Cleavage of ether, ester, and tosylate C(sp3)-O bonds by an iridium complex, initiated by oxidative addition of C-H bonds. Experimental and computational studies

Article

pubs.acs.org/JACS

Cleavage of Ether, Ester, and Tosylate C(sp³)−O Bonds by an Iridium

Complex, Initiated by Oxidative Addition of C−H Bonds.

Experimental and Computational Studies

Sabuj Kundu, Jongwook Choi, David Y. Wang, Yuriy Choliy, Thomas J. Emge, Karsten Krogh-Jespersen,*

and Alan S. Goldman*

Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, New Brunswick, New Jersey 08903,

United States

S

* Supporting Information

ABSTRACT: A pincer-ligated iridium complex, (PCP)Ir

(PCP = κ³-C₆H₃-2,6-[CH₂P(t-Bu)₂]₂), is found to undergo

oxidative addition of C(sp³)−O bonds of methyl esters (CH₃−

O₂CR′), methyl tosylate (CH₃−OTs), and certain electron-

poor methyl aryl ethers (CH₃−OAr). DFT calculations and

mechanistic studies indicate that the reactions proceed via

oxidative addition of C−H bonds followed by oxygenate

migration, rather than by direct C−O addition. Thus, methyl

aryl ethers react via addition of the methoxy C−H bond,

followed by α-aryloxide migration to give cis-(PCP)Ir(H)-

(CH₂)(OAr), followed by iridium-to-methylidene hydride migration to give (PCP)Ir(CH₃)(OAr). Methyl acetate undergoes

C−H bond addition at the carbomethoxy group to give (PCP)Ir(H)[κ²-CH₂OC(O)Me] which then aﬀords (PCP-

CH₂)Ir(H)(κ²-O₂CMe) (6-Me) in which the methoxy C−O bond has been cleaved, and the methylene derived from the

methoxy group has migrated into the PCP C_ipso−Ir bond. Thermolysis of 6-Me ultimately gives (PCP)Ir(CH₃)(κ²-O₂CR), the

net product of methoxy group C−O oxidative addition. Reaction of (PCP)Ir with species of the type ROAr, RO₂CMe or ROTs,

where R possesses β-C−H bonds (e.g., R = ethyl or isopropyl), results in formation of (PCP)Ir(H)(OAr),

(PCP)Ir(H)(O₂CMe), or (PCP)Ir(H)(OTs), respectively, along with the corresponding oleﬁn or (PCP)Ir(oleﬁn) complex.

Like the C−O bond oxidative additions, these reactions also proceed via initial activation of a C−H bond; in this case, C−H

addition at the β-position is followed by β-migration of the aryloxide, carboxylate, or tosylate group. Calculations indicate that the

β-migration of the carboxylate group proceeds via an unusual six-membered cyclic transition state in which the alkoxy C−O bond

is cleaved with no direct participation by the iridium center.

Oxidative addition is one of the most fundamental processes

in organometallic chemistry and is a key step in many

transformations catalyzed by transition-metal complexes. The

oxidative addition of carbon−halogen and carbon−hydrogen

bonds has been and continues to be extensively studied.³⁷⁻⁴⁰In

a seminal study in 1978, Ittel and Tolman showed that the

Fe(0) intermediate Fe(dmpe)₂(dmpe = Me₂PCH₂CH₂PMe₂)

undergoes oxidative addition reactions with anisole or methyl

benzoate, aﬀording Fe(dmpe)₂(CH₃)(OPh) or

Fe(dmpe)₂(CH₃)(O₂CPh), respectively (eq 1).^41,42Since

then, however, although several examples have been reported

of oxidative addition with cyclic ethers or sp²C−O

bonds,^15,43−48reports of oxidative addition of sp³C−O

bonds in ethers and esters have been extremely limited.⁴⁹

In 2008, Chirik et al. reported that (^iPrPDI)Fe(N₂)₂cleaves

the C−O bonds in alkyl-substituted esters via a binuclear

oxidative addition (eq 2).⁴⁷Other notable examples of

1. INTRODUCTION

Carbon−oxygen bonds are ubiquitous in nature and are among

the most fundamental linkages in organic chemistry. Accord-

ingly, transformations of C−O bonds to other functional

groups and the removal of oxygen-containing groups rank

among the most attractive challenges in modern catalysis.

Increasing interest in the conversion of biomass to fuel as an

alternative to petroleum-based feedstocks has spurred research

in C−O bond activation as this will require removal of oxygen

from the “overfunctionalized” biomass.¹⁻¹⁰Alternatively, partial

deoxygenation of biomass could lead to higher value chemical

products. Recent progress in transition-metal catalyzed C−C or

C−X bond coupling reactions, achieved by activating sp²or

benzylic-type sp³C−O bonds in ethers¹¹⁻²³and esters²⁴⁻³²has

also elevated interest in C−O bond activation. However, while

numerous studies have been reported involving cleavage of

activated C−O bonds such as those in allyl esters or

ethers,³³⁻³⁶the catalytic activation of sp³C−O bonds has

been relatively unexplored and remains a signiﬁcant challenge.

Received: December 31, 2012

Published: March 7, 2013

5127

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

In an initial experiment in this study, the reaction of (PCP)Ir

with anisole, the simplest methyl aryl ether, was investigated.

Treatment of 1 or 1′ with anisole (a; 1.05 equiv) in p-xylene-

d

₁₀at room temperature resulted in an immediate color change

from dark red to red (eq 3; PR₂= P^tBu₂for all structures

shown). Selectively proton-decoupled ³¹P NMR spectroscopy

revealed one major signal, a doublet at 68.2 ppm (J_HP= 12.2

Hz). A hydride resonance in the ¹H NMR spectrum appears as

a sharp triplet (J_HP= 14.4 Hz) at −45.03 ppm, characteristic of

a square pyramidal ﬁve-coordinate complex in which the

hydride occupies the axial site.⁶⁷These signals were assigned to

the C−H addition product 2a. The hydride resonance is sharp

at room temperature, in contrast with the previously reported

aryl hydride complex (PCP)Ir(Ph)(H) in which the hydride

signal is sharp only at or below ca. −40 °C, due to rapid

reversible reductive elimination of benzene.⁵⁶DFT calculations

(see Supporting Information, SI, for details) indicate that C−H

bond addition to give complex 2a is 3.7 kcal/mol more

exergonic than benzene C−H addition, presumably at least in

part because of a weak favorable interaction between the

methoxy group and the iridium center in 2a (computed Ir−O

distance = 3.02 Å).

Heating the reaction mixture for 3 h at 90 °C results in the

formation of cyclometalated product (PCP)Ir(κ²-CH₂OC₆H₄)

(3a) in quantitative yield (eq 3). Characterization of 3a by

NMR spectroscopy and X-ray crystallography has been

reported,⁶⁴and analogous cyclometalated adducts have been

reported previously by Wilkinson,⁶⁸Carmona,^67,69,70and

Perutz and co-workers.⁷¹DFT calculations indicate that the

key step in the reaction of 2a to aﬀord 3a (and H₂), cleavage of

the methoxy C−H bond, proceeds via an Ir(V) intermediate

which readily loses H₂. The reaction is calculated to be

endergonic (ΔG_rx= 7.0 kcal/mol at P = 1 atm, T = 25 °C), but

presumably the released dihydrogen is readily scavenged by the

excess oleﬁnic hydrogen acceptor present under the reaction

conditions.

In an eﬀort to prevent formation of products like 2a and 3a,

and thereby allow the formation of C−O addition products

instead, we investigated the reaction of 1 with 2,6-dimethyl

anisole and 3,5-dimethyl anisole (in which the ortho C−H

bonds are substituted or sterically hindered, respectively).

Heating the reaction mixture at 100 °C, however, only

produced a complex mixture of unidentiﬁed products. In

order to favor C−O addition, we then turned to electron-

withdrawing triﬂuoromethyl groups to block the ortho C−H

bonds. Despite the steric bulk of such groups, however, when 1

equiv. of (PCP)Ir(TBV)(H) (1) was treated with 3,5-

bis(triﬂuoromethyl)anisole (b) in p-xylene at 40 °C, the

ortho C−H activation product 2b formed cleanly (eq 4; see SI

for NMR spectroscopic characterization). Apparently, although

the presence of an ortho methyl group strongly disfavors

addition to (PCP)Ir,⁵⁶the even greater steric eﬀects of the

unactivated sp³C−O bond cleavage include Carmona and

Paneque’s seminal discovery of the rearrangement reactions of

methyl aryl ethers by a tris(pyrazolyl)borate iridium complex,⁵⁰

reports by Ozerov and Grubbs of cleavage of the tert-butyl-

oxygen and benzyl-oxygen bonds of methyl ethers by a PNP-

pincer Ir complex,^51,52and Jones’ report of the cleavage of the

methoxy C−O bond of methyl-acetate by a diphosphine

platinum complex.³²

The pincer-ligated complex (PCP)Ir (PCP = κ³-C₆H₃-2,6-

[CH₂P(t-Bu)₂]₂) and its derivatives are the most eﬃcient

alkane dehydrogenation catalysts developed to date.⁵³⁻⁵⁵This

species also activates sp²C−H bonds of arenes or vinyl

moieties,⁵⁶C−X (X = halogen) bonds,⁵⁷O−H bonds of water

and alcohols,^58,59and the N−H bond of aniline.⁶⁰Recently, we

reported the oxidative addition of C(sp³)−F bonds by

(PCP)Ir.⁶¹

As part of our eﬀorts to activate unreactive bonds in small

molecules, we have investigated C−O bond cleavage by

(PCP)Ir. Herein, we report that reactions of (PCP)Ir with

alkyl esters, certain electron-poor alkyl aryl ethers, and alkyl

tosylates result in oxidative addition or other cleavage reactions

of the C(sp³)−O bond. Results from DFT calculations as well

as experimental evidence indicate that C−O bond cleavage in

these substrates proceeds via initial activation of a C−H bond,

rather than direct C−O insertion or nucleophilic attack. These

mechanistic ﬁndings have obvious implications for the

microscopically reverse C−O bond formation reactions.^62,63

Parts of this study have been previously communicated.⁶⁴

2. RESULTS AND DISCUSSION

2.1. Reactions of (PCP)Ir with Methyl Oxygenates

MeOAr, MeOTs, MeOAc, and MeO₂CPh. 2.1.1. Reaction of

(PCP)Ir with Methyl Aryl Ethers. The reactive 14e⁻three-

coordinate species (PCP)Ir may be generated by loss of oleﬁn

from (PCP)Ir(TBV)(H) (1, TBV = 3,3-dimethyl-1-butenyl) or

(PCP)Ir(NBE) (1′, NBE = 2-norbornene), as previously

reported;^60,65,66these two approaches have been used

interchangeably in this study.

