Mechanistic studies of Pd-catalyzed regioselective aryl C-H bond functionalization with strained alkenes: Origin of regioselectivity

Article abstract of DOI:10.1002/chem.201100210

Mechanistic studies of a palladium-catalyzed regioselective aryl C-H functionalization of 2-pyrrole phenyl iodide with norbornene are presented. Kinetic and spectroscopic analyses together with crystallographic data provide evidence for intermediates in a

Full text of DOI:10.1002/chem.201100210

FULL PAPER
DOI: 10.1002/chem.201100210
ꢀ
Mechanistic Studies of Pd-Catalyzed Regioselective Aryl C H Bond
Functionalization with Strained Alkenes: Origin of Regioselectivity
David I. Chai,[a] Praew Thansandote,[b] and Mark Lautens*[a]
Abstract: Mechanistic studies of a pal-
origin of the regioselectivity is due to a
ligand exchange between Iꢀ and HOꢀ
on the norbornyl palladium complex.
These mechanistic studies also impli-
cate that either alkoxide or water is re-
sponsible for the formation of the pal-
ladacycle, but a reversible ring-open-
ing–ring-closing process of the pallada-
cycle with HX can retard the rate of
reaction of a key intermediate. The sig-
nificant aspects of the proposed mecha-
nism are discussed in detail.
ꢀ
ladium-catalyzed regioselective aryl C
H functionalization of 2-pyrrole phenyl
iodide with norbornene are presented.
Kinetic and spectroscopic analyses to-
gether with crystallographic data pro-
vide evidence for intermediates in a
proposed stepwise mechanism. On the
basis of the mechanistic studies, the
ꢀ
Keywords: C H funtionalization ·
heterogeneous catalysis · norbor-
nene · palladacycle · palladium
Introduction
ꢀ
Selective catalytic functionalization of aromatic C H bonds
ꢀ
by palladium is emerging as a powerful methodology for C
C bond formation and has obtained significant attention
within the field of synthetic chemistry.[1] This method simpli-
fies the preparation of substrates and minimizes waste pro-
duction due to the avoidance of pre-functionalization of one
or both of the coupling partners. In particular, the function-
ꢀ
alization of a C H bond ortho to an aryl halide is of funda-
mental importance in synthetic organic chemistry and has
provided new synthetic variations using this method.[2,3]
Thus, the treatment of aryl halide 1 with either an alkyl or
aryl halide containing a functional group such as an amine,
alcohol, alkene, alkyne, or aryl system in the presence of
catalytic Pd0 and excess norbornene provides functionalized
Scheme 1. Reaction of aryl halide 1 with norbornene.
In 2007, as part of an on-going project on annulation, we
reported a coupling reaction between 1-(2-halophenyl)-1H-
ꢀ
aromatic carbo- and heterocycles 2 via ortho-C H function-
pyrrole (4a) and norbornene in the presence of PdACHTUNGTRENNUNG(OAc)2
as a catalyst at 1108C leading to annulation product 6
(Scheme 2). This reaction proceeds by the carbopalladation
alization.[2] On the other hand, the ortho-C H functionaliza-
ꢀ
tion of aryl halide 1 with norbornene in the absence of cou-
pling partners leads to the formation of annulated products
2’.[3] However, if aryl halide 1 contains a nucleophilic func-
tional group adjacent to the halide, a different regioselective
annulation occurs to provide interesting cyclic systems 3
(Scheme 1).[4]
of aryl palladium(II) across norbornene, then a subsequent
[5]
ꢀ
ꢀ
C H functionalization of a pyrrole C H bond (Scheme 3).
[a] Dr. D. I. Chai, Prof. Dr. M. Lautens
Davenport Chemical Laboratories, Department of Chemistry
University of Toronto, 80 St. George Street
Toronto, ON M5S 3H6 (Canada)
Fax : (+1)416-946-8185
E-mail: mlautens@chem.utoronto.ca
[b] Dr. P. Thansandote
Department of Chemistry, University of Cambridge
Lensfield Road, Cambridge, CB2 1EW (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100210.
Scheme 2. Palladium-catalyzed annulation and domino reaction
Chem. Eur. J. 2011, 17, 8175 – 8188
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8175

Some other annulation products with other bicyclic deriva-
tives have also been isolated. Mechanistically, we proposed
that the seven-membered palladacycle Pd-6 is formed as an
intermediate in this process.
Alternatively, in a related process, we reported the syn-
thesis of tetracyclic fused pyrroles by a palladium-catalyzed,
norbornene-mediated domino reaction using pyrrole 4a and
alkyl halide 5 (Scheme 2).[2l] This methodology constructed
tion providing tetracyclic compound 7 and re-generating the
active palladium catalyst via reductive elimination.
ꢀ
Due to the formation of two different C H functionaliza-
tion products from the same substrate, the conflicting regio-
selective palladacycle formations from complex 9 became
our primary focus of interest. Thus, we began considering
the controlling factors for the formation of the five-mem-
bered palladacycle 10 over the seven-membered palladacy-
ꢀ
ꢀ
three C C bonds in one procedure including two challeng-
cle 14. In view of the potential of selective ortho C H func-
ꢀ
ing C H functionalizations of unactivated aryl ring systems.
tionalizations, this study would be significant to discovering
new catalytic systems. To develop a simple system suitable
to mechanistic studies, we decided to examine the formation
of the benzocyclobutene 15, which is formed from the palla-
dacycle 10 by a reductive elimination (Scheme 3). Herein,
we will describe detailed mechanistic studies including ki-
netic and NMR studies, to explain the regiomeric outcome
of the reaction.
The proposed mechanism is based upon the formation of
the five-membered palladacycle Pd-7 according to the mech-
anistic studies made by Catellani and Cheng (Scheme 3).[6]
As we will discuss in more detail, these two reactions share
common intermediates in their early stages, but show impor-
tant differences in their late stages.
Results and Discussion
Screening and optimization study: The initial investigation
of the benzocyclobutene reaction focused on using 2-pyrrole
substituted phenyl bromide and norbornene while varying li-
gands, bases, solvents, and temperature. Whereas reactions
with electron rich or neutral ligands such as (p-MeOC6H4)3P
or PPh3 yielded only the annulation products, the formation
of the benzocyclobutene was first observed with electron-
poor ligands such as (p-CF3C6H4)3P, but in only 1:1 selectivi-
ty of 15 and 6. However, we discovered that the reactions
with the 2-pyrrole substituted phenyl bromide and this elec-
tron poor ligand gave poor reproducibility. To our delight,
the benzocyclobutene reaction was found to be both high
yielding and reproducible using water as a co-solvent and
hydroxide ion as a base. Interestingly, the use of the electron
neutral ligand, PPh3, led to excellent selectivity in this case.
Thus, the benzocyclobutene 15 was obtained in 95% yield
Scheme 3. Proposed mechanisms of palladium-catalyzed annulation and
domino reactions. Ligands and halides are omitted for clarity.
using 10 mol% PdACTHNUTRGNEUNG(OAc)2, 22 mol% PPh3, 4 equiv norbor-
nene, and 2 equiv Bu4NOH in toluene and water (1:1).
A list of the controlling factors obtained from the optimi-
ꢀ
A simplified mechanistic hypothesis for these two reac-
tions is outlined in Scheme 3. Both catalytic cycles start with
the aryl palladium complex 8, which is formed by oxidative
addition of aryl iodide 4a. Subsequent norbornene insertion
into the resulting complex 8 provides the alkyl palladium
complex 9. At this stage, the complex 9 can give rise to two
zation studies for the C H functionalization is summarized
in Table 1 and interpreted in Table 2. Among the important
factors governing the selectivity, the temperature and base
had the most significant impact to favor the benzocyclobu-
tene formation. It seems that the pyrrole annulation reac-
tion can only proceed at temperatures higher than 908C.
Also, no benzocyclobutene products were observed with
stronger bases such as KHMDS or KOR or weaker bases
such as K3PO4 or K2CO3. It is worth noting that the forma-
tion of 15 was not observed unless the reaction was run in a
sealed system.
ꢀ
regioselective intermediates 10 or 14 by a C H functionali-
ꢀ
zation. Therefore, this C H functionalization step appears
of fundamental importance for the determination of the re-
gioselectivity. The intermediate 14 undergoes a reductive
elimination to provide the annulation product 6. The other
complex 10 couples with alkyl halide 5 presumably via a
PdIV intermediate to give complex 11. Likely due to the
steric effect of norbornene of complex 13, exerted by two
ortho-substituents, extrusion of norbornene occurs leading
to complex 12, which undergoes a cyclization to give com-
plex 13. The catalytic cycle is finally closed by direct aryla-
Development of a system for mechanistic study: Upon suc-
cessful optimization studies, we sought to develop a system
suitable for kinetic and spectroscopic analysis of the palladi-
um intermediates formed during the reaction. Unfortunate-
ly, the palladium intermediates 9 and 10 are not sufficiently
8176
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8175 – 8188

