Ethylene oligomerization using nickel-β-diimine hybrid xerogels produced by the sol-gel process

Article abstract of DOI:10.1016/j.apcata.2012.09.024

Nickel-β-diimine complexes have been used as homogeneous- and heterogeneous-phase oligomerization catalysts. The following complexes have also been heterogenized on silica using the sol-gel method: 1,5-bis(phenyl)- pentanediimine-dibromonickel(II) (A), 1,

Full text of DOI:10.1016/j.apcata.2012.09.024

Accepted Manuscript
Title: Ethylene oligomerization using nickel-␤-diimine hybrid
xerogels produced by the sol-gel process
´
Authors: Enederson Rossetto, Marcela Caovilla, Daniel
˜
Thiele, Roberto F. de Souza, Katia Bernardo-Gusmao
PII:
DOI:
Reference:
S0926-860X(12)00617-5
doi:10.1016/j.apcata.2012.09.024
APCATA 13880
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
8-8-2012
18-9-2012
21-9-2012
Please cite this article as: E. Rossetto, M. Caovilla, D. Thiele, R.F. de Souza,
˜
K. Bernardo-Gusmao, Ethylene oligomerization using nickel-rmbeta-diimine hybrid
xerogels produced by the sol-gel process, Applied Catalysis A, General (2010),
doi:10.1016/j.apcata.2012.09.024
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

*Graphical Abstract (for review)
Page 1 of 35

*Highlights (for review)
Research Highlights





Three β-diimine ligands and their corresponding nickel-complexes were synthesized;
Hybrid silica containing β-diimine groups were prepared by sol-gel process;
Hybrid silica were complexed with nickel to form heterogenized nickel-complex;
Ethylene oligomerisation using nickel--diimines were described;
Ligands nature were correlated with the catalytic activity and selectivity.
Page 2 of 35

Ethylene oligomerization using nickel--diimine
hybrid xerogels produced by the sol-gel process
Enéderson Rossetto, Marcela Caovilla, Daniel Thiele, Roberto F. de
Souza, Katia Bernardo-Gusmão*
Institute of Chemistry, Universidade Federal do Rio Grande do Sul, Av. Bento Gonçalves
9500, P.O. Box 15003, CEP 91501-970 Porto Alegre, Brazil.
Abstract
Nickel-β-diimine complexes have been used as homogeneous- and heterogeneous-
phase oligomerization catalysts. The following complexes have also been heterogenized on
silica using the sol-gel method: 1,5-bis(phenyl)-pentanediimine-dibromonickel(II) (A), 1,5-
bis(2,6-dimethylphenyl)-2,4-pentanediimine-dibromonickel(II) (B) and 1,5-bis(2,4,6-
trimethylphenyl)-2,4-pentanediimine-dibromonickel(II) (C). The synthesis of the xerogels
was
performed
employing
3-chloropropyltrimethoxysilane
(CPTMS),
tetraethylorthosilicate (TEOS) and HF. The hybrid β-diimine/silica materials are used in
combination with ethylaluminum sesquichloride (EASC) and have been shown to be
outstanding ethylene oligomerization catalyst systems. Under mild reactions conditions,
the homogeneous and heterogenized catalysts had high turnover frequencies (between
1800 and 210000 h-1). The best selectivities were obtained with the heterogenized
catalysts, which give 100 % of C4 fraction with 1-butene selectivities of up to 91 %.
Keywords: Oligomerization, Ethylene, -Diimine, Xerogel, Sol-gel process,
Organic-inorganic hybrid materials.
*Corresponding author.
Tel.: +55-51-33087240; Fax: +55-51-33087304.
E-mail address: katiabg@iq.ufrgs.br
Page 3 of 35

Graphical Abstract
Hybrid silica was synthesized by the sol-gel method and subsequently complexed
with metal to form a heterogenized nickel-β-diimine/silica complex, which was employed
in ethylene-oligomerization reactions.
INSERT FIGURE ABSTRACT
2
Page 4 of 35

1. Introduction
The oligomerization of olefins is an attractive method for producing longer chain
olefins. Olefin oligomers are used in several applications, ranging from fuel components
for better performance gasolines to intermediates for the manufacture of petrochemicals,
such as plasticizers and lubricants [1-5]. Most of the olefin-oligomerization processes use
transition metal catalysts, and a large variety of ligands containing O, N and P heteroatoms
have been described. The proper choice of transition metal, organic ligands and co-
catalysts enable some degree of tunability of the selectivity, ensuring production of the
desired olefins.
Much effort has been devoted to studying the effects of the ligand structure of
transition metal complexes on their catalytic properties in oligo/polymerization reactions
[6-9]. β-Diimine and β-diiminate ligands are easy to prepare and have several attractive
properties, including tunable electronic and steric parameters, which affect the catalytic
performance of their transition metal complexes [10,11]. The β-diimine ligands have
become more popular since the discovery by Feldman and co-workers of their use with
nickel and palladium catalysts [12,13]. The first palladium -diimine complex was
obtained from the reaction of β-iminoamine with [Pd(CH3CN)4][BF4]2. β-Diimine ligands
containing bulky aryl substituents on nitrogen atoms have been used in transition metals
complexes with nickel for the successful synthesis of polyolefins [14,15].
Homogeneous catalysts usually contain uniform and well-defined active sites,
which lead to high activities and reproducible selectivities. However, the major drawback
in using these catalysts is the difficulty in separating the catalysts, products and solvent. An
alternative to circumvent this drawback is to immobilize the homogeneous catalyst in
various media, including inorganic and organic supports [16-19].
Sol-gel chemistry can play a central role in the development of hybrid materials
because of the mild synthesis conditions provided by these processes [20]. The sol-gel
process has received much attention in materials research due to its unique advantages
such as low processing temperatures, high homogeneity of the final product and capacity to
generate materials with surface-controlled properties. Materials with these properties are
suitable for use as catalyst supports.
3
Page 5 of 35

