Cobalt(II)-based ethylene dimerization catalysts with silicon-bridged diphosphine ligands

Article abstract of DOI:10.1002/aoc.4604

A series of ligands and cobalt complexes were synthesized and characterized using NMR spectroscopy, elemental analysis and single-crystal X-ray diffraction. Complex C1 [CoCl₂{N-(2,6-diisopropylphenyl)-N-[(diphenylphosphino)dimethylsilyl]-1,1-diphenylphosphine}] features a monomeric structure with κ² coordination of the silicon-bridged diphosphine ligand to the cobalt center. Using dried methylaluminoxane (DMAO) and AlEt₃ as co-catalysts and methylcyclohexane as solvent, the Co(II) complex proved to be moderately active and highly selective for ethylene dimerization. The effects of ligand structure, catalyst loading, reaction temperature, Al/Co molar ratio, DMAO/AlEt₃ molar ratio and reaction time were investigated with respect to the catalytic activity and product selectivity of ethylene dimerization. Cobalt complex C1 exhibited moderate catalytic activity of 2.3?×?10⁵?g?(mol_Co)^?1?h^?1, with 100% selectivity towards C₄ (87% 1-C₄ in the C₄ fraction).

Full text of DOI:10.1002/aoc.4604

Received: 17 July 2018
DOI: 10.1002/aoc.4604
Revised: 25 August 2018
Accepted: 28 August 2018
F U L L P A P E R
Cobalt(II)‐based ethylene dimerization catalysts with
silicon‐bridged diphosphine ligands
Wei Wei1 | Buwei Yu2 | Fakhre Alam1 | Yongwang Huang1 | Shaoling Cheng1 | Tao Jiang1
1 Tianjin Marine Chemical Technology
A series of ligands and cobalt complexes were synthesized and characterized
using NMR spectroscopy, elemental analysis and single‐crystal X‐ray diffrac-
tion. Complex C1 [CoCl2{N‐(2,6‐diisopropylphenyl)‐N‐[(diphenylphosphino)
dimethylsilyl]‐1,1‐diphenylphosphine}] features a monomeric structure with
κ2 coordination of the silicon‐bridged diphosphine ligand to the cobalt center.
Using dried methylaluminoxane (DMAO) and AlEt3 as co‐catalysts and
methylcyclohexane as solvent, the Co(II) complex proved to be moderately
active and highly selective for ethylene dimerization. The effects of ligand
structure, catalyst loading, reaction temperature, Al/Co molar ratio, DMAO/
AlEt3 molar ratio and reaction time were investigated with respect to the cata-
Engineering Center, College of Chemical
Engineering and Material Science, Tianjin
University of Science and Technology,
Tianjin 300457, China
2 PetroChina Daqing Chemical
Engineering Research Center, Daqing
163714, China
Correspondence
Tao Jiang, Tianjin Marine Chemical
Technology Engineering Center, College
of Chemical Engineering and Material
Science, Tianjin University of Science and
Technology, Tianjin 300457, China.
Email: jiangtao@tust.edu.cn
lytic activity and product selectivity of ethylene dimerization. Cobalt complex
−1
C1 exhibited moderate catalytic activity of 2.3 × 105 g (molCo
)
h
−1, with
100% selectivity towards C4 (87% 1‐C4 in the C4 fraction).
Funding information
National Key Research and Development
Program of China, Grant/Award Number:
(2017YFB0306700); Natural Science
Foundation of Tianjin City, Grant/Award
Number: 14JCYBJC23100,
KEYWORDS
butene, cobalt complex, diphosphine ligand, ethylene dimerization
15JCYBJC48100, 16JCZDJC31600
1 | INTRODUCTION
Relative to nickel catalysts, cobalt complexes function
with low rates of isomerization and, consequently, high
product selectivity.[7] Therefore, many cobalt complexes
have been explored for use in ethylene oligomerization
during the last decade.[8–12]
The main application of 1‐butene is as a co‐monomer in
the production of certain kinds of polyethylene, such as
linear low‐density polyethylene. It has also been used as
a precursor to polypropylene resins, butylene oxide and
butanone.[1] At present, 1‐butene is produced mainly
through ethylene oligomerization – the most advanced
technology for the production of linear α‐olefins. Because
the nature of the catalytic system strongly governs the
catalytic activity and product selectivity, many metallic
complexes – especially nickel complexes – have been
investigated for ethylene oligomerization in the last
20 years.[2–5] The main problem when using nickel cata-
lysts is the high rate of isomerization during ethylene
oligomerization; in turn, this also affects the product
selectivity.[6] Therefore, many other metallic complexes
have been scrutinized for ethylene oligomerization.
