Cobalt complexes based on 2-(1H-benzimidazol-2-yl)-phenol derivatives: preparation, spectral studies, DFT calculations and catalytic behavior toward ethylene oligomerization

Article abstract of DOI:10.1080/00958972.2017.1328506

In this study, five novel Co(II) complexes of 2-(1H-benzimidazol-2-yl)-phenol derivatives (HL_x: x?=?1–5) have been synthesized and characterized. The general formula for complexes C1 and C2 is K₂[Co(HL_1,2)₂Cl

Full text of DOI:10.1080/00958972.2017.1328506

Journal of Coordination Chemistry
ISSN: 0095-8972 (Print) 1029-0389 (Online) Journal homepage: http://www.tandfonline.com/loi/gcoo20
Cobalt complexes based on 2-(1H-benzimidazol-2-
yl)-phenol derivatives: Preparation, spectral
studies, DFT calculations and catalytic behavior
toward ethylene oligomerization
Marzieh Haghverdi, Azadeh Tadjarodi, Naeimeh Bahri-Laleh & Mahdi
Nekoomanesh Haghighi
To cite this article: Marzieh Haghverdi, Azadeh Tadjarodi, Naeimeh Bahri-Laleh & Mahdi
Nekoomanesh Haghighi (2017): Cobalt complexes based on 2-(1H-benzimidazol-2-yl)-phenol
derivatives: Preparation, spectral studies, DFT calculations and catalytic behavior toward ethylene
oligomerization, Journal of Coordination Chemistry, DOI: 10.1080/00958972.2017.1328506
To link to this article: http://dx.doi.org/10.1080/00958972.2017.1328506
View supplementary material
Accepted author version posted online: 08
May 2017.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=gcoo20
Download by: [FU Berlin]
Date: 10 May 2017, At: 07:41

Publisher: Taylor & Francis
Journal: Journal of Coordination Chemistry
DOI: http://dx.doi.org/10.1080/00958972.2017.1328506
Cobalt complexes based on 2-(1H-benzimidazol-2-yl)-phenol derivatives:
Preparation, spectral studies, DFT calculations and catalytic behavior
toward ethylene oligomerization
MARZIEH HAGHVERDI†, AZADEH TADJARODI*†, NAEIMEH BAHRI-LALEH‡ and
MAHDI NEKOOMANESH HAGHIGHI‡
†
Research Laboratory of Inorganic Materials Synthesis, Department of Chemistry, Iran University of Science
and Technology, Tehran, Iran
Polymerization Engineering Department, Iran Polymer and Petrochemical Institute (IPPI), Tehran, Iran
‡
In this study, five novel Co(II) complexes of 2-(1H-benzimidazol-2-yl)-phenol derivatives (HLx:
x = 1-5) have been synthesized and characterized. The general formula for complexes C1 and C2
is K [Co(HL ) Cl ]·H O, for complex C3 K [Co(HL ) Cl ], and for complexes C4 and C5
2
1,2 2
2
2
2
3 2
2
[
Co(HL ) ]. In all complexes, the ligands are coordinated as bidentate, via one imine nitrogen
4,5 2
and the phenolate oxygen atoms. The structures of the compounds were characterized by FT-IR,
1
13
UV-Vis, H-, C-NMR spectroscopies, ICP and elemental analysis (C, H and N). The purity of
these compounds was ascertained by melting point (m.p.) and TLC. Geometry optimization of
the studied complexes was done by Gaussian09 software at B3LYP/TZVP level of theory and
satisfactory theoretical–experimental agreement was achieved for NMR and IR spectra of the
compounds. Based on the combined experimental and theoretical studies, six-coordinate
octahedral structures have been proposed for complexes C1-C3, while complexes C4 and C5
had distorted tetrahedral geometry. All complexes were activated with diethylaluminum chloride
(
Et AlCl), cobalt(II) complexes containing bulky methyl groups in the aryl moiety show high
2
-1
-1
catalytic activities (1774 kg·mol (Co)·h ) for ethylene oligomerization. The oligomers obtained
from the cobalt complexes exhibit good selectivity for linear 1-butene and 1-hexene. Results
revealed that both the steric and electronic effects of ligands strongly affect the catalytic
activities and the properties of the catalytic products.
*Corresponding author. Email: tajarodi@iust.ac.ir