ꢀ
Aryl C H Bond Functionalization
FULL PAPER
Table 1. Experiments to determine controlling factors for benzocyclobutene.[a]
from reactions of 1-(2-iodo-
phenyl)-1H-pyrrole (4a) under
the optimized reaction condi-
tions were used to compare
with the data obtained for our
model substrate. It is worth
noting that attempts to take the
kinetic study of the full catalyt-
ic cycle were unsuccessful due
to the significant formation of
by-products.
Entry
Original conditions
Changed conditions
15 [%]
6 [%]
1
2
3
4
5
6
7
toluene/H2O 1:1
808C
Bu4NOH
H2O
1008C
Cs2CO3
100
0
0–10
trace
0
90–100
30
0
Bu4NOH
KHMDS or KOR (R=Me and tBu)
toluene/H2O 1:1
aryl iodide 4a
0.30m[e]
toluene[b]
aryl bromide[c]
0.15m
80
20
Kinetic analysis of carbopalla-
dation steps: Step 1: To deter-
mine possible intermediates
and obtain mechanistic insight
for the carbopalladation step,
we monitored the reaction be-
tween aryl palladium complexes
and norbornene at variable
45–100[d]
50
0–55[d]
18
[a] Measured by 1H NMR. [b] KOH was used instead of Bu4NOH. [c] 2-Pyrrole phenyl bromide was used in
place of 4a. [d] Not reproducible. [e] Concentration (mmol) of 4a in 1 mL toluene.
[a]
ꢀ
Table 2. Controlling factors for aryl C H functionalization.
temperatures
by
1H
and
31P NMR spectroscopy. The rel-
ative rates of the carbopallada-
tion of the isolated palladium
complexes (Pd-1) were studied
with 1.5 equiv of norbornene in
Entry Reagent
Importance
CDCl3 (Figure 1). All reactions
represented here showed no or
1
solvent
choice of solvent does not control the selectivity since reactions still worked in neat
conditions on water
2
3
temperature important for selectivity; full conversion to 15 at 808C, but low selectivity at 1008C
little induction period and all
reaction rates were measured
after the reaction was initiated.
The rates of carbopalladation
conducted with Pd-1 varying
electronics and sterics are illus-
trated by curves in Figures 1
and 2. The kinetics of the car-
Bu4NOH
important for selectivity; it can act as a phase-transfer reagent as well as a base. Con-
trols the basicity of organic layer.
4
5
water
halide
low selectivity without water; removes by-products such as HX (X=OAc or I)
Aryl bromide gave un-reproducible results; aryl iodide necessary for reproducible re-
sults
6
concentration 0.3m or higher concentration required for full conversion; negligible variations in selec-
tivity.
[a] Reactions were conducted on 0.15 mmol scale with PdACHTNUTGRNEG(UN OAc)2 (10 mol%), PPh3 (22 mol%), Bu4NOH
(0.3 mmol), toluene (0.5 mL), H2O (0.5 mL), and norbornene (0.6 mmol).
bopalladation was consistent
with the reaction being second
order. The second-order kinetic
stable to isolate, presumably due to steric hindrance be-
tween the pyrrole and norbornyl moieties. Thus, we decided
to focus on synthesizing palladium complexes with unsubsti-
tuted phenyl iodides to use for our mechanistic studies.
For high-yielding reactions with minimal byproduct for-
mation, we modified the optimized reaction conditions in
Equation (1) can be obtained by considering excess norbor-
nene with respect to Pd-1, where x is the molar fraction of
Pd-1 at t (i.e., x=[Pd-1]t/ACTHNUTRGNE[NUG Pd-1]0) (see Supporting Informa-
tion for derivation).
the following ways: 1) a stoichiometric [Pd
ACHTUNGTRENN(NUG PPh3)4] catalyst
was used instead of Pd(OAc)2/PPh3 mixture to avoid the re-
AHCTUNGTRENNUNG
duction of PdII to Pd0 and the potential competitive coordi-
nation between acetate and iodide ions for the palladium
complexes. 2) The mechanistic studies were conducted from
the substrate immediately leading to each intermediate of
interest. 3) Changes in reaction conditions were made when
necessary such as switching the solvent to overcome solubili-
ty issues, lowering the temperature to avoid decomposition
of the intermediates, and changing the base to increase the
reaction rate. To correlate the proposed mechanism to the
original reaction conditions, the crude NMR data obtained
As shown in Figure 2, the plot of ln(1+(1/
time shows a linear plot with slope equal to k
ACHTUNGTRENNUNG
·CHTUNGTRENNUNG
for the carbopalladation step, but slightly deviates from the
linear plot over time. Clearly, the relative rates of Pd-1 with
different electronic properties follow the order: Pd-1b
(kapp =0.0798mꢀ1 minꢀ1) > Pd-1a (kapp =0.018mꢀ1 minꢀ1) >
Pd-1c (kapp =0.00336mꢀ1 minꢀ1) at 258C, which suggests that
Chem. Eur. J. 2011, 17, 8175 – 8188
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8177

M. Lautens et al.
ported for the Heck reaction of iodobenzene and methyl ac-
rylate in the presence of a PdACHTNUGRTNEUNG
(OAc)2/PPh3 catalytic system,[7]
suggesting that norbornene insertion is kinetically faster
than the insertion of methyl acrylate. Indeed, a reaction of a
mixture of aryl halide, alkene, and norbornene in the Catel-
lani process shows that norbornene insertion is faster than
alkene insertion unless the aryl halide exerts a large steric
effect by the presence of two ortho-substituents.[8]
This observed electronic trend indicates that the migra-
tion of the aryl group from the palladium complex onto the
alkene is viewed as a nucleophilic addition to the double
bond. Two mechanisms are the most widely accepted in the
Heck reaction:[9] the neutral mechanism, which is initiated
by the dissociation of a phosphine ligand and subsequent as-
sociation of an alkene,[9b] and the cationic mechanism, which
is initiated by the dissociation of a halide ligand and subse-
quent association of an alkene.[9c] However, since the rate of
carbopalladation was not largely affected by adding 1 equiv
of Bu4NI in our reaction, the cationic mechanism seems less
likely in this case.
With the proposed intermediate Pd-2 (Scheme 4), we
sought to detect the intermediate by variable temperature
NMR during the carbopalladation process.[10] Unfortunately,
any possible intermediates could not be detected at a variety
of temperatures presumably due to the fast migration of the
aryl group to the double bond in norbornene (k2 and/or
kꢀ1 @k1).
Figure 1. Comparison of rates of carbopalladation between Pd-1a–d and
norbornene.
Scheme 4. Proposed mechanism of carbopalladation.
Kinetic order of the reaction: Step 1: Next, we turned our
attention to analyze the kinetic orders of Pd-1, norbornene,
and PPh3 to elucidate the proposed mechanism. We chose
Pd-1b (R=4-OMe) in this study due to easier spectroscopic
analysis. Rate Equation (2) was derived for this study using
the steady state approximation where k1 and kꢀ1 represent
reversible ligand exchange and k2 represents the carbopalla-
dation step (see Supporting Information for the derivation).
It is worth noting that Pd-3 did not undergo retro-carbopal-
ladation, even at 608C, both in the presence and absence of
PPh3 unless a large steric effect exerted by ortho-substitu-
ents results in destabilization of Pd-3.
Figure 2. Linear relationship of ln(1+(1/
palladation of Pd-1a–d (42.9 mm) with norbornene (64.3 mm) in 0.7 mL
of CDCl3 at 258C; x=[Pd-1]t/A[Pd-1]0;
app =0.0798mꢀ1 minꢀ1 (R=4-
OMe), (R=H), (R=2-Me),
0.018mꢀ1 minꢀ1 0.0104mꢀ1 minꢀ1
0.00336mꢀ1 minꢀ1 (R=4-CF3).
ACHTUNGTRNE(NUNG 2·x))) versus t [min] for carbo-
G
k
electron-rich substituents favor carbopalladation. The reac-
tion with sterically hindered Pd-1d (kapp =0.0104mꢀ1 minꢀ1)
occurred slower than that of the electron-neutral palladium
complex, Pd-1a, but faster than that of the electron poor
palladium complex Pd-1c.
We first examined the rate of the carbopalladation from
Pd-1b by reacting norbornene (42.9 mm) and PPh3
Calculating those rate constants (kapp) at 258C, provided
¼
the following activation energies: DG =16.5 kcalmolꢀ1
(R=4-OMe), 17.4 kcalmolꢀ1 (R=H), 18.4 kcalmolꢀ1 (R=
CF3), 17.7 kcalmolꢀ1 (R=2-Me). These activation values are
lower than the 23 kcalmolꢀ1 activation energy previously re-
8178
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8175 – 8188

