Ortho C - H activation of haloarenes and anisole by an electron-rich iridium(I) complex: Mechanism and origin of regio- and chemoselectivity. An experimental and theoretical study

Article abstract of DOI:10.1021/om060078+

Reaction of (PNP)Ir(COE)⁺PF₆^- (1) (PNP = 2,6-bis(di-tert-butylphosphinomethyl)pyridine; COE = cyclooctene) with benzene yields a stable unsaturated square pyramidal Ir(III) hydrido-aryl complex, 2, which undergoes arene exchange upon reaction with other arenes at 50 °C. Upon reaction of 1 with haloarenes (chlorobenzene and bromobenzene) and anisole at 50 °C, selective ortho C - H activation takes place. No C - halogen bond activation was observed, even in the case of the normally reactive bromobenzene and despite the steric hindrance imposed by the halo substituent. The ortho-activated complexes (8a, 9a, and 10a) exhibited a higher barrier to arene exchange; that is, no exchange took place when heating at a temperature as high as 60 °C. These complexes were more stable, both thermodynamically and kinetically, than the corresponding meta- and para-isomers (8b,c, 9b,c, and 10b,c). The observed selectivity is a result of coordination of the heteroatom to the metal center, which kinetically directs the metal to the ortho C - H bond and stabilizes the resulting complex thermodynamically. Upon reaction of complex 1 with fluorobenzene under the same conditions, no such selectivity was observed, due to low coordination ability of the fluorine substituent. Competition experiments showed that the ortho-activated complexes 8a, 9a, and 10a have similar kinetic stability, while thermodynamically the chloro and methoxy complexes 8a and 10a are more stable than the bromo complex 9a. Computational studies, using the mPW1K exchange - correlation functional and a variety of basis sets for PNP-based systems, provide mechanistic insight. The rate-determining step for the overall C - H activation process of benzene is COE dissociation to form a reactive 14e complex. This is followed by formation of a η²_c-c intermediate, which is converted into an η²_C-H complex, both being important intermediates in the C - H activation process. In the case of chlorobenzene, bromobenzene, and anisole, η¹-coordination via the heteroatom to the 14e species followed by formation of the ortho η²_C-H complex leads to selective activation. The unobserved C - halide activation process was shown computationally in the case of chlorobenzene to involve the same Cl-coordinated intermediate as in the C - H activation process, but it experiences a higher activation barrier. The ortho C - H activation product is also thermodynamically more stable than the C - Cl oxidative addition complex.

Full text of DOI:10.1021/om060078+

3190
Organometallics 2006, 25, 3190-3210
ortho C-H Activation of Haloarenes and Anisole by an
Electron-Rich Iridium(I) Complex: Mechanism and Origin of
Regio- and Chemoselectivity. An Experimental and Theoretical Study
Eyal Ben-Ari, Revital Cohen, Mark Gandelman, Linda J. W. Shimon,
Jan M. L. Martin,* and David Milstein*
Department of Organic Chemistry, The Weizmann Institute of Science, 76100 RehoVot, Israel
ReceiVed January 24, 2006
-
Reaction of (PNP)Ir(COE)+PF6 (1) (PNP ) 2,6-bis(di-tert-butylphosphinomethyl)pyridine; COE )
cyclooctene) with benzene yields a stable unsaturated square pyramidal Ir(III) hydrido-aryl complex, 2,
which undergoes arene exchange upon reaction with other arenes at 50 °C. Upon reaction of 1 with
haloarenes (chlorobenzene and bromobenzene) and anisole at 50 °C, selective ortho C-H activation
takes place. No C-halogen bond activation was observed, even in the case of the normally reactive
bromobenzene and despite the steric hindrance imposed by the halo substituent. The ortho-activated
complexes (8a, 9a, and 10a) exhibited a higher barrier to arene exchange; that is, no exchange took
place when heating at a temperature as high as 60 °C. These complexes were more stable, both
thermodynamically and kinetically, than the corresponding meta- and para-isomers (8b,c, 9b,c, and 10b,c).
The observed selectivity is a result of coordination of the heteroatom to the metal center, which kinetically
directs the metal to the ortho C-H bond and stabilizes the resulting complex thermodynamically. Upon
reaction of complex 1 with fluorobenzene under the same conditions, no such selectivity was observed,
due to low coordination ability of the fluorine substituent. Competition experiments showed that the
ortho-activated complexes 8a, 9a, and 10a have similar kinetic stability, while thermodynamically the
chloro and methoxy complexes 8a and 10a are more stable than the bromo complex 9a. Computational
studies, using the mPW1K exchange-correlation functional and a variety of basis sets for PNP-based
systems, provide mechanistic insight. The rate-determining step for the overall C-H activation process
of benzene is COE dissociation to form a reactive 14e complex. This is followed by formation of a η2
C-C
intermediate, which is converted into an η2C-H complex, both being important intermediates in the C-H
activation process. In the case of chlorobenzene, bromobenzene, and anisole, η1-coordination via the
heteroatom to the 14e species followed by formation of the ortho η2
complex leads to selective
C-H
activation. The unobserved C-halide activation process was shown computationally in the case of
chlorobenzene to involve the same Cl-coordinated intermediate as in the C-H activation process, but it
experiences a higher activation barrier. The ortho C-H activation product is also thermodynamically
more stable than the C-Cl oxidative addition complex.
selective catalytic ortho-hydroarylation of aromatic esters,
ketones, and aldehydes based on metal coordination to the
carbonyl group.2-4 Amine, pyridyl,5 hydroxy,6 and/or other good
coordinating groups have also been utilized. Recently, selective
ortho C-H bond activation of acetophenone and nitrobenzene
by an electron-rich (PCP)Ir complex was reported.7,8 It was
shown in this case that the selectivity is governed by thermo-
dynamics rather than kinetics, i.e., stabilization of the C-H
activation product by coordination of the ortho functional group
to the metal, while coordination prior to C-H activation does
not play a significant role.
Introduction
The activation of strong C-H bonds of hydrocarbons is a
topic of much current interest. Several examples of C-H bond
activation under relatively mild conditions by transition metal
complexes are known, and comprehensive mechanistic studies
of this process have been reported.1 Such studies are important
for the design of efficient processes for the functionalization of
hydrocarbons.
An attractive goal is the selectiVe C-H bond activation of
substituted arenes or alkanes. Murai et al. reported the regio-
* To whom correspondence should be addressed. E-mail:
david.milstein@weizmann.ac.il; comartin@wicc.weizmann.ac.il.
(1) For reviews, see: (a) Lersch, M.; Tilset, M. Chem. ReV. 2005, 105,
2471. (b) Jones, W. D. Inorg. Chem. 2005, 44, 4475. (c) Labinger, J. A,
Bercaw, J. E. Nature 2002, 417, 507. (d) Goldberg, K. I, Goldman, A. S.,
Eds. ActiVation and Functionalization of C-H Bonds; ACS Symposium
Series 885; American Chemical Society: Washington, DC, 2004. (e) Jones
W. D. Acc. Chem. Res. 2003, 36, 140. (f) Slugovc, C.; Padilla-Martinez, I.;
Sirol, S.; Carmona, E. Coord. Chem. ReV. 2001, 213, 129. (g) Crabtree, R.
H. J. Chem. Soc., Dalton Trans. 2001, 2437. (h) Murai, S., Ed. Topics in
Organometallic Chemistry, Vol. 3; Springer: New York, 1999. (i) Jensen,
C. M. Chem. Commun. 1999, 24, 2443. (j) Shilov, A. E.; Shul’pin, G. B.
Chem. ReV. 1997, 97, 2879. (k) Arndtsen, B. A.; Bergman, R. G.; Mobley,
T. A.; Peterson, T. H. Acc. Chem. Res. 1995, 28, 154. (l) Jones, W. D.;
Feher, F. J. Acc. Chem. Res. 1989, 22, 91 (m) Hill, C., Ed. ActiVation and
Functionalization of Alkanes; John Wiley and Sons: New York, 1989. (n)
Crabtree, R. H. Chem. ReV. 1985, 85, 245.
(2) Kakiuchi, F.; Matsumoto, M.; Sonoda, M.; Fukuyama, T.; Chatani,
N.; Murai, S.; Furukawa, N.; Seki, Y. Chem. Lett. 2000, 750.
(3) Kakiuchi, F.; Murai, S. Acc. Chem. Res. 2002, 35, 826, and references
therein.
(4) (a) Kakiuchi, F.; Kan, S.; Igi, K.; Chatani, N.; Murai, S. J. Am. Chem.
Soc. 2003, 125, 1698. (b) Asaumi, T.; Chatani, N.; Matsuo, T.; Kakiuchi,
F.; Murai, S. J. Org. Chem. 2003, 68, 7538.
(5) Kakiuchi, F.; Igi, K.; Matsumoto, M.; Hayamizu, T.; Chatani, N.;
Murai, S. Chem. Lett. 2002, 396.
(6) Dorta, R.; Togni, A. Chem. Commun. 2003, 760
(7) (a) Guari, Y.; Sabo-Etiennes, S; Chaudret, B. J. Am. Chem. Soc. 1998,
120, 4228. (b) Guari, Y.; Castellanos, A.; Sabo-Etiennes, S; Chaudret, B.
J. Mol. Catal. A 2004, 212, 77.
(8) Zhang, X.: Kanzelberger, M.; Emge, J. T.; Goldman, A. S. J. Am.
Chem. Soc. 2004, 126, 13192.
10.1021/om060078+ CCC: $33.50 © 2006 American Chemical Society
Publication on Web 05/17/2006

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3191
Scheme 1
Scheme 2
The chemo- and regioselective activation of strong C-H
bonds in the presence of reactive substituents such as halides,
is intriguing fundamentally and of potential synthetic impor-
tance. While there are a few reports of C-H bond activation
of hydrocarbons in the presence of a C-Cl bond,9-11 in most
cases such chemoselectivity was demonstrated with systems in
which C-Cl oxidative addition is difficult, such as late transition
metal complexes of Ir and Rh having low electron density at
the metal, or early transition metals, where the mechanism
probably involves σ-bond metathesis. Moreover, coordination
of halo-arenes to metal centers through the halide substituent
is known to stabilize electron-poor, high-valent transition metal
species.12 Although examples of C-H activation of haloarenes
by soluble late transition metal complexes were reported,9,13,14
activation of the less hindered meta and para C-H bonds was
observed due to steric reasons. Selective ortho C-H activation
of fluoroarenes is known.15,16 Recently, selective ortho C-H
bond activation of chlorobenzene by a Zr complex was
reported,13 and competitive C-Cl and ortho C-H bond activa-
tion of chlorobenzene in the gas phase by FeO+ is known.9 We
have reported an electron-rich, cationic PNP-type Ir(I) system
that undergoes facile arene C-H bond activation, leading to
unsaturated, stable arene-Ir(III) hydride complexes (Scheme
1).17 This system exhibits chemo- and regioselectivity toward
ortho C-H bond activation in chloro- and bromobenzene.9 No
C-halide oxidative addition was observed. The process was
suggested to be directed by coordination of the metal to the
halide substituent.
Experimental Results
Synthesis and Structure of the Complex (PNP)Ir(COE)+,
1. The electron-rich, cationic, Ir(I) complex 117 was prepared
by addition of an equimolar amount of the PNP ligand (Scheme
2) to a yellow solution of the cationic complex [Ir(I)(acetone)2-
(COE)2]+PF6- 18 in acetone, resulting in a color change to red-
purple. 31P{1H} NMR revealed an AB quartet system at 46 ppm,
indicative of a low-symmetry complex due to the sterically
demanding environment around the metal center. Purple-red
prismatic crystals of 1 were obtained by slow evaporation of
its acetone solution. An X-ray crystallographic study revealed
a distorted square planar geometry19 (Figure 1, Table 1).
Here we report experimental and computational DFT studies
that provide new insight into the C-H bond activation process
and the reasons for the observed selectivity in the activation of
chlorobenzene, bromobenzene, and anisole mediated by (PNP)-
Ir(COE)+PF6- (1) (PNP ) 2,6-bis(di-tert-butylphosphinometh-
yl)pyridine; COE ) cyclooctene).
(9) Competitive ortho C-H and C-Cl bond activation of chlorobenzene
by FeO+ in the gas phase: Bronstrup, M.; Trage, C.; Schro¨der, D.; Schwartz,
H. J. Am. Chem. Soc. 2000, 122, 699.
(10) Vetter, A. J.; Jones, W. D. Polyhedron 2004, 23, 413.
(11) (a) Arndtsen, B. A.; Bergman, R. G. Science 1995, 270, 1970. (b)
Tellers, D. T.; Yung, C. M.; Arndtsen, B. A.; Adamson, D. R.; Bergman,
R. G. J. Am. Chem. Soc. 2002, 124, 1400. (c) Y. Fuchita, Y. Utsunomiya,
M. Yasutake, J. Chem. Soc. Dalton Trans. 2001, 2330. (d) Fan, L.; Parkin,
S.; Ozerov, O. V. J. Am. Chem. Soc. 2005, 127, 16772.
(12) (a) Crabtree, R. H.; Faller, J. W.; Mellea, M. F.; Quirk, J. M.;
Organometallics 1982, 1, 1361. (b) Burk, M. J.; Crabtree, R. H. J.
Organomet. Chem 1988, 341, 495. (c) Kulawiec, R. J.; Crabtree, R. H.;
Coord. Chem. ReV. 1990, 99, 89.
Figure 1. Structure of complex 1 (ORTEP drawing at 50%
probability level; hydrogen atoms and the counteranion are omitted
for clarity). For bond lengths and angles see Table 1.
Table 1. Selected Bond Lengths and Angles for 1
selected bond
bond length (Å)
Ir(1)-P(4)
Ir(1)-P(3)
Ir(1)-N(5)
2.3376(15)
2.3370(13)
2.103(5)
(13) Wu, F.; Dash, A. K.; Jordan, R. F. J. Am. Chem. Soc. 2004, 126,
15360.
selected bond
bond angle (deg)
(14) (a) Cho, J. Y.; Tse, M. K.; Holmes, D.; Maleczka, R. E., Jr.; Smith,
M. R., III. Science 2002, 295, 305. (b) Ishiyama, T.; Takagi, J.; Ishida, K.;
Miyaura, N.; Anastasi, N. R.; Hartwig, J. F. J. Am. Chem. Soc. 2002, 124,
390.
(15) (a) Clot, E.; Oelckers, B.; Hugo, A. K.; Eisenstein, O.; Perutz, R.
N. Dalton Trans. 2003, 4065. (b) Renkema, K. B.; Bosque, R.; Streib, W.;
Maseras, F.; Eisenstein, O.; Caulton, K. G. J. Am. Chem. Soc. 1999, 121,
10895.
N(5)-Ir(1)-P(4)
N(5)-Ir(1)-P(3)
C(1)-Ir(1)-P(3)
C(2)-Ir(1)-P(3)
C(1)-Ir(1)-P(4)
C(2)-Ir(1)-P(4)
79.84(4)
90.2(4)
85.6(3)
94.3(4)
94.4(4)
85.5(6)
Reaction of [Ir(COE)2Cl]2 with sodium tetrakis[3,5-bis(trifluo-
romethyl)phenyl]borate (NaBArF) in acetone followed by ad-
(16) Selmeczy, D. A.; Jones, D. W.; Partridge, G. M.; Perutz, R. N.
Organometallics, 1994, 13, 522.