The possibility of combining properties of organic and inorganic components in a
unique composite material is a distinct characteristic of these materials because their
properties are related not only to the chemical nature of the inorganic and organic
components but also to their synergy. Therefore, the interface between inorganic and
organic constituents strongly influences their properties. Organic-inorganic hybrid
materials thus represent a creative alternative to design new materials and compounds and
allow for the development of materials for innovative applications [21]. In this context, one
attractive use is the immobilization of transition metal complexes in inorganic networks by
sol-gel processes, producing heterogeneous hybrid-catalyst precursors. The production of
ethylene oligomers using these heterogeneous catalytic systems provides a compelling
answer to the demand for environmentally friendly processes by allowing for the recycling
of the catalyst in contrast to the homogeneous-phase methods.
Despite the considerable interest in the application of hybrid systems as catalysts,
surprisingly few studies have been devoted to describing their behavior. Important
examples are the reports on the use of Fe, Re, Ti and Mo hybrid materials produced via
sol-gel methods in epoxidation reactions [22-26] and Zr hybrid materials in ethylene
polymerization [27-29]. Ethylene polymerization with nickel-β-diimine complexes
immobilized in sol-gel hybrid materials was described in 2008 [30]; however, to the best of
our knowledge, there are no examples of organic-inorganic hybrid materials prepared by
the sol-gel method used for ethylene oligomerization available in the current literature.
We describe herein the synthesis of nickel-β-diimine complexes, their
heterogenization in a hybrid silica-based matrix using the sol-gel process and their catalytic
activities in ethylene oligomerization.
2. Experimental
All experiments were performed under an argon atmosphere using standard
Schlenk-tube techniques. The solvents were deoxygenated by refluxing over appropriate
drying agents (toluene, hexane, ethyl ether and benzene on sodium benzophenone and
dichloromethane and acetonitrile on phosphorous pentoxide) under argon and distilled
immediately prior to use. Methanol, ethanol and tetrahydrofuran (THF) were used without
4
Page 6 of 35

further purification. Aniline, 2,6-dimethylaniline and 2,4,6-trimethylaniline were purchased
from Aldrich and distilled under reduced pressure prior to use. 3-
Chloropropyltrimethoxysilane (CPTMS), tetraethylorthosilicate (TEOS), sodium hydride
(NaH) and anhydrous nickel bromide (NiBr2) were purchased from Aldrich. p-
Toluenesulfonic acid (Vetec) was distilled on benzene by the Dean-Stark technique.
Sodium carbonate was purchased from Vetec and used without further purification.
Ethylaluminum sesquichloride (Al2Et3Cl3, EASC) was supplied by Akzo Nobel and used
with previous dilution with toluene. Solids were dried under reduced pressure.
Elemental analyses were carried out in a Perkin Elmer M CHN Analyzer 2400.
Infrared analyses were performed in an FT-IR Shimadzu Prestige-21 instrument using self-
supported pellets (without the use of KBr) of 40-50 mg of the sample. The results were
obtained from the accumulation of 32 scans. The 1H NMR spectra of the samples dissolved
in CDCl3 were obtained in an Inova 300 MHz spectrometer. UV-Vis absorption of the
samples was measured in a spectral range from 200 nm to 800 nm using a Varian Carry
100 spectrophotometer equipped with a photomultiplier detector and diffuse reflectance
accessory. Thermogravimetric analyses were performed on a TA Instruments TGA Q50.
The samples were heated at 10 °C/min from 20 °C to 800 °C under nitrogen flow. The
morphology and particle size of the products were investigated using a JEOL/EO JSM
6060 scanning electron microscope (SEM) operating at 20 kV. After sample pretreatment
for 3 h at 80 °C, the specific surface areas of the samples were determined by nitrogen
adsorption-desorption using a Micrometrics TriStar II 3020. The crystallinity was
evaluated by X-ray diffraction (XRD) on a Siemens D500 diffractometer with Cu K
radiation (l = 1.54056 Å). The nickel content of the solids was determined by flame atomic
absorption spectrometry (FAAS). The analysis was performed on a Perkin Elmer A atomic
absorption spectrometer with a hollow nickel cathode lamp (λ = 232 nm) and air-acetylene
flame (10 L/min: 2.5 L/min). The samples were prepared by treating 15 mg of the
heterogenized complex with 2 mL of HCl, 6 mL of HNO3 and 5 mL of HF, adding the
mixture to teflon autoclaves and subsequently using a digester for 24 h at 150 ºC. After
cooling, the samples were diluted to 50 mL.
2.1. Ligands
As previously published [31], 1,5-bis(phenyl)-pentanediimine, A, was synthesized
by reacting acetylacetone and aniline in the presence of HCl as the catalyst (Figure 1). A
5
Page 7 of 35

