Stoichiometric and catalytic oxygen activation by trimesityliridium(III)

Article abstract of DOI:10.1021/ic025700e

Trimesityliridium(III) (mesityl = 2,4,6-trimethylphenyl) reacts with O₂ to form oxotrimesityliridium(V), (mes)₃Ir=O, in a reaction that is cleanly second order in iridium. In contrast to initial reports by Wilkinson, there is no evidence for substantial accumulation of an intermediate in this reaction. The oxo complex (mes)₃Ir=O oxidizes triphenylphosphine to triphenylphosphine oxide in a second-order reaction with ΔH? = 10.04 ± 0.16 kcal/mol and ΔS? = -21.6 ± 0.5 cal/(mol·K) in 1,2-dichloroethane. Triphenylarsine is also oxidized, though over an order of magnitude more slowly. Ir(mes)₃ binds PPh₃ reversibly (K_assoc = 84 ± 3 M^-1 in toluene at 20°C) to form an unsymmetrical, sawhorse-shaped four-coordinate complex, whose temperature-dependent NMR spectra reveal a variety of dynamic processes. Oxygen atom transfer from (mes)₃Ir=O and dioxygen activation by (mes)₃Ir can be combined to allow catalytic aerobic oxidations of triphenylphosphine at room temperature and atmospheric pressure with overall activity (～60 turnovers/h) comparable to the fastest reported catalysts. A kinetic model that uses the rates measured for dioxygen activation, atom transfer, and phosphine binding describes the observed catalytic behavior well. Oxotrimesityliridium does not react with sulfides, sulfoxides, alcohols, or alkenes, apparently for kinetic reasons.

Full text of DOI:10.1021/ic025700e

Inorg. Chem. 2002, 41, 4815−4823
Stoichiometric and Catalytic Oxygen Activation by Trimesityliridium(III)
Bridey Grant Jacobi, David S. Laitar, Lihung Pu, Michael F. Wargocki, Antonio G. DiPasquale,
Kevin C. Fortner, Stephany M. Schuck, and Seth N. Brown*
Department of Chemistry and Biochemistry, 251 Nieuwland Science Hall,
UniVersity of Notre Dame, Notre Dame, Indiana 46556-5670
Received May 7, 2002
Trimesityliridium(III) (mesityl ) 2,4,6-trimethylphenyl) reacts with O to form oxotrimesityliridium(V), (mes)3IrdO, in
2
a reaction that is cleanly second order in iridium. In contrast to initial reports by Wilkinson, there is no evidence for
substantial accumulation of an intermediate in this reaction. The oxo complex (mes)3IrdO oxidizes triphenylphosphine
q
q
to triphenylphosphine oxide in a second-order reaction with ∆H ) 10.04 ± 0.16 kcal/mol and ∆S ) −21.6 ± 0.5
cal/(mol‚K) in 1,2-dichloroethane. Triphenylarsine is also oxidized, though over an order of magnitude more slowly.
Ir(mes)3 binds PPh3 reversibly (Kassoc ) 84 ± 3 M-1 in toluene at 20 °C) to form an unsymmetrical, sawhorse-
shaped four-coordinate complex, whose temperature-dependent NMR spectra reveal a variety of dynamic processes.
Oxygen atom transfer from (mes)3IrdO and dioxygen activation by (mes)3Ir can be combined to allow catalytic
aerobic oxidations of triphenylphosphine at room temperature and atmospheric pressure with overall activity (∼60
turnovers/h) comparable to the fastest reported catalysts. A kinetic model that uses the rates measured for dioxygen
activation, atom transfer, and phosphine binding describes the observed catalytic behavior well. Oxotrimesityliridium
does not react with sulfides, sulfoxides, alcohols, or alkenes, apparently for kinetic reasons.
Scheme 1. Oxygen Atom Transfer-Based Strategy for Catalyzed Air
Oxidations
Introduction
Molecular oxygen is an extremely desirable oxidant
because of both its low monetary cost and its low environ-
mental impact. While dioxygen is a thermodynamically
potent oxidant, its reactions have an unfortunate combination
of being both sluggish and unselective. Catalysis of the
reactions of O2, usually with transition metals, has been
widely used to overcome these problems.1
One particularly simple and attractive mode of catalysis
that can be envisioned involves formation of a terminal oxo
moiety by reaction of O2 with a reduced metal center,
followed by oxygen atom transfer2 from the oxometal
complex to a substrate (Scheme 1). While there are numerous
well-characterized examples of both the former and the latter
reactions individually, the number of cases of well-defined
systems that can combine both reactions to achieve catalytic
reactions at reasonable rates is still rather small. Both
thermodynamic and kinetic considerations suggest that this
combination will be challenging to achieve. Thermodynami-
cally, to allow both legs of the cycle to proceed, the strength
of the metal-oxygen multiple bond is constrained to be
strong enough to be formed exothermically from O2, but not
so strong that substrates cannot be oxidized. Kinetically,
dioxygen is an electron-poor substance that reacts most
favorably with electron-rich metal centers.3 However, the
reactivity of oxometal complexes in oxygen atom transfer
reactions (with typical electron-rich substrates) is strongly
* Author to whom correspondence should be addressed. E-mail:
Seth.N.Brown.114@nd.edu.
(1) Barton, D. H. R.; Martell, A. E.; Sawyer, D. T., Eds. The ActiVation
of Dioxygen and Homogeneous Catalytic Oxidation; Plenum Press:
New York, 1993.
(3) (a) Jones, R. D.; Summerville, D. A.; Basolo, F. Chem. ReV. 1979,
79, 139-179. (b) Niederhoffer, E. C.; Timmons, J. H.; Martell, A. E.
Chem. ReV. 1984, 84, 137-203.
(2) Holm, R. H. Chem. ReV. 1987, 87, 1401-1449.
10.1021/ic025700e CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/16/2002
Inorganic Chemistry, Vol. 41, No. 18, 2002 4815

Jacobi et al.
by following the absorbance at 620 or 630 nm (λmax for (mes)3-
IrdO is 620 nm, ꢀ ) 2000 M-1 cm-1).
dependent on the electrophilic character of the oxo com-
plex.4,5 Thus a catalyst acting by the mechanism of Scheme
1 must strike a balance between nucleophilic character in
its reduced form and electrophilic character in its oxidized
form and is unlikely to excel at both oxidative and reductive
steps.
