Catalytic oxidation of primary aromatic alcohols using half sandwich Ir(III), Rh(III) and Ru(II) complexes: A practical and theoretical study

Article abstract of DOI:10.1016/j.molcata.2016.06.017

The complexes [Cp*IrCl(N-(pyridin-2-ylmethylene)aniline)]PF₆ (1), [Cp*RhCl(N-(pyridin-2-ylmethylene)aniline)]PF₆ (2), (where Cp*?=?1,2,3,4,5-pentamethylcyclopentadiene) and [η⁶-areneRuCl(N-(pyridin-2-ylmethylene)aniline)]PF₆ (3) have been synthesized and the structure and purity of these were confirmed by single crystal XRD and elemental analyses. Iridium and rhodium complexes exhibit the P21/n space group, the ruthenium complex the P21/c space group and all three complexes show the expected pseudo octahedral “piano-stool” geometry. The catalytic performance of these complexes for the dehydrogenation of primary alcohols to their respective aldehydes with different bases and solvents was investigated. The complexes of iridium and ruthenium give good conversions in different alkaline solutions. Density functional theory was applied to determine the respective MO energy levels, bond lengths, bond angles and binding energies of all the metal complexes. It was also used to study the activity, stability and intermediates of the complexes. A Gibbs free energy (ΔG) DFT calculation was carried out to help understand the reaction mechanism/catalytic cycle of the Rh complex (2). The energy barrier for oxidation of aromatic alcohols by the rhodium hydride complex is much lower (10.32?kcal/mol) than the barrier for hydride transfer of the corresponding Rh benzyloxo species (15.19?kcal/mol), in agreement with mechanisms proposed for related systems.

Full text of DOI:10.1016/j.molcata.2016.06.017

Accepted Manuscript
Title: Catalytic Oxidation of Primary Aromatic Alcohols
Using Half Sandwich Ir(III), Rh(III) and Ru(II) Complexes: A
Practical and Theoretical study
Author: Saravanan Thangavel Subramaniam Boopathi N.
Mahadevaiah Ponmalai Kolandaivel Pramod Pansuriya Holger
B. Friedrich
PII:
S1381-1169(16)30230-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.06.017
MOLCAA 9923
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
20-4-2015
10-6-2016
17-6-2016
Please cite this article as: Saravanan Thangavel, Subramaniam Boopathi,
N.Mahadevaiah, Ponmalai Kolandaivel, Pramod Pansuriya, Holger B.Friedrich,
Catalytic Oxidation of Primary Aromatic Alcohols Using Half Sandwich Ir(III), Rh(III)
andRu(II)Complexes:APracticalandTheoreticalstudy, JournalofMolecularCatalysis
A: Chemical http://dx.doi.org/10.1016/j.molcata.2016.06.017
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Highlights





Ir(III), Rh(III) and Ru(II) complexes show pseudo octahedral “piano-tool” geometry
Catalytic activity for benzyl alcohol oxidation demonstrated by all three complexes
Ir and Ru complexes show best conversion of primary aromatic alcohols to aldehydes
DFT studies carried out on the complexes
Proposed reaction mechanism is supported by DFT
1

Catalytic Oxidation of Primary Aromatic Alcohols Using Half Sandwich
Ir(III), Rh(III) and Ru(II) Complexes: A DFT Study
a
b
a
b
Saravanan Thangavel , Subramaniam Boopathi , N. Mahadevaiah , Ponmalai Kolandaivel ,
a
*a
Pramod Pansuriya , Holger B. Friedrich
a
School of Chemistry and Physics, University of KwaZulu-Natal, Durban 4000, South Africa.
b
Department of Physics, Bharathiar University, Coimbatore-641 046, Tamilnadu, India.
*
2
Corresponding author: E-mail address: friedric@ukzn.ac.za; Tel: +27 31 2603107; Fax: +27 31
603091.
Graphical abstract
2

Catalytic Oxidation of Primary Aromatic Alcohols Using Half Sandwich
Ir(III), Rh(III) and Ru(II) Complexes: A Practical and Theoretical study.
a
b
a
b
Saravanan Thangavel , Subramaniam Boopathi , N. Mahadevaiah , Ponmalai Kolandaivel ,
a
*a
Pramod Pansuriya , Holger B. Friedrich
a
School of Chemistry and Physics, University of KwaZulu-Natal, Durban 4000, South Africa.
b
Department of Physics, Bharathiar University, Coimbatore-641 046, Tamilnadu, India.
*
2
Corresponding author: E-mail address: friedric@ukzn.ac.za; Tel: +27 31 2603107; Fax: +27 31
603091.
Abstract
The complexes [Cp*IrCl(N-(pyridin-2-ylmethylene)aniline)]PF6 (1), [Cp*RhCl(N-
pyridin-2-ylmethylene)aniline)]PF (2), (where Cp* =1,2,3,4,5-pentamethylcyclopentadiene)
(
6
6
and [η -areneRuCl(N-(pyridin-2-ylmethylene)aniline)]PF (3) have been synthesized and the
6
structure and purity of these were confirmed by single crystal XRD and elemental analyses.
Iridium and rhodium complexes exhibit the P21/n space group, the ruthenium complex the P21/c
space group and all three complexes show the expected pseudo octahedral “piano-stool”
geometry. The catalytic performance of these complexes for the dehydrogenation of primary
alcohols to their respective aldehydes with different bases and solvents was investigated. The
complexes of iridium and ruthenium give good conversions in different alkaline solutions.
Density functional theory was applied to determine the respective MO energy levels, bond
lengths, bond angles and binding energies of all the metal complexes. It was also used to study
the activity, stability and intermediates of the complexes. A Gibbs free energy ꢀꢁꢂꢃDFT
calculation was carried out to help understand the reaction mechanism / catalytic cycle of the Rh
complex (2). The energy barrier for oxidation of aromatic alcohols by the rhodium hydride
complex is much lower (-8.50 Kcal/mol) than the barrier for hydride transfer of the
corresponding Rh benzyloxo species (21.41 Kcal/mol), in agreement with mechanisms proposed
for related systems.
3

