Heterolytic benzene C-H activation by a cyclometalated iridium(III) dihydroxo pyridyl complex: synthesis, hydrogen-deuterium exchange, and density functional study

Article abstract of DOI:10.1021/om900039s

We report the synthesis of the pincer-cyclometalated (NNC ^Bu)Ir(III) dihydroxo pyridyl complex 6, which catalyzes hydrogen-deuterium (H/D) exchange between water and benzene in the presence of base (TOF = ~6 x 1O^-3 s^-1 at

Full text of DOI:10.1021/om900039s

Organometallics 2009, 28, 5293–5304 5293
DOI: 10.1021/om900039s
Heterolytic Benzene C-H Activation by a Cyclometalated Iridium(III)
Dihydroxo Pyridyl Complex: Synthesis, Hydrogen-Deuterium
Exchange, and Density Functional Study
Steven K. Meier,† Kenneth J. H. Young,† Daniel H. Ess,§,‡ William J. Tenn, III,†
Jonas Oxgaard,‡ William A. Goddard, III,*,‡ and Roy A. Periana*,†,§
†Department of Chemistry, Loker Hydrocarbon Institute, University of Southern California, Los Angeles,
California 90089, §Department of Chemistry, The Scripps Research Institute, Jupiter, Florida 33458, and
‡Materials and Process Simulation Center, Beckman Institute (139-74), Division of Chemistry and Chemical
Engineering, California Institute of Technology, Pasadena, California 91125.
Received January 17, 2009
We report the synthesis of the pincer-cyclometalated (NNCt-Bu)Ir(III) dihydroxo pyridyl complex
6, which catalyzes hydrogen-deuterium (H/D) exchange between water and benzene in the presence
of base (TOF = ∼6 ꢀ 10-3 s-1 at 190 °C). Experimental and density functional theory (B3LYP)
studies suggest that H/D exchange occurs through loss of pyridine followed by benzene coordination
and C-H bond activation by a heterolytic substitution mechanism to give a phenyl aquo complex,
which may dimerize. Exchange of H2O for D2O followed by the microscopic reverse of CH activation
leads to deuterium incorporation into benzene. Synthesis of the μ-hydroxo phenyl dinuclear complex
[(NNCt-Bu)Ir(Ph)(μ-OH)]2 (9) also catalyzes H/D exchange with a turnover frequency (TOF = ∼7 ꢀ
10-3 s-1 at 190 °C) similar to that for 6.
Introduction
Hydrocarbon CH activation followed by heteroatom
functionalization has great potential for installing diverse
functionality into C-H bonds.1 The most efficient hydro-
carbon hydroxylation catalysts utilize platinum, palladium,
mercury, and gold complexes and operate by a sequence of
electrophilic C-H bond activation2 followed by reductive
functionalization to generate the C-OR functionality. Be-
cause of their electrophilic character, these previous genera-
tions of hydroxylation catalysts are highly susceptible to
water and alcohol inhibition and therefore require strong
acid solvents to minimize this inhibition.2a
To thwart this inhibition, we decided to modify the highly
effective Pt(bipyrimidine)L2 motif into a less electrophilic3
catalyst by increasing the electron density at the metal center
Figure 1. Progression toward less electrophilic CH activation
catalysts.
by using a more electron donating ligand and replacing
platinum with iridium (Figure 1). Previously, our group
reported the synthesis of a Pt(NNCt-Bu)TFA complex
(NNCt-Bu = 6-phenyl-4,40-di-tert-butyl-2,20-bipyridine and
TFA = trifluoroacetate) that promotes H/D exchange of
benzene and trifluoroacetic acid with an activation barrier of
∼30 kcal/mol.3c More importantly, we have recently re-
ported that the (NNCt-Bu)Ir(TFA)2(C2H4) complex cataly-
tically activates and functionalizes methane in trifluoroacetic
acid.4 This system was developed on the basis of the results
that treatment of (NN)(NC)IrIII(Me)OTf (NN = κ2-4,40-
di-tert-butyl-2,20-bipyridine, NC = κ2-(N,C-3)-6-phenyl-2,20-
bipyridine, OTf = trifluoromethanesulfonate) with PhI-
(TFA)2 (PITFA) results in efficient oxy functionalization
to generate the methyl ester.5
*To whom correspondence should be addressed. E-mail: rperiana@
scripps.edu (R.A.P.); wag@wag.caltech.edu (W.A.G.).
(1) (a) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507. (b) Waltz, K.
M.; Hartwig, J. F. Science 1997, 277, 211. (c) Crabtree, R. H. Dalton Trans.
2001, 2951. (d) Conley, B. L.; Tenn, W. J., III; Young, K. J. H.; Ganesh, S. K.;
Meier, S. K.; Ziatdinov, V. R.; Mironov, O. A.; Oxgaard, J.; Gonzales, J.;
Goddard, W. A., III; Periana, R. A. J. Mol. Catal. A 2006, 251, 8.
(2) (a) Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.;
Fuji, H. Science 1998, 280, 560. (b) Periana, R. A.; Mironov, O.; Taube,
D. J.; Bhalla, G.; Jones, C. J. Science 2003, 301, 814. (c) Periana, R. A.;
Taube, D. J.; Evitt, E. R.; Loffler, D. G.; Wentrcek, P. R.; Voss, G.; Masuda, T.
Science 1993, 259, 340. (d) Jones, C. J.; Taube, D.; Ziatdinov, V. R.; Periana,
R. A.; Nielsen, R. J.; Oxgaard, J.; Goddard, W. A., III Angew. Chem., Int. Ed.
2004, 43, 4626. (e) Periana, R. A.; Mironov, O. A.; Taube, D. J.; Gamble, S.
Chem. Commun. 2002, 2376.
(3) (a) Zhong, H. A.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.
2002, 124, 1378. (b) Periana, R. A.; Bhalla, G.; Tenn, W. J., III; Young,
K. J. H.; Liu, X. Y.; Mironov, O. A.;Jones, C. J.; Ziatdinov, V. R. J. Mol. Catal.
A 2004, 220, 7. (c) Young, K. J. H.; Meier, S. K.; Gonzales, J. M.; Oxgaard, J.;
Goddard, W. A., III; Periana, R. A. Organometallics 2006, 25, 4734.
(4) Young, K. J. H.; Oxgaard, J.; Ess, D. H.; Meier, S. K.; Stewart,
T.; Goddard, W. A., III; Periana, R. A. Chem. Commun. 2009, 3270.
(5) (a) Young, K. J. H.; Mironov, O. A.; Periana, R. A. Organome-
tallics 2007, 26, 2137. (b) Young, K. J. H.; Yousufuddin, M.; Ess, D. H.;
Periana, R. A. Organometallics 2009, 28, 3395.
r
2009 American Chemical Society
Published on Web 09/02/2009
pubs.acs.org/Organometallics