5128

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

NMR spectroscopy as well as X-ray crystallography revealed the

quantitative formation of (PCP)Ir(CH₃)(OC₆F₅) (4c) result-

ing from oxidative addition of the C(sp³)−O bond;⁶⁴no side-

products or intermediates were observed during the reaction

process. 4-methoxy-2,3,5,6-tetraﬂuorotoluene (d) likewise

reacted with 1 immediately at room temperature and

analogously aﬀorded (PCP)Ir(CH₃)(2,3,5,6-tetraﬂuoro-4-

methyl phenoxide) (4d) in quantitative yield.⁶⁴

To probe the scope of the C−O addition, reactivity with

various other ﬂuorine-substituted methyl aryl ethers was

investigated; results are summarized in Table 1. All reactions

were carried out at 100 °C in p-xylene solution, except

reactions of arenes c and d which were conducted at room

temperature. Arenes e and f, also ﬂuorinated ortho to the

methoxy group, gave C−O oxidative adducts (4) as the major

products although yields were lower than obtained with the

more highly ﬂuorinated arenes c and d. In the case of 2,4,6-

triﬂuoroanisole (g), two rotamers of the aryl C−H adduct,

(PCP)Ir(H)(2,4,6-triﬂuoro-3-methoxyphenoxide), were the

only products obtained, even after heating for two days at

100 °C (eq 5); no C−O oxidative addition product, viz.

(PCP)Ir(CH₃)(2,4,6-triﬂuorophenoxide), was observed. This

result is consistent with the favorable eﬀect of ortho-F

substituents (of which there are two in this case) on the

stability of metal aryl complexes as demonstrated by Perutz,

Jones, Eisenstein, and co-workers.⁷¹⁻⁷⁵

triﬂuoromethyl group are more than compensated for by its

favorable electronic eﬀects.⁷¹⁻⁷⁵

In view of the known reversibility of aryl C−H activation by

(PCP)Ir,⁵⁶we subjected complex 2b to thermolysis; to our

gratiﬁcation, after 5 h at 80 °C the formation of (PCP)Ir-

(CH₃)(3,5-bis(triﬂuoromethyl)phenoxide) (4b) was observed

in 65% yield, along with complex 3b in 35% yield (eq 4).⁶⁴

Both 3b and 4b were characterized by NMR spectroscopy, and

the structure of 4b was conﬁrmed by X-ray crystallography.⁶⁴

To our knowledge, complex 4b is only the second reported

example of intermolecular oxidative addition to a transition

metal of an alkyl carbon−oxygen bond in simple ethers (the

ﬁrst example being that in eq 1).^41,42

The presence of C−H bonds ortho to the methoxy group

(substrates h, i, and j) resulted in complete conversion to

cyclometalated C−H addition products analogous to complex

3a, which was formed in the case of anisole. The structures of

complexes 3h and 3j were conﬁrmed by X-ray crystallography

(see Figures S1 and S2 of the SI). The stability of these

The formation of cyclometalated product 3b led us to

investigate the use of substrates in which ortho C−H bond

activation is fully prevented, while the electron-poor character

of the aryl ring is maintained or increased. When 1 was treated

with 1.05 equiv. of pentaﬂuoroanisole (c) at room temperature,

an immediate color change from dark red to red was observed.

a

Table 1. Oxidative Addition of the C−O Bond in Various Fluorine-Substituted Methyl Aryl Ethers

a

b

General reaction conditions: (PCP)Ir(TBV)(H) (20−25 μmol) + 1.05 equiv. of substrate in p-xylene solution at 100 °C for 15−20 h. Room

c

temperature. 30 h.

5129

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

A kinetic isotope eﬀect, KIE_obs= k_ArOCH3/k_ArOCD3= 4.3(3)

for the overall reaction 6, was determined at room temperature

from the product distribution as an average of three

independent experiments. We note that this experiment reﬂects

information regarding only the nature of the transition state for

the critical product-determining step,⁷⁶speciﬁcally only the

relative energies of the two isotopomeric TSs (relative to the

free ethers). In particular, it may be noted that the observed

KIE is independent of whether or not (PCP)Ir and ether are

complexed in the lowest energy (resting) state. This point is

illustrated in Scheme 1, which indicates that the relative rates of

reaction of CH₃OAr and CD₃OAr will depend (as per the

Curtin-Hammett principle) only on ΔΔG^‡(eq 7); because of

the nature of a competition experiment, ΔG^‡for formation of

either isotopomer relates to the same resting state (or states).

derivatives of 3 is presumably increased by the ﬂuorine

substitution of the ring.⁷¹⁻⁷⁵

‡

ΔG_CX3= G_TS‐X− G_RS(X = H or D)

(7a)

2.1.2. Mechanism of Ether C−O Bond Addition. As noted

above, reports of oxidative addition of ether C(sp³)−O bonds

are limited, the only previous intermolecular example being the

Ittel-Tolman system (eq 1). In that case, the reaction probably

proceeds via an S_N2-type nucleophilic attack by the very

electron-rich Fe(0) center on the electrophilic anisole methyl

group, thus accounting for formation of the trans addition

product.^28,29Milstein and co-workers proposed two diﬀerent

activation mechanisms for C−O bond cleavage in aryl methyl

ether pincer precursors by Rh, Pd, and Ni complexes.^45,46They

suggested that electron-rich, nucleophilic Rh complexes cleave

sp³C−O bonds via a concerted, three-centered transition state

(direct C−O addition), whereas electron-deﬁcient Pd or Ni

complexes were proposed to cleave sp³C−O bonds via

electrophilic interaction with the oxygen atom.

For the current set of reactions, nucleophilic attack did not a

priori seem likely as (PCP)Ir is not expected to be a strong

nucleophile. Rather, the reactivity of (PCP)Ir is dominated by

the vacant coordination site trans to the aryl group (essentially

an empty site in a d⁸square planar conﬁguration), with some

contribution from the ﬁlled d-orbitals of π-symmetry. The most

obvious mechanism consistent with this electronic conﬁg-

uration would be the direct, concerted, insertion of Ir into the

C−O bond. However, in view of the proclivity of (PCP)Ir

toward addition of C−H bonds, we investigated the eﬀect of

isotopic substitution of the methoxy hydrogen atoms.

ΔΔG^‡= (G_TS‐D− G_RS) − (G_TS‐H− G_RS

= G_TS‐D− G_TS‐H

)

(7b)

The high observed kinetic isotope eﬀect, k_ArOCH3/k_ArOCD3

=

4.3(3), is inconsistent with either a direct C−O addition or an

S_N2-type mechanism.^77,78Instead, it implies that the rate-

determining transition state features one or more C−H bonds

that have been cleaved or at least signiﬁcantly weakened. The

high KIE_obsindicates, particularly in view of the propensity of

(PCP)Ir to undergo oxidative addition of C−H bonds, that the

C−O addition proceeds, rather unexpectedly, via addition of a

methoxy C−H bond. The simplest pathway to proceed from a

C−H addition intermediate to the C−O addition product is

that shown in Scheme 2.

DFT calculations (using M06 functionals, see Computational

Methods) on the full (nontruncated) complexes were

conducted for both the mechanism proposed in Scheme 2

and for a direct C−O addition mechanism. Using 4-MeO-p-

C₆F₄Me as the substrate aryl ether (substrate d, Table 1), direct

oxidative addition (3-centered transition state) of (PCP)Ir to

the O−Me bond would encounter a barrier that is quite high

(G_TS= 33.4 kcal/mol relative to the ether and free three-

coordinate (PCP)Ir). The reaction pathway proposed in

Scheme 2 is calculated to be more favorable by almost 18

kcal/mol (Figure 1).

A mixture of 4-methoxy-2,3,5,6-tetraﬂuorotoluene and its

deuterated methoxy analogue (at least 7 fold excess of each

substrate) was added to a p-xylene solution of 1. An immediate

color change from dark red to red was observed and two

resonances due to (PCP)Ir(CH₃)(2,3,5,6-tetraﬂuoro-4-methyl-

phenoxide) (4d) and (PCP)Ir(CD₃)(2,3,5,6-tetraﬂuoro-4-

methylphenoxide) (4d-d₃), respectively, appeared in the ³¹P

NMR spectrum at δ 47.26 and 47.48 ppm (eq 6). The product

ratio of 4d to 4d-d₃was determined by integration of the two

³¹P NMR resonances and by integration of the Ir−CH₃and the

Initial activation of a methoxy group sp³C−H bond

(transition state G_TS= 8.2 kcal/mol; all values of G are given

relative to free (PCP)Ir and the free organic species) yields a

ﬁve-coordinate alkyl hydride intermediate (G = −4.5 kcal/mol).

The TS for α-aryloxy elimination from this species is calculated

to have a free energy G_TS= 14.6 kcal/mol and leads to the

methylidene hydride aryloxide (G = 2.8 kcal/mol). Subsequent

1,2-migration of hydride from Ir to the carbene (G_TS= 15.8

kcal/mol) aﬀords the observed product 4d (G = −26.4 kcal/

mol). The diﬀerence in TS energies calculated for the 1,2-

hydride migration (15.8 kcal/mol) and for α-aryloxy elimi-

nation (14.6 kcal/mol) is small and probably not reliable

considering the level of precision expected of the calculations

when comparing species of a fairly diﬀerent nature. Relative to

the reactant (PCP)Ir(TBV)(H) (1) (G = −4.7 kcal/mol), the

calculated overall free energy activation barrier is thus 19−21

kcal/mol, which is consistent with a reaction that proceeds

rapidly at room temperature (predicted half-life of approx-

imately 10²s at 25 °C). In contrast, the calculated barrier for

1

IrOC₆F₄CH₃signals in the H NMR spectrum.

5130

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

a

Scheme 1. Hypothetical Free Energy Diagrams for the CH₃OAr/CD₃OAr Competition Reaction

a

In the diagram on the left, the implied resting state for iridium is not complexed with the ether (e.g., (PCP)Ir(TBV)(H)) while on the right, iridium

and ether are associated in the resting state [e.g., (PCP)Ir(CH₃OAr) or (PCP)Ir(H)(CH₂OAr)]. In either case, the observed KIE will reﬂect only

the energy diﬀerence between the isotopomeric TSs for the rate-determining step and the free ethers (ΔΔG^≠= G_TSD− G_TSH).

Scheme 2. Proposed Mechanism for Formation of 4 from 1 and Methyl Aryl Ethers

Figure 1. Calculated Gibbs free energies (kcal/mol) for reaction of (PCP)Ir and CH₃−OAr, Ar = p-C₆F₄Me. Values of G are given relative to free

(PCP)Ir and CH₃−OAr. The free energies correspond to a reference state of 1 M concentration for each species participating in the reaction and T =

298.15 K. Potential energies, enthalpies, entropies and free energies for reaction of (PCP)Ir + CH₃−OAr, Ar = p-C₆F₄Me are summarized in Table

S1 of the SI.

direct C−O addition (relative to (PCP)Ir(TBV)(H)) is G^‡=

38.1 kcal/mol (corresponding to a half-life of ca. 10¹⁵s).