Keywords: Benzimidazolylphenols; Cobalt(II) complexes; Characterization; DFT calculations;
Ethylene oligomerization
1
. Introduction
α-Olefins are major industrial reactants that are extensively used in the preparation of detergents,
lubricants, plasticizers, oil field chemicals, and comonomers used for branched polyolefins [1].
The novel late transition metal catalysts compared with traditional Ziegler-Natta catalysts of
early transition metals possess advantages not only in catalytic activity and selectivity, but also
in their potential for tolerating heteroatom functionalities [2]. Over the past decade, progress has
been made in ethylene reactivity by late-transition metal complex catalysts. Regarding iron and
cobalt systems based on bis(arylimino)pyridines, those initially discovered by the groups of
Brookhart and Gibson individually are well known [3].
After that, relative research works have been focused on modifying those catalysts’
models through synthesizing numerous derivatives of their complexes and also devising
alternative models of these catalysts [4]. However, there are few models of iron and cobalt
complex catalysts performing good catalytic activities toward ethylene oligomerization and
polymerization in comparison with various models of nickel complexes. Among different models
of these catalysts, the catalysts containing bi- and tridentate ligands for polymerization and
oligomerization of ethylene to α-olefins has attracted much attention.
Among bidentate ligands, there are reports about the use of benzimidazolyl-phenol and
its derivatives as chelating agents with various transition metal complexes [5-10], that as well as
having good catalytic activities toward ethylene oligomerization. Studies on characterization of
these types of the compounds and their complexes are rather scarce. Therefore, the preparation of
these compounds becomes more important [11]. A general method for their synthesis is the
condensation of o-phenylenediamines with 2-hydroxy aromatic aldehydes [12].
In this work, the synthesis and characterization of the Co(II) complexes with various
1
H-benzimidazol-2-yl-phenol derivatives are reported. Density functional theory (DFT)
calculations of the electronic properties of the compounds were performed which were in good
agreement with the experimental results. Also, the catalytic behavior of Co(II) complexes in
ethylene reactivity was investigated. Relatively good catalytic activities toward ethylene
oligomerization could be observed in the presence of diethylaluminum chloride (Et AlCl).
2

2. Experimental
2
.1. Chemistry and apparatus
All manipulations of air- and moisture-sensitive compounds were performed under nitrogen
using standard Schlenk techniques. Toluene was refluxed over sodium-benzophenone and
distilled under nitrogen prior to use. Diethylaluminum chloride (Et AlCl) was purchased from
2
Aldrich Chemicals. All other chemicals were obtained commercially and used without
purification. Elemental analyses were carried out in a Perkin Elmer model 2400 series 2 (made in
USA) automatic carbon, hydrogen, nitrogen analyzer. Determination of metal percentage was
1
performed using ICP (3410ARL model made in Swiss). NMR spectra (300 MHz for H and
1
3
7
5 MHz for C) were run on a Bruker AVANCE-300 instrument (Germany) in
dimethylsulfoxide (DMSO-d ) using tetramethylsilane (TMS) as internal standard. IR spectra
6
were recorded in KBr disks on a Shimadzu FT-IR 8400 spectrometer (Japan) from 400-
-1
4
000 cm . UV-Vis spectra were taken on a Shimadzu UV-1700 spectrophotometer (Japan) in
DMSO solution. GC analyses were performed with a Varian CP-3800 gas chromatograph
equipped with a flame ionization detector and a 100 m (0.25 mm i.d., 0.5 µm film thickness)
CP-Sil PONA CB. The yield of oligomers was calculated by referencing with the mass of the
solvent on the basis of the prerequisite that the mass of each fraction was approximately
proportional to its integrated areas in the GC trace. Selectivity for linear 1-alkenes was defined as
(amount of linear 1-alkenes of all fractions)/(total amount of oligomer products) in percent.
Melting points were determined using an electrothermal melting point apparatus. The completion
of reactions was monitored by TLC.
2
.2. Synthesis of the ligands: General procedure
The HL to HL ligands (see scheme 1 for labels) were prepared according to previous literature
1
5
procedures [13]. In a typical procedure, 2-(5,6-dimethyl-1H-benzimidazol-2-yl)-phenol (HL5)
was obtained by the reaction of 2-hydroxybenzaldehyde (1.83 g, 15.0 mmol) with sodium
metabisulfide (1.60 g, 8.5 mmol) at room temperature in 20 mL of ethanol and water,
respectively. The reaction mixture was stirred vigorously and more ethanol was added. The
mixture was kept in a refrigerator for several hours. Then, it was filtered and sodium hydroxyl-
(
2-hydroxyphenyl)methansulfonate salt was obtained from the crude extract. This crude salt and

4,5-dimethylbenzene-1,2-diamine (0.54 g, 4.0 mmol) in 10 mL of dimethylformamide (DMF)
were refluxed for 4 h at 130 °C in an oil bath. The completion of reaction was monitored by
TLC. Then the mixture was poured in an ice bath. After that, the reaction was quenched by
adding water and obtained ligand filtered and recrystallized from ethanol. This method was
selected because of its ease, high yield and short synthesis time. This intermediate compound
(namely, bisulfite compound of the aldehyde) was utilized to catalyze the benzimidazole
synthesis since 1,2-phenylenediamine derivatives and aldehydes do not give cyclization reaction
by themselves under mild conditions and possibly might lead to Schiff bases. 2-(5,6-Dimethyl-
1
H-benzimidazol-2-yl)-phenol was used as ligand for the synthesis of the complexes.
2
2
.3. Synthesis of the cobalt complexes C1-C5
.
.3.1. K [Co(HL ) Cl ] H O (C1). To a schlenk tube that was purged three times with nitrogen,
2
1 2
2
2
a solution of HL (0.106 g, 0.5 mmol) and KOH (0.028 g, 0.5 mmol) in methanol was added.
1
Then, a solution of CoCl ·6H O (0.059 g, 0.25 mmol) in methanol was added. The color of the
2
2
solution changed immediately. The reaction mixture was stirred for 24 h at room temperature
under nitrogen. Completion of the reaction was monitored by TLC. The resulting precipitation
was separated by filtration, washed with diethyl ether and dried in vacuum. Brick red powder,
-1
yield: 62%. IR (KBr disc, cm ): 3425 v(N-H), 3180 v(C-H), 1625 v(C=N), 1533 v(C=C), 1247
v(C-O), 667, 574 v(Co-O), 450 v(Co-N). Anal. Calcd. for C H N O K Cl Co: C, 48.45; H,
2
6
20
4
3
2
2
3
.10; N, 8.69; Co, 9.15. Found: C, 47.23; H, 3.77; N, 9.93; Co, 9.40%.
.
.3.2. K [Co(HL ) Cl ] H O (C2). C2 was prepared in a manner similar to that described for
2
2
2 2
2
2
-1
C1. Greenish brown powder, yield: 41.29%. IR (KBr disc, cm ): 3436 v(N-H), 3155 v(C-H),
625 v(C=N), 1550 v(C=C), 1245 v(C-O), 667, 597 v(Co-O), 424 v(Co-N). Anal. Calcd. for
C H N O K Cl Co: C, 43.76; H, 2.52; N, 7.85; Co, 8.26. Found: C, 40.48; H, 1.83; N, 7.30;
1
26
18
4
3
2
4
Co, 8.57%.
2
.3.3. K [Co(HL ) Cl ] (C3). C3 was prepared in a manner similar to that described for C1.
2
3 2
2
-1
Light violet powder, yield: 43%. IR (KBr disc, cm ): 3440 v(N-H), 3116 v(C-H), 1633 v(C=N),
552 v(C=C), 1251 v(C-O), 671, 578 v(Co-O), 422 v(Co-N). Anal. Calcd. for
1