Thiele and de Souza[13] prepared a series of Co(II)‐
based catalysts bearing N^N^N‐type ligands and obtained
catalytic activities of up to 3.4 × 105 g (molCo −1 h−1, with
)
excellent selectivity (96.9%) towards 1‐butene. Caovilla
and co‐workers[14] reported a series of cobalt β‐diimine
complexes for ethylene oligomerization that achieved
−1
high catalytic activity (2.1 × 106 g (molCo
)
h
−1) and
high selectivity (84%) towards 1‐butene. Transition metals
supported by diphosphine ligands have been investigated
extensively for use in selective ethylene oligomeriza-
tions.[15–17] These catalytic systems can exhibit good
catalytic performance, due to their singly and weakly
coordinating moieties that can readily dissociate to create
Appl Organometal Chem. 2018;e4604.
wileyonlinelibrary.com/journal/aoc
© 2018 John Wiley & Sons, Ltd.
1 of 7
https://doi.org/10.1002/aoc.4604

2 of 7
WEI ET AL.
a vacant site for coordination, thereby enhancing the cat-
alytic activity.[18–21] Incorporation of a large electroposi-
tive silyl moiety can also substantially influence the
structure of the complex and control the stereochemistry
of the reaction. Indeed, our group recently reported effi-
cient catalytic systems for ethylene tri‐/tetramerization
based on Cr(III) complexes stabilized by silicon‐bridged
non‐symmetric diphosphine ligands (PNSiP).[22] Herein,
we report the results of a detailed study of cobalt
catalysts, featuring silylamine‐bridged non‐symmetric
diphosphine ligands, for ethylene oligomerization.
6H), 0.41 (d, J = 2.0 Hz, 6H). 31P NMR (162 MHz,
C6D6): 52.30–52.14 (d, N&bond;P), −43.18 to −43.34 (d,
Si&bond;P). 13C NMR (101 MHz, C6D6): 148.32, 142.44,
142.40, 139.06, 138.87, 137.71, 137.69, 137.53, 137.52,
136.13, 136.12, 136.09, 135.96, 135.91, 135.88, 135.85,
129.69, 128.90, 128.83, 128.61, 128.41, 128.16, 126.84,
125.12, 29.37, 29.34, 26.61, 26.60, 24.03, 2.71, 2.63, 2.56.
Anal. Calcd for C38H43NP2Si (%): C, 75.59; H, 7.18; N,
2.32. Found (%): C, 75.55; H, 7.16; N, 2.28.
N‐Cyclopentyl‐N‐((diphenylphosphanyl)
dimethylsilyl)‐1,1‐diphenylphosphanamine (L2): 1H
NMR (400 MHz, C6D6): 7.69 (t, 4H), 7.56 (t, 4H), 7.08
(m, 12H), 3.82–3.68 (m, 1H), 1.86–1.68 (m, 2H), 1.46 (m,
4H), 1.26 (m, 2H), 0.53 (s, 6H). 31P NMR (162 MHz,
C6D6): 51.96 (s, N&bond;P), −53.55 to −53.88 (d,
Si&bond;P). 13C NMR (101 MHz, C6D6): 140.90, 140.88,
140.70, 140.69, 137.30, 137.27, 137.12, 137.09, 135.49,
135.48, 135.31, 135.31, 133.08, 132.88, 128.86, 128.80,
128.74, 128.69, 128.42, 128.18, 63.09, 34.57, 34.55, 32.21,
23.79, 3.20, 3.08, 2.96. Anal. Calcd for C31H35NP2Si (%):
C, 72.77; H, 6.90; N, 2.74. Found (%): C, 72.72; H, 7.11;
N, 2.52.
2 | EXPERIMENTAL
2.1 | Materials and Instrumentation
All manipulations related to air‐ and moisture‐sensitive reac-
tions were performed under a nitrogen atmosphere in a glove
box. Diphenylphosphorus chloride, diphenylphosphine,
dichlorodimethylsilane, n‐butyllithium (2.4 M in n‐hexane)
and CoCl2 were purchased from Aldrich and used without
further purification. Lithium diphenylphosphine was pre-
pared according to a literature procedure.[23] Polymeriza-
tion‐grade ethylene was obtained from Tianjin Summit
Specialty Gases (China). AlEt3 (1.0 M in methylcyclohexane)
and methylaluminoxane (MAO; 1.4 M in toluene) were pur-
chased from Albemarle. Dried MAO (DMAO) was prepared
by removing all of the volatile compounds from MAO under
vacuum at 40°C for 6 h. Methylcyclohexane, toluene, CH2Cl2
and n‐hexane were dried over appropriate drying agents and
degassed under a nitrogen atmosphere prior to use.