5294 Organometallics, Vol. 28, No. 18, 2009
Meier et al.
Scheme 1. Synthesis of 6
Several groups have developed systems that are competent
for CH activation in aqueous or mixed organic/aqueous
media. The most well-known system that operates in water
is the Shilov6 catalyst that uses K2PtCl4. Gunnoe7 and Lau8
in separate work have also demonstrated that ruthenium
hydro(trispyrazolyl)borate complexes promote H/D ex-
change between water and arene substrates. More examples
include the work of Milstein and Leitner,9 who utilized a
ruthenium pincer complex, the classic Bergman10 (Cp)-
Ir(PMe3)Cl2 complex, and more recently Goldberg’s11
(PNP)Rh(OPh) complex. Carmona and Poveda12 have also
shown that H/D exchange occurs between tetrahydrofuran
and water using a (TpMe2)IrH4 catalyst (TpMe2 = hydro(tris-
(3,5-dimethylpyrazolyl)borate). In addition, our own work
previously showed that the (κ2-O,O-acac)2Ir(OH)(OH2)
complex (acac = acetylacetonate) catalyzes H/D exchange
between benzene and water.13
Here we report the synthesis and experimental study of
benzene and water H/D exchange catalyzed by a dihydroxo
iridium(III) pyridine complex, (NNCt-Bu)Ir(OH)2Py (6).
Density functional calculations are also reported that were
used to explore several possible mechanisms for H/D ex-
change.
Results and Discussion
Synthesis. Scheme 1 shows the synthesis of the (NNCt-Bu)-
Ir(OH)2Py complex. With (NNCt-Bu)Ir(Et)(Cl)NCCH3 (1)4
as the starting material, the chloride was replaced with
the more labile trifluoroacetate group by stirring complex 1
for 2 days in the presence of silver trifluoroacetate to yield
(NNCt-Bu)Ir(Et)(TFA)NCCH3 (2) in 78% yield after recrys-
tallization. The ethyl group was also replaced by stirring
complex 2 under argon in neat trifluoroacetic acid for 12 h at
room temperature. After removal of the acid under reduced
pressure and purification by column chromatography, a
yellow, air-stable solid was obtained in 60% yield as the
complex (NNCt-Bu)Ir(TFA)2NCCH3 (3). Fluorine NMR
analysis shows only one fluorine resonance, suggesting that
the trifluoroacetate groups are oriented trans to one another.
Direct routes to (NNCt-Bu)Ir(OH)2L (L = H2O, NCCH3)
were unsuccessful, due to the insolubility of 3 in water.
Therefore, the trifluoroacetate ligands were replaced with
methoxo ligands, since we knew that 3 was soluble in
methanol. The dimethoxo precursor complex was advanta-
geous because the dimethoxo product can easily be identified
by 1H NMR due to the presence of the methyl protons of the
methoxo groups. The reaction of 3 in methanol with 16 equiv
(6) (a) Goldshleger, N. F.; Tyabin, M. B.; Shilov, A. E.; Shteinman,
A. A. Zh. Fiz. Khim. 1969, 43, 2174. (b) Shilov, A. E.; Shteinman, A. A.
Coord. Chem. Rev. 1977, 24, 97.
(7) (a) Feng, Y.; Lail, M.; Barakat, K. A.; Cundari, T. R.; Gunnoe,
T. B.; Petersen, J. L. J. Am. Chem. Soc. 2005, 127, 14174. (b) Feng, Y.;
Lail, M.; Foley, N. A.; Gunnoe, T. B.; Barakat, K. A.; Cundari, T. R.; Petersen,
J. L. J. Am. Chem. Soc. 2006, 128, 7982.
(8) Leung, C. W.; Zheng, W.; Wang, D.; Ng, S. M.; Yeung, C. H.;
Zhou, Z; Lin, Z.; Lau, C. P. Organometallics 2007, 26, 1924.
€
(9) Prechtl, M. H. G.; Holscher, M.; Ben-David, Y.; Theyssen, N.;
Loshcen, R.; Milstein, D.; Leitner, W. Angew. Chem., Int. Ed. 2007, 46,
2269.
(10) (a) Klei, S. R.; Golden, J. T.; Tilley, T. D.; Bergman, R. G. J. Am.
Chem. Soc. 2002, 124, 2092. (b) Klei, S. R.; Tilley, T. D.; Bergman, R. G.
Organometallics 2002, 21, 4905.
(11) (a) Kloek, S. M.; Heinekey, D. M.; Goldberg, K. I. Angew.
Chem., Int. Ed. 2007, 46, 4736. (b) Hanson, S. K.; Heinekey, D. M.;
Goldberg, K. I. Organometallics 2008, 27, 1454.
ꢀ
(12) Gutierrez-Puebla, E.; Monge, A.; Paneque, M.; Poveda, M. L.;
Taboada, S.; Trujillo, M.; Carmona, E. J. Am. Chem. Soc. 1999, 121,
346.
(13) (a) Tenn, W. J., III; Young, K. J. H.; Bhalla, G.; Oxgaard, J.;
Goddard, W. A., III; Periana, R. A. J. Am. Chem. Soc. 2005, 127, 14172.
(b) Tenn, W. J., III; Young, K. J. H.; Oxgaard, J.; Nielsen, R. J.; Goddard,
W. A., III; Periana, R. A. Organometallics 2006, 25, 5173.
1
of sodium methoxide was followed by H NMR. After the

Article
Organometallics, Vol. 28, No. 18, 2009 5295
Table 1. Percentage of Benzene Isotopologues Resulting from Control (Left) and Catalytic (Right) Reactions at 180 °Ca
control reactionb
catalytic reactionc
isotopologue
277 min
1980 min
3060 min
isotopologue
277 min
1980 min
3060 min
C6H6
99.9
99.7
0.24
99.04
0.93
C6H6
99.81
90.67
8.91
0.38
0.01
0
87.62
11.77
0.55
0.04
0
C6H5D1
C6H4D2
C6H3D3
C6H2D4
C6H1D5
C6D6
0
0
0
0
0
0
C6H5D1
C6H4D2
C6H3D3
C6H2D4
C6H1D5
C6D6
0.15
0.01
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
a Concentrations are given on the basis of total volume of solution present. b H/D exchange analysis. Conditions: 0.16 M KOD, 0.5 mL of D2O, 0.5 mL
of benzene-H6. c H/D exchange analysis. Conditions: 2.13 mM of 6, 0.16 M KOD, 0.5 mL of D2O, 0.5 mL of benzene-H6.
addition of the sodium methoxide, the Ir-NCCH3 peak
could no longer be observed by 1H NMR. Additional heat-
ing at 70 °C led to multiple intractable products. The
instability of 3 in the presence of base is likely due to the
reaction of the NCCH3 with base. Therefore, NCCH3 was
substituted with a more stable ligand by heating complex 3 in
neat pyridine overnight at 100 °C. Unreacted pyridine was
removed under reduced pressure, and purification of the
product by column chromatography yielded the air-stable
orange solid (NNCt-Bu)Ir(TFA)2Py (4) in 94% yield. Com-
plex 4 was not soluble in water, which ruled out direct routes
to a hydroxo complex. Therefore, 4 was heated at 70 °C for
24 h in methanol containing 16 equiv of sodium methoxide.
The product, (NNCt-Bu)Ir(OMe)2Py (5), was obtained by
removal of the excess sodium methoxide and methanol, and
it was purified by column chromatography, resulting in a
black solid in 35% yield. Complex 5 was immediately con-
verted to (NNCt-Bu)Ir(OH)2Py (6) after purification. Com-
plex 6 was obtained by heating 5 in a 1:1 mixture of THF and
H2O at 85 °C for 2 days. Complex 6 was obtained in 21%
yield as a black solid by removal of the solvent and purifica-
tion by column chromatography. Complex 6 was fully
inability to stir the solution in the NMR tube. The thermal
stability tests carried out in Schlenk tubes resulted in pre-
cipitation of a tan solid within several hours of heating the
solution at 150 °C. 1H NMR analysis of the tan solid, using
deuterated acetone, revealed a broad range of aromatic and
tert-butyl NMR peaks, indicating a mixture of products.
Catalytic H/D Exchange. Benzene and water H/D ex-
change was carried out using a 1:1 mixture of benzene-H6
and D2O, resulting in an approximately 2.0-3.5 mM solu-
tion of catalyst 6. Deuterium incorporation into benzene was
monitored up to 5% deuterium incorporation at 160-190 °C
by analysis of the benzene phase by GC-MS using a decon-
volution program (see the Experimental Section). Initial H/
D exchange studies between water and benzene without
added KOD showed little to no H/D exchange over a 48 h
period (less than 10 turnovers at 180 °C). Knowing that
added base stabilizes 6, 20 equiv of KOD was added to the
reaction mixture. H/D exchange was observed to be steady
over time with a turnover frequency (TOF) of 2.0 ꢀ 10-3 s-1
at 180 °C. Control reactions were carried out with the
same amount of KOD, and all H/D exchange results are
reported with background corrections. Table 1 gives the
analysis of the catalytic H/D exchange compared to the
background.
Figure 2 shows a stirring rate study for H/D exchange
between benzene-H6 and D2O using 6 as a catalyst at 190 °C
to determine if the reaction was mass transfer limited. The
highest temperature used for H/D exchange studies was
chosen because the chemical rates would be the most com-
petitive with the rate of diffusion. Figure 2 shows that for the
range of 300-1000 rpm the catalyst is not mass transfer
limited.14
Similar turnover frequencies (TOF) were observed for
benzene/water H/D exchange with varying amounts of
KOD (Figure 3). However, KOD concentration dependence
cannot be ruled out, because it appears that the catalyst is
operating in the benzene phase, where the amount of base is
likely saturated. When the reaction mixture is loaded into a
Schlenk flask, the aqueous phase is red due to the presence of
the catalyst and the benzene phase is colorless. Once the
reaction mixture is heated at 180 °C, the benzene layer turns
black. If the reaction is stopped after a short period of time,
the benzene phase is black and the aqueous phase is colorless,
indicating that the catalyst likely resides in the benzene
phase. To further confirm this, we carried out the H/D
1
characterized by H and 13C NMR, high-resolution mass
spectrometry, and elemental analysis. The -OH protons
for 6 were observed in the 1H NMR at δ -2.71 ppm in CDCl3.
Attempts to obtain a crystal structure were unsuccessful.
Stability. The stability of 6 in water was tested by heating a
5.0 mM solution of 6 in D2O under argon inside of a J. Young
NMR tube. As the solution was heated at 100 °C over a
1
period of 8 h, two new species were observed by H NMR
(see the Supporting Information), and attempts to isolate
these two species by preparative thin-layer chromatography
failed. During the course of this reaction the concentration of
free pyridine did not increase. Hydroxo ligand loss could not
be ruled out. The loss of the hydroxo group could potentially
produce a dinuclear species of the type [(NNCt-Bu)IrPy(μ-
OH)]22þ (see Density Functional Theory Investigation). Ad-
dition of KOD and further heating did not result in the
conversion of the two unknown species back to 6. The two
unknown species produced by heating 6 in H2O are also
observed in the resulting crude mixture from the synthesis of
6. Hypothesizing that loss of a hydroxide group was resulting
in the formation of these byproducts, we repeated the
stability reaction in the presence of 4 equiv of KOD. After
1
2 h at 130 °C only 6 was observed by H NMR (see the
Supporting Information). At 150 °C the concentration of 6
slightly decreases relative to an external standard and small
amounts of solid were observed on the wall of the NMR
tube. Similar studies were also carried out in Schlenk tubes to
confirm that the charred material was not produced from the
(14) However, these stirring studies were performed using PTFE stir
bars rather than an overhead-driven stirrer with an impeller. As a result,
at high stir rates it is possible that the solution is vortexing rather than
efficiently mixing.