Reactions involving deoxygenation of (mes)3IrdO were initiated
by injecting a small volume of a deoxygenated solution of the oxo
complex into 2 mL of a solution of known concentration of the
reducing agent (triarylphosphine or -arsine). Final concentrations
of (mes)3IrdO were between 2 × 10-4 and 5 × 10-4 M. These
reactions were performed under pseudo-first-order conditions, with
reductant in at least 10-fold molar excess over iridium. Rate
constants were obtained by least-squares fits to the equation ln|Af
- A| ) -kt + ln|Af - A0|, which were linear for at least four
half-lives. Activation parameters were determined from plots of
ln(k/T) vs (1/T).9
We were therefore struck by the report by Hay-Motherwell
and Wilkinson that the homoleptic trimesityliridium(III)6
reacts with O2 to form oxotrimesityliridium(V), a unique
example of a terminal oxo complex of iridium.7,8 We
reasoned that this high-valent iridium complex would have
a high propensity for reduction to iridium(III), and thus a
relatively weak metal-oxygen bond. We therefore hoped it
would readily oxidize a variety of substrates by oxygen atom
transfer, allowing it to act as a catalyst for air oxidations.
Here we report the scope and mechanism of the stoichio-
metric and catalytic oxidation reactions of (mes)3IrdO. We
also describe the complex fluxional behavior of sawhorse-
shaped four-coordinate (mes)3Ir(PPh3), which is present
during oxidation reactions of PPh3 catalyzed by (mes)3Ird
O.
For the oxygenation experiments, solutions of trimesityliridium-
(III) were generated by treating (mes)3IrdO in the drybox with a
slightly substoichiometric amount of triphenylphosphine (usually
0.9 equiv) and allowing the deoxygenation to go to completion (at
least 2 h at room temperature). NMR analysis of such solutions
shows that the only phenyl resonances are those due to free
triphenylphosphine oxide, indicating that this procedure generates
free (mes)3Ir. Oxygenation was initiated by injecting a small volume
of this stock solution into septum-capped cuvettes containing 2 mL
of dichloromethane that had been saturated with either air or pure
dioxygen by bubbling of the gas for at least 10 min. Given the
reported solubilities of dioxygen in chlorinated solvents (8-10 mM
at 1 atm of O2)10 and [Ir] < 5 × 10-4 M, these conditions ensure
that the dissolved dioxygen concentration should change <15%
during the reaction and so mass transport of O2 into the liquid should
not affect the measured rates. Second-order rate constants were
determined by nonlinear least-squares fitting, using the program
Kaleidagraph, of the absorbance data to the equation A ) Af + (A0
- Af)/(1 + (A0 - Af)(kt/∆ꢀ630)) (∆ꢀ630 ) -1700 M-1 cm-1). Rate
constants were found to be invariant with iridium concentration
over a range of 1.2 × 10-4 to 5.0 × 10-4 M.
Experimental Section
Materials and Methods. Unless otherwise noted, all procedures
were carried out on the benchtop. Dichloromethane and dichloro-
ethane used in the kinetics studies were dried over 4 Å molecular
sieves, followed by CaH2, and stored in an inert-atmosphere
glovebox before use. All other reagents were commercially available
and used without further purification. NMR spectra were measured
on a General Electric GN-300 or a Varian VXR-300 or -500 FT-
NMR spectrometer.
Oxotrimesityliridium(V) was prepared by a variation on the
literature procedure.7 A crude solution of trimesityliridium(III) in
hexane was generated according to the literature procedure.6 This
brown solution was then exposed to the air and stirred vigorously
overnight, turning green. The green solution was gravity-filtered,
stripped to dryness on the rotary evaporator, and purified by
repeated chromatography on silica gel, eluting with 5% Et2O/
hexane. The pure fractions (as judged by TLC) were collected and
the material crystallized from acetone/water or acetonitrile/water,
and the purity of the recrystallized material was confirmed by NMR.
Yields of this modified procedure are typically 5%, which is lower
than that reported by Wilkinson6,7 but similar to the yields obtained
in our hands on repeating the literature procedure.
Generation and Characterization of (mes)3Ir(PPh3). In a
typical experiment, a solution of trimesityliridium(III) was generated
in an NMR tube sealed to a ground glass joint by treating a solution
of (mes)3IrdO in CD2Cl2 in the drybox with a stoichiometric
amount of triphenylphosphine-d15 (Aldrich) and allowing deox-
genation to go to completion. After deoxygenation, 1 equiv of
triphenylphosphine was added to the brown-red solution of tri-
mesityliridium(III). The NMR tube was then affixed to a glass joint
with a Teflon valve, taken out of the drybox, cooled to -78 °C,
and flame sealed under reduced pressure. The tube was stored at
-20 °C when not being analyzed, since the phosphine adduct
decomposes over the course of about a day at room temperature.
1H NMR (CD2Cl2, -98 °C; a refers to the phenyl or mesityl group
in the mirror plane established at higher temperatures; b and b′
refer to the phenyl or mesityl groups related by that mirror plane):
δ 0.552 (s, 3H, o-CH3, mesb); 0.998 (s, 3H, o-CH3, mesa); 1.730
(s, 3H, o-CH3, mesa); 1.828 (s, 3H, o-CH3, mesb′); 1.965 (s, 3H,
p-CH3, mesb); 2.006 (s, 3H, p-CH3, mesb′); 2.040 (s, 3H, p-CH3,
mesa); 2.174 (s, 3H, o-CH3, mesb); 2.390 (s, 3H, o-CH3, mesb′);
4.851 (s, 1H, ArH, mesb); 5.517 (t, 7.5 Hz, 1H, o-Phb); 6.087 (s,
1H, ArH, mesa); 6.216 (s, 1H, ArH, mesb′); 6.267 (s, 1H, ArH,
mesb); 6.379 (t, 7.5 Hz, 1H, m-Phb); 6.443 (s, 1H, ArH, mesb′);
6.552 (s, 1H, ArH, mesa); 6.955 (t, 7.5 Hz, 1H, p-Phb); 7.017 (t,
7.5 Hz, 1H, m-Phb); 7.156 (br, 2H, m-Phb′); 7.189 (t, 7.5 Hz, 1H,
o-Phb); 7.214 (br, 1H, o-Pha); 7.401 (t, 7.5 Hz, 1H, p-Phb′); ∼7.5
Kinetics. UV-visible data were collected on a Beckman DU-
7500 diode-array spectrophotometer equipped with a multicell
transport block. The temperature was regulated by a circulating
water/ethylene glycol mixture and was measured by a thermocouple
inserted in the cell block. Solutions were prepared in the drybox in
1-cm quartz cells fitted with septum caps. Reactions were monitored
(4) Seymore, S. B.; Brown, S. N. Inorg. Chem. 2000, 39, 325-332.
