Mechanisms of the PtCl2-catalyzed intramolecular cyclization of o -isopropyl-substituted aryl alkynes for the synthesis of indenes and comparison of three sp3 C-H bond activation modes

Article abstract of DOI:10.1021/jo500839c

Chatani and He respectively reported an efficient way to synthesize indenes through PtCl₂ catalyzed sp³ C-H bond activation. Interestingly, the R group (R = H or Br) in the alkyne moiety of the substrates in Chatanis experiments migrates to the C3 position in indenes, whereas the R group (R = Ar) stays in the original C2 position of final indenes in Hes experiments. DFT calculations indicated that there are two competing pathways a and c for the cyclization reaction. Pathway a involves [1,2]-R migration, [1,5]-H shift, and 4π-electrocyclization, giving the indenes with the R group at the C3 position. Pathway c takes place through irreversible [1,5]-H shift/cyclization and [1,2]-H shift, generating indenes with the R group at the C2 position. DFT calculations found that, when R = H or Br, pathway a is favored. When R = alkyl group, the [1,2]-R migration is difficult and pathway c is preferred. When R = Ar, DFT calculations predicted and experiments verified that both pathways a and c occur to give two indene products. Comparison of different models of sp³ C-H activations has been presented to guide further understanding and prediction of new C-H bond activations.

Full text of DOI:10.1021/jo500839c

Article
pubs.acs.org/joc
Mechanisms of the PtCl2‑Catalyzed Intramolecular Cyclization of
o‑Isopropyl-Substituted Aryl Alkynes for the Synthesis of Indenes
and Comparison of Three sp3 C−H Bond Activation Modes
Yi Wang,†,§ Wei Liao,†,§ Genping Huang,‡ Yuanzhi Xia,‡ and Zhi-Xiang Yu*,†
†Beijing National Laboratory for Molecular Sciences (BNLMS), Key Laboratory of Bioorganic Chemistry and Molecular Engineering
of Ministry of Education, College of Chemistry, Peking University, Beijing 100871, China
‡College of Chemistry and Materials Engineering, Wenzhou University, Wenzhou, Zhejiang Province 325035, China
S
* Supporting Information
ABSTRACT: Chatani and He respectively reported an efficient
way to synthesize indenes through PtCl2 catalyzed sp3 C−H bond
activation. Interestingly, the R group (R = H or Br) in the alkyne
moiety of the substrates in Chatani’s experiments migrates to the
C3 position in indenes, whereas the R group (R = Ar) stays in the
original C2 position of final indenes in He’s experiments. DFT
calculations indicated that there are two competing pathways a
and c for the cyclization reaction. Pathway a involves [1,2]-R
migration, [1,5]-H shift, and 4π-electrocyclization, giving the
indenes with the R group at the C3 position. Pathway c takes place
through irreversible [1,5]-H shift/cyclization and [1,2]-H shift,
generating indenes with the R group at the C2 position. DFT
calculations found that, when R = H or Br, pathway a is favored.
When R = alkyl group, the [1,2]-R migration is difficult and pathway c is preferred. When R = Ar, DFT calculations predicted and
experiments verified that both pathways a and c occur to give two indene products. Comparison of different models of sp3 C−H
activations has been presented to guide further understanding and prediction of new C−H bond activations.
Scheme 1. Cyclization of Terminal Alkynes Reported by
Yamamoto4 and Liu5
INTRODUCTION
■
Designing, discovering, and developing new C−H bond
functionalization reactions is one of the most intensively
investigated research fields in today’s science of synthesis.1
Understanding the mechanisms of the reported C−H bond
functionalization reactions is very critical for advancing the
related field, considering that the mechanistic insights obtained
experimentally and/or computationally can provide useful hints
and guidance for optimizing and developing new reactions and
catalysts.2 Unfortunately, even though many elegant C−H
functionalization reactions have been developed, only a small
number of these reactions have been investigated mechanisti-
cally.3 For example, recently, Yamamoto,4 Liu,5 Chatani,6 and
He7 made great contributions in the field of the sp3 C−H
functionalization (Schemes 1 and 2),8 but mechanistic
understandings of how these reactions take place and what
factors affect their different reactivity and regiochemistry have
not been well studied.9 Herein, we report our DFT and
experimental investigations of the mechanisms and regiochem-
istry of these PtCl2-catalyzed C−H functionalization reactions.
In 2006, Yamamoto reported the PtBr2-catalyzed cyclization
reaction of 1-ethynyl-2-(1-alkoxybut-3-enyl)benzenes, giving
functionalized indenes (Scheme 1).4 However, simple alkyl
substituted compounds without methoxy groups were inappli-
cable even at high temperature (120 °C). Liu found that the
similar reaction can also be catalyzed by Ru catalyst (Scheme
1).5 Recently, Chatani’s group and He’s group independently
made significant advances in this field (reactions 1−3, Scheme
2). Chatani and co-workers6 found that PtCl2, PtCl4, or
[RuCl2(CO)3]2 can serve as the general catalyst for the
cyclization of 1-alkyl-2-ethynylbenzenes under relatively mild
conditions (30−80 °C), leading to benzylic C−H bond
functionalization products. It was demonstrated for the first
time that simple tertiary C−H bonds can participate in this type
of catalytic cyclization reaction to form all-carbon quaternary
Received: April 14, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
reached if both PtCl2 and CuBr were used as catalysts (see their
original report for details).7
Scheme 2. Cyclization Reactions Reported by Chatani
Group6 and He Group7
To our surprise, the proposed mechanisms for reactions 1−3
by Chatani and He are quite different (pathways a and b,
Scheme 3). Chatani6 proposed that, for substrate S with R = H
or Br, the cyclization reaction starts from coordination of PtCl2
to the triple bond in S, leading to the formation of Pt−π-alkyne
complex A, which can be viewed as an equilibrium structure of
zwitterionic B (We name this equilibrium form as an endo vinyl
cation, which is opposite to another equilibrium form, the exo
vinyl cation H. Also B and H can be regarded as carbene
species.).10 Subsequently, the hydrogen or bromine at the
terminal alkyne migrates to the adjacent internal β-carbon to
form a vinyldiene platinum intermediate C,11 which activates
the sp3 C−H bond at the benzylic carbon. A [1,5]-H shift
transforms the platinum vinylidene C into a Pt−carbene
complex D. Complex D can be viewed as a metallohexatriene,
which then undergoes electrocyclization to form metallacycle E.
Subsequent reductive elimination from E gives the final product
P-I. Because of the existence of [1,2]-R migration, the migrated
R (H or Br) was incorporated at the C3 position in the final
indene products.
In contrast, He7 proposed a mechanism without [1,2]-R
shifts for substrates with R = Ar and alkyl groups (pathway b).
In pathway b, complex B, an endo vinyl cation, undergoes a
[1,4]-H shift to give Pt-carbene complex F, which then
furnishes product P-II through cyclization and reductive
elimination. In P-II, the R group is incorporated at the C2
position of the indene product.