5296 Organometallics, Vol. 28, No. 18, 2009
Meier et al.
exchange of benzene and D2O using 6. The reaction was
carried out at 190 °C for 17 h, and the benzene phase was
completely removed and analyzed by GC-MS, resulting in a
TOF of 7.5 ꢀ 10-3 s-1 (23.2% C6H5D and 2.8% C6H4D2)
with the results corrected for background H/D exchange.
New benzene containing no catalyst was added to the
remaining aqueous solution, and the reaction mixture was
heated for another 17 h at 190 °C under argon. Analysis of
the benzene phase showed no H/D exchange, which indicates
that no catalyst is present in the aqueous phase.
H/D exchange studies with 6 showed an inverse depen-
dence of pyridine through a 1/[Py] study conducted at 190 °C
(see the Supporting Information). Figure 4 shows an Eyring
plot for H/D exchange of benzene and water over a tem-
perature range of 160-190 °C. Linear regression analysis
gives an activation enthalpy (ΔHq) of 33((2) kcal/mol.
Phenyl Complex. Carrying out the stoichiometric reaction
under the catalytic conditions used for H/D exchange yielded
multiple products. One of the possible products from C-H
activation of benzene is (NNCt-Bu)Ir(Ph)(OH)Py (7). Our
group previously reported a similar complex, (NNCt-Bu)Ir-
(Ph)(Cl)Py (8), which was used as a precursor for the
attempted synthesis of 7 (Scheme 2).4 Complex 8 was heated
in THF at 70 °C for 8 h with 4 equiv of cesium hydroxide.
Rather than the predicted complex 7, the μ-hydroxo di-
nuclear complex [(NNCt-Bu)Ir(Ph)(μ-OH)]2 (9) was isolated
by recrystallization.
The dinuclear complex was confirmed by synthesis
from the previously reported μ-chloro dinuclear complex
[(NNCt-Bu)IrPh(μ-Cl)]2 (10).4 Heating 10 at 70 °C for 8 h
with 4 equiv of cesium hydroxide in THF followed by
recrystallization using CH2Cl2 and pentane also gave 9, a
black, air-stable solid in 38% yield. Complex 9 was char-
acterized by 1H and 13C NMR, high-resolution mass spectro-
metry, elemental analysis, and X-ray crystallography
(Figure 5).
Figure 2. Stirring rate study for H/D exchange between ben-
zene-H6 and D2O in the presence of 6.
Since the chloro group could be replaced with a more labile
group, such as trifluoroacetate, it might be possible to
synthesize 7 under milder conditions. Therefore, the
(NNCt-Bu)Ir(Ph)(TFA)Py species 11 was synthesized from
8 using silver trifluoroacetate in CH2Cl2. This reaction was
sluggish at room temperature, taking approximately 7 days
to run to completion. After silver chloride and solvent were
removed, 11 was obtained by recrystallization as a reddish
orange, air-stable solid in 85% yield. Attempts to synthesize
7 from 11 proved unsuccessful. Reacting 11 in THF-d8 with 4
equiv of cesium hydroxide at room temperature over 2 days
produced 9 in 49% yield by proton NMR using internal
standardization. These reactions suggest that the ΔG value
Figure 3. Plot of turnover frequency vs equivalents of KOD.
Determination of base dependence on H/D exchange.
Figure 4. Eyring plot for H/D exchange by 6 with benzene-H6 and D2O.

Article
Organometallics, Vol. 28, No. 18, 2009 5297
Figure 5. ORTEP diagram of 9. Thermal ellipsoids are given at the 50% probability level, with hydrogens, a molecule of water, and
˚
CH2Cl2 omitted for clarity. Selected bond distances (A): Ir(1)-O(1), 2.093(8); Ir(1)-C(1), 1.99(14); Ir(1)-N(1), 2.150(9); Ir(1)-N(2),
1.964(10). Selected bond angles (deg): Ir(1)-O(1)-Ir(2), 101.7(3); N(2)-Ir(1)-O(1), 170.7(4).
Scheme 2. Synthesis of Possible Phenyl Intermediate
1
for formation of 9 from 7 is favorable and 9 likely efficiently
enters the H/D exchange catalytic cycle. In fact, 9 catalyzes
H/D exchange between benzene and water with a TOF
of 7.0 ꢀ 10-3 s-1 at 190 °C. This TOF is very similar to
that observed on starting with 6 (TOF = 6.4 ꢀ 10-3 s-1
at 190 °C).
chemical shift in the H NMR of the catalytic reaction is
at δ 8.5 ppm. Known mononuclear complexes used as
precursors during synthesis have a characteristic chemical
shift for the ortho proton on the pyridyl (NNCt-Bu) ligand
that appears at δ 9 ppm or further downfield. The two phenyl
dinuclear complexes 9 and 10, [(NNCt-Bu)Ir(Ph)(μ-X)]2 (X =
Cl, OH), have their farthest downfield chemical shift no
greater than δ 8.5 ppm in the 1H NMR. This suggests that the
product mixture likely contains various dinuclear com-
plexes. Further support for this comes when 6 is heated in
a J. Young NMR tube containing H2O, KOH, and C6D6.
Monitoring the NMR reaction over time shows that 6
essentially goes away over the course of 4 h at 160 °C with
the formation of free pyridine. The pyridine formation was
also confirmed by GC-MS. The formation of free pyridine in
the reaction suggests that OH-bridged dinuclear complexes
We attempted to directly produce 9 from 6 by carrying out
the H/D exchange reaction on a preparative scale. Analysis
1
of the H NMR spectrum of the catalytic reaction did not
reveal the presence of 6 or the NNCt-Bu ligand, and the
presence of 9 is difficult to determine, due to the large
number of aromatic peaks from the multiple products pro-
duced in the reaction. Attempts to isolate 9 from this mixture
failed both by recrystallization and by separation on column
chromatographic supports due to the instability of 9 on
chromatographic material. However, the furthest downfield

