Electronic and medium effects on the rate of arene C-H bond activation by cationic Ir(III) complexes

Article abstract of DOI:10.1021/ja011809u

A detailed mechanistic study of arene C-H activation in CH₂Cl₂ solution by Cp*(L)IrMe(X) [L = PMe₃, P(OMe)₃; X = OTf, (CH₂Cl₂)BAr_f; (BAr_f = B[3,5-C₆H₃ (CF₃)₂]₄)^-] is presented. It was determined that triflate dissociation in Cp*(L)IrMe(OTf), to generate tight and/or solvent-separated ion pairs containing a cationic iridium complex, precedes C-H activation. Consistent with the ion-pair hypothesis, the rate of arene activation by Cp*(L)IrMe(OTf) is unaffected by added external triflate salts, but the rate is strongly dependent upon the medium. Thus the reactivity of Cp* (PMe₃)IrMe(OTf) can be increased by almost 3 orders of magnitude by addition of (n-Hex)₄NBAr_f, presumably because the added BAr_f anion exchanges with the OTf anion in the initially formed ion pair, transiently forming a cation/borate ion pair in solution (special salt effect). In contrast, addition of (n-Hex)₄NBAr_f to [Cp*PMe₃Ir(Me)CH₂Cl₂][BAr_f] does not affect the rate of benzene activation; here there is no initial covalent/ionic preequilibrium that can be perturbed with added (n-Hex)₄NBAr_f. An analysis of the reaction between Cp*(PMe₃)IrMe(OTf) and various substituted arenes demonstrated that electron-donating substituents on the arene increase the rate of the C-H activation reaction. The rate of C₆H₆ activation by [Cp*(PMe₃)Ir(Me)CH₂Cl₂] [BAr_f] is substantially faster than [Cp*(P(OMe)₃)Ir(Me)CH₂Cl₂] [BAr_f]. Density functional theory computations suggest that this is due to a less favorable preequilibrium for dissociation of the dichloromethane ligand in the trimethyl phosphite complex, rather than to a large electronic effect on the C-H oxidative addition transition state. Because of these combined effects, the overall rate of arene activation is increased by electron-donating substituents on both the substrate and the iridium complex.

Full text of DOI:10.1021/ja011809u

Published on Web 01/24/2002
Electronic and Medium Effects on the Rate of Arene C-H
Bond Activation by Cationic Ir(III) Complexes
David M. Tellers, Cathleen M. Yung, Bruce A. Arndtsen, Dan R. Adamson, and
Robert G. Bergman*
Contribution from the DiVision of Chemical Sciences, Lawrence Berkeley National Laboratory,
and Center for New Directions in Synthesis, Department of Chemistry,
UniVersity of California, Berkeley, California 94720
Received July 24, 2001
Abstract: A detailed mechanistic study of arene C-H activation in CH2Cl2 solution by Cp*(L)IrMe(X) [L )
PMe3, P(OMe)3; X ) OTf, (CH2Cl2)BArf; (BArf ) B[3,5-C6H3(CF3)2]4)-] is presented. It was determined that
triflate dissociation in Cp*(L)IrMe(OTf), to generate tight and/or solvent-separated ion pairs containing a
cationic iridium complex, precedes C-H activation. Consistent with the ion-pair hypothesis, the rate of
arene activation by Cp*(L)IrMe(OTf) is unaffected by added external triflate salts, but the rate is strongly
dependent upon the medium. Thus the reactivity of Cp*(PMe3)IrMe(OTf) can be increased by almost 3
orders of magnitude by addition of (n-Hex)4NBArf, presumably because the added BArf anion exchanges
with the OTf anion in the initially formed ion pair, transiently forming a cation/borate ion pair in solution
(special salt effect). In contrast, addition of (n-Hex)4NBArf to [Cp*PMe3Ir(Me)CH2Cl2][BArf] does not affect
the rate of benzene activation; here there is no initial covalent/ionic preequilibrium that can be perturbed
with added (n-Hex)4NBArf. An analysis of the reaction between Cp*(PMe3)IrMe(OTf) and various substituted
arenes demonstrated that electron-donating substituents on the arene increase the rate of the C-H activation
reaction. The rate of C6H6 activation by [Cp*(PMe3)Ir(Me)CH2Cl2][BArf] is substantially faster than [Cp*-
(P(OMe)3)Ir(Me)CH2Cl2][BArf]. Density functional theory computations suggest that this is due to a less
favorable preequilibrium for dissociation of the dichloromethane ligand in the trimethyl phosphite complex,
rather than to a large electronic effect on the C-H oxidative addition transition state. Because of these
combined effects, the overall rate of arene activation is increased by electron-donating substituents on
both the substrate and the iridium complex.
Introduction
and the electronic character of the C-H activation transition
state.5-9
There is much interest in furthering our understanding of
C-H bond breaking reactions because they hold great poten-
tial for developing new industrial processes and synthetic
methodologies.1-3 In 1993 we reported an unusual iridium(III)
complex, Cp*(PMe3)IrMe(OTf) (1), that undergoes a mild and
selective reaction with C-H bonds in arenes, alkanes, and other
functionalized organic molecules (eq 1).4
Complex 1 is both electronically and coordinatively saturated,
yet it requires only mild conditions for benzene C-H activation
(t1/2 ) 2 h at 25 °C in CH2Cl2, eq 1). The mechanism of C-H
activation with 1 continues to be the focus of current theoretical
and experimental work.10-16
Since mechanisms of C-H bond activation typically require
a vacant coordination site on the metal, the simplest mechanistic
(5) Tilset and co-workers have performed detailed experimental studies on
related cationic platinum systems. See the four references immediately
following this note.
(6) Johansson, L.; Tilset, M. J. Am. Chem. Soc. 2001, 123, 739.
(7) Heiberg, H.; Johansson, L.; Gropen, O.; Ryan, O. B.; Swang, O.; Tilset,
M. J. Am. Chem. Soc. 2000, 122, 10831.
We now wish to offer a more complete investigation of the
mechanism of C-H bond activation by this compound, includ-
ing a discussion of solvent and salt effects on the reaction rate
(8) Johansson, L.; Tilset, M.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.
2000, 122, 10846.
(9) Johansson, L.; Ryan, O. B.; Tilset, M. J. Am. Chem. Soc. 1999, 121, 1974.
(10) Luecke, H. F.; Bergman, R. G. J. Am. Chem. Soc. 1998, 120, 11008.
(11) Strout, D. L.; Zaric, S.; Niu, S.; Hall, M. B. J. Am. Chem. Soc. 1996, 118,
6068.
(12) Su, M.-D.; Chu, S.-Y. J. Am. Chem. Soc. 1997, 119, 5373.
(13) Hinderling, C.; Plattner, D. A.; Chen, P. Angew. Chem., Int. Ed. Engl. 1997,
36, 243.
(14) Hinderling, C.; Feichtings, D.; Plattner, D. A.; Chen, P. J. Am. Chem. Soc.
1997, 119, 10793.
(15) Alaimo, P. J.; Bergman, R. G. Organometallics 1999, 18, 2707.
(1) Arndtsen, B. A.; Bergman, R. G.; Mobely, T. A.; Peterson, T. H. Acc.
Chem. Res. 1995, 28, 154.
(2) Davies, J. A.; Watson, P. L.; Liebman, F.; Greenberg, A. SelectiVe
Hydrocarbon ActiVation; VCH: Toledo, 1990.
(3) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. Angew. Chem., Int. Ed. 1998,
37, 2181.
(4) Burger, P.; Bergman, R. G. J. Am. Chem. Soc. 1993, 115, 10462.
9
1400 VOL. 124, NO. 7, 2002 J. AM. CHEM. SOC.
10.1021/ja011809u CCC: $22.00 © 2002 American Chemical Society

Effects on the Rate of Arene C−H Bond Activation
A R T I C L E S
Scheme 1
Scheme 2
hypothesis for the reaction of saturated complex 1 involves prior
dissociation of triflate, forming an unsaturated cationic 16 e-
iridium fragment which then reacts to cleave the C-H bond
(Scheme 1).
Consistent with this mechanism, it was shown that replacing
the triflate ligand in 1 with the more weakly coordinating CH2-
Cl2 ligand and BArf [BArf ) B(3,5-C6H3(CF3)2)4-] counterion
increases the reactivity of the complex.17 The resulting com-
pound 3 showed unprecedented reactivity, activating benzene
at -30 °C (eq 2) to form [Cp*(PMe3)IrPh(CH2Cl2)][BArf] (4).
investigated and are less clearly understood. Phenomena such
as the “common ion effect” are complicated when applied to
nonpolar media, since ions created in these systems are likely
to be highly associated and thus do not exist as free ions to any
great extent. The possible presence of several associated ionic
species (solvent-separated ion pairs, tight ion pairs, higher
aggregates) complicates the mechanistic investigation of such
processes.31
The studies discussed in this report were largely carried out
on the reactions of iridium triflate 1 with arenes in dichloro-
methane solvent, both with and without added salts. In addition
to the studies involving medium effects, we also examined the
electronic requirements for Ir(III) aryl C-H bond activation.
These effects were studied both via modification of the metal
complex and by studying a range of disubstituted arenes. To
facilitate these experiments, the phosphite analogue of 1, Cp*-
(P(OMe)3)IrMe(OTf) (5), was synthesized and investigated, as
was the borate salt of this Ir(III) complex, [Cp*(P(OMe)3)IrMe-
(CH2Cl2)][BArf] (6).
Single-crystal X-ray analysis of 3 demonstrated that in the solid
state, the borate anion is dissociated from the iridium fragment,
with a chlorine-bound CH2Cl2 molecule occupying a coordina-
tion site on the metal center.
The quantitative mechanistic study of the reaction shown in
eq 1 is complicated by the reaction’s sensitivity to its environ-
ment. The study of medium effects in metal-mediated processes
is an important but poorly understood area of chemistry.18-24
Solvent and salt effects in organic SN1 reactions have been more
extensively studied, but almost exclusively in relatively polar
and/or protic solvents such as water, alcohols, and acetic acid.