C H N O K Cl Co: C, 51.38; H, 3.36; N, 8.56; Co, 9.01. Found: C, 57.05; H, 3.74; N, 9.41;
28
22
4
2
2
2
Co, 10.30%.
2
.3.4. [Co(HL ) ] (C4). C4 was prepared in a manner similar to that described for C1. Greenish
4
2
-1
brown powder, yield: 50%. IR (KBr disc, cm ): 3419 v(N-H), 3056 v(C-H), 1625 v(C=N), 1550
v(C=C), 1242 v(C-O), 675, 534 v(Co-O), 457 v(Co-N). Anal. Calcd. for C H N O Cl Co: C,
2
6
14
4
2
4
5
0.73; H, 2.27; N, 9.10; Co, 9.58. Found: C, 49.34; H, 1.98; N, 8.87; Co, 9.28%.
2
.3.5. [Co(HL ) ] (C5). C5 was prepared in a manner similar to that described for C1. Light
5
2
-1
brown powder, yield: 53%. IR (KBr disc, cm ): 3431 v(N-H), 3058 v(C-H), 1625 v(C=N), 1548
v(C=C), 1249 v(C-O), 690, 553 v(Co-O), 478 v(Co-N). Anal. Calcd. for C H N O Co: C,
3
0
26
4
2
6
7.55; H, 4.87; N, 10.50; Co, 11.05. Found: C, 65.18; H, 4.51; N, 10.02; Co, 11.44%.
2
.4. Computational details
DFT calculations were performed with the Gaussian 09 package using the B3LYP (hybrid GGA
functional of Becke-Lee, Parr, and Yang [13,14]) level of theory, since as reported before, in the
molecular simulation of transition metal containing systems better agreement with the
experimental results can be obtained with this method [15-19]. In all cases, the electronic
configuration of the molecular systems was described with the triple-ζ basis set augmented with
one polarization functions of Ahlrichs and coworkers (TZVP keyword in Gaussian) [20]. For all
the computed cobalt complexes, the quartet containing three unpaired electrons was the
electronic ground state and thus open shell calculations were performed.
The vibrational frequency calculations were performed to ensure that the optimized
geometries represent the global minima and that there are only positive eigen-values [21-23].
The optimized geometry of compounds furnished the total energy E.
2
.5. General procedure for ethylene oligomerization
Ethylene oligomerizations and polymerizations were performed in a 1L stainless steel autoclave
equipped with a mechanical stirrer and a temperature controller. Toluene, the desired amount of
diethylaluminum chloride (Et AlCl), and toluene solution of the catalyst precursor were added to
2
the reactor in this order under a nitrogen atmosphere; the total volume was 110 mL. When the

desired reaction temperature was reached, ethylene at 20 atm pressures was introduced to start
the reaction, and the ethylene pressure was maintained by a constant feed of ethylene during the
reaction period. After 30 min, the reaction was stopped, and the autoclave was cooled. Then the
pressure was released and a small amount of the reaction solution was collected, and the organic
layer was analyzed as quickly as possible by gas chromatography (GC) to determine the
composition and mass distribution according to GC spectrum of oligomers. Then the remaining
reaction mixture was quenched with HCl-acidified ethanol (5%) in order to collect polyethylene
obtained. However, the polyethylene was not formed.
3
. Results and discussion
1
H-Benzimidazol-2-yl-phenol derivatives (HL –HL ) were prepared according to our reported
1
5
procedures [13]. The cobalt (II) complexes were obtained by mixing the methanol solution of
CoCl ·6H O and the corresponding ligands with KOH at room temperature. The precipitation
2
2
was immediately formed, which indicated the coordination of 2-(1H-benzimidazol-2-yl)-phenol
to metal complexes is fast. The precipitation of the metal complexes were separated by filtration,
washed with diethylether and dried in vacuum. The cobalt complexes (C1–C5) were obtained in
good yields (43–62%). These complexes are stable in both solution and solid state. The
analytical data and physical properties of the ligands and the cobalt complexes are summarized
in table S1.
These compounds were characterized by FT-IR and UV-Vis spectra, ICP and elemental
analysis (C, H and N). It has been tried hard in getting the single crystals of all complexes but the
suitable crystals were not found in order to determine crystal structure of these compounds.
Therefore, further investigation for direct molecular structures of complexes by DFT calculations
was performed. Theoretical calculations involving geometry optimization, vibrational
1
13
frequencies, stability of the compounds, H- and C-NMR spectra, as well as the various bond
lengths and angles generated from the optimized structures have been done at the DFT/B3LYP
level of theory.
3
.1. Geometrical parameters
The optimized geometry of HL is shown in figure 1 as a representative ligand structure. The
2
various bond lengths and angles generated from the structures of the optimized ligand (HL ) and
2