mixture of 10.3 mL of acetylacetone (100 mmol) and 18.0 mL of aniline (200 mmol) was
placed in a 250 mL Schlenk tube in an ice bath, and 8.3 mL of hydrochloric acid was added
slowly under stirring. A yellow-brown solution was formed. After 12 h under stirring, the
reaction mixture was filtered, and the solid was washed with hexane. The solid was
dissolved by the addition of 8 mL of CH2Cl2, 50 mL of H2O and 20 mL of triethylamine.
The aqueous phase was extracted with ether and combined with the organic phase. After
the solvent was evaporated under reduced pressure, the crude product was recrystallized
with methanol, giving 8.07 g (32.2 mmol) of ligand A in a 32 % yield.
Ligands B, 1,5-bis(2,6-dimethylphenyl)-2,4-pentanediimine, and C, 1,5-bis(2,4,6-
trimethylphenyl)-2,4-pentanediimine, were similarly synthesized, but p-toluenesulfonic
acid was used as the catalyst [32]. For ligand B, 4.4 mL (42.8 mmol) of acetylacetone, 11.1
mL of 2,4-dimethylaniline (89.9 mmol) and 42.8 mmol of p-toluenesulfonic acid in 100
mL of benzene were added to a Schlenk tube. For ligand C, 5.7 mL (55.5 mmol) of
acetylacetone, 16.4 mL (116.6 mmol) of 2,4,6-trimethylaniline and 55.5 mmol of p-
toluenesulfonic acid in 100 mL of benzene were added to a Schlenk tube. In both cases, the
suspension was refluxed for 24 h, and the water formed was removed using the Dean-Stark
apparatus. The solvent was removed under reduced pressure. Then, 60 mL of CH2Cl2 and
80 mL of a saturated solution of sodium carbonate were added. The aqueous phase was
extracted with CH2Cl2. After the solvent was evaporated under reduced pressure, the solids
were recrystallized with methanol to give 35.5 mmol of B and 39.5 mmol of C, which
corresponded to 83 % and 92 % yields, respectively.
INSERT FIGURE 1
1
The results from the H NMR, IR, UV and CHN analyses of the ligands are listed
below.
1
(A) H NMR δ (ppm): 12.72 (s, 1H, N-H), 7.30 (t, 4H, J = 7.8 Hz, m-ArH), 7.06 (t, 2H,
J=7.5 Hz, p-ArH), 6.97 (d, 4H, J=7.8 Hz, o-ArH), 4.89 (s, 1H, β-CH), 2.02 (s, 6H, α-CH3).
IR: C=N 1555 cm-1. UV: 305 and 380 nm. CHN Calcd. for C17H18N2: C = 81.56 %, H =
7.25 %, N = 11.19 %; Found: C = 81.11 %, H = 7.29 %, N = 11.05 %.
6
Page 8 of 35

1
(B) H NMR δ (ppm): 12.20 (s, 1H, N-H), 7.03-6.94 (m, 6H, m-ArH+, p-ArH), 4.88 (s,
1H, β-CH), 2.16 (12H, s, o-CH3 Ar), 1.69 (s, 6H, α-CH3). IR: C=N 1545 cm-1. UV: 289 and
353 nm. CHN Calcd. for C21H26N2: C = 82.31 %, H = 8.55 %, N = 9.14 %; Found: C =
81.88 %, H = 8.82 %, N = 9.17 %.
1
(C) H NMR δ (ppm): 12.15 (s, 1H, N-H), 6.85 (s, 4H, m-ArH), 4.85 (s, 1H, β-CH), 2.25
(6H, s, p-CH3 Ar), 2.11 (s, 12H, o-CH3 Ar), 1.68 (s, 6H, α-CH3). IR: C=N 1550 cm-1. UV:
286 and 365 nm. CHN Calcd. for C23H30N2: C = 82.59 %, H = 9.04 %, N = 8.37 %; Found:
C = 82.68 %, H = 9.40 %, N = 8.37 %.
2.2 Nickel-β-diimine complexes
The nickel-β-diimine complexes corresponding to ligands A-C were obtained by
reacting Ni(MeCN)2Br2 with the corresponding ligands as shown in Figure 2.
INSERT FIGURE 2
In a Schlenk tube, a solution of 6 mmol of the ligand dissolved in 20 mL of CH2Cl2
was added to a suspension of 3.06 g (10.2 mmol) of Ni(MeCN)2Br2 in 40 mL of CH2Cl2
and maintained under stirring for 5 days at room temperature. The mixture was filtered,
and the solvent was removed under reduced pressure, resulting in a purple solid with 20-32
% yield.
The Ni-A complex gives IR: 1594 cm-1 (νC=N); CHN Calcd. for C17H18Br2N2Ni: C
= 43.55 %, H = 3.87 %, N = 5.98 %; Found: C = 45.30 %, H = 3.98 %, N = 5.90 %. The
Ni-B complex gives IR: 1620 cm-1 (νC=N); CHN Calcd. for C21H26Br2N2Ni: C = 48.05 %,
H = 4.99 %, N = 5.34 %; Found: C = 48.05 %, H = 5.05 %, N = 5.27 %. The Ni-C
complex gives IR: 1619 cm-1 (νC=N); CHN Calcd. for C23H30Br2N2Ni: C = 49.96 %, H =
5.47 %, N = 5.07 %; Found: C = 50.36 %, H = 5.52 %, N = 5.12 %.
2.3 Hybrid -diimine/silica xerogels and nickel--diiminesil/silica
The xerogel syntheses were carried out according to procedures adapted from the
literature [25, 30] for the sol-gel method.
7
Page 9 of 35

Ligands A, B and C (6 mmol) were activated with sodium hydride (0.21 g, 9 mmol)
in 20 mL of CH2Cl2 for 30 min at 35 C (step 1 in Figure 3). Then, a solution of 3-
chloropropyltrimethoxysilane (CPTMS) (1.1 mL, 6 mmol) in 5 mL of a toluene and THF
mixture was added. The mixture was stirred under argon and maintained under reflux for 3
h. The solution was then cooled and centrifuged at 5000 rpm for 20 min, and the
supernatant was used as the sol–gel chelating organic precursor in the gel formation
process (step 2, Figure 3). Afterwards, tetraethylorthosilicate (TEOS) (50 mL), ethanol (50
mL), water (16 mL) and HF (1 mL) were added under stirring to the precursor solution.
The gelation occurs by the fluoride catalytic process at pH 4. The mixture was stored for a
week for gelation and solvent evaporation. The resulting xerogel was washed using 20 mL
of each of the following solvents: toluene, THF, CH2Cl2, ethanol, water and diethyl-ether.
The xerogel was dried for 4 h in an oven at 60 C (step 3 in Figure 3). The resulting hybrid
silicas were named HA, HB and HC for hybrid silica containing the corresponding
ligands.
To remove residual moisture, the solid phase was dried in vacuum (10-2 mbar) at 80
°C for 24 h. Then, a suspension of 1 mole of Ni(CH3CN)2Br2 [33] per mole of ligand is
dissolved in 20 mL of CH2Cl2, added to 1.0 g of -diiminesil/silica and maintained under
stirring for 5 days at room temperature. The solid phase was filtered and extensively
washed with CH3CN (4 aliquots of 30 mL). The solid phase, Ni(II)-(ligand)--
diiminesil/silica (step 4, Figure 3), was dried at 60 C for 1 h. The heterogenized
complexes were named Ni-HA, Ni-HB and Ni-HC when obtained from the HA, HB and
HC hybrid silicas, respectively.
INSERT FIGURE 3
2.4 Ethylene-oligomerization runs
Ethylene-oligomerization experiments were carried out in a 450 mL Parr stainless
steel autoclave equipped with magnetically driven mechanical stirring, a thermocouple and
a pressure gauge. The reaction temperature (10 ) was controlled with a thermostatic bath.
8
Page 10 of 35