3192 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.
Scheme 3
Scheme 4
dition of 2 equiv of the PNP ligand resulted in formation of
complex (PNP)Ir(COE)+BArF- in 80% yield (Scheme 2).
The unique reactivities of these complexes toward C-H
activation of benzene, haloarenes, and anisole were investigated
experimentally and theoretically. Both complex 1 and complex
1a were studied, and the same reactivity was observed.
Reaction of (PNP)Ir(COE)+ with Benzene. Heating of
complex 1 at 50 °C for 1 h (or at 25 °C for 48 h) in benzene
resulted in formation of the unsaturated aryl hydride complex
2 (Scheme 1). 31P NMR of this complex revealed a doublet at
Pr3)2(H)(C6H5)(Cl).21 Recently, Goldman reported a closely
related unsaturated complex, (PCP)Ir(H)(C6H5) (PCP ) 2,6-
(tBu2PCH2)2C6H3).22 This thermally unstable complex was
characterized in solution and exhibited fast arene exchange even
at low temperature, which followed a dissociative mechanism.22
Intuitively, it seems that the main difference between these
isoelectronic systems is the electron density at the iridium
centers. Since a lower electron density is expected at the iridium
center of the cationic (PNP)Ir system, that should result in more
facile reductive elimination and lower thermal stability, the
observed higher stability of the PNP system is surprising. It
could be argued that a reductive addition process23 took place,
increasing the electron density on the (PNP)Ir system after
addition of the C-H bond. To address the question of relative
electron density on the cationic Ir(I) and Ir(III) PNP-based
systems (and to compare it to the relevant PCP-based neutral
Ir complex), the corresponding carbonyl complexes were
prepared. Addition of 1 equiv of CO to a DMSO solution of
(PNP)Ir(H)(C6H5) (2) resulted in formation of (PNP)Ir(H)-
(C6H5)(CO) (4) (Scheme 4).
2
54 ppm with JHP ) 13 Hz, characteristic of cis hydrogen-
phosphorus coupling. 1H NMR exhibited a triplet at -44 ppm,
indicative of a hydride located trans to a vacant coordination
site. Orange prismatic crystals were obtained by slow evapora-
tion of a benzene solution of 2. Single-crystal X-ray analysis
revealed a slightly distorted square pyramidal geometry around
the metal center, the hydride being located at the axial site.17
Although complex 2 is coordinatively unsaturated, it exhibited
high stability, showing no decomposition upon heating as a solid
at 100 °C for 24 h. However, when heated at 60 °C in m-xylene,
arene exchange took place, resulting in the m-xylyl complex 3
(Scheme 3), which was synthesized independently by heating
1 in m-xylene at 50 °C (Scheme 1). Furthermore, when complex
2 was heated at 60 °C in C6D6, disappearance of the hydride
ligand and appearance of a deuteride were observed, indicating
formation of (PNP)Ir(D)(C6D5), 2a (Scheme 3).
The 31P{1H} NMR spectrum of 4 revealed a signal at 45 ppm,
1
and H NMR exhibited the hydride ligand as a triplet at -7.5
ppm, representing a large downfield shift compared to the
corresponding hydride signal of complex 2 (-44 ppm). Such a
shift indicates coordination of the CO trans to the hydride ligand.
In addition, 13C{1H} NMR revealed a characteristic signal at
183 ppm for the Ir-CO bond.
Complex 2 represents a rare example of a thermally stable
coordinatively unsaturated M(III) d6 hydrido-aryl complex. Such
species are proposed as reactive intermediates in several
transition metal-catalyzed processes.3,6,20 Werner reported the
only example of a crystallographically characterized, thermally
stable, unsaturated Ir(H)(R) (R ) aryl, alkyl) complex, Ir(Pi-
A comparison of the CO absorption frequency of (PCP)Ir-
(CO)(H)(C6H5)22 (1973 cm-1) with that of (PNP)Ir(H)(C6H5)-
(CO) (4) (2005 cm-1) indicates that the electron density at the
cationic Ir(III) center in 4 is lower, as expected. Moreover, the
CO stretching frequencies of (PNP)Ir(CO) 5 (1964 cm-1) and
4 (2005 cm-1) indicate a significantly lower electron density
on the Ir(III) center of 4. This indicates that C-H oxidative
addition rather then reductive addition took place. Complex 5
(17) Ben-Ari, E.; Gandelman, M.; Rozenberg, H.; Shimon, L. J. W.;
Milstein, D. J. Am. Chem. Soc. 2003, 125, 4714.
(18) Preparation of this complex in solution was reported: Bosch, M.;
Werner, H. Eur. J. Inorg. Chem. 2001, 3181. Recently we isolated it as a
yellow solid: Dorta, R.; Rozenberg, H.; Shimon, L. J. W.; Milstein, D. J.
Am. Chem. Soc. 2002, 124, 188.
(19) Due to disorder of the cyclooctene ligand, it was impossible to
determine its bond lengths.
(20) (a) Doughty, D. H.; Pignolet, L. H. J. Am. Chem. Soc. 1978, 100,
7083.
(21) Werner, H.; Ho¨hn, A.; Dziallas, M. Angew. Chem., Int. Ed. Engl.
1986, 12, 1090.
(22) Kanzelberger, M.; Singh, B.; Czerw, M.; Krogh-Jespersen, K.;
Goldman, A. S. J. Am. Chem. Soc. 2000, 122, 11017.
(23) Crabtree, R. H.; Quirk, J. M. J. Organomet. Chem. 1980, 199, 99.

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3193
Scheme 5
was prepared by the facile reaction of complex 1 with 1 equiv
of CO (Scheme 4).
indicates that there is very little back-donation from the metal
to the nitrile group.
Interestingly, elimination of benzene from complex 4 occurred
only when it was heated to a temperature as high as 110 °C for
48 h, resulting in quantitative formation of (PNP)Ir(CO), 5
(Scheme 4). The octahedral (PNP)Ir(H)(C6H5)(CO) complex,
4, like Goldman’s (PCP)Ir(H)(C6H5)(CO) complex, is kinetically
stabilized toward reductive elimination of benzene, which is
expected to involve ligand loss.24 Complex 5 did not exhibit
C-H oxidative addition reactivity, as no reaction took place
when it was heated in anisole at 110 °C. Noteworthy, addition
of acetonitrile to complex 2 resulted in the formation of complex
6, which was crystallographically characterized, exhibiting an
octahedral structure with the nitrile ligand coordinated trans to
the hydride ligand in an end-on fashion (Figure 2, Table 2). A
comparison of the C-N triple bond length of free acetonitrile
(1.158 Å)25 with the C(2)-N(1) bond length of 1.134 Å
Complex 6 does not eliminate benzene upon heating in C6D6
even at 70 °C. However, at higher temperature it exhibits arene
exchange (Scheme 5).
The kinetic isotope effect of the C-H activation process of
benzene was determined by dissolving (PNP)Ir(COE)+PF6- (1)
in a 1:1 molar ratio of C6H6:C6D6 and leaving the solution at
25 °C for 48 h under kinetic control,26 resulting in a kH/kD
)
1.0. The lack of kinetic isotope effect indicates that C-H
cleavage occurs after the rate-determining step. According to
DFT calculations (vide infra), the reaction pathway involves
an early transition state where the rate-determining step is
dissociation of the cyclooctene ligand to form a 14e (PNP)Ir
intermediate. Conversion of an η2
benzene complex to an
C-C
agostic η2
complex involves the second highest barrier,
C-H
followed by the C-H cleavage step (vide infra).
Reaction of (PNP)Ir(COE)+ (1) with Aryl Halides. When
complex 1 was heated in fluorobenzene at 50 °C for 1 h, three
regioisomers of (PNP)Ir(H)(C6H4F)+PF6- (7) were formed. 31P-
{1H} NMR revealed signals at 55.2, 58, and 52.3 ppm in a ratio
of 2:2:1, assigned to 7a, 7b, and 7c, respectively (vide infra).
1H NMR revealed three signals in the hydride region in the
same ratio at -41, -45, and -42.5 ppm, respectively. An NOE
experiment was carried out on a mixture of ortho- and para-
isomers (obtained as described below), to assign the observed
hydrides and the ortho-aryl proton resonances to isomers 7a
and 7c. Irradiation of the signal at -41 ppm resulted in NOE
enhancement of a signal at 7.10 ppm, while irradiation of the
signal at -42.5 ppm gave NOE enhancement of the signal at
7.25 ppm corresponding to two protons. Together with H-H
correlation spectroscopy (COSY), the signals at -41, -45, and
-42.5 ppm were assigned as the ortho, meta, and para C-H-
activated fluorobenzene complexes 7a, 7b, and 7c, respectively.
Thus, the C-H activation process of fluorobenzene is not
selective. Prolonged heating for 48 h at 60 °C resulted in a
mixture of the ortho-activated and the para-activated fluoroben-
zene complexes 7a and 7c in ratio of 1.8:1, respectively. Heating
at 70 °C or higher for 48 h resulted in a similar mixture of o-
and p-isomers in a ratio of 2.3:1, respectively (Scheme 6).
Figure 2. Structure of 6 (ORTEP drawing at 50% probability level;
hydrogen atoms and counteranion are omitted for clarity). For
selected bond lengths and angles see Table 2.
Table 2. Selected Bond Lengths and Angles for 6
selected bond
bond length (Å)
Ir(1)-C(21)
Ir(1)-P(3)
Ir(1)-P(2)
Ir(1)-N(1)
Ir(1)-N(4)
C(2)-N(1)
C(2)-C(3)
2.072(2)
2.3291(9)
2.3304(9)
2.153(2)
2.157(2)
1.134(1)
1.46(2)
Selective C-H activation of fluorobenzene at the ortho-
position was reported and was studied theoretically as well as
experimentally.15,16,27 The ortho-isomer was shown to be the
thermodynamically most stable one in the case of a Re
complex,15a and the kinetically and thermodynamically most
stable in the case of Os complex.15b Moreover, it is known that
the ortho-positions in haloarenes are the most acidic ones,
especially in fluoroarenes.28
selected bond
bond angle (deg)
N(1)-Ir(1)-P(3)
N(1)-Ir(1)-C(21)
N(4)-Ir(1)-C(21)
N(4)-Ir(1)-N(1)
N(1)-C(2)-C(3)
C(21)-Ir(1)-P(3)
97.78(6)
95.83(3)
178.11(8)
82.44(9)
178.5(3)
98.25(7)
(25) Handbook of Chemistry and Physics, 57th ed.; CRC Press: Cleve-
land, OH, 1976.
(26) By kinetic control we mean that there is no significant exchange
with benzene in solution under these conditions.
(24) (a) Milstein, D. J. Am. Chem. Soc. 1982, 104, 5227. (b) Milstein,
D. Acc. Chem. Res. 1984, 17, 221. (c) Basato, M.; Longato, B.; Morandini,
F.; Bresadola, S. Inorg. Chem. 1984, 23, 3972. (d) Rosini, G. P.; Wang,
K.; Patel, B.; Goldman, A. S. Inorg. Chim. Acta 1998, 270, 537
(27) Clot, E.; Besora, M.; Maseras, F.; Me´gret, C.; Eisenstein, O.;
Oelckers, B.; Perutz, R. N. Chem. Commun. 2003, 490.
(28) Schlosser, M.; Marzi, E.; Cottet, F.; Bu¨ker, H. H.; Nibbering, N.
M. Chem. Eur. J. 2001, 7, 3511.

3194 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.
Scheme 6
Scheme 7
Scheme 8
While complex 1 exhibited no kinetic preference, and low
thermodynamic preference, for the ortho C-H bond of fluo-
robenzene, ortho C-H selectivity was observed with chlo-
robenzene. Thus, heating complex 1 in chlorobenzene at 50 °C
for 1 h resulted in the o-, m-, and p-isomers of (PNP)Ir(H)-
(C6H4Cl) complexes in a ratio of 4.6:2:1, assigned as 8a-c,
respectively (vide infra), at a point where 1 was fully consumed
(Scheme 7). 31P{1H} NMR revealed a signal at 51 ppm
corresponding to the ortho-isomer and a broad signal at 54 ppm
corresponding to the meta- and para-isomers. 1H NMR revealed
a hydride signal at -34 ppm, which is characteristic of M-H
trans to an occupied coordination site (ortho-isomer, see below),
and at -41 and -44 ppm for the meta- and para-isomers.
Irradiating the hydride signal in 1H NMR at -41 ppm gave an
NOE enhancement (relative to the aryl protons) at 7.35 ppm.
Irradiating the signal at -44 ppm resulted in an NOE enhance-
ment of the aryl protons at 7.45 ppm. Together with COSY
data, the signals at -41 and -44 ppm were assigned to the
hydride ligands of the meta and para C-H-activated chloroben-
zene complexes 8b and 8c, respectively. Monitoring the reaction
by NMR revealed at 10% conversion a ratio 8a:8b:8c of 4:2:1,
indicating that complex 1 has a kinetic preference for the
activation of the C-H bond ortho to the chlorine atom.
Remarkably, when this reaction mixture was heated for 48 h,
complex 1 was totally consumed and complexes 8b and 8c were
fully conVerted to the ortho-activated complex 8a (Scheme 7),
indicating that this complex is the most kinetically and
thermodynamically favored product under these conditions.
Yellow prismatic crystals were obtained by slow diffusion
of pentane into a chlorobenzene solution of complex 8a. An
X-ray analysis of 8a reveals that the Ir atom is located in the
center of a distorted octahedron.29 The Cl atom is coordinated
to the metal, and as predicted from the NMR data, it is located
trans to the hydride. Although the Ir-Cl bond length of 2.816
Å is significantly shorter than the sum of the van der Waals
radii of Ir and Cl atoms, it is longer than other reported Ir-Cl
bonds in coordinated chloroalkanes.12c This elongation is
probably due to some strain in the four-membered ring, as
indicated by the Ir-Cl-C angle of 72°. Noteworthy, the C-Cl
bond (1.78 Å) is significantly elongated compared to the one
in free C6H5Cl (1.700 Å),25 suggesting π-back-donation from a
filled metal d-orbital to the empty C-Cl π*-orbital.
Interestingly, complex 8a is the most stable isomer despite
the steric hindrance imposed by the chlorine atom. The
intramolecular coordination of the Cl atom of activated chlo-
robenzene can explain the higher thermodynamic stability of
8a versus the other isomers 8b and 8c, in which such
coordination is impossible. Halocarbon coordination to transition
metals is well documented,12,30 although examples of chloro-
and bromoarene coordination are not common.30 Remarkably,
reaction of 1 with bromobenzene is similar to that of chlo-
robenezene, despite the higher steric hindrance imposed by the
Br atom and the expected high reactivity of Ar-Br toward
oxidative addition to electron-rich Ir(I) (Scheme 8).
31P{1H} NMR follow-up at 50 °C revealed a signal corre-
sponding to the ortho C-H activation complex 9a at 49 ppm,
in addition to a broad signal at 51-51.5 ppm, corresponding to
the m- and p-activated complexes 9b,c. 1H NMR revealed three
hydride signals at -29, -42, and -45 ppm, corresponding to
(29) For detailed crystallographic information see the Supporting Infor-
mation of ref 17.
(30) (a) Butts, M. D.; Scott, B. L.; Kubas, G. J. J. Am. Chem. Soc. 1996,
118, 1970. η1-Coordination of chlorobenzene: (b) Korolev, A. V.; Delpech,
F.; Dagorne, S.; Guzei, I. A.; Jordan, R. F. Organometallics 2001, 20, 3367.
(c) Kowalczyk, J. J.; Agbossou, S. K.; Gladysz, J. A. J. Organomet. Chem.
1990, 397, 333.

