Mechanistic Study of Direct Arylation of Indole Using Differential Selectivity Measurements: Shedding Light on the Active Species and Revealing the Key Role of Electrophilic Substitution in the Catalytic Cycle

Article abstract of DOI:10.1021/acs.organomet.8b00216

Differential selectivity of the direct arylation of indole with aryl halides under competing and noncompeting conditions with a varying set of reaction parameters was determined using phase trajectories. The results described herein allow for conclusions to be drawn regarding the character of active complexes (cationic, neutral, or anionic) as well as realization of the indole electrophilic substitution in the catalytic cycle using the ligand-free catalytic system.

Full text of DOI:10.1021/acs.organomet.8b00216

Article
Cite This: Organometallics XXXX, XXX, XXX−XXX
Mechanistic Study of Direct Arylation of Indole Using Differential
Selectivity Measurements: Shedding Light on the Active Species and
Revealing the Key Role of Electrophilic Substitution in the Catalytic
Cycle
Anna A. Kurokhtina, Elizaveta V. Larina, Elena V. Yarosh, Nadezhda A. Lagoda,
and Alexander F. Schmidt*
Department of Chemistry, Irkutsk State University, K. Marx Str., 1, 664003 Irkutsk, Russia
*
S Supporting Information
ABSTRACT: Differential selectivity of the direct arylation of
indole with aryl halides under competing and noncompeting
conditions with a varying set of reaction parameters was
determined using phase trajectories. The results described
herein allow for conclusions to be drawn regarding the
character of active complexes (cationic, neutral, or anionic) as
well as realization of the indole electrophilic substitution in the
catalytic cycle using the ligand-free catalytic system.
INTRODUCTION
(R = Ar) generate π-complex 4 (Scheme 1b, B). In addition,
■
three alternative pathways for indole conversion may proceed
Direct arylation of C−H bonds in (hetero)aromatics by aryl
halides is an attractive synthetic methodology for obtaining
building blocks for the synthesis of biologically active
/
as follows: (i) formation of σ-complexes 5 and 5 via Heck-
/
type carbopalladation (Scheme 1b, C, C ); (ii) formation of σ-
/
1
complexes 7 and 7 via electrophilic substitution (Scheme 1b,
substances. Despite significant progress in synthetic proce-
/
/
/
D, F and D , F ); (iii) formation of σ-complex 7 via concerted
nonelectrophilic metalation−deprotonation (CMD, Scheme
dures using direct arylation, many “blank spots” remain from a
mechanistic viewpoint. To date, a comprehensive description
of the reaction mechanism of direct C−H arylation of
heteroaromatics has not been reported. To the best of our
knowledge, all mechanistic hypotheses that have been
proposed so far can be categorized in two distinct mechanistic
proposals. We attempted to present all mechanistic possibilities
using the most common mechanistic scheme (Scheme 1).
There is a general consensus regarding the participation of aryl
halides in the direct arylation of heteroaromatics via oxidative
addition to Pd(0) (Scheme 1a, A). Therefore, both
mechanistic hypotheses for the formation of C2- and C3-
arylated indoles being the most common products of the
reactions, including C−H activation of indole molecule with
/
2,3
1
b, E ) that should lead to the prior formation of C2-
palladated intermediates following the formation of the C2-
3
,5,6
/
/
arylated products.
Intermediates 5, 5 or 7, 7 are then
/
converted to C2- or C3-arylated products (Scheme 1d, H, H
or I, I ) accompanied by the regeneration of Pd(0) 1, which
/
can react with aryl halide in the next catalytic cycle (Scheme
1
a, A). Efforts to verify the noncooperative mechanism,
including the alternative possibilities of indole activation
/
/
/
(
Scheme 1b, C, C , D, D , E ), have been reported for several
types of aromatic and heteroaromatic substrates, and
3
,5−11
conflicting results have been obtained.
The second mechanistic proposal includes the cooperation
of two distinct metal fragments generated in merging catalytic
cycles, as proposed by Hartwig et al. based on the results of
2
−4
aryl halide,
include the oxidative addition of aryl halides to
Pd(0). If only one regioisomer formation is considered (for
instance, the C2-arylated indole), then the first variant of the
mechanism involves one of the following sequences: A ⇒ B ⇒
12
extensive study of the direct arylation of pyridine N-oxide.
This proposal was supported by the studies of the direct
1
3
/
/
/
/
/
/
/
arylation of simple arenes, as well as their oxidative
C ⇒ H or A ⇒ B ⇒ D ⇒ F ⇒ I or A ⇒ B ⇒ E ⇒ I
Schemes 1a, b, and d). All the sequences mentioned apply to
14
coupling. When considering the formation of only one
(
the so-called noncooperative reaction mechanism (can be also
called a linear mechanism from a kinetic viewpoint).
Accordingly, the indole and product of oxidative addition 3
Received: April 12, 2018
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1. Alternative Mechanistic Hypotheses for the Formation of C2- and C3-Arylated Products (Where L Represents
Neutral or Anionic Ligands)
regioisomer (for instance, the C2-arylated indole as before),
the second variant of the reaction mechanism includes two
the second (or intermediate between 1 and 2) reaction order
13,14
in the catalyst precursor concentration.
It should be noted
(
1
instead of one) sequences to form the catalytic cycle (Scheme
that for the reaction of indoles with aryl iodides considered in
this study, the reaction order in [Pd] was considerably higher
than unity (1.6). However, high reaction order in the catalyst
precursor concentration can arise from nonlinearity of the
catalytic cycle steps (in accordance to Hartwig’s mechanism,
the step G and G/ where two palladium-containing
intermediates react) and from nonlinear processes occurring
outside the catalytic cycle, such as the in situ formation of
catalytically active Pd nanoparticles. It should be noted here
that we did not aim to discriminate cooperative and
noncooperative mechanisms by the approach used. The kinetic
approach described herein based on the measurements of
differential selectivity allows for information to be obtained
regarding particular steps of the catalytic reaction irrespective
of any distinct hypothesized sequence of steps and exact nature
of active species.