Adding an external salt usually increases the rate of such a
dissociative reaction,25-30 in many cases by increasing the
effective polarity of the medium. Reaction rate enhancement
then occurs by a shift in an initial covalent/ionic dissociation
equilibrium toward further dissociation at higher ionic strengths
(the “normal salt effect”).31 Reactions that involve ionic
dissociation in nonpolar solvents have been less extensively
Results and Discussion
Synthesis of Compounds. The synthesis of the phosphite
triflate complex 5 (Scheme 2) was accomplished by a route
analogous to that used for the phosphine complex 1.4 With Cp*-
(P(OMe)3)IrMe2, the reaction of 1 equiv of HOTf can be
controlled to produce 5 as the major product. This is in contrast
to Cp*(PMe3)IrMe2, which more readily undergoes disubstitu-
tion, producing the corresponding bis-triflate complex. However,
the conproportionation between Cp*(P(OMe)3)IrMe2 and Cp*-
(P(OMe)3)Ir(OTf)2 (Scheme 2) was still the preferred synthetic
route to the trimethyl phosphite complex 5, since higher yields
and cleaner initial products were obtained. Phosphite complex
5 does decompose slowly at room temperature, but it is more
stable than the higher molecular weight iridium phosphites
synthesized by Simpson, which readily decompose at room
temperature by intramolecular C-H activation of the phosphite
group.32 The phosphite borate complex 6 was generated via
metathesis with NaBArf. It can be characterized in solution, but
was found to be unstable when isolated both in the solid state
and in solution at room temperature.
(16) Klei, S. R.; Tilley, T. D.; Bergman, R. G. J. Am. Chem. Soc. 2000, 122,
1816.
(17) Arndtsen, B. A.; Bergman, R. G. Science 1995, 270, 1970.
(18) Castellani, M. P.; Hesse, E. T.; Tyler, D. R. Organometallics 1994, 13,
399.
(19) Poli, R.; Owens, B. E.; Linck, R. G. Inorg. Chem. 1992, 31, 662.
(20) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1996, 118,
5961.
(21) Trost, B. M.; Bunt, R. C. J. Am. Chem. Soc. 1998, 120, 70.
(22) Brainard, R. L.; Nutt, W. R.; Lee, T. R.; Whitesides, G. M. Organometallics
1988, 7, 2379.
(23) Peters, R. G.; White, S.; Roddick, D. M. Organometallics 1998, 17, 4493.
(24) Cheng, T.; Brunschwig, B. S.; Bullock, R. M. J. Am. Chem. Soc. 1998,
120, 13121.
(25) Allen, A. D.; Tidwell, T. T.; Tee, O. S. J. Am. Chem. Soc. 1993, 115,
10091.
(26) Bockman, T. M.; Hubig, S. M.; Kochi, J. K. J. Am. Chem. Soc. 1998, 120,
2826.
(27) Loupy, A.; Tchoubar, B.; Astruc, D. Chem. ReV. 1992, 92, 1141.
(28) Hossain, M. A.; Schneider, H. Chem.-Eur. J. 1999, 5, 1284.
(29) Winstein, S.; Klinedinst, P. E.; Robinson, G. C. J. Am. Chem. Soc. 1961,
83, 885.
(30) Goodson, F. E.; Wallow, T. I.; Novak, B. M. J. Am. Chem. Soc. 1997,
119, 12441.
(31) Loupy, A.; Tchoubar, B. Salt Effects in Organic and Organometallic
Chemistry; VCH: Weinheim, Germany, 1992; Chapter 2.
Kinetic Studies. Our first kinetic studies were carried out
on the C-H activation reaction of benzene with trimethylphos-
phine-substituted methyl triflate 1 in dichloromethane solution.
We considered the mechanism outlined in Scheme 3 (Scheme
4 for P(OMe)3) as a starting point for analysis of the data. With
the knowledge that step 2 is irreversible, the rate law predicted
for this mechanism is given in eqs 3 and 4, assuming the steady-
(32) Simpson, R. D. Organometallics 1997, 16, 1797.
9
J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1401

A R T I C L E S
Tellers et al.
Scheme 3
Scheme 4
Figure 1. Plot of observed rate constant (kobs) vs [C6H6] and [C6D6] for
the reaction of 1 (1.1 mM) at 10 °C. PPNOTf (1.8 mM) was also present.
state approximation and a fast, essentially complete final
recombination step.
k1k2[C6H6]
rate ) kobs[1] )
rate ) kobs[5] )
[1]
[5]
(3)
(4)
k-1[OTf] + k2[C6H6]
k4k5[C6H6]
k-4[OTf] + k5[C6H6]
At sufficiently high benzene concentrations, we would predict
saturation behavior in arene for this mechanism (kobs ) k1 at
high R-H concentrations). However, complex 1 does not show
this saturation behavior at elevated benzene concentrations in
dichloromethane at 10 °C (Figure 1). While the reaction rate
does increase as expected at low benzene concentrations, it
actually decreases at benzene concentrations greater than 3 M.
We believe this is because at these very high concentrations of
benzene, the effective polarity of the solvent is perturbed,
becoming substantially less polar than pure CH2Cl2. This would
shift the initial ionization equilibrium toward the covalent
complex 1 (Scheme 3), inhibiting the overall rate. At the sugges-
tion of a referee, at 2.6 M benzene concentration cyclohexane-
d12, instead of additional benzene, was added to the reaction
mixture to determine whether this would produce rate retardation
similar to that observed at higher concentrations of benzene.
The kobs without added C6D12 was measured to be 2.32 × 10-4
s-1 (2.6 M C6H6, 10 °C). When 0.5 and 1.0 M C6D12 were
added, the reaction rates were measured to be 1.43 × 10-4 and
9.59 × 10-5 s-1, respectively. Addition of C6D12 to the reaction
mixture thus decreases the kobs, supporting our explanation of
the decrease in rate observed at high benzene concentrations.
In addition, a C6H6/C6D6 isotope effect of 3.5 was measured at
1.0 M benzene concentration, demonstrating that kinetic satura-
tion in [C6H6] is not achievable under these conditions with
complex 1. At substrate saturation, there should be little or no
isotope effect observed, since the rate would be substrate
independent and would only depend on k1.
Figure 2. Plot of observed rate constant (kobs) vs [C6H6] and [C6D6] for
the reaction of 5 (12.0 mM) at 30 °C.
A similar result was obtained when the phosphite triflate
complex 5 was treated with varying amounts of benzene in
dichloromethane at 30 °C (Figure 2). The reaction rate begins
to decrease at concentrations greater than 2 M, and an isotope
effect (C6H6/C6D6, 1.0 M benzene concentration) of 3.6 is
observed. In addition, the overall rate of reaction of 5 with
benzene is substantially slower than that for 1. We suggest that
this result is due to the fact that P(OMe)3 is more electron-
withdrawing than PMe3, slowing the initial triflate dissociation
step (k4) in the phosphite complex 5 (vide infra) (Scheme 4).
Salt Concentration Dependence on the Reactivity of 1, 3,
and 5. To probe the effect of ionic strength on the C-H
activation reaction, the reaction of the triflate complex 1 with
benzene in dichloromethane was run in the presence of variable
amounts of [(PPh3)2N][B(3,5-C6H3(CF3)2)4] (PPNBArf). At
higher [PPNBArf] (>20 mM), the kinetic analysis was com-
plicated by the rapid formation of PPNOTf as a precipitate
during the reaction. Thus the triflate complex 1 not only reacted
with benzene but also simultaneously underwent a metathesis
reaction, to generate the more reactive CH2Cl2/borate complex
3 (Scheme 5).
The reaction of the methyl triflate complex 1 with PPNBArf
without benzene present was shown by NMR analysis to also
proceed within minutes to the CH2Cl2/borate complex 3.
Similarly, the phenyl triflate complex 2 reacts with PPNBArf
9
1402 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Effects on the Rate of Arene C−H Bond Activation
A R T I C L E S
Figure 4. Plot of observed rate constant (kobs) vs [(n-Hex)4NBArf] for the
reaction of 5 (0.013 M) and C6H6 (0.25 M) at 10 °C in CH2Cl2.
Figure 3. Plot of observed rate constant (kobs) vs [(n-Hex)4NBArf] for the
reaction of 1 (1.90 mM) and C6H6 (51.2 mM) at 25.2 °C in CH2Cl2.
Scheme 6
Scheme 5
Scheme 7
to form the less stable phenyl CH2Cl2/borate complex 4. A
summary of the reactions of 1 with PPNBArf and benzene in
dichloromethane is shown in Scheme 5.
salt approaches the rate of the BArf complexes 3 and 6 (e.g.,
for 6, kobs ) 3.4 × 10-3 s-1), but the complications at high salt
concentrations prevented the observation of further approach
toward this expected limit.
This metathesis, precipitation, and decomposition behavior
prevented us from collecting reliable kinetic data in the presence
of PPNBArf. To circumvent these problems, a more hydrophobic
salt, (n-Hex)4NBArf, was used. In this case no precipitate is
observed, and we believe that with (n-Hex)4NBArf, neither the
methyl triflate complex 1 nor the phenyl triflate complex 2
undergoes extensive metathesis to form the corresponding CH2-
Cl2/borate complexes at room temperature (i.e., the position of
the equilibrium strongly favors the covalent iridium triflate
complex 1). The reaction of 1 with benzene using variable
concentrations of (n-Hex)4NBArf proceeds cleanly even at
extremely high (n-Hex)4NBArf concentrations (up to 350 mM).
The plot of kobs versus [(n-Hex)4NBArf] is shown in Figure 3.
The added salt strongly accelerates the rate, and the plot is linear
over a wide range of (n-Hex)4NBArf concentrations.