Keywords: Catalytic oxidation; N, N’ bidentate ligand; half sandwich metal complexes; primary
alcohols; DFT.
4

1
. Introduction
Oxidation of primary aromatic alcohols to their respective aldehydes is an important
chemical transformation in synthetic chemistry [1]. Alcohols are generally inactive and require
harsh methods to activate the hydroxyl group present [2-5]. In transition metal chemistry, the
metal can mediate the dehydrogenation of alcohols to yield the aldehydes [5-9] and there are
many reports on the oxidation of alcohols to aldehydes and ketones. Indeed, metal catalyzed
versions of the oxidation reaction are recognized as one of the greener methods [10-13]. There
are several water soluble homogeneous metal complexes that have been reported for the
oxidation of alcohols, such as those of copper, palladium and gold [14, 15]. Many classical
methods and reagents are available for these reactions, like “activated DMSO” methods,
NaOCl/TEMPO (TEMPO = 2,2,6,6-tetramethyl-1-piperidinyloxyl), hypervalent iodine reagents
and chromium and manganese oxides [16-18].
To explore the metal complex systems, researchers also have developed eco-friendly and
efficient catalytic metal complex systems for the oxidation of alcohols using less toxic oxidants
such as oxygen, hydrogen peroxide and acetone [19-21].
In a homogeneous catalytic system, the borrowing of hydrogen and hydrogen auto
transfer reactions usually occur via the formation of a metal hydride intermediate to complete the
catalytic cycle [22, 23]. In recent years the hydrogen gas evolved alcohol dehydrogenation
process has generated great interest from the view point of atom economy [22, 23]. There are
different homogeneous catalytic systems which effectively dehydrogenate primary alcohols to
their respective aldehydes, which include metals such as iridium, rhodium and ruthenium [24-
2
6]. In the present work, we report comparative catalytic studies using iridium, rhodium and
ruthenium in N, N’ functional bidentate N-(pyridin-2-ylmethylene)aniline ligand complexes for
the oxidation of primary aromatic alcohols to their respective aldehydes using inexpensive bases
such as Cs CO , K CO , Na CO and NaHCO . We also report a DFT modelling study of the
2
3
2
3
2
3
3
complexes, HOMO-LUMO energy differences and reactivity.
2
. Experimental Section
5

2
.1. Methods and analysis
Unless otherwise noted, all manipulations were performed using standard Schlenk tube
techniques under argon atmosphere. The reagents and solvents were purchased from Sigma-
Aldrich, Fluka and Merck and used without further purification. The solvents were dried by
standard procedures and distilled prior to use. Gas chromatography (GC) analysis was performed
using a Perkin-Elmer Clarus 500 GC with a HP-Pona-50 m x 0.2 mm, 0.5 micron capillary
column.
Single-crystal X-ray diffraction data were collected on a Bruker KAPPA APEX II DUO
diffractometer using graphite-monochromated Mo-K radiation (λ = 0.71073 Å). Data collection
was carried out at 173(2) K. Temperature was controlled by an Oxford Cryostream cooling
system (Oxford Cryostat). Cell refinement and data reduction were performed using the program
SAINT [27]. The data were scaled and absorption corrections performed using SADABS. The
structure was solved by direct methods using SHELXS-97 and refined by full-matrix least-
2
squares methods based on F using SHELXL-97 [28]. For the structures and graphics, ORTEP-3
[
29] and DIAMOND [30] programs were used. The crystals of all the metal complexes were
obtained by slow evaporation of a mixture of hexane and acetonitrile at ambient temperature.
The electronic absorption spectroscopy (UV-Vis) studies were recorded using a Perkin Elmer
precisely Lambda35 instrument. The TGA-DSC analyses were performed using a Thermal
-1
Analyser SDT Q600 instrument and a heating rate of 10 °C.min and nitrogen flow of 100
-1
ml.min . The metal precursors (IrCl Cp*) , (RhCl Cp*) and (RuCl C H ) were prepared
2
2
2
2
2
6
6 2
according to the reported methods [31-33].
2
.2. DFT Calculations
All the calculations were performed using the Guassian 09 program package [34]. The
structural and energetic results calculated using M05-2X were more consistent with the
experimental observations [35]. The hybrid exchange-correlation functional M05-2X, developed
by Zhao and Truhlar [36, 37], have been shown to describe weak bonding interaction better than
the other density functionals such as M06 classes, M05, B3LYP, and PBE functional. So, all the
complex structures were optimized at M05-2X level of theory using the LANL2DZ basis set.
Frequency calculation was carried out at the same level of theory and no negative frequency was
observed. The absence of imaginary frequencies confirmed that all complex structures had stable
6

