Nickel complexes based on hyperbranched salicylaldimine ligands: Synthesis, characterization, and catalytic properties for ethylene oligomerization

Article abstract of DOI:10.1016/j.jorganchem.2016.08.022

A series of new nickel complexes based on hyperbranched salicylaldimine ligands have been prepared in good yields and characterized by FT-IR,¹H NMR, UV, ESI-MS and TGA. Upon activation with methylaluminoxane (MAO) or diethylaluminum chloride (DEAC), these new precatalysts showed high activity in ethylene oligomerization, and notably remarkably high selectivity of high carbon olefins in the presence of DEAC. The catalytic performance was substantially affected by the solvent type, reaction conditions and the structure of catalyst. Under optimized conditions, precatalyst C1 led to TOF?=?13.65?×?10⁵?g/mol Ni·h and 21.53% selectivity for high carbon olefins. The result of thermal analysis revealed that the complex C1 was stable up to 298.4?°C. The research result of the kinetics of ethylene oligomerization found that the pressure drop of ethylene consumption sharply increased in the initial stage, however with the prolonged reaction time, the pressure drop of ethylene consumption varied a little.

Full text of DOI:10.1016/j.jorganchem.2016.08.022

Journal of Organometallic Chemistry 822 (2016) 104e111
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Nickel complexes based on hyperbranched salicylaldimine ligands:
Synthesis, characterization, and catalytic properties for ethylene
oligomerization
Jun Wang, Na Zhang, Cui Qin Li*, Wei Guang Shi, Zhi Yu Lin
Provincial Key Laboratory of Oil & Gas Chemical Technology, College of Chemistry & Chemical Engineering, Northeast Petroleum University, Daqing 163318,
Heilongjiang Province, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of new nickel complexes based on hyperbranched salicylaldimine ligands have been prepared in
good yields and characterized by FT-IR, 1H NMR, UV, ESI-MS and TGA. Upon activation with methyl-
aluminoxane (MAO) or diethylaluminum chloride (DEAC), these new precatalysts showed high activity in
ethylene oligomerization, and notably remarkably high selectivity of high carbon olefins in the presence
of DEAC. The catalytic performance was substantially affected by the solvent type, reaction conditions
and the structure of catalyst. Under optimized conditions, precatalyst C1 led to TOF ¼ 13.65 ꢀ 105 g/mol
Ni$h and 21.53% selectivity for high carbon olefins. The result of thermal analysis revealed that the
complex C1 was stable up to 298.4 ꢁC. The research result of the kinetics of ethylene oligomerization
found that the pressure drop of ethylene consumption sharply increased in the initial stage, however
with the prolonged reaction time, the pressure drop of ethylene consumption varied a little.
© 2016 Published by Elsevier B.V.
Received 1 April 2016
Received in revised form
2 August 2016
Accepted 23 August 2016
Available online 24 August 2016
Keywords:
Hyperbranched macromolecule
Salicylaldimine ligand
Nickel complex
Ethylene oligomerization
Catalytic performance
Kinetics
1. Introduction
However the very first synthesis of branched system was reported
€
by Vogtle and co-worker [11] until 1978, and used the term“cascade
€
a
-olefins can undergo many transformations, which include
macromolecules”. The innovative work of Vogtle led to the syn-
thesis of the first macromolecular hyperbranched polyphenylene
system that was reported in 1990 by Kim and Webster [12e14].
Subsequently numerous hyperbranched macromolecules with
hydrogenation, cracking, cyclization, isomerization, oligomeriza-
tion and polymerization reactions [1]. The oligomerization of
ethylene is an important catalytic process for the production of
linear
a-olefins, which are important substances extensively used
different
structure
were
synthesized
[15e23].
Metal-
in the preparation of lubricants, plasticizers, surfactants, and oil
chemicals, as well as comonomers for the synthesis of linear low-
density polyethylene, therefore this process receives great inter-
est both in academia and industry fields [2e4]. Since the discovery
of Shell Higher Olefins Process (SHOP) nickel (Ⅱ) based catalysts,
metal nickel (Ⅱ) complexes have attracted much attention because
they are less sensitive to protonic solvents and polar monomers
[5e9].
Hyperbranched macromolecules are a relatively new class of
macromolecules, which have attracted significant attention due to
their unique architectures, special chemical properties and physical
properties since Flory's seminal theoretical reported in 1952 [10].
lohyperbranched molecules can be obtained by incorporating
metal ions into infrastructure of hyperbranched molecular either as
cores, branching centres, arm connectors or termini. Metal-
lohyperbranched molecules have applications in a large number of
areas such as drug delivery, molecular electronics, sensors and
catalysis.
