Role of coordination geometry in dictating the barrier to hydride migration in d6 square-pyramidal iridium and rhodium pincer complexes

Article abstract of DOI:10.1021/ja204851x

Syntheses of the olefin hydride complexes [(POCOP)M(H)(olefin)][BAr ^f₄] (6a-M, M = Ir or Rh, olefin = C₂H ₄; 6b-M, M = Ir or Rh, olefin = C₃H₆; POCOP = 2,6-bis(di-tert-butylphosphinito)benzene; BAr^f = tetrakis(3,5- trifluoromethylphenyl)borate) are reported. A single-crystal X-ray structure determination of 6b-Ir shows a square-pyramidal coordination geometry for Ir, with the hydride ligand occupying the apical position. Dynamic NMR techniques were used to characterize these complexes. The rates of site exchange between the hydride and the olefinic hydrogens yielded δG^? = 15.6 (6a-Ir), 16.8 (6b-Ir), 12.0 (6a-Rh), and 13.7 (6b-Rh) kcal/mol. The NMR exchange data also established that hydride migration in the propylene complexes yields exclusively the primary alkyl intermediate arising from 1,2-insertion. Unexpectedly, no averaging of the top and bottom faces of the square-pyramidal complexes is observed in the NMR spectra at high temperatures, indicating that the barrier for facial equilibration is >20 kcal/mol for both the Ir and Rh complexes. A DFT computational study was used to characterize the free energy surface for the hydride migration reactions. The classical terminal hydride complexes, [M(POCOP)(olefin)H]⁺, are calculated to be the global minima for both Rh and Ir, in accord with experimental results. In both the Rh ethylene and propylene complexes, the transition state for hydride migration (TS1) to form the agostic species is higher on the energy surface than the transition state for in-place rotation of the coordinated C-H bond (TS2), while for Ir, TS2 is the high point on the energy surface. Therefore, only for the case of the Rh complexes is the NMR exchange rate a direct measure of the hydride migration barrier. The trends in the experimental barriers as a function of M and olefin are in good agreement with the trends in the calculated exchange barriers. The calculated barriers for the hydride migration reaction in the Rh complexes are ～2 kcal/mol higher than for the Ir complexes, despite the fact that the energy difference between the olefin hydride ground state and the agostic alkyl structure is ～4 kcal/mol larger for Ir than for Rh. This feature, together with the high barrier for interchange of the top and bottom faces of the complexes, is proposed to arise from the unique coordination geometry of the agostic complexes and the strong preference for a cis-divacant octahedral geometry in four-coordinate intermediates.

Full text of DOI:10.1021/ja204851x

ARTICLE
pubs.acs.org/JACS
Role of Coordination Geometry in Dictating the Barrier to Hydride
Migration in d6 Square-Pyramidal Iridium and Rhodium
Pincer Complexes
Michael Findlater, Alison Cartwright-Sykes, Peter S. White, Cynthia K. Schauer,* and Maurice Brookhart*
Department of Chemistry, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290, United States
S Supporting Information
b
ABSTRACT: Syntheses of the olefin hydride complexes
[(POCOP)M(H)(olefin)][BArf4] (6a-M, M = Ir or Rh, olefin =
C2H4; 6b-M, M = Ir or Rh, olefin = C3H6; POCOP = 2,6-bis-
(di-tert-butylphosphinito)benzene; BArf = tetrakis(3,5-trifluoro-
methylphenyl)borate) are reported. A single-crystal X-ray struc-
ture determination of 6b-Ir shows a square-pyramidal coordina-
tion geometry for Ir, with the hydride ligand occupying the apical
position. Dynamic NMR techniques were used to characterize
these complexes. The rates of site exchange between the hydride
and the olefinic hydrogens yielded ΔGq = 15.6(6a-Ir), 16.8 (6b-Ir),
12.0 (6a-Rh), and 13.7 (6b-Rh) kcal/mol. The NMR exchange data also established that hydride migration in the propylene
complexes yields exclusively the primary alkyl intermediate arising from 1,2-insertion. Unexpectedly, no averaging of the top and
bottom faces of the square-pyramidal complexes is observed in the NMR spectra at high temperatures, indicating that the barrier
for facial equilibration is >20 kcal/mol for both the Ir and Rh complexes. A DFT computational study was used to characterize
the free energy surface for the hydride migration reactions. The classical terminal hydride complexes, [M(POCOP)(olefin)H]+,
are calculated to be the global minima for both Rh and Ir, in accord with experimental results. In both the Rh ethylene and propylene
complexes, the transition state for hydride migration (TS1) to form the agostic species is higher on the energy surface than
the transition state for in-place rotation of the coordinated CꢀH bond (TS2), while for Ir, TS2 is the high point on the energy
surface. Therefore, only for the case of the Rh complexes is the NMR exchange rate a direct measure of the hydride migration barrier.
The trends in the experimental barriers as a function of M and olefin are in good agreement with the trends in the calculated
exchange barriers. The calculated barriers for the hydride migration reaction in the Rh complexes are ∼2 kcal/mol higher than for
the Ir complexes, despite the fact that the energy difference between the olefin hydride ground state and the agostic alkyl structure is
∼4 kcal/mol larger for Ir than for Rh. This feature, together with the high barrier for interchange of the top and bottom faces of the
complexes, is proposed to arise from the unique coordination geometry of the agostic complexes and the strong preference for a cis-
divacant octahedral geometry in four-coordinate intermediates.