/
/
/
/
/
/
): (i) A ⇒ G ⇒ I ; (ii) B ⇒ D ⇒ F ⇒ G or B ⇒ E ⇒
/
/
G . Intermediates 3 (R = Ar) and 7, 7 (R = X) are generated
in separate catalytic cycles and react with each other in steps G,
G (Scheme 1c), corresponding to the so-called cooperative
mechanism (can be also called as nonlinear from kinetic
viewpoint due to interaction of two different palladium-
containing intermediates). In this hypothesis, as for the
noncooperative mechanism, the aryl halide is activated by
Pd(0) complex 1 (Scheme 1a, A), and the indole is activated
by the Pd(+2) complex 3 (R = X, Scheme 1b, B). Complex 3
/
(
R = X) in this mechanism does not contain the C-bound ary/l
ligand. After the sequential transformation of 3 into 7 or 7
without aryl ligands (R = X, Scheme 1b), the reaction proceeds
with the aryl-containing complex 3 (R = Ar) (Scheme 1c, G,
/
/
G ) forming complex 7 or 7 again, but with the aryl ligand (R
=
3
Ar) accompanied by the regeneration of the Pd(+2) complex
(R = X), which can react in the next catalytic cycle with
It should also be noted that few reports have been published
regarding the nature of the active Pd complexes responsible for
product formation. A report considered ligand-free active
species in the related reaction of benzene arylation when
indole (Scheme 1b, B).
/
Complexes 7 and 7 (R = Ar) transform into C2- and C3-
/
15
arylated products and regenerate Pd(0) 1 (Scheme 1d, I, I ),
phosphine-containing catalytic systems are used. A more
16
which can react in the subsequent catalytic cycle with ArX. The
important observation supporting cooperative mechanism
proposed by Hartwig et al. is the experimental detection of
recent paper suggested the cationic character of the active
species in the direct arylation of imidazole by aryl halides and
displacement of phosphine ligands as a component of catalytic
B
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
system with active Pd complexes. In our opinion, opportunities
for mechanistic investigations, including clarification of the
character of the active species and elementary steps, can be
realized using a set of approaches based on differential
it cannot be ruled out that true catalytic species may present in
ultralow concentration being substantially lower than detection
limit of particular analytical method. Independence of
differential selectivity on the active species concentration
allows to not measuring such concentration at all because of
changes of the differential selectivity (or its invariability)
varying the reaction conditions unambiguously point to the
change (or, correspondingly, constancy) of active species
nature that does not need any additional experimental
evidence.
In addition, the identical nature of active species and
sequences of elementary steps leading to the formation of
products under similar substrates competition as well as fewer
number of steps (and, consequently, rate constants)
influencing the differential selectivity allow to considerably
simplify mathematical description of differential selectivity,
which can be important in the correct interpretation of
experimental data (see ref 17 for details). Therefore, in spite of
complication of the reaction system, the results of differential
selectivity investigations can be analyzed with more certainty.
We proposed and tested in cross-coupling reactions earlier
the approaches for determining the degree of reversibility of
the catalytic cycle steps by measuring the differential selectivity
17,18
selectivity of the reaction.
Herein, we present the results of intensive kinetic studies of
the direct arylation of indole by aryl iodides through analysis of
the differential selectivity of the reaction. Differential selectivity
was measured in two types of experiments: under competition
of two similar substrates (aryl iodides or indoles) monitoring
the products formed by competing substrates, and under
noncompetitive conditions, examining the differential regiose-
lectivity of the C2- and C3-arylated products.
EXPERIMENTAL DETAILS
General Considerations. The quantitative compositions of the
samples were determined using gas chromatography (GC)
■
(Chromatec Crystal 5000.2 instrument fitted with a flame ionization
detector [FID] and 15-m HP-5 methyl phenyl siloxane capillary
column) and GC−mass spectrometry (MS) (Shimadzu GC−MS QP-
2
010 Ultra, ionization energy of 70 eV, 0.25 μm × 0.25 mm × 30 m
GsBP-5MS column, He as the carrier gas) analyses. The recorded
mass spectra were compared with those available in the literature
(
Wiley, NIST, and NIST05 comparison libraries).
Samples for GC and GC−MS analyses were collected at different
dependence on the nature and concentrations of competing
17,20
substrates and common reagents used.
It should be noted
reaction time points. To estimate the reproducibility of the data, each
experiment was performed three times. For phase trajectories plotted,
the appropriate polynomial fitting of the experimental data was used
to be convinced in phase trajectories overlapping/changing.
that hereafter the rate ratio of competing reactions is used as a
measure of the reaction differential selectivity (see ref 17 for
extensive validation of this method).
Two alternatives (Schemes 2a and b) were considered for
the conjugation of reversible and irreversible steps in the two
competing catalytic cycles: (i) common reagent (R at Scheme
Competing Direct Arylation of Indole by Iodobenzene and
-Iodoanisole. Iodobenzene and 4-iodoanisole (1.25 mmol each),
4
indole (2.5 mmol), PdCl (0.04 mmol, 1.6 mol %), base (1.625
2
mmol), and naphthalene (0.5 mmol) as an internal standard for GC
2
) reacts in the step (Scheme 2a, Y) directly following the
and GC−MS analyses were added to 5 mL of DMF−H O mixture
2
competitive step (Scheme 2a, X); (ii) steps including the
competing substrates (Scheme 2b, X) and common reagent
(
4/1) in a glass reactor equipped with a magnetic stir bar and a
septum inlet. The reactor was placed into a preheated oil bath (140
(
Scheme 2b, Y) are separated by several other steps in the
°
C), and the reaction mixture was stirred.