We were very curious to know why different substrates have
different reaction pathways. In addition, we proposed that there
is another possible pathway, pathway c involving [1,5]-H shift/
cyclization and [1,2]-R migration (H or R migration), to
account for the experimental results from both Chatani and He
centers. They carried out some preliminary mechanistic
investigations using deuterated alkyne 1 and Br-substituted
alkyne 2 as the substrates and found that the deuterium atom
and Br atom were selectively incorporated at the C3 position
exclusively in the indene products 1-P and 2-P, respectively
(reactions 1 and 2). He and co-workers later on found that the
substituent in the alkyne part can be an aryl group, but the aryl
group was incorporated at the C2 position of the final products
(Scheme 2). This reaction’s yield was 52%, but 82% it can be
Scheme 3. Three Possible Pathways for PtCl2-Catalyzed Cyclization of Alkynes Proposed by Chatani (Pathway a),6 He
(Pathway b),7 and Us (Pathway c)
B
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
groups. In pathway c, the PtCl2−alkyne complex can be viewed
as an exo vinyl cation species H, which can undergo [1,5]-H
shift to give Complex I. Complex I then undergoes cyclization
and [1,2]-H shift or [1,2]-R shift, generating indene products
P-II or P-I, respectively.
Among the three possible pathways, three modes of C−H
activations were involved from complexes C, B, and H,
respectively (see also Scheme 4 in the later discussion). The
compared so that this information can be helpful for the future
design of C−H activation reactions.
In the present paper, we report our DFT study of the energy
surfaces of the proposed pathways a−c for substrates with R =
H, Br, Ph, and Me. Previously, Zhao and co-workers discussed
several possible pathways of the indene formation from
substrate with R = H.9 We aim here to answer which pathway,
for a specific substrate, will be favored and the reason for this
preference. We also tried to analyze the detailed potential
energy surfaces for all these substrates to get the kinetic and
thermodynamic data of these reactions and the structures of
key transition states and intermediates. When calculation
results were different from experiments, new experimental
tests were carried out to support or disprove calculation results.
In addition, we designed and studied reaction 4 (Scheme 2)
computational and experimentally to understand how a
substrate with an alkyl group in its alkyne part undergoes the
C−H activation reaction using PtCl2 as the catalyst.
Scheme 4. DFT Computed Activation Free Energies (kcal/
mol) for [1,2]-R Migration and C−H Activations in the Gas
Phase
COMPUTATIONAL METHODS
■
All calculations were performed with the Gaussian 09
program.12 Density functional theory (DFT)13 calculations
using the B3LYP functional14 were used to locate all the
stationary points involved. The 6-31G(d) basis set15 was
applied for all elements except for Pt, for which the
LANL2DZ16 basis set and pseudopotential were used. This
approach has been successfully applied to study structures and
reaction mechanisms for reactions of PtCl2 complexes and
other Pt-catalyzed cycloadditions.17 Frequency calculations at
the same level were performed to confirm each stationary point
to be either a minimum or a transition structure. Intrinsic
reaction coordinate (IRC)18 calculations were performed to
confirm the connection of each transition state to its
corresponding reactant and product. The single point energies
were calculated by the M0619 method using the same basis set
to include the contribution of dispersion energies (we found
that in the present system, B3LYP and M06 gave very similar
C−H bond is activated through [1,5]-H shift reaction from C
in pathway a and H in pathway c, while in pathway b, the C−H
bond is activated through the [1,4]-H shift from B. The
activation free energies of three modes will be calculated and
Figure 1. DFT computed energy surface of pathway a for the PtCl2-catalyzed cyclization of 1-ethynyl-2-isopropylbenzene 1.
C
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 2. DFT optimized key structures for the PtCl2-catalyzed cyclization of 1-ethynyl-2-isopropylbenzene 1 in pathway a (distances in Å).
energies, but M06 is better to compare the different competing
pathways, see the Supporting Information for details).
Solvation energies in toluene were evaluated with M06 method
by a self-consistent reaction field (SCRF) using the CPCM
model,20 where UAKS radii21 were used. Solvation calculations
were carried out on the gas-phase optimized structures. Unless
specifically mentioned, all discussed relative energies in this
paper are referred to ΔG298 K (the relative gas phase free energy
at 298 K). The relative enthalpies (ΔH298 K) in the gas phase,
and the relative Gibbs free energies in toluene solution (ΔGsol),
both at 298 K, were also given. We must point here that
geometry optimization and energy calculations in solvent using
the CPCM model did not change our understanding of the
investigated reactions (see discussion in the Supporting
Information). All figures of structures were prepared using
CYLView.22
1-IN1 can be regarded as a zwitterionic species with negative
charge at the PtCl2 part and the positive at the C3 atom. The
cationic C3 atom can be envisioned as a vinyl cation. NBO
analysis23 shows that the charge populations on C2 and C3 in
free reactant 1 are −0.21 e and −0.04 e, respectively, while
these are changed to −0.27 and 0.12 e in 1-IN1. NBO analysis
also shows that Pt−C3 is a single bond, while Pt−C2 has some
double-bond character, as demonstrated by the bond orders of
Pt−C3 (0.99) and Pt−C2 (1.4). Thus, the Pt−C2 bond can
also be described as a platinum carbene species. The favorable
generation of the unsymmetrical complex 1-IN1 could be
comprehended by the fact that the positive charges on C3 atom
can be better stabilized by the aryl group. Thus, the triple bond
in C3 position in 1-IN1 has attribute of a vinyl cation.24
[1,2]-H Shift to Generate Vinylidene Complex. Complex 1-
IN1 then undergoes a [1,2]-H shift25 process to give vinylidene
platinum intermediate 1-a-IN2. This step requires an activation
free energy of 18.6 kcal/mol in the gas phase and is endergonic
by only 1.5 kcal/mol. IRC calculations unambiguously
confirmed the connection of 1-a-TS1 to its corresponding
reactant and product. In the [1,2]-H shift transition state 1-a-
TS1, the C2−Pt bond is nearly formed while the hydrogen is
transferring from C2 to C3 with distances of C2−H2 and C3−
H2 of 1.24 and 1.37 Å, respectively. Different groups have
different barriers for the [1,2]-R migration and this difference is
one reason for the different pathways (Scheme 4. The other
migrations will be discussed later on).
RESULTS AND DISCUSSION
■
1. Mechanism of the Cyclization of 1-Ethynyl-2-
isopropylbenzene 1 (Substrate S with R = H, Scheme
3). DFT calculations revealed that pathway a is the most
favored for the cyclization of 1-ethynyl-2-isopropylbenzene 1.
In this part, we will discuss our understanding of the favored
pathway a in detail (relative energies are given in Figure 1,
while some key computed structures are given in Figure 2).
The disfavored pathways b and c of this reaction will be
discussed briefly later on.
[1,5]-H Shift. Intermediate 1-a-IN2 is then converted to the
benzyl cation intermediate 1-a-IN3 via a [1,5]-H shift transition
state 1-a-TS2. In 1-a-TS2, the bond distance of the breaking
C1−H1 bond is 1.40 Å. The computed activation free energy of
this step is only 5.2 kcal/mol, and the formation of 1-a-IN3 is a
thermodynamically favored process, which is exergonic by 2.8
kcal/mol.
Catalyst Transfer. The catalytic cycle starts from the catalyst
transfer between product-catalyst complex 1-a-IN4 (which is
generated in the previous catalytic cycle) and the substrate 1.
This process is endergonic by 5.0 kcal/mol, giving an alkyne−
Pt complex 1-IN1.
Complex 1-IN1 is a polarized complex with distances
between the coordinated Pt and two sp-hybridized carbon
atoms of 1.98 (Pt−C2) and 2.40 Å (Pt−C3), respectively
(Figure 2). This unsymmetrical coordination mode implies that
Experimentally, the [1,5]-sigmatropic hydrogen shift is
widely used in organic synthesis,26 but usually harsh reaction
D
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
conditions are required.27 For example, thermal cyclization of
2-alkyl-1-allenylbenzenes actually proceeds at 185 °C (Scheme
5).28
should be easier in the Pt−vinylidene system than the other
reactions shown in Scheme 6, which are endothermic.