minima. The HOMO and LUMO energy gap was calculated from the DFT method. The frontier
molecular orbital plot was drawn using the chemcraft program package
(
http://www.chemcraftprog.com). Using Gaussian NBO (ver. 3.1), natural population analysis
NPA) and natural bond orbital (NBO) analysis has been performed to investigate the change in
(
electron density from proton donor to proton acceptor for optimized metal complexes at the
(2)
M05-2X/LANL2DZ level of theory. Second order perturbation energy (E ) was used to
estimate the magnitude of electron delocalization [38].
2
.3. General procedure for the oxidation of benzyl alcohol
Benzyl alcohol (1 mmol), catalyst (2 mol%), base (5 mol%) and solvent (20 mmol) were
placed in a Schlenk tube under argon or nitrogen atmosphere. The mixture was stirred
magnetically under reflux over the specified time. The yield of the benzaldehyde was determined
by GC analysis using chlorobenzene as an internal standard.
2
.4. Procedure for the oxidation of benzyl alcohol derivatives catalysed by complexes 1 and 3
The benzyl alcohol derivatives (1 mmol), catalyst 1 or 3 (2 mol%), base (5 mol%) and
toluene or p-xylene (20 mmol) were added to a Schlenk tube under argon or nitrogen
atmosphere. The mixture was stirred magnetically at reflux temperature. The yields of the
products were analyzed by GC chromatography using chlorobenzene as an internal standard.
3
. Results and discussion
3
.1. Synthesis of ligand and metal complexes
The N, N’ bidentate ligand precursor N-(pyridin-2-ylmethylene)aniline and metal
complexes of iridium, rhodium and ruthenium were synthesised according to methods discussed
elsewhere [39-41]. The structure and purity of the synthesized metal complexes, along with the
ligands were confirmed by single crystal XRD, NMR, IR and electronic absorption spectral
studies. The thermal stability of the prepared compounds was confirmed by TGA analysis. The
structures of the metal complexes are shown (Fig.1).
7

Fig. 1. The metal complexes of Ir(III) (1), Rh(III) (2) and Ru(II) (3).
3
.2. X –Ray Crystallography
The crystallographic data of complexes 1-3 are given in Table 1. The ORTEP view of the
metal complexes iridium (1), rhodium (2) and ruthenium (3) are shown in Fig. 2a-c. In
complexes 1 and 2, the iridium and rhodium metal centers are coordinated via the nitrogen atoms
of the ligand, five carbon atoms of the Cp* moiety and Cl of the metal center. In complex (3), the
6
ruthenium metal is coordinated via the nitrogen atoms of the ligand, via η coordination to the
benzene ring and to chlorine. This gives pseudo-octahedral half sandwich “piano-stool”
structures around the iridium, rhodium and ruthenium metal centers in the complexes [41]. The
Cp*/arene moieties of complexes 1, 2 and 3 occupy a triangular face of an octahedron. The N1,
N2 nitrogen atoms form a five membered chelate ring with the metal center. A chloride
completes the coordination sphere. The M-Cl bond length in complexes 1-3 are 2.400(10) Å,
2
.393(8) Å and 2.393(10) Å, respectively. These values are consistent with related reported
5
5
metal complexes in literature, like [(η -C Me )IrCl(C H N-2-CH=NC H -P-NO )]PF [40], [(η -
5
5
5
5
6
4
2
6
6
C Me )RhCl(C H N-2-CH=NC H -P-Cl)]BF [40] and [(η -C H )RuCl(C H N-2-CH=NC H -P-
5
5
5
5
6
4
4
6
6
5
5
6
4
CH )]PF [41]. The M-C (Cp*/arene) bond lengths in the complexes are 2.153(3) - 2.205(3) Å
3
6
(1), 2.136(3) - 2.192(3) Å (2) and 2.171(4) - 2.192(4) Å (3), respectively. Non-covalent
interactions of C-H…F and C-H…Cl bonds in the complexes of 1, 2 and 3 are observed. In the
crystal structure of complexes 1-3 C-H…Cl and C-H…F non-covalent intra and inter-molecular
interactions are seen (Fig. 3a-c). In all three complexes, the counter ion PF is involved in C-
6
H…F non-covalent inter-molecular interactions. The P-F…π inter-molecular interaction is
observed in complexes 1 and 3 and C-H… π inter-molecule interaction is observed in complex 2
(Tables 2 & 3).
8

Table 1
Crystallographic data and summary of structural refinement of the metal complexes 1, 2 and 3.
1
2
3
Empirical formula
Formula weight
Crystal colour
Temperature/K
Crystal system
Space group
a (Å)
C H ClF IrN P
C H ClF N PRh
C H ClF N PRu
2
2
25
6
2
22 25
6
2
18 16
6
2
690.06
Orange
173(2)
600.77
541.82
Yellow
173(2)
Yellow
173(2)
monoclinic
monoclinic
monoclinic
P2 /n
1
P2 /n
1
P2 /c
1
8.694(4)
13.312(6)
20.664(9)
90.00
8.7080(4)
13.2046(3)
20.6563(8)
90.00
7.7407(3)
17.0759(4)
15.3185(6)
90.00
b (Å)
c (Å)
α (°)
β (°)
94.244(9)
90.00
94.4670(10)
90.00
101.7610(10)
90.00
γ (°)
3
Volume (Å )
2385.0(18)
4
2367.96(15)
4
1982.28(12)
4
Z
3
ρcalc (mg/mm )
1.922
1.685
1.816
-
1
μ(MoKα) [mm ]
5.837
0.962
1.068
F(000)
1336.0
1208.0
1072.0
3
Crystal size/mm
0.16 × 0.14 × 0.12
3.96 to 57.26°
81251
0.18 × 0.14 × 0.11
7.34 to 54.92°
94755
0.16 × 0.12 × 0.11
7.22 to 55.76°
94614
2
θ range for data collection
Reflections collected
Independent reflections
Data/restraints/parameters
6006[R(int) = 0.0495]
6006/0/303
5403[R(int) = 0.0806]
5403/0/303
4716[R(int) = 0.0847]
4716/0/262
1.132
2
Goodness-of-fit on F
1.034
1.072
Final R indexes [I>=2σ (I)]
R = 0.0225, wR = 0.0458
R = 0.0344, wR = 0.0692
R = 0.0450, wR = 0.0960
1
2
1
2
1
2
Final R indexes [all data]
R = 0.0374, wR = 0.0514
R = 0.0656, wR = 0.0807
R = 0.0751, wR = 0.1109
1
2
1
2
1
2
-
3
Largest diff. peak/hole (e Å )
1.31/-0.68
1.20/-0.66
1.59/-0.77
9

