Switching from Ethylene Trimerization to Ethylene Polymerization by Chromium Catalysts Bearing SNS Tridentate Ligands: Process Optimization Using Response Surface Methodology

Article abstract of DOI:10.1007/s10562-018-2571-5

Two types of chromium catalysts bearing pyridine and amine based SNS ligands under the title of (pyridine-SNS-alkyl/CrCl₃) and (amine-SNS-alkyl/CrCl₃) were synthesized. Different thiolates such as octyl, pentyl, butyl, cyclohexyl and cyclopentyl thiolates were reacted with 2,6-pyridine-dimethylene-ditosylate (PMT)/THF solution at room temperature. Then, the purified pyridine-based SNS ligands (1–5) were reacted with CrCl₃ (THF)₃ to obtain the pyridine-SNS-alkyl/CrCl₃ catalysts (6–10) in 50–70% yields. MMAO-activated pyridine-SNS-alkyl/CrCl₃ catalysts were capable of oligomerizing ethylene. Statistical experimental design was conducted using the central composite design method and surface methodology to study of the effect of important parameters such as ethylene pressure, Al/Cr ratio, catalyst concentration and the reaction temperature on 1-C₆ productivity of catalyst (7). A quadratic polynomial equation was developed to predict the 1-C₆ productivity. Ethylene oligomerization using the catalyst (7) was lead to a optimized reaction conditions, including the ethylene pressure of 19.5?bar, the temperature of 58.2?°C, the MMAO co-catalyst, Al/Cr = 841 and the catalyst concentration of 8.7?μmol. The catalytic properties for ethylene oligomerization are strongly affected by reaction temperature. The experimental results indicated the reasonable agreement with the predicted values. The transformation from ethylene trimerization to ethylenev polymerization of catalyst system (7) was occurred by exchanging the reaction pressure. Influence of ligand structure with different substitutions on sulphur atom on productivity and selectivity was investigated. 1-C₆ with the high selectivity and productivity 4318 (g 1-C₆/g Cr h) was obtained for catalyst (7). In the second part, 1-C₆ was obtained with high selectivity and productivity around 141 × 10³ (g 1-C₆/g Cr h) for amine-based catalyst. All amine-based catalysts (14–16) showed considerably higher catalytic activities compared to pyridine-based catalysts. According to the TGA analysis the thermal stability of pyridine-based catalysts was found to be higher than the amine-based catalysts. Graphical Abstract: Chromium complexes bearing pyridine and amine based SNS ligands have been synthesized and their catalytic performance in ethylene oligomerization has been investigated. A switching from ethylene trimerization to ethylene polymerization of the catalyst (7) was obtained utilizing exchanging of the ethylene pressure. [Figure not available: see fulltext.].

Full text of DOI:10.1007/s10562-018-2571-5

Catalysis Letters
https://doi.org/10.1007/s10562-018-2571-5
Switching from Ethylene Trimerization to Ethylene Polymerization
by Chromium Catalysts Bearing SNS Tridentate Ligands: Process
Optimization Using Response Surface Methodology
Majid Soheili1 · Zahra Mohamadnia1 · Babak Karimi1
Received: 12 June 2018 / Accepted: 23 September 2018
© Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract
Two types of chromium catalysts bearing pyridine and amine based SNS ligands under the title of (pyridine-SNS-alkyl/CrCl3)
and (amine-SNS-alkyl/CrCl3) were synthesized. Different thiolates such as octyl, pentyl, butyl, cyclohexyl and cyclopentyl
thiolates were reacted with 2,6-pyridine-dimethylene-ditosylate (PMT)/THF solution at room temperature. Then, the purified
pyridine-based SNS ligands (1–5) were reacted with CrCl3 (THF)3 to obtain the pyridine-SNS-alkyl/CrCl3 catalysts (6–10)
in 50–70% yields. MMAO-activated pyridine-SNS-alkyl/CrCl3 catalysts were capable of oligomerizing ethylene. Statistical
experimental design was conducted using the central composite design method and surface methodology to study of the
effect of important parameters such as ethylene pressure, Al/Cr ratio, catalyst concentration and the reaction temperature
on 1-C6 productivity of catalyst (7). A quadratic polynomial equation was developed to predict the 1-C6 productivity. Eth-
ylene oligomerization using the catalyst (7) was lead to a optimized reaction conditions, including the ethylene pressure of
19.5 bar, the temperature of 58.2 °C, the MMAO co-catalyst, Al/Cr=841 and the catalyst concentration of 8.7 µmol. The
catalytic properties for ethylene oligomerization are strongly affected by reaction temperature. The experimental results
indicated the reasonable agreement with the predicted values. The transformation from ethylene trimerization to ethylenev
polymerization of catalyst system (7) was occurred by exchanging the reaction pressure. Influence of ligand structure with
different substitutions on sulphur atom on productivity and selectivity was investigated. 1-C6 with the high selectivity and
productivity 4318 (g 1-C6/g Cr h) was obtained for catalyst (7). In the second part, 1-C6 was obtained with high selectiv-
ity and productivity around 141×103 (g 1-C6/g Cr h) for amine-based catalyst. All amine-based catalysts (14–16) showed
considerably higher catalytic activities compared to pyridine-based catalysts. According to the TGA analysis the thermal
stability of pyridine-based catalysts was found to be higher than the amine-based catalysts.