(5) (a) DuMez, D. D.; Mayer, J. M. Inorg. Chem. 1998, 37, 445-453.
(b) Brown, S. N.; Mayer, J. M. J. Am. Chem. Soc. 1996, 118, 12119-
12133. (c) DuMez, D. D.; Mayer, J. M. Inorg. Chem. 1995, 34, 6396-
6401.
(6) Hay-Motherwell, R. S.; Wilkinson, G.; Hussain-Bates, B.; Hursthouse,
M. B. J. Chem. Soc., Dalton Trans. 1992, 3477-3482.
(7) Hay-Motherwell, R. S.; Wilkinson, G.; Hussain-Bates, B.; Hursthouse,
M. B. Polyhedron 1993, 12, 2009-2012.
(8) Terminal imido (NR) complexes of iridium are known: Glueck, D.
S.; Wu, J. X.; Hollander, F. J.; Bergman, R. G. J. Am. Chem. Soc.
1991, 113, 2041-2054.
(9) Frost, A. A.; Pearson, R. G. Kinetics and Mechanism, 2nd ed.;
Wiley: New York, 1961.
(10) Wilhelm, E.; Battino, R. Chem. ReV. 1973, 73, 1-9.
4816 Inorganic Chemistry, Vol. 41, No. 18, 2002

Catalytic Oxygen ActiWation by Trimesityliridium(III)
Table 1. Rate of Oxygen Atom Transfer from (mes)3IrdO to
Phosphines and Arsines (CH2Cl2, 25 °C)
(v br, 2H, m-Pha); 7.504 (t, 7.5 Hz, 1H, p-Pha); 7.604 (br, 2H,
o-Phb′); 8.620 (br, 1H, o-Pha).
atom transfer (M-1 s-1
)
Catalytic Air Oxidation of PPh3. In a typical run, triph-
enylphosphine (130.1 mg, 0.496 mmol) in 2.5 mL of toluene was
added to a dark blue-green solution of (mes)3IrdO (5.5 mg, 0.0096
mmol, 2 mol %) in 2.5 mL of toluene in a 15-mL two-necked
round-bottom flask. Oxygen gas, which was presaturated with
toluene by passing it through a gas dispersion tube in a 100-mL
two-necked flask filled with toluene, was admitted to the reaction
vessel through a needle. Oxygen was bubbled through the reaction
mixture at a moderate rate and vented through a water bubbler.
After addition of the triphenylphosphine, 0.2 mL of the reaction
mixture was syringed out and added to a mixture of CDCl3 (0.3
mL) mixed with 50 µL of concentrated aqueous HCl (to decompose
the trimesityliridium) in an NMR tube. Aliquots were taken and
quenched analogously throughout the course of the reaction, and
each aliquot was then analyzed by 31P{1H} NMR. Kinetic simula-
tions were performed by numerical integration (Runge-Kutta) of
the coupled differential equations with Microsoft Excel.
substrate
PPh3
P(p-C6H4Cl)3
P(p-C6H4CH3)3
P(p-C6H4OCH3)3
k
3.95 (5)
3.76 (6)
6.33 (12)
10.88 (14)
0.058 (2)
0.111 (17)
a
P(o-C6H4CH3)3
AsPh3
a Measured at 21.9 °C.
listed in Table 1. Electron-donating substituents on the
triarylphosphine increase the rate of oxidation, but only
slightly. The modest Hammett F value of -0.29 ( 0.10
(Figure S2, Supporting Information) has the same sign but
is smaller in magnitude than is typically observed for oxygen
atom transfer to triarylphosphines by metal-oxo complexes
(compare, for example, atom transfer from TpReOCl2, F )
-0.474). Bulky triarylphosphines such as tri-ortho-tolylphos-
phine react much more slowly than their unhindered coun-
terparts. Triphenylarsine reacts substantially more slowly than
triphenylphosphine, as is expected given the smaller driving
force for triphenylarsine oxidation.11 The rate difference of
a factor of 36 is similar to the 61-fold faster oxidation of
PPh3 over AsPh3 by (tetramesitylporphyrinato)dioxoruthenium-
(VI),12 while much smaller differences are seen in peroxo-
metal oxidations.13
Results
Reactivity of Oxotrimesityliridium(V). Oxotrimesityl-
iridium(V) ((mes)3IrdO; mes ) 2,4,6-trimethylphenyl) is
quantitatively reduced by triphenylarsine and by a variety
of triarylphosphines to give homoleptic trimesityliridium-
(III) and the corresponding arsine or phosphine oxide (eq
1). The rates of these reactions may be conveniently
In contrast to its ready oxidation of phosphines and arsines,
(mes)3IrdO is inert to all commonly occurring organic
functional groups. Thus, dimethyl sulfide, dimethyl sulfoxide,
styrene, norbornene, 2,3-dihydrofuran, allyl alcohol, and
benzaldehyde do not react with (mes)3IrdO over a period
of one week at room temperature. In certain cases, the ability
of the reduced (mes)3Ir species to deoxygenate organic
substrates has been tested, and it too is found to be unreactive
to substrates such as styrene oxide and Me2SO. However,
the more strongly oxidizing trimethylamine-N-oxide has been
reported to convert (mes)3Ir to (mes)3IrdO.7
Reaction of Trimesityliridium(III) with Dioxygen. Tri-
mesityliridium is very air-sensitive, being oxidized rapidly
to (mes)3IrdO by O2.7 The kinetics of this oxygen uptake
process may be monitored spectrophotometrically by mea-
suring the growth of the absorbance of the iridium(V)
complex at 630 nm. The kinetics observed in this manner
show a clean second-order dependence on iridium(III)
concentration (Figure 1). Individual kinetic runs fit extremely
well to second-order kinetics, and the rate constants obtained
in this manner are independent of the total iridium concen-
tration (1.2 × 10-4 to 5.0 × 10-4 M). In air, the observed
second-order rate constant is 31 ( 2 M-1 s-1. The reaction
performed under 1 atm of pure O2 is much faster (Figure 1),
with a corresponding rate constant of 240 ( 60 M-1 s-1.
