H43N3-
O12F36P6Ir2: C, 23.42; H, 2.17; N, 2.10. Found: C, 23.21; H, 2.07;
N, 1.99. Crystals suitable for X-ray diffraction were grown from
concentrated Et2O solutions layered with pentane at -35 °C as
colorless blocks.
Preparation of Ir2II,II(tfepma)3(µ-C6D4)(C6D5)D (2-d10). Prepa-
ration of 2-d10 proceeded in a similar manner to that of complex 2.
A 220 mg amount of Ir20,II(tfepma)3Cl2 (0.115 mmol, 1 equiv) was
dissolved in 5 mL of THF, and the solution was frozen. In a separate
vial, 2.1 equiv of freshly prepared C6D5MgBr (482 µL, 0.5 M THF
solution) was diluted with 1 mL of THF and the solution was frozen.
Immediately upon thawing the Grignard was added to the iridium
complex dropwise, affecting a color change to light yellow. Workup
proceededasoutlinedfor2,giving209mg(90%)ofIr2II,II(tfepma)3(µ-
C6D4)(C6D5)D (2-d10) as a colorless solid. 1H NMR (CD3CN)
δ/ppm: 2.56 (t, 6.4 Hz, 3H, -NMe), 2.58 (t, 9.8 Hz, 3H, -NMe),
2.78 (t, 9.8 Hz, 3H, -NMe), 3.28 (m, 1H, -OCH2CF3), 3.38 (m,
1H, -OCH2CF3), 4.3–4.9 (m, 20H, -OCH2CF3), 4.13 (m, 1H,
-OCH2CF3), 5.306 (m, 1H, -OCH2CF3). 31P{1H} NMR (CD3CN)
δ/ppm: 26.29 (dddd, 318.0 Hz, 32.4 Hz, 21.5 Hz, 9.9 Hz), 31.84
(dddd, 924.5 Hz, 32.4 Hz, 26.1 Hz, 18.7 Hz), 56.50 (ddm, 318.0
Hz, 113.2 Hz), 65.10, (ddd, 113.2 Hz, 23.6 Hz, 9.9 Hz), 83.3 (dm,
∼200 Hz), 93.33 (dddd, 924.5 Hz, 195.4 Hz, 26.1 Hz, 21.1 Hz).
2H NMR (CH3CN) δ/ppm: -13.61 (dm, 37 Hz, Ir-D), 6.5–7.2 (bm,
C-D), 7.4–8.1 (bm, C-D). IR(CD3CN) νIr-D/cm-1: 1453. Anal. Calc
for C39H33D10N3O12F36P6Ir2: C, 23.30; H, 2.66; N, 2.09. Found: C,
23.58; H, 2.29; N, 1.93.
Preparation of Ir2I,III(tfepma)3(µ-C6H4)(C6H5)H (3). A 235
mg (0.118 mmol) amount of Ir2II,II(tfepma)3(µ-C6H4)(C6H5)H was
suspended in 3 mL of C6H6 in a thick-walled glass bomb to give
a light yellow solution. The flask was then placed in an oil bath at
50 °C for 6 h, after which the oil bath was allowed to cool slowly.
Thesolventwasdecantedtoafford150mg(64%)ofIr2I,III(tfepma)3(µ-
1
C6H4)(C6H5)H (3) as colorless single crystals. H NMR (CD3CN)
δ/ppm: -9.79 (ddt, 233.3 Hz, 23.8 Hz, 6.4 Hz, 1H, Ir-H), 2.47 (t,
7.0 Hz, 3H, -NMe), 2.79 (t, 6.4 Hz, 3H, -NMe), 2.87 (t, 9.8 Hz,
3H, -NMe), 3.2–3.4 (m, 4H, -OCH2CF3), 3.75–4.01 (m, 2H,
-OCH2CF3), 4.1–4.9 (ovm, 18H, -OCH2CF3), 6.75 (t, 6.4 Hz, 1H,
C-H), 6.83–6.92 (m, 4H, C-H), 6.99 (t, 6.4 Hz, 1H, C-H), 7.20 (t,
6.4 Hz, 1H, C-H), 7.74 (bs, 2H, C-H). 31P{1H} NMR (CD3CN)
δ/ppm: 17.47 (m), 28.45 (dt, 833.3 Hz, 42.4 Hz), 87.0 (m), 90.64
(dt, 174.90 Hz, 32.39 Hz), 92.45 (dd, 213 Hz, 14.98 Hz), 97.19
(dm, 213 Hz). In C6D6 a multiplet is observed at 2.69 ppm for 3.
As in 2, the multiplet arises from the 3JH-F splitting from the three
3
fluorines of the CF3 group and the JH-P from the phosphorus of
one of the tfepma ligands. IR(CD3CN) νIr-H/cm-1: 2023. Anal.
Calc for C39H43N3O12F36P6Ir2: C, 23.42; H, 2.17; N, 2.10. Found:
C, 23.33; H, 2.08; N, 2.00. Crystals suitable for X-ray diffraction
were picked from the recrystallized product as colorless blocks.
Preparation of Ir2I,III(tfepma)3(µ-C6D4)(C6D5)D (3-d10). Ary-
lation with C6D5MgBr proceeded as in the preparation of 2-d10,
using 328 mg of Ir20,II(tfepma)3Cl2 (0.171 mmol) and 180 µL of
C6D5MgBr (0.5 M THF solution, 2.1 equiv). The material was
worked up as in 2-d10 except that after filtration through Celite the
Et2O solution was allowed to stir at ambient temperature for 16 h,
concentrated to 3 mL, and placed in a freezer at -35 °C overnight
(19) Schultz, A. J.; Carlin, R. L. Acta Crystallogr. B 1995, 51, 43–47.
(20) Schultz, A. J.; De Lurgio, P. M.; Hammonds, J. P.; Mikkelson,
D. J.; Mikkelson, R. L.; Miller, M. E.; Naday, I.; Peterson, P. F.; Porter,
R. R.; Worlton, T. G. Physica B 2006, 385–386, 1059–1061.
(21) Jacobson, R. A. J. Appl. Phys. 1986, 19, 283–286.
(22) Schultz, A. J. Trans. Am. Crystallogr. Assoc. 1987, 23, 61–69.
(23) Schultz, A. J.; Van Derveer, D. G.; Parker, D. W.; Baldwin, J. E.
Acta Crystallogr. C 1990, 46, 276–279.
(24) Larson, A. C.; Von Dreele, R. B. General Structure Analysis System
(GSAS); Los Alamos National Laboratory Report LAUR 86-748, 2004.