Competing Direct Arylation of Indole and N-Methylindole
catalytic cycle. It can be concluded that the nature and/or
concentration of any common reagent can influence the
differential selectivity of the reaction products, P1 and P2, only
if all steps in the catalytic cycle beginning with the step where
competition occurs (step X in Schemes 2a and b) and up to
the step where the common reagent participates (step Y in
Schemes 2a and b) are substantially reversible. If even one
virtually irreversible step is present between them (any step
marked with a dashed arrow in Schemes 2a and 2b), the
common reagent cannot affect the observed differential
selectivity of the reaction products (P1 and P2) formed in
the two competing catalytic cycles. The kinetic equations
describing differential selectivity and corresponding mecha-
nistic hypotheses from the literature regarding the catalytic
cycle of direct arylation (Scheme 1) were derived (see
Supporting Information, equations S1−S13). It should be
noted that all variants of conjugation in the reversible and
irreversible catalytic cycle steps where competing substrates
participate assumed the linearity of differential selectivity with
respect to the ratio of competing substrate concentrations.
Therefore, as the initial stage of hypothesis verification, the
linearity of the differential selectivity dependence on the ratio
of concentrations of competing substrates is determined (see
Sections 3.1.1 and 3.2.1 in the Supporting Information). All
equations obtained demonstrated that influence of the nature
and/or concentration of the common reagent on the
differential selectivity is only achieved when reversible steps
exist (Schemes 2a and b). Thus, measurements of differential
by Iodobenzene. Indole and N-methylindole (1.25 mmol each),
iodobenzene (2.5 mmol), PdCl (0.04 mmol, 1.6 mol %), base (1.625
2
mmol), and naphthalene (0.5 mmol) as an internal standard for GC
and GC−MS analyses were added to 5 mL of DMF−H O mixture
2
(
4/1) in a glass reactor equipped with a magnetic stir bar and a
septum inlet. The reactor was placed into a preheated oil bath (140
°
C), and the reaction mixture was stirred.
Noncompeting Direct Arylation of Indole by Iodobenzene.
Indole and iodobenzene (2.5 mmol each), PdCl (0.04 mmol, 1.6 mol
2
%
), base (1.625 mmol), and naphthalene (0.5 mmol) as an internal
standard for GC and GC−MS analyses were added to 5 mL of DMF−
H O mixture (4/1) in a glass reactor equipped with a magnetic stir
2
bar and a septum inlet. The reactor was placed into a preheated oil
bath (140 °C), and the reaction mixture was stirred.
RESULTS AND DISCUSSION
■
Competing reaction methods are widely used as a mechanistic
19
tool for investigation of complex reactions. In spite of visible
complications of the reaction arising when additional substrate
is introduced, the artificial multiroute character with measure-
ments of differential selectivity have distinct advantages over
noncompeting one-route reactions with corresponding meas-
urements of catalytic activity (for a detailed description of the
advantages, see ref 17). Thus, experimental detection and
kinetic control of active species is usually complicated task
that, in addition, cannot be solved unambiguously due to a role
of species detected by any physic-chemical technique can be as
positive (catalytically active) as negative (inactive). In addition,
C
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
17
Scheme 2. Catalytic Reactions between Common Reagent
regarding the concentration of reactants. The approach
consists of plotting so-called phase trajectories of competing
reaction considering the dependences of one product yield
versus another. The slope of the tangent to any point of the
phase trajectory represents the ratio of the rates of competing
reactions, which is also the parameter describing the
differential selectivity (for a detailed substantiation, see 17).
Therefore, any discrepancy in the trajectories of the reaction
when a parameter is changed unambiguously indicates the
differential selectivity variation. On the other hand, coinci-
dence of phase trajectories indicates a constant differential
selectivity at each moment of the reaction proceeding. It is
important that phase trajectories coincidence (i.e., have the
same differential selectivities) becomes possible along with
considerable changes in the rates of competing reaction
product formation resulting from changes in active species
concentration. Therefore, by comparing phase trajectories, it is
possible to estimate the influence of individual factors on the
differential selectivity of the reaction. When two similar
substrates (i.e., aryl iodides or indoles) were used (Scheme
(
R) and Two Similar Substrates (Sub1 and Sub2)
Competing for the Common Catalyst (Xcom) and Formation
of Products P1 and P2
a
3
) to measure the differential selectivity of competing
Scheme 3. Competing Direct Arylation of (a) Indole by
Iodobenzene and 4-Iodoanisole and (b) Indole and N-
Methylindole by Iodobenzene
a
The common reagent (R) (loaded to the reactor or generated in situ
by the additional catalytic cycle (marked by grey) participates (a) in
the catalytic cycle step directly following the step where Sub1 and
Sub2 compete or (b) in the catalytic cycle step that is distant from the
Sub1/Sub2 competition step. Dashed arrows indicate the step
reversibility where the nature and/or concentration of reagent R
can influence the differential selectivity of the reaction products.
selectivity of competing reactions with varying natures and
concentrations of common reagents can yield information
about the degree of reversibility of particular catalytic cycle
steps.
It should be noted that this method is valid for verifying
mechanistic hypotheses irrespective of the number of catalytic
cycles (only one for noncooperative mechanism or several
cycles for cooperative mechanism). In the latter mechanism,
the common reagent is not loaded into the reactor but is
generated in situ from the additional catalytic cycle (marked by
gray in Schemes 2a and b). If any changes occur in this cycle
under the varying reaction conditions resulting in changes of
concentration and/or nature of the generated common
intermediate, the influence on the reaction differential
selectivity becomes possible only if the all steps in the
competing catalytic cycles (beginning at the steps where
competition occurs up to the steps where the generated
common intermediate participates) are substantially reversible,
as for the mechanistic hypotheses based on the noncooperative
mechanism.
substrates, the phase trajectories could be plotted using the
sums of the corresponding C2- and C3-arylated products. The
differential selectivity of the regioisomer formation can be
measured by plotting the phase trajectories using the
corresponding C2- and C3-arylated products in competing
and noncompeting reaction conditions.
Differential Selectivity in the Competition Reaction
of Two Aryl Halides with Varying Concentration and
Nature of the Indole. All direct arylation mechanistic
hypotheses include the oxidative addition of the aryl halide to
palladium (A). The virtually irreversible oxidative addition in
1
3
It should be noted also that measurement of differential
selectivity is a complicated task because of the differentiation
of integral data regarding substrates or products concentrations
needs for obtaining reaction rates at each reaction moment.