Scheme 1. Classical Picture of Migratory Insertion in an
Ethylene Hydride Complex
’ INTRODUCTION
Migratory insertion reactions of transition metal olefin hy-
dride and olefin alkyl complexes are key transformations in a
variety of catalytic reactions, including olefin hydrogenation, hydro-
formylation, isomerization, polymerization, and oligomerization.1
The classical view of the migratory insertion reaction of a metal
olefin hydride complex is illustrated in Scheme 1 with ethylene
hydride complex, 1, and involves the transformation of this species
to a high energy unsaturated ethyl intermediate, 3, via a transition
state, 2, in which hydrogen is bridged between the metal and the
β-carbon. The barriers to such migratory insertion processes have
often been examined by measuring the rates of hydrogen scram-
bling between the metal hydride and olefinic sites through either
dynamic NMR studies or, on slower time scales, through H/D
scrambling reactions.2
of numerous “olefin hydride” complexes (particularly late metal,
first-row, electrophilic species) often proves to be the bridged,
agostic species, 20.3ꢀ6 There is growing evidence that the
scrambling between the agostic hydrogen and the terminal
While this is an appealingly simple picture, the actual situation
is often much more complex (Scheme 2). The most stable form
Received: May 26, 2011
Published: June 24, 2011
r
2011 American Chemical Society
12274
dx.doi.org/10.1021/ja204851x J. Am. Chem. Soc. 2011, 133, 12274–12284
|

Journal of the American Chemical Society
ARTICLE
hydrogens in such species occurs by an “in-place” rotation around
CR-Cβ and not by formation of a true unsaturated metal alkyl
complex, which exhibits no β-CHꢀmetal contact.7 Similarly,
hydrogen exchange in classical olefin hydride complexes likely
occurs via formation of an agostic intermediate followed by in-
place rotation. The unsaturated metal alkyl complexes in these
cases represent high-energy species and may not be true inter-
mediates. Experimental studies are often insufficient to distinguish
the classical description (Scheme 1) from the mechanism in
Scheme 2, and computational studies have been invaluable in
refining the subtle features of these processes.8
This situation is clearly illustrated by the structure and
dynamics of the cationic Co(III) ethylene hydrides and their
Rh(III) congeners shown in Scheme 3.9ꢀ11 These d6, 18-elec-
tron complexes possess geometries which are best described as
distorted octahedral. The stable form of the cobalt complexes, 4,
is the agostic structure as shown.8,9 The barriers to scrambling
of the agostic H with the terminal hydrogens lie in the range
9.2ꢀ13.4 kcal/mol, and the scrambling process was presumed to
proceed via the 16-electron ethyl complex, 40. Computational
studies indicate such scrambling occurs by in-place rotation and
that the 16-electron species, 40, is energetically inaccessible. The
R-agostic species 400 is calculated to be more stable than the
unsaturated ethyl complex 40.7
A similar situation holds for the Rh complexes, 5.10 The global
minimum is the olefin hydride species, 5, and measured scram-
bling barriers are considerably higher (12ꢀ15 kcal/mol) than
those for the cobalt analogues. Computations show that the
scrambling occurs via the formation of the agostic species, 50,
followed by in-place rotation. The Rh 16-electron ethyl complex
is calculated to be very high in energy and less stable than the
R-agostic isomer.7a
Similarly, experimental12 and theoretical13 studies of d8, 16-
electron, square-planar complexes of the type [(diimine)MꢀR]+
(M = Ni, Pd, Pt; R = CnH2n+1, n > 1) have shown that β-agostic
structures are adopted for M = Ni or Pd, while Pt exhibits a
classical olefin hydride structure. The barriers to scrambling of
the agostic H with the terminal hydrogens increase upon
descending the group; for example, a free energy of activation
of 19.2 kcal/mol (337 K) was determined for hydride migration
in the ethylene hydride complex [[2,6-Me2C6H3)NdC(An)ꢀ
C(An)dN(2,6-Me2C6H3)]Pt(H)ꢀ(ethylene)[BArf4].12a Simi-
larly, the barriers for methyl migration in related cationic M(II)
alkyl ethylene complexes fall in the range of 13ꢀ14 kcal/mol for
Ni(II), 17ꢀ19 kcal/mol for Pd(II), and 30 kcal/mol for Pt(II).12
We report here an investigation of the synthesis and migratory
insertion of pincer-supported square-pyramidal d6, 16-electron
Ir(III) and Rh(III) olefin hydride complexes. By virtue of the
rigid meridionally coordinated pincer ligand, we are able to
experimentally determine that the barrier for scrambling of the
metal hydride with the hydrogens of the ethylene ligands via an
Scheme 2. Hydrogen Scrambling Processes in Ethylene Hy-
dride Complex, 1, and the β-Ethyl Agostic Complex, 20
Scheme 3
12275
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
in-place rotation mechanism is lower than the barrier for forma-
tion of a square-planar 14-electron ethyl complex by at least
9 kcal/mol for Rh and 5.5 kcal/mol for Ir. Experimental studies
are augmented by DFT calculations. In addition, dynamics of the
analogous propylene complexes are reported which show that
scrambling occurs preferentially via a 1,2-insertion mechanism.
’ RESULTS AND DISCUSSION: SYNTHESIS AND DY-
NAMICS OF OLEFIN HYDRIDE COMPLEXES
Synthesis and Characterization of [(POCOP)M(H)(L)]-
[BArf4] (M = Ir, L = C2H4 (6a-Ir); M = Rh, L = C2H4 (6a-Rh);
M = Ir, L = C3H6 (6b-Ir); and M = Rh, L = C3H6 (6b-Rh)). Olefin
hydride complexes can be cleanly generated by purging a mixture
of (POCOP)M(H)(Cl)14 and NaBArf4 in methylene chloride
solvent with either ethylene or propylene gas (Scheme 4). Reac-
tions are complete within 15 min, as indicated by a color change
from light red to pale orange. Removal of sodium chloride by
syringe filtration under argon and evaporation of the solvent
yields pure products as pale orange powders. However, the
propylene complex 6b-Rh is unstable at room temperature in
methylene chloride in the absence of an excess of propylene.14b
Figure 1. ORTEP diagram of [(POCOP)Ir(H)(C3H6)][BArf] (6b-Ir).
The hydrogen atoms and counterion have been excluded for clarity.
Thermal ellipsoids shown at 40% probability. Key bond distances and
bond angles: Ir(1)ꢀC(10) 2.046 Å, Ir(1)ꢀP(1) 2.338 Å, Ir(1)ꢀP(2)
2.338 Å, Ir(1)ꢀC(1) 2.255 Å, Ir(1)ꢀC(2) 2.276 Å, C(1)ꢀC(2) 1.141 Å,
P(1)ꢀIr(1)ꢀC(10) 78.57°, P(2)ꢀIr(1)ꢀC(10) 78.72°, C(2)ꢀC(1)ꢀ
Ir(1)ꢀP(2) ꢀ39.78°. Errors in metrical parameters are omitted due to
disorder in the structure.
Scheme 4. Preparation of Complexes 6a-Ir, 6a-Rh, 6b-Ir, and
6b-Rh
are some notable differences. In 6b-Ir, the iridium hydride reso-
nance appears as a broadened triplet at ꢀ42.90 ppm, analogous to
ethylene complex 6a-Ir. The aryl backbone of the ligand in 6a-Ir
has two resonances associated with it, a triplet for H4 (para to the
iridium-bound carbon) and a doublet for the two equivalent
hydrogens (H3 and H5) adjacent to H4. In the propylene complex
6b-Ir, H3 and H5 are now inequivalent and appear as two doublets
in the 1H NMR spectrum at 6.90 and 6.86 ppm. There are also now
four resonances for the tert-butyl groups, indicating that none of the
tert-butyl groups are symmetry equivalent. Consistent with these
observations, the 31P NMR spectrum exhibits two doublets (182.7
ppm, 175.0 ppm, JPꢀP = 260 Hz). It is likely that the propylene
ligand, similar to the ethylene ligand, is also rotating rapidly at room
temperature (see below); however, since none of the conforma-
tions possess a mirror plane bisecting the ligand backbone, rapid
rotation does not result in averaging the 31P signals, H3 with H5, or
any of the tert-butyl resonances.
At room temperature the bound ethylene of 6a-Ir is rapidly
rotating on an NMR time scale, as is evident by a single ethylene
resonance in the 1H NMR spectrum at 4.15 ppm in CD2Cl2. The
upfield shift of the iridium hydride at ꢀ42.8 ppm in 6a-Ir is
consistent with a five-coordinate complex in which the hydride is
situated in the apical position trans to an empty coordination site.
1
In marked contrast, the H NMR spectrum of 6a-Rh under
ambient conditions displays neither ethylene nor hydride signals,
which suggests that exchange between these sites via migratory
insertion is occurring on a time scale that broadens these signals
into the baseline. Indeed, upon cooling a CD2Cl2 solution of 6a-
Rh to temperatures below ca. 283 K in the NMR probe, a single
broad resonance is observed at 4.59 ppm, corresponding
to the bound ethylene, with concomitant appearance of the
Rhꢀhydride resonance at ꢀ28.61 ppm as a doublet of triplets
(1JRhꢀH = 51.0 Hz, 2JPꢀH = 9.0 Hz).
The molecular structure of [(POCOP)Ir(H)(C3H6)][BArf4]
(6b-Ir) was established via single-crystal X-ray diffraction. A
crystal suitable for X-ray analysis was grown by slow diffusion of
pentane into a methylene chloride solution of 6b-Ir at room
temperature under an argon atmosphere. An ORTEP diagram of
6b-Ir is shown in Figure 1. The propylene ligand is rotated from
alignment of the CꢀC double bond with the plane of the ligand
backbone. This rotation likely occurs to decrease the steric
interactions between the methyl group of the propylene ligand
and the tert-butyl substituents on phosphorus. A slight disorder-
ing in the propylene ligand prevents a meaningful discussion of
the bond lengths of the ligated olefin in this structure.