N‐((Diphenylphosphanyl)dimethylsilyl)‐N‐isopropyl‐
1,1‐diphenylphosphanamine (L3): 1H NMR (400 MHz,
C6D6): 7.73–7.64 (m, 4H), 7.59–7.49 (m, 4H), 7.15–7.00
(m, 12H), 3.75 (m, J = 13.4, 6.7 Hz, 1H), 1.08 (d,
J = 6.7 Hz, 6H), 0.57–0.39 (m, 6H). 31P NMR (162 MHz,
C6D6): 50.42 (s, N&bond;P), −52.86 to −53.20 (d,
Si&bond;P). 13C NMR (101 MHz, C6D6): 140.71, 140.70,
140.52, 140.50, 137.28, 137.25, 137.10, 137.07, 135.55,
135.55, 135.38, 135.37, 133.22, 133.02, 128.84, 128.78,
128.70, 128.64, 128.42, 128.18, 52.27, 25.82, 25.79, 25.76,
3.49, 3.38, 3.26. Anal. Calcd for C29H33NP2Si (%): C, 71.73;
H, 6.85; N, 2.88. Found (%): C, 71.72; H, 6.83; N, 2.75.
N‐((Diphenylphosphanyl)(methyl)(phenyl)silyl)‐N‐
isopropyl‐1,1‐diphenylphosphanamine (L4): 1H NMR
(400 MHz, C6D6): 7.95–7.72 (m, 4H), 7.46–7.36 (m, 2H),
7.32–7.25 (m, 2H), 7.18 (d, J = 16.4 Hz, 4H), 7.14–7.02
(m, 10H), 6.84 (s, 3H), 3.77 (td, J = 13.4, 6.4 Hz, 1H),
0.92 (d, J = 6.7 Hz, 3H), 0.84 (s, 3H), 0.82 (s, 3H). 31P
NMR (162 MHz, C6D6): 52.08 (s, N&bond;P), −56.78 to
−57.52 (d, Si&bond;P). 13C NMR (101 MHz, C6D6):
140.81, 140.64, 140.62, 140.61, 140.44, 140.41, 140.08,
140.04, 139.89, 139.85, 137.82, 137.64, 135.93, 135.88,
134.41, 134.26, 133.73, 133.52, 133.28, 133.08, 131.95,
131.77, 131.75, 130.08, 129.05, 128.99, 128.91, 128.81,
128.74, 52.48, 25.90, 25.78, 0.33. Anal. Calcd for
C34H35NP2Si (%): C, 74.56; H, 6.44; N, 2.56. Found (%):
C, 74.52; H, 6.40; N, 2.51.
1H NMR (400 MHz), 13C NMR (101 MHz) and 31P
NMR (162 MHz) spectra were recorded using a Fourier
400M NMR spectrometer (Bruker AscendIII) at 25°C.
The ligands srtuctures and their NMR spactra are given
in the Supporting Information. Samples were prepared
in C6D6 containing tetramethylsilane as the internal ref-
erence. Data are reported as follows: chemical shift in
ppm (δ), multiplicity (s = singlet; d = doublet; m = multi-
plet), coupling constant (Hz), integration. Elemental
analyses for C, H and N were performed using an Ele-
mental Vario EL analyzer. GC was performed using an
Agilent Technologies 7890A GC system.
2.2 | Ligand Synthesis
Ligands were synthesized according to a procedure
reported previously.[22]
N‐(2,6‐Diisopropylphenyl)‐N‐((diphenylphosphanyl)
dimethylsilyl)‐1,1‐diphenylphosphanamine (L1): 1H
NMR (400 MHz, C6D6): 7.98 (t, J = 7.5 Hz, 4H), 7.58
(dd, J = 10.4, 4.9 Hz, 4H), 7.14–6.96 (m, 15H), 3.58–3.46
(m, 2H), 1.11 (d, J = 6.7 Hz, 6H), 0.43 (d, J = 6.7 Hz,
2.3 | Preparation of Complexes
The cobalt complexes were prepared as shown in Fig. 1.

WEI ET AL.
3 of 7
[CoCl2] (0.0260 g, 0.200 mmol) as materials. Yield:
0.109 g (80%). Anal. Calcd for C34H35Cl2NCoP2Si (%): C,
60.27; H, 5.21; N, 2.07. Found (%): C, 60.23; H, 5.27; N,
2.01.