This rate acceleration under 1 atm of O2 of 7.7 ( 2.0 is
reasonably close to the factor of 5 difference in oxygen partial
pressure between air and pure O2, strongly suggesting that
the reaction is first order in dioxygen.
measured by monitoring the decrease in absorbance at 620
nm as the blue-green iridium(V) is converted to yellow
iridium(III). In all cases, isosbestic points are observed and
the pseudo-first-order rate constants measured in the presence
of excess reducing agent are proportional to the concentration
of the phosphine or arsine reagent. In some cases, the
reducing agents form detectable amounts of adducts with
(mes)3Ir (see below). However, since the ligands are present
in substantial excess and binding rates are invariably much
faster than rates of reduction, adduct formation does not
interfere with kinetic measurements. Oxidation of phosphines
and arsines by (mes)3IrdO is irreversible.
These observations are consistent with atom transfer taking
place in a reaction that is first order in both iridium and
substrate and which proceeds without observable intermedi-
ates. An analysis of the temperature dependence of kox for
the deoxygenation of (mes)3IrdO by PPh3 in 1,2-dichloro-
ethane from 15 to 55 °C gives activation parameters for this
reaction of ∆Hq ) 10.04 ( 0.16 kcal/mol and ∆Sq ) -21.6
( 0.5 cal/(mol‚K) (Figure S1, Supporting Information).
Second-order rate constants for the reactions of various
triarylphosphines and of triphenylarsine with (mes)3IrdO,
measured at ambient temperature in dichloromethane, are
(11) Holm, R. H.; Donahue, J. P. Polyhedron 1993, 12, 571-589.
(12) Cheng, S. Y. S.; James, B. R. J. Mol. Catal. A 1997, 117, 91-102.
(13) Abu-Omar, M. M.; Espenson, J. H. J. Am. Chem. Soc. 1995, 117,
272-280.
Inorganic Chemistry, Vol. 41, No. 18, 2002 4817

Jacobi et al.
Figure 2. Spectral evolution of the reaction between (mes)3Ir (initial
concentration 2.5 × 10-4 M) and O2 in air-saturated CH2Cl2. Spectra were
taken at 0, 120, 600, and 2200 s; arrows show the direction of absorbance
change.
Figure 1. Reaction of (mes)3Ir with O2 in air (open circles) or under 1
atm of O2 (open squares). The solid line is the fit to second-order kinetics.
The iridium concentration is 1.2 × 10-4 M. Inset: Plots of 1/(A - Af) for
the same two kinetic runs.
(conditions where formation of a peroxo species is most
1
unlikely) results in H NMR resonances due to free OPPh3
and mesityl resonances at the population-weighted average
positions for iridium(III) and iridium(V) (Figure S3, Sup-
porting Information). Apparently new NMR signals therefore
appear in incompletely oxygenated solutions of (mes)3Ir as
a result of this exhange phenomenon rather than because an
intermediate is present. Details of this unprecedentedly fast
oxygen atom exchange will be discussed elsewhere.15
Binding of Triphenylphosphine to Trimesityliridium-
(III). Solutions of (mes)3Ir in CH2Cl2 containing an excess
of PPh3 at room temperature show a single, slightly
broadened resonance in the 31P{1H} NMR. That the broaden-
ing is due to weak, reversible binding of PPh3 to the iridium
center is made evident upon cooling the solution, whereupon
the 31P resonance broadens further, decoalesces, and splits
into resonances at δ -6.2 ppm for free PPh3 and δ 2.7 ppm
for (mes)3Ir(PPh3) (Figure S4, Supporting Information), with
phosphine exchange becoming slow on the NMR time scale
below about -30 °C. One equivalent of OPPh3 is also
The observed rate law (rate ) k[Ir(III)]2[O2]) is consistent
with a reaction mechanism involving weak preequilibrium
complexation of O2 by one iridium center, followed by rate-
determining reduction of this complex by another molecule
of (mes)3Ir to give 2 equiv of (mes)3IrdO (eq 2). (A direct
termolecular reaction is consistent with the kinetics but is
chemically implausible.) While we have no structural
information about the intermediate, an η2 complex seems
more likely, given the extensive precedent for this mode of
coordination to iridium.14 The mechanism proposed in eq 2
is essentially identical with that suggested by Hay-Mother-
well, Wilkinson, and co-workers.7 The previous study also
reported the observation of new NMR signals on treatment
of (mes)3Ir with O2, which were assigned to a labile
intermediate iridium peroxide. In fact, any intermediates
present in this reaction do not accumulate to a significant
extent, as demonstrated by the scrupulous adherence to
second-order kinetics throughout the course of the reaction
and by the presence of tight isosbestic points at 492, 415,
and 330 nm (Figure 2). The new NMR signals seen at
intermediate stages of the reaction by Wilkinson may be
plausibly explained by the surprising observation that
degenerate intermetal oxygen atom transfer between (mes)3Ir
and (mes)3IrdO is fast on the NMR time scale. For example,
anaerobic titration of (mes)3IrdO with PPh3 in CDCl3
1
observed by 31P and H NMR, as the complex is generated
by addition of excess PPh3 to (mes)3IrdO. The weak binding
of phosphine and the moderate thermal instability of the
complex (it decomposes in solution over the course of about
a day at room temperature) have prevented us from isolating
a pure solid sample of the adduct, and all studies were carried
out on samples generated in situ.
Only 1 equiv of PPh3 binds to (mes)3Ir, even at -98 °C.
Rates of phosphine exchange are independent of [PPh3],
suggesting that phosphine exchange is dissociative, with kdiss
) 1.5 × 104 s-1 at 23 °C. Analysis of the temperature
dependence of the NMR line shape from 235 to 295 K
(Figure S5) gives ∆Hq ) 15.41((0.24) kcal/mol and ∆Sq )
+12.4((0.9) cal/(mol‚K) for phosphine dissociation. From
the amounts of free and bound phosphine that are observed
at higher temperatures (as judged from the position of the
averaged 31P NMR signal), one can estimate thermodynamic
parameters for phosphine binding (Figure S6) as ∆H° )
-15.7((0.9) kcal/mol and ∆S° ) - 44((3) cal/(mol‚K);
(14) (a) Antinolo, A.; Fonseca, I.; Ortiz, A.; Rosales, M. J.; Sanz-Aparicio,
J.; Terrenos, P.; Torrens, H. Polyhedron 1999, 18, 959-968. (b) Perera,
S. D.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1998, 2887-2891.
(c) Barbaro, P.; Bianchini, C.; Laschi, F.; Midollini, S.; Moreti, S.;
Scapacci, G.; Zanello, P. Inorg. Chem. 1994, 33, 1622-1630.
(15) Fortner, K. C.; Laitar, D. S.; Muldoon, J.; Pu, L.; Brown, S. N.