Cationic 4π-Electrocyclization Step. Chatani proposed that
intermediate D in pathway a undergoes six-electron electro-
cyclization to give Pt(IV) complex E, which then gives the final
indene product via reductive elimination reaction. Similar
mechanism has been proposed by Liu in the Ru-catalyzed
indene and analogue synthesis.5 However, all efforts to locate
such a 6π-electron electrocyclization transition state failed.
Instead, we can locate C1−C2 bond formation transition state
1-a-TS3. IRC calculations found such a transition state leads to
indene−PtCl2 complex 1-a-IN4, in which PtCl2 coordinates to
the alkene’s double bond. This cationic 4π-electrocyclization
step is exergonic by 34.2 kcal/mol and is very facile with an
activation free energy of 3.3 kcal/mol. This facile cyclization
reaction can be envisioned as a 4π-electrocyclization because 1-
a-IN3 can be regarded as a vinyl-substituted benzyl cation. For
comparison, we computed the simple vinyl benzyl cation’s
cyclization of 1D-a-IN3 and 1E-a-IN3, finding that these
processes are easy with activation free energies of 8.2 and 8.7
kcal/mol, respectively (Scheme 7).
Scheme 5. Thermal Cyclization of 2-Alkyl-1-allenylbenzene
Why is the [1,5]-H shift in the present vinylidene system so
easy? To answer it, we computed the activation barriers of
several [1,5]-H shift processes (Scheme 6). Calculations
Scheme 6. Different [1,5]-H Shift Processes
Scheme 7. Different 4π-Electrocyclizations
showed that the [1,5]-H shift of allenyl- and alkenyl-substituted
precursors (1B-a-IN2 and 1C-a-IN2) have activation free
energies of 32.3 and 41.3 kcal/mol, about 25−30 kcal/mol
higher than [1,5]-H shift in the Pt−vinylidene system. We
attributed the lower activation barrier for the Pt−vinylidene
system to the following two reasons. One is that this Pt−
vinylidene is polarized and can be regarded as a zwitterionic
species with positive charge at C2 atom and negative charge at
the PtCl2 moiety. Consequently, [1,5]-hydride shift for this
system is easier for the shifting hydride to the cationic C2 atom.
The second reason is that the [1,5]-H shift is endothermic for
1B-a-IN2 and 1C-a-IN2, while for the Pt−vinylidene system 1-
a-IN2, this is exothermic. On the basis of the Hammond
Postulate,29 the [1,5]-H shift reaction which is exothermic
In summary, the overall potential energy surface of pathway a
shows that the first [1,2]-hydride shift is the rate-limiting step
of the cyclization (Figure 1) and the activation free energy for
the cyclization reaction is 23.6 kcal/mol in the gas phase. The
whole catalytic cycle is highly exergonic by 30.5 kcal/mol.
Why Pathways b and c are Not Favored? Besides pathway
a, the reaction could also occur via pathway b, in which a [1,4]-
H shift first occurs to give 1-b-IN2. Then 1-b-IN2 gives the
final product via the 4π-electron electrocyclization reaction
(Scheme 8). This mechanism has been ruled out by deuterium
labeling experiments.6 Calculations also found that this pathway
is not favored compared to pathway a because the energy
E
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 8. Computed Activation Energies of Some Key Steps in Pathways b and c of Substrate 1
Figure 3. DFT computed energy surface for the PtCl2-catalyzed cycloaddition of 1-(bromoethynyl)-2-isopropylbenzene 2 in pathway a.
barrier of [1,4]-hydride shift via 1-b-TS1 is as high as 30.2 kcal/
mol from 1-IN1, 11.6 kcal/mol higher than that required in
pathway a.
energy of the [1,2]-H shift via 1-c-TS2 would be 35.6 kcal/mol
from 1-c-IN2 to 1-a-IN4, probably due to the very stable of
platinum−carbene intermediate 1-c-IN2. The irreversible [1,5]-
H shift in the gas phase in pathway c is disfavored by only 2.1
kcal/mol than that of the [1,2]-H shift in pathway a, but this is
disfavored further to 3.4 kcal/mol in toluene, suggesting that
pathway c is completely suppressed in solution compared to
pathway a.
From the above Results and Discussion, we can find that for
1-ethynyl-2-isopropylbenzene, the preferred pathway a involves
the first formation of the vinylidene intermediate 1-a-IN2 via a
[1,2]-H shift of the acetylenic hydrogen. This step is the rate-
determining step and requires an activation free energy of 23.6
kcal/mol. Then a [1,5]-H shift of benzylic hydrogen occurs to
form the vinyl benzyl cation intermediate, which could be
transformed into the product−catalyst complex easily via the
The direct insertion of π-alkyne−Pt into a benzylic C−H
bond via pathway c can also be excluded. As shown in Scheme
8, the first step for formation of cyclopentane platinum carbene
intermediate 1-c-IN2 through 1-c-TS1 via an irreversible [1,5]-
H shift is predicted to be more difficult than the [1,2]-H shift
via 1-a-TS1, with an activation free energy of 20.7 kcal/mol
from 1-IN1 to 1-c-TS1. This process can be viewed as the
[1,5]-H shift to generate a zwitterionic species, which quickly
cyclizes to give 1-c-IN2 (Scheme 8). The cyclization step can
also be viewed as a 4π-electron electrocyclization. The
formation of 1-c-IN2 from 1-IN1 is dramatically exergonic by
39.3 kcal/mol as the result of the formation of a new C−C
bond in this process. In addition, the computed activation free
F
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 4. DFT optimized key structures for the PtCl2-catalyzed cyclization of 1-(bromoethynyl)-2-isopropylbenzene 2 in pathway a (distances in Å).
Figure 5. DFT computed energy surfaces of pathways a and c for the PtCl2-catalyzed cycloaddition of 1-isopropyl-2-(phenylethynyl)benzene 3-
C6H5.
direct 4π-electron electrocyclization. Pathway b involving [1,4]-
H shift and pathway c involving [1,5]-H shift both require
higher activation free energies and are not favored compared to
pathway a.
system are different from the reaction of 1-ethynyl-2-
isopropylbenzene, where complex 1-a-IN2 undergoes [1,5]-H
shift and 4π-electron electrocyclization separately to give 1-a-
IN4. In the present system, 2-a-IN2 can form 2-a-IN4 in one
step via a [1,5]-H shift 2-a-TS2, requiring an activation free
energy of 3.8 kcal/mol (Figure 4). This suggests that the
intermediate generated by the [1,5]-H shift step is not a
minimum in the reaction but it directly undergoes the 4π-
electron electrocyclization to give 2-a-IN4.30 It is interesting to
note that the second step of [1,5]-H shift has a negative barrier
in toluene solution, suggesting that this step is very fast.
Calculations found that the rate-determining step in pathway
a is the [1,2]-Br shift via transition state 2-a-TS1. The overall
activation free energy for the pathway a is 15.9 kcal/mol. This
2. Mechanism of the Cyclization of 1-(Bromoethynyl)-
2-isopropylbenzene 2 (Substrate S with R = Br, Scheme
3). Figure 3 gives the DFT-computed energy surface of the
Pt(II)-catalyzed intramolecular cyclization of 1-(bromoethyn-
yl)-2-isopropylbenzene 2 in the favored pathway a, which also
starts from catalyst transferring to the triple bond of substrate
from catalyst−product complex 2-a-IN4, followed by a [1,2]-Br
shift to give vinylidene−platinum complex 2-a-IN2 with an
activation free energy of 15.9 kcal/mol (for some key
structures, see Figure 4). The following steps in the present
G
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 6. DFT optimized key structures for the PtCl2-catalyzed cycloaddition of 1-isopropyl-2-(phenylethynyl)benzene 3-C6H5 in pathways a and c
(distances in Å).
is similar to the reaction of 1, which has the [1,2]-H shift step
as the rate-determining step with an activation energy of 23.6
kcal/mol. This suggests that 2 is more reactive than 1 under the
catalysis of PtCl2. Calculation results are consistent with the
experimental observations (Scheme 2).6
We also computed the energy surfaces for pathway b and c.