complex (C2) showed good resemble with the X-ray structural investigations of similar ligands
and complexes reported [24-27]. The bond lengths and angles of the ligand (HL ) and complex
2
(C2) are given in table 1. The optimized structure of C2 is shown in figure 2.
3
.2. FT-IR spectra
Vibrational properties of the ligands and complexes have been performed by using the DFT
calculations with B3LYP/TZVP level. The prominent bands in the IR spectra of the cobalt
complexes are presented in table S2.
For instance, the measured and calculated FT-IR spectra of the HL ligand and its related
2
complex, C2, are compared in figure 3(a,b) and figure 4(a ,b), respectively. In the FT-IR spectra
-1
-1
-1
of C1-C5, the ꢀ(N-H) (~1477cm ), ꢀ(C=C) (~1550 cm ), ꢀ(C=N) (~1633 cm ) and ꢀ(C-O)
-1
(
~1251cm ) bonds shifted to lower wavenumbers [28], in comparison with their corresponding
-1
-1
-1
free ligands HL1-5 (ꢀ(N–H) ~1490 cm , ꢀ(C=C) ~1591 cm , ꢀ(C=N) ~1637 cm [29] and
-1
ꢀ
(C-O) ~1298 cm ), thereby indicating a coordinative interaction between the imino nitrogen
atom and the central metal. Also, ꢀ(N-H), ꢀ(C=C), ꢀ(C=N) and ꢀ(C-O) in predicted theoretical
-1
-1
-1
-1
spectra of the complexes are calculated at ~1480 cm , ~1552 cm , ~1633 cm and ~1255 cm ,
respectively. In the IR spectra of C1-C5 complexes, the imine-nitrogen coordination could also
-1
be confirmed by appearance of a weak band located at the low wavenumbers (~457 cm ), which
-1
may be assigned to (Co-N) [30, 31]. In the high frequency region, bands at 2854-3180 cm are
attributed to the stretching vibrations of aromatic ꢀ(C-H) for C1-C5 complexes [32, 33], and the
-1
corresponding peaks in theoretical spectra are calculated at the range of 3010-3172 cm . The
-1
sharp or medium bands in the complexes in the ~738-985 cm range are due to the out-of-plane
deformation bands for the aromatic C-H [5]. In the IR spectra of C1-C5 complexes,
deprotonation and subsequent involvement of the phenoxyl group in metal coordination could
-1
also be supported by the appearance of new bands at a lower frequency (~574, ~667 cm ) region
assignable to (Co-O). These results are in agreement with the theoretical calculations. The
experimental and calculated IR spectra of ligands and complexes agree extremely well with
respect to their peak frequencies, the shapes of the bands, and band intensities [34].

3
.3. UV-vis spectra
The UV-vis spectra of the ligands and the complexes were measured from 200-800 nm in DMSO
and are given in table 2. According to the spectra, the lower-wavelength bands (210-300 nm)
correspond to the π→π* transitions of the aromatic rings. The bands in the 310-360 nm region
are due to n→π* transitions [9]. The electronic spectrum of C1 shows a band at 360 nm related
to L→Co charge-transfer and two weak bands at 380 and 440 nm which may be assigned to
4
4
4
4
4
4
T g(F)→ A g(F) and T g(F)→ T g (P) transitions, respectively. T g(F)→ T g(F) transition is
1
2
1
1
1
2
not observed in the spectrum of the present complex. In most cases, it is rarely observed as it is
inherently weak due to an orbital selection rule [35]. The spectra of C2 and C3 are like the
spectrum of C1. They are characteristic for high-spin octahedral geometry for the C1-C3
complexes. The electronic spectra of complexes C4 and C5 are of little help in the present case,
since the d→d transitions are masked by the strong charge-transfer bands [36]. In the electronic
spectra of C4 and C5, there is a band at 390 and 370 nm, respectively, that may be caused by the
combination of oxygen→metal, nitrogen→metal charge-transfers (L → Co charge transfer).
DFT theoretical studies of UV-vis spectra on the synthesized complexes were performed
using the B3LYP/TZVP level of theory. The corresponding simulated absorption curves are
plotted in figure 5 and the λmax data are collected in table 2. As shown in figure 5, the calculated
absorption spectra showed strong intense features around 400 and 390 nm for C4 and C5 and
weak intense features around 390, 378 and 410 nm for C1 to C3, respectively. Obtained λmax
values were in the range of the amounts obtained by UV-vis instrument.
3
1
.4. NMR spectra of the ligands
13
1
H- and C- NMR data of the ligands are given in table 3. For instance, the H NMR spectra of
HL , HL and HL ligands exhibit only a broad singlet for the OH and the amine NH protons
3
4
5
(
13.14, 12.5 and 13.11 ppm for HL , HL and HL , respectively). This combination results from
3 4 5
strong intramolecular hydrogen bonding between the amine nitrogen with double bond and
phenolic hydrogen atoms [37]. The HL and HL ligands give two singlets for OH and NH
1
2
protons (13.07 ppm and 9.44 ppm for HL , 13.26 ppm and 12.76 ppm for HL , respectively). In
1
2
1
the H-NMR spectra of HL (having one methyl substituent at the 5-position on the
3
benzimidazole ring) and HL (having two methyl substituents at the 5,6-positions on the
5
benzimidazole ring) show a singlet at 2.42 and 2.32 ppm, respectively. Position of protons and