Graphical Abstract
Chromium complexes bearing pyridine and amine based SNS ligands have been synthesized and their catalytic performance
in ethylene oligomerization has been investigated. A switching from ethylene trimerization to ethylene polymerization of
the catalyst (7) was obtained utilizing exchanging of the ethylene pressure.
P=15 bar
6420 g 1-C6/g Cr.h
N
Al/Cr=900, T= 80 °C, Cat=3 µm
MMAO, ethylene
Cl
S
Cr
S
R
Cl
R
Cl
R=Pentyl
P=27 bar
Polyethylene
101282 g PE/g Cr.h
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s10562-018-2571-5) contains
supplementary material, which is available to authorized users.
Extended author information available on the last page of the article
Vol.:(0123456789)
1 3

M. Soheili et al.
Keywords Ethylene trimerization · Chromium catalyst · 1-Hexene · Selctivity · Tridentate ligands
the amine-SNS-Alkyl/CrCl3 catalysts was developed by sev-
eral scientists, including McGuinness [30] and Mohamadnia
[21].
1 Introduction
Currently ethylene oligomerization has attracted commercial
and academic media attentions affording a wide spectrum
of linear α-olefins (LAOs) follow a Schulz-Flory or a Pois-
son type of distribution [1]. LAOs in the range of C4–C20
are the most important compounds produced by metal-cat-
alysed ethylene oligomerization processes and used more
than 3 million tons each year [2]. These valuable commod-
ity chemicals significantly used as precursors in many areas
of industry, such as detergents, plasticizer alcohols, and as
co-monomer in the production of linear low density poly-
ethylene (LLDPE) [3]. A main problem of scientists and
various industries is the non-selective Schulz-Flory distribu-
tion of oligomers requiring further processes to separate the
preferred products [4]. The light fraction of LAOs, such as
1-butene, especially 1-hexene, and 1-octane are mainly used
for the production of LLDPE [5].
Methylaluminoxane (MMAO)-activated pyridine-
SNS-alkyl/CrCl3 catalysts have been tested for the ethyl-
ene oligomerization. Pyridine-SNS-alkyl/CrCl3 catalysts
(Alkyl = octyl, pentyl and cyclopentyl) have not been
reported yet. For the first time, the central composite method
was used to design and optimization of the ethylene trimeri-
zation experiments of pyridine-SNS-pentyl/CrCl3 catalyst
using response surface method (Desing-Expert®Version
7.0.0). The effects of various oligomerization conditions
such as ethylene pressure, Al/Cr ratio, reaction temperature
and catalyst concentration on ethylene trimerization were
systematically investigated and the 1-C6 productivity was
studied as the experimental response. A literature review
revealed that design and optimization of the ethylene tri-
merization experiments for this type of catalysts using the
response surface method is novel and has not been reported
previously. Ethylene oligomerization with amine-SNS-
alkyl/CrCl3 catalysts were tested and compared with pyr-
idine-SNS-alkyl/CrCl3 catalysts. Finally, 1-C6 productivity
and thermal stability of these two types of catalysts were
compared.
