Isolation and Characterization of Regioisomers of Pyrazole-Based Palladacycles and Their Use in α-Alkylation of Ketones Using Alcohols

Article abstract of DOI:10.1021/acs.organomet.7b00478

Regioisomers of 3,5-diphenyl-1-(4-(trifluoromethyl)phenyl)-1H-pyrazole-based palladacycles (1 and 2) were synthesized by the aromatic C-H bond activation of N/3-aryl ring. The application of these regioisomers as catalysts to enable the formation of α-alkylated ketones or quinolines with alcohols using a hydrogen borrowing process is evaluated. Experimental results reveal that palladacycle 2 is superior over palladacycle 1 to catalyze the reaction under similar reaction conditions. The reaction mechanisms for the palladacycles 1 and 2 catalyzed α-alkylation of acetophenone were studied using density functional theoretical (DFT) methods. The DFT studies indicate that palladacycle 2 has an energy barrier lower than that of palladacycle 1 for the alkylation reaction, consistent with the better catalytic activity of palladacycle 2 seen in the experiments. The palladacycle-phosphine system was found to tolerate a wide range of functional groups and serves as an efficient protocol for the synthesis of α-alkylated products under solvent-free conditions. In addition, the synthetic protocol was successfully applied to prepare donepezil, a drug for Alzheimer's disease, from simple starting materials.

Full text of DOI:10.1021/acs.organomet.7b00478

Article
pubs.acs.org/Organometallics
Isolation and Characterization of Regioisomers of Pyrazole-Based
Palladacycles and Their Use in α‑Alkylation of Ketones Using Alcohols
Ramesh Mamidala, Shaikh Samser, Nishant Sharma, Upakarasamy Lourderaj,*
and Krishnan Venkatasubbaiah*
School of Chemical Sciences, National Institute of Science Education and Research (NISER), HBNI, Bhubaneswar, 752050 Orissa, India
S
* Supporting Information
ABSTRACT: Regioisomers of 3,5-diphenyl-1-(4-(trifluoromethyl)phenyl)-
1H-pyrazole-based palladacycles (1 and 2) were synthesized by the aromatic
C−H bond activation of N/3-aryl ring. The application of these regioisomers
as catalysts to enable the formation of α-alkylated ketones or quinolines with
alcohols using a hydrogen borrowing process is evaluated. Experimental
results reveal that palladacycle 2 is superior over palladacycle 1 to catalyze
the reaction under similar reaction conditions. The reaction mechanisms
for the palladacycles 1 and 2 catalyzed α-alkylation of acetophenone were
studied using density functional theoretical (DFT) methods. The DFT
studies indicate that palladacycle 2 has an energy barrier lower than that of
palladacycle 1 for the alkylation reaction, consistent with the better catalytic
activity of palladacycle 2 seen in the experiments. The palladacycle−phosphine
system was found to tolerate a wide range of functional groups and serves as
an efficient protocol for the synthesis of α-alkylated products under solvent-free conditions. In addition, the synthetic protocol
was successfully applied to prepare donepezil, a drug for Alzheimer’s disease, from simple starting materials.
approach for the α-alkylation of ketones is to use alcohols as an
alkylating agent.4 Over the past decade, several groups have
reported various metal-catalyzed methods for the α-alkylation of
ketones using alcohols as an alkylating agent.5−12 For instance,
Chao and co-workers reported the α-alkylation of ketones
by trialkylamines under heterogeneous Pd/C catalysis.13
A homogeneous pincer-type Pd-NHC catalyst was reported for
the α-alkylation of acetophenone using benzyl alcohol at 125 °C,
giving a mixture of alcohol and ketone as products.14 More
recently, Beller and co-workers elegantly reported the
manganese-based catalytic system for the α-alkylation of
ketones.12 However, many of these catalytic systems suffer
from low product selectivity, low yield, and high reaction
temperature. We have also been pursuing the synthesis of
1,3,5-triphenyl-based palladacycles and their application
toward catalysis.15 We recently described the synthesis of
the 3,5-diphenyl-1-(2-(trifluoromethyl)phenyl)-1H-pyrazole-
based palladacycle and its application toward N-alkylation of
amines.15b In the study reported here, we describe the isolation
of regioisomers of palladacycles, their characterization, and
activity toward α-alkylation of ketones. As alluded vide supra,
the reactivity difference in regioisomers, especially the pallada-
cycles, has not been examined. We also discuss the activity of
regioisomers of the palladacycles as catalysts, which have been
studied by experiments and density functional theory (DFT)
calculations.
INTRODUCTION
■
Palladacycles are an important family of organometallic
compounds. The interest in the use of palladacycles has been
due to their potential applications in various fields such as
catalysis, medicinal science, and organic synthesis.1 Since the
discovery of the application of palladacycles in catalysis by
Hermann and co-workers in 1995, intense research has been
focused on their synthesis, structural analysis, and application
toward catalysis.1a Among the several available methods for the
preparation of palladacycles, ortho-metalation of C−H bonds
is an attractive methodology.2 While C−H bond activation has
been thoroughly studied, the recognition of chemically similar
C−H bonds have turned out to be an arduous task. Different
approaches have been identified to achieve C−H bond activation
of chemically similar environments.3 In many instances, the
regioisomers have been isolated and structurally characterized.