1076 Organometallics, Vol. 27, No. 6, 2008
Esswein et al.
to furnish 241 mg (70%) of 3-d10 as a colorless solid. Complex
3-d10 can also be prepared by gently heating solutions of 2-d10 in
an analogous manner to the procedure employed to give complex
1
3 from 2. H NMR (CD3CN) δ/ppm: 2.38 (m, 1H, -OCH2CF3),
2.47 (t, 6.7 Hz, 3H, -NMe), 2.79 (t, 7.3 Hz, 3H, -NMe), 2.87 (t,
9.8 Hz, 3H, -NMe), 3.15–3.35 (ovm, 3H, -OCH2CF3), 3.75–4.02
(m, 2H, -OCH2CF3), 4.1–4.9 (ovm, 18H, -OCH2CF3). 2H NMR
(CH3CN) δ/ppm: -9.70 (dm, 34 Hz, Ir-D), 6.5–7.5 (bm, C-D),
7.79 (bm, C-D). 31P{1H} NMR (CH3CN) δ/ppm: 22.48 (m), 33.56
(ddd, 874.0 Hz, 173.7 Hz, 31.7 Hz), 91.5 (m), 95.57 (ddd, 874.0
Hz 173.7 Hz, 31.7 Hz), 97.57 (dd, 212.3 Hz, 16.5 Hz), 102.14 (ddd,
212.3 Hz, 31.7 Hz, 26.6 Hz). IR(CD3CN) νIr-D/cm-1: 1415. Anal.
Calc for C39H33D10N3O12F36P6Ir2: C, 23.30; H, 2.66; N, 2.09. Found:
C, 23.79; H, 2.09; N, 1.95.
Preparation of Ir2II,II(tfepma)3(H)4 (4). A 374 mg amount of
Ir20,II(tfepma)3Cl2 was dissolved in 4 mL of THF with stirring, and
the solution was frozen. In a separate vial, 2.1 equiv of C6H5MgBr
(410 µL, 1.0 M THF solution) was diluted with 1 mL of THF, and
the solution was frozen. Reaction and workup proceeded as for
complex 2. The resulting Et2O solution (5 mL) was loaded into a
thick-walled glass bomb with a Teflon valve and the vessel attached
to a high-vacuum line. The solution was freeze–pump–thawed (1
× 10-5 Torr) for three cycles followed by the introduction of 1
atm of hydrogen, which was passed through a U-tube immersed in
liquid nitrogen prior to charging the reaction vessel. The reaction
vessel was then sealed, shaken vigorously for 10 min, and allowed
to stand overnight. A colorless precipitate formed and was dissolved
by the addition of 2 mL Et2O. The resulting solution was filtered
and placed in a freezer (-35 °C) overnight to afford 170 mg (47%)
of Ir2II,II(tfepma)3(H)4 (4) as a colorless solid. Dynamic behavior
Figure 1. Solid state structure of Ir2II,II(tfepma)3(µ-C6H4)(C6H5)H
(2). Ellipsoids are drawn at the 50% probability level with -NMe
and -CH2CF3 groups removed for clarity.
-80 °C): 102.5 (m), 110.2 (m), 113.0 (m), 115.4 (m), 118.1 (m),
121.5 (m). IR (CD3CN) νIr-H/cm-1: 2036 (sharp), 1966 (br
shoulder). Anal. Calc for C27H37N3O12F36P6Ir2: C, 17.53; H, 2.02;
N, 2.27. Found: C, 17.64; H, 2.04; N, 2.18. 2 may also serve as a
synthon of 4, but the hydrogenation competes with isomerization
of 2 to 3. In the interest of maximizing the yield of 4, the
hydrogenation was initiated from freshly prepared solutions of 2.
Neutron diffraction experiments were performed at the IPNS at
Argonne National Laboratories (Vide supra) on large single crystals
of 4 (2 × 2 × 1 mm3; 6.8 mg). Hydride ligand fractional
occupancies refine to a value of unity for each ligand, suggesting
that the hydrides are not disordered over the bimetallic core in the
solid state.
1
is evident in the H NMR spectrum of 4. Whereas complexes 2
and 3 exhibit sharp, well-defined, hydride and -NMe resonances,
those of 4 in THF-d8 are considerably broadened. Four -NMe
resonances are observed at 2.59, 2.73, 2.86, and 2.90 ppm. These
results indicated the presence of two products in solution. Integration
reveals a 2:1 ratio for the peaks at 2.73 and 2.90 ppm. A 2:1 ratio
is also identified for the resonances at 2.86 and 2.59 ppm. The two
sets of peaks integrate in a 3:2 ratio, in favor of the peaks at 2.73
and 2.90 ppm. The hydride region also indicates multiple compo-
nents with resonances at -11.51 and -11.64 ppm observed as broad
singlets, and a resonance centered at -14.40 ppm as overlapping
broad multiplets. Integration over all -NMe resonances and all
hydrides gave a 9:4 ratio, as expected for the Ir2II,II(tfepma)3(H)4
formulation. Crystals suitable for X-ray diffraction were grown from
a subsequent reaction following the same procedure as above except
that the reaction mixture in Et2O was allowed to stand for several
days. Large colorless blocks of syn-Ir2II,II(tfepma)3(H)4 deposited
from solution. These crystals cracked easily and crystallinity was
lost at temperatures below -80 °C, necessitating data collection at
elevated temperatures. Consequently significant disorder for the
-CH2CF3 groups of the tfepma ligands is observed. The crystals
not used for X-ray analysis were loaded into a J. Young resealable
NMR tube, which was attached to a high-vacuum line. THF-d8 was
then vacuum transferred into the tube while immersed in liquid
nitrogen. The tube was warmed until the THF thawed, and then it
was immediately transferred into the magnetic field of an NMR
spectrometer maintained at -80 °C. The 1H NMR spectrum showed
the presence of multiple species in solution and neither the hydride
nor the -NMe resonances exhibited noticeable sharpening. 1H
NMR (THF-d8, 20 °C) δ/ppm: -14.65-14.20 (ovm, 3.5H, Ir-H),
-11.51 (bs, 2H, Ir-H), -11.64 (bs, 1.5H, Ir-H), 2.59 (b, 1.5H,
-NMe), 2.70 (bs, 6H, -NMe), 2.86 (b, 3.9H, -NMe), 2.90 (t, 6.7
Results
Synthesis of Ir2II,II(tfepma)3(µ-C6H4)(C6H5)H (2). The
Ir20,II(tfepma)3Cl2 (1) complex provides two convenient sites
for functionalization via the halides of the IrII center. Indeed,
aryl Grignard reagents react with 1 readily to effect the following
conversion,
(3)
The identity of the reaction product 2 was revealed by single crystal
X-ray diffraction analysis. The solid state structure of 2 is presented
in Figure 1, and selected bond lengths and angles are listed in Table
2. The core of the complex is comprised of two octahedral IrII
centers with an Ir-Ir bond distance of 2.7472(4) Å. The pseudo-
octahedral geometry about each iridium center is similar, with
a terminal phenyl and hydride distinguishing the coordination
spheres of Ir(1) and Ir(2), respectively. The presence of a hydride
ligand trans to P(6) is signified by a conspicuously vacant
coordination site ( C(5)-Ir(2)-P(4) ) 162.71(11)° and
P(5)-Ir(2)-Ir(1) ) 166.95(3)°) in the early refinement cycles.