So nowadays, both industry and academia are focused
on the selective processes for the production of LAOs espe-
cially 1-hexene [6]. The catalytic oligomerization of ethyl-
ene for the preparation of LAOs has increased tremendously
over the last decade [7]. In the most widely used industrial
processes for the preparation of LAOs such as the two-step
Ziegler stoichiometric process (INEOS), one-step the Shell
higher olefins process (SHOP), the Ziegler process (Chev-
ron-Phillips Chemicals), Idemitsu and SABIC processes [8],
the homogeneous catalysts are widely used. A vast number
of selective ethylene trimerization catalyst based on chro-
mium, titanium, zirconium, tantalum, and hafnium have
been developed [9].
2 Experimental
2.1 General
All reactions were carried out under an inert atmosphere of
argon or nitrogen using Schlenk techniques and glovebox
equipped with a purifier unit. The cocatalyst Modified Meth-
ylaluminoxane (MMAO) (7 wt% in toluene), CrCl3 (THF)3,
2,6-pyridinedimethanol and molecular sieve were obtained
from the Aldrich (Germany). Butanethiol, 1-pentanethiol,
1-octanethiol, cyclopentanethiol, cyclohexanethiol, ethanol,
THF, toluene, p-Toluenesulfonyl chloride, Na2CO3, MgSO4,
NaCl, Na2SO4, sodium and NaOH were purchased from
Merck (Germany). Argon and ethylene provided by Bandar
Imam Petrochemical Company (Iran) were purified by pass-
ing through NaOH, activated silica gel and molecular sieves
3 Å columns. Fourier transform infrared (FT-IR) spectros-
copy was carried out by means of a Bruker Vector 22 FT-IR
Among all of the transition metal-based oligomeriza-
tion catalysts, chromium catalysts have been reported for
both selective and non-selective ethylene oligomerization
processes [10]. The first Cr-catalysed selective trimeriza-
tion of ethylene to 1-hexene has been reported by Manyik
[11]. The other chromium-based catalysts include Phillips
[12], Mitsubishi [13], BP Chemicals [14], and Sasol [15].
In the past decade, several ethylene oligomerization chro-
mium catalysts bearing NˆNˆN [16], PˆNˆP [17, 18], SˆNˆS
[19–21], PˆN [22], NˆNˆP [23], NˆOˆN and NˆSˆN [24],
NˆPˆN [25], PˆNˆC [26], NˆO [27], and PˆC [28], ligands
have been reported and their catalytic behaviour in ethylene
oligomerization have also been explored.
spectrometer (Germany) in the region of 400–4000 cm−1
.
So in the present work, two types of novel chromium
catalysts bearing pyridine and amine based SNS ligands
under the titles of pyridine-SNS-alkyl/CrCl3 and amine-
SNS-alkyl/CrCl3 have been prepared and evaluated as cata-
lysts for ethylene oligomerization. The pyridine-SNS-Alkyl/
CrCl3 catalysts initially developed by Gambarotta [29], and
The samples were prepared by mixing with KBr powder in
the form of pellets at ambient conditions and measured in
the transmission mode. UV–Vis absorbance spectra of the
samples were measured using UV/Vis spectrophotometer
(Pharmacia Biothech Ultrospec 4000) and acquired between
1 3

Switching from Ethylene Trimerization to Ethylene Polymerization by Chromium Catalysts Bearing…
200 and 800 nm. Thermal analysis of sample was performed
by NETZSCH STA 409 PC 141H Luxx (Germany) with a
heating rate of 20 K min−1 under N2 atmosphere. Tempera-
ture range was set at 25–900 °C. Prior to analysis, different
was obtained in yield of 76% and high purity (>98% pure
by NMR). IR (KBr) (υmax cm−1): 3060 (CHAromatic), 2910
(CHaliphatic), 1600, 1356, 1167 (C–O), 1065, 1032 (S=O).
1H NMR (400 MHz, CDCl3): δH 2.47 (6H, s, CH3), 5.08
(4H, s, py-CH2-O), 7.35 (6H, d, py and ArH), 7.73 (1H,
t, py), 7.82 (4H, d, ArH). 13C NMR (100 MHz, CDCl3): δ
21.7, 71.3, 121.4, 128.1, 129.9, 132.7, 137.9, 145.2, 153.5.
Elemental analysis: calculated for C21H21NO6S2 (found): C
56.36(56.26), H 4.73(5.02), N 3.13(3.07), O 21.45(20.63)
and S 14.33(15.02).