ꢀ
Aryl C H Bond Functionalization
FULL PAPER
(21.4 mm) with varying Pd-1b concentration (21.4 mm to
85.7 mm) in CDCl3 at 258C. The reaction was monitored by
1H NMR, and the initial rate was determined at each con-
centration of Pd-1b. As described in Figure 3, the order in
Pd-1b was determined to be first order, which is consistent
with the rate equation involving a monomeric palladium
species.
Figure 5. A plot of initial rate [mminꢀ1] vs. concentration of PPh3: In-
verse-order dependence.
which is only a slight deviation from an inverse first order
due to k2 value [r / {kꢀ1
[PPh3] + k2}ꢀ1, see Equation (2)].
ACHTUNGTRENNUNG
Proposed mechanism of the carbopalladation: Step 1: The
close fit between experimental and predicted kinetic data
allows us to support the proposed mechanism shown in
Scheme 4. For example, Equation (2) predicts the reaction
to be first order in Pd-1b, first order in norbornene, and in-
verse order in PPh3, consistent with the experimental kinetic
data. The reaction is first initiated by a reversible ligand ex-
change between the phosphine ligand and norbornene, then
a subsequent carbopalladation to form Pd-3a.
Figure 3. A plot of initial rate [mminꢀ1] vs concentration of Pd-1b: First-
order dependence.
The order in norbornene was next determined by analo-
gous procedure, using Pd-1b (42.9 mm), PPh3 (21.4 mm), and
varying norbornene concentration (21.4 mm to 85.7 mm) in
CDCl3 (Figure 4). As expected, a first order dependence in
norbornene was observed.
X-ray crystal structure of complex Pd-3e: Pure palladium
complex Pd-3e was synthesized from the corresponding aryl
iodide and norbornene using [PdACTHNUTRGNEUNG(PPh3)4] in THF at 608C in
good yields (see Scheme 6 below). A relatively fast crystalli-
zation was necessary due to the decomposition of the com-
plex, which occurred during prolonged storage in solution at
room temperature (~12 h). Crystals suitable for X-ray analy-
sis were obtained in a few hours by dissolving Pd-3e in
CH2Cl2 followed by the addition of MeOH.
The X-ray structure clearly shows that the aryl ring and
the palladium are positioned cis and exo to the norbornyl
group (Figure 6). The palladium possesses a slightly distort-
ed square-planar center with the PPh3, Iꢀ, norbornyl group,
and p-system of the aryl ring, in which the ipso-carbon (C8)
of the p-system occupies one of the four coordination sites
(14.38 dihedral angle between I-Pd-C8 and P-Pd-C1 planes).
The ortho-carbon (C13) is out-of the plane (38.68 dihedral
angle between I-Pd-C13 and P-Pd-C1 planes). The h2 coor-
dination to palladium can attribute to the longer bond
Figure 4. A plot of initial rate [mminꢀ1] vs concentration of norbornene:
First-order dependence
ꢀ
Finally, the order of PPh3 was examined using Pd-1b
(42.9 mm), norbornene (42.9 mm), and varying the PPh3 con-
centration (21.4 mm to 85.7 mm) (Figure 5). Qualitatively,
the reaction rate decreased with increasing [PPh3], suggest-
ing an inverse order dependence on phosphine ligands. A
nonlinear least-squares fit calculated from Excel function to
length of C8 C13, which is measured to be 1.41 ꢁ compared
ꢀ
with 1.39 ꢁ for the other aryl C C bonds. In addition, the
ꢀ
ipso carbon C8 Pd bond length is measured to be 2.42 ꢁ,
ꢀ
while C13 Pd has 2.52 ꢁ, presumably due to stronger back-
donation to C8 from palladium.[11] These bond lengths are
2
ꢀ
slightly longer than previously reported h -alkene palladium
the equation: r=a
N
bonds.[12]
Chem. Eur. J. 2011, 17, 8175 – 8188
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8179

M. Lautens et al.
Scheme 6. Syntheses of norbornyl palladium complexes.
With the palladium complexes in hand, we next sought to
probe the Hammett relationship to describe the rate acceler-
Figure 6. ORTEP structure of Pd-3e as determined by X-ray crystallo-
graphic characterization.
ꢀ
ation by electronic effects in the C H functionalization step.
For this study, we switched the base from Bu4NOH to
KOPh to increase the reaction rate; full conversion was ob-
tained in 5 days with Bu4NOH compared to 30 min with
KOPh. However, the instability of the electron-poor palladi-
um complex Pd-3 f and the large rate fluctuation without
stirring in a NMR tube resulted in poor kinetic data
(Table 3).
It is known that the dynamic NMR behavior of an analo-
gous palladium complex results in a mixture of two isomers
due to rotation of the aryl group.[6e] Interestingly, in solid
state, the aryl group adapts a conformation in which the me-
thoxy group is positioned near the palladium. It is worth
noting that the H on C9 seems to experience more steric
effect with the norbornyl group than the H on C13, which is
positioned away from the norbornyl group (Figure 6). This
assumption suggests that the complex possessing an ortho-
substituted aryl group would adapt a structure such as Pd-A
(Scheme 5). However, the conformation Pd-B should be a
ꢀ
Table 3. Kinetic data for palladacycle formation via C H functionaliza-
tion.[a]
ꢀ
key for the C1 H functionalization leading to the palladacy-
cle. Presumably, an equilibrium exists between these two
structures in solution. A detailed structural analysis of this
complex is on-going and will be published in due course.
Entry
Substrate
Product
t [h]
Conversion [%][b]
1
2
3
4
5
Pd-3a
Pd-3a
Pd-3e
Pd-3e
Pd-3 f
Pd-5a
Pd-5a
Pd-5e
Pd-5e
Pd-5 f
0.5
1
0.5
1
67[c]
86[c]
67
86
0.5
100[d]
[a] Reactions were conducted with Pd-3 (0.03 mmol), KOPh (0.09 mmol)
and PPh3 (0.054 mmol) in CH2Cl2 (2.0 mL). [b] Measured by 1H NMR.
[c] Small amount of decomposed products found. [d] Large amount of
decomposition products found.
Scheme 5. Proposed structure of the 2-substituted phenyl norbornyl palla-
dium complex.
Thus, we turned our attention to NMR spectroscopic anal-
ysis for the detection of possible intermediates. 31P NMR
spectroscopic analysis of Pd-3a in CDCl3 showed one singlet
signal at 38.9 ppm (Figure 7a). This signal was broadened
upon the addition of 1 equiv of PPh3 at 258C, which could
be a result of fast equilibrium between Pd-3a and a new pal-
ladium complex (Figure 7b). However, the resonances are
sharpened to a single signal at low temperature (Figure 7d),
suggesting that the phosphine ligand on Pd-3a exchanges
with free phosphine at 258C, but not at ꢀ508C.
NMR analysis of palladacycle formation: Step 2: Pure palla-
dium complexes Pd-3a and Pd-3e were synthesized from
their corresponding aryl iodides and norbornene using [Pd-
ACHTUNGTRENNUNG(PPh3)4] in THF at 608C in good yields (Scheme 6). Howev-
er, the electron-poor palladium complex Pd-3 f could not be
prepared by this one-step method due to poor stability of
the product under the conditions. Since we had observed re-
action between Pd-1c and norbornene in CDCl3 by NMR,
we thought to use this two-step method for the synthesis of
Pd-3 f. Indeed, these sequential reactions gave relatively
pure complex Pd-3 f in 80% yield, but with incomplete con-
version.
We next examined 31P NMR of Pd-3a with KOPh in the
absence of PPh3 in CDCl3 (Figure 8). When Pd-3a and
excess KOPh were combined at ꢀ408C, a new broad peak
8180
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8175 – 8188

ꢀ
Aryl C H Bond Functionalization
FULL PAPER
Figure 7. 31P NMR study of reaction between Pd-3a (30 mm) and PPh3
(60 mm) in CDCl3. a) 31P NMR of Pd-3a at 258C. b) 31P NMR of Pd-3a
taken at 10 min after addition of PPh3 at 258C. c) 31P NMR of Pd-3a
taken at 10 min after cooling to 08C. d) 31P NMR of Pd-3a taken at 10
min after cooling to ꢀ508C.
Figure 9. 31P NMR study of reaction between Pd-3a (30 mm), KOPh
(150 mm), and PPh3 (60 mm) in CDCl3. a) 31P NMR taken at 20 min after
mixing of Pd-4 and PPh3 in CDCl3 cooled in liquid nitrogen, ꢀ508C. b)
31P NMR taken at 10 min after warming to ꢀ208C. c) 31P NMR taken at
10 min after warming to 08C.
least ꢀ208C to mix the reagents due to high viscosity. We
believe that triphenylphosphine oxide was formed due to a
small amount of water present in either Pd-3a or the sol-
vent. The result showed an immediate transformation to the
palladacycle Pd-5a upon warming up to 08C, with resonan-
ces appearing at d 27.0 and 22.2 ppm as two sets of doublet
with J=17.8 Hz. On the basis of these NMR studies, we
postulate that the new palladium complex Pd-4 is a reactive
species leading to the palladacycle Pd-5a.
Next we sought to deduce a structure of the new palladi-
um complex Pd-4. In addition, incomplete conversion from
Pd-3 made structural determination difficult. Thus, the
1H NMR of Pd-4 formed in-situ from Pd-3a was analyzed at
1
maximum conversion. This H NMR corresponding to Pd-4
shows a close match with one of Pd-3a, suggesting structural
similarity (Figure 10).
Figure 8. 31P NMR study of reaction between Pd-3a (30 mm) and KOPh
(150 mm) in CD2Cl2. a) 31P NMR taken at 20 min after mixing of Pd-3a
and KOPh in CD2Cl2 cooled in liquid nitrogen, ꢀ408C. b) 31P NMR
taken 10 min after warming to ꢀ208C. c) 31P NMR taken 10 min after
warming to 08C. d) 31P NMR taken 10 min after warming to 108C. e)
31P NMR taken 10 min after warming to 258C.
was observed at 30.7 ppm. Upon warming to 258C, this new
signal grew to a 1:2.4 ratio corresponding to 70% conver-
sion. Unfortunately, attempts to reach full conversion by in-
creasing the amount of base, higher temperature, or increas-
ing concentration of the reaction resulted in decomposition
of this species, Pd-4.
A sample containing Pd-4 and a small amount of Pd-3
was treated with excess PPH3 to follow its conversion to Pd-
5. The reaction of the new complex Pd-4 was investigated
by 31P NMR by subjecting excess PPh3 in CDCl3 to a mix-
ture containing Pd-4 and a small amount of Pd-3a at ꢀ508C
(Figure 9). Some Pd-5a was present before the NMR experi-
ment was started since it was necessary to warm up to at
Figure 10. Comparison of 1H NMR spectra of Pd-4 and Pd-3a at ꢀ0.1 to
3.6 ppm. a) 1H NMR of mixture of Pd-4 and Pd-3a at maximum conver-
sion, 70%. b) 1H NMR of pure Pd-3a; * indicates peaks for Pd-4 and ›
indicates peaks for Pd-3a.
Chem. Eur. J. 2011, 17, 8175 – 8188
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8181