The [1,4]-H shift step in pathway b and the [1,5] hydride shift
step in pathway c require an activation free energy of 25.0 and
20.4 kcal/mol in the gas phase, respectively. Because both
pathways b and c need higher activation barriers than pathway a
has, 1-bromoethynyl-2-isopropylbenzene prefers pathway a to
generate indene product.
3. Mechanism of Cyclization of 1-Isopropyl-2-
(phenylethynyl)benzene 3-C6H5 (Substrate S with R =
Ph, Scheme 3). We computed the energy surfaces of phenyl
substituted compound 3-C6H5 for pathways a, b, and c (Figures
5 and 6). The [1,4]-H shift for pathway b has an activation
energy of 30.2 kcal/mol, which is significantly higher than those
in pathways a and c, so pathway b can be ruled out for further
consideration (Scheme 4 and Figure 5).
bond is weaker than the ordinary C−H bond in the previous
case.
There is a possible [1,2]-Ph shift step in pathway c to
compete with the [1,2]-H shift from intermediate 3-c-IN2, but
this step can be ruled out due to its higher activation energy
(this reaction is difficult than the [1,2]-H shift by 4.8 kcal/mol,
see Figure 5). Calculations suggested that in pathway c, the
[1,2]-H shift is the rate-determining step with the activation
free energy of 26.4 kcal/mol in the gas phase.
Calculations indicated that the [1,5]-hydride shift and
cyclization in pathway c is irreversible, and this step is easier
than the rate-determining [1,2]-H shift in pathway a by 2.0
kcal/mol (in solution, the energy difference is reduced to 1.5
kcal/mol), suggesting that the C2-phenyl substituted product
3P-C6H5 is the major product, while C3-phenyl substituted
product 3P′-C6H5 is the minor product (Figure 5). In He’s
experimental result, only 3P-C6H5 was isolated and reported
(Scheme 2, reaction 3).7 To test whether our calculation
prediction was correct or not, we repeated this experiment
(Scheme 9 and Table 1).
To our delight, we found that both products 3-P and 3-P′
can be obtained with a ratio of 3.0:1 when Ar  Ph (Figure 7).
This suggests that the energy preference for the pathway c over
pathway a is 0.9 kcal/mol. Our computed energy difference
here is overestimated by 0.6 kcal/mol in the solution (1.1 kcal/
mol in the gas phase). We further reasoned that this energy
In pathway a, the first step of [1,2]-Ph migration reaction is
the rate-determining step, which requires an activation free
energy of 19.7 kcal/mol. The second step is a concerted [1,5]-
hydride shift and cationic 4π-electrocyclization reaction, which
is easy with an activation free energy of 5.8 kcal/mol (for ligand
transfer reactions, see the Supporting Information).
But to our surprise, we found that pathway c is favored over
pathway a by about 2.0 kcal/mol. The first step in pathway c is
[1,5]-hydride shift and cationic 4π-electrocyclization, requiring
an activation free energy of 17.7 kcal/mol. This step is
exergonic by 29.3 kcal/mol. The final step in pathway c is the
[1,2]-H shift to give indene−PtCl2 complex 3-c-IN3 via
transition state 3-c-TS2. This step requires an activation free
energy of 26.4 kcal/mol.31 Compared to the [1,2]-H shift for 1-
ethynyl-2-isopropylbenzene 1, which requires an activation free
energy of 35.6 kal/mol (Scheme 8), the present [1,2]-H shift is
much easier. This is because the phenyl group can stabilize the
transition state through conjugation. Besides, the benzylic C−H
Scheme 9. Cyclizations of Aryl Substituted Alkynes 3
(Reaction Yields and the Ratios of Products Are Given in
Table 1)
H
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 1. Experimental and Computational Study of the Product Distribution for PtCl2 Catalyzed Cyclization of 3 with Different
Ar Groups
DFT predicted ratio in solution at 120 °C
(3-P:3-P′)
experimenally measured ratio
total yield of 3-P and
3-P′ (%)
a
b,
c
b,c
entry
Ar
3-a-TS
3-c-TS
(3-P:3-P′)
1
2
3
C6H5
17.0 (19.7)
15.2 (18.4)
16.7 (19.2)
15.5 (17.7)
15.0 (17.4)
16.0 (18.0)
6.8:1
1.3:1
3.0:1
2.6:1
1.9:1
43
63
70
4-tBuC6H4
4-MeOC6H4
2.4:1
a
b
The reaction was catalyzed by 10 mol% PtCl2 in toluene (see Scheme 9). The computed free energy in kcal/mol in toluene was relative to that of
c
3/PtCl2 complex. The value in the parentheses is the computed activation free energy in the gas phase.
1
Figure 7. H NMR of mixture from the cyclization reaction (36 h) of 3-C6H5 catalyzed by PtCl2 at 120 °C after removing catalyst by flash
chromatography on silica gel.
difference could be changed by introducing different sub-
stituents in the aryl group, and consequently, the ratio of the
two products could be changed, too. Calculations indicated that
the energy preference of pathway c over a was about 1 kcal/mol
when a t-Bu or a MeO group was introduced in the aryl group,
suggesting that more 3-P′ would be generated (Table 1). Our
new experiments also supported the calculation results because
the ratios of 3-P to 3-P′ in these two cases were 2.6:1 and 1.9:1,
respectively.
Experimentally, He and co-workers used both PtCl2 and
CuBr as the catalysts. At this moment, we do not know how the
CuBr can facilitate the reaction. But we found that using He’s
reaction conditions, both products 3-P and 3-P′ were obtained,
in contrast to their previous reports (see the Experimental
Section). This suggested that pathways a and c could both take
place when both catalyst of PtCl2 and additive of CuBr were
used. Further investigation on how CuBr affects the reaction
will be carried out.
4. Mechanism of the Cyclization of 1-Isopropyl-2-
(prop-1-ynyl)benzene 4-CH3 (Substrate S with R = Me,
Scheme 3). Experimentally, He and co-workers carried out the
indene synthesis using R = alkyl group in their optimized
reaction conditions (PtCl2, CuBr) but not under the PtCl2
catalysis conditions. We were interested to know what could
happen for this system using PtCl2 as the only catalyst. Our
DFT calculations found that for reaction 4, the pathway c is the
most favored (Figures 8, 9 and Scheme 4). The [1,5]-H shift/
cyclization via 4-c-TS1 and [1,2]-H shift step via 4-c-TS2
require activation free energies of 21.5 and 25.4 kcal/mol,
respectively. The corresponding [1,2]-Me shift for pathway a
requires an activation free energy of 31.6 kcal/mol, indicating
that C3 substituted product is difficult to be generated. We
carried out an experiment to test this, showing that our
calculations agreed with the new experiment. The cyclo-
isomerization of 4-CH2CH2Ph (substrate 4 with R =
CH2CH2Ph) gave a single C2 substituted product (see Scheme
2 and Experimental Section).
I
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 8. DFT computed energy surface of pathway c for the PtCl2-catalyzed cycloaddition of 1-isopropyl-2-(prop-1-ynyl)benzene 4-CH3.