1
samples were oven-dried at 60 °C for about 24 h. The H
NMR and 13C NMR spectra of ligands were recorded with
a Bruker 400 MHz Ultrashield TM NMR instrument (Ger-
1
many) at room temperature. For H NMR and 13C NMR
analysis, the samples were dissolved in CDCl3. Elemental
analysis was performed using a Vario EL III CHNOS ele-
mental analyser. The trimerization reaction mixtures were
analysed using a gas chromatograph (VARIAN CP 3800)
with an FID detector.
2.3 Preparation of the Pyridine‑Based SNS Ligands
(Pyridine‑SNS‑Alkyl) (1–5)
2.2 Preparation of the Precursor
2.3.1 2,6‑Bis(CH3(CH2)3SCH2)Pyridine (Pyridine‑SNS‑Butyl)
(1)
2,6‑Pyridine‑Dimethylene‑Ditosylate (PMT)
The title compound, PMT was synthesized according to
Komarsamy et al. [31] method with relatively few modifi-
cations. In a 100 mL round bottom flask (RBF) with a stir
bar, NaOH (0.862 g, 21 mmol) and 2,6-pyridine dimethanol
(0.3 g, 2.1 mmol) were dissolved in a mixture of THF–water
(1:1) (16 mL) and the resulting solution was cooled to 0 °C.
Then, P-toluenesulfonyl chloride (TSCl) (0.8 g, 4.2 mmol)
dissolved in THF (8 mL) was added and the reaction mix-
ture stirred for 15 min at 0 °C. Then it was stirred at room
temperature for about 4 h. Water (21 mL) was added and
the mixture was extracted with distilled DCM (8 mL) three
times. The organic layer was washed with saturated NaCl
solution (brine) to remove the bulk of the water. Then, anhy-
drous Na2SO4 drying agent was added to eliminate all traces
of water. Then the organic layer was evaporated and diethyl
ether was added to the residue and the crystalline precipi-
tate was collected by filtration after cooling in a refrigerator
for 3 h. It was washed with diethylether (three times) and
dried under diminished pressure. PMT (white crystalline)
Pyridine-SNS-butyl ligand was prepared according to the
literature procedures with a slight modification (Scheme 1)
[32]. Under argon atmosphere, butanethiol (0.406 g,
4.5 mmol) in ethanol (1 mL) was added dropwise to the
stirred solution of sodium metal (0.103 g, 4.5 mmol) in
ethanol (1.5 mL) at 0 °C and the solution was stirred for
40 min. The prepared mixture was added dropwise to the
solution of PMT (1 g, 2.25 mmol) in ethanol and THF
(1:2) (3 mL) and stirring was continued for 5 min. Then
the mixture was heated under reflux overnight. The sol-
vent was removed under vacuum and diethyl ether was
added in order to remove the unreacted materials. The
organic layer was washed with aqueous Na2CO3 (8 w/v%)
followed by deionized water (2 × 30 mL), dried over
anhydrous MgSO4, filtered, concentrated under vacuum
and purified using column chromatography using ethyl
acetate:hexane (8:2) to give compound 1 as a yellow oil
(0.31 g, 49%). IR (KBr) (υmax cm−1): 3060 (CHAromatic),
1
2927 (CHaliphatic), 1589, 1573, 1454, 813, 748. H NMR
Scheme 1 General procedure
for synthesis of pyridine-SNS-
alkyl with different R substitu-
N
°
Stiring at 78 C
Solvent evaporation
OTS
°
OTS
tions
R SH + Na (NaOH)/( EtOH/THF)
Diethyl ether was used for extraction
12 hour
0 C for 40 min
R=
Butyl (1)
Pentyl (2)
Octyl (3)
Cyclopentyl (4)
Cyclohexyl (5)
Washing with aqueous Na2CO3 solution
Dring with MgSO4
Vacuum evaporation
Filtration
N
S
S
Column chromatography method
R
R
R=
Insoluble residue
(1) (Yellow oil (49%))
(2) (White semi-crystalline (70%))
(3) (White semi-crystalline (60%))
(4) (Orange oil (60%))
(5) (Milky white oil (70%))
1 3

M. Soheili et al.
(400 MHz, CDCl3): δH 0.85 (6H, t, CH3), 1.30 (4H, m,
S(CH2)2CH2CH3), 1.50 (4H, m, SCH2CH2CH2CH3), 2.40
(4H, t, SCH2(CH2)2CH3), 3.88 (4H, s, py–CH2S), 7.28
(2H, d, py), 7.63 (1H, t, py). 13C NMR (100 MHz, CDCl3):
δ 13.66 (S(CH2)3CH3), 22.17 (S(CH2)2CH2CH3), 29.92
(SCH2CH2CH2CH3), 31.27 (SCH2(CH2)2CH3), 38.00
(py–CH2–S), 121.02 (CH–py), 137.22 (CH–py), 158.46
(C–py). Elemental analysis: calculated for C15H25NS2
(found): C 63.55(63.44), H 8.89(8.92), N 4.94(5.13), and
S 22.62(22.51).