The problem can be solved by using special equipment that
direct arylation of simple arenes and cross-coupling
17,20
reactions
was previously demonstrated. However, for the
direct arylation of heteroaromatics, irreversibility has not yet
been unequivocally demonstrated. To determine the reversi-
bility of the oxidative addition step, a set of experiments using
competing iodobenzene and 4-iodoanisole with varying nature
and concentrations of common reactants was performed
(Scheme 3a). To measure the differential selectivity of aryl
halides, the phase trajectories should be plotted using the sums
of C2- and C3-arylated indoles forming from each competing
2
1
allows for the direct measurement of reaction rates.
However, this equipment is expensive and cannot measure
the rates of consumption/accumulation of several substrates/
products. A more simple approach, allowing for the estimation
of differential selectivity is achieved using integral kinetic data
D
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
aryl halide. It means that C2- and C3-arylated products
formation proceeds after the selectivity-determining step where
aryl halides compete (regularities of the reaction selectivity in
C2- and C3-arylated indoles see below in corresponding
section).
As in all mechanistic hypotheses (Scheme 1), the step
directly following (A) includes the interaction of an
arylpalladium intermediate 3 (R = Ar) with indole (common
reagent, noncooperative mechanism (B), Scheme S1 in the
/
Supporting Information) or with intermediates 7, 7 (R = X)
generated in the additional catalytic cycle (common reagent,
/
cooperative mechanism G, G ) Scheme S2). Both variants are
described by Scheme 2a and eqs S1−S3. According to eq S1
and eq S2, the reversibility of the catalytic cycle steps where
competing aryl halides participate (A) becomes possible only
when the concentration and/or nature of the common reagent
Figure 1. Phase trajectories of the reaction of indole with competing
iodobenzene and 4-iodoanisole (Scheme 3a) varying the base nature
(
PdCl , DMF−H O = 4/1, 140 °C, reaction time 4−6 h).
2
2
(
indole in noncooperative or indole-containing intermediates
, 7 (R = X) in cooperative mechanisms) influences
/
7
differential selectivity. To assess the possible reversibility of
aryl halide oxidative addition, experiments with varying
concentrations of indole (0.5−1 M) and its nature (indole
was substituted with 5-methoxyindole) were performed. For all
experiments, the phase trajectories coincided (Figure S2 in the
Supporting Information), indicating independence of differ-
ential selectivity on concentration and nature of the common
reagent, validity of eq S3, and virtual irreversibility of the
oxidative addition (A). The conclusion is valid not only for
noncooperative but also for cooperative mechanisms because
changes in the concentration and/or nature of the indole
Interestingly, the use of acetates and carbonates of Na and K
suggested that differential selectivity is sensitive only to the
anion of the base and not to the cation. Based on the data
obtained upon varying the nature and concentration of the
indole, virtual irreversibility of the oxidative addition was
established. Consequently, there is a minimum of one virtually
irreversible step (i.e., oxidative addition) between the aryl
halides competition step (A) and those in which the base
/
/
/
participates (F, F , E , H, H ). Thus, the base cannot influence
the differential selectivity of the competing aryl halides because
of the reversibility of the catalytic cycle steps. The observed
changes in differential selectivity of indole arylation with
competing aryl halides upon changing the anion of the base
can be explained by changes in the nature of the active species.
The base anions enter the coordination sphere of the Pd(0) 1
complex participating in oxidative addition (A). The possibility
of forming a Pd(0) anionic complex in related cross-coupling
reactions was demonstrated under model experimental
(
common reagent in noncooperative mechanism) should lead
to changes in the concentration/nature of the indole-
containing intermediates 7 and 7 (R = X) (common reagent
/
in cooperative mechanism), which reacts with intermediates of
the competing catalytic cycles. Thus, lack of such influence is
in agreement with the virtual irreversibility of the oxidative
addition. Note, that despite complete coincidence of the phase
trajectories (and, consequently, the differential selectivities),
22,23
conditions by Amatore and Jutand
and under real catalytic
−
2
24
the reaction rates considerably differed (from 1.1 × 10 to 2.0
conditions. In such case, the nature of the active species and,
consequently, the differential selectivity should be sensitive to
presence of other anions, including those with low basicity that
can coordinate to Pd(0). Indeed, even when only moderate
amounts of NaBr or NaCl (20 equiv to Pd) were added to the
reaction vessel, differential selectivity was changed slightly but
reproducibly (each experiment was repeated for three times).
Changes of the phase trajectories were confirmed by the fact
that different polynomials were needed to describe them
(Figure 2). In our opinion, this indicated the entrance of the
halide anion presenting in the catalytic system or forming from
aryl halide conversion (endogenous halide anions) in the active
Pd complex that can be either soluble molecular compounds or
situated at the surface of heterogeneous catalyst (including
nanoparticles) and its participation in the catalytic cycle step
where two aryl halides competed, i.e., their oxidative addition
to Pd(0) complexes.
−
2
×
10 M/min). These results highlight the advantages of
using differential selectivity over catalytic activity measure-
ments, which are traditionally used for kinetic studies.
Differential Selectivity in the Competition Reaction
of Two Aryl Halides with Varying Concentration and
Nature of the Base. Because the direct arylation of indole
requires a base to proceed, it should be considered as an
essential common reagent for the reaction with aryl halides
competition. According to the mechanistic hypotheses
(
Schemes 1b and c), in the noncooperative mechanism, the
base is necessary in the catalytic cycle step that proceeds after
/
/
/
the indole enters the catalytic cycle (F, F , E , H, H ) and does
not participate in oxidative addition (A). As per Hartwig’s
hypothesis, changes in differential selectivity with competing
aryl halides could arise from changes in the concentration/
/
nature of common intermediate 7 or 7 (R = X) (Scheme S2)
due to variation in the base concentration/nature. Thus, the
possible influence of the base concentration and nature on the
differential selectivity with competition of two aryl halides was
investigated. The phase trajectories did not change with
varying concentration of NaOAc (Figure S3) but changed
when it was replaced with carbonates (Figure 1). These
changes were slight but reproducible (each experiment was
repeated for three times). In addition, the phase trajectories
obtained needed different polynomials to be described.