In a typical homogeneous reaction run, 20 µmol of the catalytic precursor and 60
mL of toluene were added to the reactor under argon, followed by the addition of a solution
of EASC. The reactor was pressurized with ethylene, and the temperature was adjusted to
10 °C with a thermostatic bath circulation. The Al/Ni molar ratios varied from 30 to 200,
and the ethylene pressure ranged from 5 atm to 15 atm. After 30 min, the reaction was
stopped, and the mixture was cooled to -30 ºC and analyzed immediately by gas
chromatography.
Heterogeneous-medium runs were performed, and the results were compared to the
results obtained with the homogeneous-phase complex. In a typical heterogeneous-medium
experiment, the reactor was charged with a suspension of the desired catalytic precursor
containing 6.5 µmol or 20 µmol of nickel and 60 mL of toluene saturated with ethylene,
and the temperature was adjusted to 10 °C. The pressure was raised to 15 atm under
continuous stirring. The amount of alkylaluminum co-catalysts introduced into the reactor
was calculated to obtain an aluminum to nickel molar ratio (Al/Ni) between 100 and 1200.
After 30 min, the reaction was stopped, and the mixture was cooled and analyzed
immediately by gas chromatography.
Recycle experiments were carried out in a 100 mL double-walled glass reactor
containing a magnetic stirring bar with a constant supply of neat gaseous ethylene at 5 atm
and a thermocouple to measure the temperature. The reaction temperature was regulated to
10 °C by an external-circulation ethanol bath. In a typical experiment, the reactor was
charged with a solution of the desired catalytic precursor (20 µmol) in 50 mL of toluene
saturated with ethylene. The reactor was purged with ethylene, and the alkylaluminum
solution is added in amounts calculated to obtain an aluminum to nickel molar ratio (Al/Ni)
of 200. After 30 min, the reaction was stopped, and the mixture was cooled and analyzed
by gas chromatography. After the first reaction, the products were removed from the
reactor through a cannula and collected at -20 °C. Catalyst recycling was accomplished by
adding another 50 mL of toluene and continuing the reaction.
In all cases, chromatographic analyses were performed on a Varian 3400CX gas
chromatograph equipped with a Petrocol DH capillary column (methyl silicone, 100 m
long, 0.25 mm ID, 0.5 μm film thickness). Analysis conditions were 36 °C for 15 min,
followed by heating at a rate of 5 °C/min up to 250 °C. The products were identified by the
method of co-injection of standards, and cyclohexane was used as the internal standard for
9
Page 11 of 35

quantification. The TOF values, defined as moles of converted ethylene per mole of pre-
catalyst per reaction time (in h), show a variation of ±12 %, as determined by at least three
independent experiments performed at each condition.
3. Results and discussion
The formation of the Ni-A, Ni-B and Ni-C complexes from the corresponding
parent ligands A, B and C is accompanied by an enhancement of the infrared C=N
stretching frequencies by 36, 41 and 25 cm-1, respectively. Figure 4 gives the IR spectrum
of the Ni-C complex as an example.
INSERT FIGURE 4
The energy associated with the CN vibrations of the imine ligands is expected to
decrease upon simple metal complexation. The observed energy increase indicates a
change in the conjugated structure of the free ligand to a non-conjugated structure of the
ligand when coordinated to the metal, giving the corresponding Ni-β-diimine complexes.
The measured elemental compositions of the homogeneous complexes are
consistent with the calculated values for Ni-B and Ni-C; however, a higher %C is observed
for Ni-A, most likely due to the presence of residual ligands.
The synthesis of the HA, HB and HC hybrid silicas containing the corresponding ligands
in their composition is performed in three steps. The starting point is the activation of the
ligands with NaH to form the sodium salt, which is monitored by the release of the H2
formed (step 1 of Figure 3). The synthesis of the organic precursor by reaction with
CPTMS follows (step 2 of Figure 3), and finally, the polycondensation with TEOS in
presence of HF completes the procedure, forming the hybrid silica (step 3 of Figure 3). The
incorporation of the organic material is monitored by elemental analysis.
The amount of ligand incorporated into the silica and the ligand:Si ratio is
calculated from the nitrogen content obtained by elemental analysis as shown in Table 1.
10
Page 12 of 35