2.3.4 2,6‑Bis((CH2)4CHSCH2)Pyridine
(Pyridine‑SNS‑Cyclopentyl) (4)
The title ligand (4) was synthesized according to the
method reported by Temple et al. [29] with some modifica-
tions (Scheme 1). Under argon atmosphere, a solution of
cyclopentanethiol (0.46 g, 4.5 mmol) in ethanol (5 mL) was
added dropwise to a solution of sodium hydroxide (0.18 g,
4.5 mmol) in ethanol (7.5 mL) and stirred for 40 min. Then it
was added dropwise to the solution of PMT (1 g, 2.25 mmol)
in THF (12 mL). After 24 h stirring at room temperature,
the solvent was removed under vacuum and the residue was
twice extracted to remove unreacted materials by adding
DCM/deionized water (1:1) (8 mL). The combined aque-
ous layers were washed with DCM (3 × 20 mL), and the
combined organic layers were dried over anhydrous MgSO4,
concentrated and purified using column chromatography to
give an orange oil (4) (0.39 g, 60%). IR (KBr) (υmax cm−1):
3085 (CHAromatic), 2954 (CHaliphatic), 1590, 1572, 1451,
813, 703. 1H NMR (400 MHz, CDCl3): δH 1.4–1.6 (8H, m,
SCH(CH2)2(CH2)2), 1.95–2.1 (8H, m, SCH(CH2)2(CH2)2),
3–3.2 (2H, m, SCH(CH2)4), 3.87 (4H, s, py–CH2–S), 7.27
(2H, d, py), 7.62 (1H, t, py). 13C NMR (100 MHz, CDCl3):
δ 24.89 SCH(CH2)2(CH2)2, 33.68 SCH(CH2)2(CH2)2, 38.37
SCH(CH2)4, 43.56 (py–CH2–S), 121.01 (CH–py), 137.18
(CH–py), 158.57 (C–py). Elemental analysis: calculated
for C17H25NS2 (found): C 66.40(65.77), H 8.19(10.49), N
4.55(4.05) and S 20.85(19.69).
2.3.2 2,6‑Bis(CH3(CH2)4SCH2)Pyridine (Pyridine‑SNS‑Pentyl)
(2)
Pyridine-based ligand with pentyl substituent (2) was
synthesized according to the method 1 with the fol-
lowing molar ratio of starting materials, pentanethiol
(0.62 g, 6 mmol), sodium (0.138 g, 6 mmol), PMT
(1.34 g, 3 mmol) to give compound 2 as a white semi-
crystalline solid (0.1 g, 70%). IR (KBr) (υmax cm−1): 3050
(CHAromatic), 2924 (CHaliphatic), 1589, 1573, 1454, 820, 748.
1H NMR (400 MHz, CDCl3): δH 0.88 (6H, t, S(CH2)4CH3),
1.2–1.50 (8H, m, S(CH2)2(CH2)2CH3), 1.6–1.8 (4H, m,
SCH2CH2(CH2)2CH3), 2.58 (4H, t, SCH2(CH2)3CH3), 4.16
(4H, s, py–CH2–S), 7.57 (2H, d, py), 7.98 (1H, t, py). 13
C
NMR (100 MHz, CDCl3): δ 13.97 (S(CH2)4CH3), 22.25
(S(CH2)3CH2CH3), 28.89 (S(CH2)2CH2CH2CH3), 30.94
(SCH2CH2(CH2)2CH3), 32.29 (SCH2(CH2)3CH3), 34.89
(py–CH2–S), 123.19 (CH–py), 141.41 (CH–py), 157.07
(C–py). Elemental analysis: calculated for C17H29NS2
(found): C 65.54(65.24) H 9.38(9.44) N 4.50(4.90) and
S 20.58(20.42).