Figure 9. DFT optimized key structures for the PtCl2-catalyzed cycloaddition of 1-isopropyl-2-(prop-1-ynyl)benzene 4-CH3 (distances in Å).
Scheme 10. Competing Pathways of PtCl2-Catalyzed Cyclization of o-Isopropyl Substituted Aryl Alkynes and How Substituents
Affect the Regiochemistry
5. Competition of [1,2]-R Migration and [1,5]-H shift is
Key to Determine the Reaction Pathway. Through DFT
calculations, three modes of sp3 C−H activation shown in
Schemes 3 and 4 have been investigated. We found the most
efficient C−H activation is from intermediate C (vinylidene
complex) to D in pathway a with activation free energy of
about 5 kcal/mol. The C−H activation via [1,5]-H shift from H
to I in pathway c usually requires activation free energies of
J
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
through a pad of Celite. The filtrate was concentrated in a vacuum, and
the residue was purified by flash chromatography on silica gel (eluted
with PE) to afford the product as a white solid (532.3 mg, 95%). Rf =
0.45 (PE), mp: 50−52 °C. 1H NMR (400 MHz, CDCl3): δ 7.60−7.55
(m, 3H), 7.47−7.44 (m, 2H), 7.38−7.34 (m, 2H), 7.26−7.21 (m, 1H),
3.71−3.60 (m, 1H), 1.41−1.39 (m, 15H). 13C NMR (100 MHz,
CDCl3): δ 151.5, 150.5, 132.3, 131.3, 128.6, 125.6, 125.4, 125.0, 122.3,
120.7, 93.3, 87.7, 34.8, 31.8, 31.3, 23.2. IR (neat): 2969, 1516, 1482,
1468, 1367, 1270 cm−1. HRMS (ESI): calcd for C21H25 (M + H)+,
277.1951; found, 277.1958.
about 17−20 kcal/mol. The most difficult C−H activation is
through [1,4]-H shift from endo vinyl cation in pathway b, with
activation free energies of about 25−30 kcal/mol. Due to these
reasons, pathway b is not operative for the C−H activation.
Because C−H activations in pathways a and c are easy, the
competition between pathways a and c is determined by the
[1,2]-R migration step and [1,5]-H shift. The results shows that
easier [1,2]-migration (when R = H and Br) favors pathway a
while difficult migration (when R = alkyl group) prefers
pathway c. When R = aryl group, pathways a and c have very
close activation energies due to the competitive [1,2]-Ar
migration and [1,5]-H shift, and consequently two products
can be obtained (Scheme 10). This prediction was supported
by new experiments.
General Cyclization Procedure Using PtCl2 as Catalyst
(Schemes 2 and 9). PtCl2 (10 mol% to the substrate) was charged
in a base-washed, oven-dried Schlenk flask under an atmosphere of
nitrogen, and then a solution of the alkyne substrate in dried toluene
was added. The reaction mixture was stirred at 120 °C for 36 h. After
being cooled to room temperature, the mixture was concentrated and
the residue was purified by flash column chromatography with silica
gel to afford the mixture of cycloaddutcs.
CONCLUSIONS
■
1-Isopropyl-2-(phenylethynyl)benzene (3-C6H5)7 (23.0 mg, 0.10
mmol), PtCl2 (3.0 mg, 0.011 mmol), toluene (1.5 mL), eluent: PE,
We scrutinized computationally and (in some cases) exper-
imentally the mechanisms of platinum(II)-catalyzed intra-
molecular cyclization of ortho-substituted aryl alkynes involving
a key step of sp3 C−H bond activation. We evaluated the
feasibility of three pathways proposed by Chatani, He, and us
for four different substrates with the different R substituents (R
= H, Br, Ph, and Me) in the alkyne part. Calculations found
that for substrates with R = H or Br, these reactions favor
pathway a to undergo [1,2]-H and [1,2]-Br migrations from the
endo vinyl cations, generating Pt−vinylidene complexes. Then
facile [1,5]-H shifts and 4π-electrocyclization (in two-step or
one-step) furnish the indene products. H and Br atoms in this
pathway migrate from the original terminal position of the
alkyne to the C3 position of the indenes. For substrate with R =
Me, the reaction reacts through pathway c, starting from
irreversible [1,5]-H shift from exo vinyl cation species. The
competing pathway a via [1,2]-methyl migration is very difficult
and cannot take place. Consequently, the alkyl group is still at
the C2 position of the indene products. For substrates with R =
Ar, DFT calculations indicated that both pathways a and c can
take place, leading to two indene products. Major product with
Ph at C2 position was obtained through pathway c, and the
minor product with Ph at C3 position was obtained through
pathway a. The prediction was further supported by the new
experiments. The in-depth understanding of the present three
C−H activation modes will be helpful for the future design of
new C−H activation reactions.
7
total yield of cycloadducts (10.0 mg, 43%), ratio of 3P-C6H5 /3P′-
32
33
1
C6H5 = 3.0:1 (determined by H NMR).
1-((4-tert-Butylphenyl)ethynyl)-2-isopropylbenzene (3-tBuC6H4)
(29.3 mg, 0.12 mmol), PtCl2 (3.7 mg, 0.014 mmol), toluene (1.5
mL), eluent: PE, total yield of cycloadducts (18.4 mg, 63%), ratio of
3P-tBuC6H4/3P′-tBuC6H4 = 2.6:1 (determined by H NMR).
1
1-Isopropyl-2-((4-methoxyphenyl)ethynyl)benzene (3-
MeOC6H4)34 (31.4 mg, 0.13 mmol), PtCl2 (3.5 mg, 0.013 mmol),
toluene (1.3 mL), eluent: PE to PE/EA = 100:1, total yield of
7
cycloadducts (22.0 mg, 70%), ratio of 3P-MeOC6H4 /3P′-MeOC6H4
= 1.9:1 (determined by H NMR).
1
1-Isopropyl-2-(4-phenylbut-1-ynyl)benzene (4-CH2CH2Ph)7 (40.6
mg, 0.16 mmol), PtCl2 (4.3 mg, 0.016 mmol), toluene (3.0 mL),
eluent: PE, conversion 71%, yield of 4P-CH2CH2Ph7 (6.4 mg, brsm
22%). This reaction is not clean because we obtained a mixture of
substrate and product and the yield of product was determined by 1H
NMR from the isolated mixture of substrate and product. In He’s
experiment using both PtCl2 and CuBr, the reaction yield was 71%.7
General Cyclization Procedure using PtCl2 as Catalyst with
Additive CuBr (He’s Original Procedure).7 PtCl2 (10 mol % to the
substrate) and CuBr (2 equiv to the substrate) were charged in a base-
washed, oven-dried Schlenk flask under an atmosphere of nitrogen.
Then a solution of the alkyne substrates in dried toluene was added.
The solution was then stirred at 120 °C for 36 h. After being cooled to
room temperature, the mixture was concentrated and the residue was
purified by flash column chromatography with silica gel to give the
cycloaddition products.
1-Isopropyl-2-(phenylethynyl)benzene (3-C6H5)7 (23.0 mg, 0.10
mmol), PtCl2 (3.0 mg, 0.011 mmol), CuBr (28 mg, 0.20 mmol),
toluene (1.5 mL), eluent: PE, total yield of cycloadducts (17.7 mg,
77%), ratio of 3P-C6H5/3P′-C6H5 = 3.0:1 (determined by 1H NMR).