The organic matter is incorporated into the support at 20-40 %, and the ligand:Si molar
ratio is between 1:100 and 1:180. The hybrid material produced with ligand C has low
organic content (HC), most likely due to the steric hindrance of ligand C.
INSERT TABLE 1
Table 1. Characteristics of the hybrid β-diimine/silica xerogels and corresponding nickel
complexes.
Ligand Organic content Ligand:Si Specific areab Heterog.
Nickel contentb
mmol/g
mmol/g
0.128
m2/g
176
complex
1:120
1:100
1:180
0
0.110
HA
HB
HC
Wa
Ni-HA
0.150
0.080
0.000
206
191
Nd
0.187
0.025
0.000
Ni-HB
Ni-HC
“Ni”-W
Reaction conditions: 6 mmol ligand, 225 mmol TEOS, 1 mL HF. a Control: silica obtained without addition
of the ligand.
b Determined by BET.
Figure 5 shows the infrared spectra of the HA, HB and HC hybrid silicas. The
control silica W shows the silica overtones bands at 1984 and 1877 cm-1. The hybrid silica
contains two components: an inorganic one and an organic one. The organic component
can be identified by the deformation bands of the N-H group at 1540 cm-1. The band
related to the stretching of the C=N group is superimposed on the H-O-H deformation
band appearing at 1625 cm-1, which originates from the adsorbed water in the silica.
INSERT FIGURE 5
The electronic spectra of the free ligands show two bands at 305 nm and 380 nm for
ligand A, 289 nm and 353 nm for ligand B and 286 nm and 365 nm for ligand C. The
spectra of the hybrid silicas exhibit only one band at 320 nm for the HA hybrid silica, 318
11
Page 13 of 35

nm for the HB hybrid silica and 320 nm for the HC hybrid silica. These bands are shifted
from 305 nm to 320 nm in the HA hybrid material, from 286 nm to 318 nm in HB and
from 289 nm to 320 nm in HC when the corresponding free ligands are compared to the
hybrid silicas. This behavior can be attributed to the fact that in the hybrid material, the
ligand is trapped, and the electronic transitions are less energetic than when the ligand is in
the free state. The bands at 380 nm for the HA hybrid material, at 353 nm for HB and 365
nm for HC disappear when the ligand isn´t free, which can be attributed to the lower
mobility of the intermolecular hydrogen bond when supported in the hybrid material.
The specific areas of the hybrid silicas were calculated by the BET method, giving
176 m2/g for the HA hybrid silica, 206 m2/g for the HB hybrid silica and 191 m2/g for the
HC hybrid silica. These results are similar to those reported by Brinker and Scherer [34]
for silica gel. The isotherms are type IV [35], characteristic of mesoporous materials, and
show type H3 hysteresis, which indicates the presence of plate-like particles that give rise
to slit-shaped pores.
INSERT FIGURE 6
The X-ray diffractograms of the xerogels showed only a broad peak at 2θ between
16° and 26°, suggesting that the hybrid materials are basically amorphous.
The TGA analysis is more instructive and shows a decomposition temperature of
260 ºC for the free ligands, which increases to 400 ºC when the ligands are anchored in the
inorganic support. This increase in decomposition temperature demonstrates the significant
increase in the stability of these ligands upon immobilization on the inorganic support.
The nickel complexes on hybrid silica corresponding to Ni-A, Ni-B and Ni-C were
prepared by reacting the corresponding hybrid silica with a solution of nickel bromide. The
amount of heterogenized nickel was determined by FAAS and is shown in the last column
of Table 1. In the case of ligand A, the amount of nickel incorporated into the Ni-HA
catalyst (0.11 mmol/g) is very close to the amount of ligand present in the parent HA silica
(0.12 mmol/g). In the Ni-HB catalyst, the amount of nickel incorporated (0.187 mmol/g) is
greater than the amount of ligand present in the HB silica (0.15 mmol/g), indicating the
presence of Ni species deposited on the silica surface in addition to those complexed with
12
Page 14 of 35

the heterogenized ligand. In the case of the Ni-HC catalyst, the amount of incorporated
nickel (0.025 mmol/g) was lower than the amount of ligand present in the HC silica, most
likely due to the steric hindrance of the ligand, which decreases the metal’s access to the
binding site and thus limits its incorporation into the catalyst.
Oligomerization runs
The Ni-A, Ni-B and Ni-C nickel complexes have been used as catalysts in
combination with ethylaluminum sesquichloride, EASC, for the oligomerization of
ethylene. The Ni-A, Ni-B and Ni-C complexes showed activity for ethylene
oligomerization, and their performance is affected by the catalyst structure and the reaction
conditions. The effects of the nickel precursor structure, amount of alkylaluminum
compound (Al/Ni ratio) and ethylene pressure on the activity and selectivity for ethylene
oligomerization are shown in Table 2.
Table 2. Ethylene oligomerization results for nickel-β-diimine catalysts in the
homogeneous phase.
Entry
Complex Pressure Al/Ni TOF
SC4
SC6
SC8+
(%)
-C4
(atm)
(103 h-1) (%)
(%)
(%)
1
2
3
4
5
6
7
8
15
15
15
15
15
15
5
100
100
100
30
210
18.3
18.7
4.3
78
97
96
94
100
97
99
97
32
87
87
93
95
91
93
90
20
3
4
6
-
2
-
-
-
-
-
-
-
Ni-A
Ni-B
Ni-C
Ni-C
Ni-C
Ni-C
Ni-C
Ni-C
50
6.1
200
100
100
16.5
5.4
3
1
3
10
9.7
Complex: 20 µmol, T = 10 °C, reaction time = 0.5 h, solvent = toluene (60 mL), co-catalyst = EASC.
The Ni-A complex, which contains a phenyl ring without substituents, shows the
highest activity (210 x 103 h-1) with good selectivity towards C4 products (78 %). The Ni-B
and Ni-C complexes show lower activities than Ni-A but greater selectivity towards C4
13
Page 15 of 35