The hydride was eventually located crystallographically but was
1
Hz, 3H, -NMe), 4.05 – 4.86 (ovm, 40H, -OCH2CF3). H NMR
(THF-d8, -80 °C) δ/ppm: -14.35 (ovm, 3.5 H, Ir-H), -11.58 (bs,
1.4 H, Ir-H), -11.405 (bs, 2H, Ir-H), 2.55 (b, 1.5H, -NMe), 2.64
(bs, 6H, -NMe), 2.88 (b, 4H, -NMe), 3.02 (t, 6.7 Hz, 3H, -N
Me), 4.05–4.90 (ovm, 40H, -OCH2CF3). 31P{1H} NMR (THF-d8,

Intramolecular C-H Bond ActiVation
Organometallics, Vol. 27, No. 6, 2008 1077
Table 2. Selected Bond Lengths and Angles for
2.125(4) Å). The constrained four-atom ring created by the
bridging benzyne and iridium atoms distorts the octahedral
geometry of the Ir2II,IIcore (Ir(1): C(15)-Ir(1)-Ir(2) )
71.5(2)°, Ir(2): C(10)-Ir(2)-Ir(1) ) 71.0(3)°) and the angles
around the Ir-C atom connections (C(10): ( Ir(10)-
C(10)-C(15) ) 108.1(6)° and C(11)-C(10)-Ir(2) ) 131.6(7)°,
C(15): Ir(1)-C(15)-C(10))109.1(6)°and C(14)-C(15)-Ir(1)
) 131.6(7)°). Notably the Ir(2)-P(6) bond length of 2.3197(11)
Å is markedly longer than the corresponding Ir(1)-P(2) bond
length of 2.2770(10) Å, reflecting the significant trans influence
of the hydride ligated to Ir(2). All other bond lengths and angles
within the terminally bound and bridging aromatic rings are
normal.
Ir2II,II(tfepma)3(µ-C6H4)(C6H5)H (2)
Bond Lengths (Å)
Ir(1)-Ir(2)
Ir(1)-C(4)
Ir(1)-C(10)
Ir(2)-C(5)
Ir(1)-P(1)
2.7472(4) Ir(1)-P(2)
2.092(4) Ir(1)-P(3)
2.125(4) Ir(2)-P(4)
2.117(4) Ir(2)-P(5)
2.2798(10) Ir(2)-P(6)
2.2770(10)
2.2652(11)
2.2532(10)
2.2228(10)
2.3197(11)
Bond Angles (deg)
P(5)-Ir(2)-Ir(1)
C(5)-Ir(2)-P(4)
C(4)-Ir(1)-Ir(2)
C(5)-Ir(2)-Ir(1)
P(2)-Ir(1)-C(10)
166.95(3)
P(1)-N(1)-P(2)
99.63(18)
69.09(4)
120.9(12)
99.05(18)
68.74(4)
162.71(11) P(1)-Ir(1)-P(2)
71.34(11) P(3)-N(2)-P(4)
71.38(11) P(5)-N(3)-P(6)
167.87(11) P(5)-Ir(2)-P(6)
1
In solution, 2 retains its solid state structure. An H NMR
Dihedral Angles (°)
P(3)-Ir(1)-Ir(2)-P(4) 3.68(4)
C(4)-Ir(1)-Ir(2)-C(5) 2.54(16)
spectrum of 2 obtained in THF-d8 reveals a hydride resonance
at –13.73 ppm as an apparent doublet of doublet of triplets (ddt).
Coupling of the hydride to its trans (2JP-H ) 244.8 Hz) and cis
(2JP-H ) 20.4 Hz, 10.2 Hz) phosphorus neighbors is observed.
The observed splitting pattern for this hydride suggests the
involvement of an additional coupling, either a long-range JH-P
through the Ir-Ir bond or an additional coupling to a 31P nucleus
through the PNP ligand backbone. In either case, the hydride
resonance for 2 should appear as a doublet of doublet of doublet
of doublets (dddd); however the apparent ddt can be rationalized
by similar coupling to two chemically distinct 31P nuclei. The
resonances for the -NMe groups on the diphosphazane ligand
appear as three triplets at 2.74 ppm (3JP-H ) 8.9 Hz), 2.82 ppm
(3JP-H ) 10.2 Hz), and 2.83 ppm (3JP-H ) 6.4 Hz). The doubly
chelating and one bridging binding motif observed in the solid
state for 2 allows for the assignment of the resonances at 2.74
and 2.83 ppm as those that chelate the iridium centers based
on 3JP-H coupling constants. The observed coupling constants for
the chelating tfepma ligands are larger than that of the bridging
tfepma -NMe resonance by ∼2 Hz. The trifluoroethoxy proton
resonances present a characteristically complicated pattern of
overlapping multiplets. The aryl region consists of seven well-
defined resonances. The absence of line broadening in this region
suggests that the coordinated phenyl group on Ir(1) is most likely
freely rotating and is not locked into a single configuration due to
steric congestion with the -OCH2CF3 groups of the tfepma ligand.
Sharp resonances are to be expected for the bridging benzyne. Six
distinct resonances are observed in the 31P{1H} NMR spectrum,
which is well defined but complicated owing to the observation of
2JP-P cis couplings through the PNP ligand backbone and also for
the chelating tfepma ligands through the iridium centers.
fixed at an idealized distance of 1.600 Å in later refinement
cycles. The presence of one bridging and two chelating tfepma
ligands represents the first occurrence of this motif for this ligand
system; the more typical presentation of three tfepma ligands
is for two to bridge the bimetallic core and one to chelate a
single metal. The flexibility of the tfepma PNP backbone is
indicated by the ligand’s ability to accommodate both the tight
bite angle required for chelation ( P(1)-N(1)-P(2) ) 69.09(4)°
and P(5)-N(3)-P(6) ) 68.74(4)°) and the expanded angle
required for bridging a dinuclear core ( P(3)-N(2)-P(4) )
120.9(2)°). The most prominent structural feature of 2 is the
µ-η2-benzyne ligand that spans the diirdium core. Benzyne
complexes are typically encountered when the ligand is bound
to mononuclear metal centers of early and late metals25,26 or
higher nuclearity clusters composed of metals in low oxidation
states.27 The observation of benzyne groups bound to two metals
is uncommon. Benzyne adsorbed on an Ir{100} surface appears
to assume a µ-benzyne bonding mode.28 At the molecular level,
[Ir2(µ-o-C6H4)(η5-C5H5)2(CO)2],29 [Ir2(µ-o-C6H4)(µ-PPh2)(µ-
H)(η5-C5H5)2],30 [Ir2(µ-OH)(µ-PAr2)(µ-C6H3R)(η5-C5Me5)2-
(CO)2] (Ar ) C6H5, p-CH3C6H4 and R ) H, CH3),31 and [Fe2(µ-
o-C6H4)(CO)8]32 all present the benzyne in a bridge position
with the C-C bond of the benzyne moiety parallel to the M-M
axis. Recently a dipalladium complex [(dcpe)Pd(µ-o-C6H4)Pd-
(dcpe)] was reported33 in which the benzyne unit is twisted 46°
from the Pd-Pd bond axis. In 2, the C(4)-C(5) bond length of
1.405(5) Å is more closely akin to the distances observed in
clusters as opposed to monomeric complexes. The benzyne may
be viewed as an aromatic C6H42- moiety in which two mutually
ortho protons are substituted by iridiums from the bimetallic
core. To compare, the Ir(1)-C(4) and Ir(2)-C(5) bond lengths
of 2.092(4) and 2.117(4) Å, respectively, are similar and only
marginally shorter than the sp2-bound phenyl (Ir(1)-C(10) )
The deuterated product, 2-d10, is readily accessed using
C6D5MgBr in place of C6H5MgBr. The solid state structure of
2-d10 is completely analogous to 2 (see Tables 1 and S2.8). The
presence of the hydride, although implicated by crystallographic
studies and the 1H NMR, is unambiguously shown with infrared
spectroscopy. An Ir-H stretching frequency at 2069 cm-1 shifts
to 1453 cm-1 upon deuteration; a simple Hooke’s law calcula-
tion, assuming no change in force constants, predicts a shift to
1463 cm-1 for these stretches upon isotopic substitution of
deuterium for hydrogen. Using the observed stretching frequen-
cies, the deviations in Ir-H/D force constants upon isotopic
substitution is estimated to be fIr-H/fIr–D ) 1.02 for 2.