We previously reported the synthesis of hyperbranched nickel
complex derived from 1.0 generation (1.0G) hyperbranched sali-
cylaldimine ligand. The complex was evaluated as a catalyst pre-
cursor in ethylene oligomerization with MAO as a co-catalyst. The
results showed that the complex possessed a middle catalytic ac-
tivity of 5.59 ꢀ 105 g/mol$Ni$h with a selectivity for high carton
olefins (C10eC18) of 91% [24]. Based on above study, we designed
and synthesized a series of novel hyperbranched salicylaldimine
nickel complexes in this paper. The influence of the catalytic sys-
tems and catalytic conditions on the catalytic activity and products
* Corresponding author.
E-mail address: dqpilicuiqin@126.com (C.Q. Li).
http://dx.doi.org/10.1016/j.jorganchem.2016.08.022
0022-328X/© 2016 Published by Elsevier B.V.

J. Wang et al. / Journal of Organometallic Chemistry 822 (2016) 104e111
105
distribution were also studied.
2. Experimental section
4H,CH2), 3.583 (t, 4H, CH2), 8.313 (s, 2H, CH), 7.288 (d, 2H, Ar-H),
6.989 (t, 2H, Ar-H), 7.260 (t, 2H, Ar-H), 6.915 (d, 2H, Ar-H), 7.213
(s, 2H, OH). lmax (nm): 224, 253, 314.
Ligand L3 was prepared according to the method described for
L1 using 1.0G hyperbranched macromolecule S3 (5.81 g, 0.01 mol)
and salicylaldehyde (4.30 mL, 0.04 mol). Ligand L2 was obtained as
a yellow powder in 19% yield. IR (KBr, cmꢂ1): 3300 (s), 2916 (s),
1638 (s), 1423 (m), 1581 (w), 1278 (m), 751 (m). 1H NMR (400 MHz,
2.1. Materials and general considerations
All synthetic manipulations were performed under a nitrogen
atmosphere using standard Schlenk techniques. Toluene, cyclo-
hexane and chlorobenzene were analytical grade and dried by
refluxing over sodium/benzophenone and distilled under nitrogen
prior to use. Methylaluminoxane (MAO, 10 wt% in toluene) was
purchased from Sigma-Aldrich. Diethylaluminum chloride (DEAC,
25 wt% in hexane) were obtained from Aladdin and used as
received. Salicylaldehyde was provided by Tianjin Guangfu Fine
Chemical Industry Research Institute. Nickel Chloride Hexahydrate
was purchased from Tianjin Jizhun chemical reagent co., Ltd. 1,4-
CDCl3, ppm): d: 0.886 (t, 3H,CH3), 1.226 (m, 14H, CH2), 1.413 (m, 2H,
CH2), 1.571 (m, 2H, CH2), 1.698 (m, 2H, CH2), 2.417 (m, 4H, CH2),
2.734 (t, 2H, CH2), 2.339 (t, 4H, CH2), 6.865 (m, 2H,NH), 3.254 (m,
4H,CH2), 3.583 (t, 4H, CH2), 8.314 (s, 2H, CH), 7.293 (d, 2H, Ar-H),
7.046 (t, 2H, Ar-H), 7.263 (t, 2H, Ar-H), 6.921 (d, 2H, Ar-H), 7.228
(s, 2H, OH). lmax (nm): 226, 253, 316.
2.3. Synthesis of metal complexes
butanediamine,
1,6-hexamethylenediamine
and
1,8-
octanediamine were obtained from Tianjin Kermel Chemical Re-
agent co., Ltd. The above reagents were used without any further
purification. 1.0G hyperbranched macromolecules were synthe-
sized according to the literature procedures [25] and the yields of
them were above 80%.
The methanol solution (15 mL) of NiCl2$6H2O (4.75 g, 0.02mol)
was added dropwise to the methanol solution (15 mL) of L1 (5.31 g,
0.01 mol) under nitrogen atmosphere. The reaction mixture was
stirred for 24 h at 25 ꢁC. The precipitate was obtained after ether
added to the resultant solution. The resulting precipitate was
filtered off, washed three times with ether and dried in vacuum to
furnish the pure product as a green power in 80% yield (Scheme. 1).
IR (KBr, cmꢂ1): 3260 (w), 2925 (s), 1628 (m),1474 (m),1315 (m), 759
(m). lmax (nm): 221, 240. ESI-MS: m/z: 734 [MþH]þ.
Infrared spectra were recorded in a KBr disc matrix using a
Bruker Vector 22 IR spectrophotometer over the range of
4000e450 cmꢂ1 1H-NMR spectra were recorded with Bruker-
.