For example, Solan and co-workers demonstrated the prefer-
ential palladation of the naphthalene ring using N-donors as
directing ligands.3h Recently, Wendt and co-workers3i reported
metal-controlled regioselectivity in the directed aromatic
substitution of 2-(1-naphthyl)pyridine. Although the regioisomers
of palladacycles have been reported, in many instances, their
application for catalysis has not been examined.
α-Alkylation of carbonyl compounds is one of the most
fundamental reactions used to form a C−C bond. Traditionally,
the α-alkylation of ketones was achieved by the reaction of
ketones with electrophiles such as alkyl halides. However, this
method involves the use of toxic organohalides and formation of
a large amount of inorganic salts as waste. An alternative, greener
Received: June 23, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Palladacycle 1 (Figure 2) crystallizes in the triclinic space group
RESULTS AND DISCUSSION
■
P1, whereas palladacycle 2 (Figure 3) crystallizes in the
̅
The pyrazole ligand was readily synthesized by the condensation
of 1,3-diphenylpropane-1,3-dione and 4-(trifluoromethyl)-
phenyl)hydrazine using a reported procedure.15 The ligand
was fully characterized by 1H, 13C, 19F NMR, and HRMS. Next,
typical cyclopalladation reaction in acetic acid was carried out
to obtain the acetate-bridged dimeric complex (Scheme 1).
The resultant solid was crystallized from dichloromethane and
n-hexane. When we analyzed the crystals in a microscope,
we noticed two different types of crystals. The two forms were
then separated by fractional crystallization. The first crop was
isolated as a light yellow crystal with 39% yield (palladacycle 1),
whereas the third crop was isolated as yellowish green crystals
with 43% yield (palladacycle 2). The 1H NMR in CDCl3 showed
distinct peaks for both the isomers, notably the pyrazole
hydrogen resonates at 6.06 and 6.26 ppm for the palladacycles
1 and 2, respectively (Figure 1). The 19F NMR spectra of
palladacycles 1 and 2 show singlets at −62.4 and −63.6 ppm,
respectively. The structures of palladacycles 1 and 2 were
further unambiguously established by X-ray diffraction analysis.
orthorhombic space group Pccn. X-ray analysis reveals that
both the palladacycles 1 and 2 are dimeric in nature, bridged
together by acetates with the pyrazole ligands aligned in a
staggered fashion. The geometry around the palladium atom
in both structures exhibits a distorted square planar geometry.
The Pd−C (1.978(3) Å for 2; 1.940 (3) Å for 1) and Pd−N
(2.024(3) Å for 2; 2.015(3) Å for 1) distances are comparable
with those of the reported cyclometalated palladium com-
pounds.15 The Pd−C distance in palladacycle 1 is shorter than
the corresponding distance in palladacycle 2. However, the
Pd···Pd distance in palladacycle 2 is slightly shorter than the
Pd···Pd distance observed in palladacycle 1.
To explore the optimum reaction conditions for the alkylation
of acetophenone with benzyl alcohol, we have screened various
bases and temperatures using palladacycle 2 and the ancillary
ligand P(2-Fur)3P(C4H3O)3 under solvent-free conditions.
Of the different conditions screened, LiOH at 80 °C resulted in
a quantitative yield of the desired product using palladacycle 2
Scheme 1. Synthesis of Palladacycles 1 and 2
Figure 1. 1H NMR spectra of palladacycles 1 and 2 (only the aromatic region is shown for clarity).
B
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. Molecular structure of palladacycle 1 (30% probability thermal ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond lengths (Å)
and bond angles (degree): Pd1−C15 1.940(3), Pd1−N1 2.015(3), Pd1−O1 2.132(2), Pd1−O3 2.035(2), Pd1−Pd2 2.8885(3), C15−Pd1−N1
80.68(12), C15−Pd1−O1 176.02(11), N1−Pd1−O1 99.54(10), C15−Pd1−O3 92.91(11), N1−Pd1−O3 171.27(10), O3−Pd1−O1 87.27(9),
C15−Pd1−Pd2 97.60(9), N1−Pd1−Pd2 103.77(7), O1−Pd1−Pd2 78.47(6), O3−Pd1−Pd2 82.85(7).
Figure 3. Molecular structure of palladacycle 2 (30% probability thermal ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond lengths (Å)
and bond angles (degree): Pd1−C9 1.978(3), Pd1−N2 2.024(3), Pd1−O1 2.151(2), Pd1−O2 2.052(2), Pd1−Pd1* 2.848(5), C9−Pd1−N2
80.48(12), C9−Pd1−O1 177.01(11), N2−Pd1−O1 99.69(10), C9−Pd1−O2 93.74(12), N2−Pd1−O2 173.03(10), O2−Pd1−O1 86.29(10),
C9−Pd1−Pd1* 98.22(9), N2−Pd1−Pd1* 101.90(8), O1−Pd1−Pd1* 78.81(6), O2−Pd1−Pd1* 82.69(7).
(Table S1, entry 4, Supporting Information). However, use of
the palladacycle 1 gave relatively low yield (Table S1, entries 1
and 16, Supporting Information). When the reaction was per-
formed using commercially available palladium sources such as
Pd(OAc)2, PdCl2, and Pd2(dba)3, no product formation was
observed under similar reaction conditions (Table S1, entries 13,
14, and 15 respectively, Supporting Information).