Manuscript in preparation.
4818 Inorganic Chemistry, Vol. 41, No. 18, 2002

Catalytic Oxygen ActiWation by Trimesityliridium(III)
and three pairs of methyl groups (δ 0.56 and 1.83, δ 1.97
and 2.01, and δ 2.17 and 2.39) undergo pairwise exchange.
All five pairwise exchanges occur at the same rate, indicating
that they are due to a common chemical process. Two
aromatic hydrogens (δ 6.09, 6.56) and three methyl groups
(δ 1.00, 1.74, 2.04) are unaffected by the exchange process
and remain sharp until the onset of phosphine dissociation
at ca. -30 °C. This pattern of exchange indicates that some
motion introduces a mirror plane on the NMR time scale,
rendering two mesityl groups equivalent, although top and
bottom remain inequivalent and do not exchange. Line shape
analysis of this “rocking” motion gives activation parameters
of ∆Hq ) 11.19((0.25) kcal/mol and ∆Sq ) +6.0((1.1)
cal/(mol‚K) (Figure S9).
At low temperature, all three phenyl groups of the PPh3
ligand are inequivalent, and further changes in the spectra
can be attributed to slowing of rotation rates about the three
P-C bonds, with one phenyl group showing effectively
stopped rotation at -98 °C, one showing a moderate rotation
rate (k175K ) 170 s-1), and one still rotating rather rapidly
(decoalescence is not observed even at 175 K). Curiously,
it is the unique phenyl group whose environment is not
affected by the rocking motion of the mesityls that shows
the middle rotation rate. It is also apparent that this phenyl
group does not exchange with the other two phenyl groups
up to the onset of PPh3 dissociation (note for example that
the coalesced resonance due to the two ortho protons of this
phenyl group at δ 7.94 ppm are still distinct from the other
four ortho protons at δav ) 6.9 ppm even at -28 °C). Thus,
rotation about the Ir-P bond, which would interconvert all
three phenyl groups, must be slower than phosphine dis-
sociation. The rates and activation parameters of the observed
fluxional processes in (mes)3Ir(PPh3) are summarized in
Table 2.
Triphenylarsine forms a similar unsymmetrical adduct
(mes)3Ir(AsPh3), though it appears to bind more weakly than
does triphenylphosphine. Tri-o-tolylphosphine does not bind
to (mes)3Ir, even at -80 °C, nor do the phosphine oxides
OPPh3 or OP(o-C6H4CH3)3.
Catalysis of Air Oxidations by Oxotrimesityliridium-
(V). Since (mes)3IrdO reacts with triphenylphosphine and
triphenylarsine to form (mes)3Ir, and since (mes)3Ir reacts
with O2 to regenerate (mes)3IrdO, it is clear that (mes)3Ird
O should be a catalyst for the air oxidation of Ph3P or Ph3-
As, albeit with some potential inhibition due to ligand binding
to (mes)3Ir (Scheme 2). Indeed, these substrates are converted
quantitatively to the corresponding oxides in the presence
of air or O2 and catalytic amounts of (mes)3IrdO. Catalysis
proceeds in a variety of solvents, but chlorinated solvents
cause decomposition of the trimesityliridium. In chloroform,
decomposition takes place quite rapidly, effectively halting
catalysis after a few turnovers, but even in CH2Cl2 or 1,2-
dichlorobenzene there is noticeable decomposition after <20
turnovers. In contrast, catalysis in benzene or toluene can
proceed to >50 turnovers without noticeable catalyst de-
composition, as monitored by NMR integration against an
internal standard. No phosphine oxidation is observed in the
absence of catalyst. For phosphine oxidation, the resting state
Figure 3. 1H NMR spectra (500 MHz) (aryl region) of (mes)3Ir(PPh3) in
CD2Cl2 from -28 to -98 °C. Peaks due to excess PPh3 are marked with
an asterisk.
at 295 K, Keq for binding is 97((7) M-1. UV-visible titration
data in toluene at 293 K (Figure S7) give similar values (Keq
) 84 ( 3 M-1).
Although the 31P{1H} NMR spectrum of (mes)3Ir(PPh3)
1
is invariant below about -30 °C, the H NMR undergoes
substantial changes as the temperature is lowered to -98
°C (aryl region, Figure 3; methyl region, Figure S8). In the
low-temperature spectra, the presence of six inequivalent
aromatic protons and nine inequivalent methyl groups
indicates that all mesityl groups are inequivalent and that
(mes)3Ir(PPh3) is therefore completely unsymmetrical (C1
symmetric). However, as the temperature is raised, two pairs
of aromatic hydrogens (δ 4.85 and 6.22, δ 6.27 and 6.44)
Inorganic Chemistry, Vol. 41, No. 18, 2002 4819

Jacobi et al.
Table 2. Dynamics of (mes)3Ir and (mes)3Ir(PPh3) in CD2Cl2
a As measured for (mes)3Ir generated in situ from (mes)3IrdO and excess
P(o-tol)3; Wilkinson et al. report essentially identical values (∆Gq ) 13.8
kcal/mol at 287 K) for Ir(mes)3 in CD2Cl2 in the absence of P(o-tol)3 and
OP(o-tol)3.6 b Extrapolated.
Figure 4. Air oxidation of PPh3 catalyzed by (mes)3IrdO. (a) Experimental
data (toluene, 1 atm of O2, 20 °C). In each run the P:Ir ratio is 50:1; iridium
concentrations range from 0.24 to 4.5 mM. (b) Kinetic simulations of the
catalytic reaction using the model in Scheme 2 and the following rate and
Scheme 2. Proposed Mechanism for (mes)3IrdO-Catalyzed Aerobic
Oxidation of Triphenylphosphine
equilibrium constants: kox ) 3.95 M-1s -1, kO ) 155 M-1 atm-1 s-1, and
2
Kassoc ) 84 M-1
.
simulations using the rate and equilibrium constants mea-
sured for the individual steps (Figure 4b; kox ) 3.95 M-1
s-1, kO ) 155 M-1 atm-1 s-1, as measured in CH2Cl2, and
2
Kassoc ) 84 M-1, as measured in toluene). The balance
between phosphine inhibition and oxygen uptake under these
conditions means that the overall turnover frequency is
relatively insensitive to total concentration in these experi-
ments, and is about 60 h-1. The agreement of the experi-
mental data with the predictions of the kinetic model based
on independent measurements of the individual steps (which
therefore has no adjustable parameters) is remarkably good,
especially considering the differences one might expect based
on the differing solvents (toluene vs dichloromethane). The
only serious discrepancy is the consistent presence of a
significantly elevated amount of phosphine oxidation in the
first few minutes. The origin of this small initial burst of
oxidation is unknown.
of the catalyst is iridium(III), as judged by the yellow-orange
color of the reacting solution, which reverts to blue-green
on consumption of the triphenylphosphine.