The free energies calculated for the 1,2-hydride migration

and α-aryloxy elimination TSs diﬀer only by 1.2 kcal/mol

(Figure 1). The computed KIEs, however, are signiﬁcantly

dependent on which TS eﬀectively determines the rate, with

k_ArOCH3/k_ArOCD3values of 4.24 or 7.23 predicted depending on

whether α-aryloxy elimination or 1,2-hydride migration is rate-

determining. The former value is in excellent agreement with

the experimentally determined KIE_obs, k_OCH3/k_OCD3= 4.3(3); it

therefore seems likely that α-aryloxy migration is in fact rate-

determining.⁷⁹The overall reaction may thus be viewed as a

pre-equilibrium between (PCP)Ir (or its precursor, 1) and

(PCP)Ir(CH₂OAr)(H) (eq 8), followed by rate-determining

1,2-OAr migration (eq 9). Accordingly, the overall KIE_obsmay

be regarded as the product of the EIE for eq 8 and the KIE for

eq 9.⁸⁰

The EIE (at 25 °C) calculated for formation of the

aryloxymethyl hydride (EIE₈= K_8‑H/K_8‑D) is 3.12 (ΔΔG₈=

−0.674 kcal/mol) while KIE₉(k_9‑H/k_9‑D) is calculated as 1.36

(ΔΔG^‡= −0.182 kcal/mol). The sum of these energy

9

diﬀerences is necessarily equal to the diﬀerence of the energies

of the isotopomeric TSs for OAr migration relative to the free

ethers (see Scheme 1; ΔΔG^‡= ΔΔG₈+ ΔΔG^‡= −0.856

9

5131

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

which is indicative of C−H addition occurring prior to carbonyl

oxygen coordination.⁹⁴

Quite surprisingly, heating the resulting solution of 5 at 80

°C for 5 h resulted in formation of 6-Ph, in which a methylene

unit derived from the carbomethoxy group has apparently

inserted into the PCP aryl-Ir bond (eq 10). The Ir-CH₂-Ar

group of 6-Ph gave rise to a triplet at δ 2.28 ppm (³J_PH= 9.8

1

Hz) in the H NMR spectrum, and a triplet at δ −17.67 ppm

(²J_CP= 5.2 Hz) in the ¹³C NMR spectrum. These values are

similar to those reported by Milstein et al. for complex 8 (eq

kcal/mol), corresponding to an overall calculated KIE_obsof

4.24. EIEs for alkane C−H addition at ambient temperature are

typically⁸¹above 2.0 but our calculated value of EIE₈= 3.12 is

unusually high. We attribute this to particularly low Ir−H(D)

bending frequencies due to the shallow energy surface for

deformation of the ligand arrangement in the equatorial plane

of the ﬁve-coordinate d⁶complex.⁸²The value of 1.36 for KIE₉

may seem high for a secondary KIE but is well precedented by

organic S_N1 (solvolysis) reactions for which values above 1.2

per H/D atom are common.⁸³⁻⁹⁰It seems quite plausible that

OAr migration in eq 9 (KIE calculated as 1.17 per H/D) would

aﬀord a secondary KIE comparable to that resulting from the

dissociation of oxygen-bound anions (e.g., tosylates) from the

11).^95,96In the H NMR spectrum the hydride appeared as a

1

triplet at δ −32.76 ppm (²J_PH= 12.5 Hz). Single crystal X-ray

analysis conﬁrmed the structure of 6-Ph (Figure 2).

Heating complex 6-Ph at 125 °C in p-xylene for 9 h led to

the formation of the methyl benzoate C(sp³)-O oxidative

addition complex 7-Ph (eq 10). The Ir-CH₃signal of 7-Ph

appears as a triplet at δ 1.45 ppm (³J_PH= 4.8 Hz) and as a

triplet at δ −29.1 ppm (²J_CP= 4.6 Hz) in the ¹H and ¹³C NMR

spectra, respectively. Complex 7-Ph was crystallized from n-

hexane, and the structure was conﬁrmed by X-ray crystallog-

raphy (Figure 2).

Milstein has reported that complex 8, with a (PCP-CH₂)IrH

motif analogous to that of 6-Ph, aﬀords the iridium methyl

complex 9 upon heating (eq 11).⁹⁵The formation of 7-Ph from

6-Ph can be explained analogously, as resulting from reductive

elimination of (PCP−CH₂) and H from 6-Ph, followed by

C−C bond cleavage to give complex 7-Ph.

corresponding alkyl derivatives. Thus, the value of KIE_obs

,

4.3(3), in conjunction with the DFT calculations, is supportive

of the pathway of Scheme 2 with 1,2-aryloxide migration as the

rate-determining step.

2.1.3. Reactions of (PCP)Ir with Methyl Esters. As with

ethers, certain esters underwent clean oxidative addition of

C−O bonds by (PCP)Ir. In all such cases, the strongerand

typically much less reactiveof the two ester C−O single

bonds was cleaved, i.e., the alkoxy C−O bond, rather than the

acyl C−O linkage.^{27,32,91−93}

Since the acyl carbon lacks C−H bonds, this observation is of

course consistent with a reaction mechanism related to that

proposed for ethers and discussed above. Moreover, and in

contrast with the ether additions, for some of the ester C−O

addition reactions we have been able to observe and isolate

apparent intermediates in a pathway initiated by C−H

activation.

The room-temperature reaction of 1 with methyl acetate, like

that with methyl benzoate, resulted in a cyclometalated

product, but in contrast to the methyl benzoate reaction,

C−H bond addition occurred at the carbomethoxy group,

aﬀording complex 10-cis (eq 12). The hydride resonance of 10-

Addition of methyl benzoate (1.15 equiv.) to 1 in p-xylene-

d₁₀at room temperature resulted in the rapid formation of

(PCP)Ir(H)[κ²-(C₆H₄C(O)OMe)], 5, the product of addition

of the aryl C−H bond ortho to the ester group and

coordination of the ester carbonyl oxygen (eq 10). The

1

cis appears in the H NMR spectrum as a triplet (²J_PH= 16.1

Hz) at δ −25.98 ppm; this relatively upﬁeld shift is consistent

with the coordination geometry in which the hydride is trans to

the ester carbonyl oxygen.⁹⁴The resonances of the IrCH₂O

group appear at δ 67.19 ppm (t, ²J_PC= 6.6 Hz) in the ¹³C NMR

spectrum and 6.74 ppm (t, ³J_PH= 7.9 Hz, 2H) in the ¹H NMR

spectrum, while a singlet attributable to the acetyl methyl group

appears at δ 1.73 ppm in the ¹H NMR spectrum. Formation of

complex 10-cis is reversible at room temperature; when

complex 10-cis-d₃was formed from the reaction of 1 with

methyl acetate-d₃(CH₃CO₂CD₃), it underwent exchange upon

treatment with an excess of perprotio methyl acetate to yield

perprotio 10-cis.

structure of 5 was determined by NMR spectroscopy, most

1

notably the presence of a signal in the H NMR spectrum at δ

−9.24 ppm (²J_PH= 18.1 Hz) indicative of a hydride

coordinated trans to an aryl group.⁹⁴The structure was

conﬁrmed by X-ray crystallography (Figure 2). The formation

of complex 5 proceeds in analogy with the previously reported

reaction of (PCP)Ir with acetophenone including, in particular,

the selectivity for formation of the trans-C−H addition product,

5132

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

Figure 2. X-ray structures of (PCP)Ir(H)[κ²-(C₆H₄C(O)OMe)] (5), (PCP−CH₂)Ir(H)(κ²-O₂CPh) (6-Ph), and (PCP)Ir(CH₃)(κ²-O₂CPh) (7-

Ph).

Thermolysis of complex 10-cis at 80 °C, like complex 5,

results in the formation of a product, 6-Me, in which the CH₂

group derived from the carbomethoxy group has inserted into

the PCP aryl-Ir bond (eq 12). The Ir-CH₂-Ar group of 6-Me

aﬀorded triplets at δ 2.17 ppm (³J_PH= 9.5 Hz) in the ¹H NMR

spectrum and δ −17.94 ppm (²J_CP= 5.1 Hz) in the ¹³C NMR

Complete conversion of 6-Me yielded an approximately 1:1

mixture of 7-Me and 11. Complex 7-Me showed a signal in the

¹H NMR spectrum attributable to the iridium bound methyl

group at δ 1.30 ppm (Ir-CH₃), as well as a singlet associated

with the acetyl group at δ 1.82 ppm. Complex 11 was

characterized by NMR spectroscopy and by X-ray crystallo-

graphic analysis of a crystal obtained by slow evaporation from

a solution of 11 in hexanes (Figure 3).

In contrast with methyl benzoate and methyl acetate, methyl

triﬂuoroacetate reacted rapidly with 1 at room temperature to

give the methylene-bridged-PCP product (PCP-CH₂)Ir(H)(κ²-

O₂CCF₃), 6-CF₃. Heating a C₆D₆solution of 6-CF₃at 100 °C

for 30 h gave the complex (PCP)Ir(CH₃)(O₂CCF₃) (7-CF₃).

2.1.4. Mechanism of Methyl Ester C−O Bond Addition. A

DFT study of the pathway for the reaction of methyl acetate

with (PCP)Ir was conducted. Two (PCP)Ir/CH₃CO₂Me

adducts were located (see Table S2 of the SI); in the lowest

energy adduct, with G = −4.9 kcal/mol (energy values are

relative to methyl acetate plus free three-coordinate (PCP)Ir

unless indicated otherwise), the ester coordinates to Ir only

through its carbonyl oxygen (Ir−O2(carbonyl) = 2.27 Å, Ir−

O1(methoxy) = 4.49 Å). This precoordination of the substrate

does not appear to play any particular role in the reaction

pathway described below, other than possibly competing with

(PCP)Ir(TBV)(H) (G = −4.7 kcal/mol) as the resting state;

the structure of the adduct is unrelated to that of the TS for the

subsequent C−H addition reaction.

The calculations (see Figure 4) indicate a low barrier to

C−H addition, unassisted by either oxygen atom, followed by a

facile chelation to give intermediate 10-cis; the free energy of

the TS is calculated to be 8.4 kcal/mol (relative to methyl

acetate plus free three-coordinate (PCP)Ir), or 13.3 kcal/mol

above the O-coordinated species (which is comparable to the

energy of the (PCP)Ir(TBV)(H) precursor). These values are

consistent with the rapid formation of 10-cis observed in the

reaction of methyl acetate with a (PCP)Ir precursor at room

temperature. α-Carboxylate migration from 10-cis in a manner

analogous to the α-aryloxide migration that occurs with

(PCP)Ir(H)(CH₂OAr), whereby the α-oxygen migrates to

give cis-(PCP)Ir(H)(CH₂)(O₂CMe), is calculated to have a

very high barrier, 33.2 kcal/mol (G_TS= 26.0 kcal/mol relative

to free (PCP)Ir plus MeCO₂Me). A much more facile

migration of carboxylate from 10-cis to aﬀord cis-(PCP)Ir(H)-

(CH₂)(O₂CMe) proceeds via an electrocyclization with a ﬁve-

membered cyclic transition state TS-10−12-cis (Scheme 3).

1

spectrum, while the hydride gave rise to a H NMR signal at

−32.90 ppm (J_PH= 12.5 Hz). These diagnostic features are all

clearly very similar to those found in the spectra of 6-Ph, and

the structure of 6-Me was conﬁrmed crystallographically

(Figure 3).

Figure 3. X-ray structures of (PCP−CH₂)Ir(H)(κ²-O₂CMe) (6-Me)

and cyclometalated (PCP)Ir(acetate), [(κ⁴-C₆H₃-2-(CH₂P^tBu₂)-6-

(CH₂P^tBu(CMe₂CH₂))]Ir(O₂CMe) (11).

Heating a p-xylene solution of 6-Me at 125 °C produced 7-

Me, in analogy with the reaction of 7-Ph, but this reaction was

accompanied by formation of a coproduct 11 (eqs 13 and 14).

5133

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

formation of (PCP−CH₂)Ir(H)(O₂CMe), which is observed

prior to the C−O addition product.