400 MHz NMR at 400 MH in CDCl3 with tetramethyl silane (TMS) as
an internal reference. The UVevisible spectra were carried out on a
UV-1700 ultravioletevisible spectrophotometer by using methanol
as a solvent. Electrospray ionization mass spectrometry (ESI-MS)
datas were collected on an micrOTOF-Q II mass spectrometer. The
thermal gravimetric analysis (TGA) was carried out on a Diamond
TG/DTA Perkin-Elmer SII Instrument over the range of 28e800 ꢁC
with the heating rate of 10 ꢁC/min. Gaschromatography (GC) ana-
lyses of oligomers were performed on a Fuli GC9720 equipped with
Complex C2 was prepared from ligand L2 (5.87 g, 0.01 mol) and
NiCl2$6H2O (4.75 g, 0.02 mol) following the procedure described
for complex C1 as a green power in 76% yield. IR (KBr, cmꢂ1): 3407
(w), 2925 (s), 1621 (m), 1447 (m), 1317 (m), 757 (m). lmax (nm): 223,
239 ESI-MS: m/z: 791 [MþH]þ.
Complex C3 was prepared from ligand L3 (6.43 g, 0.01 mol) and
NiCl2$6H2O (4.75 g, 0.02 mol) following the procedure described
for complex C1 as a green power in 71% yield. IR (KBr, cmꢂ1): 3385
(w), 2925 (s), 1621 (m),1446 (m),1315 (m), 756 (m). lmax (nm): 222,
241. ESI-MS: m/z: 847 [MþH]þ.
a flame ionization detector (FID) and a 50 m (0.2 mm i.d., 0.5
film thickness) HP-PONA column.
mm
2.2. Synthesis of ligands
2.4. General procedure for ethylene oligomerization reaction
Methanol was added to the mixture of 1.0G hyperbranched
macromolecule S1 (4.69 g, 0.01 mol) and Na2SO4 (3 g) which was
added to remove the water in the Schiff base reaction under a ni-
trogen atmosphere at 25 ꢁC and stirred for 15 min, then the sali-
cylaldehyde (4.30 mL, 0.04 mol) was added dropwise to the
solution. The mixture was refluxed for 12 h at 65 ꢁC. The mixture
was filtered to remove Na2SO4 and attained the ligand solution. The
precipitate was obtained after the ligand solution crystallized for
15 d at ꢂ30 ꢁC. The resulting precipitate was filtered off, and then
washed three times with ether. The product was dried in vacuum to
afford the ligand L1 as a yellow powder in 24% yield (Scheme 1). IR
(KBr, cmꢂ1): 3414 (s), 2917 (s), 1635 (s), 1581 (w), 1436 (m), 1280
Ethylene oligomerization reactions were carried out in a 250 mL
stainless steel reactor with magnetic stirring. The reactor was
heated under vacuum for 2 h at 160 ꢁC and subsequently allowed to
cool to the room temperature. The reactor was flushed with
ethylene three times. Solvent, the desired amount of co-catalyst,
and solution of the metal complex (0.7 mmol/mL, 10 mL) (The to-
tal volume was 50 mL) were added to the reactor in this order
under an ethylene atmosphere, when the desired reaction tem-
perature was reached, the ethylene pressure was increased to
desired value, and maintained at this level by constant feeding of
ethylene. After 30 min, the reaction was stopped by releasing the
excess ethylene. A small amount of the reaction solution was
collected, the reaction was terminated by the addition of 5%
aqueous hydrogen chloride, and then this mixture was analyzed by
gas chromatography (GC) to determine the distribution of oligo-
mers obtained.
(m), 753 (m). 1H NMR (400 MHz, CDCl3, ppm):
d: 0.874 (t, 3H,CH3),
1.228 (m, 14H, CH2), 1.413 (m, 2H, CH2), 1.572 (m, 2H, CH2), 1.680 (m,
2H, CH2), 2.405 (m, 4H, CH2), 2.714 (t, 2H, CH2), 2.321 (t, 4H, CH2),
6.860 (m, 2H,NH), 3.241 (m, 4H,CH2), 3.582 (t, 4H, CH2), 8.312 (s, 2H,
CH), 7.288 (d, 2H, Ar-H), 6.976 (t, 2H, Ar-H), 7.258 (t, 2H, Ar-H),
6.935 (d, 2H, Ar-H), 7.216 (s, 2H, OH). lmax (nm): 231, 252, 315.
Ligand L2 was prepared according to the method described for
L1 using 1.0G hyperbranched macromolecule S2 (5.25 g, 0.01 mol)
and salicylaldehyde (4.30 mL, 0.04 mol). Ligand L2 was obtained as
a yellow powder in 25% yield. IR (KBr, cmꢂ1): 3301 (s), 2917 (s),
1639 (s), 1424 (m), 1582 (w), 1274 (m), 751 (m). 1H NMR (400 MHz,
3. Results and discussion
3.1. Characterization of metal complexes
3.1.1. FT-IR spectroscopy
IR spectra for the hyperbranched salicylaldimine ligands and
metal complexes are shown in Fig. 1. For the ligands, the absorption
peaks at around 3301 cmꢂ1 were due to the stretching vibration of
the N-H group. The characteristic peak at around 2917 cmꢂ1
CDCl3, ppm): d: 0.875 (t, 3H,CH3), 1.229 (m, 14H, CH2), 1.413 (m, 2H,
CH2), 1.591 (m, 2H, CH2), 1.680 (m, 2H, CH2), 2.406 (m, 4H, CH2),
2.716 (t, 2H, CH2), 2.322 (t, 4H, CH2), 6.861 (m, 2H,NH), 3.242 (m,

106
J. Wang et al. / Journal of Organometallic Chemistry 822 (2016) 104e111
Scheme 1. Synthetic routes of metal complexes.