Catalytic oxidation of PPh3 with 2 mol % (mes)3IrdO in
toluene under 1 atm of O2 at 20 °C was monitored over time
by 31P{1H} NMR at a variety of concentrations (Figure 4a).
Dioxygen activation is rate-determining and is second order
in iridium, so its rate slows at lower iridium concentrations.
However, phosphine binding is stronger at higher phosphine
concentrations, and thus phosphine inhibition is less impor-
tant at lower phosphine concentrations. If the molar ratio of
phosphorus:iridium is held constant but the overall concen-
tration is varied, then these two effects oppose each other.
The net result, at 2 mol % iridium with [Ir] in the range
0.24-4.5 mM, is that the initial turnover frequencies decrease
slightly with increasing concentration, due to increased
phosphine inhibition. However, turnover frequencies increase
markedly at later times in the higher concentration runs as
the phosphine inhibition is relieved. These phenomena are
apparent in both the experimental runs and in kinetic
Discussion
Structure and Dynamics of (mes)3Ir(PPh3). Low-tem-
perature NMR spectroscopy clearly shows that the iridium-
(III) phosphine adduct (mes)3Ir(PPh3) has an unsymmetrical
(C1) structure (Figure 3). The triphenylarsine adduct shows
similar spectral features. The low symmetry excludes a
structure where the phosphine (or arsine) ligand binds on
the 3-fold axis of (mes)3Ir, and instead is most consistent
with a sawhorse geometry related to an octahedron by
4820 Inorganic Chemistry, Vol. 41, No. 18, 2002

Catalytic Oxygen ActiWation by Trimesityliridium(III)
removal of two cis ligands. This geometry appears to be
universal in structurally characterized nominally four-
coordinate d6 complexes such as Pt(C6Cl5)4,16 Rh(C6Cl5)3-
(py),17 [Rh(C6Cl5)3X]- (X ) Cl, C6Cl5) and [Rh(COC6Cl5)2-
(C6Cl5)Cl]-,18 [IrH2(PtBu2Ph)2]+ and [IrH(η2-C6H4PtBu2)(Pt-
Bu2Ph)]+,19 [Ru(R)(CO)(PtBu2Me)2]+,20 and RuCl2(PPh2[2,6-
Me2C6H3])2.21 In these structurally characterized complexes,
both vacant sites are generally occupied either by agostic
C-H bonds or o-Cl atoms, with the exception of [IrH(η2-
C6H4PtBu2)(PtBu2Ph)]+, where only one agostic interaction
can be achieved because of geometric constraints.19b How-
ever, calculations indicate that the sawhorse structure should
still be preferred even in the absence of agostic interac-
tions.19,22
suggests some energetic role for agostic methyl groups.
Likewise, while we have no definitive evidence for agostic
interactions in the phosphine adduct (mes)3Ir(PPh3), the
1
presence of upfield-shifted methyl groups in the H NMR
(δ 0.55, 1.00) suggest that one or two of the o-methyl groups
may be interacting with iridium. Unusually upfield reso-
nances for one of the phenyl groups (Phb, δ 5.52 and 6.38)
suggest that some interaction between that phenyl ring and
the iridium (agostic C-H or η2-aryl) may also be possible.
Such anomalous chemical shifts must be treated with extreme
caution as possible indicators of direct interaction with a
metal center; note for example the far upfield shift of one
of the mesityl aromatic hydrogens (δ 4.85), which cannot
be due to a direct interaction of the metal with the meta
position of the ring and which instead is presumably a
through-space effect of the ring current of one of the
phosphine phenyl groups. Nonetheless, the presence of some
interaction between Phb and the iridium would explain the
otherwise puzzling observation that P-C bond rotation in
that phenyl group is substantially slower than that in the two
other groups, including the unique phenyl group that is
presumably interdigitated between two of the mesityl groups.
Three independent dynamic processes can be observed in
Agostic interactions are likely present in (mes)3Ir(PPh3)
as well. Wilkinson and co-workers argued that agostic
interactions are absent in (mes)3Ir and (mes)3Rh because they
were unable to find any stigmata of their presence such as
upfield shifts in the 1H NMR, reduced values of JCH, or low-
frequency C-H stretches in the IR, and indeed concluded
that the trimesityls are planar in solution, in contrast to their
pyramidal structures in the solid state.6,23 The latter conclu-
sion is clearly erroneous, since the appearance of two equal-
intensity aromatic peaks and three equal-intensity methyl
peaks in the low-temperature NMR spectra of the trimesityls
is consistent only with a C3-symmetric static structure and
incompatible with any of the possible point groups of a planar
trimesityl complex (D3h, D3, C2V, C2, or C1). Pyramidalization
at iridium does not in and of itself require the presence of
agostic interactions with the o-methyl groups of the aryl
ligands (for example, calculations support a pyramidal
structure for nonagostic RhH3 in the gas phase24). However,
the fact that there is a substantial barrier to Ir-C rotation in
Ir(mes)3, with the ortho and aromatic resonances sharp and
distinct below -20 °C (∆Gq ) 13.4 kcal/mol at 250 K),
while the corresponding resonances in four-coordinate Od
Ir(mes)3 are equivalent even at -80 °C, is difficult to
reconcile with a purely steric barrier to bond rotation and
1
(mes)3Ir(PPh3) by H NMR at temperatures below those
required to promote phosphine dissociation (Table 2). Bond
rotation around each of the three P-C bonds takes place at
a different rate and two of these processes can be partially
or fully frozen out at low temperatures. At higher temper-
atures, a “rocking” motion takes place that reverses the sense
of the helical twist of the C3-symmetric (mes)3Ir fragment
and thus creates a time-averaged mirror plane. Equally
interesting are two transformations that are not observed.
Rotation about the Ir-P bond, which would interconvert the
three phenyl groups, does not occur at a significant rate until
the onset of phosphine dissociation. This “locking” of the
phosphine substituents is not surprising, given the extreme
crowding to be expected in (mes)3Ir(PPh3). Also unobserved
in (mes)3Ir(PPh3) is formation of a pseudotetrahedral species,
which would have the effect of causing exchange among all
mesityl (and phenyl) groups. Since the tetrahedral adduct
would presumably be favored sterically, this suggests a strong
electronic preference for the sawhorse-shaped geometry.