products (96-97 %). These results suggest that the steric hindrance of the methyl groups is
responsible for the decrease in ethylene accessibility and consequently for the observed
decrease in the reaction activity.
The Ni-C complex was chosen for optimization of the reaction parameters for the
oligomerization tests.
Initially, the influence of the Al/Ni ratio on the activity and the selectivity of the
catalyst was evaluated. The activity of the catalyst increases continuously with the Al/Ni
ratio up to a value of 100 (entries 3, 4 and 5 in Table 2), a condition in which the maximum
activation of the Ni-C complex is attained. A further increase in the amount of co-catalyst
up to Al/Ni = 200 (entry 6) decreases the activity of the catalyst, most likely due to catalyst
decomposition and the corresponding reduction in the number of active species.
The activity increases linearly with the ethylene pressure (as shown by entries 5, 7
and 8 in Table 2), indicating a first-order dependence in accordance with a Cossee-
Arlmann chain-growth mechanism [36].
These observations suggest that the activity trends observed in Table 2 for the Ni-A,
Ni-B and Ni-C catalysts are related to electronic effects instead of steric effects. The larger
electronic effect of the ligand compared to its steric hindrance is also consistent with the
results for α-keto-β-diimine nickel catalysts for ethylene polymerization [37, 38].
The selectivities of the Ni-B and Ni-C catalysts for C4 products and 1-butene are
very similar. The Ni-A catalyst exhibits much low selectivities compared to Ni-B and Ni-
C. The decrease in the C4 selectivity of the Ni-A catalyst can be attributed to a
consequence of the co-oligomerization reaction of ethylene with the olefin reaction
products of this catalyst (produced after 1-butene insertion at the metal-hydride bond),
which has already been observed for cobalt and iron catalysts [39].
Comparison of the results described in the literature with those for the nickel
complexes with β-diimine ligands herein described shows that the catalysts in this paper
have similar activities but much higher selectivities towards C4 products and, more
importantly, towards α-olefins [40-42]. Zhang [43] used β-diimine complexes for ethylene
oligo/polymerization with MAO as an activator and obtained oligomers in the range from
14
Page 16 of 35

C4 to C8 and high molecular weight polyethylene with an activity of approximately 18 x
103 h-1. Li [44] studied the effect of CF3 substituents on the electrophilicity of catalysts
used for ethylene oligo/polymerization. In this study, MAO was used as an activator, and
oligomers ranging from C4 to C8 were obtained simultaneously with polyethylene with an
activity of 350 x 103 h-1.
Oligomerization using EASC like co-catalyst were very active. The EASC is a
known choice was based at literature. Eberhardt at al described EASC like the most
efficient activator to their nickel complex because this co-catalyst is both Lewis acid and
alkylating agent. [45].
The reactions using these nickel complexes heterogenized in hybrid silica were
performed at similar conditions to those used for the homogeneous-phase runs to compare
their behaviors. Different Al/Ni ratios (100 to 1200) and different amounts of catalyst (6.5
µmol and 20 µmol) have been tested, and the results are shown in Table 3.
Table 3. Results of ethylene oligomerization by heterogenized nickel-β-diimine/silica
complexes.
a
Entry
Complex
Ni-HA
Ni-HB
µmol
20
Al/Ni
TOF
(103·h-1)
1.8
2.2
12.8
2.5
4.3
2.3
10.3
13.8
13.3
0
-C4
(%)
92
96
91
1
2
3
4
5
6
7
8
100
200
400
100
200
400
300
600
94
20
100
100
95
93
96
6.5
--
Ni-HC
W
9
1200
10b
-
c
Complex: T = 10 °C, pressure = 15 atm, reaction time = 0.5 h, solvent = toluene (60 mL), co-catalyst =
a
b
c
EASC. SC4 = 100 %. Silica synthesized without the presence of an organic precursor. Same volume as
entry 2.
15
Page 17 of 35

The mechanism of the catalytic reaction can be described as involving an alkyl-
nickel species, 2, (obtained by alkylation with the EASC co-catalyst) or a nickel-hydride
species, 5, (obtained by ethylene insertion at the nickel-carbon bond on species 2 followed
by -elimination and 1-butene descoordination). Both species are formed from the Ni-A,
Ni-B, Ni-C, Ni-HA, Ni-HB or Ni-HC complexes (Figure 7). Ethylene insertion on species
2 gives 3. -elimination on species 3 gives 4. The intermediate 4 can give 1-butene via
dissociation or species 6 via 1-butene insertion at the nickel-hydride bond. Compound 6
can give 2-butene from -elimination associated with the dissociation reaction.
INSERT FIGURE 7
The three nickel-β-diimine/silica complexes have the same selectivity for C4
fraction (100 %) and similar selectivities for α-C4 (above 90 %).
The best activities for the Ni-HA, Ni-HB and Ni-HC complexes were obtained
with Al/Ni ratios of 400, 200 and 600, respectively. The Ni-HC catalyst showed the best
activity (13.8 x 103 h-1) among the catalysts tested. Busico et al. [46] demonstrated that a
lower catalyst concentration coupled with a greater amount of active species available for
olefins provides higher catalytic activity. The Ni-HA catalyst showed a higher activity of
12.8 x 103 h-1 compared to the Ni-HB catalyst, which had an activity of 4.3 x 103 h-1.
Comparing the results for the homogeneous nickel-β-diimine complex in entries 1,
2 and 3 of Table 2 with the results for the heterogenized nickel-β-diimine/silica complex
shows that heterogenization is accompanied by a considerable decrease in catalytic
activity. This fact can be explained by the limited accessibility of the metal centers to the
olefin in the heterogenized complexes, which leads to a smaller number of available active
species and lower catalytic activity as described in other heterogenized systems [47,48]. It
was also noted that the heterogenized complexes required a greater amount of the
alkylaluminum activator than the homogeneous complex. This result is most likely due to
the presence of hydroxyl groups on the support surface of the heterogenized catalyst,
which consume part of the alkylaluminum. The heterogenized complexes show better
16
Page 18 of 35