the assignments of the peaks in benzimidazole moiety and phenolic ring moiety in HL1-5 are
different, the reason of this may be an electron donating (CH substituents) or electron
3
withdrawing group (Cl substituents) at the same position on the benzimidazole ring. It is known
that the chloro groups on aromatic ring moves the resonance of the protons downfield, while the
1
methyl groups on ring move the resonance of these protons upfield. H-NMR spectra of the
1
3
ligands are shown in figure S1. C-NMR spectra of synthesized ligands were mentioned earlier.
1
3
For example, in the C-NMR spectrum of the HL ligands, the signals at 157.71-158.98 ppm
1
-5
and 150.83-153.77 ppm are attributed to carbon atoms bonded to the OH oxygen atom and
imidazole (C=N) carbon atom, respectively [5]. The low-signals of HL and HL , 21.29 ppm and
3
5
19.96 ppm, are due to methyl carbon atoms, respectively. The other signals are related to the
benzimidazole and phenolic ring carbon atoms.
1
13
H- and C-NMR chemical shift calculations of the HL1 ligands were carried out using
-5
1
the B3LYP/TZVP level of theory as used for geometry optimizations. The predicted H- and
1
3
C-NMR chemical shifts were derived from the equation δ = Σo – Σ, where δ is the chemical
shift, Σ is the absolute shielding and Σo is the absolute shielding of the standard (TMS) [38]. The
chemical shift values obtained were compared with the experimental data listed in table 3. The
results showed fair agreement between experimental and computed chemical shifts.
3
.5. Thermodynamic stability calculation
The most important molecular orbitals are the frontier molecular orbitals, i.e. Highest Occupied
Molecular Orbitals, HOMOs, and Lowest Unoccupied Molecular Orbitals, LUMOs. In fact,
HOMO and LUMO energy gap determines the way the molecule interacts with other species. In
other words, the frontier orbitals energy gap helps characterize the kinetic stability and chemical
reactivity of the molecule so that the larger HOMO-LUMO gap always refers to lower chemical
reactivity and higher kinetic stability [39]. The calculated HOMO-LUMO gaps of complexes C1,
C2, C3, C4 and C5 at B3LYP/TZVP level of theory were 3.1, 3.0, 3.1, 3.8 and 3.9 eV,
respectively. For these values we would say that although all the complexes are definitely stable,
C4 (figure 7) and C5 (figure 6) show higher kinetic stability.
According to the geometry optimization of the structures of C1-C3 with DFT
calculations, Co is located on an octahedral center with the C=N nitrogen and hydroxy oxygen
atoms of benzimidazole moiety and phenol ring, respectively, along with two chloride ions and

in complexes C4 and C5 is placed on a distorted tetrahedral center with one imine nitrogen and
the phenolate oxygen atoms (molecular symmetry: D for C1-C3 and C for C4 and C5).
2
h
2v
3
.6. Ethylene oligomerization
At first, toluene was transferred to the fully dried reactor under a nitrogen atmosphere. The
required amount of diethylaluminum chloride (Et AlCl) as cocatalyst was then injected into the
2
reactor via a syringe. Then, the desired cobalt complexes (10 µmol) were suspended in toluene
(10 mL). This mixture was transferred to a 1L Büchi laboratory autoclave under inert atmosphere
and thermostated at 30 °C. An ethylene pressure of 20 bar was applied and the mixture was
stirred for 30 min, then the ethylene flow was stopped. After cooling down, the ethylene pressure
was carefully released. A sample (1 mL) was taken from the cooled mixture by syringe and
applied for the GC analysis. The relative ratios of the oligomer fractions could be taken from the
gas chromatogram. In all cases, these catalysts generate butenes and hexene as oligomeric
products and the distribution of oligomers does not follow Schulz-Flory rules. No odd carbon
number oligomers were detected in the GC analysis [40].
3
.6.1. Ethylene oligomerization by complexes C1-C5. The catalytic activities of C1-C5 for
ethylene oligomerization have been carried out in the presence of Et AlCl. Complexes C1 and
2
C2 produced C and C , whereas C3-C5 produced C as main oligomeric product and showed
4
6
4
high selectivity for 1-butene. Regarding the oligomer distribution, the bulkier R1, R2
substituents increase the content of C as given in table 4.
4
Their catalytic activities were significantly affected by the ligands environment, and their
activity varied in the order C5 > C3 > C4 > C2 > C1. As it is clear, among the complexes, C1
had the lowest catalytic activity and C5 had the highest activity. The catalysts with electron-
withdrawing group (C2 and C4) showed lower activity than the catalysts with electron-donating
group. This phenomenon is caused by the solubility factors of cobalt complexes. The complexes
with ligands having additional methyl-substituents enhance their solubility, but complexes with
ligands containing halides have lower solubility. The similar phenomena were also observed
within complexes ligated by their analogues of the ligands [27].