Differential Selectivity in the Competition Reaction
of Two Indoles with Varying the Concentration and
Nature of Aryl Halide. All mechanistic hypotheses for direct
arylation include π-complex formation via reaction of
intermediate 3 with the indole molecule (B). Experiments
with competing indole and N-methylindole (Scheme 3b) with
varying nature and concentrations of the common reactants
were performed. To measure differential selectivity of indoles,
the phase trajectories should be plotted using the sums of C2-
E
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Nature of the Base. In accordance with all mechanistic
hypotheses, the base (the common reagent) participates in a
/
/
step (F, F or H, H ) separate from that where the two indoles
compete (B). This is described by Schemes 2b and S3 and eqs
S4−S6. Additionally, for the mechanistic hypothesis including
/
the CMD step, the base participates in step (E ) directly
following the step where the two indoles compete (B), as
described in Schemes 2a and S5 and eqs S7−S9. It should be
emphasized that the influence of the nature of the base on
differential selectivity can arise from changes in the additional
catalytic cycle where the common reagent is generated for the
cooperative mechanism (Scheme S4). However, if changes in
the base lead to changes in the concentration of common
reagent 3 (R = Ar), it would be unable to influence the
differential selectivity of the competing indoles. Otherwise, a
change in differential selectivity of the competing indoles
should be observed under varying aryl halide concentration
and nature (see above section). Consequently, if the base
influences the steps of the additional catalytic cycle where
common intermediate 3 forms (sequence of the steps A ⇒ G
Figure 2. Phase trajectories of the reaction of indole with competing
iodobenzene and 4-iodoanisole (Scheme 3a) using NaOAc as the
base without (●) and with additives of NaCl (▲) or NaBr (□) (20
equiv to Pd) (PdCl , DMF−H O = 4/1, 140 °C, reaction time 3−6
2
2
h).
and C3-arylated indoles forming from each competing indole.
As for two aryl halides competition C2- and C3-arylated
products formation proceeds after the selectivity-determining
step where indoles compete which, accordingly to all
mechanistic hypotheses, is the π-complex formation (B)
/
/
⇒
I or A ⇒ G ⇒ I ), no influence on the proceeding of the
cycles where indoles compete for 3 (R = X) should be
observed. Therefore, all the above reasoning is valid for both
the cooperative and noncooperative mechanisms.
Differential selectivity was not sensitive to base concen-
tration (Figure S6) but was extremely sensitive to its nature
(
regularities of the reaction selectivity in C2- and C3-arylated
indoles, see below in corresponding section).
(
Figure 3). Using carbonates instead of acetates led to the
In accordance with all mechanistic hypotheses (Scheme 1b),
the step directly following that where indoles participate (B)
does not include the interaction of the π-complex 4 with aryl
halide (common reagent, noncooperative mechanism, Scheme
S3) or with intermediate 3 (R = Ar) generated in the
additional catalytic cycle (common reagent, cooperative
mechanism, Scheme S4). Both variants are described by
Scheme 2b and eqs S4−S6. According to eqs S4 and S5, the
considerable reversibility of all the steps in the catalytic cycle
between the indole competition (B) and the step where the
common reagent (aryl halide or intermediate 3 (R = Ar))
/
participates (A or G, G , respectively) influences differential
selectivity. By varying the concentration of the aryl halide
(
0.25−0.75 M) and its nature (iodobenzene was replaced by 4-
iodoanisole), it was shown that the phase trajectories coincided
Figure S5). This indicates the independence of differential
(
selectivity with respect to the concentration and nature of the
aryl halide and validates eq S6. The results also indicate the
presence of at least one virtually irreversible step in the
catalytic cycle between the steps where the indole and
common reagent (or common intermediate) participate
Figure 3. Phase trajectories of the reaction of iodobenzene with
competing indole and N-methylindole (Scheme 3b) varying the base
nature (PdCl , DMF−H O = 4/1, 140 °C, reaction time 4−6 h).
2
2
(
Scheme 2b). This agrees with the possible irreversibility of
formation of phenylindoles dominating compared to that of N-
methylphenylindoles. The differential selectivity was also
dependent on the base anion but not on its cation, as
observed for the aryl halide competition. The observed
changes in differential selectivity with variation of the base
anion can be explained by changes of the nature of the active
species through the coordination of base anions to Pd(+2)
complex 3. The complex 3 then reacts with the competing
indoles (B), similar to the above-mentioned situation where
the base influenced the reaction selectivity of the aryl halides
competition. The differential selectivity could also be
influenced by the reversibility of the catalytic cycle steps
between the indoles competition and base participation. In the
former case the nature of active species should be sensitive to
the presence of other anions in the catalytic systems, including
those with low basicity that can coordinate to Pd(+2). Indeed,
/
/
one or more steps (C − I, C − I ). In the noncooperative
mechanism, the irreversible step(s) should be where stable
/
/
arylated indoles are formed (H, H , or I, I ). However, in
Hartwig’s mechanism (cooperative mechanism), the irrever-
sible steps should be accompanied by the formation of arylated
products realized in the additional catalytic cycle (where the
common intermediate is generated) separate from the
competing catalytic cycles (Scheme S4). Consequently, phase
trajectory coincidence in Hartwig’s mechanism should arise
from the irreversibility of another step. As such, the virtually
irreversible step with excess base is the formation of σ-
/
/
complexes 7 and 7 via the CMD step (E ) or electrophilic
/
substitution (F, F ).