Similar behavior is observed when the reaction of phosphite
triflate complex 5 with benzene is run in the presence of variable
concentrations of (n-Hex)4NBArf (Figure 4). At relatively low
salt concentrations (<200 mM), the reaction rate increased
linearly with the concentration of salt. At higher salt concentra-
tions, the reaction rate still increased, but began to deviate from
linear behavior. The reaction rate could not be measured with
accuracy at high salt concentrations (>400 mM), because the
solubility limit of the salt in CH2Cl2 was reached.33 The reaction
rate for both 1 and 5 in the presence of added (n-Hex)4NBArf
The increase in reaction rate for both the PMe3 and the
P(OMe)3 systems observed above demonstrates the operation
of a salt effect. One explanation for this observation is that the
added (n-Hex)4NBArf directly assists in ionization of the triflate
iridium bond (normal salt effect) (Scheme 6, complexes 1 and
5 are generalized as Cp*(L)IrMe(OTf) (7)).34 Although normal
salt effects on organic solvolysis reactions are not as large as
those in polar solvents, rate increases of >103 have been seen
in relatively nonpolar solvents such as diethyl ether and
acetone.34,35
However, our preferred rationale for this effect is that by
introducing a source of borate anion, the iridium triflate complex
7 is activated by transiently forming a borate complex in
solution. As depicted in Scheme 7, dissociation of the triflate
anion from 7 produces ion pair 8. Subsequent exchange between
the triflate and BArf anions results in the BArf complex 9. Both
ion pairs 8 and 9 are assumed to react very rapidly and
irreversibly with benzene. Yet with (n-Hex)4NBArf salt, the
equilibrium shifts toward the BArf complex 9 and away from
the triflate ion pair 8 and its covalent complex 7 and, as a result,
the reaction rate increases. This effect would be analogous to
the special salt effect observed by Winstein and co-workers in
the solvolysis reactions of alkylsulfonates in polar solvents (e.g.,
(33) Additionally, at such high salt concentrations the n-hexyl resonances of
the added salt began to obscure the P(OMe)3 resonances, making the
acquisition of reliable data difficult.
(34) Winstein, S.; Friedrich, E. C.; Smith, S. J. Am. Chem. Soc. 1964, 86, 305.
(35) For a theoretical treatment of salt effects in nonpolar solvents, see: Perrin,
C. L.; Pressing, J. J. Am. Chem. Soc. 1971, 93, 5705.
9
J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1403

A R T I C L E S
Tellers et al.
Scheme 8
Scheme 9
Figure 5. Plot of observed rate constant (kobs) vs [(n-Hex)4NOTf] for the
reaction of 1 (1.1 mM) with 1.92 M C6H6 at 10 °C.
acetic acid) where addition of perchlorate anion results in a
marked increase in the reaction rate.36-38 They attributed the
rate increase to formation of a more reactive perchlorate species
via exchange of the sulfonate anion for the perchlorate anion
(analogous to our exchange of OTf for BArf).39
change in the reaction rate.41 We believe it is highly unlikely
that these effects would exactly cancel each other. Instead, it is
more likely that (n-Hex)4NOTf is also ion-paired and, as a result,
does not change the ionic strength of the solvent significantly.42-45
In this scenario, no triflate inhibition is observed because
benzene reacts directly with a tight or solvent-separated ion pair
(Scheme 8, eq 5). This explanation is further supported by our
observation that added PPNOTf also has no effect on the rate
of benzene activation by phosphite 5 (Scheme 9, eq 6).46
When (n-Hex)4NBArf is added to a mixture of the preformed
phosphine CH2Cl2/borate 3 and benzene, it does not affect the
(already rapid) rate of reaction, nor is the reactivity of the
phosphite CH2Cl2/borate 6 affected by added (n-Hex)4NBArf.
This result is consistent with our model since these CH2Cl2/
borate complexes already exist as ion pairs in solution, so there
is no initial covalent-ionic preequilibrium in these systems that
can be perturbed with added (n-Hex)4NBArf.8,40
k7k8[C6H6]
rate ) kobs[1] )
rate ) kobs[5] )
[1]
[5]
(5)
(6)
k-7 + k8[C6H6]
Qualitative reactivity studies of 1 with alkanes and ethers in
the presence of (n-Hex)4NBArf indicate that similar rate
accelerations with added (n-Hex)4NBArf salt occur for all
substrates investigated. The increase in observed rate again spans
almost 3 orders of magnitude upon the addition of (n-
Hex)4NBArf. Such an extreme salt effect is likely a result of
the stabilization of the ion pair by the highly polar added
noncoordinating salts.35
k9k10[C6H6]
k-9 + k10[C6H6]
Solution Behavior of 1 and 3. In the solid state, complex 3
has a single chlorine-bound dichloromethane molecule coordi-
nated to the iridium center. In CH2Cl2 solution, it is likely that
a CH2Cl2 molecule would also occupy a coordination site of 3.
In CD2Cl2 solution, the CD2Cl2/borate complex 3 shows discrete
1H NMR chemical shifts clearly different from those assigned
to the triflate analogue 1. These large spectral differences are
accounted for by our postulate that complex 1 exists predomi-
nately as a covalent uncharged inner-sphere triflate, while com-
plex 3 exists predominantly as a cationic covalent Ir-CH2Cl2
complex/borate ion pair.
Dichloromethane is a very weak donor ligand,17 and bound
CH2Cl2 or CD2Cl2 likely undergoes rapid exchange with CD2-
Cl2 solvent via a dynamic solvation equilibrium. Consistent with
this assumption, at room temperature the 1H NMR spectrum of
a CD2Cl2 solution containing both the triflate complex 1 and
the CD2Cl2/borate salt 3 does not show resolved peaks for
either compound. However, at lower temperatures the peaks
Effect of Added Triflate. When the reaction of triflate
complex 1 with benzene was performed with variable amounts
of added (n-Hex)4NOTf, the observed rate remained constant
(Figure 5). This is not in agreement with preequilibration of 1
with free ions, which would predict an inverse dependence of
kobs on [OTf] (Scheme 3, eq 3). The added (n-Hex)4NOTf could
also potentially serve to increase the effective solvent polarity
and increase the rate of reaction. One explanation is that there
are actually two counteracting effects: (a) inhibition of the
ionization equilibrium by the common ion effect (as would be
predicted from Scheme 1) and (b) a (fortuitously) compensating
increase in the degree of ionization of 1 due to the increased
effective polarity of the solution, thereby resulting in no net
(36) Winstein, S.; Clippinger, E. J. J. Am. Chem. Soc. 1956, 78, 2784.
(37) Winstein, S.; Robinson, G. C. J. Am. Chem. Soc. 1958, 80, 169.
(38) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry; Harper and Row: New York, 1987; Chapter 4.
(39) For the classical special salt effect observed with organic substrates in polar
solvents, there is a large nonlinear increase in the rate constant at low salt
concentrations. At higher salt concentrations, the rate constant growth slows
down, becoming linear, similar to that observed for a normal salt effect.
We believe we do not see a change in the slope of the plot for phosphine
1 because under our conditions, the special salt effect regime appears to
extend to much higher concentrations of salt in the relatively nonpolar
solvent CH2Cl2.
(40) Recently, Johansson et al. have explored the reaction of diimine-based
platinum cations with benzene in trifluoroethanol. In analogy to our results
with 3, they found that the addition of salts to the reaction mixture had no
effect on the reaction rate. In addition, benzene coordination was found to
be rate determining, and the C-H activation step most likely occurs via
oxidative addition (Pt(II) to Pt(IV)). See ref 8.
(41) To probe this possibility further, an experiment was performed with 1 using
variable ratios of (n-Hex)4NBArf:(n-Hex)4OTf but with a constant total
amount of salt ((n-Hex)4NBArf + (n-Hex)4NOTf ) constant). While the
results of these experiments did show inhibition at higher concentrations
of triflate, the constant total salt concentration is a rather poor approximation
for constant ion strength in dichloromethane, and the results cannot be
quantitatively analyzed.
(42) For a discussion of the interactions between the PPN cation and different
anions, see the two references immediately following this note.
(43) Tilset, M.; Zlota, A. A.; Folting, K.; Caulton, K. G. J. Am. Chem. Soc.
1993, 115, 4113.
(44) Alunni, S.; Pero, A.; Reichenbach, G. J. Chem. Soc., Perkin Trans. 1998,
1747.
(45) See ref 29 for the observation of a relatively small linear salt effect in
acetic acid solution with a common ion present.
(46) Consistent with this observation, compound 1 is a very weak electrolyte in
CH2Cl2. See ref 4.
9
1404 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Effects on the Rate of Arene C−H Bond Activation
A R T I C L E S
q
Scheme 10
for both compounds can be resolved (Tc ) -71 °C, ∆Gc
)
11.8 kcal mol-1).
Identities of the Reactive Species in CH2Cl2. We believe
that the triflate complex 1 and the CD2Cl2/borate salt 3 lead to
common or similar cationic Ir-CH2Cl2 intermediates. Two main
pieces of evidence suggest that ion pairs (rather than free ions
or the covalent triflate) are the reactive species in dichloro-
methane (Schemes 8 and 9). The first involves the lack of any
common ion effect when (n-Hex)4NOTf is added to the reaction
mixture. This rules out the specific rate laws in eqs 3 and 4
which predict an inverse dependence on triflate concentration.
The second piece of evidence stems from the fact that compound
5 reacts with benzene at a much slower rate than compound 1
does. The ion-pair mechanism illustrated in Scheme 8 predicts
that a shift in the initial ionization equilibrium to the left (e.g.,
by replacing PMe3 with P(OMe)3) should slow the overall
reaction rate. This also is consistent with our observations. In
summary, the lack of a (n-Hex)4NOTf dependence coupled with
differences in reaction rates between 1 and 5 suggest that
dissociation of triflate to generate a solvent-separated ion pair
occurs prior to rate-limiting C-H activation.
Finally, it is important to ask whether the iridium cation reacts
with benzene as the 18 e- solvate or as the 16 e- unsaturated
cation; that is, does arene activation occur in a dissociative or
an associative manner? We currently favor dissociative loss of
dichloromethane prior to arene cleavage. This is based primarily
upon results obtained with the hydridotris(pyrazolyl)borate (Tp)
analogue of 3, [Tp(PMe3)Ir(Me)CH2Cl2][BArf], where we have
evidence supporting the intermediacy of an unsaturated 16 e-
iridium cation.47,48 However, it should be noted that Gladysz
and co-workers examined the mechanism of dichloromethane
substitution in [CpPPh3Re(NO)ClCH2Cl][BF4].49 They con-
cluded that the substitution likely occurred in an associative
manner, via either an 18 e- bent nitrosyl complex or, relevant
to our system, an 18 e- cyclopentadienyl slipped (η3) complex.
Recently, Tilset and Johansson have noted that the square planar
cationic platinum complex [(diimine)PtCH3(L)][OTf] activates
C-H bonds in an associative manner.6 We are currently
attempting to address this question with further studies on the
Cp* system.