Kinetics of Migratory Insertion Processes. The rate of site
exchange between the hydride and ethylene hydrogens of 6a-Ir
was determined using line-broadening techniques. As the tem-
perature is raised to 320 K (C6D5Cl solution), both the ethylene
and hydride resonances broaden (Figure 2). To verify that the
observed broadening was due to insertion, a spin-saturation
transfer experiment was performed. At 276 K, when the hydride
signal was irradiated, the ethylene signal decreased 49%, clearly
Both ethylene complexes have a geometry that requires the
tert-butyl groups syn to the hydride to be distinct from those anti
to the hydride. Indeed, two tert-butyl signals (seen as virtual
triplets due to strong phosphorusꢀphosphorus coupling) are
1
observed in the H NMR spectra for both 6a-Ir and 6a-Rh,
suggesting the existence of a vertical mirror plane bisecting the
ligand backbone. This analysis is supported by a single 31P
resonance at 182.0 ppm for the two equivalent phosphorus
atoms in 6a-Ir and a doublet centered at 205.3 ppm (1JRhꢀP
105 Hz) for 6a-Rh.
=
The propylene complexes 6b-Ir and 6b-Rh display some
characteristics similar to the ethylene complexes; however, there
12276
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
analysis of the propylene complexes, 6b-Ir and 6b-Rh, does
yield information concerning the rotational barrier in these more
hindered systems. The variable-temperature NMR spectra in
CD2Cl2 of 6b-Ir are shown in Figure 3. The propylene vinylic
and methyl signals as well as the IrꢀH resonance all broaden
below ca. 223 K, and at 175 K there is clear evidence for a set of
signals for both a minor isomer and a major isomer. Signals
marked 1aꢀ4a correspond to the major isomer, while signals
marked 1bꢀ4b correspond to the minor isomer, with the isomer
ratio equal to ca. 0.066 at 175 K. Both hydride resonances
at ꢀ42.56 (minor) and ꢀ43.20 (major) ppm indicate that both
isomers are five-coordinate species with a vacant coordination
site trans to the hydride ligand, ruling out a minor species
involving solvent coordination. The observation of two distinct
hydride resonances in a ratio of 0.056 at ꢀ26.72 (minor) and
1
ꢀ28.38 (major) ppm in the H NMR spectrum of 6b-Rh at
193 K suggests a similar situation for the rhodium complex.
Structures of the two possible rotamers of 6b-Ir are shown in
Scheme 5; the syn rotamer is the isomer that was crystallographically
characterized. Using line-broadening techniques together with the
experimentally determined Keq value, the rate constant for conversion
of the minor rotamer to the major rotamer, kmin-to-maj, wasdetermined
to be 16 sꢀ1 (205 K, ΔGq = 9.1 kcal/mol), with kmaj-to-min = 1.2 sꢀ1
(ΔGq = 10.2 kcal/mol).
For the 6b-Ir, insertion of propylene into the MꢀH bond
could lead to two possible agostic intermediates, a linear species
(from 1,2-insertion) and a branched species (from 2,1 insertion),
as shown in Scheme 6. Spin-saturation-transfer NMR experi-
ments (Figure 4, 296 K) were used to determine the favored
mode of insertion. Spectrum E shows the 1-H vinyl region as well
as the hydride region of the spectrum, with peak assignments
given at the top of the figure. Site exchange between a pair of
resonances is indicated by a decrease in intensity of the compa-
nion signal upon saturation of an individual resonance. Spectrum
A of Figure 4 shows that when the resonance for Hc is irradiated,
only the signal for Hb decreases in intensity. Complementary
results are seen in spectrum B, showing that irradiation of Hb
only reduces the intensity of the Hc signal. Spectrum C shows
that irradiation of Ha only decreases the hydride signal intensity,
Hd, and conversely in spectrum D, irradiation of Hd only
decreases Ha. The pairwise exchange of Hb/Hc and Ha/Hd
supports the proposal that the insertion of propylene into the
IrꢀH bond preferentially forms the linear (1,2-insertion)
product.
As mentioned above, the propylene complex has both “top/
bottom” and “side-to-side” inequivalence. Insertion of propylene
into the IrꢀH bond results in a species with an average mirror
plane bisecting and perpendicular to the ligand plane, as ob-
served in the variable-temperature 1H NMR spectra. The H3 and
H5 hydrogens of the ligand backbone merge into one doublet
from two, and the two pairs of tert-butyl resonances broaden and
merge into two triplets. Even at elevated temperatures, the “top”
and “bottom” faces do not equilibrate, consistent with observa-
tions for the ethylene complexes 6a-Ir and 6a-Rh. Spin-satura-
tion-transfer NMR techniques were used to measure the rate of
insertion of propylene into the MꢀH bond in 6b-Ir and 6b-Rh.
Using the spinꢀlattice relaxation time in combination with the
extent of signal reduction upon irradiation, standard equations
can be applied to determine exchange rates (see Supporting
Information for details). Using data from irradiation of Ha at
6.50 ppm, the rate of exchange via migratory insertion was
estimated to be 2.6 sꢀ1 at 296 K, corresponding to a free energy
Figure 2. High-temperature 1H NMR stacked plots of the ethylene
(a) andhydride(b) resonanceof[(POCOP)Ir(H)(C2H4)][BArf4] (6a-Ir).
showing that the observed line broadening of the hydride and
ethylene signals is due to exchange through a migratory insertion
process.
Since the population ratios are 4:1, the hydride resonance
broadens at 4 times the rate of the ethylene signal. Applying the
slow exchange approximation (k = πΔW) to line widths at half-
heights for these signals yields a rate for exchange of IrꢀH into
the ethylene ligand of 58 sꢀ1 at 310 K and 148 sꢀ1 at 320 K,
corresponding to ΔGq = 15.6 kcal/mol. Similar values are
obtained from analysis of the broadening of the ethylene signal
when a statistical correction is applied.15
Notably, the two resonances for the tert-butyl signals do not
average in a similar fashion. The tert-butyl resonances remain
sharp up to temperatures as high as 383 K (line width change less
than 3 Hz, conservatively), indicating a rate of averaging of less
than 10 sꢀ1 and a barrier to averaging the “top” and “bottom”
faces of the complex of greater than ca. 21 kcal/mol. In other
words, there is a high barrier for formation of a symmetrical
14-electron ethyl complex, which can collapse to either face of
the pincer ligand and re-form the stable ethylene hydride. This is
addressed in more detail below when describing the DFT results.
The behavior of the rhodium ethylene hydride complex 6a-Rh
is similar to that of 6a-Ir except that the barrier to scrambling is
lower. Using the same analysis of line broadening of the hydride
resonance as for 6a-Ir, the rate constant for exchange of the
RhꢀH with the ethylene hydrogens is 9.0 sꢀ1 at 229 K and
276 sꢀ1 at 253 K, corresponding to ΔGq = 12.0 kcal/mol. As for
6a-Ir, the tert-butyl resonances remain sharp up to 373 K,
indicating a barrier greater than 20.5 kcal/mol for face-to-face
switching of the ethyl group.15
Low-temperature NMR analysis of the iridium and rhodium
ethylene complexes, 6a-Ir and 6a-Rh, provided no information
concerning the barrier to ethylene rotation. However, NMR
12277
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
Figure 3. Variable-temperature stacked 1H NMR spectra of [(POCOP)Ir(H)(C3H6)][BArf4] (6b-Ir).
Scheme 5. Two Possible Rotamers of
[(POCOP)Ir(H)(C3H6)][BArf4] (6b-Ir)
address is the large barrier for exchanging the top and bottom
faces of the complex via the alkyl agostic intermediate.