Rotation by 180° around the Ir-CH₂OAc bond of 10-cis can

lead to an isomer, 10-trans, in which H and the Ir-coordinated

CH₂OAc carbon are situated mutually trans. Although we have

not been able to locate a proper TS (i.e., a stationary point on

the PES with just one negative eigenvalue of the Hessian

matrix) for this process, constrained dihedral angle optimiza-

tions indicate that the potential energy barrier separating the

cis_H−CH2complex 10-cis from 10-trans is less than 14 kcal/mol,

and isomerization should therefore be facile. Alternatively, loss

of methyl acetate from 10-cis has a calculated barrier of 15.6

kcal/mol (consistent with the facile exchange noted above and

observed with 10-cis-d₃and CH₃CO₂CH₃); rapid readdition of

the C−H bond to (PCP)Ir, followed by collapse of the C−H

addition product to produce 10-trans would thus have an

overall barrier of only 15.6 kcal/mol.

The geometries of the ﬁve-membered Ir−C−O−CO rings

in 10-cis and 10-trans (Scheme 4) are quite similar, with

slightly longer distances in each case for the respective bond

trans to the hydride versus trans to the PCP ipso carbon

(calculated Ir−C bond lengths (Å) are 2.144 and 2.166, while

Ir−O bond lengths are 2.272 and 2.255 in 10-cis and 10-trans,

respectively). 10-trans is higher in energy than 10-cis by 5.6

kcal/mol, in accord with the fact that this isomer is not

observed experimentally. This cis-trans conformer energy

diﬀerence could arise because in 10-trans the coordinating

group of the chelate with the much stronger trans-inﬂuence

(the oxymethylene carbon versus the carbonyl oxygen) is

positioned trans to the strong trans-inﬂuence hydride ligand.

The calculated barrier for α-carboxylate migration from 10-

trans is very high (G_TS= 31.4 kcal/mol), whereas cyclic

carboxylate migration is calculated to occur with a modest

barrier (G_TS= 14.5 kcal/mol), leading to trans-(PCP)Ir(H)-

(CH₂)(O₂CMe) (12-trans, Figure 5). The corresponding

transition state, TS-10−12-trans (Scheme 4), is structurally

very similar to TS-10−12-cis (Scheme 3); for example, it is

very “late” in its geometrical parameters and the CH₂unit is

signiﬁcantly rotated toward the “horizontal” plane (Scheme 4;

<C(PCP)−Ir−C(CH₂)−H1 = 42°; <C(PCP)−Ir−C(CH₂)−

H2 = −144° in TS). Subsequent hydride migration to the

methylidene unit, however, is of course not possible in the case

of the trans isomer.

Figure 4. Calculated Gibbs free energies (kcal/mol) for reaction of

(PCP)Ir and CH₃−OAc via a hypothetical “cis-pathway”. Values of G

are given relative to free (PCP)Ir and CH₃−OAc. The free energies

correspond to a reference state of 1 M concentration for each species

participating in the reaction and T = 298.15 K. Potential energies,

enthalpies, entropies and free energies for (PCP)Ir + CH₃−OAc

reactions are summarized in Table S2 of the SI.

This migration TS is very “late” in its structural parameters.

The Ir−O2 (carbonyl oxygen) bond is preserved in the cyclic

TS (Ir−O2 = 2.23 vs 2.25 Å in the product), while the C−O1

bond is eﬀectively broken (2.92 Å in the TS vs 3.04 Å in the

product). The Ir−C bond is very close to double bond length

(1.925 vs 1.921 Å in the product), and the CH₂group has

rotated ca. 57° toward its preferred “horizontal” orientation, i.e.,

in the plane of Ir and the nonphosphorus coordinating groups

(<H−Ir−C(CH₂)−H1 = 32°; <H−Ir−C(CH₂)−H2 = −146°).

The reaction coordinate at the TS shows substantial torsional

motion of the CH₂and O2C(O1)Me units in opposite

directions. TS-10−12-cis is relatively low in free energy at G_TS

= 11.1 kcal/mol or only 3.1 kcal/mol above the methylidene

complex to which it leads (12-cis); perhaps this is not

surprising as this is eﬀectively a “6-electron” retro-electro-

cyclization reaction. The resulting methylidene intermediate,

12-cis, is at G = 8.0 kcal/mol, or 15.2 kcal/mol above the free

energy of 10-cis, and thus this reaction is quite reversible

(Figure 4).

While hydride-to-methylidene migration is not geometrically

possible for methylidene complex 12-trans, the barrier to

migration of the PCP aryl group to the methylidene (i.e.,

methylidene insertion into the Ir−C(ipso) bond) is quite low,

7.5 kcal/mol (G_TS= 17.8 kcal/mol; Figure 5). This step yields

the experimentally observed species 6-Me and is very exergonic

(ΔG = −30.0 kcal/mol); therefore, this step is eﬀectively

irreversible as the back reaction has a calculated barrier G^‡=

37.5 kcal/mol. The rate determining step for the reaction of 10-

Subsequent to the ﬁve-membered cyclic carboxylate

migration, migration of the hydride to methylidene would

produce the observed C−O oxidative addition product; this

hydride migration has G_TS= 21.7 kcal/mol (26.6 kcal/mol

above the (PCP)Ir/ester adduct). However, while this “cis-

pathway”, as shown in Figure 4, oﬀers a fairly low barrier to the

ﬁnal C−O addition product, it does not account for the initial

Scheme 3

5134

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

Scheme 4

Figure 5. Calculated Gibbs free energies (kcal/mol) for reaction of (PCP)Ir and CH₃−OAc via the “trans-pathway”, which is proposed to be the

actual reaction path (as opposed to the hypothetical “cis-pathway” shown in Figure 4). Values of G are given relative to free (PCP)Ir and CH₃−OAc.

The free energies correspond to a reference state of 1 M concentration for each species participating in the reaction and T = 298.15 K. Potential

energies, enthalpies, entropies and free energies for (PCP)Ir + CH₃−OAc reactions are summarized in Table S2 of the SI.

Scheme 5

cis to give 6-Me (eq 12) thus appears to be this aryl-to-

methylidene migration (occurring via a reversible isomerization

to 10-trans); the overall barrier (from 10-cis for which G =

−7.2 kcal/mol) via this “trans-pathway” is calculated to be

approximately 25 kcal/mol. This is in excellent agreement with

experiment, which indicates an activation barrier of ca. 27 kcal/

mol (a half-life of 1 h at 80 °C corresponds to ΔG^‡= 26.8 kcal/

mol).

(PCP−CH₃)Ir(O₂CMe), the product of C−H reductive

elimination from (PCP−CH₂)Ir(H)(O₂CMe) (6-Me) is

calculated to be thermodynamically quite accessible with G =

−3.1 kcal/mol (Figure 5). We have searched extensively for a

TS of reasonably low energy for this C−H reductive

elimination from (6-Me). If the coordination sphere around

Ir is otherwise kept intact, then the computed barrier for direct

energy pathway (Scheme 5 and Table S2 of the SI), however, in

which the hydride undergoes trans-migration prior to C−H

elimination. This proceeds via O−H elimination to give a

coordinated acetic acid ligand which can rotate around the Ir−

O(carbonyl) bond and then undergo O−H addition (Scheme

5). The TS for the initial protonation of acetate (O−H

elimination) has a free energy G_TS= 7.9 kcal/mol (ΔG^‡= 27.6

kcal/mol). A series of constrained dihedral angle optimizations

(angle Ir−O−C−OH was varied) indicate that rotation of the

acetic acid about the Ir−O bond may be accomplished via a TS

with G_TS≈ 11 kcal/mol, while for the subsequent O−H

addition, G_TS= −5.6 kcal/mol. (As the rotation involves a

presumably very acidic O−H group, the gas-phase calculated

energy for this process should perhaps be viewed as an upper

limit.) C(PCP−CH₂)−H elimination from the resulting isomer

is facile (G_TS= 4.7 kcal/mol; Table S2 of the SI), giving (PCP−

CH₃)Ir(O₂CMe). Oxidative addition of the PCP−CH₂C−C

bond addition in (PCP−CH₃)Ir(O₂CMe) yields the ﬁnal

product, 7-Me; the TS for this reaction is at 8.6 kcal/mol

relative to the free reactants. Thus, the highest energy TS for

the overall process is calculated to be no more than ca. 11 kcal/

C−H elimination is calculated to be exceedingly high, G_TS

=

32.1 kcal/mol; since G_6Me= −19.7 kcal/mol, this implies a

barrier ΔG^‡≈ 52 kcal/mol. A pathway in which a phosphine

group is detached from the Ir center oﬀers a TS that is of

considerably lower energy, G_TS≈ 17 kcal/mol (ΔG^‡≈ 37 kcal/

mol), but this is still quite high. The calculations reveal a lower

5135

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

Scheme 6. Formation of a Carbene Complex from the Reaction of (PCP)Ir with Methoxymethyl Acetate and Probable Reaction

Pathway

mol (if acetic acid rotation is rate-limiting) but no less than 8.6

kcal/mol (if C−C addition is rate-limiting). The overall barrier

for transformation of 6-Me (G = −19.7 kcal/mol) to the

observed C−O addition product 7-Me is therefore calculated to

be in the range from 28 to 31 kcal/mol. This is in very good

agreement with the experimental observation that the reaction

proceeds to completion in several hours at 125 °C (a half-life of

1 h at 125 °C corresponds to ΔG^‡= 30.3 kcal/mol).

2.1.5. Observation of a Model Carbene Complex. In an

eﬀort to observe the key carbene complex intermediate, we

studied the reaction of (PCP)Ir with methoxymethyl acetate,

MeOCH₂OAc, anticipating that the presence of a methoxy

group could potentially stabilize the methylidene intermediate.

When a half equivalent of MeOCH₂OAc was treated with

Figure 6. X-ray structures of (PCP)Ir(Me)(OTs) (16) and (PCP)-

Ir(H)(OTs) (19).

(PCP)Ir(TBV)(H) (1) at room temperature, the formation of

to the Ir−CH₃group in 16 appear at δ 2.19 ppm (t, J_PH= 5.0

(PCP)IrC(H)(OMe) (13) and (PCP)Ir(H)(OAc) (14-Me)

1

Hz) in the H NMR spectrum and at δ −25.4 ppm in the ¹³C

was observed in approximately a 1:1 ratio (Scheme 6).

NMR spectrum, positions similar to the resonances observed in

Although we were not able to separate complex 13 from the

complexes 4 and 7.