Reaction Trends Using Disubstituted Arenes as Sub-
strates. The studies with added salt support the idea that the
active iridium species in arene C-H activation is cationic and,
therefore, possibly electrophilic. We decided to investigate
whether the substrate’s electronic properties would reflect this,
predicting that electron-rich arenes should react faster. This was
accomplished by comparing the rates of reaction of several 1,2-
disubstituted arenes with complex 1. We chose 1,2-disubstitution
because most of these substrates react cleanly to yield a single
product, resulting from attack at the 4-position of the arene
(Scheme 10). 1,2-Difluorobenzene is an exception and yields
two products, in about a 1:2 ratio. Characterization of this
mixture by NMR spectroscopy indicates that they are the 1,2,3-
and 1,3,4-trisubstituted products, respectively. In this case, it is
postulated that the small size of the fluorine atom permits
activation of the C-H bond at the 3-position.
Table 1. Observed Rate Constants for the Reaction of 1 (1.90
mM) with 1,2-Disubstituted Arenes (51.2 mM) in CH2Cl2 at
14.1 °Ca
substrate
kobs (s-1
)
tetrahydronaphthalene
xylene
0.000844
0.001131
0.000533
0.000166
benzeneb
dichlorobenzene
a PPNBArf (20.0 mM) was also present. b Corrected for the number of
hydrogens.
The observed rate constants for the reaction between 1 and
the various substituted arenes are shown in Table 1. Benzene
was statistically corrected by a factor of (2/6) due to the
difference in the number of C-H bonds available for activation
(six C-H bonds available on benzene vs two accessible C-H
bonds on a disubstituted arene). While this system does not show
completely linear Hammett-type behavior, there is a clear trend
indicating that electron-withdrawing substituents decrease the
reactivity of the substrate, as would be expected if the arene
acts as a nucleophile.50,51
A competition experiment carried out by treating a 10-fold
excess 1:1 mixture of 1,2-dichlorobenzene and o-xylene with
the CH2Cl2/borate complex 3 gave almost exclusively (>90%
by NMR spectroscopy) the o-xylene activated product. This
indicates that the CH2Cl2/borate complex 3, like triflate complex
1, also prefers electron-rich substrates.
Comparison of the Reactivity of the CH2Cl2/Borate
Complexes 3 and 6. Interpretation of substituent effects at the
metal center is complicated because modification of the complex
has the potential to affect several different steps in the reaction
(47) Tellers, D. M.; Bergman, R. G. Can. J. Chem. 2001, 79, 525.
(48) Because 1 reacts quickly with benzene, and for solubility reasons, the
reactions must be performed in dichloromethane, we have so far not been
able to obtain experimental evidence supporting either mechanism.
(49) Dewey, M. A.; Zhou, Y.; Liu, Y.; Gladysz, J. A. Organometallics 1993,
12, 3924.
(50) Periana and co-workers have noted that methane bisulfate, the product of
methane functionalization in their platinum system, is likely stable to
overoxidation because of the electron-withdrawing character of the bisulfate
group. See the following reference.
(51) Periana, R. A.; Taube, D. J.; Gamble, S.; Taube, H.; Satoh, T.; Fujii, H.
Science 1998, 280, 560.
9
J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1405

A R T I C L E S
Tellers et al.
Table 2. Calculated Energies (kcal/mol) for the Reaction of
expected, the alkane complex is somewhat lower in energy than
the separated cation and free methane. Significantly, the alkane
complex/oxidative addition product energy gap is very similar
for the PMe3 and P(OMe3)3 systems.55 These calculations
suggest that the rate difference observed between the phosphine
3 and phosphite 6 cannot be attributed solely to electronic
differences in the C-H activation transition state.
[Cp(L)IrMe]+ with Methanea
energy
energy
L ) PMe3
(kcal/mol)
L ) P(OMe)3
(kcal/mol)
[Cp(L)IrMe]+ + CH4
[Cp(L)IrMe(H-CH3)]+
[Cp(L)Ir(Me)2H]+
0
[Cp(L)IrMe]+ + CH4
[Cp(L)IrMe(H-CH3)]+
[Cp(L)Ir(Me)2H]+
0
-3
4.8
-0.3
6.5
a Energy of “[Cp(L)IrMe]+ + methane” has been arbitrarily set to zero.
Even if they are imprecise in an absolute sense, the results
of the DFT calculations suggest it is likely that the rate
difference between phosphine 3 and phosphite 6 is the result
of the relative differences in dissociation of the dichloromethane
ligand. Because the iridium center in phosphite 6 is less electron-
rich, the rate of dissociation of dichloromethane is undoubtedly
slower. This likely contributes to the slower rate of C-H
activation by phosphite 6. Overall, it is clear from this work
that use of more electron-releasing ancillary ligands (i.e.,
phosphine vs phosphite) lowers the barrier to loss of the leaving
group (OTf in 1; CH2Cl2 in 3) and, as a result, provides more
facile access to the complex responsible for C-H activation.
Whatever the exact cause, it is clear that the oVerall rate of
C-H actiVation is inhibited by electron-withdrawing substitu-
ents on both the substrate and the complex. This atypical pattern
appears to contrast the standard “electrophile-nucleophile”
pattern of reactivity that exists for most organic and organo-
metallic even-electron bimolecular reactions.
mechanism. For the triflate complexes 1 and 5, both the initial
preequilibrium step and the actual C-H bond activation step
can be influenced by the electronic and steric differences caused
by changes in electron density at the metal. While the preequi-
librium effects were discussed earlier, we wished to determine
the effect of such a substitution upon the actual C-H bond-
breaking step. In an attempt to accomplish this, we compared
the reactivity of the CH2Cl2/borate complexes 3 and 6, where
we know that dissociation of bound dichloromethane (k11 in eq
7, k12 in eq 8) is fast in both cases.
Conclusion
The reactions of the phosphine CH2Cl2/borate (3) and
phosphite CH2Cl2/borate (6) complexes with benzene in CD2-
Cl2 were both followed at 0 °C by NMR spectroscopy. These
experiments showed that the phosphite complex (kobs ) (3.0 (
0.3) × 10-4 s-1) is less reactive than the phosphine complex
(kobs ) (8.4 ( 0.8) × 10-3 s-1) by a factor of nearly 30. Since
the phosphite ligand is actually less sterically demanding than
the phosphine ligand,52 this suggests that the diminished
reactivity of the phosphite complex is due to an electronic rather
than a steric effect, resulting from less electron density at the
metal center. In parallel studies, we have found that replacing
the Cp* ligand with the hydridotris(3,5-dimethylpyrazolyl)borate
(TpMe2) ligand has given similar results.53 The TpMe2 ligand was
shown to be less electron-donating than Cp*, and the resulting
complex is less reactive than the Cp* complex 1. However, in
this case, it is more difficult to separate electronic from steric
effects since the TpMe2 ligand is also bulkier than the Cp* ligand.
It is tempting to attribute this effect to the C-H activation
transition state. However, we believe that the more likely
explanation is that decreasing the electron density at the metal
center affects k12/k-12 relative to k11/k-11 (eqs 7 and 8); that is,
the preequilibrium is shifted to the left, thereby decreasing the
overall rate of C-H activation.48 This explanation was rein-
forced by DFT calculations performed using Jaguar54 at the
B3LYP/LACVP**+ level of theory on the reaction of methane
with the model system [Cp(L)IrMe]+ (L ) PMe3, P(OMe)3).
Optimized geometries and energies were calculated for [Cp-
(L)IrCH3]+/CH4, the alkane complex [Cp(L)IrCH3(H-CH3)]+,
and the iridium(V) complex [Cp(L)Ir(CH3)2H]+. The relative
energies of these systems (Table 2) were in good accord with
those reported by Hall and co-workers.11 In both cases, as
The C-H activation reactions of triflate complex 1 with arene
and alkane substrates almost certainly occur after dissociation
of triflate, leading to a reactive cationic iridium/CH2Cl2 solvate
(Scheme 8). The reaction is sensitive to medium effects;
decreasing the effective polarity of the CH2Cl2 solvent by adding
large amounts of benzene decreases the overall rate of reaction.
Alternatively, the reactivity of 1 can be increased by almost 3
orders of magnitude by addition of the hydrophobic salt (n-
Hex)4NBArf. We believe that this is due to a special salt effect,
with addition of salt causing triflate/BArf exchange and resulting
in the formation of the more reactive iridium-borate ion pair.
Added salts are well known to increase the rates of ionic
reactions in polar solvents.31 Quantitative studies of organo-
metallic salt effects in nonpolar solvents are relatively rare, but
salts have been used in several cases to increase rates or modify
the nature of reactions in these media.31 This approach is
advantageous because the system’s fundamental components
remain unmodified chemically. In this regard, we are currently
examining the activities of other organometallic reaction systems
which involve ionic intermediates in nonpolar solvents with
added bulky, soluble salts. As Winstein and co-workers noted
in 1964, “Carrying out such [ionic organic] reactions under ‘salt
promoted’ conditions may be useful practically and interesting
theoretically.” 34
The reactions of 1 with 1,2-disubstituted arenes indicate that
electron-withdrawing groups on the arene substrate decrease the
reaction rates. The reactivity of the phosphite complex 5 was
also compared to the reactivity of complex 1. The relative
reactivities of the phosphine and phosphite CH2Cl2/borate
complexes indicate that electron-withdrawing groups on the
(52) Tolman, C. A. Chem. ReV. 1977, 77, 313.
(53) Tellers, D. M.; Bergman, R. G. J. Am. Chem. Soc. 2000, 122, 954.
(54) Jaguar, version 4.1; Schro¨dinger, Inc.: Portland, Oregon, 2000.
(55) The transition states were not sought for these two systems because it is
unlikely that the relatively small difference in energy between the two
systems would translate into a large energy difference in the transition state.
9
1406 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Effects on the Rate of Arene C−H Bond Activation
A R T I C L E S
metal complex also decrease the oVerall rate.56-63 This unusal
pattern of reactivity, whereby a reaction is enhanced by electron-
withdrawing substituents on both reactants, is rather uncommon
but may be reasonable considering the mechanism of C-H bond
activation. We propose that this is caused by a decrease in the
preequilibrium concentration of the reactive iridium cation.
Additional work will be required on systems that are not
complicated by an initial preequilibrium to conclusively identify
the electron-density requirements of the metal center in so-called
“electrophilic reactions”.