1-((4-tert-Butylphenyl)ethynyl)-2-isopropylbenzene (3-tBuC6H4)
(31.8 mg, 0.12 mmol), PtCl2 (3.7 mg, 0.014 mmol), CuBr (40.2
mg, 0.28 mmol), toluene (1.5 mL), eluent: PE, total yield of
cycloadducts (27.4 mg, 86%), ratio of 3P-tBuC6H4/3P′-tBuC6H4 =
EXPERIMENTAL SECTION
■
General Information. Toluene was dried over Na before use.
Reaction temperatures refer to the external temperature or to the
temperature of the bath in which the reaction vessel was partially
immersed. Data for 1H NMR (400 MHz) spectra are reported as
follows: chemical shift (ppm, referenced to TMS; s = singlet, d =
doublet, t = triplet, q = quartet, dd = doublet of doublets, dt = doublet
of triplets, ddd = doublet of doublet of doublets, ddt = doublet of
doublet of triplets, m = multiplet), coupling constant (Hz), and
integration. Data for 13C NMR (100 MHz) are reported in terms of
chemical shift (ppm) relative to residual solvent peak (CDCl3: 77.0
ppm). Infrared spectra are reported in wavenumbers (cm−1). HRMS
were performed under ESI ionization technique using FT-ICR
analyzer. PE = petroleum ether, EA = ethyl acetate.
1
2.5:1 (determined by H NMR).
1-Isopropyl-2-((4-methoxyphenyl)ethynyl)benzene (3-
MeOC6H4)34 (29.2 mg, 0.12 mmol), PtCl2 (3.2 mg, 0.012 mmol),
CuBr (34.4 mg, 0.24 mmol), toluene (1.3 mL), eluent: PE to PE/EA =
100:1, total yield of cycloadducts (25.7 mg, 88%), ratio of 3P-
1
MeOC6H4/3P′-MeOC6H4 = 1.8:1 (determined by H NMR).
Test of Whether Additive of D2O or CD3OD Could Catalyze
the [1,2]-H Shift in the PtCl2 Catalyzed Cyclization. Similar to the
general cyclization procedure using PtCl2 as catalyst except for 5 equiv
of D2O or CD3OD was added to the Schlenk flask before heating. For
D2O: 1-isopropyl-2-((4-methoxyphenyl)ethynyl)benzene (3-
MeOC6H4), (27.8 mg, 0.10 mmol), PtCl2 (2.7 mg, 0.010 mmol),
toluene (1.0 mL), D2O (10.0 mg, 0.50 mmol), eluent: PE to PE/EA =
100:1, conversion 65%, total yield of cycloadducts (15.5 mg, brsm
86%), ratio of 3P-MeOC6H4/3P′-MeOC6H4 = 1.9:1 (determined by
1-((4-tert-Butylphenyl)ethynyl)-2-isopropylbenzene (3-tBuC6H4).
To a stirred mixture of 1-iodo-2-isopropylbenzene (493.4 mg, 2.00
mmol), PdCl2(PPh3)2 (70.2 mg, 0.10 mmol), and CuI (38.1 mg, 0.20
mmol) in NEt3 (3 mL), 4-tert-butylphenylacetylene (348.1 mg, 2.20
mmol) was added dropwise at 0 °C. The mixture was then stirred at
room temperature overnight. On completion, the mixture was passed
K
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
1H NMR). For CD3OD: 1-isopropyl-2-((4-methoxyphenyl)ethynyl)-
benzene (3-MeOC6H4), (27.4 mg, 0.099 mmol), PtCl2 (2.6 mg,
0.0099 mmol), toluene (1.0 mL), CD3OD (18.0 mg, 0.50 mmol),
eluent: PE to PE/EA = 100:1, total yield of cycloadducts (22.6 mg,
82%), ratio of 3P-MeOC6H4/3P′-MeOC6H4 = 1.7:1 (determined by
1H NMR).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are indebted to the generous financial support from the
Natural Science Foundation of China (21232001) and the
National Basic Research Program of China-973 Program
(2011CB808600). We thank Mr. Pei-Jun Cai and Dr. Wenbo
Xu for repeating some experiments.
1,1-Dimethyl-2-phenyl-1H-indene (3P-C6H5).7 Colorless oil; Rf =
0.35 (PE). 1H NMR (400 MHz, CDCl3): δ 7.65−7.63 (m, 2H), 7.45−
7.39 (m, 4H), 7.36−7.31 (m, 1H), 7.30−7.24 (m, 2H), 7.00 (s, 1H),
1.53 (s, 6H). 13C NMR (100 MHz, CDCl3): δ 155.9, 155.0, 141.5,
136.2, 128.4, 127.3, 127.2, 126.6, 126.0, 125.2, 121.2, 121.1, 50.8, 25.0.
IR (neat): 2969, 2928, 2864, 1494, 1471, 1445, 1352, 1028 cm−1.
HRMS (ESI): calcd for C17H17 (M + H)+, 221.1325; found, 221.1319.
1,1-Dimethyl-3-phenyl-1H-indene (3P′-C6H5).32 Colorless oil; Rf =
0.41 (PE). 1H NMR (400 MHz, CDCl3): δ 7.62−7.60 (m, 2H), 7.53−
7.51 (m, 1H), 7.48−7.43 (m, 2H), 7.42−7.35 (m, 2H), 7.30−7.27 (m,
2H), 6.43 (s,1H), 1.41 (s, 6H). 13C NMR (100 MHz, CDCl3): δ
154.3, 143.9, 141.7, 140.7, 135.8, 128.4, 127.6, 127.4, 126.3, 125.3,
121.4, 120.5, 48.3, 24.6. IR (neat): 2965, 2864, 2842, 1620, 1512,
1468, 1352, 1251, 1177, 1039 cm−1. HRMS (ESI): calcd for C17H17
(M + H)+, 221.1325; found, 221.1323.
REFERENCES
■
(1) (a) Lyons, T. W.; Sanford, M. S. Chem. Rev. 2010, 110, 1147.
(b) Satoh, T.; Miura, M. Chem.Eur. J. 2010, 16, 11212. (c) Chen, X.;
Engle, K. M.; Wang, D.-H.; Yu, J.-Q. Angew. Chem., Int. Ed. 2009, 48,
5094. (d) Park, Y. J.; Park, J.-W.; Jun, C.-H. Acc. Chem. Res. 2008, 41,
222. (e) Lewis, J. C.; Bergman, R. G.; Ellman, J. A. Acc. Chem. Res.
́
2008, 41, 1013. (f) Díaz-Requejo, M. M.; Perez, P. J. Chem. Rev. 2008,
108, 3379. (g) Davies, H. M. L.; Manning, J. R. Nature 2008, 451, 417.
(h) Herrerías, C. I.; Yao, X.; Li, Z.; Li, C.-J. Chem. Rev. 2007, 107,
2546. (i) Godula, K.; Sames, D. Science 2006, 312, 67.
2-(4-tert-Butylphenyl)-1,1-dimethyl-1H-indene (3P-tBuC6H4).
1
Light-yellow oil; Rf = 0.32 (PE). H NMR (400 MHz, CDCl3): δ
(2) (a) Tauchert, M. E.; Incarvito, C. D.; Rheingold, A. L.; Bergman,
R. G.; Ellman, J. A. J. Am. Chem. Soc. 2012, 134, 1482. (b) Ackermann,
L. Chem. Rev. 2011, 111, 1315. (c) Sawyer, K. R.; Cahoon, J. F.;
Shanoski, J. E.; Glascoe, E. A.; Kling, M. F.; Schlegel, J. P.; Zoerb, M.
C.; Hapke, M.; Hartwig, J. F.; Webster, C. E.; Harris, C. B. J. Am.