5298 Organometallics, Vol. 28, No. 18, 2009
Meier et al.
Scheme 3. Benzene and D2O H/D Exchange Reaction using 6
under Catalytic Conditions To Detect Phenyl Intermediates
the thermodynamics for solvent-ligand substitution and
dinuclear coordination were computed (Scheme 4). The
pyridine ligand in structure 12 can rearrange from the
meridional position to the apical position in 13. This
isomerization slightly favors structure 13 by -0.3 kcal/
mol and is possible if either pyridine or hydroxide dissoci-
ates. Pyridine dissociation in water to give structure 14 is
endergonic by 13.5 kcal/mol. Water coordination to gen-
erate 15 further increases the free energy to 15.5 kcal/mol,
while hydroxide coordination to generate the anionic tri-
hydroxo species 16 is slightly favorable from 14 but still
endergonic by 10.9 kcal/mol compared to 12. Outer-sphere
explicit water molecules were also used to test whether
hydrogen bonding would substantially lower the energies
of complexes 14-16. However, the typical energy difference
by using a mixed explicit/implicit solvation model was only
∼1 kcal/mol.
are forming. It should also be pointed out that during the
course of the NMR reaction the number of tert-butyl groups
increases (greater than nine sets of tert-butyl groups) over
4 h, indicating that multiple species are formed in the
reaction. To determine how much of the products in the
preparative-scale reaction contain a phenylated species, tri-
fluoromethanesulfonic acid was reacted with the product
mixture (Scheme 3). The resulting protonolysis resulted in
free benzene-H6 and was quantified by using mesitylene as an
internal standard. The benzene produced was determined to
be 40%, on the basis of the amount of 6 used. Decomposition
of the NNCt-Bu ligand was not observed in the presence of
trifluoromethanesulfonic acid. We attempted to cleave the
iridium-phenyl bond of 9 by heating a suspension of 9 in a
0.01 M KOD/D2O solution at 130-160 °C, but only the
starting material 9 was recovered. Higher temperatures
resulted in charring of the material.
Density Functional Theory Investigation. To study possible
catalytic intermediates and low-energy pathways for benzene
H/D exchange, we have employed density functional theory
(DFT) calculations. All DFT calculations were performed in
Jaguar 7.0 and 7.5.15 All reactant and transition structures
were optimized using B3LYP/LACVP** hybrid-density
functional theory. Reported free energies are the sum ΔE
(B3LYP/LACV3Pþþ**//B3LYP/LACVP**) þ ΔU þ pV
-298ΔS þ ΔG(solvation); U and S values were obtained from
B3LYP/LACVP** structures. Selected M06/LACV3Pþþ**16
functional electronic energies evaluated using B3LYP struc-
tures are also reported. M06 calculations were run using tight
convergence criteria and a very dense grid.17 Benzene (ε =
2.284, radius probe 2.6) and water (ε = 80.37, radius probe
1.4) implicit free energy of solvation corrections were applied
using the Poisson-Boltzmann solvation model. All transi-
tion structure figures were generated using CYL mol.18 The
NNC pincer ligand was modeled without the tert-butyl
groups, and structure 12 is the model of complex 6.
Although hydroxide dissociation to give the cationic five-
coordinate pseudo-trigonal-bipyramidal structure 17 is en-
dergonic by 10.3 kcal/mol, the formation of the dinuclear
species 18 from the combination of two hydroxo-pyridine
species is exergonic by ∼-13 kcal/mol. The M06 functional
predicts a slightly more exergonic dincular species of -17
kcal/mol. It is possible that substantial hydroxide concen-
tration offsets this equilibrium and prevents formation of 18
and other species.
Without substantial KOD, benzene-D2O hydrogen-
deuterium exchange is also very slow. Although complex 6
is initially soluble in water, the CH activation chemistry was
determined experimentally to occur in the benzene phase.
Although water and hydroxide have limited solubility in
benzene, it is plausible that water or hydroxide may assist or
be directly involved in the mechanism of H/D exchange.
Therefore, we have extensively explored possible C-H acti-
vation transition states with explicit hydroxide and water
molecules.
In benzene, B3LYP predicts pyridine dissociation to gen-
erate the square-pyramidal five-coordinate intermediate 14
to require 13.1 kcal/mol (ΔGM06 = 21.1 kcal/mol). Coordi-
nation of benzene via η2 binding can potentially occur cis
(19) or trans (21) to the NNC pincer ligand (Scheme 5).
However, attempts to optimize structure 21 resulted in a
dissociated benzene molecule and 14 due to the labilizing
effect of the cyclometalated NNC pincer ligand. Cis η2
binding to give 19 is 29.0 kcal/mol above 14. From 19,
C-H bond substitution may occur if the hydrogen is trans-
ferred to an inner-sphere hydroxo ligand with concomitant
Ir-C bond formation. The resulting iridium phenyl species
20 is endergonic by 7.0 kcal/mol, while the alternative ligand
arrangement (structure 22) either generated directly or re-
sulting from water dissociation is significantly more ender-
gonic with a free energy of 19.9 kcal/mol. Although 20 and 22
are endergonic, water loss followed by pyridine recoordina-
tion gives the exergonic structures 24 and 25. Importantly,
the dinuclear species 23 (Scheme 5b) formed from the various
iridium phenyl species is exergonic. Structures 25 and 23 are
predicted to be the lowest energy intermediates upon reac-
tion of 12 with benzene.
Experimentally, when 6 was heated in a solution of D2O,
several new species appeared in the proton NMR spectra.
However, addition of 4 equiv of KOD suppressed the
generation of these new species. To investigate the role of
hydroxide and identify possible species generated in water,
A search for possible transition states revealed a viable
CH activation route to structure 20 by an internal sub-
stitution (IS) transition state (TS1, Figure 6).19 In this
(15) (a) Jaguar, version 7.0; Schrodinger, LLC, New York, 2007.
(b) Jaguar, version 7.5; Schrodinger, LLC, New York, 2009.
(16) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157.
(17) Jaguar options applied: iacc=1; gdftmed=-14; gdftfin=-14;
gdftgrad=-14.
ꢀ
(18) Legault, C. Y. CYLview, 1.0b; Universite de Sherbrooke, 2009
(19) The π-complex 19 is not directly connected to 20, but rather
there is an agostic interaction with the benzene C-H bond.
(http://www.cylview.org).

Article
Organometallics, Vol. 28, No. 18, 2009 5299
Scheme 4. Relative free energies in water. Values in parentheses are relative free energies in benzene. (kcal/mol)
Scheme 5. Relative Reaction Free Energies for (a) Benzene CH Activation and (b) Dinuclear Formation
˚
four-membered metathesis-like transition structure, the
˚
2.14 A. This IS transition state is similar to the previously
reported transition states for the reactions of (κ2-O,O-
acac)2Ir(OH)(pyridine) and (κ2-O,O-acac)2Ir(OMe)-
(pyridine) with benzene.13 In TS1 the benzene hydrogen
atom is transferred to the oxygen lone pair while the
C-H bond is substantially broken to 1.33 A and the
hydrogen is transferred to the oxygen atom at a distance
˚
of 1.34 A. The newly forming Ir-Ph bond has a distance of
˚
2.32 A, and the Ir-OH bond is only slightly stretched to