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3195
Scheme 9
the ortho-, meta-, and para-isomers in a ratio of 7:2:1,
respectively, at 10% conversion. Irradiating the signal at -42
ppm resulted in NOE enhancement between the hydride and
aryl protons at 7.56 ppm. Irradiating the signal at -45 ppm
gave an NOE enhancement between the hydride and aryl proton
at 7.52 ppm. Together with COSY results, the signals at -42
and -45 ppm were assigned to the hydride ligands of the meta
and para C-H-activated bromobenzene complexes 9b and 9c,
respectively. Continued heating at 50 °C for 48 h resulted in
quantitative formation of 9a as the sole product (Scheme 8).
These results show that 9a is both kinetically and thermody-
namically favored.
Scheme 10
Slow diffusion of pentane into an anisole solution of 10a resulted
in formation of orange prismatic crystals, which were analyzed
by X-ray diffraction (Figure 3). Although there is no indication
of strong coordination of the oxygen atom to the metal center
in 10a in solution (only -0.22 ppm shift in the 1H NMR of the
OMe group as compared with anisole), there is an indication
for directionality toward the metal center in the solid state. Thus,
the angles C(8)-C(1)-Ir(1) and C(2)-C(1)-Ir(1) (131.12° and
112.5°, respectively) indicate bending toward the vacant site
trans to the hydride. This bending is negligible in complex 2,
but very significant in complexes 8a, 9a, and 10a. Moreover,
the C(2)-O(3) bond length in the crystal structure (1.397 Å) is
longer than the Ar-O bond length of free anisole (1.357 Å),31
implying interaction of the methoxy group with the metal center.
Exchange Reactions. To compare the regioselective ortho
C-H activation in chlorobenzene, bromobenzene, and anisole,
we have performed competition experiments under conditions
in which there is no exchange between the ortho C-H-activated
complexes 8a, 9a, and 10a. When complexes 8a, 9a, and 10a
were dissolved, each one separately, in a mixture of anisole
and chlorobenzene (1:1 molar ratio) and heated at 55 °C for 24
h, no reaction took place, indicating that the complexes are
kinetically stable under these conditions.
Reaction of (PNP)Ir(COE)+ with Anisole. When 1 was
heated in anisole at 50 °C for 1 h, the ortho, meta, and para
C-H-activated anisole complexes 10a-c were formed. Com-
plex 10a was the major product (93% of the resulting mixture).
Continued heating for 24 h at the same temperature resulted in
the sole formation of the ortho-isomer 10a (Scheme 9). Complex
10a gives rise to a doublet at 56.60 ppm in the 31P{1H} NMR
Heating the Ir(I) complex 1 in a mixture of chlorobenzene,
bromobenzene, and anisole (1:1:1 molar ratio) at 55 °C for 24
h resulted in a mixture of ortho-activated products in a ratio of
38% 10a, 32% 8a, and 30% 9a, indicating only a slight
preference toward the ortho C-H bond of anisole as compared
with haloarenes. No Hammett correlation based on these ratios
(which correspond to the relative rates of formation of these
complexes) and the σ and the σ+ constants was found, implying
that the overall C-H activation process is not governed by
electronic influence. Interestingly, upon heating the ortho-
activated fluorobenzene complex 7a at 55 °C for 1.5 h in a
mixture of anisole and chlorobenzene at a molar ratio of 1:1,
exchange took place, resulting in a mixture of complexes 8a,
10a, 8b,c, and 10b,c, after 45% conversion with respect to 7a.
Heating complexes 8a, 9a, and 10a (1:1:1 molar ratios) in a
mixture of anisole:chlorobenzene:bromobenzene (molar ratio of
1:1:1) under thermodynamically controlled conditions, i.e., 70
°C for 5 days,32 a new mixture of the o-isomers was formed.
Figure 3. Structure of 10a (ORTEP drawing, 50% probability
level; hydrogen atoms and counteranion are omitted for clarity).
For selected bond lengths and angles see Table 3.
Table 3. Selected Bond Lengths and Angles for 10a
selected bond
bond angle (deg)
selected bond
bond length (Å)
N(2)-Ir(1)-P(4)
C(8)-C(1)-Ir(1)
P(4)-Ir(1)-P(3)
P(3)-Ir(1)-H(1)
C(1)-Ir(1)-P(4)
N(2)-Ir(1)-P(3)
N(2)-Ir(1)-C(1)
N(2)-Ir(1)-H(1)
C(2)-O(3)-C(4)
O(3)-C(2)-C(1)
C(5)-C(2)-C(1)
C(5)-C(2)-O(3)
C(2)-C(1)-Ir(1)
C(8)-C(1)-Ir(1)
C(8)-C(1)-C(2)
P(4)-Ir(1)-H(1)
C(1)-Ir(1)-H(1)
C(1)-Ir(1)-P(3)
81.99(10)
131.6(4)
164.04(4)
90.8(14)
96.55(13)
83.13(11)
177.7(17)
79.5(13)
116.7(4)
111.0(4)
122.6(5)
126.3(5)
112.5(3)
131.1(2)
115.9(4)
80.9(14)
102.1(13)
98.48(13)
Ir(1)-H(1)
Ir(1)-C(1)
Ir(1)-P(3)
Ir(1)-P(4)
Ir(1)-N(2)
C(2)-O(3)
O(3)-C(4)
O(3)-Ir(1)
1.55(4)
2.038(4)
2.3380(13)
2.3149(13)
2.139(4)
1.397(6)
1.425(6)
2.76(2)
(31) Seip, H. M.; Seip, R. Acta Chem. Scand. 1973, 27, 4024.
(32) The ratios were observed after 4 days at 70 °C and were constant
for the next 24 h. Moreover, a small amount of C-H activation products
resulting from the methoxy group of anisole were observed (23% from the
total mixture after 5 days).
1
spectrum and a triplet at -42 ppm in the H NMR spectrum.

3196 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.
Figure 4. Relative free energies [kcal/mol] for the mechanism of C-H activation of benzene and associative benzene exchange (with
benzene) by the (Me-PNP)Ir+ system at the mPW1k/SDD//mPW1k/SDB-cc-pVDZ level of theory.
31P{1H} NMR revealed three signals at 56, 52, and 49 ppm,
corresponding to complexes 10a, 8a, and 9a, respectively, in
39%, 36%, and 25%, respectively (Scheme 10). This implies
that the anisole- and chlorobenzene-activated complexes 8a and
10a are the most stable thermodynamically, with a slight
preference to the anisole complex 10a.
32 kcal/mol. No transition state was found for the ligand
dissociation process. An associative mechanism to generate the
η2-complex from the (PNP)Ir(COE)+ (C1) cannot be ruled
out: however, no transition state was located for an associative
mechanism. After the T-shaped complex C2 is formed, facile
coordination of benzene takes place to form the relatively stable
η2
intermediate, C3. The η2
arene can easily exchange
C-C
C-C
Theoretical Results
positions along the aromatic ring with a barrier of 7.8 kcal/
mol; however, in the case of benzene this process is degenerate,
leading to the same η2-arene complex, C3. To activate the arene
C-H bond, complex C2 should first be converted to the aryl
To gain further insight into the mechanism of the aryl C-H
activation process in the (PNP)Ir(COE)+PF6- (1) system, DFT
calculations were performed. Geometry optimizations were
carried out at the mPW1k/SDD level of theory, and on the
optimized structures single-point energy calculation at the
mPW1k/SDB-cc-pVDZ level of theory were performed, to
increase the accuracy of the results. First, calculations were
carried out on the H-PNP(COE)+ system, with no substituents
on the phosphine arms, and then, Me substituents were added
to the phosphines of the optimized structures in order to simulate
better the electronic and steric effects of the realistic system.
Finally, where necessary to increase the accuracy of the results,
tBu substituents were added and geometry optimization were
carried out on the full system at the (mPW1k/LANL2DZ+P//
ONIOM(mPW1k/LANL2DZ:mPW1k/LANL1MB)) level of
theory. In the calculations we considered three systems: ben-
zene, chlorobenzene, and anisole (PNP)Ir+ systems.
η2
product, C4, with a barrier of 10.4 kcal/mol, following
C-H
by C-H bond activation to form C5.33 No C-H activation was
found to take place directly from the arene η2C-C complex C3.
The C-H activation product, C5, is square pyramidal with the
hydride ligand trans to an empty coordination site, as in the
experimental X-ray structure. It is the thermodynamically most
stable product in the system, which can be generated in a
stepwise manner from the (PNP)Ir(COE)+ complex C1 with
relatively low barriers. The rate-limiting step in this process
was found to be not the C-H activation step but rather the
dissociation of the COE ligand from the initial complex to form
the highly unstable (PNP)Ir+ T-shaped complex C2. From the
computational model suggested here it appears that the (PNP)-
Ir(COE)+ complex C1 is more stable than the C-H activation
product, (PNP)Ir(Ph)(H)+, C5. However, it was shown experi-
mentally that the tBu analogue of complex C5 is the thermo-
dynamically most stable product of the reaction of (PNP)Ir-
(COE)+ with benzene. Thus, to verify the reason for the
discrepancy between experimental and computational results,
tBu substituents on the phosphines were added to the compu-
tational model, and geometry optimizations were carried out
for the realistic (tBu-PNP)Ir+ system using the ONIOM two-
layer method. Consequently, the results from the ONIOM
calculations revealed that the cationic (tBu-PNP)Ir(Ph)(H)+
C-H Activation of Benzene by Me-PNP/Ir+. We describe
here the mechanism of arene C-H activation by the (Me-PNP)-
Ir+ system (a similar mechanism was found for the (H-PNP)-
Ir+ system; however, lower barriers were found for the Me
system, which is more closely related to the experimental system
both electronically and sterically). C-H activation in benzene
is also a reference point for further description of C-H
activation in chlorobenzene and anisole cases. Both a η2
C-H
intermediate and η2-arene intermediates are implied. The energy
profile for the mechanism found for C-H activation in benzene
by the (PNP)Ir+ system is described in Figure 4.
Only one reaction sequence leading to the C-H-activated
(Me-PNP)Ir(H)(Ph)+ complex C5 was found (Figure 4). The
stepwise arene C-H bond activation starts with an initial
dissociation of cyclooctene (COE) to form the T-shaped
complex C2. Dissociation of COE was found to be the rate-
determining step of the reaction, with a dissociation energy of
(33) The transition state leading to C5 appears to rest below the
intermediate, C4 (∆Gq ) -0.4 kcal/mol). It should be noted that using the
mPW1K/SDD-optimized geometry for the mPW1K/SDB-cc-pVDZ energy
profile is an approximation that can lead to errors, especially in cases of
small energy differences. Furthermore, the correction to the energy used to
calculate ∆G298 is based upon the rigid rotor harmonic oscillator approxima-
tion.

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3197
Table 4. Comparison between Selected Bond Lengths (Å)
and Angles (deg) of the Calculated and Experimental
(X-ray) tBu-PNP/Ir(Ph)(H) Structures
experimental
calculated
Ir-H
1.66
2.043
2.142
2.32
88
94
178
166
1.53
2.027
2.15
2.36
92
91
177
167
Ir-CPh
Ir-N
Ir-P
CPh-Ir-H
N-Ir-H
N-Ir-CPh
P-Ir-P
complex is more stable than the cationic (tBu-PNP)Ir(COE)+
complex by 8.8 kcal/mol.
Structure of the C-H-Activated Complex 4. The calculated
structure of the (tBu-PNP)Ir(Ph)(H)+ complex 4 (tBu derivative
of C5) (at the mPW1k/LANL2DZ+P//ONIOM(mPW1k/LAN-
L2DZ:mPW1k/LANL1MB level of theory) is in good agreement
with the experimental X-ray structure. Comparison between
selected bond lengths and angles is presented in Table 4.
Comparison between PNP- and PCP-Ir(Ph)(H) Systems.
To address the question of electron density on the Ir(III)
complexes, the PCP and PNP complexes (Me-PCP)Ir(Ph)(H)
and (Me-PNP)Ir(Ph)(H)+ were calculated and compared for
charge densities and bond lengths (see results in Figure 5). It
can be seen that the Ir-H bond distance is identical in both
systems (1.53 Å). Additionally, the natural charges on both
hydrides (+0.13 for both systems) and Ir centers (+0.09 for
the PNP and +0.05 for the PCP) are comparable. Thus, the
nature of the H ligand is similar in the two systems (i.e.,
hydridic), and the complex is Ir(III) both for PNP and PCP
systems. However, when comparing the Ir-Ph bonds, it can
be seen that in the PCP system the Ir-CPh bond length is 0.06
Å longer (2.09 Å) than in the PNP system (2.03 Å). Similarly,
the Cipso-Ir bond in the (PCP)Ir system is 0.06 Å shorter than
the analogous N-Ir in the (PNP)Ir+ system. Similar results were
obtained when comparing the realistic (tBu-PCP)Ir and (tBu-
PNP)Ir+ systems (Figure 5). Similar bond lengths and angles
for the (PCP)Ir system were obtained by Krogh-Jespersen et
al.46 in calculations using the B3LYP functional. Thus, it can
be suggested that in both systems the C-H activation reaction
Figure 5. Comparison between selected NPA (natural population
analysis) charges and bond lengths (Å) of the (PNP)Ir(Ph)(H)+ and
the (PCP)Ir(Ph)(H)+ systems (Me and tBu substituents on the
phosphines).
is oxidative, leading to the formation of Ir(III) complexes. The
trans effect of the PCP-aryl groups is probably responsible for
the elongation and weakening of the Ir-Cipso bond trans to it.
This effect lowers the overall stability of the (PCP)Ir(H)Ph
complex in comparison to the (PNP)Ir(H)Ph+ system.
C-H Activation of Chlorobenzene by (Me-PNP)Ir+. The
C-H activation process of chlorobenzene is expected to have
a mechanism similar to the one found for benzene C-H
activation. However, in contrast to benzene, in chlorobenzene
there are three choices for C-H activation, leading to the
formation of three metal aryl complexes, ortho, para, and meta
to Cl. There is also a possibility of C-Cl activation in this
system. The experimental work shows a remarkable preference
for the ortho-Cl position C-H activation both kinetically and
thermodynamically. The kinetic reason for this discrimination
is suggested to be precoordination of the halide to the metal
center, directing the activation toward the ortho-position of the
arene. The thermodynamic preference is, at least partially, a
result of stabilization of the ortho C-H-activated complex by
additional coordination of the halide to the metal center, as
observed in the X-ray structure of the product.17 DFT calcula-
tions were performed in order to address both the kinetic and
the thermodynamic issues. The analysis of the computational
(34) The six different η2-arene complexes (11, 9, 18, and 10) can be
viewed as three pairs of enantiomers (18 and 10; 9 (ortho); 11 (meta)). In
the case of meta (11) and ortho (9), each one of the η2-arene complexes
can lead to two different η2C-H complexes since the metal center can equally
reach both C-H bonds of the two coordinated carbons. To avoid
complication, we chose all the η2-complexes to be different (by giving them
different numbers) and each one to lead to only one η2
complex (with
C-H
complex).
η2-arene complex 10 not leading to any η2
C-H
(35) TSs for association and dissociation of ligands are generally difficult
to locate on the bottom-of-the-well energy surface, as the maximum on the
finite-temperature free energy surface is often purely entropic in origin,
unless significant geometric rearrangement is involved.
(36) C-H activation in the methoxy group is beyond the scope of this
study.
(37) No transition state was located for this rotation around the Ir-C
bond; however, it is expected to be low in energy. The experimental results
support the assumption that this rotation barrier is low since only one meta
and one ortho C-H activation complex were observed by NMR, suggesting
that the rotation between the isomers is facile.
(38) Hall, C.; Perutz, R. N. Chem. ReV. 1996, 96, 3125.
(39) (a) Lavin, M.; Holt, E. M.; Crabtree, R. H. Organometallics 1989,
8, 99. (b) Toner, A. J.; Grundemann, S.; Clot, E.; Limbach, H. H.;
Donnadieu, B.; Sabo-Ettienn, S.; Chaudret, B. J. Am. Chem. Soc. 2000,
122, 6777.
(40) Vigalok, A.; Uzan, O.; Shimon, L. W. J.; Ben-David, Y.; Martin,
J. M. L.; Milstein, D. J. Am. Chem. Soc. 1998, 120, 12539.
(41) Rybtchinski, B.; Cohen, R.; Martin, J. M. L.; Milstein, D. J. Am.
Chem. Soc. 2003, 125, 11041.
(43) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry; University Science
Books: Mill Valley, CA, 1987.
(44) Zhu, K. M.; Achord, P. D.; Zhang, X. W.; Krogh-Jespersen, X.
W.; Goldman, A. S. J. Am. Chem. Soc. 2004, 126, 13044.
(45) Since the associative mechanism for arene exchange is affected by
the steric hindrance of the ligands, it is anticipated that for the realistic
system (tBu-PNP/Ir) this barrier will be even higher in energy than the one
for the computed Me-PNP/Ir system.
(42) Rybtchinski, B.; Oevers, S.; Montag, M.; Vigalok, A.; Rozenberg,
H.; Martin, J. M. L.; Milstein, D. J. Am. Chem. Soc. 2001, 123, 9064.
(46) Krogh-Jespersen, K.; Czerw, M.; Kanzelberger, M.; Goldman, A.
S. J. Chem. Inf. Comput. Sci. 2001, 41, 56.