Chem. Soc. 2010, 132, 1848. (d) Potavathri, S.; Pereira, K. C.;
Gorelsky, S. I.; Pike, A.; LeBris, A. P.; DeBoef, B. J. Am. Chem. Soc.
2010, 132, 14676.
(3) Nakamura, I.; Bajracharya, G. B.; Wu, H.; Oishi, K.; Mizushima,
Y.; Gridnev, I. D.; Yamamoto, Y. J. Am. Chem. Soc. 2004, 126, 15423.
(4) Bajracharya, G. B.; Pahadi, N. K.; Gridnev, I. D.; Yamamoto, Y. J.
Org. Chem. 2006, 71, 6204.
(5) (a) Odedra, A.; Datta, S.; Liu, R.-S. J. Org. Chem. 2007, 72, 3289.
(b) Datta, S.; Odedra, A.; Liu, R.-S. J. Am. Chem. Soc. 2005, 127,
11606.
(6) Tobisu, M.; Nakai, H.; Chatani, N. J. Org. Chem. 2009, 74, 5471.
(7) Yang, S.; Li, Z.; Jian, X.; He, C. Angew. Chem., Int. Ed. 2009, 48,
3999.
7.58−7.56 (m, 2H), 7.43−7.41 (m, 2H), 7.36−7.34 (m, 2H), 7.26−
7.18 (m, 2H), 6.98 (s, 1H), 1.50 (s, 6H), 1.35 (s, 9H). 13C NMR (100
MHz, CDCl3): δ 155.6, 155.2, 150.2,141.6, 133.0, 126.7, 126.6,
125.30, 125.29, 125.0, 121.1, 120.9, 50.7, 34.5, 31.3, 25.2. IR (neat):
2972, 2872, 1516, 1471, 1367, 1274, 1117, 1024 cm−1. HRMS (ESI):
calcd for C21H25 (M + H)+, 277.1951; found, 277.1953.
3-(4-tert-Butylphenyl)-1,1-dimethyl-1H-indene (3P′-tBuC6H4).
1
Light-yellow oil; Rf = 0.40 (PE). H NMR (400 MHz, CDCl3): δ
7.55−7.52 (m, 3H), 7.47−7.45 (m, 2H), 7.39−7.37 (m, 1H), 7.29−
7.22 (m, 2H), 6.39 (s, 1H), 1.39 (s, 6H), 1.37 (s, 9H). 13C NMR (100
MHz, CDCl3): δ 154.5, 150.5, 143.7, 141.9, 140.6, 133.0, 127.3, 126.3,
125.5, 125.3, 121.5, 120.7, 48.3, 34.6, 31.4, 24.8. IR (neat): 2969, 2864,
1516, 1471, 1412, 1367, 1270, 1117, 1024 cm−1. HRMS (ESI): calcd
for C21H25 (M + H)+, 277.1951; found, 277.1953.
2-(4-Methoxyphenyl)-1,1-dimethyl-1H-indene (3P-MeOC6H4).7
1
Light-yellow oil; Rf = 0.45 (PE/EA = 50:1). H NMR (400 MHz,
CDCl3): δ 7.58−7.54 (m, 2H), 7.35−7.33 (m, 2H), 7.26−7.18 (m,
2H), 6.96−6.92 (m, 2H), 6.89 (s, 1H), 3.84 (s, 3H), 1.48 (s, 6H). 13C
NMR (100 MHz, CDCl3): δ 158.9, 155.5, 154.9, 141.7, 128.7, 128.3,
126.6, 124.9, 124.6, 121.1, 120.8, 113.8, 55.3, 50.7, 25.1. IR (neat):
2959, 2918, 2849, 1633, 1508, 1469, 1249, 1178, 1034 cm−1. HRMS
(ESI): calcd for C18H19O (M + H)+, 251.1430; found, 251.1431.
3-(4-Methoxyphenyl)-1,1-dimethyl-1H-indene (3P′-MeOC6H4).
(8) For selected recent transition-metal-catalyzed benzylic sp3 C-H
activation reactions, see: (a) Minami, Y.; Yamada, K.; Hiyama, T.
Angew. Chem., Int. Ed. 2013, 52, 10611. (b) Bhunia, S.; Ghorpade, S.;
Huple, D. B.; Liu, R.-S. Angew. Chem., Int. Ed. 2012, 51, 2939. (c) Xia,
J.-D.; Deng, G.-B.; Zhou, M.-B.; Liu, W.; Xie, P.; Li, J.-H. Synlett 2012,
23, 2707. (d) Zhao, S.-C.; Shu, X.-Z.; Ji, K.-G.; Zhou, A.-X.; He, T.;
Liu, X.-Y.; Liang, Y.-M. J. Org. Chem. 2011, 76, 1941. (e) Zeng, X.;
Ilies, L.; Nakamura, E. J. Am. Chem. Soc. 2011, 133, 17638. (f) Shu, X.-
Z.; Ji, K.-G.; Zhao, S.-C.; Zheng, Z.-J.; Chen, J.; Lu, L.; Liu, X.-Y.;
Liang, Y.-M. Chem.Eur. J. 2008, 14, 10556.
1
Light-yellow oil; Rf = 0.49 (PE/EA = 50:1). H NMR (400 MHz,
CDCl3): δ 7.54−7.51 (m, 2H), 7.49−7.47 (m,1H), 7.38−7.36 (m,
1H), 7.26−7.23 (m, 2H), 6.98−6.95 (m, 2H), 6.33 (s, 1H), 3.85 (s,
3H), 1.37 (s, 6H). 13C NMR (100 MHz, CDCl3): δ 159.1, 154.4,
142.9, 142.0, 140.2, 128.7, 128.4, 126.3, 125.2, 121.4, 120.5, 113.9,
55.3, 48.2, 24.7. IR (neat): 2959, 2926, 1600, 1509, 1464, 1248, 1177,
1035 cm−1. HRMS (ESI): calcd for C18H19O (M + H)+, 251.1430;
found, 251.1429.
(9) Zhao and coworkers reported the mechanism of cyclization of 1-
ethynyl-2-isopropylbenzene, but Br, aryl, methyl substituted com-
pounds were not discussed, and pathway c proposed by us was not
considered, either. See Li, Z.-F.; Fan, Y.; DeYonker, N. J.; Zhang, X.;
Su, C.-Y.; Xu, H.; Xu, X.; Zhao, C. J. Org. Chem. 2012, 77, 6076.
(10) (a) Davies, H. M. L.; Dick, A. R. Top. Curr. Chem. 2010, 292,
303. (b) Davies, H. M. L.; Beckwith, R. E. J. Chem. Rev. 2003, 103,
2861. (c) Davies, H. M. L.; Antoulinakis, E. G. Org. Lett. 2000, 2, 4153.
(11) (a) Lynam, J. M. Chem.Eur. J. 2010, 16, 8238. (b) Fiukamizu,
K.; Miyake, Y.; Nishibayashi, Y. Angew. Chem., Int. Ed. 2009, 48, 2534.
(c) Trost, B. M.; McClory, A. Chem.Asian J. 2008, 3, 164. (d) Liu,
R.-S. Synlett 2008, 801. (e) Bruneau, C.; Dixneuf, P. H. Angew. Chem.,
Int. Ed. 2006, 45, 2176. (f) Caskey, S. R.; Stewart, M. H.; Johnson, M.
J. A.; Kampf, J. W. Angew. Chem., Int. Ed. 2006, 45, 7422. (g) Nevado,
C.; Echavarren, A. M. Synthesis 2005, 167.