Fig. 1. IR spectra of the hyperbranched salicylaldimine ligands and nickel complexes.
corresponded to the asymmetric stretching vibration of eCH2e in
the alkyl chain. The peak at around 1638 cmꢂ1 were attributable to
the stretching vibration of C]N. The absorption peaks at around
1581 cmꢂ1 were assigned to the stretching vibration of benzene
ring, which indicated that the Schiff base reaction was happened
between 1.0G hyperbranched macromolecules and salicyladehyde.
Compared with the IR spectra of ligands, the C]N stretching vi-
bration in complexes C1eC3 were shifted to lower frequencies at
around 1621 cmꢂ1, indicating an effective coordination interaction
between the imino nitrogen atom and the nickel center.
around 6.935e7.288 ppm. In addition, the signal for the protons in
hydroxyl connected to benzene occurred around 7.261 ppm. Above
signal for the protons indicated that the Schiff base reaction
occurred between 1.0G hyperbranched macromolecules and
salicyladehyde.
3.1.3. UV spectroscopy
UV spectra of hyperbranched salicylaldimine ligands and nickel
complexes are shown in Fig. 3. The absorption spectra of hyper-
branched salicylaldimine ligands were characterized by three
intense absorption bands. The bands at around 231 nm and 252 nm
assigned to the R band of n/
of benzene ring respectively, and the band at around 315 nm cor-
responded to the R band of n/
*transition for C]N. After nickel
p
*transition for C]O and the B band
3.1.2. 1H NMR spectroscopy
1H NMR spectra of the hyperbranched salicylaldimine ligands
are showen in Fig. 2. The signal for the protons in the methyl and
methylene groups which belong to the alkyl chains could be found
p
center coordinating with the imino nitrogen atom, the R band of
the C]O and the band B of the benzene ring were blue shifted to
at around 0.874e1.680 ppm. The signal for the protons in
a-
around 221 nm and 240 nm, and the R band of n/
C]N disappeared.
p
*transition for
methylene connected to tertiary amine group was evidenced at
around 2.405 ppm and 2.714 ppm, respectively. The signal for the
protons in methylene connected to carbonyl group was exhibited at
around 2.231 ppm. The signal for the protons in methylene group
connected to secondary amine group was evidenced by the peak at
around 3.241 ppm. The signal for the protons in the secondary
amine group was indicated at around 6.860 ppm. The signal for the
protons in the methylene connected to imine and benzene occurred
around 8.312 ppm. The signal for the protons in aryl rings occurred
3.1.4. ESI-MS spectroscopy
ESI mass spectra for the metal nickel complexes are shown in
Fig. 4. The ESI mass spectra of C1, C2 and C3 showed molecular ion
peaks at m/z: 734, 791 and 847, respectively. Further molecular
fragmentation was consistent with various fragments of the com-
plexes. Fragmentation peaks due to elimination of Ni were

J. Wang et al. / Journal of Organometallic Chemistry 822 (2016) 104e111
107
Fig. 2. 1H NMR spectra of the hyperbranched salicylaldimine ligands.
Fig. 3. UV spectra of the hyperbranched salicylaldimine ligands and nickel complexes (a: UV spectra of the hyperbranched salicylaldimine ligands. b: UV spectra of the nickel
complexes).
observed at 676, 732 and 788 in the ESI mass spectra of all the
complexes. These values were in good agreement with the pro-
posed composition for the metal nickel complexes.
mass loss of about 2.5% for complex C1. This was a drying stage up
to 298.4 ꢁC, and surface moisture and inherent moisture were
emitted in this stage. The second stage occurred in the rage of
298.4e507.7 ꢁC, and the weight loss was 54.7%, which may attri-
bute to the degradation of the 1.0G hyperbranched macromolecule.