A facile pathway for reaction 14 is calculated, analogous to

the α-aryloxide migration mechanism for CH₃−OAr addition.

product mixture, complete NMR characterization of 13 was

possible, since analytically pure 14-Me could be prepared from

the reaction of 1 with ethyl acetate (vide infra). As a diagnostic

The computed free energy proﬁle is presented in Figure 7. A

feature, the IrC−H resonance appears at δ 14.89 ppm as a

direct C−O addition mechanism is calculated to have a much

1

triplet in the H NMR spectrum and δ 254.8 ppm in the ¹³C

higher barrier with G_TS= 37.5 kcal/mol. The free-energy proﬁle

NMR spectrum, chemical shifts very similar to the resonances

shown in Figure 7 (more detailed energetics are presented in

reported for other iridium Fischer-type carbene com-

Table S3 of the SI) is quite similar to that calculated for

plexes.^69,97,98The formation of both products 13 and 14-Me

oxidative addition of the C−O bond of p-CH₃−OC₆F₄Me

is presumably attributable to the formation of a half equivalent

(Figure 1). Reaction 14 aﬀording (PCP)Ir(CH₃)(OTs) is

slightly more exergonic (G = −28.6 kcal/mol) than the

of hydrido acetate carbene complex (15 or its cis isomer). This

methoxy-stabilized analogue of 12 does not undergo phenyl-to-

analogous reaction with p-MeOC₆F₄Me (G = −26.4 kcal/mol).

carbene (or hydride-to-carbene) migration but instead

The barriers for α-OTs migration and 1,2-H migration are

computed as essentially equal (G_TS= 15.7 and 15.8 kcal/mol,

respectively). The overall barrier is calculated as 21.5 or 21.6

kcal/mol, respectively, consistent with the experimentally

observed rapid rate of reaction at ambient temperature.

In analogy to the experiments with methyl aryl ethers, the H/

D kinetic isotope eﬀect for oxidative addition of methyl tosylate

to (PCP)Ir was measured. When a mixture of CH₃OTS and

eliminates acetic acid, which rapidly reacts with remaining 1

to aﬀord 14-Me.

2.1.6. Reaction of (PCP)Ir with Methyl Tosylate and

Mechanism. Methyl tosylate (CH₃OTs) also reacts with

(PCP)Ir to undergo oxidative addition of the sp³C−O bond.

When 1.1 equiv. of CH₃OTs was added to a p-xylene solution

of 1 at room temperature, an immediate color change from

dark to bright red was observed. The reaction produces a

CD₃OTs (at least 10-fold excess of each) was treated with

quantitative yield of (PCP)Ir(CH₃)(OTs) (16), whose

(PCP)Ir(TBV)(H) in p-xylene solution, two singlets, attribut-

structure was conﬁrmed by NMR spectroscopy as well as X-

able to (PCP)Ir(CH₃)(OTs) (16) and (PCP)Ir(CD₃)(OTs)

ray crystallography (eq 15, Figure 6). Resonances attributable

(16-d₃) were observed in the ³¹P NMR spectrum. From an

average of three such experiments the isotope eﬀect was

determined to be KIE_obs= k_CH3OTs/k_CD3OTs= 2.4(2). As with

the methyl aryl ether, the observed isotope eﬀect is quite

signiﬁcant and clearly indicates that C−H bond activation is

involved in or prior to the rate-determining step of oxidative

C−O bond addition. The DFT-calculated overall KIE resulting

from trideuteration (OCH₃→ OCD₃) is 2.9 if α-OTs

5136

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

Figure 7. Calculated Gibbs free energies (kcal/mol) for reaction of (PCP)Ir and CH₃−OTs. Values of G are given relative to free (PCP)Ir and

CH₃−OTs. The free energies correspond to a reference state of 1 M concentration for each species participating in the reactions and T = 298.15 K.

Potential energies, enthalpies, entropies, and free energies for (PCP)Ir + CH₃−OTs reactions are summarized in Table S3 of the SI.

migration is the rate-determining step, as opposed to 6.7 if

hydride-migration to the carbene intermediate is the rate-

determining step. The experimentally determined value of

C₆F₄Me) and 18 in a 1:1 ratio. Upon thermolysis of this

reaction mixture at 80 °C for two days, 18 is fully converted to

17-C₆F₄Me (eq 16).⁶⁴

k_CH3/k_CD3= 2.4(2) is clearly in good agreement with the

(PCP)Ir(TBV)(H) (1) also reacts with isopropyl phenyl

ether to produce (PCP)Ir(H)(OPh) (17-Ph) and propene (eq

17); 17-Ph was identiﬁed by ³¹P NMR spectroscopy, and

former case and, accordingly, we propose that α-OTs migration

is the rate-determining step for reaction 14.

1

2.2. Reactions of (PCP)Ir with Higher Alkyl Oxygen-

ates R−OAr, R−OTs, and R−OAc (R = ethyl or isopropyl).

2.2.1. 1,2-H-OR Elimination Reactions (R = Ar, Ac, and Ts).

Reactions of aryl ethers, esters, and tosylate bearing alkyl

groups higher than methyl have also been investigated.

Treatment of 1 equiv. of ethyl phenyl ether with (PCP)Ir-

(TBV)(H) (1) at room temperature initially produced a

mixture of intermediates resulting from C−H bond activation

of the ether phenyl group. After approximately ﬁve hours at

room temperature, this mixture was converted to an equimolar

mixture of (PCP)Ir(H)(OPh) (17-Ph) and (PCP)Ir(C₂H₄)

(18) (eq 16). Upon heating this mixture at 125 °C for two

propene was identiﬁed by H NMR spectroscopy and GC

analysis.

Somewhat surprisingly, however, the reaction of (PCP)Ir

with n-propyl phenyl ether does not proceed at room

temperature, even after three days. At 60 °C, 17-Ph is formed

but in only ∼30% yield along with a complex mixture of

byproducts. Likewise, the reaction with n-butyl phenyl ether

requires heating to 60 °C and does not proceed cleanly, giving

17-Ph in a yield of ca. 50%. In both cases, (PCP)IrH₂was

observed as a side product in the course of the reaction,

probably generated by dehydrogenation of the alkyl chain of

the alkyl phenyl ethers.

Addition of ethyl acetate to a p-xylene solution of

(PCP)Ir(TBV)(H) (1) at room temperature resulted in an

1

immediate color change from dark red to brown. ³¹P and H

NMR spectroscopy revealed formation of an equimolar mixture

of (PCP)Ir(H)(OAc) (14-Me) and (PCP)Ir(ethylene) (18).

Upon further heating of the reaction mixture at 80 °C, 18 was

converted to 14-Me and ethylene (eq 18). The structure of 14-

Me was conﬁrmed by X-ray crystallography (Figure S-8 of the

SI) and NMR spectroscopy. The phosphorus resonance in the

PCP ligand appears at δ 60.4 ppm as a doublet in the ³¹P NMR

spectrum, and Ir−H manifests at δ −29.8 ppm as a triplet in the

¹H NMR spectrum.

days, full conversion to (PCP)Ir(H)(OPh) was observed (eq

16). The structure of 17-Ph, previously reported by our

group,^64,99was determined by NMR spectroscopy and X-ray

crystallography.

In the case of 4-ethoxy-2,3,5,6-tetraﬂuorotoluene, the

analogous reaction, at room temperature, immediately aﬀords

(PCP)Ir(H)(2,3,5,6-tetraﬂuoro-4-methyl phenoxide) (17-

The reaction of 1 with ethyl benzoate proceeded analogously

to give 14-Ph (see SI for spectroscopic and crystallographic

characterization, Figure S-9 of the SI), but went to completion

5137

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

at room temperature. The structure of 19 was determined by

NMR spectroscopy and X-ray crystallography (Figure S-11 of

the SI). Upon further heating of the reaction mixture at 80 °C

for several hours, 18 was converted to 19.

2.2.2. Mechanistic Study of 1,2-H-OR Elimination Reac-

tions (R = Ar, Ac, and Ts). A priori, the above-described

reactions of ethyl and higher alkyl aryl ethers, acetates and

tosylate could proceed via direct C−O bond addition, followed

by β-H elimination of the resulting alkyl group. In light of the

results of the computational and mechanistic studies of the

methyl derivatives, however, this would seem unlikely; this

conclusion is reinforced by the results of experiments described

below.

at room temperature (3 h). Compound 1 rapidly reacted with

isopropylacetate to give complete conversion to 14-Me and

propene (eq 19). Presumably, the lesser stability of the propene

analogue of 18 allows it to quickly react with isopropylacetate.

When a mixture of 4-methoxy-2,3,5,6-tetraﬂuorotoluene and

4-ethoxy-2,3,5,6-tetraﬂuorotoluene was reacted with (PCP)Ir

(eq 21), aryloxide and ethylene complexes 17-C₆F₄Me and 18

The reaction of 1 with n-propylacetate, in contrast, did not

quickly lead to (PCP)Ir(H)(OAc). (PCP)IrH₂and several

unidentiﬁed intermediates were observed in the course of the

reaction, although after ca. 20 h at room temperature 14-Me

was the only observed product. Other alkyl acetates underwent

dehydrocarboxylation as indicated in Scheme 7, although yields

were lower than those obtained with ethyl- or isopropylacetate.

Analogously to the reactions with ethyl aryl ether or ethyl

acetate (eqs 16 and 18) the reaction of 1 with ethyl tosylate

gave an equimolar mixture of (PCP)Ir(H)(OTs) (19) and 18

(eq 20). The reaction went cleanly to completion immediately

were formed exclusively (the same products that are obtained

from pure 4-ethoxy-2,3,5,6-tetraﬂuorotoluene). There was no

evidence of any reaction with 4-methoxy-2,3,5,6-tetraﬂuoroto-

luene.¹⁰⁰

Closely related to the experiment of eq 21 with ethers,

addition of an excess of a methyl acetate/ethyl acetate solution

to a p-xylene solution of complex 10-cis at room temperature

resulted exclusively in the immediate formation of 14-Me and

18 (eq 22), i.e., the same products that result from the reaction

of pure ethyl acetate. This implies, ﬁrst, that the methyl acetate

C−H addition is reversible, in accord with the calculated energy

barriers (Figures 4 and 5) as well as the isotope exchange

experiment between 10-cis-d₃and perprotio methyl acetate.

The fact that ethyl acetate and ethyl aryl ether are so much

more reactive than their methyl analogues argues strongly

Scheme 7. Reaction of (PCP)Ir with Alkyl Acetates

5138

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

Scheme 8. Proposed Mechanism for 1,2-H-OR Elimination of Ethyl Oxygenates (R = Aryl, Tosyl, and Acyl)

against pathways involving direct C−O oxidative addition; it

would seem improbable that direct addition of an O−CH₃

bond would be much less facile than O−C₂H₅addition.

Moreover, a direct C−O addition would certainly not explain

why n-propyl acetate and ether, but not the i-propyl analogues,

are much less reactive than their ethyl analogues.

The high level of reactivity of the ethyl and i-propyl species

can be explained with a pathway which, like the reactions of the

methyl species, is initiated by C−H addition at a methyl group,

but is then followed by β-oxygenate elimination. The analogous

pathway for n-propylacetate, n-propyl phenyl ether, and n-butyl

phenyl ether would require addition of a secondary C−H bond,

which is less favorable for transition metal complexes in

general,³⁸and for (PCP)Ir in particular.⁵⁵These results are

thus all consistent with the pathway shown in Scheme 8 for the

1,2-H−OR eliminations.^101,102A similar mechanism has been

proposed by Chirik et al. in the reaction between a bis(indenyl)

zirconium complex and diethyl ether, resulting in an equimolar

mixture of Zr(H)(OEt) and Zr(ethylene) complexes.¹⁰¹

The apparent generally greater reactivity of the ethyl vs

methyl oxygenates may be explained by generally greater ease

of β- versus α-oxygen elimination. Direct comparison between

the rates of α- and β-oxygen elimination are rare. However,

Brookhart and co-workers have reported¹⁰³that a Ni complex

bearing a β-acetate group decomposes at −34 °C via acetate

elimination, while a similar Ni complex with an α-acetate

moiety is stable up to much higher temperature (>70 °C).