2.4 | X‐ray Crystallography
The X‐ray structural data for complex C1 were
obtained using a Bruker SMART APEX CCD diffrac-
tometer and graphite‐monochromated Mo Kα radiation
(λ = 0.71073 Å). The structure was resolved through
direct methods and refined through the full‐matrix
least‐squares method on F 2 using the SHELXTL pro-
gram. Hydrogen atoms were considered in calculated
positions.
FIGURE 1 Synthesis and structure of cobalt complexes.
2.3.1 | [CoCl2{N‐(2,6‐diisopropylphenyl)‐N‐
[(diphenylphosphino)dimethylsilyl]‐1,1‐
diphenylphosphine}] (C1)
A solution of L1 (0.127 g, 0.210 mmol) in toluene (20 ml)
was added to a suspension of [CoCl2] (0.0260 g,
0.200 mmol) in toluene (30 ml). The mixture was stirred
for 18 h at room temperature until it formed a stable green
color. The solvent was evaporated under vacuum; the solid
residue was washed three times with n‐hexane and dried
under vacuum to afford a green powder. Yield: 0.127 g
(87%). Anal. Calcd for C38H43Cl2NCoP2Si (%): C, 62.21;
H, 5.91; N, 1.91. Found (%): C, 62.17; H, 5.87; N, 1.75.
2.5 | General Oligomerization Procedure
A 140 ml transparent glass reactor was heated in an oven
for 2 h at 105°C prior to use. A magneton was placed
inside the reactor, which was then heated in an oil bath
to the temperature required for the oligomerization reac-
tion. The reactor was equipped with a thermocouple, a
pressure meter and needle valves for injection. The reac-
tor was purged under vacuum and then charged with
high‐purity nitrogen; this process was repeated three
times. The operation described above was then repeated
using ethylene. Next, solutions of the catalyst precursor
and the DMAO/AlEt3 co‐catalyst in methylcyclohexane
were injected into the reactor and then the reactor was
charged with 1.0 MPa of ethylene. The reaction was run
for the required time (generally 30 min) and then the
reactor was depressurized and the reaction quenched
using acidic ethanol. A weighed amount of n‐heptane
was added as an internal standard and the liquid products
were analyzed using an Agilent 7890A GC equipped with
a flame ionization detector and an HP‐5 capillary column
(length, 60 m; diameter, 0.32 mm; film thickness,
0.25 μm). Analysis conditions: working at 35°C for
10 min; heating at 10°C min−1 to 280°C and then main-
taining that temperature for 10 min; injector temperature,
250°C; detector temperature, 300°C.
2.3.2 | [CoCl2{N‐(cyclopentyl)‐N‐
[(diphenylphosphino)dimethylsilyl]‐1,1‐
diphenylphosphine}] (C2)
This complex was prepared using the same synthetic
approach as for C1 with L2 (0.107 g, 0.210 mmol) and
[CoCl2] (0.0260 g, 0.200 mmol) as materials. Yield:
0.106 g (83%). Anal. Calcd for C31H35Cl2NCoP2Si (%): C,
58.04; H, 5.50; N, 2.18. Found (%): C, 58.10; H, 5.57; N,
2.15.
2.3.3 | [CoCl2{N‐(isopropyl)‐N‐
[(diphenylphosphino)dimethylsilyl]‐1,1‐
diphenylphosphine}] (C3)
This complex was prepared using the same synthetic
approach as for C1 with L3 (0.102 g, 0.210 mmol) and
[CoCl2] (0.0260 g, 0.200 mmol) as materials. Yield:
0.105 g (85%). Anal. Calcd for C29H33Cl2NCoP2Si (%): C,
56.60; H, 5.40; N, 2.28. Found (%): C, 56.67; H, 5.47; N,
2.22.
3 | RESULTS AND DISCUSSION
3.1 | Molecular Structure of Complex C1
The molecular structure of complex C1 was determined
using X‐ray diffraction (Fig. 2). A single crystal of C1
was obtained after slowly diffusing n‐hexane into a
CH2Cl2 solution of C1. The crystallographic data and
structural refinement parameters (Table 1) confirmed
the almost ideal tetrahedral geometry of this cobalt
2.3.4 | [CoCl2{N‐((diphenylphosphanyl)
(methyl)(phenyl)silyl)‐N‐isopropyl‐1,1‐
diphenylphosphanamine}] (C4)
This complex was prepared using the same synthetic
approach as for C1 with L4 (0.115 g, 0.210 mmol) and

4 of 7
WEI ET AL.
silyl‐substituent from methyl to phenyl (C4) leads to a
considerable change in the catalytic properties. Although
the catalytic activity was enhanced for C4, the selectivity
towards 1‐butene decreased and a wide distribution was
observed. The formation of more cis‐/trans‐isomers when
using catalysts C2–C4 was attributed to the change in the
electronic effect in the ligand framework.