5300 Organometallics, Vol. 28, No. 18, 2009
Meier et al.
Figure 6. Potential benzene C-H bond activation transition states (a) cis to NNC pincer ligand and (b) trans to NNC pincer ligand.
hydroxo group remains σ-ligated, which is different from a
classic σ-bond metathesis mechanism.20 Although the cyclo-
metalated NNC pincer ligand is electron donating, the iridium
center remains electrophilic with a Mulliken charge of 0.46e to
induce Ir-Ph bond formation while the nucleophilic iridium
hydroxide fragment abstracts the protic benzene hydrogen.
The Mulliken charge on the benzene H atom is 0.37e. The
benzene carbon atom and the iridium hydroxide group are
both negatively charged in TS1: -0.20e and -0.62e, respec-
tively. Because of the electrophilic/nucleophilic character of
this substitution transition state, the benzene C-H bond is
activated by an ambiphilic mechanism.20e
differences in free energies of solvation, it is unlikely that water
is directly involved in CH activation in the benzene phase.
No transition states were located for direct deprotonation of
benzene by an external hydroxide with concomitant Ir-Ph
bond formation. Alternatively, transition states TS3 and TS6
were located, which utilize an outer-sphere hydroxide to
deprotonate the Ir-OH bond during CH activation. This
hydroxide stabilization substantially alters the transition-state
geometry compared to TS1. TS3 is earlier along the reaction
˚
pathway; the forming Ir-Ph bond is longer (2.49 A), and the
˚
breaking C-H bond is shorter (1.22 A). This transition state
may best be described as the reaction of benzene with an iridium
˚
˚
The activation free energy for TS1 is 33.6 kcal/mol relative
to structure 14; the overall activation free energy from 12 is
46.7 kcal/mol. This is slightly too large to fit with experiment.
The M06 functional predicts a more reasonable free energy
barrier of 25.4 kcal/mol relative to 14 and 38.5 kcal/mol
relativeto12.21 A similar transitionstatefor forming structure
22 via TS4 was also located but is 5.1 kcal/mol higher due
to the trans NNC ligand effect and steric crowding by the
NNC ligand, which results in twisting of the incoming phenyl
group.
A water molecule may directly assist in CH activation by
shuttling a proton from benzene to the hydroxo group. How-
ever, these six-membered “water-assisted” transition states,
TS2 and TS5, are 5.6 and 10.9 kcal/mol above TS1. In
combination with the penalty for H2O to shuttle from water
into benzene, estimated as ∼6 kcal/mol on the basis of
oxide. The Ir-O bond shortens from 2.14 A in TS1 to 1.99 A in
TS3, and the new hydroxide-hydrogen bond is nearly fully
˚
formed at 1.03 A. However, despite the large effect of the
hydroxide on the transition-state geometry, TS3 and TS6
are 9.5 and 13.3 kcal/mol above TS1. These transition states
are even less viable, considering the very poor solubility of
hydroxide in benzene and differences in solvation free energy.
As further evidence against water and/or hydroxide direct
participation in CH activation, we also considered the
energies for TS1-TS3 in water. Relative to structure 14,
the activation free energies are 24.7, 36.6, and 57.6 kcal/mol
for TS1-TS3. This eliminates the possibility of water and
hydroxide participation, but TS1 is 8.9 kcal/mol lower in
energy in water than in benzene. The resulting iridium phenyl
aquo structure 20 and dinuclear structure 23 are both
significantly more exergonic in water: -7.9 and -34.2
kcal/mol relative to 14. However, as shown by the indepen-
dent synthesis of 9/23, this complex is not soluble in water.
Although it is beyond the scope of this current investigation,
there remains the possibility that CH activation and H/D
exchange may occur at the water-benzene interface whereby
solubility and solvation effects are optimally balanced.
We have also explored whether the NNC pincer ligand
is a spectator ligand or active participant in H/D exchange.
(20) (a) Oxgaard, J.; Tenn, W. J., III; Nielsen, R. J.; Periana, R. A.;
Goddard, W. A., III Organometallics 2007, 26, 1565. (b) Ess, D. H.;
Bischof, S. M.; Oxgaard, J.; Periana, R. A.; Goddard, W. A., III Organome-
tallics 2008, 27, 6440. (c) Ryabov, A. D. Chem. Rev. 1990, 90, 403.
(d) Canty, A. J.; van Koten, G. Acc. Chem. Res. 1995, 28, 406. (e) Ess,
D. H.; Nielsen, R. J.; Goddard, W. A., III; Periana, R. A. J. Am. Chem. Soc.
2009, 131, 11686.
(21) MPW1K/LACV3P++** predicts
a free energy barrier of
25.5 kcal/mol, similar to that for M06.

Article
Organometallics, Vol. 28, No. 18, 2009 5301
Figure 7. Proton transfer transition states to the NNC ligand.
Scheme 6
Figure 8. Benzene substitution and oxidative addition transi-
tion states with NNC pyridine ligand dissociation.
37.9 kcal/mol relative to 14, making this pathway 4.3 kcal/mol
higher than the pathway utilizing TS1. The direct route to 27
via TS8 or TS9 (Figure 7) requires 3.5 and 8.8 kcal/mol more
free energy than TS7.
Addition of the benzene C-H bond across the cyclometalated
Ir-C bond would give the iridium phenyl species 26
(Scheme 6). No oxidative Ir(V) hydride intermediate was
located between 14 and 26. The transition state, TS7, is similar
to other so-called “oxidative hydrogen migration” (OHM)
transition states where there is significant oxidative character
but no intermediate.22 Inthismetal-assistedσbondmetathesis
transition state the newly forming Ir-Ph bond and breaking
An important issue to explore is the influence and neces-
sity of the pincer-coordinated cyclometalated NNC ligand
for CH activation. It is possible that the pyridine portion of
the NNC ligand could decoordinate from the iridium center
to generate a more electrophilic metal center. In the ground
state, pyridine dissociation requires 11-14 kcal/mol of free
energy. The corresponding transition state, TS10 (Figure 8),
has an activation free energy of 47.8 kcal/mol. This is 14.2
kcal/mol above the fully coordinated transition state TS1,
showing that there is no beneficial transition state effect for
pyridine to decoordinate, and so it prefers to remain ligated
during CH activation. It is also conceivable that pyrdine
decoordination would open two cis sites to allow for oxida-
tive addition to an Ir(V) hydride species. Although this
transition state, TS11 (ΔGq = 45.2 kcal/mol), is lower in
energy than TS10, where no formal oxidation takes place, it
is still too high to be competitive with TS1.
On the basis of the activation free energies for all of the
transition states explored, catalytic H/D exchange between
benzene and water likely occurs via an internal substitution
pathway, as outlined in Scheme 7. With significant hydro-
xide concentration, the starting structure 12 will not convert
into the lower energy dinuclear species 18. From 12, the π-
complex 19 is generated via pyridine loss and benzene
coordination. Rearrangement to interaction with the C-H
bond followed by cleavage with a hydroxo group gives the
iridium phenyl species 20. Water-D2O exchange may occur
at this point and then undergo the microscopic reverse of CH
˚
Ir-C bond lengths are very similar: 2.13 and 2.09 A, respec-
tively. However, the hydrogen atom is not equally shared, due
to the constraints imposed by the NNC ligand bonding. The
˚
breaking C-H bond is significantly more broken (2.06 A) and
˚
the forming bond C-H bond is much shorter (1.51 A). The
OHM transition state TS7 (see Scheme 6 and Figure 7) is only
0.8 kcal/mol above TS1. Neither water nor hydroxide assis-
tance lowers the energy of TS7. From 26, one pathway
whereby H/D exchange may occur is by coordination of
D2O to the vacant site followed by intramolecular proton-
ation to generate the trihydroxo species 27. Water coordina-
tion to 26 is favorable by 10.0 kcal/mol. However, the free
energy barrier for phenyl protonation of 26 to 27 is 19.4 and
(22) (a) Oxgaard, J.; Muller, R. P.; Goddard, W. A., III; Periana,
R. A. J. Am. Chem. Soc. 2004, 126, 352. (b) Webster, C. E.; Fan, Y.; Hall,
M. B.; Kunz, D.; Hartwig, J. F. J. Am. Chem. Soc. 2003, 125, 858.
(c) Hartwig, J. F.; Cook, K. S.; Hapke, M.; Incarvito, C. D.; Fan, Y.; Webster,
C. E.; Hall, M. B. J. Am. Chem. Soc. 2005, 127, 2538. (d) Ng, S. M.; Lam,
W. H.; Mak, C. C.; Tsang, C. W.; Jia, G.; Lin, Z.; Lau, C. P. Organometallics
2003, 22, 641. (e) Lam, W. H.; Jia, G.; Lin, Z.; Lau, C. P.; Eisenstein, O.
Chem. Eur. J. 2003, 9, 2775. (f) In these references OHM transition states
have been referred to as metal-assisted σ-bond metathesis, oxidatively added
transition state, and σ-complex assisted metathesis. For a recent review see:
Perutz, R. N.; Sabo-Etienne, S. Angew. Chem., Int. Ed. 2007, 46, 2578.