They also reported that a Pd complex with an α-acetate

substituted methyl group decomposed at a higher temperature

than the analogous compound with an α-triﬂuoroacetate (60

°C vs 10 °C).¹⁰³This ﬁnding is consistent with our result that

(PCP)Ir eﬀects C−O oxidative addition of methyl triﬂuoro-

acetate at room temperature, whereas the acetoxymethyl

iridium complex (10) formed from the reaction with methyl

acetate requires much higher temperature (70 °C) to undergo

further reaction.

Figure 8. Calculated Gibbs free energies (kcal/mol) for reactions of

(PCP)Ir and C₂H₅−OAr, Ar = phenyl (bold) and p-C₆F₄Me (italics).

Values of G are given relative to free (PCP)Ir and C₂H₅−OAr, Ar =

phenyl (bold) and p-C₆F₄Me (italics). The free energies correspond to

a reference state of 1 M concentration for each species participating in

the reactions and T = 298.15 K. Potential energies, enthalpies,

entropies and free energies for (PCP)Ir + C₂H₅−OAr, Ar = phenyl

and p-C₆F₄Me, reactions are summarized in Table S4 of the SI.

for the phenoxide and 2,3,5,6-tetraﬂuoro-4-methylphenoxide,

respectively. The binding enthalpies of the resulting coordi-

nated ethylene are modest (9.0 and 10.3 kcal/mol for

ethoxybenzene and 4-ethoxy-2,3,5,6-tetraﬂuorotoluene, respec-

tively) and loss of ethylene is a slightly exergonic process

(ΔG_C2H4‑loss= −5.9 and −3.6 kcal/mol for ethoxybenzene and

4-ethoxy-2,3,5,6-tetraﬂuorotoluene, respectively). A conven-

tional TS for ethylene loss could not be located on the

potential energy surface; however, the barriers may be very

approximately estimated by equating barrier heights with

binding enthalpies. This procedure leads to approximate

G_{TS,C2H4‑loss}free energies of 1.6 and −6.9 kcal/mol for

ethoxybenzene and 4-ethoxy-2,3,5,6-tetraﬂuorotoluene, respec-

tively, values which are signiﬁcantly lower than the preceding

TSs.

In accord with the above discussion, the barrier to direct

C(sp³)−O addition calculated for ethoxybenzene is prohib-

itively high, ΔG^‡

= 40.8 kcal/mol (ca. 46 kcal/mol relative

to 1). For 4-ethoxy-2,3,5,6-tetraﬂuorotoluene the calculated

C−O

barrier is somewhat lower, but still very high, ΔG^‡

= 35.0

C−O

kcal/mol (ca. 40 kcal/mol relative to 1). In contrast, the

activation energy barriers are much lower for a reaction

pathway (Figure 8) initiated by addition of a terminal sp³C−H

bond from the ethoxy unit, followed by β-aryloxy elimination

(more detailed energetics are presented in Table S4 of the SI).

The computed C−H activation barrier heights for the

ﬂuorinated and parent substrates are virtually identical (ΔG^‡

= 9.7 and 9.8 kcal/mol for ethoxybenzene and 4-ethoxy-2,3,5,6-

tetraﬂuorotoluene, respectively). The energies of the β-aryloxy

migration transition states, however, are strongly reduced by

the ﬂuorine substituents on the aryl ring: ΔG_TS= 14.0 kcal/mol

for ethoxybenzene vs 7.6 kcal/mol for 4-ethoxy-2,3,5,6-

tetraﬂuorotoluene. This large diﬀerence in TS energies is

apparently driven by an even larger diﬀerence in energies of the

β-elimination products, G = −7.5 kcal/mol and −17.2 kcal/mol

The computed reaction energy proﬁle for the reaction of

ethyl acetate with (PCP)Ir is shown in Figure 9 (more detailed

energetics are presented in Table S5 of the SI). The lowest TS

for C−H addition has G_TS= 11.0 kcal/mol relative to free

reactants. Subsequent to C−H addition the carbonyl oxygen

undergoes chelation to give the κ²-acetoxyethyl hydride

complex 20.

Intermediate 20 (G = 4.1 kcal/mol) undergoes β-acetate

migration to give ethylene complex 21 with a TS that is only

12.0 kcal/mol above free reactants. The nature of TS-20−21,

whose geometry is indicated in Scheme 9, is quite unusual. The

C2−O1 bond is largely broken in TS-20−21, but unlike in a

typical β-elimination TS, there is clearly no bonding developing

between the iridium center and either C2 or O1. Indeed,

although TS-20−21 leads to an ethylene complex with typically

5139

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

Figure 9. Calculated Gibbs free energies (kcal/mol) for reaction of (PCP)Ir and C₂H₅OAc. Values of G are given relative to free (PCP)Ir and

C₂H₅OAc. The free energies correspond to a reference state of 1 M concentration for each species participating in the reactions and T = 298.15 K.

Potential energies, enthalpies, entropies, and free energies for reaction of (PCP)Ir and C₂H₅OAc are summarized in Table S5 of the SI.

Scheme 9. Geometry of TS-20−21

short Ir−C bond distances (2.26 Å for Ir−C2), the Ir−C2

distance in the TS is actually slightly longer, at 3.19 Å, than the

distance of 3.13 Å found in the preceding intermediate, 20.

Following the intrinsic reaction coordinate leads toward the

formation of an ethylene complex, although it is one in which

the ethylene ligand is oriented in the “horizontal” plane (the

plane containing the Ir atom and the three other non-

phosphorus coordinating atoms); the ethylene unit in 21 is

actually oriented along the P−Ir−P axis (perpendicular to the

“horizontal” plane), presumably to reduce steric congestion in

the “horizontal” plane. The free energy diﬀerence between the

two conformers with diﬀerent ethylene orientations is, however,

only 1.2 kcal/mol.

Subsequent ethylene elimination from 21 is an exergonic,

entropy-driven process (ΔG_elim‑C2H4= −4.5 kcal/mol), and the

ethylene binding enthalpy is fairly low (∼8.8 kcal/mol).

Despite a thorough geometrical search, we were not able to

locate a conventional TS for ethylene loss from the (PCP)Ir

unit on the potential energy surface. Given the weak binding

however, the TS for ethylene loss is presumably much lower in

energy than the preceding TS for acetate migration (TS-20−

21, Figure 9). Thus acetate migration is presumably irreversible

and is probably the rate-determining step. The overall barrier

computed for reaction 17 in that case is only ca. 17 kcal/mol,

consistent with the very rapid rate observed at room

temperature.

3. CONCLUSIONS

Examples of C(sp³)−O bond cleavage promoted by transition-

metal complexes are limited despite expanding interest in this

area. We ﬁnd that (PCP)Ir oxidatively adds the C(sp³)−O

bond of various methyl oxygenates (esters, tosylates, and

electron-poor aryl ethers). This reaction proceeds via C−H

addition, followed by α-OR migration (OR = OAr, OTs, OAc,

or OC(O)Ar). In the case of OR = carboxylate, the α-migration

proceeds via a cyclic transition state in which the alkoxy group

C−O bond is broken, while an Ir−C π-bond is formed and,

formally, a dative Ir−O (carbonyl O) bond is converted to a

covalent Ir−O bond. Derivatives with higher alkyl groups

possessing β-C−H bonds react with (PCP)Ir to undergo a 1,2-

dehydro-oxygenation to give (PCP)Ir(H)(OR) and the

corresponding oleﬁn. This reaction also is initiated by the

addition of what is generally considered an “unreactive” C−H

bond, in this case the C−H bond positioned β to the C−O

bond undergoing cleavage; this is followed by β-oxygenate

elimination. For the ethers and tosylates, the β-oxygenate

elimination is apparently a straightforward process. In contrast,

the TS for the acetates is a six-membered cyclic species in

which the C−O bond is cleaved without any direct

participation from the iridium center. On the basis of

considerations of microscopic reversibility, the surprising

mechanistic implications of this study presumably also apply

to potential C−O bond-making reactions. Eﬀorts to employ the

principles elucidated in this work toward the development of

catalytic processes are currently underway.

5140

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

4. COMPUTATIONAL METHODS

Article

(14) Kakiuchi, F.; Usui, M.; Ueno, S.; Chatani, N.; Murai, S. J. Am.

Chem. Soc. 2004, 126, 2706−2707.

(15) Ueno, S.; Mizushima, E.; Chatani, N.; Kakiuchi, F. J. Am. Chem.

Soc. 2006, 128, 16516−16517.

(16) Guan, B.-T.; Xiang, S.-K.; Wang, B.-Q.; Sun, Z.-P.; Wang, Y.;

Zhao, K.-Q.; Shi, Z.-J. J. Am. Chem. Soc. 2008, 130, 3268−3269.

(17) Guan, B.-T.; Xiang, S.-K.; Wu, T.; Sun, Z.-P.; Wang, B.-Q.;

Zhao, K.-Q.; Shi, Z.-J. Chem. Commun. 2008, 1437−1439.

(18) Tobisu, M.; Shimasaki, T.; Chatani, N. Chem. Lett. 2009, 38,

710−711.

DFT calculations employed the recently developed M06 func-

tionals;¹⁰⁴an eﬀective core potential and valence basis set of triple-ζ

quality on the Ir atom (LANL08),¹⁰⁵and basis sets of at least split

valence plus polarization quality on all main group atoms.¹⁰⁶The

calculations were performed on the actual molecular species used in

the experiments, i.e., truncations were not made to the substrates or to

the (PCP)Ir fragment, which retained the bulky t-Bu groups on P. We

did not consider solvent eﬀects explicitly in the calculations, since all

experiments were conducted in simple nonpolar hydrocarbons (e.g., p-

xylene), and speciﬁc solvent dependence was not encountered in any

of the experiments. Indeed, a few selected calculations in which the

general eﬀects of a surrounding solvent were included via a continuum

dielectric model revealed only very minor changes in reaction barrier

heights (<1 kcal/mol). A reference state of T = 298 K and 1 M

concentration of all reacting species was used. A complete description

of the computational methodology is available in the SI.

(19) Shimasaki, T.; Konno, Y.; Tobisu, M.; Chatani, N. Org. Lett.

2009, 11, 4890−4892.

(20) Yu, D.-G.; Li, B.-J.; Shi, Z.-J. Acc. Chem. Res. 2010, 43, 1486−

1495.

(21) Alvarez-Bercedo, P.; Martin, R. J. Am. Chem. Soc. 2010, 132,

17352−17353.

(22) Sergeev, A. G.; Hartwig, J. F. Science 2011, 332, 439−443.

(23) Tobisu, M.; Yamakawa, K.; Shimasaki, T.; Chatani, N. Chem.

Commun. 2011, 47, 2946−2948.

ASSOCIATED CONTENT

* Supporting Information

■

(24) Guan, B.-T.; Wang, Y.; Li, B.-J.; Yu, D.-G.; Shi, Z.-J. J. Am. Chem.

Soc. 2008, 130, 14468−14470.

S

Experimental details and procedures; spectral data for all

complexes characterized; crystallographic structures and details

for complexes 3h, 3j, 5, 6-Ph, 6-Me, 7-Ph, 11, 14-Me, 14-Ph,

16, and 19; computational details, molecular energies, and

geometries. This material is available free of charge via the

Internet at http://pubs.acs.org.