2.3.5 2,6‑bis((CH2)5CHSCH2)Pyridine
(Pyridine‑SNS‑Cyclohexyl) (5)
According to the procedure for preparation of 4, a milky-
white oil (5) (yield: 70%) was synthesized. IR (KBr) (υmax
cm−1): 3058 (CHAromatic), 2854 (CHaliphatic), 1588, 1572,
1
2.3.3 2,6‑Bis(CH3(CH2)7SCH2)Pyridine (Pyridine‑SNS‑Octyl)
(3)
1449, 814, 704. H NMR (400 MHz, CDCl3): δH 1.1–1.4
(4H, m, SCH(CH2)4CH2, 1.6–1.8 (8H, m, SCH(CH2)2(C
H2)2CH2), 1.9–2.0 (8H, m, SCH(CH2)2(CH2)3), 2.55–2.7
(2H, m, SCH(CH2)5), 3.86 (4H, s, py–CH2–S), 7.27 (2H,
d, py), 7.61 (1H, t, py). 13C NMR (100 MHz, CDCl3): δ
25.84 SCH(CH2)4CH2, 26.02 SCH(CH2)2(CH2)2CH2, 33.43
SCH(CH2)2(CH2)3, 36.55 SCH(CH2)5, 43.32 (py–CH2–S),
120.89 (CH–py), 137.20 (CH–py), 158.72 (C–py). Elemental
analysis: calculated for C19H29NS2 (found): C 68.00(66.28),
H 8.71(10.79), N 4.17(3.84), and S 19.11 (19.09).
According to the procedure 1, a white semi-crystalline com-
pound (3) (yield: 60%) was synthesized. IR (KBr) (υmax
cm−1): 3065 (CHAromatic), 2921 (CHaliphatic), 1584, 1465,
1
820. H NMR (400 MHz, CDCl3): δH 0.89 (6H, t, CH3),
1.1–1.50 (20H, m, S(CH2)2(CH2)5CH3), 1.6–1.8 (4H, m,
SCH2CH2(CH2)5CH3), 2.5 (4H, t, SCH2CH2(CH2)5CH3),
3.84 (4H, s, py–CH2–S), 7.3 (2H, d, py), 7.64 (1H, t, py).
13C NMR (100 MHz, CDCl3): δ 14.01 (S(CH2)7CH3),
22.6 (S(CH2)6CH2CH3), 28.90 (S(CH2)5CH2CH2CH3),
29.20 (S(CH2)3(CH2)2(CH2)2CH3), 29.30 (S(CH2
)2CH2(CH2)4CH3), 31.70 (SCH2CH2(CH2)5CH3),
31.80 (SCH2(CH2)6CH3), 38.10 (py–CH2–S), 121.10
(CH–py), 137.30 (CH–py), 158.50 (C–py). Elemental anal-
ysis: calculated for C23H41NS2 (found): C 69.81(70.11), H
10.44(10.49), N 3.54(3.35) and S 16.21(16.05).
2.4 General Procedure for the Synthesis
of 2,6‑Bis(AlkylSCH2)Pyridine/CrCl3 Catalysts
(Pyridine‑SNS‑Alkyl/CrCl3) (6–10)
All pyridine-SNS-alkyl/CrCl3 catalysts were synthesized
according to the method reported by Sandro Gambarotta
et al. [29] CrCl3(THF)3 (0.12 mmol) was added to a solution
1 3

Switching from Ethylene Trimerization to Ethylene Polymerization by Chromium Catalysts Bearing…
of pyridine-SNS-alkyl ligands (0.126 mmol) in dry toluene
(5 mL) at room temperature and stirred for 40 min (Scheme
S1). A visible colour change to green was observed. The
catalyst was recovered by centrifuging and decantation of
the reaction mixture. It was then washed with n-hexane
(3×5 mL) and dried under vacuum.
added to the stirred solution of bis(2-chloroethyl)amine
hydrochloride (8.33 mmol) in ethanol (16 mL) at 0 °C. After
stirring at 0 °C for 2 h followed by room temperature for
16 h, the mixture was filtered, the filtrate evaporated using
rotary evaporator, the residue taken up with dry n-hexane
and diethyl ether respectively and filtered again. Finally, the
solvent was evaporated under vacuum to give the amine-
based ligands.
2.4.1 Pyridine‑SNS‑Butyl/CrCl3 (6)
The catalyst (6) was green solid with 61% yield. IR (KBr)
(υmax cm−1): 3100 (CHaromatic), 2900 (CHaliphatic), 1603,
1577, 1459, 889, 794. Elemental analysis calculated for
C15H25Cl3CrNS2 (found): C 40.78(41.10), H 5.70(5.91), N
3.17(2.97), S 14.51(14.34), Cl 24.07 (−), and Cr 11.77 (−).