The third stage occurred beyond 507.7 ꢁC, and the weight lost was
about 26.7%. At about 640.0 ꢁC, the complex decomposed
completely. This part of weightlessness was due to the thermal
decomposition of salicylaldehyde. When the temperature exceeded
3.1.5. TG analyses
The thermal stability of the complex C1 was investigated by the
thermal gravimetric analysis (TGA). As showed in Fig. 5, the thermal
degradation process of the catalyst precursor can be subdivided
into three stages based on the TGA profile. The first stage showed a

108
J. Wang et al. / Journal of Organometallic Chemistry 822 (2016) 104e111
Fig. 4. MS spectra of the hyperbranched salicylaldimine nickel complexes.
solvent in order to probe the effect of the co-catalyst on these
ethylene oligomerization reactions. Table 1 shows a summary of
the ethylene oligomerization data obtained. From the results, it was
clear that the type of aluminium co-catalysts had a significant effect
on the catalytic activities of the precatalysts and product distribu-
tion. When activated with MAO, all the complexes showed lower
activities. However when activated with DEAC, the catalyst showed
higher activity. The products were mainly low carbon olefins acti-
vating with MAO and DEAC. In addition, as can be seen from Table 1
that the content of high carbon olefins with MAO as co-catalyst was
less than with DEAC as co-catalyst. For example, using MAO as co-
catalyst, the content of high carbon olefins were from 2.19% to 8.72%
(Entries 1e3) compared to from 3.21% to 16.25% (Entries 4e6) using
DEAC as co-catalyst.
3.2.2. Effect of solvent on the catalytic activity and oligomer
distribution
Considering the role of solvent in determining the catalytic ac-
tivity and product distribution in ethylene oligomerization re-
actions, the catalytic behaviour of complexes C1eC3 was further
investigated using cyclohexane and chlorbenzene solvents as
shown in Table 2. The oligomerization reaction in chlorbenzene
solvent produced more activities compared to the same reactions in
cyclohexane. This could be attributed to improved solubility of the
precatalysts in chlorobenzene leading to higher catalytic activity.
GC analyses showed that the products were mainly of C4 oligomers
with cyclohexane and chlorbenzene as solvents and the content of
C4 were all above 80%.
Fig. 5. Thermal gravimetric curve of the nickel complex C1.
640.0 ꢁC, the thermal gravimetric curve went to mild and was
almost unchanged. The result of thermal analysis revealed that the
complex C1 was stable up to 298.4 ꢁC.
3.2. Evaluation of the metal complexes as catalysts for ethylene
oligomerization reactions
The effect of the nature of solvent on catalytic activity in
ethylene oligomerization is apparent from Tables 1 and 2. For
example, catalytic activities of complex C1 of 2.01 ꢀ 105 g/(mol
Ni$h), 5.80 ꢀ 105 g/(mol Ni$h) and 6.69 ꢀ 105 g/(mol Ni$h) were
reported in cyclohexane, toluene and chlorobenzene, respectively.
This enhanced catalytic activity in chlorobenzene could be attrib-
uted to improved solubility of the precatalysts in chlorobenzene
compared to in cyclohexane and toluene, which are less polar
solvents. As revealed in Tables 1 and 2, the solvent significantly
affect the catalytic activity, however, no obvious influence on the
product distribution could be observed.
The catalytic activities of complexes C1eC3 were evaluated in
ethylene oligomerization using methylaluminoxane (MAO) and
diethylaluminum chloride (DEAC) as co-catalysts in three different
solvents: toluene, cyclohexane and chlorobenzene. In all cases, the
complexes showed appreciable catalytic activities, and GC analyses
confirmed that the products were mainly of low carbon olefins. A
notable observation was the dependence of catalytic activity and
product distribution on the nature of co-catalyst and solvent used
as well as reaction parameters. In the subsequent sections, we
present detailed analyses of the products and discussion of our
findings.
3.2.1. Effect of co-catalyst on the catalytic activity and oligomer
distribution
3.2.3. Effect of reaction parameters on the catalytic activity and
oligomer distribution
Complexes C1eC3 were activated by MAO and DEAC in toluene
Complex C1 was typically investigated in detail under a range of

J. Wang et al. / Journal of Organometallic Chemistry 822 (2016) 104e111
109
Table 1
Effect of co-catalysts on the catalytic activity and product distribution.a
Entry
Catalysts
Co-catalysts
Activity 105 g/mol Ni$h
Distribution of oligomers (%)
b
c
C4
C6
C8
ꢃC8
ꢄC10
1
2
3
4
5
6
C1
C2
C3
C1
C2
C3
MAO
MAO
MAO
DEAC
DEAC
DEAC
1.92
1.06
0.99
5.80
5.43
4.92
53.02
71.99
81.25
71.63
86.47
87.60
11.62
10.09
9.40
9.74
6.36
7.33
32.43
15.21
7.16
2.38
1.41
1.81
97.07
97.29
97.81
83.75
94.24
96.74
2.93
2.71
2.19
16.25
5.76
3.26
a
Reaction conditions: 7 m
mol catalyst, Al/Ni 500, toluene 50 mL, 25 ꢁC, 0.5 MPa ethylene, 30 min.
Less than 8 carbon atoms in oligomers, low carbon olefins.