M. Lautens et al.
On the basis of these NMR studies, we postulate that Pd-
activation may proceed. Those scenarios can rationalize the
KIE observed.
4 is an active palladium species, formed from Pd-3a by a
ligand exchange between Iꢀ and PhOꢀ, which undergoes the
[13,14]
ꢀ
C H functionalization to form the palladacycle Pd-5a.
Proposed mechanism of palladacycle formation: Step 2: For
ꢀ
Catellani suggested a similar ligand exchange process in her
palladium-catalyzed C H functionalizations, several reac-
mechanistic study on the C H functionalization of the nor-
tion pathways[17–20] have been proposed and, among those,
electrophilic aromatic substitution and concerted metalla-
tion–deprotonation (CMD) are the most widely accepted.
ꢀ
bornyl palladium complex with chloride and pyridine as li-
gands.[6a] In this report, she noted that the complex was
stable at ꢀ508C, but underwent palladacycle formation at
ꢀ308C with excess pyridine.
ꢀ
Comparatively, the oxidative C H insertion process is rare
in the absence of a directing group and Heck-like processes
cannot provide the palladacycles such as Pd-5a.
Kinetic isotope effect experiment of palladacycle formation:
ꢀ
Step 2: To further study the C H functionalization process,
we sought to measure the H/D kinetic isotope effect (KIE)
for the deuterated substrates Pd-3a[D] and Pd-3a[D5] [Eqs.
(3) and (4)]. The Pd-3a[D] was synthesized from the palla-
dacycle by treatment of DCl (see Section on reversibility of
palladacycle for synthesis). No kinetic isotope effect was ob-
served in the competition of Pd-3a and its pentadeuterated
complex Pd-3a[D5]. This suggests that the rate-limiting step
ꢀ
does not involve C H bond breakage. However, a primary
isotope effect of 4.2 was determined from mono-ortho-deu-
terated complex Pd-3a[D]. In fact, this unusual KIE has
been previously reported by Buchwald for the development
of oxindole synthesis from a-chloroacetanilides via Pd-cata-
lyzed C H functionalization (Scheme 7).[15] In this report, he
ꢀ
The data assembled for the palladacycle formation allow
us to propose a mechanism. As shown in Scheme 8, the re-
action is initiated by a ligand exchange, which seems to be a
rate limiting step (k3 < k4, k6) based on the observed kinetic
ꢀ
isotope effect (KIEinter =1.0 and KIEintra =4.2). The C H
functionalization may proceed by an electrophilic aromatic
substitution to give a cationic five-membered palladacycle,
which is subsequently deprotonated to provide the pallada-
cycle Pd-5a. The observed intramolecular KIE of 4.2 is rea-
sonable if the k4 step is reversible and fast relative to the de-
protonation step. A second possible mechanism is a CMD-
type mechanism via s-metathesis or termolecular electro-
philic (SE3) mechanisms. Considering the slow ligand ex-
change step, this CMD mechanism also should explain the
observed KIE values.
ꢀ
Scheme 7. Buchwaldꢂs palladium-catalyzed C H functionalization for ox-
indole synthesis.[15]
suggested that the first oxidative addition (kA) is slow rela-
ꢀ
tive to the C H functionalization step and is rate-determin-
ing. This can explain the intermolecular KIE. Despite the
rate-limiting oxidative addition, the alkyl palladium complex
ꢀ
ꢀ
B can still choose either C D bond or C H bond breakage,
ꢀ
resulting in a primary KIE. Then, the C H functionalization
step may proceed by either an electrophilic or a Heck-type
mechanism if carbopalladation of the aromatic ring (kB and
ꢀ
kC) is fast and reversible. Alternatively, a direct C H bond
Scheme 8. Proposed mechanism of palladacycle formation.
8182
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8175 – 8188

ꢀ
Aryl C H Bond Functionalization
FULL PAPER
Table 4. Scope of benzocyclobutene reaction with aryl iodides and
D2O.[a]
X-ray crystal structure of Pd-5a: Single crystals suitable for
X-ray analysis were obtained from a mixture of CH2Cl2 and
MeOH. An ORTEP drawing of Pd-5a is shown in
Figure 11. The palladium centre possesses square-planar ge-
ometry with dihedral angle of 7.988 between P2-Pd-C1 and
ꢀ
ꢀ
P1-Pd-C13 planes. The bond lengths Pd P1 and Pd P2 are
2.41 and 2.38 ꢁ, respectively.
Entry
Conditions
T [8C]
16 [%]
15a [%]
1
2
A
A
A
B
B
80
15 (no D)
45 (1:1 D/H)
89 (no D)
5 (no D)
0
0
0
75
10
100
100
80
3[b]
4
5
100
90 (1:1 D/H)
[a] Condition A: 4 (0.15 mmol), norbornene (0.6 mmol), 10 mol% Pd-
(OAc)2, 22 mol% PPh3, Cs2CO3 (0.3 mmol), toluene (1.5 mL), D2O
(0.3 mmol); condition B: (0.15 mmol), norbornene (0.6 mmol),
10 mol% Pd(OAc)2, 22 mol% PPh3, KOD (0.3 mmol), toluene (0.5 mL),
AHCTUNGTRENNUNG
4
AHCTUNGTRENNUNG
D2O (0.5 mL); [b] with CD3OD (0.3 mmol) instead of D2O.
mation for the detection of ring-opening product with D2O).
This result suggests that our biphasic and hydroxide-contain-
ing system for the benzocyclobutene reaction can reduce the
rate of the ring-opening reaction by removing HI formed
during the ring-closing reaction.
Figure 11. ORTEP structure of Pd-5a as determined by X-ray crystallo-
graphic characterization.
Reversibility of palladacycle: Step 2: While we conducted
the optimization study, we found an interesting reaction,
suggesting the possibility for reversible palladacycle forma-
tion. The compound 16 was isolated in an H/D ratio of 1:1
when the reaction was performed under either the pyrrole
annulation conditions (A) in the presence of 2 equiv D2O or
the benzocyclobutene reaction conditions (B) at 1008C
(Table 4, entries 2 and 5). However, no deuterium was incor-
porated when the reaction was conducted with CD3OD at
1008C (Table 4, entry 3). In fact, a stoichiometric reaction
conducted under the anhydrous conditions (with Cs2CO3 as
the base) revealed that the palladacycle was present in only
trace amount. This result suggested that the palladacycle
can only be formed with water present.[16] In addition, no
deuterium was observed in 16 when the reaction was per-
formed at 808C under both conditions (Table 4, entries 1
and 4), indicating that the reversibility only happens if suffi-
cient energy is provided (i.e., >1008C).
NMR analysis of reductive elimination: Step 3: Since little
or no reductive elimination occurred at temperatures lower
than 708C in any solvent system, this study was conducted
in [D8]toluene at 808C. The palladacycle Pd-5a was only
partially soluble below 608C, but became soluble at 808C.
The relative rate of the reductive elimination was unaf-
fected by the addition of hydroxide or water, suggesting that
this step occurs directly from the palladacycle. However, if
the reaction was run at 40–608C in the absence of water,
Pd-5a slowly decomposed to palladium black; whereas no
decomposition was detected in the presence of water. Thus,
it seems that even though water encourages palladacycle re-
versibility, it also increases the stability of the palladacycle.
In addition to Pd-5a, Pd-5e (R1 =OMe, R2 =H) was pre-
pared to study electronic effects on the reductive elimina-
tion step. Surprisingly, Pd-5e did not produce the benzocy-
clobutene in [D8]toluene at 808C, presumably because this
electron-rich palladacycle is too stable. Indeed, the reactions
using 4-iodoanisole under the optimized catalytic and stoi-
chiometric reaction conditions did not provide correspond-
ing benzocyclobutenes, even though the palladacycle was
detected in the crude NMR analysis (see Section on the
Scope of the reaction with other bicyclic systems and aryl
halides for details).
To further support this reversibility hypothesis, we con-
ducted a reaction between the palladacycle Pd-5a and either
D2O or DCl [Eq. (5)]. The result showed that the palladacy-
cle underwent a ring opening reaction with both D2O and
DCl, but in low conversion with D2O (see Supporting Infor-
Relatively fast reductive elimination at this temperature
led to unreliable kinetic data. Thus, we studied this reaction
by qualitative NMR analysis. However, no intermediates or
by-products were detected in this study. Thus, we propose
that the reductive elimination occurs directly from Pd-5a.
Unexpected new palladium complex: A few palladium(II)
hydroxo-bridged dimers have been synthesized, character-
Chem. Eur. J. 2011, 17, 8175 – 8188
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8183