The catalytic performances of the palladacycles 1 and 2 were
evaluated using the model substrates acetophenone and benzyl
alcohol. The reactions were performed under solvent-free
conditions at 60 °C using 1 mol % of catalyst. The catalytic per-
formances of the palladacycles 1 and 2 are compared in Figure 4.
At 60 °C using 1 mol % of catalyst loading, palladacycle 2
displayed superior activity over palladacycle 1. Although both the
Figure 4. Reactivity study of palladacycles using acetophenone and
palladacycles were active for the α-alkylation of acetophenone
benzyl alcohol at 60 °C. Reaction conditions: 0.5 mmol acetophenone,
with benzyl alcohol, we chose palladacyle 2 for further studies
0.6 mmol benzyl alcohol, 25 mol % of LiOH, 1 mol % of palladacycle
because of its superior activity over palladacycle 1. In order to
1 (or) 2, and 2 mol % of P(2-Fur)3. Conversions were analyzed using
know the catalytic efficiency of palladacycle 2, we examined a
2.0 g (16.7 mmol) scale reaction by loading 0.001 mol % of
catalyst at 130 °C for 48 h. Palladacycle 2 exhibited high turnover
number (TON) up to 32000 (66%).
Encouraged by these results, we have evaluated the substrate
scope and limitations using palladacycle 2 for the alkylation of
1H NMR by taking p-xylene as the internal standard.
various ketones using different aryl and aliphatic primary
alcohols. In most cases, electron-rich as well as electron-deficient
substituted acetophenones or benzyl alcohols were successfully
alkylated to the desired products in good to excellent isolated
C
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Table 1. Scope of the α-Alkylation of Aryl Ketones Using Aryl and Aliphatic Alcohols
a
b
Reaction conditions: 1 mmol ketone, 1.2 mmol alcohol, 0.25 mmol LiOH, 1 × 10−2 mmol palladacycle 2, 2 × 10−2 mmol P(2-Fur)3. Reactions
c
d
were performed at 100 °C for 24 h, and 2 mmol alcohol was used. Reactions were performed using palladacycle 1 for 12 h. Reactions were
performed using palladacycle 2 for 12 h. 3 mmol alcohol was used. ND = not detected.
e
yields under mild reaction conditions (Table 1, entries 1−9).
The ortho-substituted benzylic alcohols or aryl ketones (Table 1,
entries 10, 14, 15, 16, and 18) as well as aliphatic ketones (Table 1,
entry 11) were selectively alkylated to the corresponding
α-alkylated ketones. Substrates having two α-carbons that are
prone to potential alkylation generally showed selectivity issues.
Interestingly, we found that the present catalytic system offers
a regioselective alkylation in the case of 2-pentanone and
1-(4-methoxyphenyl)propan-2-one with benzyl alcohol (Table 1,
entries 12 and 13). The present catalytic system was also
successfully applied to the α-alkylation of biologically important
steroid β-pregnenolone. In general, β-pregnenolone under metal
reaction conditions is susceptible to isomerization of the CC
bond to produce progesterone, which has an α,β-unsaturated
ketone moiety.16 Under the optimized conditions, the reaction of
β-pregnenolone with 4-methyl benzyl alcohol proceeded
smoothly to give the corresponding α-alkylated product (Table 1,
entry 17) in 78% yield, without affecting the other functional
groups such as CC and −OH.
In addition, we also explored this methodology to substrates
such as aliphatic alcohols or methylene ketones at slightly higher
temperature. Aliphatic alcohols such as 1-butanol, ethanol,
2-phenyl ethanol, and 1-octanol could be used as the electrophilic
partner for obtaining the corresponding ketones (Table 2, entries
19−22). The alkylation of cyclic or methylene ketones such as
1-tetralone, 5,6-dimethoxy-1-indanone, or propiophenone resulted
in the corresponding alkylated products in good to excellent
isolated yields (Table 2, entries 23−26). Good to excellent yields
D
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Table 2. Scope of the α-Alkylation of Aryl or Heteroaryl Ketones Using Aryl or Heteroaryl Alcohols
a
b
Reaction conditions: 1 mmol ketone, 2 mmol alcohol, 0.25 mmol LiOH, 1 × 10−2 mmol palladacycle 2, 2 × 10−2 mmol P(2-Fur)3. Reactions were
c
d
performed at 100 °C for 24 h. Reactions were performed at 120 °C for 24 h. Reactions were performed at 120 °C for 48 h, and 2 mol % of
palladacycle 2 and 25 mol % of LiOtBu were used. Reactions were performed using palladacycle 1 for 12 h. Reactions were performed using
palladacycle 2 for 12 h. 3 mmol alcohol was used.