Table 3 presents the data summarizing the effect of the
loading of complex C1 on the catalytic activity and prod-
uct selectivity. Maintaining the Al/Co molar ratio con-
stant and decreasing the loading of complex C1 from 4.8
to 1.2 μmol caused the catalytic activity and 1‐butene
selectivity to both increase significantly. At a catalyst load-
ing of 1.2 μmol, complex C1 exhibited good activity of
−1
1.6 × 105 g (molCo
)
h
−1, with 79.5% selectivity towards
FIGURE 2 Molecular structure of C1 (showing 50% thermal
ellipsoids). Hydrogen atoms and one CH2Cl2 solvent molecule
have been omitted for clarity. Selected bond lengths (Å) and angles
(°): P(1)–N(1), 1.700(3); P(1)–Co(1), 2.437(10); P(2)–Co(1), 2.385(11);
Co(1)–Cl(1), 2.210(11); Co(1)–Cl(2), 2.215(11); P(2)–Si(1), 2.276(13);
Si(1)–N(1), 1.766(3); P(1)–Co(1)–P(2), 94.04(4); P(1)–Co(1)–Cl(1),
116.08(4); P(2)–Co(1)–Cl(2), 109.68(4); P(2)–Si(1)–N(1), 104.2(10);
Co(1)–P(1)–N(1), 106.44(9); Co(1)–P(2)–Si(1), 95.48(5); Si(1)–N(1)–
P(1), 121.0(15).
1‐butene. Higher loadings of the complex C1 provided
poorer catalytic activity and product selectivity. These
results suggest that the precatalyst (complex C1) was only
partially activated by the co‐catalyst when the former was
present at relatively high concentrations. Furthermore, a
TABLE 1 Crystallographic parameters for C1
Identification code
Empirical formula
Formula weight
Temperature
Shelxl
complex. Selected bond lengths and bond angles of the
complex are listed in the caption to Fig. 2. Single‐crystal
X‐ray analysis revealed that the asymmetric unit of com-
plex C1 consisted of one ligand and one unit of cobalt
dichloride (Fig. 2); that is, the two phosphorus atoms of
the ligand coordinated with the cobalt center to form a
bidentate mononuclear complex.
C39H45Cl4CoNP2Si
818.52
113(2) K
Wavelength
0.71073 Å
Monoclinic
P2(1)/c
Crystal system
Space group
Unit cell dimensions
a = 9.2248(18) Å; α = 90°
b = 18.561(4) Å; β = 97.12(3)°
c = 23.096(5) Å; γ = 90°
3924.1(13) Å3
4, 1.385 Mg m−3
0.851 mm−1
1700
3.2 | Studies of Ethylene Oligomerization
The target ligands (L1–L4) were coordinated with CoCl2
to produce the precatalysts C1–C4 as green powders
(Fig. 1). Our goal for this study was to obtain preliminary
insight into the effects of the catalyst structure, catalyst
loading, reaction temperature, Al/Co molar ratio, co‐cata-
lyst (DMAO/AlEt3) molar ratio and reaction time on the
process of ethylene oligomerization.
In order to investigate the effect of N‐ and Si‐substitu-
ents on the catalytic properties, the cobalt complexes C1–
C4 were subjected used for selective ethylene oligomeri-
zation experiments. Using DMAO/AlEt3 as co‐catalyst at
45°C and 1 MPa ethylene pressure, the catalytic proper-
ties of precatalysts C1–C4 changed with changing N‐
and Si‐substituents (Table 2). Catalyst C1 bearing more
bulky N‐substituent shows the highest selectivity towards
1‐butene. Changing the N‐substituent from more bulky
2,6‐diisopropylphenyl to less bulky cyclopentyl and iso-
propyl moieties leads to a decline of 1‐butene selectivity
from 79.5 to 32.4 and 27.4%, respectively. Changing the
Volume
Z, calculated density
Absorption coefficient
F (000)
θ range for data collection
Limiting indices
1.78–25.02°
−10 ≤ h ≤ 10, −22 ≤ k ≤ 22,
−27 ≤ l ≤ 25
Reflections collected/unique
Completeness to θ = 25.02
Refinement method
31 235/6914 [R(int) = 0.0635]
99.9%
Full‐matrix least‐squares on F 2
6914/0/439
Data/restraints/parameters
Goodness‐of‐fit on F 2
1.076
Final R indices [I > 2σ(I)]
R indices (all data)
R1 = 0.0533; wR2 = 0.1242
R1 = 0.0665; wR2 = 0.1326
1.094 and −0.602 e Å−3
Largest diff. peak and hole

WEI ET AL.