3198 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3199
Figure 6. Detailed graphs, each representing a distinct pathway for the formation of the different C-H activation products (ortho, meta,
or para conformers) including an interchange pathway leading from one η2-arene complex to a different η2-arene one.
results describes the mechanism of C-H activation in chlo-
robenzene taking into account all the possible isomers of the
C-H-activated complexes and intermediates involved. On the
basis of the mechanistic study we interpret the experimental
results also taking the possibility of C-Cl activation into
account. As in the benzene system, geometry optimizations were
first carried out on the (H-PNP)Ir(H)(C6H4Cl)+ system at the
mPW1k/SDD level of theory, then Me substituents were added
to the optimized geometries and the structures were reoptimized.
On these optimized geometries, single-point energies were
calculated at the mPW1k/SDB-cc-pVDZ level of theory in order
to increase the accuracy of the results. The mechanism of C-H
activation in chlorobenzene mediated by the (Me-PNP)Ir+
system was found to be similar to the one in benzene. However,
it should be noted that in addition to the different isomers of
C-H-activated complexes (two ortho C-H isomers, two meta
C-H isomers, and one para C-H isomer) obtained in the C-H
activation process of chlorobenzene, it involves also different
isomers for the intermediates. Thus, there are five isomers for
η2
complexes, four isomers for η2-arene complexes,34 and
C-H
various transition states connecting between them. Consequently,
five pathways were found for C-H activation taking place from
η2-complexes, one pathway found for C-H activation taking

3200 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.
place from a Cl-coordinated complex, and one pathway was
found for C-Cl activation. Detailed graphs are given in Figure
6.
further to give eventually the endo-ortho-C-H-activated com-
plex C20; or (c) formation of the ortho-η2
complex, C7,
C-H
with a lower barrier of ∆Gq ) 16 kcal/mol, leading directly to
the most stable ortho-C-H activation complex C8.
The C-H activation of chlorobenzene proceeds, as in benzene
C-H activation, through the formation of a η2
complex,
Since the pathway leading to the C-Cl activation complex
has the highest barrier of all three, it is the most improbable
one to occur. Therefore, as the Cl-coordinated complex C6 is
formed, it will most likely follow the C-H activation pathway
rather than the C-Cl activation one, supporting the experimental
results. In the (Me-PNP)Ir+ system there is no significant
thermodynamic preference for formation of the Cl-coordinated
complex C6 over formation of other Ir(I) arene complexes,
especially the η2-arene species, which have stability comparable
to that of C6. Nevertheless, it is expected that as the bulk around
the metal center increases, the more sterically demanding η2-
arene intermediates will become less stable with respect to the
less sterically hindered Cl-coordinated intermediate. To verify
this postulation, the stability of the Cl-coordinated complex C6
with respect to the η2-arene intermediates should be calculated
for the realistic system. Thus, ONIOM calculations of the tBu
analogues of complexes C6 and C9 were performed. Conse-
quently, it was found that for the (tBu-PNP)Ir+ system the Cl-
coordinated complex is more stable than the η2-arene one (the
analogue of the most stable η2-arene for the (Me-PNP)Ir+ model
system, complex C9) by 12.7 kcal/mol. Thus, the Cl-coordinated
complex in the realistic system is substantially more stable than
the rest of the Ir(I) intermediates, probably due to reduced steric
interference in the case of Cl coordination, in comparison to
the η2-arene complex. Therefore it is the most probable one to
be formed in higher concentrations in the experimental system.
As a result, the lowest barrier pathway taking place from the
Cl-coordinated complex C6 leading to the ortho-η2C-H complex
C7 might explain also the kinetic preference observed experi-
mentally for the ortho-C-H complex C8. Thus, when the Cl-
coordinated complex is formed, it might lead to the formation
of the ortho C-H-activated complex. Therefore, coordination
of chlorobenzene to the metal center via the halide may result
in the selective ortho C-H bond oxidative addition with no
formation of the C-Cl activation product during the reaction.
No TS was located for the formation of the Cl-coordinated
complex from the T-shaped (PNP)Ir+ (or from any other
intermediate), but it is expected to have a lower barrier than
for the formation of an η2-arene complex since the product is
less sterically hindered and less geometrically constrained.35
C-H
leading to the corresponding C-H-activated complex. However,
since there are five different η2C-H complexes, each of the five
could be obtained from a distinct η2-arene complex or, in the
case of ortho-η2C-H complex C7, also from the Cl-coordinated
complex C6. In contrast to the degenerate η2-arene site
interchange process in benzene, here each site interchange (out
of six different ones) leads eventually to a different C-H
activation isomer, since there is only one η2
complex that
C-H
could be formed from each η2-arene complex.34 The η2-arene
complexes are directing the metal center toward the C-H bond
to be cleaved, since precoordination of the arene is necessary
to bring the C-H bond into close proximity to the metal center.
However, where chlorobenzene is concerned, the Cl-coordinated
complex C6 could also play a role in predirecting to C-H
activation in the ortho-position of chlorobenzene (vide infra).
It can be clearly seen from Figure 6 that both ortho C-H
activation products (C8 and C20) are the thermodynamically
most stable products and that the exo-isomer complex C8, with
the Cl moiety coordinated to the metal center, is substantially
more stable than the rest of the isomers (by 8-12 kcal/mol),
explaining the experimentally observed thermodynamics prefer-
ence of the ortho-Cl isomer. The rotation around the Ir-C bond
to convert between the two computationally found ortho-isomers
C8 and C20 is expected to have a very low barrier, and no
transition state was located for this rotation. The experimental
results support the assumption that this rotation barrier is low
since only one meta and one ortho C-H activation complex
were observed by NMR, suggesting that the rotation between
the isomers is facile. In general, the more stable isomer is
observed.
We considered also a possibility of C-Cl oxidative addition.
The C-Cl-activated complex C21 is about 9.5 kcal/mol less
stable than the ortho C-H activation complex C8. This is most
likely due to the lack of stabilization by additional coordination
to the empty site of the metal center of complex C21, as in the
C-H-activated complex C8. ONIOM calculations of the tBu
derivatives of the ortho-C-H and the C-Cl-activated complexes
support the conclusion that indeed the (tBu-PNP)Ir(H)(C6H4-
Cl)+ complex is more stable than the (tBu-PNP)Ir(Cl)(Ph)+ (by
7.5 kcal/mol).
As can be seen from Figure 6, the kinetic barriers for η2-
arene to η2-arene interchange, η2-arene to η2C-H, and C-H
activation reactions are similar for all the five different pathways
leading to the C-H activation products from the η2-arene
complexes. The highest barrier in each of the five similar
Based on the DFT calculations, the C-Cl activation complex
C21 has thermodynamic stability similar to that of other C-H-
activated complexes observed transiently in the experimental
system (meta and para C-H activation isomers, C13, C17, and
C15); however, it was not observed experimentally. Thus, the
reason that complex C21 was not observed is probably kinetic
rather than thermodynamic. To understand the kinetics of its
formation, the transition states leading to the C-Cl- and the
C-H-activated complexes were compared. C-Cl bond activa-
tion was found to take place exclusively and directly from the
Cl-coordinated complex C6. No direct activation of the C-Cl
bond was found to take place either from an η2-complex such
as C10 or C18 or from an ortho-Cl η2C-H complex such as C7
or C19. Thus, to cleave the C-Cl bond, the Cl-coordinated
complex C6 should first be formed. After its generation,
complex C6 can follow three pathways (Figure 6): (a) oxidative
addition of the C-Cl bond to form complex C21 (with a barrier
of ∆Gq ) 21 kcal/mol); (b) formation of the η2-arene complex
C18 with a lower barrier (∆Gq ) 18 kcal/mol), which can react
pathways is the formation of the η2
bond from the corre-
C-H
sponding η2-arene complex. These barriers are all around 11
kcal/mol, with the smallest TS (C10-C7) barrier of 6 kcal/
mol being the one that leads to the formation of ortho-η2
C-H
complex C7. This probably results from the relatively unstable
η2-arene complex C10, preceding the formation of C7 (probably
due to a steric interference between the Cl atom and the
phosphine ligand).34 The interchange barriers between any two
η2-arene complexes are also similar (∼8-10 kcal/mol) with the
exception of one slightly higher barrier (∆Gq ) 12 kcal/mol),
corresponding to interchange between the two most sterically
hindered η2-arene conformers C10 and C18 (Figure 6f). Thus,
essentially all the sites on the Cl-benzene ring are kinetically
accessible for coordination through an η2-arene interchange and
conversion to the corresponding η2
complexes. Conse-
C-H

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3201
complex to η2
or η2
complexes (∆Gq ≈ 15 kcal/mol).
Table 5. Comparison between Selected Bond Lengths (Å)
and Angles (deg) of the Calculated and Experimental
Structure of the ortho-Cl C-H Activation Complex
C-C
C-H
Slightly lower barriers were found for the formation of the η2
complexes from the η2-arenes (8-12 kcal/mol high). All η2
C-H
C-H
complexes have similar stability; the same is true for the η2-
arene complexes, with the exception of the sterically hindered
endo (C25) and exo (C28) η2-arenes, which are about 5 kcal/
mol higher in energy than the others. In these η2-arene
complexes the steric repulsion between the methoxy moiety and
the phosphine ligand is influencing the relative stability of the
complexes. The barriers for transforming from one η2-arene
complex to another are also relatively low (4-8 kcal/mol, with
the exception of the ones leading to the formation of the
sterically hindered η2-arenes), similarly to those in the chlo-
robenzene activation profile. The lowest barriers found for the
X-ray
mPW1k
Ir-Cl
2.816
1.524
2.045
2.159
1.78
2.33
103
82
72
63.84
2.812
1.534
2.026
2.119
1.857
2.39
Ir-H
Ir-C1
Ir-N
C2-Cl
Ir-P
CAr-Ir-H
N-Ir-H
Ir-Cl-C2
Cl-Ir-CAr
96.2
91.07
71.74
65.5
quently, all the sites can be activated with relatively small
barriers. Furthermore, the back reaction barriers (C-H reductive
elimination) from all C-H-activated complexes apart from C7
(∆Gq ) 25.2 kcal/mol) are low as well (∆Gq ≈ 10-15 kcal/
mol). Thus, in the case of meta and para C-H-activated
complexes (C17, C13, and C15), the microscopic reverse
reaction, C-H reductive elimination, can take place and by
subsequent transformations can eventually lead to the formation
of the most stable product 7. This ortho C-H-activated complex
may be regarded as a thermodynamic sink since the C-H
reductive elimination from it is much higher in energy than any
other transformation in the system including the C-H reductive
elimination from other C-H-activated complexes (C13, C15,
C17, and C20).
Structure of the ortho-Cl C-H Activation Complex. The
tBu derivative of complex C8 was calculated by the ONIOM
method at the mPW1k/LANL2DZ+P//ONIOM(mPW1k/LAN-
L2DZ:mPW1k/LANL1MB) level of theory. The resulting
structure is in good agreement with the experimental X-ray
structure.17 Comparison between selected bond lengths and bond
angles of the calculated and the experimental structures can be
seen in Table 5.
C-H oxidative addition take place from the η2
complexes
C-H
(∆Gq < 2 kcal/mol).
Considering the kinetics, as in the chlorobenzene case, in the
Anisole system, too, the direct pathway leading from the
O-coordinated complex C22 to the ortho η2
complex C23
C-H
is kinetically accessible (∆Gq ) 14.7 kcal/mol) and lower than
the alternative one leading from C22 to the η2-arene complex
C25 (∆Gq ) 17.0 kcal/mol). Similarly to the chlorobenzene
system, in the calculated (Me-PNP)/Ir model system, the
O-coordinated complex C22 appears to be less stable than most
of the η2-arene complexes (C29, C30, C33, and C36). To verify
the stability of the O-coordinated complex relative to the η2-
arene complexes in the realistic system, we carried out ONIOM
calculations on the tBu-substituted analogues of complexes C22
and C30 (which is the most stable η2-arene complex). It was
found that the O-coordinated complex is more stable than the
t-Bu η2-arene analogue of C30 by 7.7 kcal/mol (while in the
Me-PNP system it is less stable by 3 kcal/mol). The significant
stabilization of the O-coordinated complex with respect to the
η2-arene one will most probably result in much higher concen-
tration of this intermediate in the reaction mixture. At the actual
experimental conditions (heating to 70 °C), the barrier leading
from the O-coordinated complex C22 to the η2C-H complex C23
(∆Gq ) 14.7 kcal/mol) can be overcome. Accordingly, since
the O-coordinated complex can preferentially lead to the
Our computational system is in good agreement with the
experimental one regarding the kinetics and the thermodynamics
of the chlorobenzene reaction with the cationic (PNP)Ir+ system.
The computational study suggests a mechanism involving the
formation of different isomers of η2-arene and η2C-H complexes
eventually leading to the thermodynamically most stable ortho-
Cl C-H activation complex C8.
formation of the ortho η2
complex C23, the formation of
C-H
the corresponding ortho C-H activated complex C24 might
be kinetically favored over the formation of the rest of the
isomers. The Ir-CAr bond in this endo-isomer can be rapidly
rotated to give the thermodynamically most stable product
C27.37 The slightly higher energy pathway taking place from
the O-coordinated complex C22, leading to an η2-arene complex
C25, will preferentially lead to the formation of C27 as well
and may contribute to its kinetic preference over the other
isomers of the C-H activation complexes.
Structure of the ortho C-H Activated Complex. A
comparison between the X-ray structure of the ortho-methoxy
C-H activated anisole complex and the calculated ones at two
levels of theory is presented in Table 6. The computed structure
of (tBu-PNP)Ir(H)(C6H4OMe)+ (calculated at the ONIOM-
(mPW1k/LANL2DZ:mPW1k/LANL1MB) level of theory) closely
resembles the experimental X-ray structure. The Ir-X (X )
H, N, P) bond lengths are similar to those found by X-ray
analysis, although the Ir-O bond is slightly shorter in the
computed structure than the X-ray one (2.60 vs 2.76 Å).
Nevertheless, as in the X-ray structure, the methoxy group in
the computed structure is not directly coordinated to the Ir atom
but rather directed toward it. To confirm that the oxygen atom
is not coordinated to the Ir center, an additional calculation of
this structure using a different functional (ONIOM(B97-1/
C-H Activation of Anisole by Me-PNP/Ir+. Anisole
exhibits experimentally thermodynamic and kinetic preference
for the formation of an ortho-methoxy-C-H activation complex,
as in the case of chlorobenzene. To assess the mechanism of
such selectivity, we investigated computationally the arene C-H
activation in anisole.36 As in the chlorobenzene system,
geometry optimizations were first carried out on the H-(PNP)-
Ir(anisol)+ system at the DFT mPW1k/SDD level of theory,
then Me substituents were added and the structures were
reoptimized. Single-point energies were calculated on the
optimized geometries at the mPW1k/SDB-cc-pVDZ level of
theory to increase the accuracy of the results.
The arene C-H activation of anisole is very similar to the
one of chlorobenzene, involving multiple isomers of intermedi-
ates and transition states. Detailed graphs for each C-H
activation site are presented in Figure 7.
In agreement with the experiment, the ortho-methoxy C-H-
activated complex C27 was found to be the thermodynamic
product, stabilized by 6-10 kcal/mol relative to the other C-H
activation complexes. The kinetics observed for C-H activation
processes in anisole is similar to that found for chlorobenzene.
The highest barriers are the ones leading from the O-coordinated