Finally, understanding the electronic requirements of C-H
activation is potentially useful for gaining insight into the
mechanism of C-H activation by other so-called “electrophilic”
C-H activation reactions, such as those recently observed in
analogous cationic platinum systems.8,9,20,51,64 A great deal of
work has focused upon modeling the Shilov system3 by utilizing
complexes of the type [L2PtMe(solvent)]+. To obtain a greater
understanding of these systems and compare their behavior with
that of the cationic Cp*- and Tp-substituted iridium systems, it
would be useful to determine what electronic properties are
important for both the substrate and the metal. Recent progress
has been made on this problem;65 as observed in our Ir system,
the rate of C-H activation at platinum is also enhanced by
increasing electron donation to the metal center.
Kinetic studies were performed using a Hewlett-Packard 8453 UV-
vis spectrometer or by NMR spectroscopy using a Bruker AMX-300
spectrometer (300 MHz for 1H spectra) or a Bruker AMX-400
spectrometer (see above). All reaction mixtures were prepared under
N2 and kept under static N2 during the kinetic analyses. For UV-vis
kinetics experiments, the disappearance of 1 was monitored by change
in absorbance at 314 nm. The disappearance of 5 was monitored
at 300 nm. For NMR kinetics experiments, the disappearance of the
Ir-Me resonances of 1 and 2 and the P(OMe)3 resonances of 5 and 6
was monitored. Time versus absorbance data were plotted using
Kaleidograph and fit to an exponential decay with a variable t∞ value.
Calibration of total sample volumes was done using a volumetric flask.
The concentration versus absorption plots for complexes 1, 2, 5, and 6
follow the Lambert-Beer law in CH2Cl2 for the concentration regimes
studied (<5 mM).
Melting points were obtained with samples in sealed tubes under
N2 and are uncorrected. The compounds NaBArf,67 PPNOTf,68
[Cp*IrCl2]2,69 Cp*(PMe3)Ir(Cl)2,69 Cp*(PMe3)Ir(OTf)2,69 Cp*(PMe3)-
IrMe2,4 and Cp*(PMe3)IrMe(OTf)4 were prepared using literature
procedures.
[(CH3(CH2)5)4N][B(3,5-C6H3(CF3)2)4]. In a 20 mL scintillation vial
containing 220 mg (0.57 mmol) of (n-Hex)4NCl was added NaBArf
(510 mg, 0.58 mmol) and CH2Cl2 (10 mL). The solution was stirred
for 24 h, after which it was filtered through Celite, and the solvent
was removed in vacuo. This provided (n-Hex)4NBArf as a crude tan
solid. The solid was dissolved in a minimum of CH2Cl2 (∼1 mL), and
pentane (∼1 mL) was added. This mixture was cooled to -70 °C for
3 h, and the resulting clear, colorless crystals were removed by filtration
and dried under vacuum. Recrystallization in an analogous manner
yielded a white solid (370 mg, 75% yield). 1H NMR (400 MHz, CD2-
Cl2, 298 K): δ 7.72 (s, 8H, o-B(3,5-C6H3(CF3)2)4), 7.55 (s, 4H, p-B(3,5-
C6H3(CF3)2)4), 3.03 (m, 8H, (CH3(CH2)4CH2)4NH), 1.61 (m, 8H,
(CH3(CH2)3CH2CH2)4N), 1.35 (m, 24H, (CH3(CH2)3CH2CH2)4N), 0.90
(t, 12H, 3JH-H ) 6.9 Hz, (CH3(CH2)3CH2CH2)4N). 13C{1H} NMR (101
MHz, CD2Cl2, 298 K): δ 162.1 (q, 1JB-C ) 50.1 Hz, i-B(Ar′4)), 134.8
Experimental Section
General. Unless otherwise noted, all reactions and manipulations
were performed in a Vacuum Atmospheres circulating inert atmosphere
glovebox or using standard Schlenk techniques. Glassware was dried
in an oven at 150 °C before use. Unless otherwise noted, reagents were
purchased from commercial suppliers and used without further purifica-
tion. Merck silica gel, 60 Å, 230-400 mesh, grade 9385 was used in
chromatography unless otherwise noted. Silated silica gel (Silica Gel
60), 63-200 µm particle size, was obtained from EM Science. Diethyl
ether and THF were distilled from sodium/benzophenone ketyl under
N2 prior to use. Hexanes, pentane, benzene, and CH2Cl2 were passed
through activated alumina under N2, then sparged with N2 and stored
over 4 Å molecular sieves.66 Deuterated solvents were degassed by
freezing, evacuating, and thawing (3x), and were then dried over 4 Å
sieves and stored under N2.
2
1
(s, o-B(Ar′4)), 128.8 (q, JF-C ) 29.8 Hz, m-B(Ar′4 )), 124.9 (q, JF-C
3
) 271.9 Hz, CF3), 117.4 (septet, JF-C ) 4.0 Hz, p-B(Ar′4)), 59.2 (s,
(CH3(CH2)4CH2)4N), 30.9 (s, (CH3(CH2)3CH2CH2)4N), 25.8 (s, (CH3-
(CH2)2CH2(CH2)2)4N), 22.2 (s, (CH3CH2CH2(CH2)3)4N), 21.8 (s,
(CH3CH2(CH2)4)4N), 13.4 (s, (CH3(CH2)2CH2(CH2)2)4N). 19F NMR (377
MHz, CD2Cl2, 298 K): -62.8 (s). Anal. Calcd for C56H76BF24N: C,
54.68; H, 6.23; N, 1.14. Found: C, 54.76; H, 6.03; N, 1.13. mp 99-
100 °C. UV-vis (CH2Cl2): λmax ) 308 nm, ꢀmax ) 925 L mol-1 cm-1
.
Unless otherwise indicated, NMR spectra were obtained using a
Bruker AMX-400 MHz spectrometer (400 MHz for 1H spectra, 100.6
MHz for 13C{1H} spectra, 162.1 MHz for 31P{1H} spectra, 376.5 MHz
for 19F spectra). Chemical shifts are reported in parts per million (δ)
relative to residual protiated solvent, coupling constants are reported
in Hertz (Hz), and integrations are reported in number of protons. Unless
otherwise noted, samples for NMR analysis were prepared using CD2-
Cl2 as the solvent.
[(CH3(CH2)5)4N][O3SCF3]. In a 20 mL scintillation vial containing
220 mg (0.56 mmol) of (n-Hex)4NCl was added AgOTf (160 mg, 0.62
mmol) and CH2Cl2 (10 mL). The solution was stirred for 24 h in
darkness, after which it was filtered through Celite, and the solvent
was removed in vacuo over 1 day. This provided (n-Hex)4NOTf as a
clear, colorless liquid. The compound was cooled to -70 °C, and a
1
white crystalline solid formed (280 mg, 99% yield). H NMR (400
MHz, CD2Cl2, 298 K): δ 3.14 (m, 8H, (CH3(CH2)4CH2)4N), 1.62 (m,
8H, (CH3(CH2)3CH2CH2)4N), 1.35 (m, 24H, (CH3(CH2)3CH2CH2N),
3
0.91 (t, 12H, JH-H ) 6.9 Hz, (CH3(CH2)3CH2CH2)4N). 13C{1H}
(56) See the seven references immediately following this note for studies of
electronic effects on oxidative addition reactions.
NMR (101 MHz, CD2Cl2, 298 K): δ 59.0 (s, (CH3(CH2)4CH2)4N), 31.1
(s, (CH3(CH2)3CH2CH2)4N), 25.9 (s, (CH3(CH2)2CH2(CH2)2)4N), 22.3
(s, (CH3CH2CH2(CH2)3)4N), 21.9 (s, (CH3CH2(CH2)4)4N), 13.6 (s,
(CH3(CH2)2CH2(CH2)2)4N). 19F NMR (377 MHz, CD2Cl2, 298 K): δ
-79 (s). Anal. Calcd for C25H52F3NO3S: C, 59.61; H, 10.40. Found:
C, 59.63; H, 10.63. mp ∼18 °C. UV-vis (CH2Cl2): λmax ) 308 nm,
(57) Ugo, R.; Pasini, A.; Fusi, A.; Cenini, S. J. Am. Chem. Soc. 1972, 94, 7364.
(58) Hascall, T.; Rabinovich, D.; Murphy, V. J.; Beachy, M. D.; Friesner, R.
A.; Parkin, G. J. Am. Chem. Soc. 1999, 121, 11402.
(59) Abu-Hasanayn, F.; Goldman, A. S.; Krogh-Jespersen, K. J. Phys. Chem.
1993, 97, 5890.
(60) Sakaki, S.; Biswas, B.; Sugimoto, M. Organometallics 1998, 17, 1278.
(61) van de Kuil, L. A.; Luitjes, H.; Grove, D. M.; Zwikker, J. W.; van der
Linden, J. G. M.; Roelofsen, A. M.; Jenneskens, L. W.; Drenth, W.; van
Koten, G. Organometallics 1994, 13, 468.
(62) Johnson, C. E.; Eisenberg, R. J. Am. Chem. Soc. 1985, 107, 3148.
(63) Sargent, A. L.; Hall, M. B.; Guest, M. F. J. Am. Chem. Soc. 1992, 114,
517.
(64) Fang, X.; Scott, B. L.; Watkin, J. G.; Kubas, G. Organometallics 2000,
19, 4193.
ꢀmax ) 925 L mol-1 cm-1
.
[(PPh3)2N][3,5-C6H3(CF3)2)4]. In a 20 mL scintillation vial contain-
ing PPNCl (200 mg, 0.35 mmol) was added NaBArf (320 mg, 0.36
mmol) and CH2Cl2 (10 mL). The solution was stirred for 24 h, after
(65) Zhong, H. A.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2002, 124,
1378-1399.
(66) Alaimo, P. J.; Peters, D. W.; Arnold, J.; Bergman, R. G. J. Chem. Educ.
2001, 78, 64.
(67) Brookhart, M.; Grant, B.; Volpe, A. F. Organometallics 1992, 11, 3920.
(68) Obendorf, D.; Peringer, P. Anorg. Chem. Org. Chem. 1986, 41, 79.
(69) Stang, P. J.; Huang, Y. H.; Arif, A. M. Organometallics 1992, 11, 231.