e
f
g
were also achieved in the case of heteroaryl ketones or heteroaryl
alcoholsas coupling partners atrelativelymild conditions (Table 2,
entries 27, 28, and 29). Whereas it has been a challenge to
alkylate nonactivated methylene ketones with less reactive
aliphatic alcohols, in our case, the alkylation of methylene
ketones with aliphatic alcohols was carried at moderate
temperatures (Table 2, entry 31). To increase the adaptability
of our catalytic system, we applied our strategy to estrone and
trans-dehydroandrosterone. The reaction of estrone with benzyl
alcohol and trans-dehydroandrosterone with p-methylbenzyl
alcohol proceeded smoothly to the corresponding alkylated
products in 78 and 76% yields, respectively (Table 2, entries
30 and 32). It should be noted that selective monobenzylation of
trans-dehydroandrosterone was achieved without isomerization
under the optimized reaction conditions. It is worth mentioning
that our protocol offers the α-alkylation of β-pregnenolone,
estrone, and trans-dehydroandrosterone without the use of any
protection and deprotection of other functional groups.12
Another important biologically active molecule is donepezil,
which is used in the treatment of Alzheimer’s disease.17 Using
palladacycle 2, we were able to synthesize donepezil starting from
N-benzyl-4-piperidinemethanol and 5,6-dimethoxyindanone in
moderate yield (46%, Scheme 2), which is marginally higher than
the yields reported by Glorius and co-workers6d using a Ru-based
catalytic system. We further tested the synthetic utility of the
catalyst for the alkylation of 4-piperidinemethanol with benzyl
alcohol to obtain N-benzyl-4-piperidinemethanol. The yields
were comparable or better than the yields reported by Yamada
and co-workers using an Ir-based catalytic system.18
Quinolines are an important class of heterocyclic compounds
used in the design of pharmacologically active compounds that
are used as anti-inflamatory, antimalarial, and antiasthmatic
drugs.19 They can be prepared by using conventional Friedlander
̈
annulation reaction from 2-aminobenzaldehyde and ketones.20
However, the low stability of 2-aminobenzaldehyde and self-
condensation byproduct limits the use of this method. Modified
Friedlander quinoline synthesis using more stable 2-aminobenzyl
̈
alcohol under hydrogen-borrowing conditions has been reported
as a versatile method.21 Using palladacycle 2, with the optimized
conditions for the α-alkylation of ketones mentioned above,
E
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 2. Synthesis of Donepezil and N-Benzyl-4-piperidinemethanol
Table 3. Synthesis of Quinolines Using 2-Aminobenzyl
Alcohol and Ketones
Reaction of lithium alkoxide with in situ formed phosphine-
coordinated palladium species (PPX) generated the palladium
alkoxide species (PPX-BA) which underwent β-hydride
elimination to generate the Pd−H (PPX-H) intermediate and
the aldehyde. The in situ generated base-catalyzed α,β-unsaturated
carbonyl compound accessed the Pd−H, which upon reaction
with alcohol releases the product. Control experiments were
performed to validate the proposed mechanism (Scheme S1 and
Figure S1). Reaction of acetophenone with benzaldehyde under
the optimized reaction conditions resulted in only α,β-unsaturated
compound in 95% yield. Reaction of α,β-unsaturated compound
with benzyl alcohol in the presence of palladacycle 2 (1 mol %),
P(2-Fur)3 (2 mol %), and LiOH (25 mol %) at 80 °C for 12 h
resulted in 80% of the desired product. These results support our
proposed mechanism. To probe the high activity of palladacycle
2 over 1 for the α-alkylation of ketones using alcohols, the
mechanism of the catalytic reactions was investigated using DFT
methods (Supporting Information). The free energy profiles for
the reaction between acetophenone and benzyl alcohol using
both the palladacycles (1 and 2)−phosphine PP1 and PP2
were mapped. Optimized geometries of both of these catalysts
(Figure 5) suggest that they have a planar geometry around Pd.
In PP1, the distances Pd−P, Pd−O, and Pd−N are 2.3, 2.11, and
2.15 Å, respectively. On the other hand, in PP2, Pd−P, Pd−O,
and Pd−N distances are 2.54, 2.06, and 2.18 Å respectively.
Palladacycles PP1 and PP2 have a free energy difference of
12.03 kcal/mol, with PP1 being more stable than PP2. The initial
step in the catalytic reactivity is the reaction of LiOH with benzyl
alcohol (BA), giving lithium benzyloxide (Li-BA). The free
energy change during the reactions was computed with respect to
the reactants PP1 + Li-BA and PP2 + Li-BA.
The free energy profile for the catalytic cycle of PP1 is given
in Figure 6, and the structures of the stationary points obtained
on the free energy path along with free energy values (Figure S2
and Table S2) are given in the Supporting Information. In the
presence of base (LiOH), BA is converted to lithiated benzyl
alcohol (Li-BA) with H2O as the side product. Li-BA then
reacts with the catalyst PP1 to form an intermediate PP1-Int1,
which has a free energy of −19.8 kcal/mol with respect to
the reactants. In complex PP1-Int1, the Pd−O(Li-BA) and
Li(Li-BA)−O(OAc) distances are 4.14 and 1.84 Å, respectively.
PP1-Int1 is converted to another complex PP1-Int2 that
involves the transfer of Li atom from Li-BA to the OAc group
along with the formation of a new bond between the Pd and O atom
of Li-BA through a transition state PP1-TS1. Pd−O(Li-BA)
and Li−O(OAc) distances in PP1-TS1 are 2.54 and 1.78 Å,
respectively. In PP1-Int2, the distances Pd−O(Li-BA) and
Li−O(OAc) are 2.11 and 1.92 Å, respectively. Then lithium
a
a
Reaction conditions: 1 mmol 2-aminobenzyl alcohol, 1.2 mmol ketone,
0.25 mmol LiOH, 1 × 10−2 mmol palladacycle 2, and 2 × 10−2 mmol
P(2-Fur)3.