A description of the computational methodology is given in
the Experimental Section. The optimized singlet ground- and
transition-state structures for the iridium ethylene and propylene
analogues are shown in parts a and b, respectively, of Figure 5,
and selected bond distances and angles for both the rhodium and
iridium analogues are reported in Table 1. The lowest energy
triplets lie >28 kcal/mol higher in energy, suggesting that the
hydride migration reactions take place on the singlet surface.
The structures of the [(POCOP)M(H)(CH2C(H)(R)]+
complexes are based on a square-pyramidal geometry in which
the hydride ligand sits in the apical position and the alkene ligand
is positioned trans to the aryl carbon of the pincer ligand (Cp)
(Figure 5). The alkene is rotated by 42° (Ir) and 29° (Rh) with
respect to the plane of the pincer ligand. In computations on a
trimmed complex in which the tert-butyl groups were replaced by
Me groups, the alkene carbon atoms lie in the plane of the pincer
ligand for the lowest energy conformer, indicating that steric
requirements of the tert-butyl groups dictate the observed con-
formation of the alkene ligand. In the calculated structures of the
[M(POCOP)(CH2CH2R)]+ β-agostic complexes, 7a-M (M = Rh
or Ir, olefin = C2H4) and 7b-M (M = Rh or Ir, olefin = C2H3Me)
barrier of 16.8 kcal/mol. This value is similar to the value of
15.6 kcal/mol determined for migratory insertion in 6a-Ir.
Similar analysis of the dynamic behavior of the rhodium ana-
logue, 6b-Rh, via irradiation of the Hc signal at 3.64 ppm yields a
rate of exchange estimated to be 4.9 sꢀ1 at 247 K, corresponding
to a free energy barrier of 13.7 kcal/mol.
Computational Studies. To gain additional insight into the
experimentally observed rate of exchange between the hydride
and alkene ligands in the [(POCOP)M(H)(CH2C(H)(R)]+
pincer complexes, a computational study was undertaken. In
addition to obtaining structural and energetic information on the
unobserved alkyl agostic species relative to the alkene hydride
ground state, an additional feature that the computational studies
12278
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
Scheme 6. The Two Possible Modes of Propylene Insertion into the MꢀH Bond in 6b-Ir and 6b-Rh
Figure 4. Spin-saturation-transfer experiment for 6b-Ir.
(Figure 5), the coordinated CꢀH bond is positioned trans to the
pincer carbon (Cp). The appended alkyl carbon atom is centered in
the cleft between the two tert-butyl groups and distorted away from
the apical position in an ideal square pyramid (CpꢀMꢀCR =
123ꢀ129°).
ethylene (8 kcal/mol (Ir) and 10 kcal/mol (Rh)) and propylene
(10 kcal/mol (Ir) and 12 kcal/mol (Rh)) systems. The hydrogen
atom in the transition state for hydride migration approaches the
fourth coordination site trans to the pincer arene ligand (see
Figure 5), leaving the axial sites effectively vacant like in a square-
planar geometry. The higher calculated hydride migration barrier
for Rh may be related to the higher barrier for passing through a
square-planar geometry calculated for Rh in comparison to Ir in
model [M(POCOP-Me)H]+ complexes (see discussion below).
The experimentally observed NMR exchange between the
hydride resonanceand the olefin CꢀH resonancerequires hydride
migration together with interchange of the coordinated CꢀH
bonds via an in-place rotation.6 Based on the calculated energy
profiles (Figure 6), the experimental barrier for the Rh systems is a
direct measure of the barrier for hydride migration, while for Ir,
the experimental barrier is the sum of the ground-state energy
The energy profiles for the hydride migration and in-place
rotation reactions are shown in Figure 6, and the calculated
energies are summarized in Table 2. In accord with experiment,
the olefin hydride complexes, 6, are calculated to be the ground
states for both the Rh and Ir systems. The agostic species, 7, lie
6 kcal/mol (Ir) and 2 kcal/mol (Rh) higher in energy for the
ethylene complexes and 7 kcal/mol (Ir) and 5 kcal/mol (Rh)
higher in energy for the propylene complexes. Despite the larger
energy differences between the olefin hydride and alkyl agostic
species for Ir, the calculated barriers for hydride migration to
form the agostic species are higher for Rh than Ir for both the
12279
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
Figure 5. Optimized ground- and transition-state structures for (a) the [(POCOP)Ir(ethylene)(H)]+ system and (b) the [(POCOP)Ir(propylene)-
(H)]+ system.
Table 1. Bond Distances (Å) and Angles (°) for Calculated Structures
compound
MꢀP1
MꢀP2
MꢀCp
MꢀCR
MꢀCβ
MꢀH1
CRꢀCβ
CβꢀH1a
CpꢀMꢀCR (H)b
CpꢀMꢀCβ
MꢀCRꢀCβ
6a-Ir
2.35
2.36
2.35
2.37
2.36
2.36
2.35
2.36
2.34
2.37
2.36
2.38
2.35
2.38
2.36
2.37
2.35
2.36
2.35
2.37
2.35
2.36
2.36
2.35
2.34
2.37
2.36
2.37
2.34
2.38
2.37
2.36
2.04
1.98
2.00
1.96
2.03
1.98
2.01
1.96
2.02
1.97
2.00
1.95
2.01
1.98
2.01
1.95
2.23
2.04
2.15
2.02
2.24
2.06
2.15
2.03
2.27
2.01
2.17
2.00
2.28
2.02
2.16
2.01
2.25
2.37
2.28
2.64
2.32
2.39
2.32
2.68
2.28
2.37
2.27
2.60
2.38
2.38
2.32
2.64
1.52
1.88
1.60
2.81
1.52
1.85
1.60
2.80
1.49
1.93
1.55
2.78
1.49
1.92
1.55
2.75
1.39
1.49
1.42
1.53
1.39
1.49
1.42
1.53
1.38
1.50
1.40
1.52
1.37
1.50
1.40
1.52
89
129
145
124
87
7a-Ir
1.18
1.56
1.10
168
178
159
83
77
95
TS6a-Irf7a-Ir
0
TS7a-Irf7a -Ir
6b-Ir
7b-Ir
1.19
1.60
1.10
134
144
125
85
171
175
160
83
79
97
TS6b-Irf7b-Ir
0
TS7b-Irf7b -Ir
6a-Rh
7b-Rh
1.15
1.59
1.10
123
151
117
83
162
172
153
84
75
95
TS6a-Rhf7a-Rh
0
TS7a-Rhf7a -Rh
6b-Rh
7b-Rh
1.16
1.64
1.10
126
148
120
164
173
155
84
78
96
TS6b-Rhf7b-Rh
0
TS7b-Rhf7b -Rh
a A typical calculated CꢀH bonding distance is 1.09 Å. b CpꢀMꢀH angle for the alkene complexes.
difference between the olefin hydride and agostic species
(ΔE6-Irf7-Ir) and the barrier for in-place rotation. The calculated
barriers are in reasonable agreement with the experimental
barriers (see Table 2), and the trends (as functions of metal
and olefin) in the experimental data are reproduced.
lies ca. 10 kcal/mol to higher energy than the minimum with
CpꢀIrꢀCR = 128°. This net 10 kcal/mol energy difference in-
cludes a 3 kcal/mol stabilization arising from adoption of a
staggered configuration by the ethyl ligand. A similar analysis for
the rhodium system yields a 5 kcal/mol energy difference for the
structure with a CpꢀRhꢀCR angle of 90° and a RhꢀCRꢀCβ
angle of 104°.