9
J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1407

A R T I C L E S
Tellers et al.
which it was filtered through Celite, and the solvent was removed in
vacuo. This provided PPNBArf as a crude tan solid. The solid was
dissolved in a minimum of CH2Cl2 (∼1 mL), and pentane (∼1 mL)
was added. The mixture was cooled to -70 °C for 3 h, and the resulting
white crystals were removed by filtration and dried. Recrystallization
in an analogous manner yielded a white solid (370 mg, 75% yield). 1H
NMR (400 MHz, CDCl3, 298 K): δ 7.71 (m, 8H, o-Ar′4B), 7.59 (m,
6H, p-Ph3P), 7.49 (s, 4H, p-Ar′4B), 7.43 (m, 24H, o-Ph3P). 13C{1H}
) 10.8 Hz, 9H, PMe3). 13C{1H} NMR (101 MHz, CD2Cl2, 298 K): δ
147.6 (d, 2JP-C ) 14 Hz, i-C6H3Cl2), 137.8 (bs, o-C6H3Cl2), 136.6 (bs,
o-C6H3Cl2), 131.7 (bs, C-Cl), 129.7 (s, m-C6H3Cl2), 126.8 (s, C-Cl),
94.1 (d, 2JP-C ) 4 Hz, C5Me5), 14.7 (d, 1JP-C ) 38 Hz, PMe3), 9.7 (s,
C5Me5). 31P{1H} NMR (162 MHz, CD2Cl2, 298 K): δ -22.8. 19F NMR
(377 MHz, CD2Cl2, 298 K): δ -77.0 (s). MS (EI) m/z: 698 (M+).
Anal. Calcd for C20H27O3IrPF3Cl2S: C, 34.39; H, 3.40. Found: C,
34.29; H, 3.76.
1
NMR (101 MHz, CDCl3, 298 K): δ 162.3 (q, JBC ) 50.1 Hz,
i-B(Ar′4)), 134.6 (s, o-B(Ar′4)), 133.8 (s, m-PPh3), 132.1 (d, p-PPh3,
Cp*(PMe3)Ir(3,4-C6H3F2)OTf and Cp*(PMe3)Ir(2,3-C6H3F2)OTf.
The addition of 1,2-C6H4F2 (0.54 g, 4.7 mmol) to 1 (98 mg, 0.17 mmol)
in 5 mL of CH2Cl2 resulted in the slow darkening of the reaction
solution. After 24 h, the solvent was removed in vacuo to provide a
2JP-C ) 4.1 Hz), 129.5 (d, o-PPh3, 3JP-C ) 4.1 Hz), 128.8 (q, 2JF-C
)
29.7 Hz, m-B(Ar′4 )), 124.9 (q, 1JF-C ) 271.9 Hz, CF3), 117.3 (septet,
3JF-C ) 3.8 Hz, p-B(Ar′4)). 19F NMR (377 MHz, CDCl3, 298 K): δ
-62.5 (s). 31P{1H} NMR (162 MHz, CD2Cl2, 298 K): 21.6 (s). Anal.
Calcd for C68H42NBF24P2: C, 58.26; H, 3.02; N, 1.00. Found: C, 58.31;
H, 2.98; N, 1.09. mp 170-172 °C. UV-vis (CH2Cl2): λmax ) 308
1
mixture of isomers (ca. 1:2 ratio, with less than 5% impurities by H
NMR spectroscopy) as an orange solid in a 100% crude yield (120
mg, 0.17 mmol). The isomers could not be separated, and so the mixture
1
was characterized only by H and 19F NMR spectroscopy.
nm, ꢀmax ) 925 L mol-1 cm-1
.
1
Major Isomer: H NMR (400 MHz, CD2Cl2, 298 K): δ 7.0 (m,
4
Cp*(PMe3)Ir(3,4-C6H3(CH3)2)OTf. The addition of 1,2-C6H4Me2
(0.64 g, 6.0 mmol) to 1 (160 mg, 0.27 mmol) in CH2Cl2 (5 mL) resulted
in the slow darkening of the reaction solution. After 24 h, the solvent
was removed in vacuo to provide crude 12 as an orange solid.
Crystallization from a mixture of pentane (∼1 mL) and CH2Cl2 (∼1
mL) at -70 °C led to 12 in 86% yield (160 mg, 0.23 mmol). 1H
NMR (400 MHz, CD2Cl2, 298 K): δ 6.9 (m, 3H, Ir-(3,4-C6H3Me2)),
2.20 (s, 3H, Ir-(3,4-C6H3Me2)), 2.16 (s, 3H, Ir-(3,4-C6H3Me2)), 1.62
(d, 15H, 4JP-H ) 1.6 Hz, C5Me5), 1.45 (d, 9H, 2JP-H ) 10.7 Hz, PMe3).
3H, Ir-(3,4-C6H3F2)), δ 1.64 (d, 15H, JP-H ) 1.5 Hz, C5Me5), δ 1.47
2
(d, 9H, JP-H ) 11.4 Hz, PMe3). 31P{1H} NMR (162 MHz, CD2Cl2,
298 K): δ -23.8 (s). 19F NMR (377 MHz, CD2Cl2, 298 K): δ -79 (s,
OTf), δ -118 (s, C6H3F2), δ -142 (s, C6H3F2).
1
Minor Isomer: H NMR (400 MHz, CD2Cl2, 298 K): δ 7.0 (m,
4
3H, Ir-(2,3-C6H3F2)), 1.66 (d, 15H, JP-H ) 1.7 Hz, C5Me5), 1.54 (d,
9H, 2JP-H ) 11.7 Hz, P(Me3). 31P{1H} NMR (162 MHz, CD2Cl2, 298
K): δ -24.0 (s). 19F NMR (377 MHz, CD2Cl2, 298 K): δ -78 (s,
OTf), -141 (s, C6H3F2), -149 (s, C6H3F2).
3
13C{1H} NMR (101 MHz, CD2Cl2, 298 K): δ 133.8 (d, JP-C ) 3.1
Cp*(P(OMe)3)Ir(Cl)2. [Cp*IrCl2]2 (502 mg, 0.63 mmol) was
dissolved in dry CH2Cl2 (120 mL) in a Schlenk flask with a stir bar.
To this dark red solution was added P(OMe)3 (0.20 g, 1.6 mmol), and
the resulting mixture was allowed to stir under a static N2 atmosphere
for 2 h. The solvent was removed in vacuo providing a soft orange
solid which was dissolved in CH2Cl2 and chromatographed in air on a
short SiO2 column (3 cm × 2 cm) using 75:25 Et2O:CH2Cl2. The dark
orange fraction was collected, and the solvent was removed in vacuo
to provide Cp*(P(OMe)3)Ir(Cl)2 as a deep orange solid (605 mg, 92%
yield). 1H NMR (400 MHz, CD2Cl2, 298 K): δ 3.81 (d, 3JP-H ) 15.8
Hz, i-C6H3Me2), 128.1 (s, C6H3Me2), 122.8 (s, C6H3Me2), 122.1 (s,
C6H3Me2), 94.1 (s, C5(Me)5), 19.4 (s, C6H3Me2), 18.5 (s, C6H3Me2),
1
14.5 (d, JP-C ) 50 Hz, PMe3), 9.1 (s, C5(Me)5). 31P{1H} NMR (162
MHz, CD2Cl2, 298 K): δ -23.6 (s). 19F NMR (377 MHz, CD2Cl2,
298K): δ -78.7 (s). Anal. Calcd for C22H33F3IrO3PS: C, 40.17; H,
5.06. Found: C, 39.95; H, 4.76.
Cp*(PMe3)Ir(C6H3(C4H8))OTf. The addition of 1,2,3,4-tetrahydro-
naphthalene (71 mg, 0.54 mmol) to 1 (61 mg, 0.11 mmol) in CH2Cl2
(5 mL) resulted in the slow darkening of the reaction solution. After
24 h, the solution was concentrated (1 mL) and passed through a plug
of silated silica gel, making sure to collect only the orange band to
obtain the product free of excess tetrahydronaphthalene. The resulting
solution was concentrated in vacuo down to an orange foam which
was rinsed with cold pentane (3 × 2 mL). The resulting solid was then
extracted with pentane (4 × 5 mL). The light yellow pentane solution
was concentrated (10 mL) and cooled to -40 °C. Orange crystals of
Cp*(PMe3)Ir(C6H3(C4H8)) (46 mg, 0.067 mmol) were isolated in 63%
4
Hz, 6H, P(OMe)3), 1.67 (d, JP-H ) 1.7 Hz, 15H, C5Me5). 13C{1H}
NMR (101 MHz, CD2Cl2, 298 K): δ 94.4 (s, C5Me5), 53.8 (s, P(OMe)3),
8.34 (s, C5Me5). 31P{1H} NMR (162 MHz, CD2Cl2, 298 K): δ 82.2
(s). Anal. Calcd for C13H24Cl2IrO3P: C, 29.89; H, 4.63. Found: C,
29.80; H, 4.63.
Cp*(P(OMe)3)Ir(Me)2. Cp*(P(OMe)3)Ir(Cl)2 (260 mg, 0.50 mmol)
was suspended in THF (∼14 mL) with stirring, and MeLi (∼1.4 M in
Et2O, 1.0 mL) was added dropwise. Over 16 h, the solution turned
light yellow. The solution was filtered through a medium porosity fritted
funnel containing 3 g of alumina(III). The solids were washed with
∼20 mL of additional Et2O, and the yellow filtrates were collected
and evaporated to dryness. The resulting yellow crystals were dissolved
in pentane (20 mL) and passed through another fritted funnel containing
3 g of alumina(III), yielding a clear filtrate. The solvent was removed
in vacuo, yielding white crystalline Cp*(P(OMe)3)Ir(Me)2 (130 mg,
1
yield. H NMR (500 MHz, CD2Cl2, 298 K): δ 6.92 (bm, 2H, C6H3-
(C4H8)), 6.75 (d, 2JH-H ) 7.5 Hz, 1H, C6H3(C4H8)), 2.73 (bs, 2H, C6H3-
(C4H8)), 2.67 (bs, 2H, C6H3(C4H8)), 1.73 (bs, 4H, C6H3(C4H8)), 1.65
(d, 4JP-H ) 1.5 Hz, 15H, C5Me5), 1.48 (d, 2JP-H ) 11 Hz, 9H, PMe3).