5302 Organometallics, Vol. 28, No. 18, 2009
Scheme 7. Free Energy Profile for IS Catalytic H/D Exchange
Meier et al.
activation. After water loss and prior to D2O coordination it
is also likely that the dinuclear structure 23 is formed in a
reversible process. Structures 12 and 23 are nearly equal in
free energy, and therefore, formation of this dinuclear spe-
cies does not increase the free energy span for H/D exchange.
benzene were dried over sodium/benzophenone and distilled
under argon. Dichloromethane (stabilizer removed with sulfuric
acid) was dried over P2O5 and distilled under argon. Complexes
1, 8, and 10 were prepared according to the literature.4 Chro-
matotron (centrifugal thin-layer chromatography) plates were
made using aluminum oxide neutral (type E) that was purchased
from EMD.
Conclusion
Preparation of (NNCt-Bu)Ir(Et)(TFA)NCCH3 (2). Under an
inert atmosphere, CH2Cl2 (30 mL) was added to a Schlenk flask
containing 1 (620.4 mg, 0.9674 mmol) and silver trifluoroacetate
(235.0 mg, 1.064 mmol). The mixture was stirred in the dark for
2 days, followed by filtration over Celite to remove the silver
chloride. The solvent was removed under reduced pressure, and
the product was obtained by recrystallization from CH2Cl2 and
In summary, we have reported the synthesis of a water-
soluble cyclometalated dihydroxo iridium(III) pyridyl
complex, 6. This complex is capable of H/D exchange
between water and benzene in the presence of base. Experi-
mental and density functional studies suggest that H/D
exchange occurs through disassociation of pyridine to gen-
erate a cis-dihydroxo complex, which then heterolytically
activates the benzene C-H bond by an ambiphilic substitu-
tion transition state.
1
pentane at -30 °C, resulting in a 78.2% (543.8 mg) yield. H
NMR (CDCl3, 400 MHz): δ 8.87 (d, 1H, 3J = 6.0 Hz), 7.85 (d,
1H, 4J = 1.6 Hz), 7.63 (d, 1H, 4J = 1.4 Hz), 7.58 (d, 1H, 4J = 1.4
Hz), 7.53 (d, 1H, 3J = 7.4 Hz), 7.50 (d, 1H, 3J = 8.0 Hz), 7.46
(dd, 1H, 3J = 6.0 Hz, 4J = 1.9 Hz), 7.14 (t, 1H, 3J = 7.3 Hz),
6.99 (t, 1H, 3J = 7.0 Hz), 2.58 (s, 3H, -NCCH3), 1.44 (s, 9H),
1.42 (s, 9H), 0.87 (m, 1H, Ir-CH2CH3), 0.58 (m, 1H,
Experimental Section
3
Ir-CH2CH3), 0.09 (t, 3H, -CH2CH3, J = 7.9 Hz). 13C{1H}
General Considerations. Unless otherwise noted, all rea-
ctions were performed using standard Schlenk techniques
(argon) or in a MBraun glovebox (nitrogen). GC-MS analyses
were performed on a Shimadzu GC-MS QP5000 instrument
(version 2) equipped with a cross-linked methyl silicone
NMR (CDCl3, 100 MHz): δ 168.4, 162.3, 162.2 (O2CCF3,
2JC-F = 35.4 Hz), 161.5, 158.0, 156.6, 152.8, 151.0, 146.0,
134.0, 130.8, 124.7, 123.8, 121.1, 118.7, 116.4, 116.31 (COCF3,
1JC-F = 294 Hz), 114.7, 113.9, 35.51, 35.49, 31.07, 30.73, 16.6,
4.62, -15.42. 19F NMR (CD2Cl2, 376 MHz): δ -74.79 (s, 3F).
High-resolution ESI for C30H35N3O2F3Ir: calcd mass [M -
TFA]þ m/z 606.2460; found m/z 606.2400.
1
gum capillary column (DB5). H and 13C NMR spectra were
collected on a Varian 400 Mercury Plus spectrometer and
referenced to residual protiated solvent. Fluorine reso-
nances were referenced to CFCl3 or hexafluorobenzene. All
coupling constants are reported in hertz (Hz). Mass spectro-
metry analyses were performed at the UC Riverside mass
spectrometry lab. Elemental analyses were performed by Desert
Analytical Laboratory, Inc., Tucson, AZ. X-ray crystallo-
graphic data were collected on a Bruker SMART APEX CCD
diffractometer.
Materials. Trifluoroacetic acid (99%, Aldrich), cesium hy-
droxide (99%, Aldrich), sodium methoxide (95%, Aldrich), and
Drisolv pyridine (99.8%, EMD) were all used as received. All
solvents were reagent grade or better. Tetrahydrofuran and
Preparation of (NNCt-Bu)Ir(TFA)2NCCH3 (3). Under an inert
atmosphere, complex 2 (1.644 g, 2.289 mmol) was stirred over-
night with trifluoroacetic acid (60 mL) in a Schlenk flask. The
solvent was removed under reduced pressure, and the resulting
brown, oily residue was redissolved in CH2Cl2. Compound 3
was purified using column chromatography (basic alumina),
and it was eluted as a yellow band with 95% CH2Cl2/5% ethyl
acetate in a 60.2% (1.103 g) yield. 1H NMR (CD2Cl2, 400 MHz):
δ 9.32 (d, 1H, 3J = 5.8 Hz), 8.08 (d, 1H, 4J = 1.8 Hz), 7.88 (d,
1H, 4J = 1.5 Hz), 7.85 (d, 1H, 4J = 1.5 Hz), 7.72 (m, 3H), 7.32
(dt, 1H, 3J = 7.7 Hz, 4J = 1.4 Hz), 7.17 (dt, 1H, 3J = 7.6 Hz,