(25) Li, B.-J.; Li, Y.-Z.; Lu, X.-Y.; Liu, J.; Guan, B.-T.; Shi, Z.-J. Angew.

Chem., Int. Ed. 2008, 47, 10124−10127.

(26) Quasdorf, K. W.; Tian, X.; Garg, N. K. J. Am. Chem. Soc. 2008,

130, 14422−14423.

(27) Li, Z.; Zhang, S.-L.; Fu, Y.; Guo, Q.-X.; Liu, L. J. Am. Chem. Soc.

2009, 131, 8815−8823.

(28) Li, B.-J.; Li, Y.-Z.; Lu, X.-Y.; Liu, J.; Guan, B.-T.; Shi, Z.-J. Angew.

Chem., Int. Ed. 2009, 48, 1351.

AUTHOR INFORMATION

Corresponding Author

alan.goldman@rutgers.edu; kroghjes@rutgers.edu

■

(29) Li, B.-J.; Xu, L.; Wu, Z.-H.; Guan, B.-T.; Sun, C.-L.; Wang, B.-

Q.; Shi, Z.-J. J. Am. Chem. Soc. 2009, 131, 14656−14657.

(30) Shimasaki, T.; Tobisu, M.; Chatani, N. Angew. Chem., Int. Ed.

2010, 49, 2929−2932.

Notes

The authors declare no competing ﬁnancial interest.

(31) Huang, K.; Yu, D.-G.; Zheng, S.-F.; Wu, Z.-H.; Shi, Z.-J.

Chem.Eur. J. 2011, 17, 786−791.

ACKNOWLEDGMENTS

■

(32) Manbeck, K. A.; Kundu, S.; Walsh, A. P.; Brennessel, W. W.;

Jones, W. D. Organometallics 2012, 31, 5018−5024.

(33) Lu, Z.; Ma, S. Angew. Chem., Intl. Ed. 2008, 47, 258−297.

(34) Trost, B. M.; Van, V. D. L. Chem. Rev. 1996, 96, 395−422.

(35) Trost, B. M.; Crawley, M. L. Chem. Rev. 2003, 103, 2921−2943.

(36) Fang, P.; Raj Chaulagain, M.; Aron, Z. D. Org. Lett. 2012, 14,

2130−2133.

We thank David Laviska for helpful discussion regarding the

aryl C−H addition reactions. We thank the National Science

Foundation for support of this work through Grant CHE No.

1112456 and the Extreme Science and Engineering Discovery

Environment (XSEDE), which is supported by National

Science Foundation grant number OCI-1053575.

(37) Barrios-Landeros, F.; Hartwig, J. F. J. Am. Chem. Soc. 2005, 127,

6944−6945.

REFERENCES

■

(38) Some reviews of alkane C-H bond activation by organometallic

complexes: (a) Jones, W. D. Science 2000, 287, 1942−1943.

(b) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507−514.

(c) Goldberg, K. I.; Goldman, A. S., Eds. Activation and

Functionalization of C−H Bonds; ACS Symposium Series 885;

American Chemical Society: Washington, DC, 2004. http://rutchem.

rutgers.edu/∼agoldman/TOC_ACS_Symposium_Series_885.htm

(d) Crabtree, R. H. Chem. Rev. 2010, 110, 575. (e) Balcells, D.; Clot,

E.; Eisenstein, O. Chem. Rev. 2010, 110, 749−823.

(39) Hartwig, J. F. In Organotransition Metal Chemistry; University

Science Books: Sausalito, CA, 2010, pp 261−320.

(40) Crabtree, R. H. In The Organometallic Chemistry of the Transition

Metals, 4th ed.; John Wiley & Sons: New York, 2005, pp 159−182 and

364−370.

(1) Lichtenthaler, F. W.; Mondel, S. Pure Appl. Chem. 1997, 69,

1853−1866.

(2) Schlaf, M. Dalton Trans. 2006, 4645−4653.

(3) Arceo, E.; Marsden, P.; Bergman, R. G.; Ellman, J. A. Chem.

Commun. 2009, 3357−3359.

(4) Schlaf, M.; Ghosh, P.; Fagan, P. J.; Hauptman, E.; Bullock, R. M.

Adv. Synth. Catal. 2009, 351, 789−800.

(5) Zakzeski, J.; Bruijnincx, P. C. A.; Jongerius, A. L.; Weckhuysen, B.

M. Chem. Rev. 2010, 110, 3552−3599.

(6) Xu, X.; Li, Y.; Gong, Y.; Zhang, P.; Li, H.; Wang, Y. J. Am. Chem.

Soc. 2012, 134, 16987−16990.

(7) Nichols, J. M.; Bishop, L. M.; Bergman, R. G.; Ellman, J. A. J. Am.

Chem. Soc. 2010, 132, 12554−12555.

(8) Sergeev, A. G.; Webb, J. D.; Hartwig, J. F. J. Am. Chem. Soc. 2012,

134, 20226−20229.

(41) Ittel, S. D.; Tolman, C. A.; English, A. D.; Jesson, J. P. J. Am.

Chem. Soc. 1978, 100, 7577−7585.

(9) He, J.; Zhao, C.; Lercher, J. A. J. Am. Chem. Soc. 2012, 134,

20768−20775.

(42) Tolman, C. A.; Ittel, S. D.; English, A. D.; Jesson, J. P. J. Am.

Chem. Soc. 1979, 101, 1742−1751.

(10) Hanson, S. K.; Wu, R.; Silks, L. A. P. Angew. Chem., Intl. Ed.

2012, 51, 3410−3413.

(43) Zhu, Y.; Smith, D. A.; Herbert, D. E.; Gatard, S.; Ozerov, O. V.

Chem. Commun. 2012, 48, 218−220.

(11) Wenkert, E.; Michelotti, E. L.; Swindell, C. S. J. Am. Chem. Soc.

1979, 101, 2246−2247.

(44) Bonanno, J. B.; Henry, T. P.; Neithamer, D. R.; Wolczanski, P.

T.; Lobkovsky, E. B. J. Am. Chem. Soc. 1996, 118, 5132−5133.

(45) van der Boom, M. E.; Liou, S.-Y.; Ben-David, Y.; Vigalok, A.;

Milstein, D. Angew. Chem., Int. Ed. Engl. 1997, 36, 625−626.

(12) Wenkert, E.; Michelotti, E. L.; Swindell, C. S.; Tingoli, M. J. Org.

Chem. 1984, 49, 4894−4899.

(13) Dankwardt, J. W. Angew. Chem., Int. Ed. 2004, 43, 2428−2432.

5141

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

(46) van der Boom, M. E.; Liou, S.-Y.; Ben-David, Y.; Shimon, L. J.

W.; Milstein, D. J. Am. Chem. Soc. 1998, 120, 6531−6541.

(47) Trovitch, R. J.; Lobkovsky, E.; Bouwkamp, M. W.; Chirik, P. J.

Organometallics 2008, 27, 6264−6278.

(71) Carbo, J. J.; Eisenstein, O.; Higgitt, C. L.; Klahn, A. H.; Maseras,

F.; Oelckers, B.; Perutz, R. N. J. Chem. Soc., Dalton Trans. 2001, 1452−

1461.

(72) Evans, M. E.; Burke, C. L.; Yaibuathes, S.; Clot, E.; Eisenstein,

O.; Jones, W. D. J. Am. Chem. Soc. 2009, 131, 13464−13473.

(73) Tanabe, T.; Brennessel, W. W.; Clot, E.; Eisenstein, O.; Jones,

W. D. Dalton Trans. 2010, 39, 10495−10509.

(74) Clot, E.; Besora, M.; Maseras, F.; Megret, C.; Eisenstein, O.;

Oelckers, B.; Perutz, R. N. Chem. Commun. 2003, 490−491.

(75) Clot, E.; Megret, C.; Eisenstein, O.; Perutz, R. N. J. Am. Chem.

Soc. 2009, 131, 7817−7827.

(76) In this case, the product-determining step is presumably also the

rate-determining step (vide infra). However, if, for example,

elimination of TBE is rate-determining, the competition experiment

would be equally informative regarding the subject of interest, i.e.,

mechanism of the reaction of (PCP)Ir with ether; the observed KIE

would be the same as in the (presumably actual) case of reversible

TBE loss prior to reaction of (PCP)Ir with ether.

(48) Komiya, S.; Suzuki, J.; Miki, K.; Kasai, N. Chem. Lett. 1987,

1287−90.

(49) For an unusual example of reductive elimination of C(sp³)−O

bonds from a palladium center, see: Marquard, S. L.; Hartwig, J. F.

Angew. Chem., Intl. Ed. 2011, 50, 7119−7123.

(50) (a) Lara, P.; Paneque, M.; Poveda, M. L.; Salazar, V.; Santos, L.

L.; Carmona, E. J. Am. Chem. Soc. 2006, 128, 3512−3513.

(b) Conejero, S.; Paneque, M.; Poveda, M. L.; Santos, L. L.;

Carmona, E. Acc. Chem. Res. 2010, 43, 572−580. (b) Santos, L. L.;

Mereiter, K.; Paneque, M. Organometallics 2013, 32, 565−569.

(51) Whited, M. T.; Grubbs, R. H. Acc. Chem. Res. 2009, 42, 1607−

1616.

(52) Whited, M. T.; Zhu, Y.; Timpa, S. D.; Chen, C.-H.; Foxman, B.

M.; Ozerov, O. V.; Grubbs, R. H. Organometallics 2009, 28, 4560−

4570.

(77) Villano, S. M.; Kato, S.; Bierbaum, V. M. J. Am. Chem. Soc. 2006,

128, 736−737.

(53) Gupta, M.; Hagen, C.; Flesher, R. J.; Kaska, W. C.; Jensen, C. M.

Chem. Commun. 1996, 2083−2084.

(54) Xu, W.; Rosini, G. P.; Gupta, M.; Jensen, C. M.; Kaska, W. C.;

Krogh-Jespersen, K.; Goldman, A. S. Chem. Commun. 1997, 2273−

2274.

(78) Yamataka, H.; Ando, T. J. Phys. Chem. 1981, 85, 2281−2286.

(79) On the basis of calculations using PBE functionals, as opposed

to the M06 functionals used in this work, we previously (ref 64)

calculated a larger energy diﬀerence in favor of OAr migration, leading

us to suggest that H-to-methylidene migration was the rate-

determining step. We were, however, unable to easily reconcile the

observed KIE (4.3) with the KIE thus calculated (7.2). In view of this

dependence found on the particular functionals employed, and at the

suggestion of an anonymous reviewer, we have also investigated the

use of a range of other commonly applied functionals (M06-L, M06−

2X, PBE, TPSS, TPSSh, B3LYP) to calculate the energies of the two

key TS’s (the TS’s for α-aryloxy elimination and for Ir-to-methylene

hydride migration) and the predicted KIEs. With all functionals used,

the two TSs are calculated to be fairly close in energy (within 4 kcal/

mol of each other). The use of M06−2X leads to the prediction, in

agreement with our conclusion, that the α-aryloxy elimination TS is

higher in energy (albeit by only 1.1 kcal/mol) than the TS for Ir-to-

methylene hydride migration. Full results of these calculations are

given in Table S6 of the Supporting Information.