Since the aryl groups remain distinguishable until the onset
of phosphine dissociation, one can conclude that the tetra-
hedral complex is disfavored by at least 12 kcal/mol relative
to the sawhorse-shaped adduct.
Oxygen Atom Transfer Reactivity of (mes)3IrdO.
Given Wilkinson’s report that (mes)3Ir reacted rapidly with
dioxygen to form (mes)3IrdO,7 and given the high oxidation
state of iridium in (mes)3IrdO, we anticipated that (mes)3Ird
O would be an effective catalyst for the air oxidation of
organic substrates. This potential is indeed realized for the
oxidation of reactive substrates such as triphenylphosphine
and triphenylarsine. Key mechanistic features of the catalytic
air oxidation of triphenylphosphine are kinetically termo-
lecular oxygen uptake (second order in Ir(III), first order in
O2), bimolecular oxygen atom transfer from (mes)3IrdO to
PPh3, and the inhibition of oxidation by phosphine binding
(16) Fornie´s, J.; Menjo´n, B.; Sanz-Carrillo, R. M.; Toma´s, M.; Connelly,
N. G.; Crossley, J. G.; Orpen, A. G. J. Am. Chem. Soc. 1995, 11,
4295-4296.
(17) Garc´ıa, M. P.; Mart´ınez, A. P.; Jime´nez, M. V.; Siurana, C.; Oro, L.
A.; Lahoz, F. J.; Tiripicchio, A. Inorg. Chim. Acta 2000, 308, 51-
58.
(18) Garc´ıa, M. P.; Jime´nez, M. V.; Cuesta, A.; Siurana, C.; Oro, L. A.;
Lahoz, F. J.; Lo´pez, J. A.; Catala´n, M. P.; Tiripicchio, A.; Lanfranchi,
M. Organometallics 1997, 16, 1026-1036.
(19) (a) Cooper, A. C.; Streib, W. E.; Eisenstein, O.; Caulton, K. G. J.
Am. Chem. Soc. 1997, 119, 9069-9070. (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.
(20) Huang, D.; Streib, W. E.; Bollinger, J. C.; Caulton, K. G.; Winter, R.
F.; Scheiring, T. J. Am. Chem. Soc. 1999, 121, 8087-8097.
(21) Baratta, W.; Herdtweck, E.; Rigo, P. Angew. Chem., Int. Ed. 1999,
38, 1629-1631.
(22) Ujaque, G.; Cooper, A. C.; Maseras, F.; Eisenstein, O.; Caulton, K.
G. J. Am. Chem. Soc. 1998, 120, 361-365.
(23) (a) Hay-Motherwell, R. S.; Hussain-Bates, B.; Hursthouse, M. B.;
Wilkinson, G. J. Chem. Soc., Chem. Commun. 1990, 1242-1243. (b)
Hay-Motherwell, R. S.; Koschmieder, S. U.; Wilkinson, G.; Hussain-
Bates, B.; Hursthouse, M. B. J. Chem. Soc., Dalton Trans. 1991,
2821-2830.
(24) Landis, C. R.; Firman, T. K.; Root, D. M.; Cleveland, T. J. Am. Chem.
Soc. 1998, 120, 1842-1854,
Inorganic Chemistry, Vol. 41, No. 18, 2002 4821

Jacobi et al.
Table 3. Oxygen Atom Transfer Reactivity of (mes)3IrdO and
(mes)3Ir
to iridium(III) (Scheme 2). The reasonably good agreement
between the observed catalytic rates and the behavior
predicted from measurements of the individual steps indicates
that this model effectively describes the principal pathway
for oxidation. Other potential pathways, such as direct attack
of PPh3 on a transient iridium peroxide or air oxidation of
the phosphine adduct (mes)3Ir(PPh3), may therefore be
concluded to play little or no role in the reaction. Air
oxidation of triphenylphosphine is a relatively easily cata-
lyzed reaction, and many examples of oxidation catalysis
both by oxometal species25 and by other metal complexes26
have been reported. It is worth noting that (mes)3IrdO does
appear to be among the most active catalysts reported to date
among those catalysts with a significant lifetime.27 For
example, it is more active than the fastest oxo catalysts whose
rates have been reported at or near room temperature (e.g.,
MoO2(S2CNPr2), ∼15 turnovers/h at 40 °C25f) and similar
to palladium- or platinum-based catalysts (for example, Pd-
(OAc)2/THF, ∼55 turnovers/h at 30 °C26b).
a Data from ref 11. b Value given is that for deoxygenation of propylene
oxide. c Reference 7. d Value given is that for deoxygenation of PhCHdN-
(O)tBu.
But in general, (mes)3IrdO displays surprisingly little
reactivity toward organic reducing agents. The oxygen atom
transfer reactivity of the complex is summarized in Table 3.
Strong reducing agents such as PPh3 and AsPh3 (or aqueous
dithionite7) will deoxygenate (mes)3IrdO, and strong oxidiz-
ing agents such as Me3NO or O2 will oxygenate (mes)3Ir.7
One may use these reactions to bracket the strength of the
iridium-oxygen bond above that of Me3NO and below that
of Ph3AsO, giving a very approximate IrdO bond dissocia-
tion energy of about 85 ( 15 kcal/mol. However, within
this 30-kcal/mol gap, no reactivity is observed, and many
other substrates whose oxidations are undoubtedly thermo-
dynamically favorable (Me2SO, allyl alcohol) are also inert.
Given the lack of reactivity in both oxidative and reductive
directions, it must be concluded that the impediment is kinetic
in origin.
degree of polarity in its reaction with phosphines (F ) -0.29
( 0.10 for reactions of substituted triarylphosphines). The
electron-donating mesityl groups fit the iridium admirably
for dioxygen activation, but are less suitable for encouraging
subsequent oxygen atom transfer.
The geometric predilections of the iridium center may also
play a role in retarding oxygen atom transfer. Oxygen atom
transfer is an inner-sphere reaction, and the initial product
is not the free oxidized substrate but its metal complex. For
example, phosphine oxide complexes are universally invoked
as initial products of the addition of phosphines to oxometal
complexes, and can often be observed as the kinetic products
of such reactions.4,28 Typical d0 or d2 metal oxo complexes
can readily accommodate a phosphine oxide in the site
occupied by the oxo ligand prior to the oxygen atom transfer.