Therefore, when the substituents were bulkier, cobalt pro-catalysts performed higher
activity. It was believed that bulkier substituents might be helpful for both the solubility of
complexes and stability of active species due to less exposure to impurity reactants [41].
Also, to clarify the correlation between the calculated effective net charge on the Co
center and the observed activities of different complexes, NBO (Natural Bond Orbital) analysis
was performed. In this regard, for all complexes, an electronic structure containing three
unpaired electrons was considered since they showed lower energy values among other spin
states. The values for NBO charges were varied from 0.944 to 0.951, 1.075, 1.064 and 1.087 for
C1 to C5, respectively. The variation trend between net charge and catalysis activity for
synthesized complexes was predicted that their activities increase along with higher net charge
on active center. This is in accordance with the finding of other authors [42, 43].
In addition, we compared our work with the literature [44-46]. In our work, bidentate
ligands were used instead of tridentate, because bidentate ligands have high thermal stability and
activity. Also, the oligomerization process is an exothermic reaction hence bidentate ligands due
to high thermal stability are suitable for these reactions. But, in the literature, more tridentate
ligands have been used.
As well as, in our work, the synthesized complexes could be easily activated with a low
amount of cocatalyst (Et AlCl) (Al/Co = 200), and exhibited good activities and selectivity for
2
ethylene oligomerization, while, in the literature, the amount of cocatalyst was very high (Al/Co
=
1000) in the ethylene oligomerization and even in the low amount of cocatalyst, activity was
low.
Furthermore, in the present work, in the oligomerization process, wax or polymer was not
formed at all and the final product is just oligomer and indicates that these catalysts are suitable
for oligomer production. But, in the literature, in addition to oligomer, low amounts of polymers
have been formed
4
. Conclusion
Co(II) complexes containing 2-(2-hydroxyaryl)-1H-benzimidazole derivatives were synthesized
1
and characterized. The structures of the ligands and complexes were characterized by FT-IR, H-
1
3
and C-NMR and UV-vis spectra, ICP, and elemental analysis (C, H and N). All the analytical
and spectral data are in agreement with theoretical calculations. According to the geometry

optimization obtained from DFT calculations, apart from C4 and C5 (with distorted tetrahedral
geometry), all complexes are completed with two chloride ions adopting a six-coordinate
octahedral geometry. Also, all complexes had the same number of three unpaired electrons and
subsequently were in high spin. Also, for these catalysts essentially dimerization and
trimerization were achieved, with a high selectivity for 1-butene and 1-hexene.
Supplementary material
Cartesian coordinates of all the species discussed in the text to together with spectral data for the
synthesized ligands are available free of charge via the Internet at the publisher’s website.
Acknowledgements
This work was supported by Iran University of Science and Technology, Iran Polymer and
Petrochemical Institute and Petrochemical Research & Technology Company. Naeimeh Bahri-
Laleh thanks MOLNAC (www.molnac.unisa.it) for its computer facilities.
References
[
1]
W.H. Sun, S. Jie, S. Zhang, W. Zhang, Y. Song, H. Ma, J. Chen. Organometallics, 25,
66 (2006).
6
[
2]
W. Hang, Y. Weidong, J. Tao, L. Binbin, X. Wenqing, M. Jianjiang, H. You liang. J.
Chin. Sci. Bull., 47, 616 (2002).
[
[
[
[
3]
4]
5]
6]
L. Xiao, R. Gao, M. Zhang, Y. Li, X. Cao, W.H. Sun. Organometallics, 28, 2225 (2009).
W.H. Sun, S. Zhang, W. Zuo. C. R. Chimie, 11, 307 (2008).
A. Tavman, I. Boz, A. Seher Birteksöz. Spectrochim. Acta, Part A, 77, 199 (2010).
A. Tavman, S. Ikiz, A. Funda. Bagcigil, N. Yakutözgür, S. Ak. Russ. J. Inorg. Chem., 55,
2
15 (2010).
A. Tavman, S. Ikiz, A. Funda Bagcigil, N. Yakutözgür, S. Ak. J. Serb. Chem. Soc., 74,
37 (2009).
[
7]
5
[
[
[
8]
9]
A. Tavman. Transition Met. Chem., 32, 172 (2007).
A. Tavman, I. Boz, A. Seher Birteksöz, A. Cinarli. J. Coord. Chem., 63, 1398 (2010).
10] A. Tavman, S. Ikiz, A. Funda Bagcigil, N. Yakutözgür, S. Ak. J. Bull. Chem. Soc.
Ethiop., 24, 391 (2010).