The transition state for in-place rotation for agostic 7a is
0
pictured in Figure 5a (TS7a-Irf7a -Ir). The interaction between
the coordinated methyl group and Ir is considerably diminished
in the transition state, as signified by an increase in the IrꢀCβ
distance from 2.37 to 2.64 Å and an opening up of the
IrꢀCRꢀCβ angle from 83° to 95°. To obtain an estimate of
the strength of the agostic interaction, a potential energy scan was
carried out for a trimmed version of 7a-Ir, in which the t-Bu
groups were replaced by Me groups. The CpꢀIrꢀCR angle was
stepped from 90° to the 128° value for the minimum energy
structure with a geometry optimization at each step. The
structure with a CpꢀIrꢀCR = 90° and IrꢀCRꢀCβ = 107°,
in which the ethyl ligand adopts a staggered configuration,
The DFT calculations also agree well with the experimental
results in the propylene system. NMR evidence was obtained for
two rotamers of the propylene ligands in 6b-Ir. The crystal-
lographically characterized rotamer corresponds to the lowest
energy rotamer determined computationally, which is pictured in
Figure 5. The energy difference between the two rotamers of ∼1
kcal/mol is in reasonable agreement with the experimental
rotamer ratio (15:1 at 175 K). The calculated barrier for
propylene rotation in 6b-Ir is 10 kcal/mol, which agrees well
with the experimental value of 10 kcal/mol obtained from line-
broadening measurements. Hydride migration to the major
12280
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
Figure 6. Energy profile for the [(POCOP)M(H)(olefin)]+ system.
Table 2. Calculated Ground-State and Transition-State Energies (kcal/mol)
M
olefin
ML(olefin)(H)+
ML(agostic)+
TS1a
TS2b
exchange barrierc
expt ΔGq
Ir
C2H4
0.0
0.0
0.0
0.0
6.2
2.1
7.9
5.0
8.3
9.8
6.5
5.1
6.9
4.5
12.7
9.8
15.6
12.0
16.8
13.7
Rh
Ir
C2H4
C2H3Me
C2H3Me
10.5
12.2
14.8
12.2
Rh
a Transition state for hydride migration. b Transition state for in-place rotation. c For Ir, exchange barrier = ΔE(ML(agostic)+ ꢀ ML(olefin)(H)+) +
TS2. For Rh, exchange barrier = TS1.
rotamer yields a primary alkyl agostic complex, while hydride
migration to the minor rotamer yield a secondary alkyl agostic
complex. The secondary alkyl agostic complex lies ∼5 kcal/mol
higher in energy than the primary agostic complex for both Rh
and Ir, and the barrier for hydride migration to the minor rotamer
(12 kcal/mol) is higher than the barrier for hydride migration to
the major rotamer (10 kcal/mol). These computational results
are consistent with the experimental result that only exchange
resulting from hydride migration to the major rotamer is
observed. Full details of calculations for the minor isomer are
given in the Supporting Information.
ligand about the MꢀC and CꢀC bonds to adopt the appropriate
conformation for forming a new agostic interaction.
To explore the energetics of exchanging the top and bottom
faces of the complex, calculations were carried out for a model
hydride analogue, [M(POCOP-Me)H]+ (8-M; M = Rh and Ir),
with a trimmed ligand in which the t-Bu substituents are replaced
by Me groups. The barrier for interchanging the hydride ligand
between the two axial sites via a planar transition state (eq 1)
on the singlet surface is 41 kcal/mol for 8-Ir and 53 kcal/mol for
8-Rh.
A surprising feature in the NMR studies of the pincer alkene
hydride complexes is that exchange of the top and bottom faces
of the pincer complexes is not experimentally observed, and a
lower limit to the barrier of ∼20 kcal/mol is placed on both of
these systems. The most reasonable pathway for interchanging
the top and bottom faces of the pincer complex would involve de-
coordination of the agostic CꢀH bond to form a four-coordinate
complex. The lowest energy structure for a four-coordinate d6
pincer complex is based on a cis-divacant octahedron (eq 1),16
with the alkyl group occupying one of the axial positions
perpendicular to the plane of the POCOP ligand. Interchange
of the top and bottom faces of these complexes would require the
alkyl ligand to pass through a planar transition state to the
opposite face of the complex, followed by rotation of the alkyl
The barriers for equilibrating the two faces on the triplet surface
are lower (7 kcal/mol for 8-Ir and 6 kcal/mol for 8-Rh), but
the triplets lie considerably higher in energy than the singlet
(28 kcal/mol for 8-Ir and 31 kcal/mol for 8-Rh). Given these
large calculated barriers, it is not surprising that no exchange is
observed between the top and bottom faces in the ethylene hy-
dride complexes. For the d6 configuration, conversion of the
12281
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
Cs cis-divacant octahedral structure to the C2v square-planar transition
state is associated with a large reduction in the HOMOꢀLUMO
gap and corresponding destabilization.
and Pd(II) complexes are more reactive than Pt(II) analogues. The
present study suggests that this trend is reversed in d6 five-coordinate,
square-pyramidal Rh and Ir olefin hydride complexes. This observa-
tion could have significant implications for catalytic reactions involv-
ing migratory insertions of d6 square-pyramidal complexes. Other
complexes are under investigation to probe the generality of these
observations.
’ SUMMARY
Square-pyramidal cationic d6 olefin hydride complexes of the
type [(POCOP)M(H)(olefin)][BArf4] (6a-M, M = Ir or Rh,
olefin = C2H4; 6b-M, M = Ir or Rh, olefin = C3H6) have been
synthesized and characterized by dynamic NMR spectroscopy.
Rotation of the ethylene ligand is too fast to be measured by 1H
’ EXPERIMENTAL SECTION
General Considerations. All manipulations were carried out
using standard Schlenk, high-vacuum, and glovebox techniques. Argon
was purified by passage through columns of BASF R3-11 (chemalog)
and 4 Å molecular sieves. Pentane and methylene chloride were passed
through columns of activated alumina and deoxygenated by purging with
N2. Benzene was dried over 4 Å molecular sieves and degassed to remove
both oxygen and nitrogen. NMR spectra were recorded on Bruker DRX
400, AMX 300, and 500 MHz instruments and are referenced to residual
protio solvent peaks. 31P chemical shifts are referenced to an external
1
NMR spectroscopy. The H spectra at ꢀ100 °C of propylene
complexes 6b-Ir and 6b-Rh show two rotamers in a ca. 15:1 ratio.
The barriers to interconversion of rotamers were determined
using NMR techniques and are around 17 kcal/mol for 6b-Ir and
14 kcal/mol for 6b-Rh.
The rates of exchange between the hydride signal and the
olefinic H signal were calculated for the ethylene and propylene
complexes using line-broadening or spin-saturation transfer NMR
techniques. The measured exchange barriers for the ethylene com-
plexes are lower than for the propylene complex for both Rh and Ir
(ΔGq = 15.6 (6a-Ir), 12.0 (6a-Rh), 16.8 (6b-Ir), and 13.7 (6b-Rh)
kcal/mol), and the barriers for the iridium complexes were higher
than for the rhodium complexes for both pairs of complexes. No
interchange between the top and bottom faces of the olefin hydride
complexes is observed by NMR spectroscopy, setting a lower limit to
the barrier of ∼20 kcal/mol.