More than 10 carbon atoms in oligomers, high carbon olefins.
b
c
Table 2
Effect of solvents on the catalytic activity and product distribution.a
Entry
Catalysts
Solvent
Activity 105 g/mol Ni$h
Distribution of oligomers (%)
b
c
C4
C6
C8
ꢃC8
ꢄC10
1
2
3
4
5
6
C1
C2
C3
C1
C2
C3
cyclohexane
cyclohexane
cyclohexane
chlorbenzene
chlorbenzene
chlorbenzene
2.01
0.96
0.83
6.69
6.13
5.96
81.87
80.88
83.36
82.59
84.47
84.43
6.79
7.07
6.16
7.46
6.82
6.21
1.96
4.24
8.77
3.97
3.79
6.80
90.62
92.19
98.29
94.02
95.08
97.44
9.38
7.81
1.71
5.98
4.92
2.56
a
Reaction conditions: 7 m
mol catalyst, co-catalyst DEAC, Al/Ni 500, solvent 50 mL, 25 ꢁC, 0.5 MPa ethylene, 30 min.
Less than 8 carbon atoms in oligomers, low carbon olefins.
More than 10 carbon atoms in oligomers, high carbon olefins.
b
c
reaction conditions, such as reaction temperature, molar ratio of
co-catalyst to catalyst and ethylene pressure. The oligomerization
results are summarized in Table 3. As revealed in Table 3, the re-
action parameters significantly affected the catalytic activity and
the product distribution.
With complex C1, the reaction temperature significantly
affected the catalytic activity. Increasing the temperature from
15 ꢁC to 35 ꢁC (Entries 1e3 in Table 3), the catalytic activity of C1
increased obviously. However, when the reaction temperature was
further increased (Entries 3e4 in Table 3), the activity decreased.
The highest activity (10.96 ꢀ 105 g/mol Ni$h, entry 3 in Table 3) was
observed at 35 ꢁC. This may be ascribed to decomposition of the
active species and lower ethylene solubility at higher temperature
[26]. Moreover, the selectivity for high carbon olefins decreased at
elevated temperatures (Entries 1e4 in Table 3), indicating that
greater than chain growth rate.
The amount of DEAC played an important role on the catalytic
performance. Increasing the Al/Ni molar ratio from 300 to 1000
(Entries 5e8 in Table 3), the catalytic activity of C1 increased
obviously and the highest activity (16.08 ꢀ 105 g/mol Ni$h, entry 7
in Table 3) was observed at Al/Ni molar ratio of 1000, which may be
attributed to the fact that DEAC scavenged adventitious water and
impurities in the solvent at low Al/Ni molar ratio and the nickel
complex required more co-catalyst to be activated. However, when
Al/Ni molar ratio was further increased (Entry 5 in Table 3), the
activity decreased. The decreased in oligomerization activity might
be caused by the impurities in DEAC such as alkyl aluminium,
which led to the deactivation of catalytic sites [27]. Not only the
activity was sensitive to the Al/Ni molar ratio, but also the selec-
tivity for high carbon olefins was alterant in different ratios. The
distribution of the oligomers had a trend to move to the high car-
bon number olefins with increasing of Al/Ni molar ratio, and the
under the actions of the
b-hydrogen elimination and the chain
propagation in the oligomerization process, chain transfer rate was
Table 3
Effect of reaction parameters on the catalytic activity and product distribution.a
Entry
Catalyst
Pressure (MPa)
Al/Ni
Temperature (ꢁC)
Activity (105 g/mol Ni$h)
Distribution of oligomers (%)
b
c
C4
C6
C8
ꢃC8
ꢄC10
1
2
3
4
5
6
7
8
C1
C1
C1
C1
C1
C1
C1
C1
C1
C1
C1
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.1
0.3
1.0
500
500
500
500
300
700
1000
1500
500
500
500
15
25
35
45
35
35
35
35
35
35
35
2.24
5.80
10.96
9.93
10.23
13.98
16.08
13.65
1.15
71.44
71.63
72.91
74.94
81.58
70.98
70.93
66.45
77.97
81.33
72.42
7.08
9.74
4.33
2.38
1.83
2.07
1.69
1.68
1.70
1.81
4.20
2.00
2.61
82.85
83.75
88.87
92.35
97.09
88.75
1.70
78.42
90.03
90.59
95.80
17.15
16.25
11.13
7.65
14.13
15.34
13.82
16.09
13.13
10.16
7.86
2.91
11.25
14.24
21.58
9.97
9.41
4.20
9
10
11
3.74
26.27
10.26
20.77
a
Reaction conditions: 7 mmol catalyst, co-catalyst DEAC, toluene 50 mL, 30 min.
Less than 8 carbon atoms in oligomers, low carbon olefins.