2.5.1 Bis‑(2‑Butylsulfanyl‑Ethyl)‑Amine Ligand
(Amine‑SNS‑Butyl) (11)
Ligand (11) was colorless oil with 74% yield. IR (KBr) (υmax
cm−1): 3305 (NH), 2960 (CHaliphatic), 1465, 1301, 765. 1H
NMR (400 MHz, CDCl3): δH 0.93 (6H, t, CH3), 1.38–1.52
(4H, m, SC2H4CH2CH3), 1.55–1.61 (4H, m, SCH2CH2C2H5),
1.90 (1H, broad, NH), 2.30 (4H, t, SCH2C3H7), 2.57 (4H, t,
SCH2CH2NH), 2.71 (4H, t, NHCH2). 13C NMR (100 MHz,
CDCl3): δ 13.66 (CH3), 21.60 (SC2H4CH2CH3), 31.63
(SCH2CH2C2H5), 31.80 (SCH2C3H7), 32.29 (SCH2CH2NH),
48.30 (SCH2CH2NH). Elemental analysis: calculated for
C12H27NS2 (found): C 57.77 (58.46), H 10.91 (12.15), N
5.61 (6.12) and S 25.71 (23.27).
2.4.2 Pyridine‑SNS‑Pentyl/CrCl3 (7)
The catalyst (7) was green solid with 70% yield. IR (KBr)
(υmax cm−1): 3040 (CHaromatic), 2906 (CHaliphatic), 1603,
1568, 1460, 884, 797. Elemental analysis calculated for
C17H29Cl3CrNS2 (found): C 43.45(43.27), H 6.22(6.87), N
2.98(2.74), S 13.65(13.60), Cl 22.63 (−), and Cr 11.07 (−).
2.4.3 Pyridine‑SNS‑Octyl/CrCl3 (8)
2.5.2 Bis‑(2‑octylsulfanyl‑ethyl)‑amine ligand
(Amine‑SNS‑Octyl) (12)
The catalyst (8) was green solid with 70% yield. IR (KBr)
(υmax cm−1): 3035 (CHaromatic), 2858 (CHaliphatic), 1601,
1569, 1463, 895, 815. Elemental analysis calculated for
C23H41Cl3CrNS2 (found): C 49.86(49.49), H 7.46(7.73), N
2.53(2.40), S 11.57(12.09), Cl 19.19 (−), and Cr 9.38 (−).
Amine-based SNS ligand with octyl substitution (12)
was colourless oil with 74% yield. IR (KBr) (υmax
cm− 1): 3283 (NH), 2928(CHaliphatic), 1450, 1370, 1290,
1
717. H NMR (400 MHz, CDCl3): δH 0.90 (6H, t, CH3),
2.4.4 Pyridine‑SNS‑Cyclopentyl/CrCl3 (9)
1.21–1.40 (16H, m, S(CH2)3C4H8CH3), 1.54–1.63 (4H, m,
SC2H4CH2C5H10), 1.64–1.74 (4H, m, SCH2CH2C6H12),
1.85 (1H, broad, NH), 2.51 (4H, t, SCH2C7H14), 2.68 (4H,
t, SCH2CH2NH), 2.84 (4H, t, NHCH2). 13C NMR (100 MHz,
CDCl3): δ 14.22 (S(CH2)7CH3), 22.56 (S(CH2)6CH2CH3),
28.0 S(C5H10CH2C2H5), 29.21 S(CH2)4CH2(CH2)2CH3,
29.79 S(CH2)3CH2(CH2)3CH3, 31.82 (S(CH2)2CH2
(CH2)4CH3), 32.06 (SCH2CH2(CH2)5CH3), 32.35
(SCH2(CH2)6CH3), 39.27 (SCH2CH2NH), 48.37
(SCH2CH2NH). Elemental analysis: calculated for
C20H43NS2 (found): C 66.42 (65.47), H 11.98 (12.18), N
3.87 (4.82) and S 17.73 (17.53).