M. Lautens et al.
ized, and studied.[14] Experimentally, these complexes are
usually prepared from an aryl halide by treatment of
[(PPh3)2PdCl2] and KOH in biphasic conditions (benzene
and water). In these reactions, the hydroxide ion serves two
roles: to reduce PdII to Pd0 and to act as a bridging li-
gand.[14d] Previous studies have demonstrated that the reduc-
tion process proceeds first by hydroxide-induced dispropor-
tionation of [(PPh3)2PdIICl2] to highly reactive Pd0PPh3 and
triphenylphosphine oxide. Then, oxidative addition of the
aryl halide, followed by ligand exchange with hydroxide, re-
sults in the organopalladium hydroxo dimer [{(PPh3)ArPdII-
modified biphasic conditions (258C for two days) were ob-
tained for both catalytic and stoichiometric reactions (see
Supporting Information for NMR data). They clearly
showed that complexes analogous to the previously identi-
fied norbornyl palladium complex Pd-3 and palladacycle Pd-
5 were present. Moreover, a complex analogous to Pd-5 was
present as a major species. This result also suggests the re-
ductive elimination step to be the rate-limiting step since no
benzocyclobutene was obtained at 258C.
Both benzocyclobutene and pyrrole annulation reactions
are initiated by oxidative addition of Pd0 [22] to aryl iodide to
form the PdII complex Pd-7 (Scheme 9). The oxidative addi-
tion is known to be slower in the presence of alkenes by for-
ACHTUNGTRENNUNG
palladium complex from 2-pyrrole substituted phenyl iodide
resulted in isolation of an orange powder (50% yield),
which clearly shows distinct patterns from the hydroxo-
bridged dimers in H NMR. Fortunately, a good quality crys-
tal suitable for X-ray analysis could be obtained from
mation
of
unreactive
AHCTUNGTRENNUNG
alkene–palladium
complex
[Pd0(norbornene)
(PPh3)2].[23a] Then, a reversible ligand ex-
change with norbornene leads to Pd-8, followed by carbo-
palladation with norbornene to provide Pd-9 according to
the observed kinetic orders: first order to both palladium
and norbornene and inverse order to the phosphine ligand.
At this point, a ligand exchange between Iꢀ and HOꢀ leads
to Pd-10, as exemplified by an analogous alkoxide palladium
1
CH2Cl2 and MeOH (Pd-6, Figure 12). The crystal structure
complex Pd-4 was observed by both H and 31P NMR. Sub-
1
sequently, Pd-10 undergoes a cyclization to give the five-
membered palladacycle Pd-11 via either electrophilic aro-
matic substitution or concerted metallation–deprotonation
(CMD) mechanism based on the observed KIE. (KIEinter
=
1.0 and KIEintra =4.2). At this stage, the ligand exchange
ꢀ
step was determined to be slower than the C H functionali-
zation step (k10 < k11). A reductive elimination of Pd-11 can
occur to provide the benzocyclobutene 15. However, if suffi-
cient energy is provided (i.e., >1008C), the reversibility be-
Figure 12. ORTEP structure of Pd-6 as determined by X-ray crystallo-
graphic characterization.
consists of unusual dimeric units with the p-system of a pyr-
role moiety acting as a bridging ligand. This dimer shows a
ꢀ
slight deviation from C2 symmetry with the Pd1 C1 and
ꢀ
Pd1 C2 distances being 2.34 and 2.51 ꢁ, respectively, and
ꢀ
ꢀ
Pd2 C11 and Pd2 C12 distances being 2.38 and 2.55 ꢁ, re-
spectively. This difference attributes to a slightly shorter
ꢀ
bond length for C11 C12 due to less p-electron contribution
ꢀ
ꢀ
to palladium: 1.38 ꢁ for C11 C12 versus 1.42 ꢁ for C1 C2.
ꢀ
In addition, the Pd1 Pd2 bond 2.86 ꢁ is in a range of typi-
ꢀ
cal Pd Pd bond lengths for known palladium dimers, 2.7–
3.0 ꢁ.[14,21]
Summary of mechanistic studies and proposed mechanism
of benzocyclobutene formation: Our NMR studies to identi-
fy the palladium intermediates in the reactions of aryl hal-
ides with norbornene, together with kinetic studies for each
reaction parameter, have shown that the reaction occurs
through the Pd-1, Pd-3, Pd-4, and Pd-5 intermediates as
shown in Schemes 4 and 8. To establish a correlation be-
tween the reactions using 2-pyrrole substituted phenyl
iodide 4a and phenyl iodide, 31P NMR data of 4a under
Scheme 9. Proposed mechanisms of benzocyclobutene and pyrrole annu-
lation reactions based on mechanistic studies.
8184
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8175 – 8188

ꢀ
Aryl C H Bond Functionalization
FULL PAPER
Table 5. Scope of benzocyclobutene reaction with aryl iodides.[a]
tween Pd-11 and Pd-9 allows fast equilibrium prior to a pyr-
ꢀ
role C H bond functionalization to afford the seven-mem-
bered palladacycle Pd-12 (k13 < k10). Then, Pd-12 can un-
dergo a reductive elimination to produce the pyrrole annula-
tion product 6. This mechanism can explain how the deuteri-
um-labeled 16 was formed in an H/D ratio of 1:1 (Table 4,
¼
entry 5). The activation energy (DG13 ) toward the pyrrole
¼
annulation product 6 is lower than that (DG12 ) of benzocy-
clobutene 15, thus favoring 6 under the equilibrium condi-
¼
¼
tion (DG13 < DG12 ).
In case of the Pd(OAc)2/PPh3 catalytic system, we expect
Entry
Substrate
R
Product
Yield [%][b]
ACHTUNGTRENNUNG
1
2
3
4
5
6
7
8
18a
18b
18c
18d
18e
18 f
18g
18h
18i
H
2-Me
2-Et
19a
19b
19c
19d
19e
19 f
19g
19h
19i
quant.
quant.
quant
90
the mechanism to be more complex since it has been shown
that acetate ions are crucial ligands to accelerate the rate-
limiting carbopalladation step in the Heck reaction.[23]
2-iPr
2-OMe
3-OMe
4-OMe
2-CF3
3-CF3
4-CF3
3-Me
90
Scope of the reaction with other bicyclic systems and aryl
halides: With mechanistic data in hand, we sought to study
the reactivity with numerous aryl halides and bicycle sys-
tems. Poor reactivity was expected with both electron poor
and electron rich aryl halides due to the instability of elec-
tron poor palladacycles (see Table 3) and high stability of
electron rich palladacycles (see Section on NMR analysis of
reductive elimination: Step 3). As expected, both electron
rich and poor aryl halides gave no reaction with the excep-
tion of 2-substituted aryl halides such as 18e and 18h
(Table 5). It seems that the observed reactivity of 18e and
18h are primarily the result of steric influence of the sub-
stituent, increasing the rate of the reductive elimination by
relief of steric strain. Attempts to increase the reaction tem-
perature for unreactive substrates resulted in decomposition
(Table 5, entries 6–7 and 9–11).
NR[c]
NR[c]
95
9
10
11
NR[c]
NR[c]
30[c.d]
18j
18k
19j
19k
[a] Reactions were conducted with PdACHTUNGRTENUNG(OAc)2 (10 mol%), PPh3
(22 mol%), Bu4NOH (0.6 mmol), 18 (0.3 mmol), norbornene (1.2 mmol),
toluene (1 mL), H2O (1 mL). [b] Isolated yield. [c] Reactions at 1008C re-
sulted in decomposition. [d] Isolated as two regioisomers.
Table 6. Scope of benzocyclobutene reaction with 2-pyrrole substituted
aryl iodide.[a,b]
We next examined a variety of 2-pyrrole-substituted sub-
strates, including electron-rich and -poor phenyl iodide 4. In
these cases, the benzocyclobutene could not be obtained, re-
gardless of electronic effects (Table 6, entries 4–8). Howev-
er, the reaction of electron-neutral aryl iodides 4a resulted
in high yield of the corresponding benzocyclobutene 15a
with excellent regioselectivity (Table 6, entry 1). The annula-
tion reaction can also be extended to other bicyclic systems
such as norbornadiene 20b or benzonorbornadiene 20c
(Table 6, entries 2–3).
Entry
4
20
Product
Yield [%][c]
1
2
3
4
5
6
7
8
4a
4a
4a
4b
4c
4d
4e
4 f
20a
20b
20c
20a
20a
20a
20a
20a
15a
15b
15c
15d
15e
15 f
15g
15h
95–100
80
50
[d,c]
–
NR[e]
NR[e]
NR
NR
[a] Reactions were conducted with PdACHTUNGRTENUNG(OAc)2 (10 mol%), PPh3
(22 mol%), Bu4NOH (0.6 mmol), 4 (0.3 mmol), 20 (1.2 mmol), toluene
(1 mL), H2O (1 mL). [b] Reactions with other bicyclic systems such as ox-
abicycle or azabicycle were not effective. [c] Isolated yield. [d] Reprodu-
cibility problems. [e] Reactions at 1008C resulted in pyrrole annulation.
Conclusion
In summary, we carried out detailed mechanistic investiga-
ꢀ
tions and a synthetic scope of Pd-catalyzed aryl C H func-
this formation is reversible at higher temperature (>1008C).
2) Higher temperatures (>1008C) are required for pyrrole
annulation, while benzocyclobutene formation can be selec-
tively formed at low temperature (808C). 3) The biphasic,
basic conditions can slow down the reversible palladacycle
by removing HX. This study also provides important fea-
tures about the palladacycle formation: 1) The palladacycle
only can be formed in the presence of water present in tolu-
ene.[16] 2) The palladacycle formation is reversible in the
presence of water at >1008C, but irreversible at 808C. Al-
though this study does not explain the regioselectivity ob-
tionalization of phenyl iodides with strained alkenes. In this
study, we have demonstrated that either benzocyclobutenes
or pyrrole annulation products can be selectively synthe-
sized from the reaction between 2-pyrrole phenyl iodide and
norbornene at 80 or 1008C, respectively, in the presence of
water and hydroxide ions. This mechanistic study provides
the following insight into the factors that control the selec-
tivity of both the pyrrole annulation and benzocyclobutene
reactions: 1) Ligand exchange between Iꢀ and HOꢀ is re-
sponsible for five-membered palladacycle formation, but
Chem. Eur. J. 2011, 17, 8175 – 8188
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8185