3202 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3203
Figure 7. Detailed graphs, each representing a distinct pathway for the formation of the different C-H activation products (ortho, meta,
or para conformers) including a site exchange pathway leading from one η2-arene complex to a different η2-arene one.
LANL2DZ:B97-1/LANL1MB) level of theory) was performed.
Consequently, a longer Ir-O bond distance of 2.92 Å was
computed at this level of theory, confirming that the methoxy
moiety is not coordinated to the metal center. However, the
methoxy group might interact electrostatically with the metal
center (δ+ on Ir, δ- on O), stabilizing the ortho-isomer with
respect to other C-H activated isomers due to closer proximity
of the methoxy moiety to the metal center. Natural bond order
(NBO) calculations confirm that the oxygen atom is not
coordinated to the metal center (Wiberg index of 0.028 for the
Ir-O bond with respect to a larger 0.105 index for the Ir-Cl
bond in chlorobenzene aryl hydride complex). Based on the
natural population analysis (NPA) the partial charge on the
oxygen atom is -0.625, whereas on the Ir center it is +0.175,
compared to a smaller charge of -0.086 on the Cl atom and
+0.157 on the Ir in the analogous chlorobenzene complex.
Comparison between C-H Activation of Anisole, Chlo-
robenzene, and Benzene. Comparing the analogous isomers
of C-H activation of chlorobenzene and anisole (ortho-exo,
ortho-endo, meta-exo, meta-endo, and para), it can be seen in
general that their stability is similar (within 1 kcal/mol). The
anisole C-H activated complexes are slightly more stable than
the analogous chlorobenzene ones, apart from the ortho-exo
complexes, for which the chlorobenzene complex is slightly
more stable. This stability is probably gained by the stronger
coordination of the chlorine atom than of the oxygen atom to
the metal center. This is also supported by NBO calculations,
showing larger bond order for Ir-Cl than Ir-O (vide supra).

3204 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.
process1e to occur since it brings the C-H bond into close
proximity to the metal and, by that, favors η2C-H formation and
lowers the barrier for the C-H activation reaction. Formation
of (PNP)Ir(H)(Ph)+PF6 (2) was found to be a thermodynamic
sink for the overall reaction. 2 is 8.8 kcal/mol more stable than
(PNP)Ir(COE)+PF6 (1).
Table 6. Comparison between Selected Bond Lengths (Å)
and Angles (deg) for the Calculated and the X-ray
Structures for the ortho-tBu-PNP/Ir(H)(Ar) (Ar )
C6H4OCH3) Complexa
exptl (X-ray)
calcd (mPW1k)
calcd (B97)
Ir-O
2.762
1.553
2.038
2.138
2.317
2.315
1.397
1.425
2.338
164.06
177.7
112.5
81.08
55.4
2.599
1.532
2.025
2.131
2.371
2.375
1.417
1.444
2.373
163.27
174.86
108.37
82.94
58.46
2.917
1.539
2.048
2.164
2.395
2.403
1.435
1.469
2.399
164.86
178.18
115.89
77.53
53.86
Ir-H
Complex 2 is exceptionally stable for a coordinatively
unsaturated square-pyramidal iridium complex. An unsaturated,
thermally stable, hydrido-aryl trigonal bipyramidal complex, Ir-
(PiPr3)2(H)(Ph)(Cl), reported by Werner,21 is stabilized by strong
π-donation of the chloride ligand to an empty d-orbital of the
metal. This type of stabilization is absent in 2. Recently,
Goldman reported a closely related (PCP)Ir(H)(Ph) (PCP ) η
η3-2,6-(tBu2PCH2)2C6H3) complex.22 Complex 2 and (PCP)Ir-
(H)(Ph) represent a rare case of a differently charged, isoelec-
tronic couple of the same metal that undergoes C-H reductive
elimination-oxidative addition. While complex 2 exhibits high
thermal stability, the neutral (PCP)Ir(H)(Ph) exhibits a relatively
high exchange rate with other arenes even at room temperature
and is stable only at -10 °C. This seems counterintuitive, since
the electron density at the metal center in the neutral PCP system
is expected to be higher than in the cationic PNP system, which
is expected to make the latter more prone to C-H reductive
elimination. Both experiments and calculations showed that,
indeed, the electron density at the metal center in the PCP system
is higher. Thus, the CO stretching frequency of (PCP)Ir(H)-
(Ph)(CO) is 1973 cm-1 versus 2005 cm-1 of the PNP system.
According to calculations, the Ir-CPh bond in the PCP system
is longer than the Ir-CPh bond of the PNP system, which is
attributed to the higher trans influence of the aryl PCP ring
versus the pyridine PNP one and explains the lower stability of
(PCP)Ir(H)(Ph).
Ir-Cipso
Ir-N2
P1-Ir
P2-Ir
C2-O
O-Me
Ir-P
P-Ir-P
Cipso-Ir-N2
Ir-Cipso-C2
C2-O-Ir
O-Ir-Cipso
a Calculated structures are at the mPW1k/LANL2DZ+P//ONIOM(mPW1k/
LANL2DZ:mPW1k/LANL1MB) and the B97-1/LANL2DZ+P//ONIOM(B97-
1/LANL2DZ:B97-1/LANL1MB) levels of theory.
The stability of the benzene C-H activation product is
comparable to the ortho-endo-, meta-, and para-isomers of
anisole and chlorobenzene, but it is lower than the stability of
the ortho-exo complexes, probably due to the additional
coordination of the heteroatom in the chlorobenzene case or
Ir-O interaction in the anisole case. The stability of both η2-
arene and η2C-H complexes in the anisole case is slightly (1-2
kcal/mol) higher than in the chlorobenzene one. However these
differences are small, and in general, the stability is comparable
for analogous isomers of anisole and chlorobenzene, supporting
the experimental observation of slightly different ratios of their
formation in the competition experiments.
The reactivity of another closely related complex, (PNP)Ir-
(H)(Ph) (PNP ) bis(2-(diisopropylphosphino)-4-methylphenyl)-
amine), was recently reported by Ozerov et al.11d The electron
density at the Ir(III) center of this neutral complex is expected
to be higher than in the case of the cationic complex 2. As in
the case of 2, Ozerov’s complex did not exhibit fast arene
exchange at room temperature, unlike Goldman’s (PCP)Ir(H)-
(Ph), most probably as a result of the lower trans influence of
the PNP (amine) ligand compared to that of the PCP (aryl).
Discussion
C-H Activation of Benzene. Aromatic C-H bond activation
processes by late transition metals have been studied mecha-
nistically, yielding important insights.1 While aliphatic σ-com-
plexes38 have drawn great attention in view of their relevance
to saturated hydrocarbon activation, aromatic η2C-H σ-bonding
is less studied.39,40 Recently, we have reported an experimental
and computational DFT study of the C-H activation mechanism
in both aliphatic and aromatic PC-Rh systems. C-H activation
Experimental work demonstrates that heating the C-H
activation complex (PNP)Ir(Ph)(H)+ at 60 °C with other arenes
results in arene exchange.17 This process can in principle take
place by dissociative or associative mechanisms. Both mecha-
nisms are expected to take place from the Ir(I) intermediates,
complexes η2C-C C3 and η2C-H C4, rather than from the (PNP)-
Ir(Ph)(H)+ complex C5. The computational results indicate that
in both cases was found to take place through a η2
C-H
complex.41
In the current work, our experimental observations, supported
by the theoretical studies, indicate that the overall rate-deter-
mining step in the C-H activation process by (PNP)Ir-
C-H reductive elimination, leading to the η2
intermediate
-
(COE)+PF6 (1) is dissociation of COE to form the T-shaped
C-H
C4, has a relatively low barrier (∆Gq ) 11 kcal/mol), as does
the reaction C4 f C3 (∆Gq ) 3 kcal/mol). A computational
search for an associative exchange of benzene found one
transition state for the exchange of two benzene ligands in the
(PNP)Ir(I) 14e species (∆Gq ) 32 kcal/mol for Me-PNP). The
lack of kinetic isotopic effect (vide supra) in the C-H activation
process indicates that the C-H cleavage occurs after the rate-
determining step supporting the calculations. The formation of
a 14e M(I) (M ) Rh, Ir) species is suggested to be an important
step in C-H activation processes.24b,42-44 From the calculations
we can see that this species is in fact high in energy and
stabilized by coordination of the arene. Two coordination modes
are found possible for benzene, η2C-C and η2C-H, the latter being
an intermediate before the actual C-H cleavage step. The
isomerization between the two coordination modes has the
highest barrier following COE dissociation barrier (10.4 kcal/
mol), while the actual C-H bond cleavage was found to be
almost barrierless. Moreover, it is anticipated that the formation
η2
configuration (TS-C4-C4). The barrier found for the
C-H
associative exchange is ∼20 kcal/mol, which is 12 kcal/mol
higher in energy than the dissociation of benzene leading to
the T-shaped complex C2. Thus the dissociative pathway for
benzene exchange is more likely to take place in this system.45
The more bulky tBu subsituents would further favor a dissocia-
tive mechanism. Arene exchange in the similar (tBu-PCP)Ir
system was suggested to take place dissociatively based on both
computational DFT and experimental study.22,46
C-H Activation of Haloarenes and Anisole. The impor-
tance of a coordinating moiety in directing C-H activation
of the η2
complex C3 is essential for the C-H activation
C-C

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3205
toward the ortho-position of an arene was reported for carbonyl
and nitrile groups in various systems, both experimentally and
theoretically. Formation of a stable metallacycle was suggested
to account for the observed reactivity. Although in most cases
the metallacycle obtained is a stable five- or six-membered ring
complex, three- and four-membered metallacycles are also
known.47 The C-H activation process by the Ir(PNP) system
can be viewed as a cyclometalation assisted by weak coordina-
tion of the halide atom or the oxygen atom (in anisole).
Examples of chloro- and bromoarene coordination are not
common.27 Moreover, the coordination-assisted C-H activation
of those compounds is rare. In our experimental work we have
compared the activation selectivity of fluorobenzene, chloroben-
zene, bromobenzene, and anisole. In general, there is a correla-
tion between the observed kinetic selectivity, which increased
upon going from fluorobenzene to bromobenzene, and the
polarizability of the halogen atoms. As the polarizability
increases in descending the periodic table from fluorine to
bromine, the coordination ability of the halide substituent
increases.
An interesting comparison can be made with the neutral
(PNP)Ir(I) complex reported by Ozerov,11d where only 71%
selectivity toward the ortho-C-H bond of chlorobenzene was
observed. Indeed, no coordination of the chloride atom in the
C-H activation product (PNP)Ir(H)(o-C6H4Cl) was seen. Obvi-
ously, Cl coordination to a cationic Ir(III) center is more
favorable than to a neutral one. Lack of the extra stabilization
gained by coordination of the chlorine atom to the metal center
can explain the lower selectivity in that case.
According to calculations, arene C-H activation was found
to take place directly only from η2C-H complexes, whereas the
η2-arene complexes are responsible for facilitating formation
of η2C-H complexes and interchanging between different posi-
tions along the aromatic ring. However, a Cl-coordinated
intermediate can be also formed in this system, which might
favor the ortho-Cl C-H activation. Thus both η2-arene and Cl-
coordinated intermediates serve as predirecting intermediates
toward the formation of certain η2C-H complexes necessary for
C-H activation to occur. However, since the Cl-coordinated
complex was found in the calculations to be substantially more
stable than the arene η2
complexes, it might explain the
C-C
Indeed, lack of selectivity was observed experimentally in
the C-H activation of fluorobenzene with complex 1. Recently,
a theoretical and experimental study on the sensitivity of the
M-C bond strength to substituents was reported.27 It was shown
that the ortho C-H bond in fluorobenzene is the strongest C-H
bond in this molecule, and thus, the M-C bond in this position
is the strongest one, resulting in a thermodynamic preference
for C-H activation at the ortho-position. Caulton showed15b
that an Os(II) complex activates the ortho C-H bond prefer-
entially in fluoroarenes. This was explained to be a result of
preferential interaction of Os with the π-cloud ortho to the C-F
bond, making the ortho activation kinetically favored.15b In our
case a slight thermodynamic preference was found in favor of
the ortho- versus the para-isomer and even a higher preference
versus the meta-isomer. The kinetically nonselective activation
of fluorobenzene can be explained by the low coordination
ability of the fluorine substituent, and the slight thermodynamic
selectivity can be attributed to bond strength.
kinetic preference of the ortho-Cl C-H activation complex over
other C-H activation isomers, which can be obtained only from
the less stable η2
complexes.
C-C
The C-Cl bond cleavage is not observed experimentally,
although theoretically the C-Cl oxidative addition product is
as stable as the activation products of the meta and para C-H
bonds. According to the calculations, the reason is kinetic. Both
C-H and C-Cl activation proceed from the Cl-coordinated
intermediate, but the barrier to C-Cl activation is much higher,
probably as a result of steric hindrance inflicted by the tert-
butyl substituents on the phosphines. Remarkably, the normally
facile oxidative addition of the Ph-Br bond to electron-rich
Ir(I)48 was not observed, although this bond is significantly
weaker then C-H bonds and prone to nucleophilic attack. The
selective activation of anisole is governed mostly by the same
factors as in the case of chlorobenzene, with the heteroatom
(oxygen of anisole) serving as a directing moiety.49 In the case
of anisole it is less obvious why the ortho-isomer is also
thermodynamically the most stable one. Both calculations and
X-ray analysis, supported by NMR, did not show significant
coordination of the methoxy moiety to the metal center. Some
structural clues are relevant. Comparing the Ir-Cipso-C bond
angle in (PNP)Ir(H)(C6H5) (2) (125.1°), (PNP)Ir(H)(C6H4Cl)
(8a) (134.3°), and (PNP)Ir(H)(C6H4OMe) (10a) (131.1°), it is
largest in 8a and smallest in 2, implying in the case of anisole
an interaction of the oxygen atom with the metal center. Such
an interaction could be electrostatic (δ+ on Ir, δ- on oxygen).
Comparison between the Different C-H Activation Pro-
cesses. As observed in the exchange experiments, (PNP)Ir(H)-
(Ph) (2) underwent exchange with other arenes at lower
temperature relative to complexes 8a, 9a, and 10a. The lack of
ligand coordination in the empty coordination site, trans to the
hydride, contributes greatly to such reactivity. Indeed, when
acetonitrile was added as the sixth ligand, the reductive
elimination of benzene slowed. A more profound effect was
observed when the sixth ligand was CO, in which case reductive
elimination was observed only above 110 °C. C-H reductive
elimination from Rh(III) and Ir(III) octahedral complexes was
Although chelation-assisted selective ortho C-H activation
in aromatic systems is well documented4,5 and mechanistically
studied, examples of similar chelation-assisted processes in halo-
arenes are very rare.9,13,17 The Ir(I) cationic system exhibited
kinetic and thermodynamic selectivity toward the C-H bond
ortho to the halide atom (chloro- or bromobenzene) or etheral
oxygen atom (anisole). As shown experimentally and theoreti-
cally, the selectivity observed might be due to precoordination
of the halogen atom or the oxygen atom to the metal center.
Such precoordination, like the η2
arene coordination, has
C-C
an accessible pathway toward C-H activation.
In general the C-H activation of chlorobenzene is essentially
similar to that of benzene, although a much more complicated
mechanism occurs with chlorobenzene due to the lack of
symmetry and coordination through chloride. Indeed, according
to both experiment and theory, there is a distribution of several
possible C-H activation complexes, which were obtained
kinetically (ortho (C8 and C20), meta (C13 and C17), and para-
Cl (C15) C-H-activated complexes), but the ortho C-H
activation complex, C8, is the kinetically favored one. Moreover,
it is 10 kcal/mol more stable thermodynamically due to
coordination of the Cl atom to the metal center and, thus, is the
sole product observed in NMR after continued heating at
60 °C, as shown experimentally.
(48) Yamamoto, A. Organotransition Metal Chemistry; John Wiley &
Sons: New York, 1986.
(49) It is important to point out that the C-Hsp3 bond was not activated
under those conditions. However, at elevated temperatures C-Hsp3 bond
cleavage is observed. This is under study and will be reported in a future
paper.
(47) Ryabov, A. D. Chem. ReV. 1990, 90, 403.