(80) We ﬁnd it useful to express the overall KIE in terms of the

relative rates of the reaction of isotopomeric ethers with free (PCP)Ir

although free (PCP)Ir is clearly not the resting state for this system.

This yields, for each isotopomer, k_X= K_8X[(PCP)Ir] • k_9X. The key

condition on which this analysis is based is that (PCP)Ir(CH₂OAr)

(H) and (PCP)Ir(CD₂OAr)(D) are in equilibrium with each other.

This must be the case if the rate-determining step is subsequent to the

formation of these species on the reaction pathway. We express this in

terms of an equilibrium with (PCP)Ir, although it could be expressed

as an equilibrium with any other species (resting state, intermediate, or

even a purely hypothetical species); the free energy of that species will

be canceled out of the ﬁnal equation. In this case, speciﬁcally, the free

energy diﬀerence of (PCP)Ir(CH₂OAr)(H) (plus free CD₃OAr) vs

(PCP)Ir(CD₂OAr)(D) (plus free CH₃OAr) is equal to −0.674 kcal/

mol; which is necessarily equal to ΔΔG₈where EIE₈= exp(−ΔΔG₈/

RT).

(55) Liu, F.; Pak, E. B.; Singh, B.; Jensen, C. M.; Goldman, A. S. J.

Am. Chem. Soc. 1999, 121, 4086−4087.

(56) Kanzelberger, M.; Singh, B.; Czerw, M.; Krogh-Jespersen, K.;

Goldman, A. S. J. Am. Chem. Soc. 2000, 122, 11017−11018.

(57) Ghosh, R.; Emge, T. J.; Krogh-Jespersen, K.; Goldman, A. S. J.

Am. Chem. Soc. 2008, 130, 11317−11327.

(58) Morales-Morales, D.; Lee, D. W.; Wang, Z.; Jensen, C. M.

Organometallics 2001, 20, 1144−1147.

(59) Morales-Morales, D.; Redon, R.; Wang, Z.; Lee, D. W.; Yung,

C.; Magnuson, K.; Jensen, C. M. Can. J. Chem. 2001, 79, 823−829.

(60) Kanzelberger, M.; Zhang, X.; Emge, T. J.; Goldman, A. S.; Zhao,

J.; Incarvito, C.; Hartwig, J. F. J. Am. Chem. Soc. 2003, 125, 13644−

13645.

(61) Choi, J.; Wang, D. Y.; Kundu, S.; Choliy, Y.; Emge, T. J.; Krogh-

Jespersen, K.; Goldman, A. S. Science 2011, 332, 1545−1548.

(62) Some lead references and examples of C−O bond formation via

reductive elimination: (a) Han, R.; Hillhouse, G. L. J. Am. Chem. Soc.

1997, 119, 8135−8136. (b) Widenhoefer, R. A.; Buchwald, S. L. J. Am.

Chem. Soc. 1998, 120, 6504−6511. (c) Dick, A. R.; Kampf, J. W.;

Sanford, M. S. J. Am. Chem. Soc. 2005, 127, 12790−12791.

(d) Hartwig, J. F. Synlett 2006, 1283−1294. (e) Vedernikov, A. N.;

Binfield, S. A.; Zavalij, P. Y.; Khusnutdinova, J. R. J. Am. Chem. Soc.

2006, 128, 82−83. (f) Khusnutdinova, J. R.; Zavalij, P. Y.; Vedernikov,

A. N. Organometallics 2007, 26, 3466−3483. (g) Hartwig, J. F. Nature

2008, 455, 314−322. (h) Racowski, J. M.; Dick, A. R.; Sanford, M. S. J.

Am. Chem. Soc. 2009, 131, 10974−10983. (i) Vedernikov, A. N. Top.

Organomet. Chem. 2010, 31, 101−121.

(63) (a) Williams, B. S.; Holland, A. W.; Goldberg, K. I. J. Am. Chem.

Soc. 1999, 121, 252−253. (b) Williams, B. S.; Goldberg, K. I. J. Am.

Chem. Soc. 2001, 123, 2576−2587.

(64) Choi, J.; Choliy, Y.; Zhang, X.; Emge, T. J.; Krogh-Jespersen, K.;

Goldman, A. S. J. Am. Chem. Soc. 2009, 131, 15627−15629.

(65) Renkema, K. B.; Kissin, Y. V.; Goldman, A. S. J. Am. Chem. Soc.

2003, 125, 7770−7771.

(81) See, for example: (a) Parkin, G. Acc. Chem. Res. 2009, 42, 315−

325. (b) Janak, K. E.; Parkin, G. J. Am. Chem. Soc. 2003, 125, 6889−

6891 and references therein.

(66) Ghosh, R.; Zhang, X.; Achord, P.; Emge, T. J.; Krogh-Jespersen,

K.; Goldman, A. S. J. Am. Chem. Soc. 2007, 129, 853−866.

(67) Slugovc, C.; Mereiter, K.; Trofimenko, S.; Carmona, E. Angew.

Chem., Int. Ed. 2000, 39, 2158−2160.

(82) (a) Rachidi, I. E.-I.; Eisenstein, O.; Jean, Y. New J. Chem. 1990,

14, 671−677. (b) Riehl, J.-F.; Jean, Y.; Eisenstein, O.; Pelissier, M.

́

Organometallics 1992, 11, 729.

(83) Chan, J.; Tang, A.; Bennet, A. J. J. Am. Chem. Soc. 2012, 134,

1212−1220.

(84) Singh, V.; Schramm, V. L. J. Am. Chem. Soc. 2007, 129, 2783−

2795.

(85) Adcock, W.; Trout, N. A.; Vercoe, D.; Taylor, D. K.; Shiner, V.

J.; Sorensen, T. S. J. Org. Chem. 2003, 68, 5399−5402.

(86) Asperger, S.; Kukric, Z.; Saunders, W. H.; Sutic, D. J. Chem. Soc.,

Perkin Trans. 2 1992, 275−279.

(68) Jones, R. A.; Wilkinson, G. J. Chem. Soc., Dalton Trans. 1979,

472−477.

(69) Santos, L. L.; Mereiter, K.; Paneque, M.; Slugovc, C.; Carmona,

E. New J. Chem. 2003, 27, 107−113.

(70) Lara, P.; Paneque, M.; Poveda, M. L.; Santos, L. L.; Valpuesta, J.

́

E. V.; Carmona, E.; Moncho, S.; Ujaque, G.; Lledos, A.; Alvarez, E.;

Mereiter, K. Chem.Eur. J. 2009, 15, 9034−9045.

5142

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Journal of the American Chemical Society

Article

(87) Asperger, S.; Pavlovic, D.; Kukric, Z.; Sutic, D. Inorg. Chim. Acta

1990, 171, 5−9.

(88) Sutic, D.; Asperger, S.; Borcic, S. J. Org. Chem. 1982, 47, 5120−

4.

(89) Streitwieser, A.; Dafforn, G. A. Tetrahedron Lett. 1969, 10,

1263−1266.

(90) Shiner, V. J.; Rapp, M. W.; Halevi, E. A.; Wolfsberg, M. J. Am.

Chem. Soc. 1968, 90, 7171−7172.

(91) Kundu, S.; Choi, J.; Krogh-Jespersen, K.; Emge, T. J.; Goldman,

A. S. Abstracts of Papers, 237th ACS National Meeting, Salt Lake City,

UT, United States, March 22−26, 2009, INOR-668.

(92) Previous reports of bond cleavage analogous to the carboalkoxy

C−O bond were limited to allyl and vinyl esters: (a) Yamamoto, A.

Adv. Organomet. Chem. 1992, 34, 111. (b) Tatsumi, T.; Tominaga, H.;

Hidai, M.; Uchida, Y. J. Organomet. Chem. 1981, 218, 177.

(93) Examples of acyl C−O bond cleavage in esters: (a) Tatamidani,

H.; Yokota, K.; Kakiuchi, F.; Chatani, N. J. Org. Chem. 2004, 69, 5615.

(b) Yamamoto, Y.; Han, X.-H.; Ma, J.-F. Angew. Chem., Int. Ed. 2000,

39, 1965. (c) Hiraki, K.; Kira, S.-i.; Kawano, H. Bull. Chem. Soc. Jpn.

1997, 70, 1583. (d) Grotjahn, D. B.; Joubran, C. Organometallics 1995,

14, 5171. (e) Yamamoto, T.; Miyashita, S.; Naito, Y.; Komiya, S.; Ito,

T.; Yamamoto, A. Organometallics 1982, 1, 808. (f) Yamamoto, T.;

Ishizu, J.; Kohara, T.; Komiya, S.; Yamamoto, A. J. Am. Chem. Soc.

1980, 102, 3758.

(94) Zhang, X.; Kanzelberger, M.; Emge, T. J.; Goldman, A. S. J. Am.

Chem. Soc. 2004, 126, 13192−13193.

(95) Rybtchinski, B.; Vigalok, A.; Ben-David, Y.; Milstein, D. J. Am.

Chem. Soc. 1996, 118, 12406−12415.

(96) For examples of the (PCP−CH₂) ligand motifs in Rh and Ni

complexes, see: (a) Vigalok, A.; Rybtchinski, B.; Shimon, L. J. W.; Ben-

David, Y.; Milstein, D. Organometallics 1999, 18, 895. (b) van der

Boom, M. E.; Liou, S.-Y.; Shimon, L. J. W.; Ben-David, Y.; Milstein, D.

Inorg. Chim. Acta 2004, 357, 4015.

(97) Romero, P. E.; Whited, M. T.; Grubbs, R. H. Organometallics

2008, 27, 3422−3429.

(98) Whited, M. T.; Grubbs, R. H. J. Am. Chem. Soc. 2008, 130,

5874−5875.

(99) Zhang, X.; Wang, D. Y.; Emge, T. J.; Goldman, A. S. Inorg. Chim.

Acta 2011, 369, 253−259.

(100) See Supporting Information.

(101) Bradley, C. A.; Veiros, L. F.; Pun, D.; Lobkovsky, E.; Keresztes,

I.; Chirik, P. J. J. Am. Chem. Soc. 2006, 128, 16600−16612.

(102) Bradley, C. A.; Veiros, L. F.; Chirik, P. J. Organometallics 2007,

26, 3191−3200.

(103) Williams, B. S.; Leatherman, M. D.; White, P. S.; Brookhart, M.

J. Am. Chem. Soc. 2005, 127, 5132−5146.

(104) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−

241.

(105) (a) Hay, P. J. W.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.

(b) Roy, L. E.; Hay, P. J.; Martin, R. L. J. Chem. Theory Comput. 2008,

4, 1029.

(106) (a) Woon, D. E.; Dunning, T. H., Jr. J. Chem. Phys. 1993, 98,

1358. (b) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007.

(c) Kendall, R. A.; Dunning, T. H., Jr.; Harrison, J. R. J. Chem. Phys.

1992, 96, 6796.

5143

dx.doi.org/10.1021/ja312464b | J. Am. Chem. Soc. 2013, 135, 5127−5143

Article Doi

Cleavage of ether, ester, and tosylate C(sp3)-O bonds by an iridium complex, initiated by oxidative addition of C-H bonds. Experimental and computational studies

DOI: 10.1021/ja312464b

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja312464b

Full text of DOI:10.1021/ja312464b

Products guided by the article

R&D Labs maybe for 1427804-53-2

Relevant to this article

Hot Product