M. Lautens et al.
(with 74% D). M.p. 150–1608C (decomp); 1H NMR (400 MHz, CD2Cl2):
d = 7.88 (d, 1.26H; J=6.8 Hz), 7.55 (dd, 6H; J=11.6, 7.2 Hz), 7.41–7.48
(m, 3H), 7.37–7.29 (m, 9H), 3.21 (d, 1H; J=7.2 Hz), 2.60, (d, 1H; J=
9.6 Hz), 2.44 (d, 1H; J=2.0 Hz), 1.45 (brs, 1H), 1.40–1.38 (m, 1H), 1.25
(d, 1H; J=10.0 Hz), 1.18 (t, 1H; J=8.8 Hz), 0.94–0.86 (m, 2H),
0.32 ppm (m, 1H); 13C NMR (100 MHz, CD2Cl2): d = 135.0, 134.9, 134.8,
134.8, 134.7, 134.6, 131.9, 131.8, 131.5, 131.3, 130.8, 130.7, 130.6, 130.5,
130.4, 128.4, 128.3, 128.2, 128.2, 128.1, 128.0, 127.9, 127.9, 127.7, 127.6,
54.2, 43.0, 42.9, 40.7, 38.2, 36.9, 36.8, 29.7, 29.6, 29.5, 28.0, 15.0 ppm;
served in the original reaction conditions (MeCN and
Cs2CO3, Scheme 2), we believe that there are common fac-
tors controlling the regioselectivity in these two reaction
2ꢀ
conditions (MeCN/CO3
vs toluene/H2O/HOꢀ). Further
study towards coupling of the palladacycle and alkyl halides
or other oxidants is on-going and will be reported in due
course. Additionally, the reactivity of the new palladium
complex Pd-6 is on-going and will be reported in due
course.
31P NMR (161 MHz, CDCl3): d
= 35.5 ppm; ESI MS: m/z: calcd for
C31H29DPPd: 540.1177: found 540.1161 [MꢀI]+.
[Ph[D1]ACHTUNTGRNENUG(C6H10)PdACHTUTGNERNN(UGN PPh3)Cl] (Pd-3a[D]): Pd-3a[D1]-Cl (1 mmol) was
treated with KI (5 mmol) in CH2Cl2 (33.3 mLmmolꢀ1) and stirred over-
night at room temperature. The mixture was washed with water (3ꢃ
10 mL) and the organic layer dried over Na2SO4, and filtered. After the
solvent was evaporated under vacuum, the residue was treated with
ether, and the precipitates were collected by filtration. Yield: 75%
(78%D). M.p. 148–1558C (decomp); 1H NMR (200 MHz, CD2Cl2): d =
7.96 (m, 1.22), 7.63–7.64 (m, 6H), 7.56 (m, 3H), 7.44–7.35 (m, 9H), 3.30
(d, 1H; J=7.6 Hz), 2.73 (d, 1H; J=10.0 Hz), 2.55 (d, 1H; J=3.6 Hz),
1.76 (s, 1H), 1.55–1.30 (m, 3H), 1.06–0.88 (m, 2H; 0.47–0.31 ppm (m,
1H); 13C NMR (100 MHz, CD2Cl2): d = 137.1, 137.0, 136.9, 136.9, 136.8,
136.7, 136.6, 136.5, 133.9, 133.5, 133.4, 132.7, 132.5, 132.4, 132.3, 132.1,
131.4, 130.8, 130.8, 130.7, 130.7, 129.9, 129.8, 129.8, 129.5, 129.5, 129.4,
129.4, 129.3, 56.5, 56.4, 45.8, 45.7, 45.0, 44.9, 42.3, 39.6, 31.7, 31.6, 31.5,
29.6 ppm; 31P NMR (81 MHz, CD2Cl2): d = 38.9; ESI MS: m/z: calcd for
C31H29DPPd: 540.1177: found 540.1153 [M]+.
Experimental Section
General method for synthesis of benzocyclobutenes 15 and 19: Aryl
iodide 4 or 18 (0.15 mmol), PdACHTNUGTRENUNG(OAc)2 (10 mol%), and PPh3 (22 mol%)
were dissolved in toluene (0.5 mL) in a tapered microwave vial. To this
mixture was added norbornene or bicyclic derivatives (0.6 mmol) in one
portion. Then, Bu4NOH (0.2 mL, 40 wt% in H2O, 2 equiv) and water
(0.3 mL) were immediately added via syringe, sealed, then heated to
808C for 12 h in a pre-heated oil bath. Once cooled, the mixture was di-
luted with ether (0.5 mL) and water (0.5 mL) and extracted with diethyl
ether (3ꢃ1 mL). The combined organic layers were then dried (MgSO4),
filtered, and the solvent removed in vacuo to yield the crude pyrrole. The
crude material was purified by column chromatography.
General procedure for synthesis of palladacycles (Pd-5): A sealable
round-bottom flask containing norbornyl palladium complex, Pd-3
(0.5 mmol), KOPh (1.5 mmol), and PPh3 (0.9 mmol) was flushed with
argon for at least 10 min. Then, CH2Cl2 (67 mLmmolꢀ1) was added,
sealed, and stirred for 1 h. After this time, the reaction was quenched by
adding H2O (12 mLmmolꢀ1) and the organic layer was washed with
water (3ꢃ10 mL). The combined extracts were dried over Na2SO4, fil-
tered, and solvent was removed under reduced pressure. The residue was
redissolved in diethyl ether (7 mL). The product slowly precipitated out
by scratching at cold temperature. The product was collected by filtration
and washed with small amount of cold ether (2ꢃ3 mL) and cold acetone
(2ꢃ2 mL). Recrystallized from CH2Cl2 + MeOH if necessary.
Annulation product 15a: Purified by flash column chromatography
(hexane/dichloromethane 6:1) yielding the title compound as a white
solid (34.9 mg, 99%). M.p. 63–648C; 1H NMR (400 MHz, C6D6): d =
7.27–7.20 (m, 4H), 6.83–6.81 (m, 1H), 6.32 (t, 2H; J=2.19 Hz), 3.37 (d,
1H; J=3.86 Hz), 3.21 (d, 1H; J=3.84 Hz), 2.41 (dd, 1H; J=3.04 &
1.35 Hz), 2.31 (dd, 1H; J=3.00 & 1.32 Hz), 1.65–1.59 (m, 2H), 1.25–1.20
(m, 2H), 1.03–0.94 ppm (m, 2H); 13C NMR (100 MHz, C6D6): d = 148.3,
134.2, 133.9, 128.9, 118.8, 118.4, 116.2, 110.2, 50.7, 50.2, 36.6, 36.2, 32.1,
27.8, 27.7 pppm; IR (CHCl3): vmax = 2951 (s), 2869 (w), 1600 (m), 1500
(s), 1338 (m), 1078 (m), 773 (m), 723 cmꢀ1 (m); ESI MS: m/z: calcd for
C17H18N 236.1443: found 236.1427 [M+H]+.
General procedure for synthesis of norbornyl palladium complex: In a
ACHUTNGRENUN[G {C6H4}C6H10PdCAHTUNGTERNN(UGN PPh3)2] (Pd-5a): Synthesized according to the general
sealable round-bottom flask containing [PdACHTNURTGNE(UNG PPh3)4] (1 mmol) and THF
procedure. Isolated as off-white solid. Yield: 70%. M.p. 105–1108C
(decomp); 1H NMR (400 MHz, CD2Cl2): d = 7.73 (t, 6H; J=8.8 Hz),
7.53 (t, 6H; J=8.8 Hz), 7.46 (dt, 3H; J=7.2, 1.2 Hz), 7.44–7.30 (m, 9H),
7.23 (t, 6H; J=6.8 Hz), 7.10 (d, 1H; J=7.6 Hz), 6.83 (m, 2H), 6.32 (t,
1H; J=7.6 Hz), 3.18 (s, 1H), 2.90 (dd, 1H; J=14.8, 7.2 Hz), 2.43 (s, 1H),
2.27 (t, 2H; J=8.8 Hz), 1.47 (m, 1H), 1.12 (m, 3H), 0.03 ppm (m, 1H);
13P NMR (100 MHz, CD2Cl2): d = 170.7, 170.6, 141.7, 141.6, 141.5, 141.4,
135.2, 135.1, 134.9, 134.8, 134.7, 134.5, 134.2, 133.9, 129.5, 129.4, 129.1,
127.9, 127.8, 127.4, 127.3, 122.7, 122.2, 122.1, 122.0, 121.9, 121.8, 121.7,
66.9, 66.8, 66.1, 66.0, 64.0, 63.9, 48.0, 44.5, 44.4, 34.8, 32.1, 32.0, 32.0, 31.9,
30.5, 28.5 ppm; 31P NMR (161 MHz, CD2Cl2): d = 27.2 (d, J=17.8 Hz),
22.3 ppm (d, J=17.8 Hz); ESI MS: m/z: calcd for C31H30PPd: 539.1114:
found 539.1138 [M+HꢀPPh3]+.
(20 mL) was added the aryl halide (10 mmol) via syringe. After flushing
with argon for at least 10 min, norbornene (7 mmol) was added as a solid
in one portion, sealed, and heated to 608C for 8 h. After cooling down to
room temperature, solvent was removed under reduced pressure and the
residue was redissolved in diethyl ether (7 mL). The product slowly pre-
cipitated out by scratching. The product was collected by filtration and
washed with small amount of ether (2ꢃ3 mL). Recrystallized from
CH2Cl2 + MeOH if necessary.
[PhACHTUNGTRENNUNG(C6H10)PdACHTUNGTRENNUNG(PPh3)I] (Pd-3a): Synthesized according to the general
procedure using phenyl iodide. Isolated as orange solid. Yield: 70%
(45% after recrystallization); m.p. 145–1508C (decomp); 1H NMR
(400 MHz, CDCl3): d = 7.95 (brs, 2H), 7.70 (t, 6H; J=7.6 Hz), 7.56 (m,
3H), 7.44–7.35 (m, 9H), 3.31 (d, 1H; J=7.6 Hz), 2.70 (d, 1H; J=10 Hz),
2.55 (d. 1H; J=2.4 Hz), 1.77 (s, 1H), 1.51–1.43 (m, 2H), 1.33 (d, 1H; J=
10 Hz), 1.03–0.90 (m, 2H), 0.43–0.33 ppm (m, 1H); 13C NMR (100 MHz,
CDCl3): d = 135.2, 135.1, 135.0, 134.9, 131.9, 131.8, 131.4, 131.0, 130.5,
130.5, 130.1, 128.9 (br), 128.1, 128.0, 127.6 (m), 103.0, 102.9, 54.6, 54.5,
43.8, 43.7, 42.8, 42.6, 40.3, 37.8, 29.9, 29.8, 27.8 ppm; 31P NMR (161 MHz,
CDCl3): d = 38.8; ESI MS: m/z: calcd for C31H30PPd: 539.1114: found
539.1094 [MꢀI]+.
General procedure for synthesis of new palladacycle (Pd-6): A sealable
round-bottom flask containing aryl halide
4 (1.2 mmol) and [PdCl2-
AHCTUNGTRENNUNG
zene (14 mL) and aq. KOH (2.86 g KOH in 3 mL H2O) were injected in
sequence and refluxed for 3 h. After cooling to room temperature, the or-
ganic layer was washed with water (3ꢃ5 mL), and the extracts were
dried over Na2SO4. After filtration, the solvent was removed under re-
duced pressure and the residue was redissolved in acetone (7 mL). The
product slowly precipitated out by scratching. The product was collected
by filtration and washed with small amount of cold acetone (2ꢃ3 mL)
and cold acetone (2ꢃ2 mL). M.p. 150–1558C (decomp); 1H NMR
(400 MHz, CDCl3): d = 7.28 (m, 6H), 7.17 (t, 3H; J=7.2 Hz), 7.05 (d,
1H; J=3.2 Hz), 7.00 (dt, 6H; J=7.6, 1.6 Hz), 6.69 (t, 1H; J=7.6 Hz),
6.5–6.45 (m, 2H), 6.12 (dt, 1H; J=7.6, 1.2 Hz), 5.90 (m, 1H), 5.34ppm
[Ph[D1]ACHTUNGTRENNUNG(C6H10)PdACHTUNGTRENNUNG(PPh3)Cl] (Pd-3a[D]-Cl): Synthesized according to the
Echavarrenꢂs procedure.[24] To a vial containing AcCl (5 mmol) was
added deuterated methanol (5 mmol) dropwise at 08C and stirred at this
temperature for 30 min. To a 100 mL round-bottom flask containing the
palladacycle (1 mmol) in CH2Cl2 (34 mL per Pd mmol) was added D-Cl
solution (4.14 mL per Pd mmol) and stirred at ambient temperature for
3.5 h. The solvent was removed by vacuum and dilute with diethyl ether
(7 mL). The precipitates formed were collected by filtration. Yield: 89%
(q, 1H; J=2.8 Hz); 13C NMR (100 MHz, CDCl3): d
= 206.9, 150.7,
8186
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8175 – 8188