Article
Organometallics, Vol. 28, No. 18, 2009 5303
4J = 1.1 Hz), 2.81 (s, 3H), 1.55 (s, 9H), 1.51 (s, 9H). 13C{1H}
NMR (CD2Cl2, 100 MHz): δ 168.5, 165.1, 164.6, 164.1
(CO(CF3), JC-F = 35.7 Hz), 158.1, 157.6, 153.6, 147.3,
chromatography (Chromatotron) using a neutral alumina
Chromatotron plate. Complex 6 was eluted as a red band by a
slow gradient elution starting with 2% methanol/98% ethyl
acetate and increasing the eluent mixture up to 20% methanol/
80% ethyl acetate. Complex 5 was obtained as a blackish green
2
141.3, 135.8, 130.9, 125.2, 124.1, 123.9, 119.8, 118.8, 115.9,
115.6, 112.6 (CO(CF3), JC-F = 290.7 Hz), 36.0 (-C(CH3)3),
35.9 (-C(CH3)3), 31.1 (-C(CH3)3), 30.8 (-C(CH3)3), 4.5
(Ir-NCCH3). 19F NMR (CD2Cl2, 376 MHz): δ -75.09 (s,
3F). Anal. Calcd for C30H30N3O4F6Ir: C, 44.88; H, 3.77; N,
5.23; F, 14.20. Found: C, 44.96; H, 3.79; N, 5.16; F, 14.63. High-
1
solid in 21.4% (70.2 mg) yield. H NMR (CDCl3, 400 MHz):
δ 9.57 (d, 2H, 3J = 5.2 Hz), 8.28 (d, 1H, 3J = 5.8 Hz), 7.95 (d,
1H, 3J = 1.7 Hz), 7.88 (dt, 1H, 3J = 7.6 Hz, 4J = 1.8 Hz), 7.72
(d, 1H, 4J = 1.7 Hz), 7.71 (d, 1H, 4J = 1.5 Hz), 7.57 (dd, 1H, 3J
= 7.6 Hz, 4J = 1.4 Hz), 7.49 (dt, 2H, 3J = 6.9 Hz, 4J = 1.5 Hz),
7.42 (dd, 1H, 3J = 5.8 Hz, 4J = 1.9 Hz), 7.01 (dt, 1H, 3J = 7.2
Hz, 4J = 1.5 Hz), 6.96 (dt, 1H, 3J = 7.3 Hz, 4J = 1.6 Hz), 6.85
(dd, 1H, 3J = 7.0 Hz, 4J = 1.1 Hz), 1.47 (s, 9H), 1.36 (s, 9H),
-2.71 (bs, 2H, -OH). 13C{1H} NMR (CDCl3, 100 MHz): δ
168.7, 162.4, 161.6, 158.4, 157.1, 151.3, 150.3, 148.9, 147.9,
136.0, 133.3, 130.5, 125.2, 124.8, 124.00, 121.9, 119.4, 114.7,
114.1, 35.3 (-C(CH3)3), 35.2 (-C(CH3)3), 31.3 (-C(CH3)3),
30.8 (-C(CH3)3). Anal. Calcd for C29H34N3O2Ir: C, 53.68; H,
5.28; N, 6.48. Found: C, 53.12; H, 5.19; N, 6.32. High-resolution
FABþ (191Ir) for C29H34N3O2Ir: calcd. mass [M]þ m/z 647.2257,
found m/z 647.2265.
resolution FABþ 191Ir) for C30H30N3O4F6Ir: calcd mass m/z
(
[M]þ 801.1746, found m/z 801.1717.
Preparation of (NNCt-Bu)Ir(TFA)2Py (4). Complex 3 (1.103 g,
1.374 mmol) was dissolved in degassed pyridine (40 mL). The
solution was heated overnight in a Schlenk bomb at 100 °C. The
pyridine was removed under reduced pressure to give an orange
solid, which was redissolved in CH2Cl2. Complex 4 was purified
by column chromatography (basic alumina), and it was eluted
from the column as an orange band with an eluent mixture of
1
25% pentane and 75% CH2Cl2 in a 94.2% (1.088 g) yield. H
NMR (CD2Cl2, 400 MHz): δ 9.09 (dd, 2H, 3J = 5.3 Hz, 4J = 1.5
Hz), 8.53 (d, 1H, 3J = 5.5 Hz), 8.20 (d, 1H, 4J = 1.8 Hz), 8.08
(dt, 1H, 3J = 7.6 Hz, 4J = 1.6 Hz), 7.96 (d, 2H), 7.79 (dd, 1H,
3J = 7.8 Hz, 4J = 1.6 Hz), 7.66 (m, 3H), 7.22 (dt, 1H, 3J = 7.6
Hz, 4J = 1.8 Hz), 7.15 (dt, 1H, 3J = 7.6 Hz, 4J = 1.5 Hz), 6.96
(dd, 1H, 3J = 7.6 Hz, 4J = 1.2 Hz), 1.62 (s, 9H, t-Bu), 1.53 (s,
9H, t-Bu). 13C{1H} NMR (CD2Cl2, 100 MHz): δ 168.6, 164.5,
164.4, 163.6 (O2C(CF3), 2JC-F = 36.6 Hz), 160.0, 158.2, 155.0,
152.1, 150.1, 149.0, 141.5, 138.3, 133.6, 130.3, 126.3, 124.9,
Preparation of [(NNCt-Bu)Ir(Ph)(μ-OH)]2 (9). In a Schlenk
bomb, degassed THF (30 mL) was added to a mixture of 10
(150.3 mg, 0.116 mmol) and cesium hydroxide (80.5 mg, 0.479
mmol). The bomb was sealed under argon and heated at 70 °C
for 8 h. The solution was filtered over Celite to remove the
cesium chloride and excess cesium hydroxide. The Celite was
washed with CH2Cl2 to dissolve any precipitated 9. The filtrate
was then evaporated to dryness. Complex 9 was obtained in
38% yield (55.2 mg, 0.0436 mmol) as a black solid by recrys-
tallization from CH2Cl2 and pentane at -30 °C. 1H NMR
124.0, 123.7, 119.7, 115.5, 115.1, 112.7 (O2C(CF3), JC-F
=
291.3 Hz), 35.93 (-C(CH3)3), 35.90 (-C(CH3)3), 31.3
(-C(CH3)3), 30.8 (-C(CH3)3). 19F NMR (CD2Cl2, 376 MHz):
δ -75.63 (s, 3F). Anal. Calcd for C33H32N3O4F6Ir: C, 47.14; H,
3.84; N, 5.00; F, 13.56. Found: C, 46.73; H, 3.71; N, 4.83; F,
13.53. High-resolution FABþ (191Ir) for C33H32N3O4F6Ir: calcd
mass [M]þ m/z 839.1903, found m/z 839.1878.
3
(CDCl3, 400 MHz): δ 8.27 (d, 1H, J = 5.8 Hz), 7.88 (d, 1H,
4J = 1.5 Hz), 7.73 (d, 1H, 4J = 1.3 Hz), 7.69 (d, 1H, 3J = 7.7
Hz), 7.65 (d, 1H, 3J = 1.5 Hz), 7.26 (dd, 1H, 3J = 5.6 Hz, 4J =
1.8 Hz), 7.03 (dt, 1H, 3J = 7.3 Hz, 4J = 1.5 Hz), 6.75 (dt, 1H,
3J = 6.7 Hz), 6.73 (t, 1H, 3J = 7.3 Hz), 6.35 (m, 3H), 6.28 (dd,
1H, 3J = 3.7 Hz, 4J = 2.0 Hz), 1.52 (s, 9H), 1.46 (s, 9H). 13C{1H}
NMR (CDCl3, 100 MHz): δ 169.8, 161.1, 158.4, 158.1, 157.4,
156.8, 151.5, 147.7, 135.9, 133.7, 130.0, 129.8, 124.9, 124.4,
123.8, 120.2, 117.9, 114.0, 113.9, 35.3 (-C(CH3)3), 35.2
(-C(CH3)3), 31.2 (-C(CH3)3), 30.8 (-C(CH3)3), Anal. Calcd
for C60H66N4O2Ir2: C, 57.21; H, 5.28; N, 4.45. Found: C, 55.64;
Preparation of (NNCt-Bu)Ir(OMe)2Py (5). Under an inert
atmosphere, 4 (313.3 mg, 0.3726 mmol) and sodium methoxide
(330.7 mg, 6.122 mmol) were dissolved in methanol (30 mL) in a
Schlenk bomb. The bomb was sealed and heated at 70 °C for 24
h. The solvent was removed under reduced pressure, and the
resulting solid was dissolved in CH2Cl2. The solution was
filtered over Celite to remove the excess sodium methoxide.
The process of filtering out the excess sodium methoxide was
repeated until the solution could easily be filtered through a
pipet plug of Celite. Complex 5 was purified by centrifugal thin-
layer chromatography (Chromatotron) using a neutral alumina
Chromatotron plate. The product was eluted as a blackish green
band with 4% methanol/96% ethyl acetate as the eluent to yield
H, 5.37; N, 4.41. High-resolution FABþ 191Ir) for C60H66-
(
N4O2Ir2: calcd mass [M þ H]þ m/z 1261.4517, found m/z
1261.4537.
Preparation of (NNCt-Bu)Ir(Ph)(TFA)Py (11). Under an inert
atmosphere, CH2Cl2 (30 mL) was added to a Schlenk bomb
containing 8 (213.0 mg, 0.2928 mmol) and silver trifluoroacetate
(101.8 mg, 0.4610 mmol). The reaction mixture was protected
from light with aluminum foil and stirred for 1 week. The solvent
was removed under reduced pressure, and the mixture was
redissolved in CH2Cl2. The mixture was filtered over Celite to
remove the silver chloride. The filtrate was evaporated under
reduced pressure to give 11 (200.2 mg, 0.2488 mmol) as an
1
a black solid in 35.2% (88.2 mg) yield. H NMR (CDCl3, 400
MHz): δ 9.52 (dd, 2H, 3J = 7.0 Hz, 4J = 1.4 Hz), 8.27 (d, 1H,
3J = 5.8 Hz), 8.01 (d, 1H, 4J = 1.8 Hz), 7.96 (dt, 1H, 3J = 7.8
Hz, 4J = 1.4 Hz), 7.80 (dd, 2H, 3J = 6.4 Hz, 4J = 1.8 Hz), 7.63
(m, 1H), 7.58 (dt, 2H, 3J = 6.2 Hz, 4J = 1.4 Hz), 7.49 (dd, 1H,
3J = 6.0 Hz, 4J = 1.8 Hz), 7.05 (m, 3H), 2.52 (s, 6H, Ir-OCH3),
1.55 (s, 9H), 1.44 (s, 9H). 13C{1H} NMR (CDCl3, 100 MHz): δ
168.5, 162.6, 162.2, 158.1, 156.2, 151.4, 148.5, 146.5, 136.4,
132.4, 130.3, 125.7, 125.2, 125.0, 123.9, 121.6, 119.5, 114.5,
58.9 (-OMe), 35.3 (-C(CH3)3), 35.2 (-C(CH3)3), 31.1
(-C(CH3)3), 30.6 (-C(CH3)3). Anal. Calcd for C31H38N3O2Ir:
C, 55.01; H, 5.66; N, 6.21. Found: C, 55.31; H, 6.01; N, 5.93.
1
orange solid in 85.2% yield. H NMR (CD2Cl2, 400 MHz): δ
8.99 (d, 1H, 3J = 5.6 Hz), 8.06 (m, 4H), 7.87 (d, 1H, 4J = 1.8
Hz), 7.83 (dd, 1H, 3J = 7.4 Hz, 4J = 1.1 Hz), 7.79 (dd, 1H, 3J =
4
5.8 Hz, 4J = 1.6 Hz), 7.71 (d, 1H, J = 1.6 Hz), 7.66 (dt, 1H,
3J = 7.3 Hz, 4J = 1.4 Hz), 7.36 (dt, 1H, 3J = 7.5 Hz, 4J = 1.4
Hz), 7.28 (dt, 1H, 3J = 7.0 Hz, 4J = 1.4 Hz), 7.15 (t, 2H, 3J = 6.0
Hz), 6.93 (t, 2H, 3J = 7.0 Hz), 6.89 (t, 2H, 3J = 7.0 Hz), 6.84 (dt,
1H, 3J = 7.0 Hz, 4J = 1.6 Hz), 1.50 (s, 9H), 1.38 (s, 9H). 13C{1H}
NMR (CD2Cl2, 100 MHz): δ 167.0, 164.5, 164.3 (-O2CCF3,
2JC-F = 35 Hz), 163.4, 156.4, 155.7, 149.0, 148.1, 147.8, 144.4,
High-resolution FABþ 191Ir) for C31H38N3O2Ir: calcd mass
(
[M]þ m/z 675.2570, found m/z 675.2594.
Preparation of (NNCt-Bu)Ir(OH)2Py (6). Under an inert atmo-
sphere, 5 (342.2 mg, 0.5055 mmol) was dissolved in a 1:1 (24 mL
total volume) mixture of THF and water in a Schlenk bomb. The
bomb was sealed and heated at 85 °C for 2 days. Pyridine
(15 mL) was added to the reaction mixture to break up the
viscosity of the solvent, and the solvent was removed under
reduced pressure at 60 °C. The resulting reddish black solid was
then dissolved in CH2Cl2 and purified by centrifugal thin-layer
137.9, 132.2, 131.6, 130.5, 127.2, 127.0, 126.6, 125.5, 123.0,
=
1
121.6, 120.3, 117.3, 116.6, 115.3, 114.5 (-O2CCF3, JC-F
287 Hz) 35.6 (-C(CH3)3), 35.5 (-C(CH3)3), 30.4 (-C(CH3)3),
30.3 (-C(CH3)3). 19F NMR (CD2Cl2, 376 MHz): δ -75.10 (s,
3F). High-resolution ESI/APCI MS for C37H37N3O2F3Ir: calcd
mass [M - Ph]þ m/z 728.2076, found m/z 728.2078.