Differential Selectivity in the Competition Reaction
of Two Indoles with Varying the Concentration and
F
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
it was established that adding halide salts to the indole
competition reaction resulted in changes in the differential
selectivity (Figure 4). This indicates the presence of halide
Regioselectivity of the C2-/C3-Arylated Indoles in the
Reaction of Indole with Iodobenzene. Verification of the
hypotheses proposing the reversibility of the C2- and C3-
phenylated indole formation steps was performed using
analysis of the differential selectivity of C2- and C3-arylated
indoles formed from one indole and one aryl halide (instead of
on the sums of the products formed from two competing
indoles (Schemes 4 and S6)). It should be noted that using
Scheme 4. Noncompeting Direct Arylation of Indole by
Iodobenzene
Figure 4. Phase trajectories of the reaction of iodobenzene with
competing indole and N-methylindole (Scheme 3b) using Na CO as
competing conditions for this purpose should lead to the same
differential selectivity of the C2- and C3-arylated indoles
obtained under noncompeting conditions in the absence of the
second indole. This assumption was experimentally verified by
the good coincidence of the phase trajectories of the C2- and
C3-phenylated indoles under competing and noncompeting
conditions (Figure S7). The result also suggests the absence of
influence of the second indole (used under competing
conditions) on the formation of the active species responsible
for the parallel formation of the C2- and C3-phenylindoles.
In contrast to the indoles competition, where the formation
of π-complex 4 (B) is the selectivity-determining step, the
differential selectivity in noncompeting reaction is determined
by the parallel formation of C2- or C3-arylated σ-complexes 5,
2
3
the base without (●) and with additives of NaCl (
◊
), NaBr (△), KCl
(
4
⧫), KBr (▲), and NaI (○) (40 equiv to Pd) (PdCl , DMF−H O =
2
2
/1, 140 °C, reaction time 5−8 h).
anions (including endogenous iodide anions released from aryl
iodide conversion) in the Pd(+2) active complex 3 that reacts
with the indole molecule (B). Moreover, it was revealed that
the differential selectivity of the indoles competition reaction
was sensitive to the nature of the halide salt cation (Figure 4,
experiments with additives NaBr and KBr). Because the alkali-
metal cation cannot directly coordinate with the Pd center, the
most reasonable explanation of their influence on the relative
reactivity of the active species becomes possible when these
species are anionic through electrostatic interaction in the ion
pairs formed (i.e., tight ion pairs). Therefore, the rationale for
the influence of both the cation and anion of the inorganic salt
on the reaction selectivity when two indoles compete can be
the anionic nature of active Pd species (that also can be either
soluble molecular compounds or situated at the surface of
heterogeneous catalyst) in the step where indole reacted (B).
It should be noted that the anionic character of the Pd(+2)
active species was previously reported for the catalytic systems
/
/
/
/
5
, 6, 6 , or 7, 7 (depending on the hypothesis, steps C, C , or
/
/
D, D , or E ). Measurements of differential selectivity of the
C2- and C3-arylated indoles upon varying the nature of the
base allowed the determination of the (ir)reversible character
of the step where base reacts as a common reagent directly
following the parallel formation of C2- and C3-arylated
intermediates. Thus, it is valid for the hypotheses, including
carbopalladation (C, C ) or electrophilic substitution (D, D )
as described by Scheme S6 and eqs S10, S12, and S13. The
CMD step reveals that the base directly influences the rates of
the parallel formation of C2- and possible C3-arylated
/
/
25
of the related Mizoroki−Heck reaction under model and
24
/
catalytic conditions. At first glance, participation of anionic
Pd complexes in the steps where indole reacts as C-nucleophile
can seem not so probable (from formal viewpoint). However,
it was demonstrated earlier that anionic Pd(+2) complexes
were active ones in related cross-coupling reactions where
electrophilic interaction between anionic Pd(+2) and C-
intermediates (Scheme 1b, E , Scheme S6) that are described
by eqs S11−S13. The analysis of differential selectivity of C2-
and C3-products in the noncompeting reaction of iodobenzene
and indole indicated that it was dependent on the nature of the
base and, more specifically, on the nature of both its anion and
cation (Figure 5). It should be noted that when Na and K
carbonates were used, the differential selectivity significantly
changed during the reaction. Initially, the formation of the C3-
arylated product dominated, and the selectivity clearly
switched over to subsequently favor the formation of the C2-
arylated product. As it was observed for differential selectivity
in the indoles competition, the use of halide salt additives
influenced the differential selectivity of the C2-/C3-phenyl-
indoles (Figures 6 and S8). Additionally, increasing KBr when
K CO was used as the base resulted in a gradual reduction of
2
2
24,25
(
sp )C(sp ) bond realized.
Of course, one can suppose
that cationic or neutral Pd(+2) complexes may be more
reactive in such reactions; however, the regularities of
differential selectivity indicated that due to the presence of
endogenous and exogenous (added to catalytic cycle) halide
anions in the catalytic system, cationic and neutral Pd(+2)
complexes just cannot exist, and the reaction proceeded via
anionic species.
In contrast to low-basic halide anions, the influence of base
anions on the differential selectivity (Figure 3) of the two
indoles competition results from the effect on the nature of the
active species in step B. Also, the reversibility of the catalytic
cycle steps where the C2- and C3-phenylated indoles are
formed can influence the differential selectivity.
2
3
the curve of the phase trajectories where the formation of the
C3-arylated product dominated. The addition of 40 equiv of
NaI (using Na CO as the base) led to substantial suppression
2
3
of the reactions forming the C3-arylated product, and the
formation of the C2-arylated product dominated the reaction
G
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
conformed to the differential selectivity changes over the
course of the reaction where carbonate bases were used
without addition of halide salts (Figure 5). Here, the
endogenous iodide anions released as the result of aryl halide
conversion caused changes in the coordination sphere of the
Pd active complexes that led to increasing relative rate of C2-
arylated product formation (manifesting as an increase of the
phase trajectory slope). The initial dominant C3-arylated
product formation contradicts the mechanistic hypothesis that
includes the direct metalation of the indole via a non-
electrophilic CMD because it should lead to the prior
/
formation of C2-palladated intermediates (Scheme 1, E )
3
,5,6
followed by the formation of C2-arylated products.