2-aminobenzyl alcohol was reacted with acetophenone to provide
the corresponding qunoline in 89% yield (Table 3, entry 35).
Good isolated yields were obtained when we used 4-methyl
acetophenone and 4-fluoroacetophenone (Table 3, entries 36
and 37). Propiophenone, butyrophenone, and 1-tetralone were
also reacted with 2-aminobenzyl alcohol to yield the corre-
sponding quinoline derivatives in 82, 78, and 91% yield, respec-
tively (Table 3, entries 38, 39, and 40). Palladacycle 2 showed
regioselectivity toward the formation of 2-propylquinoline when
2-pentanone was used as a substrate (Table 3, entry 41).
Further, we investigated the catalytic activity of the pallada-
cycles 1 and 2 by choosing 12 h reaction time at respective
temperatures that are listed in Tables 1 (entries 1, 10, 14, and 18)
and 2 (entry 22). Palladacycle 2 exhibited better yields over
palladacyle 1. Notably, a remarkable difference was observed
when ortho-substituted acetophenone or ortho-benzyl alcohols
were used as substrates, and in such cases, palladacycle 1 was
almost unreactive or not efficient to produce the desired
products (Table 1, entries 10, 14, and 18).
It is of interest to understand the mechanism for the catalytic
reaction. As shown in Scheme 3, we have proposed a plausible
mechanism for the reaction based on our experimental results
and literature reports22 for α-alkylation of ketones with alcohols.
F
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 3. Proposed Catalytic Pathway for α-Alkylation of Ketones Using Alcohols
from benzyl CH2 to Pd atom through a transition state PP1-TS2.
PP1-TS2 has a free energy of 51.1 kcal/mol. Pd−H(−CH2
benzyl) distances in PP1-BA, PP1-TS2, and PP1-H are 3.1, 2.93,
and 1.59 Å, respectively. The intermediate benzaldehyde then
reacts with the reactant acetophenone to give the preproduct
(Pre-P) with the removal of H2O. Pre-P then reacts with PP1-H
to form an alkoxide complex PP1-Alko, which has a free energy
of −5.8 kcal/mol. PP1-Alko complex is formed by H atom
transfer from the Pd atom to the double bond of Pre-P
accompanied by the bond formation between the O atom of
Pre-P and Pd atom. This is apparent from the Pd−H bond
distances that are 1.59 Å in PP1-H and 5.1 Å in PP1-Alko. Also,
Pd−O(Pre-P) distance in PP1-Alko is 2.1 Å. In the final step,
PP1-Alko reacts with Li-BA to form lithiated alkoxide, and
Li-Alko regenerates the active species PP1-BA for the next cycle.
The free energy change for the formation of Li-Alko and
PP1-BA is −23.8 kcal/mol. Li-Alko then reacts with H2O to give
the final product of this reaction.
Figure 5. Optimized geometries of palladacycle-phosphine catalysts 1
(PP1: left) and 2 (PP2: right). Phenyl and furyl H atoms are not shown
for clarity.
The free energy profile (Figure 7) was mapped for the same
reaction using palladacycle PP2 as catalyst, and the stationary
Figure 6. Free energy profile for reaction between Li-BA with PP1.
acetate (LiOAc) dissociates from catalyst PP1, resulting in the
benzylated complex with PP1, PP1-BA. Pd−O(BA) distance in
PP1-BA is 2.06 Å. PP1-BA has a free energy of −9.2 kcal/mol.
PP1-BA is then converted to complex PP1-H (with benzalde-
hyde as the intermediate) that involves the transfer of H atom
Figure 7. Free energy profile for reaction between Li-BA with PP2.
G
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
point structures are given in the Supporting Information
(Figure S2). The first step is the formation of a complex
PP2-Int1, where Li-BA comes close to Pd metal, with
Pd−O(Li-BA) and Li−O(OAc) distances being 3.5 and 1.9 Å,
respectively. It has a free energy of −21.8 kcal/mol and is more
stable than the similar intermediate PP1-Int1 by 2 kcal/mol.
PP2-Int1 is converted to another complex PP2-Int2 through a
transition state PP2-TS1, which has a barrier of 4.9 kcal/mol.
The similar step was encountered in the case of the reaction
catalyzed by PP1 (PP1-TS1) with a barrier of 9 kcal/mol,
indicating that the reaction catalyzed by catalyst PP2 would
be more favorable than by catalyst PP1. Like PP1-TS1, this
transition state also corresponds to the transfer of Li atom from
Li-BA to the leaving group LiOAc, and the formation of a
new bond between Pd metal and O atom of Li-BA. PP2-TS1
connects to PP2-Int2 on the other side with a free energy change
of −41.5 kcal/mol. PP2-Int2 is more stable than PP1-Int2 by
10.8 kcal/mol. The leaving group LiOAc then dissociates to give
PP2-BA having a free energy change of −19.8 kcal/mol. PP2-BA
is more stabilized than PP1-BA by 10.6 kcal/mol, making the
reaction more favorable when catalyst PP2 is used. In the next
step, PP2-BA is converted to PP2-H via transition state
PP2-TS2, which is 42.4 kcal/mol higher than reactants. The
products (PP2-H) for this step using PP2 are stabilized by
18.6 kcal/mol more than the products using PP1 (PP1-H).