More than 10 carbon atoms in oligomers, high carbon olefins.
b
c

110
J. Wang et al. / Journal of Organometallic Chemistry 822 (2016) 104e111
from 0.1 MPa to 1.0 MPa (Entries 9e11 in Table 3) resulted in an
increase in activity from 1.15 ꢀ 105 g/mol Ni$h to 26.27 ꢀ 105 g/mol
Ni$h. We considered that this behaviour was principally due to
ethylene solubility dependence on pressure. In addition, the
selectivity for high carbon olefins decreased from 9.96% to 4.20%.
Decreased selectivity for high carbon olefins with the increase in
ethylene pressure could be due to the fact that the rate of chain
transfer reactions, especially the
greater than chain growth rate.
b-hydride elimination rate was
3.2.4. Effect of catalyst structure on the catalytic activity and
oligomer distribution
The influence of catalytic structure on catalyst behaviour was
also studied. From all the catalytic reactions investigated, we
observed a significant influence of catalyst structure on the cata-
lytic activity. It was apparent that the catalytic activity decreased
with the increase of molecular cavity. The complex C1 with smaller
molecular cavity displayed higher catalytic activities compared to
the other complexes (Tables 1e3). For example, in toluene with
MAO as co-catalyst, the complex C1 exhibited higher activity of
1.92 ꢀ 105 g/mol Ni$h compared to the activity of 0.99 ꢀ 105 g/mol
Ni$h displayed by the complex C3. A similar trend had also been
obtained in toluene, cyclohexane and chlorbenzene with DEAC as
co-catalyst. This could be explained that the length of carbon chain
increased with growing of molecular cavity, so the solubility of the
complex decreased, which led to the decrease of the concentration
of the effective active species in the catalytic system.
Fig. 6. Pressure drop of ethylene consumption vs reaction time for the ethylene
oligomerization catalyzed by hyperbranched salicylaldimine nickel complexes (7
catalyst, co-catalyst MAO, Al/Ni 500, toluene 50 mL, 30 min).
mmol
selectivity for high carbon olefins increased from 2.91% to 21.53%.
The effect of ethylene pressure on the catalytic activity and the
product distribution of complex C1 was examined at different
ethylene pressures. As expected, the role of ethylene pressure in
reaction system was very important. Increasing ethylene pressure
Not only the activity was sensitive to the structure of complexes,
Fig. 7. Reaction mechanism for ethylene oligomerization by hyperbranched salicylaldimine nickel complexes.

J. Wang et al. / Journal of Organometallic Chemistry 822 (2016) 104e111
111
but also there was distinct regularity in the selectivity for high
carbon olefins (Tables 1 and 2). For example, on DEAC activation in
toluene, the content of high carbon olefins of 16.25% with C1 as
precatalyst compared to the content of 5.77% with C2 as precatalyst,
and the content of high carbon olefins of 3.21% with C3 as pre-
catalyst. The trend was the same as obtained in cyclohexane and
chlorbenzene with DEAC as co-catalyst. The possible reasons for
this phenomenon were that the crimp and tangle of alkyl chain
were more likely to happen with the increase of molecular cavity,
increasing the steric hindrance of catalytic active species.
selectivity to high carbon olefins. After investigating the catalytic
condition, the highest catalytic activity was obtained as
26.27 ꢀ 105 g/mol$Ni$h, and the content of high carbon olefins
reached up to 21.53%. Catalyst structure significantly affected the
catalytic activity and oligomer distribution. With the increase of
molecular cavity, the catalytic activity and the content of high
carbon olefins decreased. In the reactions of ethylene oligomeri-
zation, the pressure drop of ethylene consumption sharply
increased in the initial stage and then varied a little with the pro-
longed reaction time. Moreover, the nickel complex had favorable
thermal stability and was thermally stable up to a temperature
higher than 298.4 ꢁC in an atmosphere of air.
3.3. The kinetics of ethylene oligomerization
The kinetics of ethylene consumption for hyperbranched sali-
cylaldimine nickel complexes catalytic systems is shown in Fig. 6.
The pressure drop of ethylene consumption increased sharply
within 200s after injecting catalyst to a solution containing co-
catalyst and ethylene monomer. However, the pressure drop of
ethylene consumption increased slowly and was smoothly kept on
after 200s. The possible reason for this phenomenon was that a
large number of active species generated after injecting the catalyst
to the catalytic system in an extremely short time, and the reaction
of ethylene insertion to active species led to the ethylene con-
sumption increasing. However, the ethylene consumption
Acknowledgement
We acknowledge the funding from the Natural Science Foun-
dation of China (No. 21576048), Petroleum Innovation Foundation
of China (No. 2014D-5006-0503) and Youth Science Foundation of
Northeast Petroleum University (2013NQ130).
References
[1] I. Coleto, R. Roldan, C. Jimenez-Sanchidrian, J.P. GomeZ, F.J. Romero-Salguero,
Catal. Today 149 (2010) 275e280.
[2] M. Peuchert, W. Keim, Organometallics 2 (1983) 594e597.