2
13C{1H} NMR (125 MHz, CD2Cl2, 298 K): δ 142.1 (d, JP-C ) 14
Hz, i-C6H3(C4H8)), 137.2 (s, C-H, C6H3(C4H8)), 136.7 (s, Cq, C6H3-
(C4H8)), 133.9 (s, C-H, C6H3(C4H8)), 131.4 (s, Cq, C6H3(C4H8)), 128.8
2
56% yield). 1H NMR (400 MHz, CD2Cl2, 298 K): δ 3.31 (d, 3JP-H
)
(s, C-H, C6H3(C4H8)), 93.6 (d, JP-C ) 4 Hz, C5Me5), 29.9 (s, C6H3-
4
(C4H8)), 29.2 (s, C6H3(C4H8)), 24.3 (s, C6H3(C4H8)), 24.3 (s, C6H3-
(C4H8)), 14.8 (d, 1JP-C ) 37.8 Hz, PMe3), 9.7 (s, C5Me5). 31P{1H} NMR
(162 MHz, CD2Cl2, 298 K): δ -22.4 (s). 19F NMR (377 MHz, CD2Cl2,
298 K): δ -77.1 (s). HRMS (EI) m/z for [C24H35O3F3PSIr]+ (M+):
calcd 684.1626, obsd 684.1621. Anal. Calcd for C24H35F3O3SIrP‚
(C5H12)0.5: C, 44.22; H, 5.74. Found: C, 44.29; H, 5.70.
22 Hz, 9H, P(OMe)3), 1.63 (d, JP-H ) 1.8 Hz, 15H, C5Me5), -0.04
3
(d, JP-H ) 5.2 Hz, 6H, IrMe2). 13C{1H} NMR (101 MHz, CD2Cl2,
298 K): δ 94.6 (s, C5Me5), 52.0 (s, P(OMe)3), 9.87 (s, C5Me5), -19.4
(d, 2JP-C ) 10 Hz, Ir-Me2). 31P{1H} NMR (162 MHz, CD2Cl2, 298K):
δ 93.7 (s). Anal. Calcd for C15H30IrO3P: C, 37.41; H, 6.28. Found: C,
37.42; H, 6.34.
Cp*(PMe3)Ir(C6H3Cl2)OTf. Addition of 1,2-C6H4Cl2 (370 mg, 2.5
mmol) to 1 (72 mg, 0.13 mmol) in CH2Cl2 (5 mL) resulted in the slow
darkening of the reaction solution. After 24 h, the solvent was removed
in vacuo to yield an orange solid. Crystallization from a mixture of
pentane (∼0.5 mL) and CH2Cl2 (∼1 mL) at -30 °C led to Cp*-
Cp*(P(OMe)3)Ir(OTf)2. Cp*(P(OMe)3)Ir(Cl)2 (880 mg, 1.7 mmol)
and AgOTf (410 mg, 3.4 mmol) were stirred in CH2Cl2 (∼90 mL)
under N2 for 12 h in darkness. The solution was cannula filtered away
from the off-white salts, and the orange supernatant was evaporated to
dryness. The resulting bright orange solid was washed with ether (3 ×
15 mL) to remove trace Ag-containing impurities, and the solids
remaining were dried under vacuum to yield Cp*(P(OMe)3)Ir(OTf)2
(1.2 g, 92% yield). Material suitable for elemental analysis was obtained
1
(PMe3)Ir(C6H3Cl2)OTf in 48% yield (43 mg, 0.061 mmol). H NMR
(400 MHz, CD2Cl2, 298 K): δ 7.38 (bs, 1H, IrC6H3Cl2), 7.14 (bm,
4
2
2H, IrC6H3Cl2), 1.65 (d, JP-H ) 1.6 Hz, 15H, C5Me5), 1.50 (d, JP-H
9
1408 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Effects on the Rate of Arene C−H Bond Activation
A R T I C L E S
by crystallization at -40 °C from a concentrated CH2Cl2 solution
Kinetics of Conversion of 1 and C6H6 to 2. A typical experiment
was performed as follows. In a glovebox, 1 (6.1 mg, 0.011 mmol) was
dissolved in CH2Cl2 (1 mL) and added to a 10 mL volumetric flask.
To this was added a CH2Cl2 (6 mL) solution of PPNOTf (12.0 mg,
0.018 mmol) and benzene (3.1 g, 0.039 mol). The solution was diluted
to 10 mL with CH2Cl2. The solution was stirred briefly and then
transferred (∼ 4 mL) by pipet to a quartz UV-vis cuvette equipped
with a Kontes high-vacuum Teflon stopcock. The cuvette was removed
from the box and immediately placed at 10.0 °C in the UV-vis cuvette
holder. Spectra were acquired at 300 s intervals over 8 h.
Kinetics of Conversion of 5 and C6H6 to Cp*(P(OMe)3)IrPh-
(OTf). A typical experiment was performed as follows. In a glovebox,
a 2 mL volumetric flask was charged with hexamethylbenzene (2.2
mg, 0.014 mmol), C6H6 (180 µL, 2.01 mmol), and diluted to 2 mL
with CD2Cl2. A 1 mL volumetric flask was charged with 5 (8.0 mg,
0.013 mmol) and diluted to 1 mL with the benzene/CD2Cl2 solution.
The orange solution was mixed thoroughly and then transferred to an
NMR tube. The tube was attached to a Cajon adapter, degassed on a
vacuum line, and flame sealed. The tube was kept at -196 °C until it
was ready to be placed in an NMR probe held at 30 ( 0.1 °C. Spectra
were acquired at 300 s intervals over 10 h.
Kinetic Experiments on the Reaction of 1 with Benzene in the
Presence of Added [(CH3(CH2)5)4N][B(3,5-C6H3(CF3)2)4]. A typical
experiment was performed as follows. In a glovebox, 16.2 mg (0.0284
mmol) of 1 was dissolved in CH2Cl2 and added to a 10 mL volumetric
flask. The solution was diluted to 10 mL with CH2Cl2. An aliquot of
the resulting solution (1.00 mL) was transferred to a 20 mL scintillation
vial containing 1.0 mL of a CH2Cl2 solution of [(n-hexyl)4N][BArf]
(115 mg, 0.094 mmol). Also transferred to the vial was 1.00 mL of a
0.154 M solution of C6H6 in CH2Cl2. The solution was stirred briefly
to ensure the salt was completely dissolved, and it was then transferred
by pipet to a quartz UV-vis cuvette equipped with a Kontes high-
vacuum Teflon stopcock. The cuvette was removed from the box and
immediately placed at 14.1 °C in the UV-vis cuvette holder. Spectra
were acquired at 100 s intervals over 4 h.
1
layered with pentane. H NMR (400 MHz, CD2Cl2, 298 K): δ 3.84
3
4
(d, 6H, JP-H ) 12.4 Hz, P(OMe)3), 1.63 (d, 15H, JP-H ) 3.1 Hz,
C5Me5). 13C{1H} NMR (99.7 MHz, CD2Cl2, 298 K): δ 94.9 (s, C5-
Me5), 53.2 (s, P(OMe)3), 9.30 (s, C5Me5). 31P{1H} NMR (162 MHz,
CD2Cl2, 298 K): δ 107.3 (s). 19F NMR (377 MHz, CD2Cl2, 298 K):
δ -78.9 (s, OTf). MS (EI): m/z 750 (M+). Anal. Calcd for C15H24F6-
IrO9PS2: C, 37.41; H, 6.28. Found: C, 37.42; H, 6.34.
Cp*(P(OMe)3)IrMe(OTf) (5). A diethyl ether (200 mL) solution
of Cp*(P(OMe)3)IrMe2 (3.2 g, 6.4 mmol) was added to a 500 mL round-
bottom flask containing Cp*(P(OMe)3)Ir(OTf)2 (4.8 g, 6.4 mmol). Upon
stirring for 1 day, all of the material had dissolved. The orange solution
was filtered through silated silica gel, washed with diethyl ether, and
the solvent was removed in vacuo. The orange solid was washed with
pentane (50 mL) and dried in vacuo to afford 5 (7.3 g, 93% yield). 1H
3
NMR (400 MHz, CD2Cl2, 298 K): δ 3.74 (d, JP-H ) 11.5 Hz, 6H,
P(OMe)3), 1.73 (d, 4JP-H ) 3.2 Hz, 15H, C5Me5), 1.20 (d, 3JP-H ) 6.5
Hz, 3H, Ir-Me). 13C{1H} NMR (101 MHz, CD2Cl2, 298 K): δ 94.9 (s,
C5Me5), 52.5 (s, P(OMe)3), 9.92 (s, C5Me5), -11.4 (s, Ir-Me). 31P{1H}
NMR (162 MHz, CD2Cl2, 298 K): δ 97.5 (s). 19F NMR (377 MHz,
CD2Cl2, 298 K): δ -78.6 (s). Anal. Calcd for C15H27F3IrO6PS: C,
29.27; H, 4.42. Found: C, 29.24; H, 4.56.
Cp*(P(OMe)3)IrPh(OTf). The addition of C6H6 (1.0 g, 13 mmol)
to 5 (230 mg, 0.370 mmol) in 7 mL of CH2Cl2 resulted in the darkening
of the reaction solution over 2 h. After 10 h, the solvent was removed
in vacuo to provide Cp*(P(OMe)3)IrPh(OTf) as an orange solid (250
mg, 99% yield). 1H NMR (400 MHz, CD2Cl2, 298 K): δ 7.64 (d, 2JH-H
) 7.4 Hz, 2H, o-C6H5), 7.30 (dd, 3JH-H ) 7.4 and 7.6 Hz, 2H, m-C6H5),
3
2
7.08 (t, JH-H ) 7.6 Hz, 1H, p-C6H5), 3.67 (d, JP-H ) 11.2 Hz, 9H,
P(OMe)3), δ 1.74 (d, JP-H ) 2.7 Hz, 15H, C5Me5). 13C{1H} NMR
(101 MHz, CD2Cl2, 298 K): δ 140.1 (d, JP-C ) 15.3 Hz, i-C6H5),
4
2
130.7 (s, o-C6H5), 128.0 (s, m-C6H5), 123.2 (s, p-C6H5), 96.1 (s, C5-
Me5), 52.5 (s, P(OMe)3), 8.94 (s, C5Me5). 31P{1H} NMR (162 MHz,
CD2Cl2, 298 K): δ 94.2 (s). 19F NMR (377 MHz, CD2Cl2, 298 K): δ
-78.5 (s). HRMS (EI) m/z for [C20H29O6F3PSIr]+ (M+): calcd
678.1004, obsd 678.1006.