In the next step, PP2-H reacts with Pre-P to form PP2-Alko as
the product with free energy change of −16.2 kcal/mol. It is more
stable than that of PP1-Alko by 10.4 kcal/mol. In the final
step, PP2-Alko reacts with Li-BA to regenerate PP2-BA
(active species) and Li-Alko with a free energy change of
−34.3 kcal/mol. Product obtained in this step is also more stable
than that obtained using catalyst PP1 by 10.5 kcal/mol. Li-Alko
can then react with water to give the final product. In addition,
another reaction path involving an isomer of palladacycle 2 with
the phosphine ligand trans to pyrazole nitrogen (PP3) was also
studied. The free energy profile for this path was found to be
similar to that for PP1 (Figure S4, Supporting Information).
Hence, this path was ruled out to account for the observed
experimental results.
more stabilized for the reaction involving 2. In addition, the
overall free energy barrier for the reaction using palladacycle 2
is stabilized by ∼9 kcal/mol compared to that of palladacycle 1.
Hence, palladacycle 2 is expected to be more favorable than
palladacycle 1, which is consistent with experimental results.
Our catalytic system broadens the scope of the hydrogen-
borrowing catalysis in alkylation reactions. Furthermore, we were
able to exploit this catalytic system for the synthesis of the
biologically active and important molecule donepezil.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organomet.
7b00478.
■
S
Experimental procedures and compound characterization
(PDF)
Accession Codes
CCDC 1555731−1555733 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge
CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
*Tel: 091-6742494165. E-mail: krishv@niser.ac.in.
■
ORCID
Ramesh Mamidala: 0000-0003-0399-4437
Shaikh Samser: 0000-0002-2669-2020
Upakarasamy Lourderaj: 0000-0002-5550-9694
Krishnan Venkatasubbaiah: 0000-0002-4858-8590
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Science & Engineering Research Board
(SERB) (SR/S1/IC-58/2012, SR/S2/RJN-49/2011 and SR/
S1/PC/0060/2010), New Delhi, and Department of Atomic
Energy (DAE) for financial support. R.M. and S.S. thank CSIR
and DST for research fellowships.
The free energies of the important intermediates and
transition states for the reaction paths using palladacycles 1
and 2 are compared in Table S2 (Supporting Information).
Of interest is the difference in free energies (ΔΔG) of the
concerned stationary points (k) along the path for the two
reactions, that is, ΔΔG = ΔGk(PP2) − ΔGk(PP1). We can see
that overall energies for reaction using PP2 are lower than that
when using PP1, with the largest stabilization of 18.6 kcal/mol
for PP2-H compared to that of PP1-H. In addition, the overall
barrier for the reaction using PP2 (2TS2 vs 1TS2) is lowered by
∼9 kcal/mol compared to that when using PP1. These findings
indicate that reaction using PP2 is expected to be favorable
compared to that with PP1, consistent with the experiments.
REFERENCES
■
̈
(1) (a) Herrmann, W. A.; Brossmer, C.; Ofele, K.; Reisinger, C.-P.;
Priermeier, T.; Beller, M.; Fischer, H. Angew. Chem., Int. Ed. Engl. 1995,
34, 1844−1848. (b) Herrmann, W. A.; Ofele, K.; Preysing, D. V.;
̈
Schneider, S. K. J. Organomet. Chem. 2003, 687, 229−248.
(c) Beletskaya, I. P.; Cheprakov, A. V. J. Organomet. Chem. 2004, 689,
4055−4082. (d) Farina, V. Adv. Synth. Catal. 2004, 346, 1553−1582.
(e) Dunina, V. V.; Gorunova, O. N. Russ. Chem. Rev. 2005, 74, 871−913.
(f) Dupont, J.; Consorti, C. S.; Spencer, J. Chem. Rev. 2005, 105, 2527−
2572. (g) Kapdi, A. R.; Fairlamb, I. J. S. Chem. Soc. Rev. 2014, 43, 4751−
4777. (h) Mo, D.-L.; Zhang, T.-K.; Ge, G.-C.; Huang, X.-J.; Ding, C.-H.;
Dai, L.-X.; Hou, X.-L. Synlett 2014, 25, 2686−2702.
(2) (a) Nonoyama, M. Transition Met. Chem. 1982, 7, 281−284.
(b) Bosque, R.; Maseras, F. Eur. J. Inorg. Chem. 2005, 2005, 4040−4047.
(c) Lyons, T. W.; Sanford, M. S. Chem. Rev. 2010, 110, 1147−1169.
(d) Engle, K. M.; Mei, T.-S.; Wasa, M.; Yu, J.-Q. Acc. Chem. Res. 2012,
45, 788−802. (e) Chu, J.-H.; Wu, C.-C.; Chang, D.-H.; Lee, Y.-M.; Wu,
M.-J. Organometallics 2013, 32, 272−282. (f) Chu, J. H.; Huang, H. P.;
Hsu, W. T.; Chen, S. T.; Wu, M. J. Organometallics 2014, 33, 1190−
1204. (g) Zhang, S.-Y.; Li, Q.; He, G.; Nack, W. A.; Chen, G. J. Am.