5 of 7
higher precatalyst loading reduced the ethylene concen-
tration – another possible reason for the lower activity.
Moreover, a large amount of the cobalt precatalyst led to
an uncontrollable temperature, due to its exothermicity,
thereby accelerating the isomerization of 1‐butene and
decreasing the selectivity towards 1‐butene.
To explore the effect of the Al/Co molar ratio on the
catalytic properties with C1, molar ratios of 100, 300,
500 and 700 were screened. As evident from Table 4
insufficient amount to form the catalytically active spe-
cies, thereby resulting in low catalytic activity. On the
other hand, an excessive amount of DMAO/AlEt3 might
have a negative influence on the formation of the
active species, by interfering in its formation or
causing its over‐reduction. Therefore, an optimal dosage
of co‐catalyst was needed to activate the catalyst effec-
tively. Furthermore, the selectivity towards 1‐butene
decreased as the Al/Co ratio increased.
(entries 1–4), the catalytic activity increased to
Methylcyclohexane was selected as a solvent for
the catalytic runs, because toluene might poison the
catalysts.[24] Although AlMe3 is often used as a co‐catalyst
for MAO,[25,26] here we introduced AlEt3 instead. The
DMAO/AlEt3 molar ratio has a marked effect on the cat-
alytic activity with C1, which decreased initially but then
increased upon decreasing the molar ratio of DMAO/
−1
1.6 × 105 g (molCo
)
h−1 upon increasing the Al/Co
molar ratio to 500, but decreased thereafter. In general,
an excessive amount of co‐catalyst is needed to effectively
activate the precatalyst. When a lower dosage of DMAO/
AlEt3 is used, a portion might be consumed by traces of
oxygen and water in the reaction system, leaving an
TABLE 2 Effect of C1–C4 on catalytic properties of ethylene oligomerization
Product selectivity (%)
Activity
(× 105 g (molCo
−1
=
Entry
Catalyst
)
h−1
)
C4
1‐C4
trans‐C4
cis‐C4
1
2
3
4
C1
C2
C3
C4
1.6
0.9
1.1
2.3
100.0
100.0
100.0
100.0
79.5
32.4
27.4
20.2
5.6
14.9
47.6
52.1
56.8
20.0
20.5
23.0
Reaction conditions: n(Cat), 1.2 μmol; n(Al)/n(Co), 500; solvent, 20 ml (methylcyclohexane); P, 1.0 MPa; time, 30 min; T, 45°C; n(DMAO)/n(AlEt3), 4:1.
TABLE 3 Effect of C1 loading on catalytic properties of ethylene oligomerization
Product selectivity (%)
Catalyst
(μmol)
Activity
(× 105 g (molCo
−1
=
Entry
)
h−1
)
C4
1‐C4
trans‐C4
cis‐C4
1
2
3
4.8
2.4
1.2
0.5
0.9
1.6
100.0
100.0
100.0
75.3
77.2
79.5
7.7
6.6
5.6
17.0
16.2
14.9
Reaction conditions: n(Al)/n(Co), 500; solvent, 20 ml (methylcyclohexane); P, 1.0 MPa; time, 30 min; T, 45°C; n(DMAO)/n(AlEt3), 4:1.
TABLE 4 Effect of Al/Co molar ratio and DMAO/AlEt3 molar ratio on ethylene oligomerization
Product selectivity (%)
Activity
−1
=
Entry
n(Al)/n(Co)
DMAO/AlEt3
(× 105 g (molCo
)
h−1
)
C4
1‐C4
trans‐C4
cis‐C4
1
2
3
4
5
6
7
100
300
500
700
500
500
500
4:1
4:1
4:1
4:1
3:2
2:3
1:4
LAa
0.3
1.6
1.0
1.7
0.9
0.4
—
100.0
100.0
100.0
100.0
100.0
100.0
80.6
79.5
67.2
80.1
80.3
87.6
6.5
5.6
10
12.9
14.9
22.8
14.4
14.0
10.7
5.5
5.7
1.7
Reaction conditions: n(Cat), 1.2 μmol; P, 1.0 MPa; solvent, 20 ml (methylcyclohexane); T, 45°C; time, 30 min; n(DMAO)/n(AlEt3) used as co‐catalyst.
aLow activity.

6 of 7
WEI ET AL.