Kinetic Experiments on the Reaction of 5 with Benzene in the
Presence of Added [(CH3(CH2)5)4N][B(3,5-C6H3(CF3)2)4]. A typical
experiment was performed as follows. A vial was charged with 5 (8.0
mg, 0.013 mmol) and 0.5 mL of CD2Cl2 to yield an orange solution.
A second vial was charged with 0.5 mL of CD2Cl2, C6Me6 (0.9 mg,
0.006 mmol), C6H6 (23 µL, 0.25 mmol), and (n-hexyl)4NBArf (24 mg,
0.020 mmol) to yield a clear solution. This clear solution was added to
the orange solution of 5 and mixed thoroughly. The mixture was
transferred to an NMR tube using a pipet. The tube was attached to a
Cajon adapter, degassed on a vacuum line, and flame sealed. The tube
was kept at -196 °C until it was ready to be placed in an NMR probe
held at 30 ( 0.1 °C. Spectra were acquired at 300 s intervals over 10 h.
Examination of Triflate Dependence for the Reaction of 1 with
Benzene. A typical experiment was performed as follows. In a
glovebox, 1 (6.1 mg, 0.011 mmol) was dissolved in CH2Cl2 (4 mL)
and added to a 10 mL volumetric flask. To this was added a CH2Cl2 (4
mL) solution of (n-Hex)4NOTf (101 mg, 0.200 mmol) and benzene
(1.5 g, 0.019 mol). The solution was diluted to 10 mL with CH2Cl2.
The solution was stirred briefly and then transferred (∼ 4 mL) by pipet
to a quartz UV-vis cuvette equipped with a Kontes high-vacuum Teflon
stopcock. The cuvette was removed from the box and immediately
placed at 10.0 °C in the UV-vis cuvette holder. Spectra were acquired
at 300 s intervals over 8 h.
[Cp*(P(OMe)3)Ir(Me)CD2Cl2][BArf] (6). In a glovebox, complex
5 (10 mg, 0.016 mmol) was dissolved in CD2Cl2 (2 mL) and cooled to
-40 °C before NaBArf (14 mg, 0.016 mmol) was added, and the
mixture was shaken intermittently over 5 min at -40 °C. The mixture
was then filtered through a precooled glass wool plug, and the solution
of 6 was used directly for kinetic experiments or characterization.
Removal of the solvent leads to degradation of 6. 1H NMR (400 MHz,
CD2Cl2, 298 K): δ 7.72 (s, 8H, o-Ar′4B), 7.57 (s, 4H, p-Ar′4B), 3.74
(d, JP-H ) 11.5 Hz, 6H, P(OMe)3), 1.71 (d, JP-H ) 2.9 Hz, 15H,
C5Me5), 1.15 (d, 3JP-H ) 4.8 Hz, 3H, Ir-Me). 31P{1H} NMR (162 MHz,
CD2Cl2, 298 K): δ 87.1 (s). 19F NMR (377 MHz, CD2Cl2, 298 K): δ
-62.9 (s).
3
4
[Cp*(P(OMe)3)Ir(Ph)CD2Cl2][BArf]. Method A: To a -40 °C
solution of 6 (6.4 mg, 0.010 mmol) in CD2Cl2 (0.6 mL) was added
C6H6 (8.1 mg, 0.10 mmol). The solution was allowed to warm to
1
room temperature, and after 20 min H NMR spectroscopy indicated
that complete conversion to [Cp*(P(OMe)3)Ir(Ph)CD2Cl2][BArf] had
occurred. As with compound 6, attempts to isolate [Cp*(P(OMe)3)Ir-
(Ph)CD2Cl2][BArf] were unsuccessful.
Method B: To a solution of Cp*(P(OMe)3)IrPh(OTf) (7.1 mg, 0.010
mmol) in CD2Cl2 (0.5 mL) was added a slurry of NaBArf (9.3 mg,
0.010 mmol) in CD2Cl2 (0.5 mL). The solution was shaken for 10 min
and transferred to a NMR tube. Complete conversion to [Cp*(P(OMe)3)-
Examination of Triflate Dependence for the Reaction of 5 with
Benzene. To a vial containing 5 (8.7 mg, 0.014 mmol) was added a
-40 °C solution of C6H6 (0.10 mL, 1.1 mmol) in CD2Cl2 (0.65 mL,
[Ir] ) 0.019 M, [C6H6] ) 1.5 M). This solution was kept at -40 °C
while a second vial containing 5 (8.3 mg, 0.014 mmol) was charged
with a -40 °C solution of C6H6 (0.10 mL, 1.1 mmol) and PPNOTf
(9.3 mg, 0.014 mmol) in CD2Cl2 (0.65 mL, [Ir] ) 0.018 M, [C6H6] )
1.5 M, [PPNOTf] ) 0.019 M). Upon mixing of this second reaction,
1
Ir(Ph)CD2Cl2][BArf] was observed as determined by H NMR spec-
troscopy. 1H NMR (400 MHz, CD2Cl2, 298 K): δ 7.76 (s, 8H, o-Ar′4B),
3
7.59 (s, 4H, p-Ar′4B), 7.25 (d, JH-H ) 7.6 Hz, 2H, o-C6H5), 7.14 (t,
3
3JH-H ) 7.1 Hz, 2H, m-C6H5), 7.02 (t, JH-H ) 7.2 Hz, 1H, p-C6H5),
3
4
3.73 (d, 6H, JP-H ) 11.3 Hz, P(OMe)3), 1.59 (d, 15H, JP-H ) 2.8
Hz, C5Me5). 31P{1H} NMR (162 MHz, CD2Cl2, 298 K): δ 82.8 (bs).
19F NMR (377 MHz, CD2Cl2, 298 K): δ -62.9 (s).
9
J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1409

A R T I C L E S
Tellers et al.
each solution was transferred to a separate NMR tube and allowed to
warm to room temperature. The reactions were monitored by 1H NMR
spectroscopy until all of 5 had been consumed (8 h). Each reaction
had a t1/2 ) 1.5 h.
B3LYP functional uses Becke’s three-parameter exchange functional72
with the LYP nonlocal correlation functional.73 The LACVP basis set74
uses the Pople 6-31 bases75 for H-Ar. Iridium is represented using the
outermost core orbitals and the Los Alamos effective core potentials.76,77
A single asterisk (*) indicates the addition of polarization functions to
carbon and phosphorus; double asterisks (**) indicate the addition of
polarization functions to hydrogen, carbon, and phosphorus. Likewise,
a single plus (+) indicates the addition of diffuse functions to carbon
and phosphorus. The energies obtained do not take into account zero-
point or Gibbs energies and are therefore reported as free energies,
∆E.
Kinetic Experiments on 1 with 1,2-Disubstituted Arenes. A typical
experiment was performed as follows. In a glovebox, 16.2 mg (0.0284
mmol) of 1 was dissolved in CH2Cl2 and added to a 10 mL volumetric
flask. The solution was diluted to 10.0 mL with CH2Cl2. A volume of
2.00 mL of this solution was transferred to a 20 mL scintillation vial
containing 1.00 mL of a 0.154 M stock solution of C6H4Cl2 in CH2-
Cl2. The solution was stirred briefly to ensure complete mixing and
was then transferred by pipet to a quartz UV-vis cuvette equipped
with a Kontes high-vacuum Teflon stopcock. The cell was removed
from the glovebox and immediately placed at 14.1 °C in the UV-vis
cuvette holder. Spectra were acquired at 200 s intervals over 12 h.
Competition Experiments of 3 or 6 with 1,2-Disubstituted Arenes.
A typical experiment was performed as follows. To 0.5 mL of a CD2-
Cl2 solution of 150 mM 1,2-dichlorobenzene and 150 mM o-xylene
was added 0.5 mL of a freshly prepared solution of 6 in CD2Cl2
(approximately 8.0 mM). The resulting mixture was sealed and allowed
to sit for 5 h at room temperature. The resulting products were analyzed
by 31P{1H} NMR spectroscopy and compared to the spectra of the
corresponding single-substrate reaction products.
Kinetic Experiments on the Reactions of 3 and 6 with Benzene.
A typical experiment was performed as follows. To 0.5 mL of a CD2-
Cl2 solution of 0.26 M benzene and C6Me6 (0.9 mg, 0.006 mmol) at
-40 °C was added 0.5 mL of a freshly prepared solution of 5 in CD2-
Cl2 (approximately 26 mM) at -40 °C. The solution was mixed
thoroughly to yield an orange solution (13 mM 5, 0.13 M C6H6), which
was transferred to an NMR tube. The tube was attached to a Cajon
adapter, degassed on a vacuum line, and flame sealed. The tube was
kept at -196 °C until it was ready to be placed in an NMR probe held
at 30 ( 0.1 °C. Spectra of 6 were aquired at 300 s intervals over 10 h,
and spectra 3 were aquired at 12 s intervals over 15 min.
Acknowledgment. This work was supported by the Director,
Office of Energy Research, Office of Basic Energy Sciences,
Chemical Sciences Division, U.S. Department of Energy,
under Contract No. DE-AC03-76SF00098. We acknowledge
Boehringer-Ingelheim Pharmaceuticals, Inc. for a graduate fellow-
ship (D.M.T.). We would also like to thank Drs. Andrew
Streitwieser, Charles Perrin, Joseph Gajewski, and J. Robin
Fulton for helpful discussions, Dr. Bernd Straub for assistance
with DFT calculations, and Drs. John Bercaw and Jay Labinger
for helpful discussions and disclosure of results prior to
publication.
Supporting Information Available: Tables containing the
kinetic data illustrated in Figures 1, 2, 3, 4, and 5 (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA011809U
(70) OpenPBS, version 2.3.1; Veridian Systems: Mountain View, California,
2000.
(71) Schaftenaar, G.; Noordik, J. H. J. Comput.-Aided Mol. Des. 2000, 14, 123.
(72) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(73) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(74) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
(75) Frisch, M. J.; Pople, J. A.; Binkley, J. S. J. Chem. Phys. 1984, 80, 3265.
(76) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
(77) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 284.
DFT Calculations. Geometries were calculated with density func-
tional theory using the B3LYP functional and the LACVP**+ basis
set in Jaguar.54 Calculations were performed using Jaguar 4.054 on either
Pentium Coppermine or AMD Athlon-based PCs running Redhat Linux
6.2 and openPBS.70 Molden was used for visualization of results.71 The
9
1410 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002