Therefore, the reaction mechanism proceeding through the
CMD step is unlikely.
Figure 5. Phase trajectories of the reaction of iodobenzene and indole
(
Scheme 4) plotted by using concentrations of C2- and C3-
phenylindoles varying the base nature (PdCl , DMF−H O = 4/1,
The observed changes of differential selectivity upon
changing the base cation (Figure 5) suggest that an anionic
2
2
/
/
140 °C, reaction time 4−6 h).
Pd(+2) active species participates in steps C, C or D, D . To
confirm that the base influence can originate from the
reversibility of the catalytic cycle steps also (Scheme 2a), it
is necessary to exclude the influence of the base as a ligand in
the Pd(+2) active species. This can be achieved by using
concentrations of additives of halide salts that are able to fully
displace the base anions from the Pd coordination sphere. The
experimental data indicate good stability of the halide
palladium complexes compared to those formed with acetate
(
under similar conditions of phosphine-free cross-coupling
23,26
reactions).
Additionally, taking into account the dependence of
differential selectivity on the nature of cations present
(
possibly resulting from formation of tight ion pairs with
anionic Pd(+2) active complexes), any difference in base and
additive salt cation should also be excluded. Therefore, we
compared differential selectivity of the C2- and C3-phenyl-
indoles in noncompeting indole arylation by iodobenzene by
adding K CO and KOAc as the base with excess of KBr
Figure 6. Phase trajectories of the reaction of iodobenzene and indole
Scheme 4) plotted by using concentrations of C2- and C3-
(
phenylindoles employing K CO as the base without additives (▲)
2
3
and with 20 (■), 40 (●), 80 (
base without additives (△) and with 80 equiv of KBr (◊) (PdCl ,
⧫
) equivalents of KBr and KOAc as the
2
3
additive (Pd/additive/base = 1/80/65). The data obtained
indicated that under conditions where all base anions should
be displaced from the Pd coordination sphere by bromide ions,
the differential selectivity differed when various bases were
used (Figure 6). This indicates the reversibility of the catalytic
cycle step where parallel formation of C2- and C3-arylated
indoles proceeds (Schemes 2a and S6 as well as eq S12).
2
DMF−H O = 4/1, 140 °C, reaction time 3−6 h).
2
from the beginning (Figure S8). These observations indicated
that the changes in the nature of the active species, namely by
additive and endogenous iodide ions coordinating the Pd
active complexes, induced major selectivity toward the C2-
arylated indole. The effect of endogenous iodide ions
/
Considering the low probability of the CMD pathway (E )
due to the domination of C3-arylated products initially during
Table 1. Summarized Data of Experiment Sets on the Influence of Common Reagents and Additives on the Differential
a
Selectivity and the Conclusions Drawn
conclusions
conditions/differential selectivity
common reagent and
additives (influence)
reversible or
irreversible
active
species
row
1
selectivity-determining step
oxidative addition (Scheme 1a, A)
measurement
competing (two aryl iodides)/on sums of indole (−) base (+)
arylated indoles
irreversible
anionic
salt (+)
2
indole coordination (Scheme 1b, B)
competing (two indoles)/on sums of
arylated indoles
aryl iodide (−) base
impossible to
make a
conclusion
anionic
b
(+) salt (+)
c
/
d
b
3
4
electrophilic substitution (Scheme 1b, D, D ) or Heck-type
carbopalladation (Scheme 1b, C, C ) or CMD (Scheme 1b, E )
noncompeting and competing (two
indoles)/on C2- and C3-regioisomers
base (+) salt (+)
reversible
reversible
anionic
anionic
/
/
/
d
b
electrophilic substitution (Scheme 1b, D, D ) or Heck-type
carbopalladation (Scheme 1b, C, C ) or CMD (Scheme 1b, E )
noncompeting and competing (two aryl base (+) salt (+)
iodides)/on C2- and C3-regioisomers
/
/
a
b
c
The extensive discussion of these conclusions can be seen in appropriate sections of the paper. Influence of both anion and cation. Making a
conclusion is impossible due to the competing steps and steps where common reagent participants are separated by several steps (Scheme 2b).
d
Most likely because it is in accordance with the revealed reversibility.
H
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
the reaction (Figures 5 and 6), other hypotheses regarding the
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.8b00216.
■
/
mechanism of C2- and C3-arylated product formation (C, C ,
*
S
/
or D, D ) are more plausible. Based on the results obtained the
electrophilic substitution of the hydrogen atom in indole (D,
/
D ) seems to be the most probable. Heck-type carbopallada-
/
tion (C, C ) accompanied by the formation of a new C-C bond
Mechanistic schemes, differential selectivity equations,
as a result of addition of an aryl moiety to the carbon atom of
the pyrrole ring under relatively mild reaction conditions does
not possesses a sufficient degree of reversibility, implying C
C bond breakage. Thus, if the reaction proceeds through
virtually irreversible carbopalladation, the base reacting in the
and experimental dependences (PDF)
AUTHOR INFORMATION
Corresponding Author
E-mail: aschmidt@chem.isu.ru.
■
*
/
following steps (H, H ) cannot influence the ratio of the
reaction rates of the parallel formation of C2- and C3-arylated
indoles. Consequently, the base would also not be able to
influence the differential selectivity of the C2- and C3-arylated
indoles under the reaction conditions when the base influence
on the active species is excluded. Therefore, the data obtained
allow discriminating of hypothesis including direct arylation
proceeding through Heck-type carbopalladation.
ORCID
Alexander F. Schmidt: 0000-0002-1881-7620
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Russian Foundation for Basic
Research (Grant 16-29-10731_ofi_m).
CONCLUSION
■
REFERENCES
In summary, analyses of differential selectivity of competing
and noncompeting reactions determined using phase trajecto-
ries of the reactions irrespective of the proposed mechanistic
hypotheses (cooperative and noncooperative) were performed.