31
H3PO4 standard. Since there is a strong 31Pꢀ P coupling in the pincer
1
complexes, many of the H and 13C signals exhibit virtual coupling and
appear as triplets. These are specified as “vt”, with the apparent coupling
simply noted as J. Elemental analyses were carried out by Atlantic Microlab,
Inc. (Norcross, GA) or Robertson Microlit Laboratories Inc. (Madison,
NJ). Allreagents, unlessotherwisenoted, werepurchasedfromcommercial
sources and used without further purification. Propylene was used as
received from National Specialty Gases (Durham, NC). Ethylene was
purchased from Matheson. The syntheses of the phosphinite complexes
(POCOP)Ir(H)(Cl) and (POCOP)Rh(H)(Cl) have been previously
described in the literature.14 The abbreviation BArf4 is used to represent
Insight into the relationship between the NMR-measured
exchange rates and the individual barriers to hydride migration
(TS1) and in-place rotation (TS2) was obtained from a DFT
study. The [M(POCOP)(olefin)H]+ complexes are calculated
to be the ground state for both Rh and Ir, in accord with
experimental results, and the energy differences between the
olefin hydride ground states and the alkyl agostic species are
∼2ꢀ4 kcal/mol larger for Ir than for Rh. For Rh, TS1 is the high
point on the energy surface, while for Ir, TS2 is the high point;
therefore, only for Rh is the experimental exchange barrier a
measure of the hydride migration barrier. The square-pyramidal
coordination geometry of the olefin hydrides, 6, and the distorted
square-pyramidal coordination for the alkyl agostic complexes, 7,
provide for some unusual trends in the barriers for these
complexes. For a four-coordinate alkyl complex, which would
be the intermediate preceding a face-to-face exchange of the alkyl
ligand, a cis-divacant octahedron is the lowest energy structure.
Computations on a model hydride complex indicate that the
barriers for passing through a square-planar transition state to the
opposite face on the singlet surface are sizable, and larger for Rh
(ΔGq = 51 kcal/mol) than for Ir (ΔGq = 41 kcal/mol). Despite
the larger ground-state energy difference calculated for Ir, the
computed barriers to hydride migration are ∼2 kcal/mol higher
for Rh than for Ir. This unusual feature may be related to the fact
that the transition state for hydride migration samples this fourth
site trans to the pincer arene ligand in a square-planar geometry,
and therefore reflects the higher barrier for passing through this
site calculated for Rh.
the counterion tetrakis(3,5-trifluoromethylphenyl)borate. The 1H and 13
C
spectral data for BArf4 in CD2Cl2 are the same for all complexes listed [1H
NMR (CD2Cl2) δ 7.72 (s, 8H, Ho), 7.56 (s, 4H, Hp); 13C{1H} NMR
(CD2Cl2) δ 162.2 (q, JCꢀB = 37.4 Hz, Cispo), 135.2 (Co), 129.3 (q, JCꢀF
=
31.3 Hz, Cm), 125.0 (q, JCꢀF = 272.5 Hz, CF3), 117.9 (Cp)] and are
therefore not reported in the characterizations below.
General Procedure for the Synthesis of [(POCOP)M(H)-
f
(olefin)][BAr4 ] (6a/b-Ir and 6a/b-Rh). In a Schlenk flask under
f
argon, 1 equiv of (POCOP)M(H)(Cl), 1.1 equiv of NaBAr4 , and excess
olefin were stirred in methylene chloride for 15 min. The reaction
mixture was then filtered under an inert atmosphere in a glovebox. The
product was isolated after removal of the methylene chloride solvent
under reduced pressure.
[(POCOP)Ir(H)(C2H4)][BArf4] (6a-Ir). The general procedure was
employed using (POCOP)Ir(H)(Cl) (0.08 mmol, 50 mg) and NaBArf4
(0.089 mmol, 77.9 mg) in CH2Cl2 (10 mL) and purging ethylene
through the reaction mixture. The color changed from pale red to orange
within 5 min, indicating the completion of the reaction. Following
filtration and solvent removal, the product was isolated as a pale orange
1
solid (104 mg, 87.7% yield). H NMR (500 MHz, CD2Cl2): δ 7.22
(t, 3JHꢀH = 8.0 Hz, 1H, 1-H), 6.92 (d, 3JHꢀH = 8.0 Hz, 2H, 2-H), 4.15
(s, 4H, C2H4), 1.34 (vt, J = 7.5 Hz, 18H, P(tBu)2), 1.27 (vt, J = 8.0 Hz,
18H, P(tBu)2), ꢀ42.83 (b, 1H, IrH). 31P{1H} NMR (162 MHz,
CD2Cl2): δ 182.0. 13C{1H} NMR (125.8 MHz, CD2Cl2): δ 167.0
(Cq, vt, J = 5.6 Hz, 2C, 3-C), 133.8 (Cq, m, 1C, 4-C), 132.3 (CH, s, 1C,
1-C), 107.1 (CH, t, JPꢀC = 6.0 Hz, 2C, 2-C), 55.1 (CH2, s, 2C, C2H4),
45.7 (Cq, vt, J = 12.1 Hz, C(CH3)3), 42.8 (Cq, vt, J = 13.5 Hz, C(CH3)3),
28.9 (CH3, vt, J = 2.1 Hz, C(CH3)3), 28.1 (CH3, vt, J = 1.9 Hz,
C(CH3)3). Elemental analysis calculated for C56H56BF24IrO2P2
(1481.98): C, 45.39; H, 3.81. Found: C, 45.18; H, 3.69.
As noted earlier, the barriers to migratory insertion reactions
of olefin hydride and olefin alkyl complexes in octahedral d6
systems and in square-planar d8 systems increase significantly
moving from first-row to third-row homologues. This trend has
important implications in catalysis; for example, catonic Ni(II)
alkyl complexes are far more reactive in olefin oligomerization
and polymerization reactions than analogous Pd(II) complexes,
[(POCOP)Rh(H)(C2H4)][BArf4] (6a-Rh). The general procedure was
employed using (POCOP)Rh(H)(Cl) (0.08 mmol, 43 mg) and Na-
BArf4 (0.089 mmol, 77.9 mg) in CH2Cl2 (10 mL) and purging ethylene
through the reaction mixture. The color changed from yellow to pale
12282
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
brown within 15 min, indicating the completion of the reaction.
Following filtration and solvent removal, the product was isolated as a
brown solid (105 mg, 94.2% yield). 1H NMR (500 MHz, CD2Cl2): δ
7.19 (t, 3JHꢀH = 8 Hz, 1H), 6.85 (d, 3JHꢀH = 7.5 Hz, 2H), 4.59 (s, 4H,
C2H4), 1.30 (vt, 18H, P(tBu)2), 1.23 (vt, 18H, P(tBu)2), ꢀ28.61
relativistic effective core potential combination for the transition
metals19,20 with a single f-type polarization function (exponent =
1.062 (Rh), 0.685 (Ir)) and the functional PBE0, the hybrid variant of
PBE that contains 25% HartreeꢀFock exchange21 for geometry opti-
mizations. The PBE0 functional was found to yield results in better
agreement with experimental data than the B3LYP22 functional in
an Ir pincer system23 and has been endorsed as one of the best-
performing functionals for late transition metal systems.24 A similar basis
set combined with the PBE0 functional was used to calculate weak
1
2
(dt, JRhꢀH = 50.5 Hz, JHꢀP = 9 Hz, 1H, Rh-H). 31P{1H} NMR
(162 MHz, CD2Cl2): δ 205.26 (d, JPꢀRh = 105.3 Hz). 13C{1H} NMR
(125.8 MHz, CD2Cl2): δ 166.26 (Ar), 130.81 (Ar), 107.01 (Ar), 72.40
(C2H4), 43.56 (C(CH3)), 41.52 (C(CH3)), 28.08 (C(CH3)), 27.44
(C(CH3)); one ArꢀC resonance not detected. Elemental analysis
calculated for C57H58BF24RhO2P2 (1406.69): C, 48.67; H, 4.16. Found:
C, 48.56; H, 3.92.