selectivities for the C4 and α-C4 fractions than the homogeneous complexes. The addition
of the silica catalyst support also acts to sterically hinder the formation of internal olefins.
Entry 10 in Table 3 presents the results of the reaction with the material without
organic matter (White) W, which shows no catalytic activity.
Figure 8 presents the results for the recycle reactions using the heterogenized Ni-
HA catalyst. The catalyst shows a decrease in activity for the first and second recycles, but
the selectivity for α-C4 is maintained.
INSERT FIGURE 8
4. Conclusions
Nickel-β-diimine complexes and nickel-β-diimine/silica are active in ethylene
oligomerization. The homogeneous-phase reaction shows the strong influence of the ligand
structure on the formation of the products. The best results in oligomerization activity and
selectivity for the homogeneous complexes are obtained with an Al/Ni molar ratio of 100.
An excess of alkylaluminum results in a decrease in the activity, and an increase in the
ethylene pressure shows first-order kinetics and affords the best activities without changing
the selectivity. The Ni-A complex has an activity of approximately 210.0 x 103 h-1, and the
Ni-B and Ni-C complexes have activities of 18.3 x 103 h-1 and 18.7 x 103 h-1, respectively.
The decrease in activity for these catalysts is most likely due to the higher steric hindrance
of the methyl groups on the ligands, which produce catalysts with high selectivities for C4
products (>95 %) and α-olefins (>85 %).
The hybrid silicas were synthesized by the sol-gel method, which resulted in good
incorporation of the organic phase. The β-diimine/silica xerogels are thermally stable. It
was possible to incorporate nickel into the resulting hybrid silica complex to form
heterogenized nickel-β-diimine/silica.
The heterogeneous catalysts had lower activity than their homogeneous analogues,
but they demonstrated much better selectivity for C4 fractions (100 %) and for α-C4
17
Page 19 of 35

fractions (>90 %). It is observed that a smaller amount of the catalyst precursor was needed
to obtain a higher catalytic activity. The silica support also prevents the formation of
internal olefins.
Recycle reactions were performed with the heterogenized Ni-A complex, which
was shown to be active and selective in the recycle reactions and is therefore promising for
large-scale use.
5. Acknowledgements
The authors thank CAPES and FAPERGS/PRONEX/CNPq for their financial
support.
6. References
[1] A. Forestière, H. Olivier-Bourbigou, L. Saussine, Oil Gas Sci. Technol. Rev. IFP 64
(2009) 649-667.
[2] M.O. de Souza, E.P. Pinto, D.J. Melchiors, R. F. de Souza, S.M.O. Einloft, K.D.S.
Bernardo, L.C. Simon, C.D.N. Wyrvalski, Propene Dimerization Process. PI9303804.
[3] Dimersol Y. Chauvin, D. Commereuc, G. Leger, J. Gaillard, N.H. Phung, Catalyst and
process for olefin oligomerization. US4362650.
[4] Y. Chauvin, R.F. de Souza, H. Olivier, Catalytic composition for biphase catalysis, in
particular using nickel complexes, and a process for the oligomerization of olefins.
US5723712.
[5] B.A. Aufdembrink, M.M. Wu, Oligomerization of ethylene with supported nickel
catalyst. US2351190.
[6] C. Bianchini, G. Giambastiani, L. Luconi, A. Meli, Coord. Chem. Rev. 254 (2010) 431-
455.
[7] G.C. Vernon, and S.K. Spitzmesser, Chem. Rev. 103 (2003) 283-315.
[8] G.J.P. Britovsek, V.C. Gibson, D.F. Wass, Angew. Chem. Int. Ed. 38 (1999) 428-447.
[9] F. Speiser, P. Braunstein, L. Saussine, Acc. Chem. Res. 38 (2005) 784-793.
[10] L. Bourget-Merle, M.F. Lappert, J.R. Severn, Chem. Rev. 102 (2002) 3031-3065.
[11] D.J. Mindiola, Angew. Chem. Int. Ed. 48 (2009) 6198-6200.
18
Page 20 of 35

[12] J. Feldman, S.J. McLain, A. Parthasarathy, W.J. Marshall, J.C. Calabrese, S.D. Arthur,
Organometallics 16 (1997) 1514-1516.
[13] M. Rahim, N.J. Taylor, S.X. Xin, S. Collins, Organometallics 17 (1998) 1315-1323.
[14] L.K. Johnson, C.M. Killian, M. Brookhart, J. Am. Chem. Soc. 117 (1995) 6414-6415.
[15] S.D. Ittel, L.K. Johnson, M. Brookhart, Chem. Rev. 100 (2000) 1169-1203.
[16] C.J. Brinker, G.W. Scherer. Sol-Gel Science: the physics and chemistry of sol-gel
processing. San Diego, 1990.
[17] R.A. Sheldon, I.W.C.E. Arends, H.E.B. Lempers, Collect. Czech. Chem. Commun. 63
(1998) 1724-1742.
[18] A.P. Wight, M.E. Davis, Chem. Rev. 102 (2002) 3589-3613.
[19] B. Meunier, Chem. Rev. 92 (1992) 1411-1456.
[20] C. Sanchez, L. Rozes, F. Ribot, C. Laberty-Robert, D. Grosso, C. Sassoye, C.
Boissiere, L. Nicole C. R., Chimie 13 (2010) 3-39.
[21] C. Sanchez, B. Julian, P. Belleville, M. Popall, J. Mater. Chem. 15 (2005) 3559-3592.
[22] T. Calvete, A.E. Gerbase, E.V. Benvenutti, J.R. Gregorio, D. Robinson, J. Sol-Gel Sci.
Technol. 50:6 (2009) 69-76.
[23] K. Dallmann, R. Buffon, Catal. Comm.1 (2000) 9-13.
[24] M. Buechler-Skoda, R. Gill, D. Vu, C. Nguyen, G. Larsen, Appl. Catal. A: Gen. 185
(1999) 301-310.
[25] M.C. Brasil, E.V. Benvenutti, J.R. Gregorio, A.E. Gerbase, Reac. Func. Pol. 63 (2005)
135-141.
[26] S. Teixeira, K. Dallmann, U. Schuchardt, R. Buffon, J. Mol. Catal. A: Chem. 182
(2002) 167-173.
[27] Y. Guo, X.Q. Zhang, W.M. Dong, J. Mol. Catal. A: Chem. 237 (2005) 45-49.
[28] J.H.Z. dos Santos, H.T. Ban, T. Teranishi, T. Uozumi, T. Sano, K. Soga, Appl. Catal.
A: Gen. 220 (2001) 287-302.
[29] J.H.Z. dos Santos, T. Teranishi, T. Uozumi, T. Sano, K. Soga, Polymer 42 (2001)
4517-4525.
[30] C.G. de Souza, M.Sc. Thesis, PPG-Química, UFRGS, Porto Alegre, 2008.
[31] L.M. Tang, Y.Q. Duan, X.F. Li, Y.S. Li, J. Organometal. Chem. 691 (2006) 2023-
2030.
[32] P.H.M. Budzelaar, R. de Gelder, A.W. Gal, Organometallics 17 (1998) 4121-4123.
[33] B.J. Hathaway, D.G. Holah, J. Chem. Soc. (1964) 2408-2416.
[34] C.J. Brinker, G.W. Scherer, Sol–gel Science, Academic Press, New York, 1990.
19
Page 21 of 35