Isomerization Chemistry. Although both the 1H and 31P{1H}
NMR of 2 indicate a static structure in solution, over time, room-
temperature solutions of 2 slowly isomerize to furnish the
Ir2I,III(tfepma)3(µ-C6H4)(C6H5)H complex 3 in near quantitative
yield. The progress of the isomerization can be monitored by
1H NMR spectroscopy and is accelerated by gentle heating at
50 °C in C6H6 for 6 h. Analysis of the resulting colorless product
(25) Bennet, M. A.; Wenger, E. Chem. Ber. Recl. 1997, 130, 1029–
1042.
(26) Jones, W. M.; Klosin, J. AdV. Organomet. Chem. 1998, 42, 147–
221.
(27) Brait, S.; Deabate, S.; Knox, S. A. R.; Sappa, E. J. Clust. Sci. 2001,
12, 139–173.
(28) Johnson, K.; Sauerhammer, B.; Titmuss, S.; Hing, D. A. J. Chem.
Phys. 2001, 114, 9539–9548.
(29) Rausch, M. D.; Gastinger, R. G.; Gardner, S. A.; Brown, R. K.;
Wood, J. S. J. Am. Chem. Soc. 1977, 99, 7870–7876.
(30) Grushin, V. V.; Vymenits, A. B.; Yanovskii, A. I.; Struchkov, Y. T.;
Vol’pin, M. E. Organometallics 1991, 10, 48–49.
(31) McGhee, W. D.; Foo, T.; Hollander, F. G.; Bergman, R. G. J. Am.
Chem. Soc. 1988, 110, 8543–8545.
(32) Bennet, M. J.; Graham, W. A. G.; Stewart, R. P., Jr. Tuggle, R.
M. Inorg. Chem. 1973, 12, 2944–2949.
(33) Retbøll, M.; Edwards, A. J.; Rae, A. D.; Willis, A. C.; Bennet,
M. A.; Wenger, E. J. Am. Chem. Soc. 2002, 124, 8348–8360.

1078 Organometallics, Vol. 27, No. 6, 2008
Esswein et al.
Table 3. Selected Bond Lengths and Angles for
Ir2I,III(tfepma)3(µ-C6H4)(C6H5)H (3)
Bond Lengths (Å)
Ir(1)-Ir(2)
Ir(1)-C(4)
Ir(2)-C(9)
Ir(2)-C(10)
Ir(1)-P(1)
2.7791(9)
2.141(5)
2.096(4)
2.111(3)
2.2793(10)
Ir(1)-P(2)
Ir(2)-P(3)
Ir(2)-P(4)
Ir(1)-P(5)
Ir(1)-P(6)
2.2585(9)
2.2225(10)
2.2478(9)
2.2585(9)
2.2878(9)
Bond Angles (deg)
C(4)-Ir(1)-P(1)
C(9)-Ir(2)-P(3)
C(4)-Ir(1)-Ir(2)
C(9)-Ir(2)-Ir(1)
P(6)-Ir(1)-Ir(2)
P(2)-Ir(1)-P(5)
151.82(10)
160.98(9)
70.22(9)
71.83(10)
157.30(2)
171.56(4)
P(1)-Ir(1)-P(2)
P(1)-N(1)-P(3)
P(3)-Ir(2)-P(4)
P(2)-N(2)-P(4)
P(5)-N(3)-P(6)
P(5)-Ir(1)-P(6)
93.42(3)
115.37(17)
102.53(3)
121.44(17)
98.81(15)
68.46(3)
Dihedral Angles (deg)
P(1)-Ir(1)-Ir(2)-P(3)
P(2)-Ir(1)-Ir(2)-P(4)
21.58(3)
12.00(3)
C(4)-Ir(1)-Ir(2)-C(9)
C(9)-Ir(2)-C(10)-C(11)
3.05(13)
57.85(30)
implies a slightly reduced force constant (fIr-H/fIr-D ) 1.03). The
terminal phenyl resides along the Ir-Ir bond axis in an axial
coordination site with a C(10)-Ir(2)-Ir(1) angle of 160.73(10)°
and an Ir(2)-C(10) bond distance of 2.111(3) Å. Whereas the
C6H42- unit of the bridging benzyne maintains a nearly planar
configuration with the diiridium core (C(4)-Ir(1)-Ir(2)-C(9)
dihedral ) 3.05(13)°), the terminal phenyl group is twisted by
57.85(30)° with respect to the plane comprised of the diiridium
benzyne unit as measured by the C(9)-Ir(2)-C(10)-C(11)
dihedral angle. Examination of the steric environments about
the terminal phenyl group in the solid state reveals an expanded
pocket relative to that observed for 2, explaining the observation
of slightly broadened resonances for the terminal phenyl protons
in the 1H NMR of 3. The bridging benzyne distorts the
octahedral geometries of the iridium centers. The angles within
the four-atom bridge are significantly contracted ( C(4)-Ir-
(1)-Ir(2) ) 70.22(9)° and C(9)-Ir(2)-Ir(1) ) 71.83(10)°)
and the iridium-carbon contacts are more asymmetric in 3
(d(Ir(1)-C(4)) ) 2.141(3) Å and d(Ir(2)-C(9)) ) 2.096(4) Å
∆(3) ) 0.045 Å vs ∆(2) ) 0.025 Å) than in 2. The shortest
contact observed for the benzyne carbons is that bonded to the
IrIII center. This is most likely due to the slightly contracted
ionic radius of IrIII vs IrII. The rearrangement of hydride and
phenyl to terminal locations on the same iridium center
represents a formal disproportionation of the valence symmetric
Figure 2. Solid state structure of Ir2I,III(tfepma)3(µ-C6H4)(C6H5)H
(3). Ellipsoids are drawn at the 50% probability level with -NMe
and -CH2CF3 groups removed for clarity.