Table 2
C-H…X intermolecular interactions in complexes 1-3.
C-H…X
D-H
H…A
D…A
D-H…A
Complex 1
C(8)-H(8a)…F(4)
0.98
2.54
2.44
2.35
2.79
2.50
2.54
3.502(4)
3.332(5)
3.093(5)
3.622(4)
3.235(4)
3.313(4)
167
151
135
146
134
138
C(10)-H(10C)…F(1) 0.98
C(11)-H(11)…F(2)
C(13)-H(13)…Cl(1)
C(14)-H(14)…F(5)
C(21)-H(21)…F(2)
Complex 2
0.95
0.95
0.95
0.95
C(10)-H(10A)…F(6) 0.98
2.43
2.35
2.79
2.51
2.54
3.341(4)
3.122(4)
3.607(3)
3.227(4)
3.315(4)
154
137
145
132
139
C(11)-H(11)…F(5)
C(13)-H(13)…Cl(1)
C(14)-H(14)…F(2)
C(21)-H(21)…F(5)
Complex 3
0.95
0.95
0.95
0.95
C(4)-H(4)…F(6)
C(17)-H(17)…F(5)
0.95
0.95
2.35
2.48
3.295(6)
3.374(6)
172
157
Table 3
Y-X…π intermolecular interactions in complexes 1-3.
Y-X…Cg(Pi-Ring)
Complex 1
X…Cg (Bond length (Å))
Y-X…Cg (Bond angles (°))
P(1)-F(1)…Cg(2)
Complex 2 (C-H…Cg)
C(7)-H(7B)…Cg(3)
Complex 3
3.111(3)
2.89
128.00(13)
137
P(1)-F(5)…Cg(1)
P(1)-F(6)…Cg(3)
3.329(4)
3.755(4)
129.22(18)
141.34(17)
1
0

Fig. 2a. ORTEP view of the metal complex 1 with atom numbering scheme. Displacement
ellipsoids are drawn at the 50% probability level and hydrogen atoms are omitted for the clarity.
Fig. 2b. ORTEP view of the metal complex 2 with atom numbering scheme. Displacement
ellipsoids are drawn at the 50% probability level and hydrogen atoms are omitted for the clarity.
1
1

Fig. 2c. ORTEP view of the metal complex 3 with atom numbering scheme. Displacement
ellipsoids are drawn at the 50% probability level and hydrogen atoms are omitted for the clarity.
Fig. 3a. Non-covalent C-H…F interactions in the metal complex 1.
1
2

Fig. 3b. Non-covalent C-H…F and C-H…Cl interactions in the metal complex 2.
1
3

Fig. 3c. Non-covalent π…π and C-H…F interactions in the metal complex 3.
3
.3. Electronic absorption spectroscopy
The electronic absorption spectra of all the metal complexes were recorded by preparing
a 10 µM (v/v DCM) solution. The resulting spectra of the metal complexes are depicted in the
wavelength range between 200 nm to 600 nm (Fig. 4). The intense low energy bands of the metal
complexes appeared at 382.92 nm (1), 369.77 nm (2), 416.27 nm (3) and the conjugated intra
ligand based π-π* and metal ligand charge transfer (MLCT) transitions were observed in the
region of 315.23 nm to 284.92 nm for (1), 324.17 nm to 274.82 nm for (2) and 313.87 nm to
2
61.25 nm for (3) [43].
1
4

2.0
1.5
1.0
0.5
0.0
Iridium complex
Rhodium complex (2)
Ruthenium complex (3)
(1)
250
300
350
400
450
500
550
600
Wavelength / nm
Fig. 4. UV-Visible spectrum of the metal complexes 1, 2 and 3 (Dichloromethane, 10µM, v/v).
3
.4. TG-DSC studies
TG patterns of all complexes are shown in Fig. 5a. In the TG traces, no degradation was
observed until ~270º C, which indicates the high inherent stability of complexes 1-3. The
calculated weight loss of the complexes at ~270 ºC was 2.28% (1), 3.12% (2) and 7.85% (3)
respectively. The DSC traces of exothermic peaks associated with the melting point of the
complexes were at 290 ºC (1), 280 °C (2) and 270 °C (3). The clear exothermic peak in the DSC
trace (Fig. 5b) further confirms the crystalline nature of the metal complexes [44].
1
5