3206 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.
shown to involve ligand loss.24 Both experiments and calcula-
tions show that coordination of the heteroatom to the metal
center after the C-H activation process increases the kinetic
barrier for reductive elimination and thus further stabilizes the
ortho-activated products in comparison to (PNP)Ir(H)(Ph).
It was shown experimentally that the three different ortho
C-H-activated complexes (PNP)Ir(H)(C6H4Cl), (PNP)Ir(H)-
(C6H4Br), and (PNP)Ir(H)(C6H4OMe) have comparable kinetic
stability. However, upon heating a solution of either complex
(PNP)Ir(H)(C6H4X) at 70 °C in a solution of anisole/chloroben-
zene or anisole/bromobenzene (1:1 molar ratio), arene exchange
occurred. Under those conditions the ortho-activated anisole
complex is the most stable one thermodynamically, followed
by the chlorobenzene and bromobenzene ortho-isomers (in that
order). The para- and meta-isomers under those conditions are
the least stable, as they disappear after prolonged heating (70
°C for 5 days). Examining the calculated bond strengths of the
ortho, meta, and para C-H bonds of the haloarenes and anisole,
it is possible to see a correlation between the bond strength
and the observed thermodynamic stability of the C-H activation
product.50 In anisole and fluorobenzene the ortho C-H bond
is strongest (compared to the other haloarenes) and the difference
in bond strength between the meta and para ones is the largest.
This correlates with an even stronger M-C bond at the ortho-
position.15 However, while in the case of anisole kinetic as well
as thermodynamic preference was observed, with fluorobenzene
no kinetic effect was observed, supporting coordination-assisted
C-H activation with anisole. Moreover, despite the fact that
the energy difference between the arene C-H bonds is lower
in chlorobenzene and bromobenzene, kinetic and thermodynamic
preference was observed toward the ortho C-H bond.
coordination, the effect being both kinetic and thermodynamic;
that is, heteroatom coordination directs the metal to the ortho-
position and stabilizes the activation product. A lower selectivity
was observed in the case of the neutral PNP complex reported
by Ozerov11d (vide supra), where no coordination of the
heteroatom to the metal center took place following C-H
activation. As shown theoretically by ONIOM calculations of
the t-Bu derivatives of C6 and C22, after COE dissociation from
complex 1 takes place, the most stable intermediates are the
heteroatom-coordinated complexes of chlorobenzene and ani-
sole. The t-Bu η2
analogues of C9 and C30, which are
C-C
similar to the η2C-C intermediate of benzene, were shown to be
less stable, pointing out the importance of heteroatom coordina-
tion in directing the C-H activation. Moreover, the calculated
ortho-activated complexes C8 and C27 are the most stable ones
thermodynamically, with the highest barriers for reductive
elimination, due to heteroatom coordination (or interaction in
the anisole case).
The unobserved C-Cl activation process was shown com-
putationally to involve the same critical Cl-coordinated inter-
mediate, as in the C-H activation process, but it experiences a
higher activation barrier, and the ortho C-H activation product
is also thermodynamically more stable than the C-Cl oxidative
addition complex. Thus, theory supports the experimentally
observed selective C-H bond activation. Examples of chlorine
and bromine as directing moieties are scarce13 and may suggest
new ways for selective functionalization of haloarenes by C-H
activation while retaining the normally more reactive C-halogen
bond.
Experimental Section
Computational Methods. All calculations were carried out using
Gaussian 98 program revision A.1151 and Gaussian 03 revision
C.0152 running on the Linux farms of Martin group and of the
Faculty of Chemistry Linux farm.
The mPW1k (modified Perdew-Wang one-parameter for kinet-
ics) exchange-correlation functional of Truhlar and co-workers53
Conclusions
An experimental and theoretical study of the C-H activation
of aromatic C-H bonds by an electron-rich (PNP)Ir(COE)+
complex was presented here. Activation of benzene involves a
dissociative mechanism via the formation of a three-coordinate
14e Ir(PNP) intermediate, formation of which is the rate-
determining step of the whole process. Two coordination modes
are found possible for benzene, η2C-C and η2C-H, both of which
(51) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98, ReVision A.11; Gaussian, Inc.: Pittsburgh,
PA, 2001.
(52) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K.
N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A.
D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A.
G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; and Pople, J. A.
Gaussian 03, ReVision C.02; Gaussian, Inc.: Wallingford, CT, 2004.
(53) Lynch, B. J.; Fast, P. L.; Harris, M.; Truhlar, D. G. J. Phys. Chem.
A 2000, 104, 4811
are involved in the C-H activation process; the η2
inter-
C-C
mediate brings the metal to the proximity of the C-H bond,
followed by its isomerization to the η2C-H intermediate, which
undergoes C-H cleavage. The isomerization between the two
coordination modes has the highest barrier following COE
dissociation barrier, while the cleavage of the C-H bond is
almost barrierless.
A rare case of two isoelectronic, differently charged systems
of the same metal, (PNP)Ir(H)(Ph)+ and (PCP)Ir(H)(Ph), which
undergo C-H oxidative addition/reductive elimination, was
examined experimentally and computationally. It was illustrated
both experimentally and theoretically that the cationic PNP
system has a lower electron density at the metal center. A higher
trans influence of the ipso carbon of the PCP system as
compared with the pyridinic nitrogen is evident theoretically
and explains the lower stability of the (PCP)Ir(H)(Ph) system.
Rare, selective C-H bond activation of haloarenes and anisole
was also studied experimentally and theoretically. Experiments
have indicated that the activation is directed by heteroatom
(50) Calculated C-H bond strength in kcal/mol at 298 K: ortho-anisole
476.2, meta-anisole 466.8, para-anisole 471.9, ortho-bromobenzene 472.5,
meta-bromobenzene 467.6, para-bromobenzene 470.6, ortho-chlorobenzene
473.9, meta-chlorobenzene 468.0, para-chlorobenzene 471.0, ortho-fluo-
robenzene 478.5, meta-fluorobenzene 468.2, para-fluorobenzene 471.8.

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3207
was employed in conjunction with the SDD and SDB-cc-pVDZ
basis sets (see below). The mPW1k functional was previously
shown54,55 to typically yield more reliable reaction barrier heights
than other exchange-correlation functionals.
The SDD basis set is the combination of the Huzinaga-Dunning
double-ú basis set on lighter elements with the Stuttgart-Dresden
basis set-relativistic effective core potential (RECP) combination56
on the transition metals. The SDB-cc-pVDZ basis set combines
the Dunning cc-pVDZ basis set57 on the main group elements with
the Stuttgart-Dresden basis set-RECP combination56 on the
transition metals, with an f-type polarization exponent taken as the
geometric average of the two f-exponents given in the Appendix
to ref 58.
generated from manual manipulation of the geometry using
MOLDEN.67 In cases where this approach failed to converge, we
used analytical second derivatives at every step.
Zero-point and RRHO (rigid rotor harmonic oscillator) thermal
corrections (to obtain ∆S and ∆G values) were obtained from the
unscaled computed frequencies.
Where necessary, the Grid ) UltraFine combination, i.e., a
pruned (99 590) grid in the integration and gradient steps and a
pruned (50 194) grid in the CPKS (coupled perturbed Kohn-Sham)
steps, was used as recommended in ref 68.
For interpretative purposes, atomic partial charges and Wiberg
bond indices69 were obtained by means of a natural population
analysis (NPA)70 at the mPW1k/LANL2DZ level for the tBu-PCP
complexes.
The energetics for our final reaction profiles obtained from the
mPW1k functional with the SDD basis set were validated by single-
point energy calculations, using the SDD reference geometries, with
the extended SDB-cc-pVDZ basis set.
Experimental Methods. General Procedures. All experiments
with metal complexes and phosphine ligands were carried out under
an atmosphere of purified nitrogen in a Vacuum Atmospheres
glovebox equipped with a MO 40-2 inert gas purifier or using
standard Schlenk techniques. All solvents were reagent grade or
better. All nondeuterated solvents were refluxed over sodium/
benzophenone ketyl and distilled under argon atmosphere. Deu-
terated solvents were used as received. All the solvents were
degassed with argon and kept in the glovebox over 4 Å molecular
sieves. Commercially available reagents were used as received.
1H, 13C, 31P, and 19F NMR spectra were recorded at 400, 100,
162, and 376 MHz, respectively, using a Bruker AMX-400 NMR
For tBu-PNP, two-layer ONIOM59 approach was employed, in
which the tert-butyl groups on the phosphorus were placed in the
outer layer and the remainder of the system was placed in the inner
layer. For technical reasons, the mPW1k functional was used for
both layers. In the anisol case, an additional ONIOM calculation
was performed with the Becke97-1 (B97-1) functional,60 as it has
been shown that in some instances this functional is more suitable
for determining molecular equilibrium properties.61 Both in mPW1k
and in B97-1 functionals, for the inner layer the valence Los Alamos
National Laboratory double-ú (LANL2DZ) basis set-RECP com-
bination was used,62 while for the outer layer, the LANL1MB basis
set-RECP combination was employed.62 (As is customary, first-
row atoms in this approach were treated by means of the Dunning-
Hay valence double-ú63 basis set.) The energetics for our final tBu-
PCP geometries were validated by single-point energy calculations,
using the ONIOM(mPW1k/LANL2DZ:mPW1k/LANL1MB) or
ONIOM(B97-1/LANL2DZ:B97-1/LANL1MB) reference geom-
etries, and at the higher level of theory mPW1k/LANL2DZ+P or
B97-1/LANL2DZ+P, respectively, where the “+P” stands for the
addition of polarization functions with exponents taken from ref
64.
1
spectrometer. All spectra were recorded at 23 °C. H NMR and
13C{1H} NMR chemical shifts are reported in ppm downfield from
tetramethylsilane. 1H NMR chemical shifts were referenced to the
residual hydrogen signal of the deuterated solvents (3.58 ppm,
tetrahydrofuran; 5.32 ppm, dichloromethane). In 13C{1H} NMR
measurements the signals of THF-d8 (67.5 ppm) and CD2Cl2 (53.8
ppm) were used as a reference. 31P NMR chemical shifts are
reported in parts per million downfield from H3PO4 and referenced
to an external 85% solution of phosphoric acid in D2O. 19F NMR
chemical shifts were referenced to C6F6 (-163 ppm). Screw-cap 5
mm NMR tubes were used in the NMR follow-up experiments.
Abbreviations used in the description of NMR data are as follows:
br, broad; s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet;
v, virtual.
Geometry optimizations for minima were carried out using the
standard Schlegel algorithm65 in redundant internal coordinates until
in the neighborhood of the solution and then continued using
analytical second derivatives.66 Optimizations for transition states
were carried out with an initial guess for the transition state being
Reaction of [Ir(COE)2(acetone)2][PF6] with PNP (PNP ) 2,6-
Bis(di-tBu-phosphinomethylene)pyridine; COE ) cyclooctene).
Preparation of [Ir(PNP)(COE)][PF6] (1). An acetone solution (2
mL) of PNP (50 mg, 0.126 mmol) was added dropwise to an
acetone solution (2 mL) of [Ir(COE)2(acetone)2][PF6]18 (85.2 mg,
0.126 mmol). Upon addition of PNP, the color turned dark purple-
red. The mixture was stirred for an additional 2 min at room
temperature, then poured into pentane, causing precipitation of
complex 1 as a pink-red solid. The solid was separated by
decantation and dried under vacuum, resulting in a quantitative yield
of complex 1.
(54) (a) Parthiban, S.; de Oliveira, G.; Martin, J. M. L. J. Phys. Chem.
A 2001, 105, 895. (b) Lynch, B. J.; Truhlar, D. G. J. Phys. Chem. A 2001,
105, 2936.
(55) Iron, M. A.; Lo, H. C.; Martin, J. M. L.; Keinan, E. J. Am. Chem.
Soc. 2002, 124, 7041.
(56) Dolg, M. In Modern Methods and Algorithms of Quantum Chem-
istry; Grotendorst, J., Ed.; John von Neumann Institute for Computing:
Ju¨lich, 2000; Vol. 1, p 479.
(57) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007.
(58) Martin, J. M. L.; Sundermann, A. J. Chem. Phys. 2001, 114, 3408.
(59) Dapprich, S.; Komaromi, I.; Byun, K. S.; Morokuma, K.; Frisch,
M. J. J. Mol. Struct. (THEOCHEM) 1999, 461, 1, and references therein.
(60) Hamprecht, F. A.; Cohen, A. J.; Tozer, D. J.; Handy, N. C. J. Chem.
Phys. 1998, 109, 6264.
[Ir(PNP)(COE)][BArF]. To an acetone suspension (5 mL) of
[Ir(COE)2-µ-Cl]2 (200 mg, 0.223 mmol) were added 3 equiv of
NaBArF (592 mg, 0.670 mmol). The slurry was stirred for 4 h at
room temperature until the color changed to red-purple. The solution
was filtered through Celite, the solvent was evaporated, and the
residue was washed with pentane (15 mL) and then washed with
ether. The pinkish solid was left under vacuum for one night,
resulting in 80% yield. 31P{1H} NMR (acetone-d6): 44.00 (d, Jpp)
318.0 Hz, P-Ir-P), 49.00 (d, Jpp ) 318.0 Hz, P-Ir-P), -150.00
(61) Boese, A. D.; Martin, J. M. L.; Handy, N. C. J. Chem. Phys. 2003,
119, 3005.
(62) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
(63) Dunning, T. H.; Hay, P. J. In Modern Theoretical Chemistry;
Schaefer, H. F., III, Ed.; Plenum: New York, 1976; Vol. 3.
(64) (a) Hollwarth, A.; Bohme, M.; Dapprich, S.; Ehlers, A. W.; Gobbi,
A.; Jonas, V.; Kohler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G.
Chem. Phys. Lett. 1993, 208, 237. (b) Ehlers, A. W.; Bohme, M.; Dapprich,
S.; Gobbi, A.; Hollwarth, A.; Jonas, V.; Kohler, K. F.; Stegmann, R.;
Veldkamp, A.; Frenking, G. Chem. Phys. Lett. 1993, 208, 111.
(65) (a) Schlegel, H. B. J. Comput. Chem. 1982, 3, 214. (b) Peng, C.;
Ayala, P. Y.; Schlegel, H. B.; Frisch, M. J. J. Comput. Chem. 1996, 17,
49.
1
(heptet, JFP ) 713.0 Hz, PF6). H NMR (acetone-d6): 1.47 (dd,
(67) Schaftenaar, G. Molden 3.6, 1999. URL: http://www.cmbi.kun.nl/
∼schaft/molden/molden.html.
(68) Martin, J. M. L.; Bauschlicher, C. W.; Ricca, A. Comput. Phys.
Commun. 2001, 133, 189.
(69) Wiberg, K. B. Tetrahedron 1968, 24, 1083.
(70) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88,
899.
(66) Stratmann R. E.; Burant J. C.; Scuseria G. E.; Frisch M. J. J. Chem.
Phys. 1997, 106, 10175.