1
by H NMR in CD3CN reveals a complex with three -NMe
resonances as sharp triplets at 2.47, 2.79, and 2.87 ppm. The
resonances integrate in a 1:1:1 ratio and exhibit 3JP-H coupling
constants of 7.0, 6.4, and 9.8 Hz, respectively. Examination of
the coupling constants indicates that the tfepma ligands have
rearranged from a doubly chelating and one bridging geometry
as in 2 to a doubly bridging and one chelating motif as shown
in eq 4,
(4)
The hydride resonance at -9.79 ppm for 2 is shifted
downfield considerably from -13.73 ppm for that in 3,
indicating that its chemical environment has been significantly
altered. The aryl region of 3, as in 2, exhibits seven resonances,
several of which overlap considerably. A broadened singlet
resonance at 7.74 ppm indicates some degree of rotational
freedom for the hydrogens on the phenyl ring. The 31P{1H}
NMR spectrum of 3 displays six distinct resonances and exhibits
II,II
I,III
Ir2
core to a two-electron mixed valence Ir2
core. The
formal oxidation states and the Ir-Ir single bond in the two-
electron mixed valence complex 3 are rationalized by a dative
2
bonding interaction from a filled dz orbital (taking the diiridum
I
2
vector as the z axis) on the Ir center into an empty dz orbital
on the IrIII center. This general bonding description has been
described in detail for two-electron mixed valence Rh2
0,II
2
dimers.34
complex splitting patterns resulting from multiple JP-P cou-
plings similar to those of 2.
The rate of the 2 f 3 conversion was monitored by 1H NMR
spectroscopy. Although sparingly soluble, C6D6 was chosen for
the reaction medium, as in this solvent the -NMe resonances
exhibit little overlap. Mixtures of 2 and 3 display six well-
defined triplets in this the region between 1.97 and 2.60 ppm
(Figure 3). The resonances at 1.97, 2.17, and 2.60 ppm are
assigned to 2, and the resonances at 2.02, 2.12, and 2.48 ppm
to 3. Examination of the isomerization end point shows no
incorporation of the deuterium labels from the C6D6 solvent
into the coordination sphere of 3. Integration of the hydride
and -NMe resonances at 2.60 ppm and 2.48 ppm, respectively,
gives the expected ratio of 3:1 in the final product. Removal of
all volatiles followed by addition of CD3CN allows for
The molecular structure of 3 is definitively identified by X-ray
structural analysis, the results of which are shown in Figure 2.
Selected bond lengths and angles are listed in Table 3. The Ir-Ir
separation in 3 lengthens slightly to 2.7791(2) Å. Consistent
3
1
with the JP-H coupling constants observed in the H NMR,
the more common ligation motif of two bridging and one
chelating tfepma is found in complex 3. The equatorial hydride
is inferred by the open coordination site trans to P(4)
( C(10)-Ir(2)-Ir(1) ) 160.73(10)° and C(9)-Ir(2)-P(3) )
160.98(9)°). The hydride was located late in the refinement and
was fixed at a bond length of 1.600 Å. Infrared spectroscopy
shows that the Ir-H stretching frequency shifts appropriately
from 2023 to 1415 cm-1 upon isotopic substitution of deuterium
for hydrogen (vIr–D(calc) ) 1426 cm-1). Analogous to complexes
2 and 2-d10, the small deviation from calculated isotopic shifts
(34) Nocera, D. G. Acc. Chem. Res. 1995, 28, 209–217.

Intramolecular C-H Bond ActiVation
Organometallics, Vol. 27, No. 6, 2008 1079
Figure 3. Segment of the 1H NMR spectrum for a partially
isomerized mixture of 2 and 3 in C6D6. The spectrum emphasizes
the six triplet -NMe resonances, with arrows indicating the course
of the spectral evolution with time. For the kinetic analysis, the
resonances at 2.60 and 2.48 ppm were employed as signatures of
2 and 3, respectively. Resonances indicated by a star are those of
methylene protons on the -OCH2F3 groups of the tfepma ligands
identified in pure samples of 2 and 3.
Figure 4. Plot of kobs vs 1000/T for the isomerization of 2 to 3
(triangles) and 2-d10 to 3-d10 (circles) from which the activation
parameters listed in Table 4 were extracted.
Whereas the iridium phenyl hydride functionality is dynamic
in 2, it is not in 3. Even with the favorable cis coordination of
a hydride and a phenyl group on the IrIII center, no evidence
for exchange of the terminal C6H5, benzyne C6H4, or hydride
with solvent is observed after extended thermolysis in C6D6 at
Table 4. Observed Rate Constants and Calculated Activation
Parameters for the Isomerization of 2 to 3 and 2-d10 to 3-d10
1
70 °C (24–48 h) as assessed by H NMR spectroscopy. At
kobs (s-1
)
temperatures greater than 80 °C, only decomposition of the
bimetallic core is observed.
22.2 °C
31.8 °C
41.5 °C
2.94 × 10-5
9.57 × 10-5
3.09 × 10-4
Hydrogenation Chemistry. Observation of an inverse iso-
tope effect for the isomerization of 2 and 2-d10 implies the
involvement of C-H(D) bonds in the transition state. Though
the intermediacy of σ-C6H6 or η2-π-C6H6 in the reductive
elimination of benzene has been debated in the literature,35,36
the simplest postulate for C-H(D) involvement in this report
entails a reductive coupling of the benzyne and hydride on a
single iridium center to generate an intermediate with either an
open coordination site or a labile σ-C6H5 or η2-π-C6H6 complex
to a low-valent Ir0 center as follows,
2
2-d10
5.14 × 10-5
2.17 × 10-4
7.52 × 10-4
activation parameters
∆Hq (kcal/mol)
21.7 ( 0.3
∆S‡q (eu)
-7.4 ( 0.9
-6.4 ( 2.4
2
2-d10
24.9 ( 0.7
integration of the -NMe resonance at 2.87 ppm versus and the
aryl region without interference from residual protio resonance
from C6D6; the expected ratio of 3:9 is obtained. Additionally,
no signals from intermediates are observed during the course
of the reaction; the only observed NMe, aryl, or hydride
resonances at any point in the reaction are assignable to either
2 or 3.
(5)
The resonances of the bridging tfepma ligands at 2.60 and
2.48 ppm appear in an uncongested portion of the spectrum
and consequently can be reliably integrated. The integrated
intensities for the peaks at 2.60 and 2.48 ppm fit a first-order
exponential decay and growth, respectively (Figure S6-S8).
The observed rate constants, listed in Table 4, agree within
experimental error. kobs increases with increasing temperature,
and a linear fit of ln(kobs/T) vs 1000/T (Figure 4) yields the
activation parameters shown in Table 4. The effects of isotopic
substitution could be ascertained using complex 2-d10. The
isomerization of 2-d10 to 3-d10 exhibits significant rate enhance-
ment relative to its nondeuterated parent complex (Table 4).
The observed rate constant of 2.17 × 10-4 s-1 at 31.8 °C in
Under the canopy of this postulate, we sought to trap a
coordinatively unsaturated intermediate. The use of a 10-fold
excess of PMe3 or tBuNC as a trapping reagent, however, gave
multiple intractable products with overlapping resonances in the
1H and 31P{1H} NMR. Hydrogen was chosen as a candidate
since previous studies show that diiridium complexes related
to 2 show a proclivity to readily activate hydrogen via vacant
coordination sites.10,11,16 We observed a smooth reaction of 2
with hydrogen to form the tetrahydride Ir2II,II(tfepma)3(H)4 (4)
with concomitant formation of two benzene molecules (Scheme
1). The syn isomer was characterized by X-ray crystallography.