[35] G. Leofanti, M. Padovan, G. Tozzola, B. Venturelli, Catal. Today 41 (1998) 207-219.
[36] C. Bianchini, G. Giambastiani, I.G. Rios, G. Mantovani, A. Meli, A.M. Segarra,
Coord. Chem. Rev. 250 (2006) 1391-1418.
[37] J.D. Azoulay, R.S. Rojas, A.V. Serrano, H. Ohtaki, G.B. Galland, G. Wu, G.C. Bazan,
Angew. Chem. Int. Ed. 48 (2009) 1089-1092.
[38] J.D. Azoulay, Z.A. Koretz, G.A. Wu, G.C. Bazan, Angew. Chem. Int. Ed. 49 (2010)
7890-7894.
[39] D. Thiele, R.F. de Souza, Catal. Lett, 138 (2010) 50-55.
[40] M.L. Mignoni, M.O. de Souza, S.B.C. Pergher, R.F. de Souza, K. Bernardo-Gusmão,
Appl. Catal. A. 374 (2010) 26-30.
[41] M. Sauthier, F. Leca, R.F. de Souza, K. Bernardo-Gusmão, L.F.T. Queiroz, L. Toupet,
R. Réau, New J. Chem. 26 (2002) 630-635.
[42] E. NelKenbaum, M. Kapon, M.S. Eisen, J. Organomet. Chem. 690 (2005) 2297-2305.
[43] J.K. Zhang, Z.F. Ke, F. Bao, J.M. Long, H.Y. Gao, F.M. Zhu, Q. Wu, J. Mol. Catal. A.
249 (2006) 31-39.
[44] Y.F. Li, M. Gao, Q. Wu, Appl. Organometal. Chem. 22 (2008) 659-663.
[45] G.G. Eberhardt, W.P. Griffin, Journal of Catalysis, 16 (1970) 245-253.
[46] V. Busico, R. Cipullo, J. Am. Chem. Soc.116 (1994) 9329-9330.
[47] A.M. Uusitalo, T.T. Pakkanen, E.I. Iskola, J. Molec. Catal. A: Chem. 177 (2002) 179-
194.
[48] A.M.R. Galletti, G. Geri, G. Sbrana, M. Marchionna, P. Ferrarini, J. Molec. Catal. A:
Chem. 111 (1996) 273-280.
20
Page 22 of 35

List of Figures:
Figure 1. β-Diimine ligands.
Figure 2. Synthesis of nickel-β-diimine complexes.
Figure 3: Hybrid materials synthesis.
Figure 4. Infrared spectra of the Ni-C complex and of ligand C.
Figure 5. Infrared spectra of the HA, HB, HC and W hybrid silicas.
Figure 6. N2 adsorption and desorption isotherms for the HC hybrid β-diimine/silica
xerogel.
Figure 7. Mechanism of ethylene dimerization catalyzed by nickel-hydride and/or nickel-
alkyl.
Figure 8. Results of the Ni-A catalyst recycle reactions.
21
Page 23 of 35

FIGURE ABSTRACT
R2
R2
R2
R2
R1
R1
Br
R1
Br
R1
R1
R1
R1
R1
N
N
N
N
Ni
Ni
Si
O
Si
N
N
Br
Si
O
HO
Si
Br
O
O
HO
R1
Si
N
N
HO
Si
O
OH
O
O
O
HO
R1
R1
R1
R1
OH
O
Si
Si
O
R1
R1
R1
Si
Si
O
O
O
O
O
O
Si
Si OH
Si
Si OH
OH
HO
Si OH
HO
HO
O
Si OH
OH
HO
O
HO
HO
O
R2
R2
O
R2
R2
Si
Si
HO
Reaction of ethylene
oligomerization
HO
R1
Ni(CH3CN)Br2
R1
R1
R1
R1
R1
N
R2
N
R2
N
R2
N
R2
Ni
R
1 Br
Br
R1
22
Page 24 of 35

FIGURE 1
Figure 1. β-Diimine ligands.
23
Page 25 of 35

FIGURE 2
Figure 2. Synthesis of nickel-β-diimine complexes.
24
Page 26 of 35

FIGURE 3
Figure 3: Hybrid materials synthesis.
25
Page 27 of 35

FIGURE 4
Figure 4. Infrared spectra of the Ni-C complex and of ligand C.
26
Page 28 of 35

FIGURE 5
Figure 5. Infrared spectra of the HA, HB, HC and W hybrid silicas.
27
Page 29 of 35