Rh HꢀC interactions in another system.25 A comparative set of
3 3 3
calculations was carried out with the LAN08(f) basis setꢀpseudo
relativistic effective core potential for the metal,26 the built-in 6-31G**
basis set for all other atoms, and the B3LYP functional. All of the major
conclusions of this work hold with either method choice. For each
metalꢀligand combination, geometries were optimized in the gas phase
for the olefin hydride [M(L)(olefin)(H)]+ and the agostic complexes
[M(L)(agostic)]+. Frequency calculations were carried out on all mini-
mum structures, and the resulting frequencies all had positive values. The
nonscaled vibrational frequencies formed the basis for the calculation of
vibrational zero-point corrections and the standard thermodynamic correc-
tions for the conversion of electronic energies to enthalpies and free
energies at 298.15 K and 1 atm. The entropy corrections for the highly
crowded POCOPꢀpropylene complexes yielded inconsistent results as a
function of metal; therefore, the ZPE-corrected electronic energies were
used for comparison to experimental data in all cases.
[(POCOP)Ir(H)(C3H6)][BArf4] (6b-Ir). The general procedure was
employed using (POCOP)Ir(H)(Cl) (0.08 mmol, 50 mg) and NaBArf4
(0.087 mmol, 77.8 mg) in CH2Cl2 (10 mL) and purging propylene
through the reaction mixture. The color changed from pale red to orange
within 5 min, indicating the completion of the reaction. Following
filtration and solvent removal, the product was isolated as a pale orange
1
solid (103 mg, 86% yield). H NMR (500 MHz, CD2Cl2): δ 7.21
(t, 3JHꢀH = 8.1 Hz, 1H, 1-H), 6.90 (d, 3JHꢀH = 7.9 Hz, 1H, 2-H), 6.86
3
3
(d, JHꢀH = 8.2 Hz, 1H, 6-H), 6.43 (m, 1H, 8-H), 5.5 (dd, JHꢀH
=
3
12.0 Hz, JPꢀH = 8.5 Hz, 1H, 7-Htrans), 3.3 (d, JHꢀH = 8.0 Hz, 1H,
7-Hcis), 1.9 (d, 3JHꢀH = 5.5 Hz, 3H, 9-H), 1.46 (d, JPꢀH = 15.3 Hz, 9H,
P(tBu)2), 1.35 (d, JPꢀH = 14.6 Hz, 9H, P(tBu)2), 1.34 (d, JPꢀH
=
15.3 Hz, 9H, P(tBu)2), 1.17 (d, JPꢀH = 14.6 Hz, 9H, P(tBu)2), ꢀ42.9
(t, JPꢀH = 11.6 Hz, 1H, IrH). 31P{1H} NMR (162 MHz, CD2Cl2): δ
182.7 (dd, JPꢀP = 260.1 Hz), 175.0 (dd, JPꢀP = 260.1 Hz). 13C{1H}
NMR (125.8 MHz, CD2Cl2): δ 167.2 (Cq, m, 2C, 3-C and 5-C), 132.4
(CH, s, 1C, 1-C), 131.3 (Cq, m, 1C, 4-C), 106.9 (CH, m, 2C, 2-C and
For each Mꢀolefin pair, the transition state for hydride migration and
the transition state for in-place rotation were optimized in the gas phase
using the synchronous transit-guided quasi-Newton (STQN) method
implemented in Gaussian. Frequency calculations yielded one imaginary
frequency for all transition states, and IRC calculations were carried out to
confirm that the transition state identified connected the correct minima.
Included as Supporting Information are a table of calculated electronic
energies, enthalpies, and free energies in the gas phase for all ground states
and transition states calculated, as well as tables of Cartesian coordinates (Å)
for the optimized structures and transition states in the gas phase.
6-C), 77.8 (CH, s, 1C, 7-C), 66.2 (CH2, s, 1C, 8-C), 45.6 (Cq, d, JPꢀH
=
22.3 Hz, 2C, C(CH3)3), 42.5 (Cq, d, JPꢀH = 20.5 Hz, 1C, C(CH3)3),
41.9 (Cq, d, JPꢀH = 20.9 Hz, 1C, C(CH3)3), 29.4 (CH3, d, JPꢀH = 3.5 Hz,
C(CH3)3), 29.2 (CH3, d, JPꢀH = 3.5 Hz, C(CH3)3), 28.4 (CH3, d,
JPꢀH = 2.9 Hz, C(CH3)3), 28.11 (CH3, d, JPꢀH = 3.5 Hz, C(CH3)3).
Elemental analysis calculated for C57H58BF24IrO2P2 (1496.01): C,
45.76; H, 3.91. Found: C, 45.64; H, 3.84.
[(POCOP)Rh(H)(C3H6)][BArf4] (6b-Rh). The general procedure was
employed using (POCOP)Rh(H)(Cl) (0.08 mmol, 43 mg) and Na-
BArf4 (0.087 mmol, 77.8 mg) in CH2Cl2 (10 mL) and purging
propylene through the reaction mixture. The color changed from pale
red to deep red-brown within 5 min. Filtration and solvent removal
resulted in regeneration of the starting material, presumably the chloride
being scavenged from the solvent. Attempts at isolating 6b-Rh free of
starting material have so far been unsuccessful. Thus, the propylene
adduct was characterized in situ and in the presence of excess propylene.
1H NMR (500 MHz, CD2Cl2): δ 7.17 (t, 3JHꢀH = 8.0 Hz, 1H, Ar-H),
’ ASSOCIATED CONTENT
S
Supporting Information. Crystal structure information
b
file for 6b-Ir, complete calculations for dynamic processes, sample
1H NMR spectra, thorough discussion of variable-temperature NMR
behavior, complete ref 17, and full details of calculations including
Cartesian coordinates of all optimized geometries. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
3
7.05 (m, 1H, Hgem), 6.80 (app t, JHꢀH = 8.0 Hz, 1H, Ar-Hmaj), 6.60
(d, 3JHꢀH = 8.2 Hz, 1H, Ar-Hmin), 5.88 (m, 1H, Htrans), 3.50 (d, 3JHꢀH
=
=
Corresponding Author
mbrookhart@unc.edu; schauer@unc.edu
8.0 Hz, 1H, Hcis), 1.86 (d, 3JHꢀH = 5.5 Hz, 3H, Me), 1.48 (d, JPꢀH
15.3 Hz, 9H, P(tBu)2), 1.30 (d, JPꢀH = 14.6 Hz, 18H, P(tBu)2), 1.20
(d, JPꢀH = 15.3 Hz, 9H, P(tBu)2), ꢀ27.28 (dt, 1H, 1JRhꢀH = 50.0 Hz,
2JHꢀP = 8.5 Hz, Rh-Hmin), ꢀ28.38 (dt, 1H, 1JRhꢀH = 50.0 Hz, 2JHꢀP
=
’ ACKNOWLEDGMENT
8.5 Hz, Rh-Hmaj). 31P{1H} NMR (162 MHz, CD2Cl2): δ 203.6maj
(dd, JRhꢀP = 86.9 Hz, JPꢀP = 224.5 Hz), 198.8min (dd, JRhꢀP = 84.4 Hz,
JPꢀP = 224.6 Hz). 13C{1H} NMR (125.8 MHz, CD2Cl2): δ 166.4 (Ar),
132.5 (Ar), 131.2 (Ar), 117.1 (Ar), 107.9 (Ar), 97.5 (br m, C(H)(Me)),
We gratefully acknowledge the financial support of the NSF
(Grant Nos. CHE-0650456 as part of the Center for Enabling New
Technologies through Catalysis (CENTC) and CHE-1010170).