100
90
80
70
60
50
40
Iridium complex
Rhodium complex (2)
Ruthenium complex (3)
(1)
0
100
200
300
400
500
Temperature / °C
-1
Fig. 5a. TGA curves of the complexes 1, 2 and 3. Heating rate 10 °C min .
1.0
0.5
0.0
Iridium complex
Rhodium complex (2)
Ruthenium complex (3)
(1)
exo
-
-
-
-
0.5
1.0
1.5
2.0
100
200
300
400
500
Temperature / ° C
-1
Fig. 5b. DSC curves of the complexes 1, 2 and 3. Heating rate 10 °C min .
3
3
.5. Catalytic activity studies
.5.1. Oxidation of primary alcohols
The oxidation of benzyl alcohol under reflux conditions was carried out with various
bases and solvents using the three complexes of iridium (1), rhodium (2) and ruthenium (3)
1
6

(Table 4). No benzyl alcohol conversion was achieved in the absence of catalysts 1-3. In the
presence of catalysts 1-3 (1 mol% or 2 mol%) and bases (5 mol%) in toluene under reflux,
benzyl alcohol to benzaldehyde conversion reached up to 90% (Table 4). In the presence of the
catalyst 1 (2 mol%) and the bases (5 mol%) Na CO , NaHCO , K CO and Cs CO , the yield of
2
3
3
2
3
2
3
benzaldehyde reached 44%, 33%, 63% and 89% respectively (Entries 3-6). In the presence of
catalysts 2 (2 mol%) and 3 (2 mol%) with Na CO as the base in toluene, the benzaldehyde yield
2
3
obtained was 49% and 44% respectively (Entries 9 and 15). For catalyst 3 with NaHCO , the
3
yield further increased to 58% (Entry 16). When catalysts 2 (2 mol%) and 3 (2 mol%) were
refluxed with K CO as the base in toluene, a 60% yield of the benzaldehyde was achieved in
2
3
both systems (Entries 11 and 17). However, overall catalyst 1 with Cs CO as base under reflux
2
3
in toluene gave the best yield to benzaldehyde (89%) (Entry 6).
When the reaction was carried out with different solvents, the benzaldehyde yield varied.
In the presence of K CO as base and p-xylene as a solvent, the maximum benzaldehyde yields
2
3
of 29% and 56% were achieved over catalysts 1 and 2 respectively (Entries 19 and 22). In the
presence of all three catalysts, the use of 1,4 dioxane and DMSO as solvents gave a lower yield
of benzaldehyde. The yield of benzaldehyde obtained from all the reactions using different
solvents and K CO as base are summarized in Table 4 (Entries 19-27). However, in the case of
2
3
catalyst 3, the yield of benzaldehyde was as high as 76% (Entry 25), likely due to the higher
solubility of catalyst 3 in p-xylene than catalysts 1 and 2.
1
7

Table 4
Screening and optimization for the oxidation of benzyl alcohol .
a
c
c
Entry
Catalyst
Base
None
None
Conversion (%)
Yield (%) TON
b
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
1
1
1
2
2
24
24
39
44
33
63
89
36
38
49
25
60
60
16
26
44
58
60
45
29
8
24
20
22
17
32
45
36
19
26
13
32
31
16
13
24
30
32
23
16
5
39
44
34
64
90
Na CO3
2
NaHCO3
K CO3
2
Cs CO3
2
b
None
None
36
38
52
26
63
62
Na CO3
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
NaHCO3
K CO3
2
Cs CO3
2
b
None
None
16
26
48
60
63
46
Na CO3
2
NaHCO3
K CO3
2
Cs CO3
2
d
K
2
K
2
K
2
K
2
K
2
CO
CO
CO
CO
CO
31
3
3
3
3
3
e
9
f
25
23
56
20
13
29
12
d
58
e
24
1
8

f
2
2
2
2
4
5
6
7
2
3
3
3
K
2
K
2
K
2
K
2
CO
CO
CO
CO
3
3
3
3
22
78
27
40
21
76
24
37
11
39
14
20
d
e
f
a
b
c
f
Benzyl alcohol (1 mmol) and catalyst (2 mol%) in toluene (20 mmol) under reflux 48 hrs.
Benzyl alcohol (1 mmol) and catalyst (1 mol%) in toluene (20 mmol) under reflux 48 hrs.
d
e
Determined by GC (average by two runs). p-xylene (20 mmol). 1,4 dioxane (20 mmol).
DMSO (20 mmol). Turnover number (TON) = [(mol of product)/(mol of catalyst)].
3
.5.2. Effect of substituents on the oxidation of benzyl alcohol over the catalysts 1 and 3
Catalyst 1 showed good activity for oxidizing substituted benzyl alcohols in the presence
of Cs CO as base and toluene as solvent (Table 5). It is evident that substrates with groups
2
3
present in the para position (Entries 2, 5, 7 and 8) were oxidized much faster than those with
ortho substituted groups present in benzyl alcohol, in the presence of base [24, 44] (Entries 3 and
6
), likely due to steric effects. The bulky ortho substitution could cause a decrease in the benzyl
alcohol oxidation rate [46] (Entries 2 and 5). On the other hand, bulky 3,4 and 2,5-OCH group
3
substituted benzyl alcohol also showed moderate conversion with catalyst 1 (Entries 9 - 11). In
contrast, para dimethylamino benzyl alcohol showed lower conversion due to the high electron
donating ability of dimethylamino group on the aromatic ring (Entry 12). The electronic nature
of the aromatic ring present in the substrate plays a big role in oxidation reactions. The efficient
catalytic activity of the iridium complex (1) could be explained by the electronic effect of the
ligand on the iridium metal. The N,N’ ligand gives a greater donor strength than in the rhodium
and ruthenium complexes [47]. The theoretical natural atomic populations study for the iridium
complex coincides well with this statement (Section 3.6). Catalyst 1 showed 100% and 97%
(Entries 2 and 8) conversion of electron rich and poor benzyl alcohols.
1
9