[
[
[
[
[
11] M. Mao, Z. Wang, J. Li, X. Song, Y. Luo. Synth. Commun., 40, 1963 (2010).
12] N. DT, G. Atyam VSSS, R. Mi, S. Raidu. J. Pharm. Chem., 2, 273 (2010).
13] C. Lee, W. Yang, R.G. Parr. J. Phys. Res, Part B, 37, 785 (1988).
14] A.D. Becke. J. Chem. Phys., 98, 1372 (1993).
15] F. Nouri-Ahangarani, N. Bahri-Laleh, M. Nekoomanesh-Haghighi. Des. Monomers
Polymers, 19, 394 (2016).
[
[
16] N. Bahri Laleh, A. Poater, L. Cavallo, S.A. Mirmohammadi. Dalton Trans., 43, 15143
(
2014).
17] N. Bahri Laleh, M. Nekoomanesh Haghighi, S.A. Mirmohammadi. J. Organomet. Chem.,
19, 74 (2012).
7
[
[
18] N. Bahri Laleh, L. Falivene, L. Cavallo. Polyolefins J., 1, 139 (2014).
19] A. Poater, L. Falivene, C.A. Urbina Blanco, S. Manzini, S.P. Nolan, L. Cavallo. Dalton
Trans., 42, 7433 (2013).
[
[
20] A. Schäfer, H. Horn, R . Ahlrichs. J. Chem. Phys., 97, 2571 (1992).
21] S. Kundu, S. Biswas, A. Sau Mondal, P. Roy, T. Kumar Mondal. J. Mol. Struct., 1100, 27
(2015).
[
[
22] M.M. Ibrahim, A.M.M. Ramadan, H.S. El Sheshtawy, M.A. Mohamed, M. Soliman, S.
Imam. J. Coord. Chem., 68, 4296 (2015).
23] R.C. Maurya, B.A. Malika, J.M. Mira, P.K. Vishwakarmaa, D.K. Rajaka, N. Jaina. J.
Coord. Chem., 68, 2902 (2015).
[
[
[
24] Q. Shi, S. Zhang, F. Chang, P. Hao, W.H. Sun. C. R. Chimie, 10, 1200 (2007).
25] L.L. Han. Acta Cryst. E, E66, o2047 (2010).
26] P. Hao, S. Zhang, W.H. Sun, Q. Shi, S. Adewuyi, X. Lu, P. Li. Organometallics, 26,
2439 (2007).
[
[
27] X. Chen, L. Zhang, T. Liang, X. Hao, W.H. Sun. C. R. Chimie, 13, 1450 (2010).
28] P.P. Hankare, L.V. Gavali, V.M. Bhuse, S.D. Delekar, R.S. Rokade. Ind. J. Chem., 43,
2578 (2004).
[
29] K. Kanmani Raja, N. Indra Gandhi, L. Lekha, D. Easwaramoorthy, G. Rajagopal. J. Mol.
Struct., 1060, 49 (2014).
[
[
30] S.S. Qian, X.L. Zhao, L.H. Pang, Z.L. You, H.L. Zhu. J. Coord. Chem., 42, 345 (2016).
31] M.M. Abd Elzaher. Appl. Organomet. Chem., 18, 149 (2004).

[
[
[
32] N. Sundaraganesan, C. Meganathan, B. Anand, C. Lapouge. Spectrochim. Acta, Part A,
6, 773 (2007).
6
33] C.A. Téllez S., A.C. Costa Jr., M.A. Mondragón, G.B. Ferreira, O. Versiane, J.L. Rangel,
G. Müller Lima, A.A. Martin. Spectrochim. Acta, Part A, 169, 95 (2016).
34] P. Saha, J. Prakash Naskar, A. Bhattacharya, R. Ganguly, B. Saha, S. Chowdhury. J.
Coord. Chem., 69, 303 (2016).
[
[
[
35] K.K. Mohammed Yusuff, R. Sreekala. Thermochim. Acta, 179, 313 (1991).
36] A. Tavman. Transition Met. Chem., 31, 194 (2006).
37] N. Sundaraganesan, B. Anand, F.-F. Jian, P. Zhao. Spectrochim. Acta, Part A, 65, 826
(
2006).
38] A.A. Dahy, E.S. Al-Farrag, N.S. Al-Hokbany, R.M. Mahfouz. J. Coord. Chem., 69, 1088
2016).
[
[
[
[
(
39] M.A.M. El-Mansy, M.M. El-Nahass, N.M. Khusayfan, E.M. El-Menyawy. Spectrochim.
Acta, Part A, 111, 217 (2013).
40] X. Tang, W.H. Sun, T. Gao, J. Hou, J. Chen, W. Chen. J. Organomet. Chem., 690, 1570
(
2005).
41] S. Song, R. Gao, M. Zhang, Y. Li, F. Wang, W. H. Sun. Inorg. Chim. Acta, 376, 373
2011).
(
[
[
[
42] W. Yang, Y. Chen, W.H. Sun. J. Macromol. React. Eng., 9, 473 (2015).
43] Y. Chen, W. Yang, R. Sha, R.D. Fu, W.H. Sun. Inorg. Chim. Acta, 423, 450 (2014).
44] J. Ma, C. Feng, S. Wang, K. Q. Zhao, W. H. Sun, C. Redshaw , G.A. Solan. Inorg. Chem.
Front., 1, 14 (2014).
[
[
45] S. Zhang, Q. Xing, W.H. Sun. RSC Adv., 6, 72170 (2016).
46] E. Yue, Y. Zeng, W. Zhang, Y. Sun, X.P. Cao, W.H. Sun. J. Sci. China Chem., 59, 1291
(2016).