These experiments allowed establishing the degree of
reversibility of selectivity-determining catalytic cycle steps
and the type of active species involved. The summarized
experimental data and the corresponding conclusions based on
them are presented in Table 1 (the extensive discussion of
these conclusions see in appropriate sections of the paper). For
the direct arylation of indoles by aryl iodides, it was
demonstrated that oxidative addition of the aryl halide is
virtually irreversible (row 1, Table 1), while the parallel
formation of C2- and C3-arylated indoles possesses substantial
reversibility (rows 3 and 4, Table 1). Considering proposals
regarding the reaction mechanism, the results presented herein
are fully consistent with the electrophilic substitution of the
indole. In contrast, mechanisms including irreversible Heck-
type carbopalladation are unlikely. Detecting changes in
differential selectivity during the reaction indicated the
domination of C3-arylated product formation at the reaction
beginning. This observation excluded mechanisms involving
direct metalation of the indole via nonelectrophilic CMD. The
type of active species participating in the selectivity-
determining step was also elucidated. The dependence of
differential selectivity of the competing aryl halides on the base
and halide anions (added to the catalytic system or
endogenous anions formed by aryl halide conversion, row 1,
Table 1) indicates the anionic nature of the Pd(0) active
complexes. The dependence of the differential selectivity of
competing indoles (row 2, Table 1) or of parallel C2-/C3-
arylated product formation (rows 3 and 4, Table 1) on the
cation and anion of the base or halide salt indicates the anionic
character of the active Pd(+2) species containing anions from
base and/or added halide salt in its coordination sphere. These
active species participate in the catalytic cycle steps where
activation of indole occurs (rows 2−4, Table 1).
■
(
1) Rossi, R.; Lessi, M.; Manzini, C.; Marianetti, G.; Bellina, F.
Synthesis 2016, 48, 3821−3862.
(
2) Bheeter, C. B.; Chen, L.; Soule, J. F.; Doucet, H. Catal. Sci.
Technol. 2016, 6, 2005−2049.
3) Lane, B. S.; Brown, M. A.; Sames, D. J. Am. Chem. Soc. 2005,
127, 8050−8057.
4) Grimster, N. P.; Gauntlett, C.; Godfrey, C. R. A.; Gaunt, M. J.
Angew. Chem., Int. Ed. 2005, 44, 3125−3129.
5) Sun, H.-Y.; Gorelsky, S. I.; Stuart, D. R.; Campeau, L.-C.;
Fagnou, K. J. Org. Chem. 2010, 75, 8180−8189.
6) Joucla, L.; Batail, N.; Djakovitch, L. Adv. Synth. Catal. 2010, 352,
2929−2936.
7) Potavathri, S.; Pereira, K. C.; Gorelsky, S. I.; Pike, A.; LeBris, A.
P.; DeBoef, B. J. Am. Chem. Soc. 2010, 132, 14676−14681.
8) Gemoets, H. P. L.; Kalvet, I.; Nyuchev, A. V.; Erdmann, N.;
Hessel, V.; Schoenebeck, F.; Noel, T. Chem. Sci. 2017, 8, 1046−1055.
9) Colletto, C.; Islam, S.; Julia-Hernandez, F.; Larrosa, I. J. Am.
(
(
(
(
(
(
(
Chem. Soc. 2016, 138, 1677−1683.
(10) Campeau, L.-C.; Parisien, M.; Jean, A.; Fagnou, K. J. Am. Chem.
Soc. 2006, 128, 581−590.
(
11) Theveau, L.; Querolle, O.; Dupas, G.; Hoarau, C. Tetrahedron
2013, 69, 4375−4380.
(
12) Tan, Y.; Barrios-Landeros, F.; Hartwig, J. F. J. Am. Chem. Soc.
2012, 134, 3683−3686.
(13) Kim, J.; Hong, S. H. ACS Catal. 2017, 7, 3336−3343.
(14) Wang, D.; Izawa, Y.; Stahl, S. S. J. Am. Chem. Soc. 2014, 136,
9
(
(
914−9917.
15) Tan, Y.; Hartwig, J. F. J. Am. Chem. Soc. 2011, 133, 3308−3311.
16) Perego, L. A.; Grimaud, L.; Bellina, F. Adv. Synth. Catal. 2016,
358, 597−609.
(17) Schmidt, A. F.; Kurokhtina, A. A.; Larina, E. V. Catal. Sci.
Technol. 2014, 4, 3439−3457.
(18) Cano, R.; Schmidt, A. F.; McGlacken, G. P. Chem. Sci. 2015, 6,
5
(
338−5346.
19) Bunnet, J. P. In Techniques of Chemistry: Investigation of Rates
and Mechanisms of Reactions. General Considerations and Reactions at
Conventional Rates, 3rd ed.; Lewis, E. S., Ed.; Wiley: New York, 1974;
Vol. 6, pp 129−209.
(20) Kurokhtina, A. A.; Larina, E. V.; Yarosh, E. V.; Schmidt, A. F. J.
Mol. Catal. A: Chem. 2016, 425, 43−54.
(
(
21) Blackmond, D. G. Angew. Chem., Int. Ed. 2005, 44, 4302−4320.
22) Amatore, C.; Carre, E.; Jutand, A.; M’Barki, M. A.; Meyer, G.
Organometallics 1995, 14, 5605−5614.
I
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
23) Amatore, C.; Jutand, A. J. Organomet. Chem. 1999, 576, 254−
78.
24) Schmidt, A. F.; Kurokhtina, A. A.; Larina, E. V.; Yarosh, E. V.;
Lagoda, N. A. Organometallics 2017, 36, 3382−3386.
25) Carrow, B. P.; Hartwig, J. F. J. Am. Chem. Soc. 2010, 132, 79−
1.
26) Schmidt, A. F.; Kurokhtina, A. A.; Larina, E. V. Russ. J. Gen.
Chem. 2011, 81, 1573−1574.
2
(
(
8
(
J
DOI: 10.1021/acs.organomet.8b00216
Organometallics XXXX, XXX, XXX−XXX