3208 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.
5JPH ) 1.8 Hz, 3JPH ) 3.8 Hz), 1.42 (dd, 5JPH ) 2.0 Hz, 3JPH ) 4.3
Hz, 18H, P-C(CH3)3), 1.44-1.50 (m, 10H, aliphatic hydrogens of
was heated at 50 °C for 4 h, after which the 31P NMR spectrum
did not change and the signals of the hydride ligand and the aromatic
1
COE), 2.55 (m, 2H, CH2-CHdCH), 4.07 (vt, JPH ) 7.9 Hz, 4H,
protons in the H NMR spectrum disappeared while appearing in
2
3
the H NMR spectrum. Leaving a solution of complex 2 in C6D6
CH2-P), 4.20 (m, 2H vinylic hydrogens of COE), 7.87 (d, JHH
)
3
at room temperature for 4 days resulted in a small exchange only.
7.8 Hz, 1H, PNP-aryl-H), 7.97 (d, JHH ) 7.6 Hz, 1H, PNP-aryl
H), 8.25 (t, JHH ) 7.7 Hz, 1H, PNP-aryl H). 19F NMR (acetone-
3
Reaction of [Ir(PNP)(COE)][PF6] (1) with CO. Formation of
[Ir(PNP)(CO)][PF6] (5). Into an acetone solution (3 mL) of
complex 1 (15 mg, 0.039 mmol) was bubbled CO in small excess
for 1 min, during which time the color changed to yellow. The
solvent was evaporated and the yellow solid was washed with
pentane and dried under vacuum overnight, resulting in pure
complex 5 in quantitative yield. Slow evaporation of the acetone
solution resulted in formation of yellow crystals suitable for X-ray
analysis. 31P{1H} NMR (CD2Cl2): 75.00 (s), -150.0 (heptet, 1JFP
) 713 Hz, PF6). 1H NMR (CD2Cl2): 1.15 (t, 3JPH ) 14.2 Hz, 36H,
P-C(CH3)3), 3.93 (m, 4H, CH2-P), 7.72 (d, JHH ) 7.6 Hz, 2H,
H-PNP-aryl), 8.0 (d, JHH ) 7.8 Hz, 2H, H-PNP-aryl). 13C{1H} NMR
d6): -73 (d, JPF ) 713.0 Hz, PF6). 13C{1H} NMR (acetone-d6):
1
30.6 (m, PC(CH3)3), 36.2-36.5 (m, CH2P), 54 (s, 2H, CHdCH),
3
5
120 (dd, JPC ) 7.4 Hz, JPC ) 10.1 Hz, PNP-aryl C), 141.6 (s,
PNP-aryl C), 164.3 (d, JPC ) 13.0 Hz, PNP aryl-C). [BArF]
2
signals: 1H NMR (acetone-d6) 7.7 (s, 8H), 8.0 (s, 4H); 19F NMR
(acetone-d6) -60 (s). Assignment of the signals was confirmed by
DEPT. Anal. Calcd for 1 (counteranion PF6-), C31H57P3F6NIr: C,
44.17; H, 6.82. Found: C, 44.25; H, 6.76.
Reaction of [Ir(PNP)(COE)][PF6] (1) with Benzene. Forma-
tion of [Ir(PNP)(H)(Ph)][PF6] (2). A solution (3 mL) of complex
1 (15 mg, 0.039 mmol) in benzene was heated at 60 °C for 1 h,
during which the color changed to orange. The solvent was
evaporated and the orange solid was washed with pentane and dried
under vacuum overnight, resulting in pure complex 2 in quantitative
yield. Compound 2 was obtained also by dissolving complex 1 in
benzene and leaving it for 4 days at room temperature. 31P{1H}
1
(CD2Cl2): 27.9-29.2 (m, P-C(CH3)3), 35.6 (t, JPC ) 12.4 Hz,
CH2P), 122.4 (m, PNP-aryl, C3,5), 139.6 (s, PNP-aryl, C4), 166.9
(br t, PNP-aryl, C2,6), 182 (br t, Ir-CO). Anal. Calcd for C24H43P3F6-
NOIr: C, 37.89; H, 5.70. Found: C, 37.80; H, 5.76.
Reaction of [Ir(PNP)(H)(Ph)][PF6] (2) with CH3CN. Forma-
tion of [Ir(PNP)(H)(Ph)(CH3CN)][PF6] (6). To a CH2Cl2 solution
(3 mL) of complex 2 (15 mg, 0.039 mmol) was added 1 equiv of
acetonitrile, resulting in a color change to pale orange. The solvent
was evaporated and the orange solid was washed with ether and
dried under vacuum overnight, resulting in pure complex 6 in
quantitative yield. Slow diffusion of pentane into a THF solution
of complex 6 allowed formation of orange crystals suitable for X-ray
analysis. 31P{1H} NMR (CD2Cl2): 37.97 (br s), -150.0 (heptet,
2
NMR (hydride coupled, CD2Cl2): 54.00 (d, JHP ) 10.0 Hz),
1
1
-150.0 (heptet, JFP ) 713 Hz, PF6). H NMR (CD2Cl2): -43.9
(vt, 2JPH ) 12 Hz, 1H, H-Ir), 1.18 (dd, 5JPH ) 6.9 Hz, 3JPH ) 14.2
Hz, 18H, P-C(CH3)3), 3.90 (m, 4H, CH2-P), 6.80 (t, JHH ) 7.3
2
Hz, 1H, aryl-H), 6.93 (t, 2JHH ) 6.2 Hz, 1H, aryl-H), 7.06 (t, 2JHH
) 7.3 Hz, 1H, aryl-H), 7.32 (d, JHH ) 7.6 Hz 2H, aryl-H), 7.7 (d,
JHH ) 7.8 Hz, 2H, H-PNP-aryl), 7.96 (t, JHH ) 7.8 Hz, 1H, H-PNP-
aryl). Assignment of signals was confirmed by COSY. NOE effect
was observed between hydride at -43.9 ppm and protons at 7.32
ppm. 13C{1H} NMR (CD2Cl2): 27.9-29.6 (m, P-C(CH3)3), 36 (m,
CH2P), 122.4 (m, PNP-aryl C3,5), 126.1-136.4 (singlets, phenyl
carbons), 139.6 (s, PNP-aryl, C4), 145 (br t, Cipso), 163.9 (s, PNP-
aryl, C2,6). Assignment of the signals was confirmed by DEPT.
Anal. Calcd for C29H49P3F6NIr: C, 42.96; H, 6.09. Found: C, 44.93;
H, 6.69.
1
1JFP ) 713 Hz, PF6). H NMR (CD2Cl2): -19.83 (br t, 1H, H-Ir),
2
1.18 (m, 36H, P-C(CH3)3), 3.73 (m, 4H, CH2-P), 6.76 (br t, JHH
) 7.3 Hz, 1H, H-aryl), 6.85 (br t, 2H, H-aryl), 7.67 (d, 2JHH ) 7.0
Hz, 2H, H-aryl), 7.5 (d, JHH ) 8.0 Hz, 2H, H-PNP-aryl), 7.78 (t,
JHH ) 7.5 Hz, 1H, H-PNP-aryl). 13C{1H} NMR (CD2Cl2): 4 (s,
CH3CN) 28.5-30.1 (m, P-C(CH3)3), 36.9 (vt,1JPC ) 9.3 Hz
P-C(CH3)3), 38.1 (vt 1JPC ) 9.3 Hz, P-C(CH3)3), 39.5 (vt,1JPC )11.0
Hz, P-C(CH3)3), 122.4 (m, PNP-aryl C3,5), 125.3-135.6 (singlets,
phenyl carbons), 139.6 (s, PNP-aryl, C4), 125 (br t, Cipso), 163.9
(s, PNP-aryl, C2,6).
Reaction of [Ir(PNP)(COE)][PF6] (1) with m-Xylene. Forma-
tion of [Ir(PNP)(H)(m-xylyl)][PF6] (3). A solution (3 mL) of [Ir-
(PNP)(COE)][PF6] (1) (15 mg, 0.039 mmol) in m-xylene was heated
at 60 °C for 1 h, during which an orange precipitate of [Ir(PNP)-
(H)(m-xylyl)][PF6] (3) was formed in quantitative yield. The solid
was washed with pentane and with ether followed by high-vacuum-
drying for one night. 31P{1H} NMR of the CD2Cl2 solution of 3
revealed one product. 31P{1H} NMR (CD2Cl2): 53.7 (br s), -150.0
Reaction of [Ir(PNP)(H)(Ph)][PF6] (2) with CO. Formation
of [Ir(PNP)(H)(Ph)(CO)][PF6] (4). Into an acetone solution (3 mL)
of complex 2 (15 mg, 0.039 mmol) was bubbled a small excess of
CO for 1 min, during which the color changed to pale yellow. The
solvent was evaporated and the pale yellow solid was washed with
pentane and dried under vacuum overnight, resulting in pure
complex 4 in quantitative yield. 31P{1H} NMR (CD2Cl2): 39 (s),
-150.0 (heptet, 1JFP ) 713 Hz, PF6). 1H NMR (CD2Cl2): -7.6 (t,
2JPH ) 12 Hz, 2H, H-Ir), 1.19 (m, P-C(CH3)3, 36H), 3.8 (m, 4H,
1
1
2
(heptet, JFP ) 713 Hz, PF6). H NMR (CD2Cl2): -43.3 (vt, JPH
) 12 Hz, 1H, H-Ir), 1.2 (m, P-C(CH3)3, 36H), 2.4 (s, 6H, aryl-
CH3), 3.84 (m, 4H, CH2-P), 6.43 (s, aryl-H), 6.90 (br s, aryl-H),
6.95 (br s, aryl-H), 7.67 (d, JHH ) 7.8 Hz, 2H, H-PNP-aryl), 7.94
(t, JHH ) 7.8 Hz, 1H, H-PNP-aryl). Assignment of signals was
confirmed by COSY. 13C{1H} NMR (CD2Cl2): 21 (s, m-xylyl-
2
CH2-P), 6.85 (br t, JHH ) 7.3 Hz, 2H, H-aryl), 6.91 (br t, 2H,
2
H-aryl,), 7.67 (d, JHH ) 7.0 Hz, 1H, H-aryl), 7.5 (d, JHH ) 8.0
1
Hz, 2H, H-PNP-aryl), 7.78 (t, JHH ) 7.5 Hz, 1H, H-PNP-aryl).
CH3), 21.5 (s, m-xylyl-CH3), 28-30 (m, P-C(CH3)3), 37 (vt, JPC
13C{1H} NMR (CD2Cl2): 29.5-31.1 (m, P-C(CH3)3), 37.9 (vt,1JPC
1
) 9.3 Hz, P-C(CH3)3), 38 (vt, JPC ) 9.3 Hz, P-C(CH3)3), 39 (vt,
1
1JPC ) 11.1 Hz, CH2P), 122.4 (m, PNP-aryl C3,5), 125.1-135.3
(singlets, phenyl carbons), 139.6 (s, PNP-aryl, C4), 137 (br t, Cipso),
163.9 (s, PNP-aryl, C2,6). Assignment of signals was confirmed
by DEPT. Anal. Calcd for C31H53P3F6NIr: C, 44.38; H, 6.37.
Found: C, 44.45; H, 6.31.
) 9.3 Hz, P-C(CH3)3), 39.1 (vt JPC ) 9.3 Hz, P-C(CH3)3), 41.5
(vt,1JPC ) 11.0 Hz, P-C(CH3)3), 122.4 (m, PNP-aryl C3,5), 125.0-
135.8 (singlets, phenyl carbons), 139.6 (s, PNP-aryl, C4), 150 (br
t, Cipso), 164.9 (s, PNP-aryl, C2,6), 186 (br t, Ir-CO). Anal. Calcd
for C30H49P3F6NOIr: C, 42.95; H, 5.85. Found: C, 42.81; H, 6.05.
Reaction of [Ir(PNP)(H)(Ph)(CO)][PF6] (4) with Anisole.
When complex 4 (15 mg, 0.0178 mmol) was heated in anisole at
80 °C for 72 h, no reaction took place. Upon heating complex 4 in
anisole at 110 °C for 48 h, elimination of benzene occurred, forming
complex 5 in quantitative yield. The same reaction was repeated
in DMSO with the same results.
Synthesis of Ir(PNP)(D)(C6D5)][PF6] (2a). Complex 2a was
synthesized in two different routs: (a) A solution (2 mL) of complex
1 (15 mg, 0.039 mmol) in C6D6 was heated at 60 °C for 1 h, during
which the color changed to orange. The solvent was evaporated
and the orange solid was washed with pentane and dried under
vacuum overnight, resulting in pure complex 2a in quantitative
yield. The 31P NMR spectrum revealed a signal at 54 ppm. 2H NMR
revealed a pattern similar to that for complex 2. (b) The second
route is by an exchange reaction. A solution of complex 2 in C6D6
Reaction of [Ir(PNP)(H)(Ph)(CH3CN)][PF6] (6) with Benzene-
d6. Heating a solution of 6 (15 mg, 0.0176 mmol) in benzene-d6 at
70 °C for 48 h resulted in no obsevable reaction. However, heating