ꢀ
Aryl C H Bond Functionalization
FULL PAPER
144.2, 144.1, 140.3, 139.2, 139.1, 139.0, 135.5, 135.4, 132.1, 131.7, 129.4,
129.4, 127.6, 127.6, 127.5, 124.0, 123.8, 123.7, 121.4, 121.4, 110.5, 107.8,
107.8, 103.9, 103.9 ppm; 31P NMR (121 MHz, CDCl3): d = 33.6 ppm; ESI
MS: m/z: calcd for C28H23NPPd: 510.0597.: found 510.0582 [M/2, mono-
meric].
[8] a) For intermolecular competition between Heck reaction and nor-
bornene carbopalladation, see a) F. Jafarpour, M. Lautens, Org. Lett.
2006, 8, 3601. For intramolecular competition between Heck reac-
tion and norbornene carbopalladation, see b) A. Martins, U. Mar-
quardt, N. Kasravi, D. Alberico, M. Lautens, J. Org. Chem. 2006, 71,
4937.
[9] a) I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000, 100, 3009;
b) C. Amatore, A. Jutand, J. Organomet. Chem. 1999, 576, 254;
c) W. Cabri, I. Candiani, S. DeBarnardis, F. Francalanci, S. Penco, J.
Org. Chem. 1991, 56, 5796.
Acknowledgements
[10] For observation of PdII alkene complexes, see a) P. S. Hanley, D.
Markovic, J. F. Hartwig, J. Am. Chem. Soc. 2010, 132, 6302; b) C. S.
Shultz, J. Ledford, J. M. DeSimone, M. Brookhart, J. Am. Chem.
Soc. 2000, 122, 6351; c) F. C. Rix, M. Brookhart, J. Am. Chem. Soc.
1995, 117, 1137.
We thank the Natural Science and Engineering Research Council of
Canada (NSERC), the Merck Frosst Centre for Therapeutic Research,
and the University of Toronto for funding. We thank Dr. Anny Jutand
(CNRS) and Prof. Mark Taylor (University of Toronto) for helpful sug-
gestions and comments. We also thank Dr. Alan Lough for X-ray crystal-
lography analysis and Dr. Tim Burrow for variable temperature NMR
studies. P.T. thanks NSERC for postgraduate scholarship (CGSM and
CGSD).
[11] C.-C. Huang, J.-P. Duan, M.-Y. Wu, F.-L. Liao, S.-L. Wang, C.-H.
Cheng, Organometallics 1998, 17, 676.
[12] a) A. Segnitz, P. M. Bailev, P. M. Maitlis, J. Chem. Soc. Chem.
Commun. 1973, 698; b) M. F. Rettig, R. M. Wing, G. R. Wiger, J.
Am. Chem. Soc. 1981, 103, 2980.
[13] For alkoxo-palladium(II) complexes, see a) M. Tanabe, K. Mutou,
N. Mintcheva, K. Osakada, Organometallics 2008, 27, 519; b) E.
Drent, J. A. M. van Broekhoven, M. J. Doyle, J. Organomet. Chem.
1991, 417, 235; c) E. Drent, P. Arnoldy, P. H. M. Budzelaar, J. Orga-
nomet. Chem. 1994, 475, 57; d) Y.-J. Kim, K. Osakada, A. Takenaka,
A. Yamamoto, J. Am. Chem. Soc. 1990, 112, 1096; e) G. M. Kap-
teijn, A. Dervisi, D. M. Grove, H. Kooijman, M. T. Lakin, A. L.
Spek, G. van Koten, J. Am. Chem. Soc. 1995, 117, 10939.
[14] The hydroxo–palladium complexes play an important role as syn-
thetic precursors. For examples of dimeric hydroxo–palladium com-
plexes, see: a) S. Kannan, A. J. James, P. R. Sharp, Polyhedron 2000,
19, 155; b) R. Schnebeck, E. Freisinger, B. Lippert, Eur. J. Inorg.
Chem. 2000, 1193; c) G. Pieri, M. Pasquali, P. Leoni, U. Englert, J.
Organomet. Chem. 1995, 491, 27; d) V. V. Grushin, H. Alper, Orga-
nometallics 1993, 12, 1890; e) M. S. Driver, J. F. Hartwig, Organome-
tallics 1997, 16, 5706; f) N. J. Ruiz, J. Cutillas, G. Samperdo, J. A.
Lꢄpez, M. Hermoso, Martꢅnez-Ripoll, J. Organomet. Chem. 1996,
526, 67; g) V. F. Kuznetsov, C. Bensimon, G. A. Facey, V. V. Grushin,
H. Alper, Organometallics 1997, 16, 97. For examples of monomeric
hydroxo-palladium complexes, see h) E. Alvaro, J. F. Hartwig, J.
Am. Chem. Soc. 2009, 131, 7858; i) M. Akita, T. Miyaji, N. Muroga,
C. Mock-Knoblauch, W. Adam, S. Hikichi, Y. Moro-oka, Inorg.
Chem. 2000, 39, 2096; j) J. Cꢆmpora, P. Palma, D. del Rio, E. Alvar-
ez, Organometallics 2004, 23, 1652.
[1] For a recent review, see: a) D. Alberico, M. E. Scott, M. Lautens,
Chem. Rev. 2007, 107, 174; b) P. Thansandote, M. Lautens, Chem.
Eur. J. 2009, 15, 5874; c) G. P. Chiusoli, M. Catellani, M. Costa, E.
Motti, N. Della Ca’, G. Maestri, Coord. Chem. Rev. 2010, 254, 456;
d) F. Bellina, R. Rossi, Tetrahedron 2009, 65, 10269.
[2] a) M. Catellani, F. Frignani, A. Rangoni, Angew. Chem. 1997, 109,
142; Angew. Chem. Int. Ed. Engl. 1997, 36, 119; b) M. Lautens, S.
Piguel, Angew. Chem. 2000, 112, 1087; Angew. Chem. Int. Ed. 2000,
39, 1045; c) M. Lautens, J. F. Paquin, S. Piguel, J. Org. Chem. 2002,
67, 3972; d) D. Alberico, J. F. Paquin, M. Lautens, Tetrahedron 2005,
61, 6283; e) C. Blaszykowski, E. Aktoudianakis, D. Alberico, C.
Bressy, D. G. Hulcoop, F. Jafarpour, A. Joushaghani, B. Laleu, M.
Lautens, J. Org. Chem. 2008, 73, 1888; f) T. Wilhelm, M. Lautens,
Org. Lett. 2005, 7, 4053; g) A. Martins, M. Lautens, Org. Lett. 2008,
10, 5095; h) A. Rudolph, N. Rackelmann, M. Lautens, Angew.
Chem. 2007, 119, 1507; Angew. Chem. Int. Ed. 2007, 46, 1485; i) A.
Rudolph, N. Rackelmann, M. O. Turcotte-Savard, M. Lautens, J.
Org. Chem. 2009, 74, 289; j) B. Mariampillai, J. Alliot, M. Li, M.
Lautens, J. Am. Chem. Soc. 2007, 129, 15372; k) P. Thansandote, M.
Raemy, A. Rudolph, M. Lautens, Org. Lett. 2007, 9, 5255; l) K. M.
Gericke, D. I. Chai, M. Lautens, Tetrahedron 2008, 64, 6002;
m) K. M. Gericke, D. I. Chai, N. Bieler, M. Lautens, Angew. Chem.
2009, 121, 1475; Angew. Chem. Int. Ed. 2009, 48, 1447; n) P. Than-
sandote, E. Chong, K.-O. Feldmann, M. Lautens, J. Org. Chem.
2010, 75, 3495; o) D. A. Candito, M. Lautens, Angew. Chem. 2009,
121, 6841; Angew. Chem. Int. Ed. 2009, 48, 6713.
[15] For similar kinetic isotope effect in aromatic palladation reactions,
see E. J. Hennessy, S. L. Buchwald, J. Am. Chem. Soc. 2003, 125,
12084 (kH/kD =4 for intramolecular KIE; kH/kD =1 for intermolecu-
lar KIE).
[16] We believe that water is not responsible for the palladacycle forma-
tion, but a small amount of hydroxide generated in-situ from water
may undergo a ligand exchange with a halide on the palladium com-
plex, which then undergoes the palladacycle formation.
[3] a) M. Catellani, Top. Organomet. Chem. 2005, 14, 174; b) M. Catel-
lani, L. Ferioli, Synthesis 1996, 769.
[4] For annulation onto an alkene, see: a) R. Grigg, P. Kennewell, A.
Teasdale, V. Sridharan, Tetrahedron Lett. 1993, 34, 153. For annula-
tion onto a nitrile, see: b) A. A. Pletnev, Q. Tian, R. C. Larock, J.
Org. Chem. 2002, 67, 9276. For annulation onto phenols and ani-
lines, see: c) M. Catellani, A. Del Rio, Russ. Chem. Bull. 1998, 47,
928; d) K. Saito, O. Katsuhiko, M. Sano, S. Kiso, T. Takeda, Hetero-
cycles 2002, 57, 1781; e) M. Lautens, J.-F. Paquin, S. Piguel, M. Dahl-
mann, J. Org. Chem. 2001, 66, 8127; f) P. Thansandote, D. G. Hul-
coop, M. Langer, M. Lautens, J. Org. Chem. 2009, 74, 1673.
ꢀ
[17] For palladium-catalyzed oxidative C H insertion, see T. Okazawa,
T. Satoh, M. Miura, M. Nomura, J. Am. Chem. Soc. 2002, 124, 5286.
ꢀ
[18] For palladium-catalyzed electrophilic C H functionalization mecha-
nism, see a) S. Pivsa-Art, T. Satoh, Y. Kawamura, M. Miura, M.
Nomura, Bull. Chem. Soc. Jpn. 1998, 71, 467; b) B. S. Lane, M. A.
Brown, D. Sames, J. Am. Chem. Soc. 2005, 127, 8050; c) C.-H. Park,
V. Ryabova, I. V. Seregin, A. W. Sromek, V. Gevorgyan, Org. Lett.
2004, 6, 1159.
[5] D. G. Hulcoop, M. Lautens, Org. Lett. 2007, 9, 1761.
[6] a) M. Catellani, G. P. Chiusoli, J. Organomet. Chem. 1992, 425, 151;
b) M. Catellani, C. Mealli, E. Motti, P. Paoli, E. Perez-Carreno, P. S.
Pregosin, J. Am. Chem. Soc. 2002, 124, 4336; c) C. Amatore, M. Cat-
ellani, S. Deledda, A. Jutand, E. Motti, Organometallics 2008, 27,
4549; d) C.-S. Li, C.-H. Cheng, F.-L. Liao, S.-L. Wang, J. Chem. Soc.
Chem. Commun. 1991, 710; e) C.-S. Li, D.-C. Jou, C.-H. Cheng, Or-
ganometallics 1993, 12, 3945; f) C.-H. Liu, C-S. Li, C.-H. Cheng, Or-
ganometallics 1994, 13, 18.
ꢀ
[19] For palladium-catalyzed Heck-like C H functionalization process,
see B. Glover, K. A. Harvey, B. Liu, M. J. Sharp, M. Tymoschenko,
Org. Lett. 2003, 5, 301.
[20] For palladium-catalyzed CMD C H functionalization process, see
ꢀ
a) D. Garcꢅa-Cuadrado, P. de Mendoza, A. A. C. Braga, F. Maseras,
A. M. Echavarren, J. Am. Chem. Soc. 2007, 129, 6880; b) D. L.
Davies, S. M. A. Donald, O. Al-Duaij, S. A. Macgregor, M. Polleth,
J. Am. Chem. Soc. 2006, 128, 4210; c) S. I. Gorelsky, D. Lapointe, K.
Fagnou, J. Am. Chem. Soc. 2008, 130, 10848; d) B. Biswas, M. Sugi-
[7] F.-G. Zhao, B. M. Bhanage, M. Shirai, M. Arai, Stud. Surf. Sci. Catal.
1999, 122, 427.
Chem. Eur. J. 2011, 17, 8175 – 8188
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8187