CONCLUSIONS
■
In summary, cyclometalation of 3,5-diphenyl-1-(4-(trifluoromethyl)-
phenyl)-1H-pyrazole with palladium acetate afforded regioselective
palladacycles 1 and 2. We successfully studied their applications
as catalysts for the alkylation of ketones using alcohols under
solvent-free conditions. Experimental studies reveal that pallada-
cycle 2 is more active than palladacycle 1. Comparison of the
energy profiles for reactions involving palladacycles 1 and 2 using
DFT methods reveals that the reactions proceed through similar
pathways. However, the intermediates and transition states are
H
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Chem. Soc. 2015, 137, 531−539. (h) Frutos-Pedreno, R.; García-
(17) (a) Sugimoto, H.; Iimura, Y.; Yamanishi, Y.; Yamatsu, K. J. Med.
Chem. 1995, 38, 4821−4829. (b) Niphade, N.; Mali, A.; Jagtap, K.; Ojha,
R. C.; Vankawala, P. J.; Mathad, V. T. Org. Process Res. Dev. 2008, 12,
731−735.
(18) Ohta, H.; Yuyama, Y.; Uozumi, Y.; Yamada, Y. M. A. Org. Lett.
2011, 13, 3892−3895.
̃
́
Sanchez, E.; Oliva-Madrid, M. J.; Bautista, D.; Martínez-Viviente, E.;
Saura-Llamas, I.; Vicente, J. Inorg. Chem. 2016, 55, 5520−5533.
(3) (a) Deeming, A. J.; Rothwell, I. P. J. Chem. Soc., Chem. Commun.
1978, 344−345. (b) Albert, J.; Granell, J.; Sales, J.; Solans, X.; Font-
Altaba, M. Organometallics 1986, 5, 2567−2568. (c) Albert, J.; Gomez,
̈
(19) (a) Khan, M. T. H. Top. Heterocycl. Chem. 2007, 11, 213−229.
(b) Narender, P.; Srinivas, U.; Ravinder, M.; Rao, B. A.; Ramesh, C.;
Harikishore, K.; Gangadasu, B.; Murthy, U. S. N.; Rao, V. J. Bioorg. Med.
Chem. 2006, 14, 4600−4609. (c) Busetti, A.; Crawford, D. E.; Earle, M.
J.; Gilea, M. A.; Gilmore, B. F.; Gorman, S. P.; Laverty, G.; Lowry, A. F.;
McLaughlin, M.; Seddon, K. R. Green Chem. 2010, 12, 420−425.
(d) Thatcher, T. H.; Luzina, I.; Fishelevich, R.; Tomai, M. A.; Miller, R.
L.; Gaspari, A. A. J. Invest. Dermatol. 2006, 126, 821−831.
(e) Klingenstein, R.; Melnyk, P.; Leliveld, S. R.; Ryckebusch, A.;
Korth, C. J. Med. Chem. 2006, 49, 5300−5308.
M.; Granell, J.; Sales, J.; Solans, X. Organometallics 1990, 9, 1405−1413.
(d) Gorunova, O. N.; Keuseman, K. J.; Goebel, B. M.; Kataeva, N. A.;
Churakov, A. V.; Kuz’mina, L. G.; Dunina, V. V.; Smoliakova, I. P. J.
Organomet. Chem. 2004, 689, 2382−2394. (e) Roiban, G.-D.; Serrano,
E.; Soler, T.; Aullon
Urriolabeitia, E. P. Inorg. Chem. 2011, 50, 8132−8143. (f) Mercier, A.;
Wagschal, S.; Guenee, L.; Besnard, C.; Kundig, E. P. Organometallics
́
, G.; Grosu, I.; Cativiela, C.; Martínez, M.;
́
́
̈
2013, 32, 3932−3942. (g) Wu, J.; Barnard, J. H.; Zhang, Y.; Talwar, D.;
Robertson, C. M.; Xiao, J. Chem. Commun. 2013, 49, 7052−7054.
(h) Cross, W. B.; Hope, E. G.; Lin, Y.-H.; Macgregor, S. A.; Singh, K.;
Solan, G. A.; Yahya, N. Chem. Commun. 2013, 49, 1918−1920.
(i) Kondrashov, M.; Raman, S.; Wendt, O. F. Chem. Commun. 2015, 51,
911−913.
(20) Cheng, C.-C.; Yan, S.-J. Org. React. 1982, 37.
(21) (a) Cho, C. S.; Ren, W. X. J. Organomet. Chem. 2007, 692, 4182−
4186. (b) Vander Mierde, H.; Van Der Voort, P.; De Vos, D.; Verpoort,
F. Eur. J. Org. Chem. 2008, 2008, 1625−1631. (c) Cho, C. S.; Ren, W. X.;
Yoon, N. S. J. Mol. Catal. A: Chem. 2009, 299, 117−120.
(22) (a) Ding, B.; Zhang, Z.; Liu, Y.; Sugiya, M.; Imamoto, T.; Zhang,
W. Org. Lett. 2013, 15, 3690−3693. (b) Gowrisankar, S.; Neumann, H.;
Beller, M. Angew. Chem., Int. Ed. 2011, 50, 5139−5143. (c) Gowrisankar,
S.; Sergeev, A. G.; Anbarasan, P.; Spannenberg, A.; Neumann, H.; Beller,
(4) (a) Dobereiner, G. E.; Crabtree, R. H. Chem. Rev. 2010, 110, 681−
703. (b) Obora, Y. ACS Catal. 2014, 4, 3972−3981. (c) Huang, F.; Liu,
Z.; Yu, Z. Angew. Chem., Int. Ed. 2016, 55, 862−875.