ortho C-H ActiVation of Haloarenes and Anisole
Organometallics, Vol. 25, No. 13, 2006 3209
3
3
at 85 °C for the same period of time resulted in an exchange
Hz,1H, aryl-H), 7.3 (d, JHH ) 7.0 Hz, 1H, aryl-H), 7.58 (d, JHH
3
reaction. The hydride and aromatic signals disappeared and the
) 6.6 Hz, 2H, H-PNP-aryl), 7.92 (t, JHH ) 6.6 Hz, 1H, H-PNP-
2
2
corresponding H signals appeared in H NMR.
aryl). 13C{1H} NMR (400 MHz, CDCl3): 28.8-29.6 (m, P-
C(CH3)3), 37.1-38.7 (m, CH2P), 121.9-142.0 (singlets, PNP-aryl
Reaction of [Ir(PNP)(COE)][PF6] (1) with Benzene-h6/
Benzene-d6. Determination of the Kinetic Isotopic Effect. A
solution (2 mL) of [Ir(PNP)(COE)][PF6] (1) (15 mg, 0.039 mmol)
in benzene-h6/benzene-d6 mixture in a molar ratio 1:1 was left at
room temperature (25 °C) for 48 h. The solution was evaporated
2
C3,5; C4 and phenyl), 163.5 (d, JPC ) 12.0 Hz, PNP-aryl C2,6),
163.3 (bs, Cipso). NOE was observed between the hydride at -34
ppm and the proton at 6.65 ppm. Selected 1H NMR data: 8b: -41
(br s, H-Ir, 1H), 7.1 (m, 1H, aryl-H), 7.35 (m, 2H, aryl-H), 7.5 (m,
3
1
1H, aryl-H). 7.58 (d, JHH ) 6.6 Hz, 2H, H-PNP-aryl), 7.92 (t,
and redissolved in CD2Cl2. H NMR was measured with a delay
3JHH ) 6.6 Hz, 1H, H-PNP-aryl). NOE effect was observed between
the hydride at -41 ppm and the protons at 7.35 ppm. 8c: -44 (br
of 25 s to ensure full spin relaxation. Two experiments were
conducted: one was done without the use of internal standard where
the calibration was performed relative to the benzylic protons of
the PNP ligand and the second was performed using TMS as
internal standard. Both experiments resulted in a 1:1 ratio of the
deuterio and protio complexes, indicating no kinetic isotope effect.
3
3
s, 1H, H-Ir), 7.0 (d, JHH ) 6.0 Hz, 2H, aryl-H), 7.45 (d, JHH
)
3
6.0 Hz, 2H, aryl-H), 7.58 (d, JHH ) 6.6 Hz, 2H, H-PNP-aryl),
3
7.92 (t, JHH ) 6.6 Hz, lH, H-PNP-aryl). NOE was observed
between the hydride at -44 ppm and the proton at 7.45 ppm. 31P
NMR (hydride coupled, CD2Cl2): 8c,8b: 53-53.5 (m, P-Ir-P).
Anal. Calcd for 8, C29H48P3F6NClIr: C, 41.21; H, 5.72. Found:
C, 41.31; H, 5.63.
Reaction of [Ir(PNP)(COE)][PF6] (1) with Fluorobenzene.
Formation of [Ir(PNP)(H)(C6H4F)][PF6] (7). A solution (2 mL)
of [Ir(PNP)(COE)][PF6] (1) (15 mg, 0.039 mmol) in fluorobenzene
was heated at 50 °C overnight, after which the solvent was
evaporated and the solid was washed with ether. 31P NMR of this
solid revealed the formation of o-, m-, and p-C-H-activated
complexes 7a,b,c, giving rise to singlets at 55.2, 58, and 52.3 ppm,
in a ratio of approximately 2:2:1, respectively. Further heating at
50 °C for 48 h resulted in a mixture of 7a and 7c in a ratio 1.8:1,
respectively. Prolonged heating at 80 °C for another 48 h resulted
in a slightly different mixture of the two isomers (2.3:1). For easier
assignment of signals to the corresponding isomers, first the mixture
of two isomers was characterized (ortho and para) using NOE and
H-COSY. 31P{1H} NMR (CD2Cl2) 7a: 55.2 (br d, 2JHP ) 11.8 Hz,
H-Ir-P-C(CH3)3), 7b: 58.0 (br d, 2JHP ) 10.3 Hz, H-Ir-P-C(CH3)3),
7c: 52.3 (br). 1H NMR (800 MHz, CD2Cl2) 7a: -41.00 (br t, 2JHP
) 11.5 Hz, H-Ir-P), 1.18 (m, 36H, PC(CH3)3), 3.50-3.90 (m, 4H,
CH2P), 6.70 (dt, 3JHH ) 8.0 Hz, 4JHH ) 1 Hz, 1H, aryl-H), 6.85 (br
t, 1JHH ) 8.0 Hz, 1H, aryl-H), 7.10 (dd, 3JHH ) 7.0 Hz, 4JHH ) 1.0
Hz, 1H, aryl-H), 7.30 (d, 3JHH ) 7.0 Hz, 1H, aryl-H), 7.66 (d, 3JHH
) 8.0 Hz, 2H, PNP-aryl H), 7.95 (t, 3JHH ) 8.0 Hz, 1H, PNP-aryl
H). NOE effect was observed between the hydride at -41.0 ppm
Reaction of [Ir(PNP)(COE)][PF6] (1) with Bromobenzene.
Formation of [Ir(PNP)(H)(C6H5Br)][PF6] (9). Complex 9 was
prepared analogously to complex 8. The ortho-activated bromoben-
zene complex 9a amounted to 70% of the mixture upon disappear-
ance of the starting complex 1 and the sole product after continued
heating at 60 °C for 3 h or longer.
Complex 9a. 31P NMR (hydride coupled, CD2Cl2): 49.0 (d, 2JPH
) 11.0 Hz). 1H NMR (CD2Cl2): -29 (t, 2JPH ) 11.0 Hz, 1H, H-Ir-
P), 1.25 (vt, 3JPH ) 7.0 Hz, 18H, C(CH 3)3P), 1.29 (vt, 3JPH ) 7.0
3
Hz, 18H, C(CH3)3P), 3.6-3.95 (m, 4H, CH2P), 6.73 (dd, JHH
)
8.0 Hz, 1H, 4JHH ) 1.0 Hz, H-aryl), 6.87 (br t, 3JHH ) 8.0 Hz, 1H,
4JHH ) 1.0 Hz, H-aryl), 7.09 (dt, 3JHH ) 7.0 Hz, 1H, H-aryl), 7.25
3
3
(br d, JHH ) 7.0 Hz, 1H, H-aryl), 7.58 (d, JHH ) 8.0 Hz, 2H,
H-PNP-aryl), 7.9 (t, JHH ) 8.0 Hz, 1H, H-PNP-aryl). 13C{1H}
3
NMR (CD2Cl2): 28.8-29.6 (m, P-C(CH3)3), 37-38.5 (m, CH2P),
120-142.3 (singlets, PNP-aryl C3,5; C4 and aryl), 164 (d, 2JPC
)
12 Hz, PNP-aryl C2,6), 163.5 (br s, Cipso). NOE was observed
between the hydride at -29 ppm and the proton at 7.25 ppm.
1
Selected NMR data for 9c: H NMR (CD2Cl2): -45 (br s, H-Ir,
2
and the proton at 7.10 ppm. 7b: -45.00 (br t, JHP ) 10.0 Hz ),
1H), 7.27 (d, 3JHH ) 5.0 Hz, aryl-H, 2H), 7.52 (d, 3JHH ) 5.0 Hz,
aryl-H, 2H). NOE was observed between the hydride at -45 ppm
and the proton at 7.52 ppm. 9b: -42 (br s, H-Ir, 1H), 7.3 (m, aryl-
H, 1H), 7.33 (m, aryl-H, 1H), 7.35 (m, aryl-H, 1H), 7.56 (m, aryl-
1.18 (m, 36H, P-C(CH3)3), 3.50-3.90 (m, 4H, CH2P), 6.9 (m, 1H,
aryl-H), 7.05 (m, 1H, aryl-H), 7.25 (t, 3JHH ) 7.0 Hz, 1H, aryl-H),
7.35 (br s, 1H, aryl-H), 7.7 (d, 3JHH ) 8.0 Hz, 2H, PNP-aryl), 8.00
3
3
3
(t, JHH ) 8.0 Hz, 1H, PNP-aryl). 7c: -42.5 (br, 1H), 1.18 (m,
H, 1H), 7.58 (d, JHH ) 6.6 Hz, 2H H-PNP-aryl), 7.92 (t, JHH )
3
36H, P-C(CH3)3), 3.50-3.90 (m, 4H, CH2P), 7.25 (d, JHH ) 7.0
6.6 Hz, lH, H-PNP-aryl). NOE was observed between the hydride
at -42 ppm and the proton at 7.56 ppm. 9b,9c: 51-51.5 (m, P-Ir-
P). Anal. Calcd for 9, C29H48P3F6NBrlIr: C, 39.15; H, 5.44.
Found: C, 39.09; H, 5.39.
Hz, 1H, aryl-H), 7.43 (d, 3JHH ) 7.0 Hz, 1H, aryl-H), 7.5 (d, 3JHH
3
) 8.0 Hz, 2H, H-PNP-aryl), 7.90 (t, JHH ) 8.0 Hz, 1H, PNP-
aryl). Assignment of the signals was done by COSY. NOE effect
was observed between the hydride at -42.5 ppm and protons at
7.25 ppm. Anal. Calcd for 7, C29H48P3F7NIr: C, 42.02; H, 5.84.
Found: C, 42.09; H, 5.81.
Reaction of [Ir(PNP)(COE)][PF6] (1) with Anisole. Formation
of [Ir(PNP)(H)(o-C7H8O)][PF6] (10). A solution (1 mL) of [Ir-
(PNP)(COE)][PF6] (1) (15 mg, 0.039 mmol) in anisole was heated
at 60 °C for 3 h. The solvent was evaporated, and the resulting
orange solid was washed consecutively with pentane and ether,
resulting in complex 10 in almost quantitative yield. Complexes
resulting from meta- and para-activated anisole were observed in
trace amounts. 31P{1H} NMR (hydride coupled, CD2Cl2): 56.60
(d, 2JPH ) 13.0 Hz). 1H NMR (CD2Cl2) -42.00 (t, 2JHP ) 13.0 Hz
H-Ir-P), 1.2 (m, 36H, P-C(CH3)3), 3.55 (s, CH3O; 3H), 3.7-4.0
Reaction of [Ir(PNP)(COE)][PF6] (1) with Chlorobenzene.
Formation of [Ir(PNP)(H)(C6H4Cl)][PF6] (8). A solution of [Ir-
(PNP)(COE)][PF6] (1) (15 mg, 0.039 mmol) in chlorobenzene (2
mL) was heated at 60 °C for 3 h. The solvent was evaporated, and
the resulting orange solid was washed consecutively with pentane
and ether, resulting in complex 8a in quantitative yield. Monitoring
the reaction by 31P NMR at 50 °C revealed after 12 min 10%
conversion of 1, resulting in a ratio 8a:8b:8c of 4:2:1. Continued
heating of complex 1 for 1 h at 50 °C resulted in formation of a
mixture containing 60% of the ortho-activated chlorobenzene 8a,
30% of the meta 8b, and 10% of the para-activated one 8c. Further
heating of this mixture for 48 h resulted in a sole product, the ortho-
activated chlorobenzene complex 8a.
(m, 4H, CH2P), 6.4 (br d, 3JHH ) 7.7 Hz, aryl-H), 6.7 (br t, JHH
6.8 Hz, aryl-H), 7.25 (d, 3JHH ) 6.8 Hz, aryl-H), 7.6 (br t, 3JHH
)
)
7.7 Hz, aryl-H), 7.65 (d, 3JHH ) 7.7 Hz, 2H, H-PNP-aryl), 7.92 (t,
3JHH ) 7.7 Hz, 1H, H-PNP-aryl). 13C{1H} NMR (CD2Cl2): 27.8-
29.9 (m, P-C(CH3)3), 35 (m, CH2P), 60 (m, CH3OPh), 122.0 (m,
PNP-aryl C3,5), 125.2-135.6 (singlets, phenyl carbons), 138.5 (s,
PNP-aryl, C4), 138 (br t, Cipso), 162.0 (s, PNP-aryl, C2,6). Anal.
Calcd for C30H51P3F6NOIr: C, 42.85; H, 6.11. Found: C, 43.06;
H, 6.03.
Complex 8a. 31P NMR (hydride coupled, CD2Cl2): 52.00 (d,
1
2
2JPH ) 13.0 Hz). H NMR (CD2Cl2): -34.00 (t, JHP ) 13.8 Hz,
1H, H-Ir-P), 1.2 (m, 36H, P-C(CH3)3), 3.2-3.8 (m, 4H, CH2P),
3
4
6.65 (dd, JHH ) 8.0 Hz, JHH ) 2.0 Hz, 1H, aryl-H), 6.91 (br t,
Lack of Reaction of [Ir(PNP)(H)(C6H4Cl)][PF6] (8a) with an
Anisole/Chlorobenzene mixture (1:1). A solution (1 mL) of [Ir-
3
4
JHH ) 8.0 Hz, 1H, aryl-H), 7.04 (dt, JHH ) 7.0 Hz, JHH ) 1.0

3210 Organometallics, Vol. 25, No. 13, 2006
Ben-Ari et al.
(PNP)(H)(C6H4Cl)][PF6] (8a) in anisole and chlorobenzene (1:1
molar ratio) was heated at 55 °C for 24 h. The resulting pale orange-
like solid was washed consecutively with pentane and ether,
residue revealed four signals, which correspond to the ortho-
activated anisole 10a, the meta- and para-isomers of activated
chlorobenzene, bromobenzene, 8b,c, 9b,c, and anisole 10b,c as well
as ortho-activated chlorobenzene 8a and ortho-activated bromoben-
zene 9a (vide supra) in a ratio of 42%, 15%, 26%, and 17%,
respectively. 1H NMR revealed four hydridic signals, which
corresponded to the four different complexes at a ratio similar to
that shown by 31P NMR. The hydrides were allowed to reach full
relaxation (D1 ) 45 s).
1
resulting in complex 8a, the starting complex. 31P{1H} NMR, H
NMR: vide supra.
Lack of Reaction of [Ir(PNP)(H)(o-C7H8O)][PF6] (10) with
a Mixture of Anisole/Chlorobenzene (1:1). A solution (1 mL) of
[Ir(PNP)(H)(o-C7H8O)][PF6] (10) in anisole and chlorobenzene (1:1
molar ratio) was heated at 55 °C for 24 h. The resulting orange
solid was washed consecutively with pentane and ether, resulting
in complex 10, the starting complex. 31P{1H} NMR, 1H NMR: vide
supra.
Acknowledgment. This work was supported by the Israel
Science Foundation, Jerusalem, Israel, by the German Federal
Ministry of Education and Research (BMBF) within the
framework of German-Israeli cooperation (DIP), by the Minerva
Foundation, Munich Germany, by the Lise Meitner-Minerva
Center for Computational Quantum Chemistry, and by the Helen
and Martin Kimmel Center for Molecular Design. D.M. is the
Israel Matz Professor of Organic Chemistry, and J.M.L.M. is
the Baroness Thatcher Professor of Chemistry.
Reaction of [Ir(PNP)(COE)][PF6] (1) with a Mixture of
Anisole/Chlorobenzene/Bromobenzene (1:1:1). (a) A solution (1
mL) of [Ir(PNP)(COE)][PF6] (1) in anisole, bromobenzene, and
chlorobenzene in 1:1:1 molar ratio was heated at 55 °C for 24 h,
and the solvent was evaporated. 31P NMR of the residue revealed
three signals, which correspond to the ortho-activated anisole 10a,
ortho-activated chlorobenzene 8a, and ortho-activated bromoben-
zene 9a (vide supra) in a ratio of 35%, 32%, and 33%, respectively.
1H NMR revealed three hydride signals, which corresponded to
the three different complexes in a ratio similar to that shown by
31P NMR. The hydrides were allowed to reach full relaxation (D1
Supporting Information Available: CIF files containing X-ray
crystallographic data for complexes 1, 6, and 10a. XYZ coordinates
of all the computed complexes. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
) 45 s). 31P {1H} NMR, H NMR: vide supra.
(b) A solution (1 mL) of [Ir(PNP)(COE)][PF6] (1) in anisole,
bromobenzene, and chlorobenzene in 1:1:1 molar ratio was heated
at 70 °C for 48 h, and the solvent was evaporated. 31P NMR of the
OM060078+