M. Lautens et al.
moto, S. Sakaki, Organometallics 2000, 19, 3895; e) D. L. Davies,
S. M. A. Donald, S. A. Macgregor, J. Am. Chem. Soc. 2005, 127,
13754; f) D. Garcꢅa-Cuadrado, A. A. C. Braga, F. Maseras, A. M.
Echavarren, J. Am. Chem. Soc. 2006, 128, 1066; g) M. Lafrance,
C. N. Rowley, T. K. Woo, K. Fagnou, J. Am. Chem. Soc. 2006, 128,
8754; h) I. Alonso, M. Alcami, P. Mauleon, J. C. Carretero, Chem.
Eur. J. 2006, 12, 4576; i) S. Pascual, P. de Mendoza, A. A. C. Braga,
F. Maseras, A. M. Echavarren, Tetrahedron 2008, 64, 6021.
Fauvarque, F. Pflꢇger, M. Troupel, J. Organomet. Chem. 1981, 208,
419.
[23] a) C. Amatore, E. Carrꢈ, A. Jutand, Y. Medjour, Organometallics
2002, 21, 4540; b) A. Jutand, Pure Appl. Chem. 2004, 76, 565; c) C.
Amatore, E. Carrꢈ, A. Jutand, M. A. M’Barki, G. Meyer, Organo-
metallics 1995, 14, 5605; d) C. Amatore, E. Carrꢈ, A. Jutand, Acta.
Chem. Scand. 1998, 52, 100; e) S. Kozuch, S. Shaik, A. Jutand, C.
Amatore, Chem. Eur. J. 2004, 10, 3072; f) C. Amatore, A. Jutand,
M. M’Barki, Organometallics 1992, 11, 3009; g) C. Amatore, E.
Carrꢈ, A. Jutand, M. A. M’Barki, Organometallics 1995, 14, 1818.
[24] B. Martꢅn-Matute, C. Mateo, D. J. Cardenas, A. M. Echavarren,
Chem. Eur. J. 2001, 7, 2341.
[21] a) P. H. M. Budzelaar, P. W. N. M. Leeuwen, C. F. Roobeek, Organo-
metallics 1992, 11, 23; b) F. K. Friedlein, K. Kromm, F. Hampel,
J. A. Gladysz, Chem. Eur. J. 2006, 12, 5267; c) X. Zhao, C. Yeung,
V. M. Dong, J. Am. Chem. Soc. 2010, 132, 5837.
[22] a) P. Fitton, M. P. Johnson, J. E. McKeon, Chem. Commun. 1968, 6;
b) P. Fitton, E. A. Rick, J. Organomet. Chem. 1971, 28, 287; c) B. E.
Mann, A. J. Musco, J. Chem. Soc. Dalton Trans. 1975, 1673; d) J. F.
Received: January 20, 2011
Published online: June 3, 2011
8188
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8175 – 8188