TABLE 5 Effect of reaction temperature and reaction time on ethylene oligomerization
Product selectivity (%)
T
(°C)
Time
(min)
Activity
(× 105 g (molCo
−1
=
Entry
)
h−1
)
C4
1‐C4
trans‐C4
cis‐C4
1
2
3
4
5
6
30
45
60
75
45
45
30
30
30
30
20
10
0.1
1.6
1.3
1.1
2.1
2.3
100.0
100.0
100.0
100.0
100.0
100.0
100.0
79.5
76.1
75.3
77.3
78.6
—
—
5.6
7.2
7.7
6.4
5.2
14.9
16.7
17.0
16.3
16.2
Reaction conditions: n(Cat), 1.2 μmol; n(Al)/n(Co), 500; P, 1.0 MPa; solvent, 20 ml (methylcyclohexane); n(DMAO)/n(AlEt3), 4:1.
AlEt3 (Table 4, entries 3, 5–7). When the molar ratio of
4 | CONCLUSIONS
DMAO/AlEt3 was 3:2, the catalytic activity reached its
highest value of 1.7 × 105 g (molCo −1 h−1 (Table 4, entry
)
Cobalt complexes C1–C4, supported by a silylamine‐
bridged non‐symmetric diphosphine ligand, have
been synthesized, characterized and investigated for
ethylene dimerization with the use of DMAO/AlEt3 as
a co‐catalyst. The steric and electronic effects of the
ligand structure have a significant effect on the selectiv-
ity of 1‐butene. The catalytic activity and product selec-
tivity increased upon decreasing the amount of catalyst,
suggesting that the presence of a large amount of catalyst
prevented the precatalyst from complete activation into
active species. Upon increasing the temperature, the
solubility of ethylene decreased in methylcyclohexane
and the rate of deactivation of the active species
increased, resulting in lower catalytic activity. Further-
more, the reaction time also greatly influenced the
catalytic activity. In all runs, C1 exhibited relatively high
catalytic activity for ethylene dimerization, providing
100% selectivity toward C4 and good selectivity towards
1‐butene. Under the optimized conditions, the catalytic
5). We suspect that too much AlEt3 deactivated the
catalytic system and resulted in low catalytic activity.
Furthermore, too little AlEt3 might not have activated
the precatalyst adequately when combined with DMAO,
thereby providing only limited alkylation capacity.
Decreasing the DMAO/AlEt3 molar ratio resulted in an
increased selectivity towards 1‐butene.
The catalytic activity of the C1/methylcyclohexane/
DMAO/AlEt3 system for ethylene oligomerization was
strongly affected by the reaction temperature (Table 5,
entries
1–4).
The
maximum
activity
of
−1
1.6 × 105 g (molCo
)
h−1 was obtained at 45°C. On the
one hand, the lower activity at higher temperatures might
be attributable to a lower concentration of ethylene in
methylcyclohexane; on the other hand, an elevated
temperature might have resulted in deactivation of the
active species, also decreasing the catalytic activity. The
selectivity towards 1‐butene reached 100% at relatively
low temperature (Table 5, entry 1), suggesting that
1‐butene might be the initial product of ethylene
oligomerization. Increasing the temperature might have
converted 1‐butene to cis‐/trans‐but‐2‐ene through
isomerization, thereby decreasing the selectivity towards
1‐butene (Table 5, entries 2–4).
activity
for
ethylene
)
dimerization
reached
−1
2.3 × 105 g (molCo
h−1 with 78.6% selectivity towards
1‐butene.
ACKNOWLEDGMENTS
The data in Table 5 (entries 2, 5–6) reveal that the
reaction time had an effect on the activity of the catalyst
with C1, but the selectivity towards 1‐butene remained
relatively stable. The catalytic activity was enhanced as
the reaction time decreased to from 30 to 10 min. This
result suggests that the catalyst might have become
This study was supported by the National Key Research
and Development Program of China (2017YFB0306700)
and the Natural Science Foundation of Tianjin City
(14JCYBJC23100, 15JCYBJC48100, 16JCZDJC31600).
ORCID
deactivated when treated for a longer period of time.
−1
The maximum activity of 2.3 × 105 g (molCo
)
h−1 was
Tao Jiang http://orcid.org/0000-0002-9854-7228
obtained at 45°C when the reaction time was 10 min
(Table 5, entry 6). Thus, it appears that the active species
were inactivated gradually upon increasing the reaction
time, with no other active centers formed during the
course of the reaction.
REFERENCES
[1] F. M. A. Geilen, G. Stochniol, Ullmann's Encyclopedia of
Industrial Chemistry, 4th ed., Weinheim 2014.

WEI ET AL.