Table 5
a
Catalytic oxidation of benzyl alcohol derivatives mediated by catalyst 1 .
c
Entry
Benzyl alcohol
derivaties
R – H
Conversion
Yield (%)
TON
c
(%)
b
1
2
3
4
5
6
7
8
9
90
89
98
80
95
89
78
89
96
70
46
62
45
50
42
50
46
40
45
49
36
24
32
17
R –4-NO2
100
b
R –2-NO2
83
R –2-OH
100
91
R –4-Cl
b
R –2-Cl
80
R –4-CH3
90
97
R –4-OCH3
R –3,4 –OCH3
R –2,5 –OCH3
R –4-OH 3–OCH3
R –4–N(CH3)2
b
72
b
1
1
1
0
1
2
48
64
b
33
30
a
b
Substrate (1 mmol) in toluene (20 mmol) under reflux 24 hrs. Substrate (1 mmol) in toluene
c
(
20 mmol) under reflux 48 hrs. Determined by GC (average by two runs). Turnover number
(TON) = [(mol of product)/(mol of catalyst)].
Catalyst 3 showed up to 96% (Entry 4, Table 6) conversion in the oxidation of primary
alcohols to their corresponding aldehydes using K CO as base with p-xylene as the solvent
2
3
(Table 6). In this catalytic system, the electron rich para substituted benzyl alcohols (Entries 7, 8
and 11) oxidized faster than electron poor para substituted benzyl alcohols (Entries 2 and 5). The
bulky groups present in the ortho position of benzyl alcohol caused lower conversion of these
2
0

substituted benzyl alcohols (Entries 3, 6 and 10). A plausible coordination site for the substrate is
the electronically unsaturated ruthenium center, which is generated after liberation of the labile
chloride atom. The base K CO is expected to promote the formation of the metal-alkoxide
2
3
species, which could then undergo β-hydride elimination to form a ruthenium-hydride complex
48, 49]. A faster oxidation rate of the ortho-hydroxy benzyl alcohol to the corresponding
[
aldehyde in both catalytic systems 1 and 3 was observed (Entry 4, Table 5 and Entry 4, Table 6).
Table 6
a
Catalytic oxidation of benzyl alcohol derivatives mediated by catalyst 3
c
c
Entry
Benzyl
alcohol Conversion (%) Yield (%)
TON
derivatives
R –H
b
1
2
3
4
5
6
7
8
9
78
45
40
96
65
50
77
85
68
51
91
76
44
37
92
63
47
76
84
64
48
90
-
39
23
20
48
33
25
39
43
34
26
46
-
b
b
R –4-NO2
R –2-NO2
R –2-OH
b
b
R –4-Cl
R –2-Cl
R –4-CH3
R –4-OCH3
R –3,4–OCH3
R –2,5-OCH3
R –4-OH 3–OCH3
R –4–N(CH3)2
b
b
1
1
1
0
1
2
b
no reaction
a
b
Substrate (1 mmol) in p-xylene (20 mmol) under reflux for 24 hrs. Substrate (1 mmol) in
c
p-xylene (20 mmol) under reflux for 48 hrs. Determined by GC (average by two runs).
Turnover number (TON) = [(mol of product)/(mol of catalyst)].
2
1

3
.5.3. Proposed catalytic reaction mechanism for oxidation of primary alcohols
A proposed reaction mechanism (supported by theoretical calculations, Section 3.6) for
the oxidation of primary alcohols by catalysts 1-3 is illustrated in Scheme 1. The initial step of
this reaction cycle is the formation of a weakly bound metal complex with benzyl alcohol (A)
formed from the free metal complex and benzyl alcohol. Step 1 of this catalytic cycle is the
generation of the metal benzyloxo species (A)TS from the active complex (A). In step 2, β-
1
hydrogen elimination occurs to form the dehydrogenated product of benzaldehyde and a metal
hydride (B) complex. After this metal-hydride step, two possible transition states could be
involved in this reaction cycle. In step 3, the metal hydride (B) intermediate promotes a
protonolysis to form a favored transition state of the metal complex (C)TS intermediate.
2
Alternatively, hydride release from the metal hydride complex (B) could form the disfavored
-
transition state involving 16e species (D)TS in step 4. The key steps of this catalytic cycle are
3
the generation of the metal hydride intermediate in step 2. A similar kind of iridium hydride
intermediate has been characterized and reported by Heiden et al. [50] and Fujita et al. [46].
Thus, the overall catalytic cycle involves two important steps: the formation of the metal
benzyloxo species (A)TS and the metal hydride species (B). The Gibbs free energy profile
1
diagram is presented in Fig. 9.
2
2

Scheme 1. Proposed reaction mechanism of an oxidation of primary alcohol by catalysts 1-3
2
3