ASSOCIATED CONTENT
* Supporting Information
■
S
NMR spectra of new compounds, full citation of Gaussian 09,
some disfavored pathways, computed energy surface using
B3LYP method, optimized Cartesian coordinates and energies.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: yuzx@pku.edu.cn.
■
(12) Frisch, M. J.; et al. Gaussian 09, revision A.02; Gaussian, Inc.:
Wallingford, CT, 2010.
(13) Parr, R. G.; Yang, W. Density-Functional Theory of Atoms and
Author Contributions
§These authors contributed equally.
Molecules; Oxford University Press: Oxford, U.K., 1989.
L
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(14) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(15) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
cyclization products came from the orgianl benzylic position (see refs
4 6, and 7). For the discussion of water-catalyzed [1,2]-H shift, see:
(a) Shi, F.-Q.; Li, X.; Xia, Y.; Zhang, L.; Yu, Z.-X. J. Am. Chem. Soc.
2007, 129, 15503. For water-catalyzed [1,2]-proton shift processes
supported by experiments and DFT calculations, see: (b) Xia, Y.;
Liang, Y.; Chen, Y.; Wang, M.; Jiao, L.; Huang, F.; Liu, S.; Li, Y.; Yu,
Z.-X. J. Am. Chem. Soc. 2007, 129, 3470. (c) Liang, Y.; Liu, S.; Xia, Y.;
Li, Y.; Yu, Z.-X. Chem.Eur. J. 2008, 14, 4361. (d) Liang, Y.; Liu, S.;
Yu, Z.-X. Synlett 2009, 905.
Molecular Orbital Theory; Wiley: New York, 1986.
(16) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
(b) Dunning, T. H., Jr.; Hay, P. J. In Modern Theoretical Chemistry;
Schaefer, H. F., III, Ed.; Plenum Press: New York, 1977; pp 1−28.
(17) (a) Huang, G.; Cheng, B.; Xu, L.; Li, Y.; Xia, Y. Chem.Eur. J.
2012, 18, 5401. (b) Jin, L.; Wu, Y.; Zhao, X. Organometallics 2012, 31,
3065. (c) Xia, Y.; Huang, G. J. Org. Chem. 2010, 75, 7842. (d) Soriano,
E.; Marco-Contelles, J. J. Org. Chem. 2007, 72, 1443. (e) Martin-
Matute, B.; Nevado, C.; Cardenas, D. J.; Echavarren, A. M. J. Am.
Chem. Soc. 2003, 125, 5757. (f) Mendez, M.; Munoz, M. P.; Nevado,
C.; Cardenas, D. J.; Echavarren, A. M. J. Am. Chem. Soc. 2001, 123,
10511.
(18) (a) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523.
(b) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154.
(19) (a) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157.
(b) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215.
(20) (a) Takano, Y.; Houk, K. N. J. Chem. Theory Comput. 2005, 1,
70. (b) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem.
2003, 24, 669. (c) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102,
1995.
(32) (a) Wagner, P. J.; Giri, B. P.; Scaiano, J. C.; Ward, D. L.; Gabe,
E.; Lee, F. L. J. Am. Chem. Soc. 1985, 107, 5483. (b) Shchukin, A. O.;
Vasil’ev, A. V. Russ. J. Org. Chem. 2007, 43, 784.
(33) When we repeated this reaction, sometimes the substrate cannot
be consumed completely and the reaction gave a mixture of substrate
and two products (conversion: 85%, brsm: 49%, ratio of 3P-C6H5/
3P′-C6H5 = 3.0:1; conversion 45%, brsm: 48%, ratio of 3P-C6H5/3P′-
C6H5 = 2.8:1). We speculated that the quality of the used PtCl2
catalyst could be a factor affecting the reaction outcomes.
(34) Zeng, X.; Ilies, L.; Nakamura, E. J. Am. Chem. Soc. 2011, 133,
17638.
(21) Barone, V.; Cossi, M.; Tomasi, J. J. Chem. Phys. 1997, 107, 3210.
(22) Legault, C. Y. CYLview, 1.0b; Universite
Sherbrooke, Quebec, Canada, 2009; http://www.cylview.org.
(23) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
899.
́
de Sherbrooke:
́
(24) We can locate another complex 1-IN1′ in a symmetric fashion,
but it is higher in energy by 3.5 kcal/mol than 1-IN1, see the
Supporting Information.
(25) (a) Bruce, M. I. Chem. Rev. 1991, 91, 197. (b) Bruce, M. I.;
Swincer, A. G. In Advances in Organometallic Chemistry; Stone, F. G. A.,
Robert, W., Eds.; Academic Press: 1983; Vol. 22, p 59.
(26) (a) Chen, L.; Zhang, L.; Lv, J.; Cheng, J.-P.; Luo, S. Chem.Eur.
J. 2012, 18, 8891. (b) Vadola, P. A.; Carrera, I.; Sames, D. J. Org. Chem.
2012, 77, 6689. (c) Sugiishi, T.; Nakamura, H. J. Am. Chem. Soc. 2012,
134, 2504. (d) Han, Y.-Y.; Han, W.-Y.; Hou, X.; Zhang, X.-M.; Yuan,
W.-C. Org. Lett. 2012, 14, 4054. (e) Cao, W.; Liu, X.; Wang, W.; Lin,
L.; Feng, X. Org. Lett. 2011, 13, 600. (f) Melchionna, M.; Nieger, M.;
Helaja, J. Chem.Eur. J. 2010, 16, 8262. (g) Jurberg, I. D.;
Odabachian, Y.; Gagosz, F. J. Am. Chem. Soc. 2010, 132, 3543.
(h) Kang, Y. K.; Kim, S. M.; Kim, D. Y. J. Am. Chem. Soc. 2010, 132,
11847. For a recent review, see (i) Haibach, M. C.; Seidel, D. Angew.
Chem., Int. Ed. 2014, 53, 5010.
(27) Roth, W. R.; Konig, J. Liebig’s Ann. Chem. 1966, 699, 24.
(28) (a) Heimgartner, H.; Zsindely, J.; Hansen, H.-J.; Schmid, H.
Helv. Chim. Acta 1973, 56, 2924. (b) For a recent mechanism study,
see Alajarin, M.; Bonillo, B.; Marin-Luna, M.; Sanchez-Andrada, P.;
Vidal, A. Chem.Eur. J. 2013, 19, 16093.
(29) Hammond, G. S. J. Am. Chem. Soc. 1955, 77, 334.
(30) (a) Rzepa, H. S.; Wentrup, C. J. Org. Chem. 2013, 78, 7565.
(b) Kraka, E.; Cremer, D. Acc. Chem. Res. 2010, 43, 591.
(31) The [1,2]-H shift could be assisted by a trace amount of water
or other proton source. To test this, we carried out the cyclo-
isomerization of 3-MeOC6H4 catalyzed by PtCl2 in the presence of
D2O and CD3OD (in ref 17f, there are discussions of the effects of
water and methanol). In both cases, there were no D atom
incorporation in the final products, suggesting that the present
[1,2]-H shift take place intramolecularly (see the Experimental Section
for details). Previously, we have shown that water-catalyzed [1,2]-H
shift in Au system needed the help of some substituent as a handle,
which can form complex with water through hydrogen bond. In the
present system, no such functional group is available in the substrate
and water or methanol-catalyzed [1,2]-H shift is not operative. In
addition, under heating at 120 °C, D2O or CD3OD could evaporate
from the reaction system and cannot participate in the reaction. Also
experiments from Yamamoto, Chatani, and He using D-substituted
substrate supported that D atoms at C2 and C3 positions of the final
M
dx.doi.org/10.1021/jo500839c | J. Org. Chem. XXXX, XXX, XXX−XXX