A view along the Ir-Ir bond axis of 4-syn is presented in Figure
5 and reveals a symmetric, triply bridged, tfepma core with the
bridging ligands twisted ∼24° about the Ir-Ir bond axis.
Selected bond lengths and angles are listed in Table 5. The core
of complex 4-syn is comprised of two octahedral iridium centers
separated by 2.8389(3) Å with overall pseudo-C2V symmetry.
D
C6D6 solution yields an inverse isotope effect of kobsH/kobs of
0.44 ( 0.1. As for 2, the activation parameters were calculated
from kinetic data measured over the range 20–40 °C and yielded
a ∆Hq of 24.9 ( 0.7 kcal/mol and a ∆Sq of -6.4 ( 2.4 eu
(Table 4). The inverse isotope effect is striking, as it indicates
significant involvement of Ir-H(D)/C-H(D) bond making/
breaking along the reaction coordinate. The isotope effect and
the lack of observable intermediates over the course of the
isomerization strongly indicate that the rate-determining step
takes the form of an Ir-H(D)/C-H(D) bond making/breaking
event, with any subsequent rearrangement steps leading to the
formation of 3 occurring fast on the NMR time scale.
(35) Johansson, L.; Tilset, M.; Labinger, J. A.; Bercaw, J. E. J. Am.
Chem. Soc. 2000, 122, 10846–10855.
(36) Albéniz, A. C.; Schulte, G.; Crabtree, R. H. Organometallics 1992,
11, 242–249.

1080 Organometallics, Vol. 27, No. 6, 2008
Esswein et al.
Scheme 1
The elongated Ir-Ir bond (relative to 2 and 3) most likely results
from the absence of the bridging benzyne and to the presence
of two strong trans-influencing terminal hydrides located in axial
coordination sites relative to the metal-metal bond. Without
the strained benzyne bridging unit, the geometry about the
iridium centers is closer to that expected for an octahedron with
cis phosphite angles in the range 95–99° (Table 5). Only a slight
decrease in the trans-spanning phosphite angles are observed
( P(1)-Ir(1)-P(3) ) 165.13(3)° and P(4)-Ir(2)-P(6) )
164.09(4)°) most likely due to the small hydride ligands in the
equatorial positions.
immediately placed into the magnetic field of an NMR
spectrometer at -80 °C. Despite these precautionary measures,
multiple products were observed and the band shape of the
hydride resonances were not significantly sharpened. These
results suggest that even at low temperatures the molecular
structure of Ir2II,II(tfepma)3(H)4 is dynamic in solution but adopts
a single conformation in the solid state. Based on results for
related rhodium hydride complexes examined in our laborato-
ries,37 the dynamic behavior of 4 in solution is most likely due
to a syn to anti isomerization, Scheme 1 bottom.
Though the hydrides of complex 4 could not be located
reliably in the difference Fourier map obtained from X-ray
diffraction, they were readily located using single-crystal neutron
diffraction. Solution of the neutron diffraction data provides the
structure presented in Figure 5 (right), consistent with that found
by X-ray crystallographic techniques, except that the two
pseudooctahedral iridium centers are completed by the presence
of four crystallographically located terminal hydride ligands.
The hydrides exhibit slightly compressed cis angles of 87.1(16)°
and 81.0(16)° for the H(1)-Ir(1)-H(2) and H(3)-Ir(2)-H(4),
respectively, and form nearly linear H-Ir-P angles with the
trans phosphite on each iridium center (see Table 5). An acute
H(2)-Ir(1)-Ir(2)-H(4) dihedral angle of 32.2(12)° confirms
the syn disposition of the hydrides about the bimetallic core.
The hydrides refine at full occupancy and disfavor the participa-
The 1H NMR of 4 reveals two sets of hydride resonances at
-11.51 and -14.25 ppm tentatively assigned to the axially and
equatorially bound (relative to the Ir-Ir internuclear axis)
hydrides, respectively. Four broadened resonances at 2.59, 2.70,
2.86, and 2.90 ppm are observed in the region typical of a
-NMe group. Integration over all the –NMe and hydride
resonances gives a ratio of 9:4, as expected for the
Ir2II,II(tfepma)3(H)4 formulation. Owing to the breadth of the
hydride signals, variable-temperature NMR studies were con-
ducted. Solutions prepared from single crystals of syn-
Ir2II,II(tfepma)3(H)4 (4-syn) were placed in a resealable NMR
tube, attached to a high-vacuum line, and THF-d8 was vacuum-
transferred into the tube while maintained at 77 K. The solution
was then allowed to warm until the solvent thawed and then
Figure 5. Single-crystal structure determinations for Ir2II,II(tfepma)3(H)4 (4) by X-ray (left) and neutron (right) diffraction methods. The
neutron data unambiguously reveal the presence of four terminal hydrides ligated in a syn disposition over the bimetallic core of 4. -NMe,
-CH2CF3 groups are omitted for clarity, with thermal ellipsoids drawn at the 50% probability level. Displacement parameters were isotropically
refined in the neutron structure, with ellipsoids drawn at the 50% probability level.

Intramolecular C-H Bond ActiVation
Organometallics, Vol. 27, No. 6, 2008 1081
Table 5. Selected Bond Lengths and Angles for syn-Ir2II,II(tfepma)3(H)4 (4-syn) obtained from X-ray and Neutron Diffraction Methods
X-ray
neutron
Bond Lengths (Å)a
X-ray
neutron
Ir(1)-Ir(2)
Ir(1)-P(1)
Ir(1)-P(2)
Ir(1)-P(3)
Ir(2)-P(4)
Ir(2)-P(5)
2.8389(2)
2.2420(9)
2.2514(9)
2.2263(8)
2.2223(9)
2.2467(9)
2.816(11)
2.278(21)
2.242(22)
2.213(20)
2.255(25)
2.193(23)
Bond Angles (deg)a
165.4(9)
97.1(8)
97.1(8)
163.0(9)
99.9(9)
Ir(2)-P(6)
Ir(1)-H(1)
Ir(1)-H(2)
Ir(2)-H(3)
Ir(2)-H(4)
2.2323(9)
2.242(19)
1.636(30)
1.647(33)
1.599(30)
1.612(32)
P(1)-Ir(1)-P(3)
P(1)-Ir(1)-P(2)
P(2)-Ir(1)-P(3)
P(4)-Ir(2)-P(6)
P(4)-Ir(2)-P(5)
165.13(3)
P(5)-Ir(2)-P(6)
H(1)-Ir(1)-H(2)
H(2)-Ir(1)-P(2)
H(3)-Ir(1)-H(4)
H(4)-Ir(1)-P(5)
95.66(3)
23.95(4)
96.6(8)
87.0(16)
174.4(12)
81.0(16)
175.5(13)
97.02(3)
97.50(3)
165.09(4)
99.84(3)
Dihedral Angles (deg)a
22.5(7)
26.1(7)
P(1)-Ir(1)-Ir(2)–P(4)
P(2)-Ir(1)-Ir(2)–P(5)
23.01(4)
25.74(4)
P(3)-Ir(1)-Ir(2)–P(6)
H(2)-Ir(1)-Ir(2)–H(4)
24.0(7)
32.2(12)
a The hydride nomenclature in the table, H(1), H(2), H(3), and H(4), corresponds to H(b), H(c), H(d), and H(a), respectively, in the cif file.