3
.6. Density Functional Theory (DFT) Calculations
DFT calculations were carried out to understand structural properties, such as type of
molecular energy levels, bond lengths, bond angles and the binding nature of the metal
complexes. A Gibbs free energy calculation on the rhodium catalytic cycle was carried out to
support the proposed mechanism given in Scheme 1. The energy difference between the HOMO
(highest occupied molecular orbitals) and the LUMO (lowest unoccupied molecular orbitals)
orbitals is called the energy gap, which is important to understand the stability of the structures
of the metal complexes. The analysis of the orbitals indicates that the electron absorption
corresponds to the transition state from the ground state to the first excited state, which mainly
describes the one electron excitation from the HOMO to the LUMO, which is correlated to the
energy gap and the chemical reactivity. A larger energy gap between HOMO and LUMO
indicates good stability of the structure and lower reactivity, and a smaller energy gap difference
implies higher reactivity of the complexes [51] (Figs. 6 and 8).
The HOMO is highly concentrated over the metal centers, benzene or the Cp* ring, but
with a small contribution from the Cl atom. Consequently, the LUMO is localized over the N1
and N2 atoms. The HOMO is constituted of a π-orbital in the Cp* ring or benzene. The LUMO is
dominated by the π* orbital of the pyridine ring. This suggests that the higher value of the
HOMO of a d orbital of the metal has a tendency to donate electrons to the p orbital of the N1,
N2 acceptor atoms. Calculated bond lengths and bond angles are given in the Table 7. The
experimentally obtained results of bond distances and bond angles are close to the theoretical
values, except for the angle C(5)-Ir(1)-Cl(1). The lone pair electrons of the Cl coordinated with
the iridium center are placed closer to the N1 and N2 atoms, as compared to the rhodium (2) and
ruthenium (3) complexes, which causes repulsion between Cl and the nitrogen atoms. Thus, the
DFT method could not predict the C(5)-Ir(1)-Cl(1) bond angle accurately, due to the presence of
the lone pair electrons of the Cl [52].
The calculated energy gaps from HOMO-LUMO for the iridium complex (1), rhodium
complex (2) and ruthenium complex (3) were 120.59 Kcal/mol, 130.67 Kcal/mol and 124.49
Kcal/mol respectively. The higher energy gap of the rhodium complex (2) indicates that it is
more stable and less reactive when compared to the iridium and ruthenium complexes. Lowest
2
4

vibrational frequency calculations have been carried out to predict the minimum energy structure
-1
with the lowest vibrational frequency for the Ir(III) complex (17.731 cm ), the Rh(III) complex
-1
-1
(
30.332 cm ) and the Ru(II) complex (3.797 cm ) and this is shown in the Supporting
Information S2.
NPA analysis describes the charge transfer from chloride or nitrogen atoms to the metal
Ir(III), Rh(III) and Ru(II)) ions, which plays a pivotal role in determining metal complex
(
properties (Fig. 7). The atomic charges of each atom in the complexes collected from NPA are
presented in the supporting information S3, the metal ions gain charges of 2.79e (Ir(III)), 2.73e
(Rh(III)) and 1.94e (Ru(II)) in the metal complexes. The positive charge of the metal ions
decreases significantly in all the complexes and especially the iridium metal ion is strongly co-
ordinated with neighboring atoms. The bidentate ligand N,N’ donor electron strength on the
iridium (N1 = -0.441 & N2 = -0.405) is higher than on rhodium and ruthenium and this would
account for the iridium complex being more reactive than the other complexes. On the
complexes 1-3, the chloride group transfers higher charges of 0.66e to the Ru(II) metal ion when
compared with lesser charge transfer from chloride (0.59e) to the Rh(III) and 0.61e to the Ir(III)
metal ion (S5, Table 3). The electron densities of the Cl group attached to the iridium (0.212) and
ruthenium (0.057) are less than that of the Cl group attached to the rhodium (0.273). These
electron density values confirm that the Rh-Cl (2) bond is stronger and this would account for it
being a less reactive complex compared to the latter two complexes [53].
The proposed reaction mechanism for the complexes is given in Scheme 1 and its frontier
orbital diagram is shown in Fig. 8. The iridium benzyloxo (A) complex has a higher energy gap
(116.59 Kcal/mol) than the iridium hydride (B) complex (106.83 Kcal/mol), which indicates that
A is more stable (less reactive) than complex B. A similar iridium hydride (B) complex has been
studied and reported as being a highly active species by Fujita et al. [46].
Gibbs free energy calculations of the rhodium catalytic cycle thus support the proposed
mechanism (Scheme 1). The result of the Gibbs free-energy ꢀꢁꢂꢃDFT calculation for the
reaction mechanism of catalyst 2 is shown in Fig. 9. Gibbs free-energy ꢀꢁꢂꢃprofile calculations
rationalize the stability and reactivity of various transformations of catalyst 2 as per the proposed
2
5

mechanism. In the reaction mechanism, the Rh-H step was processed via β-hydride elimination
TS ) with a high activation free energy barrier of 15.19 Kcal/mol. The calculated activation free
(
1
energy barrier of the Rh-H intermediate (B) was lower (10.32 Kcal/mol) than the corresponding
Rh-benzyloxo (A) species, which correlates to the rate-determining step of the proposed reaction
cycle. The Rh-H intermediate is the thermodynamically more favored stable state in the proposed
reaction mechanism. The calculated Gibbs free energy barrier for the Rh-H species (10.32
Kcal/mol) was in good agreement with a reported value for a similar species [53]. The calculated
-
activation free energy barrier of the 16e Rh complex (D)TS was higher (31.85 Kcal/mol) than
3
the respective Rh complex (C)TS (20.34 Kcal/mol), which indicates that the proposed reaction
2
-
mechanism is more favorable through the (C)TS transition state than the (D)TS 16e Rh
2
3
complex (D)TS . The calculated Gibbs free energy values of the reaction intermediates are in the
3
sequence: TS > B (stable state) < (C)TS < (D)TS , which confirms that the calculated low
1
2
3
activation free energy transition intermediates (Rh-H species) can facilitate the oxidation of
aromatic benzyl alcohols to their respective aldehydes. The Gibbs free-energy profile ꢀꢁꢂ and
the structures of the optimized rhodium catalytic cycle transition states are given in the
supporting information S14.
Fig. 6. Molecular orbital diagrams of the metal complexes 1, 2 and 3.
2
6