79.8 (br m, CH2), 43.0 (d, JPꢀH = 21.0 Hz, C(CH3)3), 42.4 (d, JPꢀH
21.5 Hz, C(CH3)3), 28.1 (d, JPꢀH = 3.5 Hz, C(CH3)3), 27.6 (d, JPꢀH
3.5 Hz, C(CH3)3), 22.8 (Meprop).
Computational Studies. All density functional theory (DFT)
calculations were performed by using the Gaussian 03 package.17 The
basis-set/functional selection was based on a prior study of methane
binding18 and consists of the built-in 6-31G** basis set for all non-
transition metal atoms, the StuttgartꢀDresden basis setꢀpseudo
=
=
’ REFERENCES
(1) See, for example: (a) Hartwig, J. Organotransition Metal Chemistry:
From Bonding to Catalysis; University Science Books: New York, 2009. (b)
Ojima, I.; Eguchi, M.; Tzamarioudaki, M. In Comprehensive Organometallic
Chemistry; Abel, E. W., Ed.; Pergamon: New York, 1995; p 39.
(2) See, for example: (a) Cross, R. J. In The Chemistry of the
MetalꢀCarbon Bond; Hartley, F. R., Patai, S., Eds.; Wiley: New York,
12283
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284

Journal of the American Chemical Society
ARTICLE
1985; Vol. 2, Chapter 8. (b) Bullock, R. M. In Transition Metal Hydrides;
Dedieu, A., Ed.; VCH: New York, 1992; p 263.
(21) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,
3865–3868.
(3) Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983,
250, 395.
(4) Brookhart, M.; Green, M. L. H.; Wong, L. L. Prog. Inorg. Chem.
1988, 36, 1.
(5) Brookhart, M.; Green, M. L. H.; Parkin, G. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 6908.
(6) Steinke, T.; Shaw, B. K.; Jong, H.; Patrick, B. O.; Fryzuk, M. D.;
Green, J. C. J. Am. Chem. Soc. 2009, 131, 10461–10466.
(7) (a) Green, M. L. H.; Wong, L. L. J. Chem. Soc., Chem. Commun.
1988, 677–679. (b) Derome, A. E.; Green, M. L. H.; Wong, L. L. New
J. Chem. 1989, 10, 747–753. (c) Bercaw, J. E.; Burger, B. J.; Green,
M. L. H.; Santarsiero, B. D.; Sella, A.; Trimmer, M.; Wong, L. L. J. Chem.
Soc., Chem. Commun. 1989, 734–736. (d) McNally, J. P.; Cooper, N. J.
Organometallics 1988, 7, 1704–1715. (e) Casey, C. P.; Yi, C. S.
Organometallics 1991, 10, 33–35. (f) Tempel, D. J.; Brookhart, M.
Organometallics 1998, 17, 2290–2296.
(22) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys. 1988,
37, 785.
(23) Ghosh, R.; Zhang, X.; Achord, P.; Emge, T. J.; Krogh-Jesperson,
K.; Goldman, A. S. J. Am. Chem. Soc. 2007, 127, 853–866.
(24) Quintal, M. M.; Karton, A.; Iron, M. A.; Boese, A. D.; Martin,
J. M. L. J. Phys. Chem. A 2006, 110, 709–716.
(25) Montag, M.; Efremenko, I.; Cohen, R.; Leitus, G.; Shimon,
L. J. W.; Diskin-Posner, Y.; Ben-David, Y.; Martin, J. M. L.; Milstein, D.
Chem.—Eur. J. 2008, 14, 8183–8194.
(26) Roy, L. E.; Hay, P. J.; Martin, R. L. J. Chem. Theory Comput.
2008, 4, 1029–1031.
(8) (a) Xu, R.; Klatt, G.; Wadepohl, H.; Koppel, H. Inorg. Chem.
2010, 49, 3289–3296. (b) Scherer, W.; Herz, V.; Bruck, A.; Hauf,
C.; Reiner, F.; Altmannshofer, S.; Leusser, D.; Stalke, D. Angew.
Chem., Int. Ed. 2011, 50, 1–6. (c) Mitoraj, M. P.; Michalak, A.;
Ziegler, T. Organometallics 2009, 28, 3727–3733 and references
therein. (d) Clot, E.; Eisenstein, O. Struct. Bonding (Berlin) 2004, 113,
1–36.
(9) Brookhart, M.; Lincoln, D. M.; Volpe, A. F., Jr.; Schmidt, G. F.
Organometallics 1989, 8, 1212.
(10) Brookhart, M.; Lincoln, D. M.; Bennett, M. A.; Pelling, S. J. Am.
Chem. Soc. 1990, 112, 2691.
(11) Brookhart, M.; Hauptman, E.; Lincoln, D. M. J. Am. Chem. Soc.
1992, 114, 10394.
(12) (a) Shiotsuki, M.; White, P. S.; Brookhart, M.; Templeton, J. L.
J. Am. Chem. Soc. 2007, 129, 4058–4067. (b) Tempel, D. J.; Johnson,
L. K.; Huff, R. L.; White, P. S.; Brookhart, M. J. Am. Chem. Soc. 2000,
122, 6686–6700. (c) Killian, C. M.; Johnson, L. K.; Brookhart, M.
Organometallics 1997, 16, 2005–2007. (d) Killian, C. M.; Johnson, L. K.;
Brookhart, M. J. Am. Chem. Soc. 1999, 121, 10634–10635. (e) Rix, F. C.;
Brookhart, M. J. Am. Chem. Soc. 1995, 117, 1137–1138. (f) Johnson,
L. K.; Killian, C. M.; Brookhart, M. J. Am. Chem. Soc. 1995,
117, 6414–6415. (g) Malinoski, J. M.; Brookhart, M. Organometallics
2003, 22, 5324–5335.
(13) (a) Stromberg, S.; Zetterberg, K.; Siegbahn, P. E. M. J. Chem.
Soc., Dalton Trans. 1997, 4147–4152. (b) Musaev, D. G.; Froese, R. D. J.;
Morokuma, K. Organometallics 1998, 17, 1850–1860. (c) Musaev, D. G.;
Morokuma, K. Top. Catal. 1999, 7, 107–123. (d) Michalak, A.; Ziegler,
T. Kinet. Catal. 2006, 47, 310–325. (e) Mitoraj, M.; Michalak, A. J. Mol.
Model. 2005, 11, 341–350. (f) Deubel, D. V.; Ziegler, T. Organometallics
2002, 21, 1603–1611. (g) Deubel, D. V.; Ziegler, T. Organometallics
2002, 21, 4432–4441. (h) Michalak, A.; Ziegler, T. Organometallics
2003, 22, 2660–2669.
(14) (a) M = Ir: Gottker-Schnetmann, I.; White, P. S.; Brookhart, M.
J. Am. Chem. Soc. 2004, 126, 1804–1811. (b) M = Rh: Salem, H.;
Shimon, L. J. W.; Leitus, G.; Weiner, L.; Milstein, D. Organometallics
2008, 27, 2293–2299.
(15) See Supporting Information.
(16) (a) Burdett, J. K. Inorg. Chem. 1975, 14, 375–382. (b) Cooper,
A. C.; Clot, E.; Huffman, J. C.; Streib, W. E.; Maseras, F.; Eisenstein, O.;
Caulton, K. G. J. Am. Chem. Soc. 1999, 121, 97–106. (c) Ujaque, G.;
Cooper, A. C.; Maseras, F.; Eisenstein, O.; Caulton, K. G. J. Am. Chem.
Soc. 1998, 120, 361–365.
(17) Frisch, M. J.; , et al. Gaussian 03, Revision E.01; Gaussian, Inc.:
Wallingford, CT, 2004.
(18) Bernskoetter, W. H.; Schauer, C. K.; Goldberg, K. I.; Brookhart,
M. Science 2009, 326, 553–556.
(19) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123–141.
(20) Martin, J. M. L.; Sundermann, A. J. Chem. Phys. 2001, 114,
3408–3420.
12284
dx.doi.org/10.1021/ja204851x |J. Am. Chem. Soc. 2011, 133, 12274–12284