Scheme 2
tion of site disordered hydrides bridging the diiridium core
in the solid state. Although dynamic in solution, the hydrides
of 4 are resistant to exchange. Solutions of 4 in THF-d8 when
exposed to 1 atm of deuterium show no incorporation of label
and maintain a static integral ratio of -NMe to hydride in the
1H NMR spectrum after 1 week.
Unlike 2, 3 did not react with hydrogen, and the complex
decomposed after days at elevated temperatures (80–100 °C,
Scheme 1). This observation is consistent with the stability of
3 and suggests that a coordinatively unsaturated intermediate
cannot be attained with hydride in an equatorial position cis to
both the benzyne and terminal phenyl ligands.
equilibrium effect as opposed to inverse kinetic isotope effect
Discussion
(i.e., KH/D < 153 and not kH/kD < 1). Within this context, two
distinct cases may give rise to an inverse isotope effect for the
isomerization of 2 to 3.46,47 In the first case (Scheme 2, left)
partial formation or breakage of the C-H(D) bond reduces the
force constant relative to free C-H(D), and thus the ZPE (zero-
point energy) for this intermediate species is presumed to fall
between that of free C-H(D) and M-H(D). If this species is
implicated in the transition state of the rate-determining step,
then the differences in ZPE should engender a slight energetic
preference (in the form of ∆Gq) for reductive coupling in the
deuterated complex. This thermodynamic preference would
Bimetallic cooperativity at two-electron mixed valence cores
promotes the activation of C-H bonds. The bimetallic core of
1 provides a source of a “masked”, electron-rich Ir0 center
adjacent to a coordinatively unsaturated IrII center. Because the
bimetallic core is unsaturated, it is able to accept ligands into
the bridging position and maintain a two-electron redox asym-
metry. The formation of the benzyne complex subsequent to
arylation is extremely fast and cannot be observed on NMR
time scales. The subsequent isomerization of the kinetic product
2, to its thermodynamic product, 3, however occurs over hours
at room temperature. This isomerization is enabled by coopera-
tion of the diiridium centers, as there is no evidence that C6H6
is released from the bimetallic core followed by its readdition
to the reduced metal center.
An inverse isotope effect for the isomerization rates of 2 and
2-d10 in C6D6 implicates the involvement of C-H(D) bond
forming/breaking reactions in the rate-determining step. Inverse
isotope effects have been observed for the reductive coupling
of alkyl hydrides38–50 and less commonly aryl hydrides.49–52
Inverse isotope effects are generally attributed to inverse
D
result in rate enhancement for this step, i.e., kRCH/kRC < 1.
(44) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1996,
118, 5961–5976.
(45) Wick, D. G.; Reynolds, K. A.; Jones, W. D. J. Am. Chem. Soc.
1999, 121, 3964–3983.
(46) Northcutt, T. O.; Wick, D. G.; Vetter, A. J.; Jones, W. D. J. Am.
Chem. Soc. 2001, 123, 7257–7270.
(47) Lo, C. H.; Haskel, A.; Kapon, M.; Keinan, E. J. Am. Chem. Soc.
2002, 124, 3226–3228.
(48) Iron, M. A.; Lo, C. H.; Martin, J. M. L.; Keinan, E J. Am. Chem.
Soc. 2002, 124, 7041–7054. Although the explanations for the observed
isotope effects have been recently disputed in the literature, see refs 49
and 50.
(37) Esswein, A. J.; Veige, A. S. ; Nocera, D. G. J. Am. Chem. Soc.
2005, 127, 16641–16651.
(49) Churchill, D. G.; Janak, K. E.; Wittenburg, J. S.; Parkin, G. J. Am.
Chem. Soc. 2003, 125, 1403–1420.
(38) Buchanan, J. M.; Stryker, J. M.; Bergman, R. G. J. Am. Chem.
Soc. 1986, 108, 1537–1550.
(50) Jones, W. D. Acc. Chem. Res. 2003, 36, 140–146.
(51) Jones, W. D.; Feher, F. J. J. Am. Chem. Soc. 1986, 108, 4814–
4819.
(39) Periana, R. A.; Bergman, R. G. J. Am. Chem. Soc. 1986, 108, 7332–
7346.
(52) Watson, W. H.; Wu, G.; Richmond, M. G. Organometallics 2006,
(40) Bullock, M. R.; Headford, C. E. L.; Hennessy, K. M.; Kegley, S. E.;
Norton, J. R. J. Am. Chem. Soc. 1989, 111, 3897–3908.
(41) Parkin, G.; Bercaw, J. E. Organometallics 1989, 8, 1172–1179.
(42) Gould, G. L.; Heinekey, D. M. J. Am. Chem. Soc. 1989, 111, 5502–
5504.
25, 930–945.
H/D
(53) The equilibrium isotope effect Keq
can be expressed as:
(43) Wang, C.; Ziller, J. W.; Flood, T. C. J. Am. Chem. Soc. 1995, 117,
1647–1648.

1082 Organometallics, Vol. 27, No. 6, 2008
Esswein et al.
Scheme 3
reaction could be facilitated by prior dissociation of one arm
of a tfepma ligand to generate a five-coordinate intermediate,
as has been postulated for reductive elimination from a variety
of PtIV complexes,54–57 followed by phenyl migration through
an µ-η1-C6H5 intermediate.58–62 Subsequent relaxation from
µ-η1-C6H5 to η1-C6H5 unifies the two pathways via a common
intermediate. From this intermediate, C-H(D) oxidative cleav-
age at the IrI center would generate the observed product 3. A
third mechanistic possibility (C) involving initial formation of
a valence symmetric Ir2II,II bridging hydride followed by valence
I,III
disproportionation to an Ir2
core without a C-H(D) bond
formation/cleavage event is thought to be less likely than
mechanisms A or B proposed in Scheme 3. The rationale begins
by noting that a bridging hydride should have a lower force
constant than a terminal hydride. This means that the ∆∆ZPE
difference between a terminal hydride/deuteride and a bridging
hydride/deuteride transition state would be very small. Following
the same logical progression as applied previously suggests that
this small ∆∆ZPE would result in smaller energetic preferences
for either the protio or deutero congeners relative to the ∆∆ZPEs
for terminal M-H(D) and C-H(D) bonds. This means that the
effects of isotopic substitution on a bridging hydride transition
state should be small. The inverse isotope effect observed in
this reaction however is large, ca. ∼0.44, suggesting that the
transition states separating 2 from 3 or 2-d10 from 3-d10 confront
significant energetic preferences resulting from isotopic substitu-
tion. This scenario is more consistent with C-H(D) bond
cleavage/formation events along the reaction coordinate.63
Unfortunately, the two favored pathways, A and B, are
kinetically indistinguishable in the present study as 2 converted
smoothly to 3 without the observation of any intermediates.