(5) Iridium: (a) Taguchi, K.; Nakagawa, H.; Hirabayashi, T.;
Sakaguchi, S.; Ishii, Y. J. Am. Chem. Soc. 2004, 126, 72−73. (b) Onodera,
G.; Nishibayashi, Y.; Uemura, S. Angew. Chem., Int. Ed. 2006, 45, 3819−
3822. (c) Morita, M.; Obora, Y.; Ishii, Y. Chem. Commun. 2007, 2850−
2852. (d) Li, F.; Ma, J.; Wang, N. J. Org. Chem. 2014, 79, 10447−10455.
(e) Frost, J. R.; Cheong, C. B.; Akhtar, W. M.; Caputo, D. F. J.;
Stevenson, N. G.; Donohoe, T. J. J. Am. Chem. Soc. 2015, 137, 15664−
15667. (f) Liu, P.; Liang, R.; Lu, L.; Yu, Z.; Li, F. J. Org. Chem. 2017, 82,
1943−1950. (g) Akhtar, W. M.; Cheong, C. B.; Frost, J. R.; Christensen,
K. E.; Stevenson, N. G.; Donohoe, T. J. J. Am. Chem. Soc. 2017, 139,
2577−2580.
M. J. Am. Chem. Soc. 2010, 132, 11592−11598. (d) Muller, J. A.; Goller,
̈
C. P.; Sigman, M. S. J. Am. Chem. Soc. 2004, 126, 9724−9734.
(6) Ruthenium: (a) Martínez, R.; Brand, G. J.; Ramon
Tetrahedron Lett. 2005, 46, 3683−3686. (b) Martínez, R.; Ramon
́
, D. J.; Yus, M.
, D. J.;
́
Yus, M. Tetrahedron 2006, 62, 8988−9001. (c) Kuwahara, T.;
Fukuyama, T.; Ryu, I. Org. Lett. 2012, 14, 4703−4705. (d) Schlepphorst,
C.; Maji, B.; Glorius, F. ACS Catal. 2016, 6, 4184−4188.
(7) Rhodium: Satyanarayana, P.; Reddy, G. M.; Maheswaran, H.;
Kantam, M. L. Adv. Synth. Catal. 2013, 355, 1859−1867.
(8) Heterogeneous palladium: (a) Kwon, M. S.; Kim, N.; Seo, S. H.;
Park, I. S.; Cheedrala, R. K.; Park, J. Angew. Chem., Int. Ed. 2005, 44,
6913−6915. (b) Cho, C. S. J. Mol. Catal. A: Chem. 2005, 240, 55−60.
(c) Yamada, Y. M. A.; Uozumi, Y. Org. Lett. 2006, 8, 1375−1378. (d) Xu,
G.; Li, Q.; Feng, J.; Liu, Q.; Zhang, Z.; Wang, X.; Zhang, X.; Mu, X.
ChemSusChem 2014, 7, 105−109. (e) Dang, T. T.; Shan, S. P.;
Ramalingam, B.; Seayad, A. M. RSC Adv. 2015, 5, 42399−42406. (f) Bai,
M.; Xin, H.; Guo, Z.; Guo, D.; Wang, Y.; Zhao, P.; Li, J. Appl. Surf. Sci.
2017, 391, 617−626.
(9) Cobalt: Zhang, G.; Wu, J.; Zeng, H.; Zhang, S.; Yin, Z.; Zheng, S.
Org. Lett. 2017, 19, 1080−1083.
(10) Iron: Elangovan, S.; Sortais, J.-B.; Beller, M.; Darcel, C. Angew.
Chem., Int. Ed. 2015, 54, 14483−14486.
(11) Osmium: Buil, M. L.; Esteruelas, M. A.; Herrero, J.; Izquierdo, S.;
Pastor, I. M.; Yus, M. ACS Catal. 2013, 3, 2072−2075.
(12) Manganese: Pena-Lopez, M.; Piehl, P.; Elangovan, S.; Neumann,
H.; Beller, M. Angew. Chem., Int. Ed. 2016, 55, 14967−14971.
(13) Yoon, C.; Kim, T. G.; Cho, C. S. Organometallics 2014, 33, 1890−
1892.
(14) Kose, O.; Saito, S. Org. Biomol. Chem. 2010, 8, 896−900.
(15) (a) Mamidala, R.; Mukundam, V.; Dhanunjayarao, K.;
Venkatasubbaiah, K. Dalton Trans. 2015, 44, 5805−5809. (b) Mamidala,
R.; Mukundam, V.; Dhanunjayarao, K.; Venkatasubbaiah, K. Tetrahe-
dron 2017, 73, 2225−2233.
(16) (a) Almeida, M. L. S.; Kocovsky, P.; Backvall, J.-E. J. Org. Chem.
1996, 61, 6587−6590. (b) Tsutsui, K.; Sakamoto, H.; Ukena, K. J.
Steroid Biochem. Mol. Biol. 2003, 85, 311−321. (c) Kawahara, R.; Fujita,
K.; Yamaguchi, R. Angew. Chem., Int. Ed. 2012, 51, 12790−12794.
I
DOI: 10.1021/acs.organomet.7b00478
Organometallics XXXX, XXX, XXX−XXX