Fig. 7. Natural atomic populations charges of complexes 1, 2 and 3.
Fig. 8. Molecular orbital diagrams of catalyst 1 reaction intermediates.
2
7

Fig. 9. Gibbs free energy profile of the Rh catalyst (2) catalyzed reaction mechanism shown in
Scheme 1.
2
8

Table 7
Selected bond lengths (Å), bond angles (°) and calculated DFT data for metal complexes 1, 2 and
3
.
Crystal
data
Crystal
data
Complex 1
Crystal data DFT
Complex 2
DFT
Complex 3
DFT
Distances (Å)
Distances (Å)
Distances (Å)
Ir(1) – C1 (Cp*)
Ir(1) – C2 (Cp*)
Ir(1) – C3 (Cp*)
Ir(1) – C4 (Cp*)
Ir(1) – C5 (Cp*)
2.174(3)
2.204(3)
2.205(3)
2.153(3)
2.171(3)
2.229
2.265
2.233
2.256
2.268
1.858
2.468
2.077
2.059
Rh(1) – C1 (Cp*)
Rh(1) – C2 (Cp*)
Rh(1) – C3 (Cp*)
Rh(1) – C4 (Cp*)
Rh(1) – C5 (Cp*)
Rh(1) – Centroid (Cp*)
Rh(1) – Cl(1)
2.157(3)
2.192(3)
2.192(3)
2.136(3)
2.157(3)
1.789
2.267
2.286
2.271
2.224
2.275
1.970
2.439
2.064
2.085
Ru(1) – C1 (arene)
Ru(1) – C2 (arene)
Ru(1) – C3 (arene)
Ru(1) – C4 (arene)
Ru(1) – C5 (arene)
2.192(4)
2.192(4)
2.185(4)
2.192(4)
2.171(4)
2.192(4)
1.681
2.42
2.405
2.387
2.393
2.374
2.388
1.766
2.426
2.063
2.049
Ir(1) – Centroid (Cp*) 1.803
Ru(1) – C6 (arene)
Ru(1) – Centroid
Ir(1) – Cl(1)
Ir(1) – N(1)
Ir(1) – N(2)
2.400(10)
2.393(8)
2.106(2)
2.119(2)
(arene)
2.096(3)
2.106(2)
Rh(1) – N(1)
Ru(1) – Cl(1)
2.393(10)
2.093(3)
2.084(3)
Rh(1) – N(2)
Ru(1) – N(1)
Ru(1) – N(2)
Angles (°)
Angles (°)
Angles (°)
N(1)-Ir(1)-Cl(1)
N(2)-Ir(1)-Cl(1)
N(1)-Ir(1)-N(2)
C(1)-Ir(1)-Cl(1)
C(2)-Ir(1)-Cl(1)
C(3)-Ir(1)-Cl(1)
C(4)-Ir(1)-Cl(1)
C(5)-Ir(1)-Cl(1)
84.76(7)
86.67(8)
76.21(10)
155.43(9)
116.77(9)
93.29(8)
103.85(9)
141.65(8)
84.251
86.710
77.117
N(1)-Rh(1)-Cl(1)
N(2)-Rh(1)-Cl(1)
N(1)-Rh(1)-N(2)
85.88(7)
87.74(6)
76.58(9)
155.14(8)
116.74(8)
92.76(8)
103.32(8)
141.06(8)
125.68
86.533
N(1)-Ru(1)-Cl(1)
N(2)-Ru(1)-Cl(1)
N(1)-Ru(1)-N(2)
C(1)-Ru(1)-Cl(1)
C(2)-Ru(1)-Cl(1)
C(3)-Ru(1)-Cl(1)
C(4)-Ru(1)-Cl(1)
C(5)-Ru(1)-Cl(1)
C(6)-Ru(1)-Cl(1)
86.34(9)
86.12(9)
76.99(12)
83.31
91.300
86.807
78.165
77.932
156.906 C(1)-Rh(1)-Cl(1)
122.115 C(2)-Rh(1)-Cl(1)
153.452
117.742
91.904
123.09(14) 120.139
94.29(13)
89.23(14)
93.505
90.532
93.316
97.505
C(3)-Rh(1)-Cl(1)
C(4)-Rh(1)-Cl(1)
100.223
136.014
124.258
112.12(16) 113.687
149.46(16) 148.727
160.77(13) 154.42
131.308 C(5)-Rh(1)-Cl(1)
Centroid (Cp*)-Ir(1)-
Cl(1)
Centroid (Cp*)-Rh(1)-
1
26.13
122.930
Cl(1)
Centroid (arene)-
Ru(1)-Cl(1)
1
27.19
127.429
2
9