Notwithstanding, either pathway relies on the cooperation of
the valence asymmetric centers of the bimetallic core.
The microscopic reverse, oxidative cleavage, would be initiated
from the σ-alkane product, which possesses intact C-H(D)
bonds. Applying the same ZPE arguments as for the reductive
coupling step would suggest a slight energetic preference for
H
the protio case and, thus, a normal isotope effect, i.e., kOC
/
D
kOC > 1. Taking a combination of these rate constants gives
an overall inverse equilibrium isotope effect. Alternatively, the
C-H(D) stretching frequency is assumed to disappear as it
becomes the reaction coordinate for the bond scission/formation.
In this case (Scheme 2, right), normal kinetic isotope effects
are expected for both the reductive coupling of the alkyl hydride
and the oxidative cleavage of the resulting σ-alkane complex.
The inverse isotope effect for the overall transformation arises
due to the juxtaposition of the two normal kinetic isotope effects
for the individual steps. In particular, when the magnitude of
the normal kinetic isotope effect for oxidative cleavage of the
σ-alkane complex is greater than that for the reductive coupling,
an overall inverse isotope effect results. In general it is difficult
to determine quantitatively whether the observed inverse isotope
effect results from of a true inverse kinetic isotope effect or an
inverse equilibrium isotope effect arising from two normal
kinetic isotope effects.46,50
In summary, we have demonstrated that ortho C-H bond
activation of coordinated phenyl groups across the metal-metal
0,II
bond in two-electron mixed valent Ir2 cores is facile. The
primary C-H bond activation product, 2, is a kinetic product,
which undergoes a room-temperature redox isomerization to
I,III
form the Ir2
core of 3 as the thermodynamic product. The
More phenomenologically, the observation of an inverse
isotope effect for alkyl hydride complexes has become a
standard measure for the intermediacy of σ-alkane complexes50
and some arene complexes.49,51 A complication for the latter is
that the precise nature of the reductive coupling product is more
ambiguous as both arene σ-C–H interactions (agostic type) and
an η2-π-arene interactions are possible as opposed to alkyl
hydrides, where only an agostic C-H σ-alkane complex is a
tenable intermediate.
rate-determining step is implicated to involve C-H bond
making/breaking step(s) on the basis of the observed inverse
isotope effects and empirically by the interception of a reaction
intermediate by hydrogen to give the tetrahydride species 4.
The kinetic data collected thus far, however, do not allow for
the inverse isotope effect to be assigned as an inverse equilib-
rium isotope effect or a true inverse kinetic isotope effect for a
single step (i.e., determination between case 1 and case 2 in
Scheme 2), though to our knowledge, this report constitutes the
first example of an inverse isotope effect involving C-H bonds
Along these lines, the observation of an inverse isotope effect
on the isomerization from 2 to 3 vs 2-d10 to 3-d10 likely
implicates an intermediate with σ-C6H5 or η2-π-C6H6 character
resulting from reductive coupling along the reaction coordinate.
Three reasonable pathways for the isomerization are shown in
Scheme 3. Along path B, the reaction is initiated with the
migration of the hydride to bridge the bimetallic core. The
hydride then reductively couples with the benzyne at the adjacent
iridium center to generate a valence symmetric d8 · · · d8 diiridium
core each ligated by a terminal phenyl group. Oxidative cleavage
(54) Crumpton, D. M.; Goldberg, K. I. J. Am. Chem. Soc. 2000, 122,
962–963.
(55) Fekl, U.; Goldberg, K. I. J. Am. Chem. Soc. 2002, 124, 6804–
6805.
(56) Jensen, M. P.; Wick, D. D.; Reinartz, S.; White, P. S.; Templeton,
J. L.; Goldberg, K. I. J. Am. Chem. Soc. 2003, 125, 8614–8624.
(57) Crumpton-Bregel, D. M.; Goldberg, K. I. J. Am. Chem. Soc. 2003,
125, 9442–9456.
(58) Evans, D. G.; Hughes, G. R.; Mingos, D. M. P.; Bassett, J. M.;
Welch, A. J. J. Chem. Soc., Chem. Commun. 1980, 1255–1257.
(59) Nobel, D.; van Koten, G.; Spek, A. L. Angew. Chem., Int. Ed. Engl.
1989, 28, 208–210.
of an ortho C-H(D) bond will generate the two-electron mixed
I,III
valence Ir2
center of 3. We note that this pathway is an
(60) Meyer, E. M.; Gamborotta, S.; Floriani, C.; Chiesi-Villa, A.;
Guastini, C. Organometallics 1989, 8, 1067–1079.
(61) Bradford, C. W.; Nyholm, R. S.; Gainsford, G. J.; Guss, J. M.;
Ireland, P. R.; Mason, R. J. Chem. Soc., Chem. Commun. 1972, 87–89.
(62) Hlavinka, M. L.; Hagadorn, J. R. Organometallics 2005, 24, 5335–
5341.
intramolecular analogue of dinuclear elimination (eq 1) and is
operative for H2 addition and generation (eq 2) at two-electron
mixed-valence cores. Alternatively, path A in Scheme 3 begins
with reductive coupling of the hydride and benzyne, generating
a two-electron mixed valence Ir20,II core. The reductive coupling
(63) We thank a reviewer for prompting this clarification.

Intramolecular C-H Bond ActiVation
Organometallics, Vol. 27, No. 6, 2008 1083
measured in a bimetallic system. Finally, the results reported
herein are noteworthy inasmuch as two-electron mixed valency
at bimetallic cores tends to render complexes inert. Bosnich
postulates1–3 in his studies of binucleating complexes that
bimetallic cores supported with a rigid ligand framework arrest
multielectron redox reactivity of mixed-valence complexes. By
employing a flexible ligand framework that permits the elec-
tronic and steric asymmetry of “multielectron” mixed valence
cores to be accommodated, C-H activation pathways that are
generally the purview of monometallic complexes become
available at bimetallic centers working in concert.
MRSEC Program of the National Science Foundation under
award number DMR 02-13282, and the Department of
Chemistry Instrument Facility (DCIF) at MIT for NMR data
collection supported by the National Science Foundation
under award number CHE-9808061 and DBI-9729592. Work
at Argonne National Laboratory was supported by the U.S.
Department of Energy, Office of Science, Office of Basic
Energy Sciences, under contract DE-AC02-06CH11357.
Supporting Information Available: Kinetic data for the isomer-
ization of 2 to 3 and 2-d10 to 3-d10, NMR probe temperature
calibration, thermal ellipsoid plots, tables of crystal data, atomic
coordinates, bond lengths and angles, anisotropic thermal param-
eters, and hydrogen coordinates for complexes 2, 2-d10, 3, and 4-syn
are available in CIF format free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. A.J.E. would like to thank Prof. Joseph
Sadighi and Prof. Theodore Betley for many helpful discus-
sions. This work was supported under grants from the
National Science Foundation (CHE-0450058). This work
made use of the Shared Experimental Facilities (for single-
crystal X-ray diffraction experiments) supported by the
OM7007748