7 of 7
[2] A. Finiels, F. Fajula, V. Hulea, Catal. Sci. Technol. 2014, 4,
Otto, M. Overett, A. M. Z. Slawin, P. Wasserscheid, S.
2412.
Kuhlmann, J. Am. Chem. Soc. 2004, 126, 14712.
[3] D. V. Gutsulyak, A. L. Gott, W. E. Piers, M. Parvez, Organome-
tallics 2013, 32, 3363.
[19] S. Harzschel, F. E. Kuhn, A. Wohl, W. Muller, M. H. Al‐Hazmi,
A. M. Alqahtani, B. H. Muller, N. Peulecke, U. Rosenthal,
Catal. Sci. Technol. 2015, 5, 1678.
[4] X. Meng, L. Zhang, Y. Chen, T. Jiang, React. Kinet. Mech. Catal.
2016, 119, 481.
[20] Y. Shaikh, K. Albahily, M. Sutcliffe, V. Fomitcheva, S.
Gambarotta, I. Korobkov, R. Duchateau, Angew. Chem. Int.
Ed. 2012, 51, 1366.
[5] T. Wang, B. Dong, Y.‐H. Chen, G.‐L. Mao, T. Jiang,
J. Organometal. Chem. 2015, 798, 388.
[21] L. Zhang, X. Meng, Y. Chen, C. Cao, T. Jiang, ChemCatChem
2017, 9, 76.
[6] V. Lecocq, H. Olivier‐Bourbigou, Oil Gas Sci. Technol. 2007, 62,
761.
[22] L. Zhang, W. Wei, F. Alam, Y. Chen, T. Jiang, Catal. Sci.
Technol. 2017, 7, 5011.
[7] J. Zhang, H. Gao, Z. Ke, F. Bao, F. Zhu, Q. Wu, J. Organometal.
Chem. 2005, 231, 27.
[23] K. Blann, A. Bollmann, H. de Bod, J. T. Dixon, E. Killian, P.
Nongodlwana, M. C. Maumela, H. Maumela, A. E. McConnell,
D. H. Morgan, J. Catal. 2007, 249, 244.
[8] Y. Chen, P. Hao, W. Zuo, K. Gao, W.‐H. Sun, J. Organometal.
Chem. 2008, 693, 1829.
[9] R. Gao, K. Wang, Y. Li, F. Wang, W.‐H. Sun, C. Redshaw, M.
Bochmann, J. Mol. Catal. A 2009, 309, 166.
[24] A. Jabri, C. B. Mason, Y. Sim, S. Gambarotta, T. J. Burchell, R.
Duchateau, Angew. Chem. Int. Ed. 2008, 47, 9717.
[10] M. Qian, M. Wang, B. Zhou, R. He, Appl. Catal. Gen. 2001, 209,
[25] S. Hansjörg, Macromol. Symp. 1995, 97, 27.
11.
[26] M. Watanabi, C. N. McMahon, C. J. Harlan, A. R. Barron,
[11] M. Zhang, R. Gao, X. Hao, W.‐H. Sun, J. Organometal. Chem.
2008, 693, 3867.
Organometallics 2001, 20, 460.
[12] Z. Wang, G. A. Solan, W. Zhang, W.‐H. Sun, Coord. Chem. Rev.
2018, 363, 92.
SUPPORTING INFORMATION
[13] D. Thiele, R. F. de Souza, J. Organometal. Chem. 2011, 340, 83.
Additional supporting information may be found online
in the Supporting Information section at the end of the
article.
[14] M. Caovilla, D. Thiele, R. F. de Souza, J. R. Gregório, K.
Bernardo‐Gusmão, Catal. Commun. 2017, 101, 85.
[15] D. Bézier, O. Daugulis, M. Brookhart, Organometallics 2017, 36,
443.
[16] Y. Huang, W. Wei, X. Meng, L. Zhang, Y. Chen, T. Jiang, Appl.
Organometal. Chem. 2018, 32, e4014.
How to cite this article: Wei W, Yu B, Alam F,
Huang Y, Cheng S, Jiang T. Cobalt(II)‐based
ethylene dimerization catalysts with silicon‐bridged
diphosphine ligands. Appl Organometal Chem.
2018;e4604. https://doi.org/10.1002/aoc.4604
[17] Y. Yang, J. Gurnham, B. Liu, R. Duchateau, S. Gambarotta, I.
Korobkov, Organometallics 2014, 33, 5749.
[18] A. Bollmann, K. Blann, J. T. Dixon, F. M. Hess, E. Killian, H.
Maumela, D. S. McGuinness, D. H. Morgan, A. Neveling, S.