[3] L. Chen, J. Hou, W.H. Sun, Appl. Cat. A Gen. 246 (2003) 11e16.
[4] W. Keim, Chem. Int. Ed. 52 (2013) 12492e12496.
[5] M.K. Ainooson, S.O. Ojwach, I.A. Guzei, L.C. Spencer, J. Darkwa, J. Org. Chem.
696 (2011) 1528e1535.
[6] R. Gao, W.H. Sun, C. Redshaw, Catal. Sci. Technol. 3 (2013) 1172e1179.
[7] A. Budhai, B. Omondi, S.O. Ojwach, C. Obuah, E.Y. Osei-Twum, J. Darkwa, Catal.
Sci. Technol. 3 (2013) 3130e3135.
decreased due to the reaction of b-hydrogen elimination and the
deactivation of active species with the prolonged reaction time.
3.4. The mechanism for ethylene oligomerization
Recently, metal complexes have gained increasing attention and
they are widely studied in catalytic processes including the process
of ethylene oligomerization. Considering the results of this study
and previous reports, a proposed reaction mechanism is presented
in Fig. 7. An active species (B) is formed upon a reaction between
hyperbranched salicylaldimine nickel complex and co-catalyst
DEAC. The mechanism involved an initiation step via the coordi-
nation of an ethylene molecular to B to give an alkyl nickel complex
(C), which was followed by a second ethylene molecule insertion
into the nickel-hydrogen bond and formation of a new alkyl nickel
complex (D). Subsequent ethylene molecules insertion led to chain
[8] S. Wang, W.H. Sun, C. Redshaw, J. Org. Chem. 751 (2014) 717e741.
[9] A. Budhai, B. Omondi, S.O. Ojwach, C. Obuah, E.Y. Osei-Twum, J. Darkwa, Catal.
Sci. Technol. 3 (2013) 3130e3135.
[10] P.J. Flory, J. Am, Chem. Soc. 74 (1952) 3364e3367.
[11] E. Buhleier, W. Wehner, F. Vogtle, Synthesis 1978 (1978) 155e158.
[12] Y.H. Kim, J. Polym. Sci. Pol. Chem. 36 (1988) 1685e1698.
[13] Y.H. Kim, O.W. Webster, J. Am. Chem. Soc. 112 (1990) 4592e4593.
[14] Y.H. Kim, O.W. Webster, Macromolecules 25 (1992) 5661.
[15] N. Yamaguchi, J.S. Wang, J.M. Hewitt, W.C. Lenhart, T.H. Mourey, J. Polym. Sci.
Part A Polym. Chem. 40 (2002) 2855e2867.
[16] A. Blencowe, L. Davidson, W. Hayes, Eur. Polym. J. 39 (2003) 1955e1963.
[17] L. Hong, Y. Cui, X. Wang, X. Tang, J. Polym. Sci. Part A Polym. Chem. 40 (2002)
344e350.
[18] P.A. Costello, I.K. Martin, A.T. Slark, D.C. Sherrington, A. Titterton, Polymer 43
(2002) 245e254.
growth to give complex (E), E could undergo b-hydrogen elimina-
[19] T.T. Hsieh, C. Tiu, G.P. Simon, Polymer 42 (2001) 1931e1939.
[20] S.E. Stiriba, H. Kautz, H. Frey, J. Am. Chem. Soc. 124 (2002) 9698e9699.
[21] C. Kavitha, K.P. Dasan, J. Coat. Technol. Res. 10 (2013) 669e678.
[22] Q. Li, H.R. Xiong, L. Pang, Q.H. Li, Y. Zhang, W.Q. Chen, Z.S. Xu, C.F. Yi, High.
Perform. Polym. 27 (2015) 426e438.
tion to yield the linear polyethylene (F) and 1-alkyl nickel complex
(C).
4. Conclusions
[23] P. Liu, A. Gu;G, Z. Liang, Q.B. Guan, L. Yuan, Prog. Org. Coat. 74 (2012)
142e150.
In this contribution, a series of new nickel complexes based on
hyperbranched salicylaldimine ligands have been synthesized in a
simple route and acceptable yields, as well as evaluated in ethylene
oligomerization reactions. Using MAO and DEAC as co-catalysts,
these complexes displayed higher catalytic activities and
[24] J. Wang, G. Yang, C.Q. Li, W.G. Shi, Chem. J. Chin. U. 35 (2014) 1536e1540.
[25] J. Wang, C.Q. Li, S.Y. Zhang, F. Sun, T. Ge, Chin. Chem. Lett. 19 (2008) 43e49.
[26] Y.Y. Wang, S.A. Lin, F.M. Zhu, H.Y. Gao, Q. Wu, Inorganica Chim. Acta 362
(2009) 166e172.
[27] G.S. Nyamato, M.G. Alam, S.O. Ojwach, M.P. Akerman, J. Organomet. Chem.
783 (2015) 64e72.