5304 Organometallics, Vol. 28, No. 18, 2009
Meier et al.
Stability Tests of 6 in D2O. In three separate J. Young NMR
tubes, a 5.0 mM solution of 6 in D2O was added. To one NMR
tube was added 2 equiv of pyridine, and to another NMR tube
was added 4 equiv of KOD. The tubes were all heated simulta-
neously and then analyzed by 1H NMR. The 1H NMR study for
6 in neat D2O and in the presence of 4 equiv of KOD can be
found in the Supporting Information.
General Procedure for H/D Exchange Studies of Water/Ben-
zene by 6. In a typical reaction, 0.50 mL of a 3.54 mM solution of
6 in D2O, 14 equiv of KOD, and 0.50 mL of benzene-H6 were
loaded into a resealable Schlenk tube under argon. The tube was
then heated anywhere from 2 to 24 h, depending on the
temperature used. After the desired heating, the Schlenk tube
was removed from the oil bath, and the benzene layer was
analyzed by GC-MS to determine the amount of deuterium
incorporation into benzene. To determine if 6 was stable under
catalytic conditions, the Schlenk tube was opened under argon
and a 0.6 μL aliquot of the benzene layer was analyzed by GC-
MS. The Schlenk tube was then resealed and heated for a longer
time. This process was repeated several times to ensure that a
plot of TON vs time was linear (see the Supporting In-
formation).
analyzed as described above. Results of the control (back-
ground) and the catalytic reaction can be seen in Table 1.
Catalyst Phase Removal. Under an inert atmosphere, 9
(8.9 mg, 7.0 mmol), KOD (0.10 mmol, 14 equiv), benzene-H6
(2.0 mL), and D2O (2.0 mL) were added in a Schlenk bomb. The
bomb was sealed and heated in an oil bath at 190 °C for 17 h, after
which the reaction was cooled to room temperature. The benzene
phase was completely removed and analyzed by GC-MS, resulting
in a TOF of 7.5ꢀ10-3 s-1 (23.2% C6H5D and 2.8% C6H4D2)
with the results corrected for background H/D exchange. Degassed
benzene-H6 containing no catalyst was added to the remaining
aqueous solution in the Schlenk bomb. The bomb was sealed and
heated at 190 °C for another 17 h. Analysis of the benzene phase
showed little to no H/D exchange (less than 0.5% benzene-H6
converted, which is within analytical error of the GC-MS).
Mass Transfer Study. Using the procedure described in Gen-
eral Procedure for H/D Exchange Studies of Water/Benzene by
6, separate Schlenk tubes were analyzed under different stirring
rates at the highest temperature that would be used in these
studies, 190 °C. The turnover frequencies observed at the
various stirring rates were within error ((1.0 ꢀ 10-3 s-1) of
what might be expected at 190 °C.
Analysis of H/D Exchange. Catalytic H/D exchange reactions
were quantified by monitoring the increase of deuterium into
C6H6 by GC-MS analyses. This was achieved by deconvoluting
the mass fragmentation pattern obtained from the MS analysis,
using a program developed with Microsoft EXCEL.23 An
important assumption made with this method is that there are
no isotope effects on the fragmentation pattern for the various
benzene isotopologues. Fortunately, because the parent ion of
benzene is relatively stable toward fragmentation, it can be used
reliably to quantify the exchange reactions. The mass range
from m/z 78 to 84 (for benzene) was examined for each reaction
and compared to a control reaction where no metal catalyst was
added. The program was calibrated with known mixtures of
benzene isotopologues. The results obtained by this method are
reliable to within 5%. Thus, analysis of a mixture of C6H6,
C6D6, and C6H5D prepared in a molar ratio of 50.6:25.3:24.1
resulted in an experimentally determined ratio of 49.4
(C6H6):23.3 (C6D6):27.3 (C6H5D1). Catalytic H/D exchange
reactions were thus run for sufficient reaction times to be able
to detect changes to >5% exchange. H/D exchange reactions of
benzene were carried out such that less than 10% deuterium
incorporation had occurred. This allowed for the kinetics to
remain pseudo first order. Keeping the deuterium incorporation
to a small amount also prevented the statistical chance that
degenerate deuterium/deuterium exchange would occur.
Background Analysis. To ensure that catalysis was occurring,
control reactions were conducted to account for the amount of
H/D exchange observed from a reaction mixture containing
D2O, KOD, and benzene-H6. The reaction conditions were
conducted similarly to those described in General Procedure
for H/D Exchange Studies of Water/Benzene by 6. However, the
solution contained no added catalyst. These solutions were
Preparative-Scale Attempt To Produce 9 from 6. Under argon,
6 mL of a 2.12 mM stock solution of 6 (0.0127 mmol) in H2O and
6 mL of benzene-H6 were added to a resealable Schlenk tube.
Aqueous sodium hydroxide was added to make the aqueous
phase a 0.050 M NaOH (aqueous) solution. The Schlenk bomb
was sealed and heated at 160 °C for 8 h. The reaction mixture was
cooled to room temperature, and 3 mL of pyridine was added to
the mixture. The tube was sealed and heated at 50 °C for 1 h prior
to removal of the solvent from the mixture under reduced
pressure with gentle heating. The mixture was redissolved in
CD2Cl2 and checked to ensure that there was no observable
benzene-H6 by 1H NMR. Trifluoromethanesulfonic acid (5 μL)
was added to protonate any iridium phenyl complexes present in
the solution. As an internal standard, 1 μL of mesitylene (0.864
mg, 0.00718 mmol) was added, producing an integration of the
mesityl-H protons of 3.00. Normalization of the 3 mesitylene
protons/mol of mesitylene relative to the 6 benzene-H6 protons/
mol of benzene (integration of 4.24) results in 0.005 07 mmol of
benzene-H6 produced. This resulted in 39.9% of benzene-H6
relative to the amount of iridium starting material.
Acknowledgment. We thank the National Science
Foundation (Grant No. CHE-0328121), Chevron Texaco
Energy Technology Co., and Scripps Florida for financial
support of this research. We thank Mr. Timothy Stewart
and Prof. Robert Bau for X-ray structure determination.
Supporting Information Available: A CIF file and tables and
figures giving X-ray structural data for 9, 1H NMR spectra for
the stability studies of 6 in D2O with and without added KOD, a
plot of turnover number vs time for H/D exchange using 6 as a
catalyst, a 1/[Py] plot for H/D exchange with 6, 1H NMR spectra
for compounds 2-6 and 9-11, and B3LYP xyz coordinates of
transition states. This material is available free of charge via the
Internet at http://pubs.acs.org.
(23) Young, K. J. H.; Meier, S. K.; Gonzales, J. M.; Oxgaard, J.;
Goddard, W. A., III; Periana, R. A. Organometallics 2006, 25, 4734
(see the Supporting Information in this paper for an Excel spreadsheet).