Thus addition of phosphine to the oxo ligand can take place
without substantial rearrangement of the coordination sphere.
In contrast, deoxygenation of d4 (mes)3IrO forms a d6 (mes)3-
Ir fragment where the preferred binding site is not along the
C3 axis where the oxo ligand was located, but rather off to
one side of the complex. The observed dynamic behavior of
(mes)3Ir(PPh3) indicates that this geometric preference is
sizable (>12 kcal/mol). The transition state for oxygen atom
transfer must exhibit a geometry unfavorable either to the
oxo reagent or to the complexed phosphine oxide product;
it cannot be optimal for both. This geometric mismatch can
obviously be overcome in sufficiently exothermic reactions
such as phosphine oxidation, but may contribute to the
relatively poor oxygen atom transfer kinetics that prevent
reactions with less easily oxidized substrates such as sulfides.
Two factors may contribute to the sluggishness of (mes)3Ird
O as an oxidant. One is that the strongly donating mesityl
groups may render the iridium center too electron-rich to be
a kinetically effective oxidant. The importance of electro-
philicity in oxidation of electron-rich substrates such as PPh3
is well-established,4. and the iridium shows only a modest
(25) (a) Hess, J. S.; Leelasubcharoen, S.; Rheingold, A. L.; Doren, D. J.;
Theopold, K. H. J. Am. Chem. Soc. 2002, 124, 2454-2455. (b) Huang,
R.; Espenson, J. H. J. Mol. Catal. A 2001, 168, 39-46. (c) Eager, M.
D.; Espenson, J. H. Inorg. Chem. 1999, 38, 2533-2535. (d) Zhu, Z.
L.; Espenson, J. H. J. Mol. Catal. A 1995, 103, 87-94. (e) Taqui
Khan, M. M.; Chatterjee, D.; Siddiqui, M. R. H.; Bhatt, S. D.; Bajaj,
H. C.; Venkatasubramanian, K.; Moiz, M. A. Polyhedron 1993, 12,
1443-1451. (f) Barral, R.; Bocard, C.; de Roch, I. S.; Sajus, L.
Tetrahedron Lett. 1972, 1693-1696.
(26) (a) Tshuva, E. Y.; Lee, D.; Bu, W.; Lippard, S. J. J. Am. Chem. Soc.
2002, 124, 2416-2417. (b) Lee, C. W.; Lee, J. S.; Cho, N. S.; Kim,
K. D.; Lee, S. M.; Oh, J. S. J. Mol. Catal. 1993, 80, 31-41. (c)
Ardizzoia, G. A.; Angaroni, M. A.; La Monica, G.; Cariati, F.; Cenini,
S.; Moret, M.; Masciocchi, N. Inorg. Chem. 1991, 30, 4347-4353.
(d) Tovrog, B. S.; Diamond, S. E.; Mares, F. J. Am. Chem. Soc. 1979,
101, 270-272. (e) Halpern, J.; Pickard, A. L. Inorg. Chem. 1970, 9,
2798-2800. (f) Wilke, G.; Schott, H.; Heimbach, P. Angew. Chem.,
Int. Ed. Engl. 1967, 6, 92-93.
(28) (a) Moyer, B. A.; Sipe, B. K.; Meyer, T. J. Inorg. Chem. 1981, 20,
1475-1480. (b) Conry, R. R.; Mayer, J. M. Inorg. Chem. 1990, 29,
4862-4867. (c) Dirghangi, B. K.; Menon, M.; Pramanik, A.; Chakra-
vorty, A. Inorg. Chem. 1997, 36, 1095-1101. (d) Chakraborty, I.;
Bhattacharyya, S.; Banerjee, S.; Dirghangi, B. K.; Chakravorty, A. J.
Chem. Soc., Dalton Trans. 1999, 3747-3753. (e) Banerjee, S.;
Bhattacharyya, S.; Dirghangi, B. K.; Menon, M.; Chakravorty, A.
Inorg. Chem. 2000, 39, 6-13. (f) Bhattacharyya, S.; Chakraborty, I.;
Dirghangi, B. K.; Chakravorty, A. Inorg. Chem. 2001, 40, 286-293.
(g) Gangopadhyay, J.; Sengupta, S.; Bhattacharyya, S.; Chakraborty,
I.; Chakravorty, A. Inorg. Chem. 2002, 41, 2616-2622.
(27) For example, porphyrinatoiron(II) complexes are extremely active
initially (>150 turnovers/h), but stop working after <30 total turnovers
due to formation of catalytically inactive µ-oxodiiron(III) complexes.
Chin, D.; La Mar, G. N.; Balch, A. L. J. Am. Chem. Soc. 1980, 102,
5945-5947.
4822 Inorganic Chemistry, Vol. 41, No. 18, 2002

Catalytic Oxygen ActiWation by Trimesityliridium(III)
Conclusions
the pseudotetrahedral oxo complex and the sawhorse-shaped
iridium(III) adduct that would be the immediate product of
atom transfer.
Trimesityliridium(III) reacts rapidly with dioxygen to form
oxotrimesityliridium(V) in a process that is second order in
iridium and first order in O2. When coupled with the ability
of (mes)3IrdO to oxidize substrates such as triphenylphos-
phine, this allows (mes)3Ir to catalyze the air oxidation of
PPh3 at 1 atm of pressure and room temperature at rates
comparable to the fastest catalysts reported to date. The
catalysis is significantly inhibited by binding of substrate to
(mes)3Ir to form a sawhorse-shaped adduct, (mes)3Ir(PPh3),
whose dynamic behavior has been probed by variable-
temperature NMR. Despite its high formal oxidation state,
(mes)3IrdO oxidizes only the most reactive oxygen atom
acceptors such as phosphines or arsines. The low kinetic
reactivity of this late metal oxo complex may be due to its
relative electron richness or to a geometric mismatch between
Acknowledgment. We gratefully acknowledge financial
support from the Camille and Henry Dreyfus Foundation
(New Professor Award), DuPont (Young Professor Award),
and the Dow Chemical Company (Innovation Recognition
Program). L.P. thanks the Bayer Corporation for a postdoc-
toral fellowship, administered by the Notre Dame Center for
Environmental Science and Technology.
Supporting Information Available: Eyring, van’t Hoff, and
Hammett plots, NMR titration data, and variable-temperature 31P-
{1H} and 1H NMR spectra (Figures S1-S9). This material is
available free of charge via the Internet at http://pubs.acs.org.
IC025700E
Inorganic Chemistry, Vol. 41, No. 18, 2002 4823