Figure captions
Figure 1. Optimized structure of HL at B3LYP/TZVP level of theory.
2
Figure 2. Optimized structure of complex C2 at B3LYP/TZVP level of theory.
Figure 3. The comparison of (a) experimental FT-IR, (b) theoretical IR, calculated at
B3LYP/TZVP level, spectra for ligand HL2.
Figure 4. The comparison of (a) experimental FT-IR, (b) theoretical IR, calculated at
B3LYP/TZVP level, spectra for complex C2.
Figure 5. Simulated UV-vis absorption spectra of the complexes at B3LYP/TZVP level of
theory.
Figure 6. Optimized structure of complex C5 at B3LYP/TZVP level of theory.
Figure 7. Optimized structure of complex C4 at B3LYP/TZVP level of theory.

Table 1. Selected bond lengths (Å) and angles (°) for HL and C2.
2
HL2
C2
Bond lengths
O(1)-C(6)
1.341
0.992
1.375
1.382
1.323
1.379
Co(51)-O(1)
Co(51)- O(8)
Co(51)-N(9)
Co(51)-N(33)
Co(51)-Cl(49)
Co(51)-Cl(50)
2.109
2.109
2.196
2.196
2.523
2.513
O(1)-H(25)
N(3)-C(7)
N(3)-C(9)
N(8)-C(7)
N(8)-C(2)
Bond angles
C(6)-O(1)-H(25)
C(7)-N(3)-C(9)
C(7)-N(8)-C(2)
N(8)-C(7)-N(3)
108.69
107.80
106.51
111.24
N(9)-Co(51)-O(1)
N(9)-Co(51)-O(8)
95.60
84.44
88.66
91.33
177.32
84.44
95.60
88.66
91.33
177.96
91.02
88.98
91.01
88.98
179.99
N(9)-Co(51)-Cl(49)
N(9)-Co(51)-Cl(50)
N(9)-Co(51)-N(33)
N(33)-Co(51)-O(1)
N(33)-Co(51)-O(8)
N(33)-Co(51)-Cl(49)
N(33)-Co(51)-Cl(50)
O(1)-Co(51)-O(8)
O(1)-Co(51)-Cl(49)
O(1)-Co(51)-Cl(50)
O(8)-Co(51)-Cl(49)
O(8)-Co(51)-Cl(50)
Cl(49)-Co(51)-Cl(50)

Table 2. UV-vis spectra data of the compounds in DMSO.
UV (nm)
experimental)
250, 290
vis (nm)
vis (nm)
Compound
(
(experimental)
(calculated)
HL1
HL2
HL3
HL4
320
-
-
-
-
220, 240, 270, 290 310, 330
210, 240, 300
300
320, 340
340, 360
HL5
C1
C2
C3
C4
C5
210, 240, 290
340
-
-
-
-
-
-
360, 380, 440
340, 380
370, 390
390
390
378
410
400
390
370

1
13
Table 3. Theoretical assignment of H-NMR and C-NMR data for the ligands optimized by
B3LYP/TZVP level: the chemical shift values (δ , ppm).
H
1
1
13
13
Proton
H-NMR
H-NMR
Carbon
C-NMR
C-NMR
Ligands
HL1
assignment (experimental) (theoretical) assignment (experimental)
(theoretical)
154.20
167.80
156.00
168.00
143.00
154.00
167.00
137.00
21.00
NH
OH
NH
OH
9.44
8.45
C=N
151.35
158.98
152.85
157.87
126.59
151.40
157.98
131.54
21.29
13.07
12.76
13.26
13.20
8.40
C-OH
C=N
HL2
HL3
12.90
C-OH
C-Cl
NH
OH
CH3
13.14
13.14
2.42
8.35
13.20
2.50
C=N
C-OH
C-CH3
-CH3
HL4
HL5
NH
OH
12.50
12.50
8.25
C=N
153.77
157.71
124.99
150.83
157.92
131.31
19.96
156.00
167.70
140.50
153.80
167.00
136.00
20.10
12.62
C-OH
C-Cl
NH
OH
CH3
13.11
13.11
2.32
8.30
13.20
2.35
C=N
C-OH
C-CH3
-CH3

a
Table 4. Results of ethylene oligomerization with C1–C5/ Et AlCl .
2
c
Tim
Oligomer distribution
Activit (%)
Pressu Yiel
Entr Precatal Temperat Al/C e
re
d
b
y
yst
ure (°C)
o
(mi
n)
y
(
atm)
(g)
C4
C6
α-C4 α-C6
1
C1
30
200 30
20
0.84 168
2.30 460
41.0 58.9 87.5 73.9
3
7
0
2
2
C2
30
200 30
20
96.6 3.33 93.1 100
7
0
3
4
5
C3
C4
C5
30
30
30
200 30
200 30
200 30
20
20
20
8.62 1724
7.97 1594
8.87 1774
100
100
100
-
-
-
100
100
100
-
-
-
a
b
-1 -1 c
General conditions: 10 µmol precatalyst, 100 mL toluene as solvent; In units of kg·mol ·h ;
Determined by GC. ΣC denotes the total amounts of oligomers.
2
7

Graphical abstract
2
8