The catalyst (9) was green solid with 63% yield. IR (KBr)
(υmax cm−1): 3035 (CHaromatic), 2913 (CHaliphatic), 1602,
1569, 1453, 794, 733. Elemental analysis calculated for
C17H25Cl3CrNS2 (found): C 43.83(44.22), H 5.41(5.82), N
3.01(2.94), S 13.76(13.30), Cl 22.83 (−), and Cr 11.16 (−).
2.4.5 Pyridine‑SNS‑Cyclohexyl/CrCl3 (10)
The catalyst (10) was green solid with 51% yield. IR (KBr)
(υmax cm− 1): 3035 (CHaromatic), 2912 (CHaliphatic), 1608,
1570, 1462, 1034, 1032. Elemental analysis calculated for
C19H29Cl3CrNS2 (found): C 46.20 (46.51), H 5.92(6.08), N
2.84(2.77), S 12.98(12.58), Cl 21.53 (−), and Cr 10.53 (−).
2.5.3 Bis‑(2‑Cyclopentylsulfanyl‑Ethyl)‑Amine Ligand
(Amine‑SNS‑Cyclopentyl) (13)
2.5 General Procedure for Synthesis
of Amine‑SNS‑Alkyl Ligands (11–13)
Ligand (13) was colorless oil with 60% yield. IR (KBr)
(υmax cm−1): 3299 (NH), 2954(CHaliphatic), 1750, (N-H),
1450, 1124, 735. 1H NMR (400 MHz, CDCl3): δH
1.36–1.56 (8H, m, SCH(CH2)2(CH2)2), 1.84–1.94 (8H,
m, SCH(CH2)2(CH2)2), 2.00 (1H, broad, NH), 2.62 (4H,
t, SCH2CH2NH), 2.76 (4H, t, NHCH2), 2.98–3.05 (2H,
All amine-based SNS ligands were synthesized accord-
ing to our previous methods [19, 21]. A solution of NaOH
(25 mmol) and thiol (25 mmol) in ethanol (25 mL) was
1 3

M. Soheili et al.
m, SCH(CH2)2(CH2)2). 13C NMR (100 MHz, CDCl3):
δ 24.72 (SCHC2H4C2H4), 31.98 (SCHC2H4C2H4),
33.90 (SCHC2H4C2H4), 43.68 (SCH2CH2NH), 48.54
(SCH2CH2NH). Elemental analysis: calculated for
C14H27NS2 (found): C 61.48 (60.72), H 9.95 (9.60), N 5.12
(6.86) and S 23.44 (22.82).
injected into the reactor to start the ethylene oligomeriza-
tion. The reaction temperature and ethylene pressure were
kept constant during the reaction. The mixture was stirred
for 30 min. The reactor was then cooled to − 10 °C. The
liquid phase was analysed by GC and solid polyethylene col-
lected by filtration, washed with MeOH, dried, and weighed.
2.6 General Procedure for Synthesis
2.7.1 Study of Catalytic Ethylene Oligomerization Using
Response Surface Methodology
of Amine‑SNS‑Alkyl/CrCl3 Catalysts (14–16)
All amine-SNS-Alkyl/CrCl3 catalysts were prepared accord-
ing our previous methods [19, 21]. The prepared ligands
(0.235 mmol) in THF (5 mL) was added dropwise to a stirred
solution of CrCl3(THF)3 (0.215 mmol) in THF (5 mL) at
room temperature and the solution kept under stirring for
10 min. Then the solvent was removed in vacuum, diethyl
ether (10 mL) added to the solid residue, cooled overnight
in a refrigerator, centrifuged, washed with diethyl ether
(3×10 mL), and dried under vacuum.
Based on our previous studies, four parameters including
ethylene pressure, Al/Cr ratio, reaction temperature and cat-
alyst concentration were identified as the key factors affect-
ing the 1-C6 productivity. For the first time, response surface
methodology (RSM) was employed to investigate the effect
of the independent variables; ethylene pressure, Al/Cr ratio,
reaction temperature and catalyst concentration. The experi-
ments were analysed using central composite design (CCD).
Design matrix was generated and results were statistically
analysed using Design Expert version 7.0.0 by Stat-Ease Inc.
(Minneapolis, USA). The center of the experimental field
was performed 5 times for CCD. Each design was evaluated
separately based on the influence of variables on 1-hexene
productivity (g 1-C6/g Cr h). The design was expressed by
polynomial regression equation to generate the model. The
model quality and significance of various factors consider-
ing the sum of squares and residual sum of squares were
estimated by analysis of variance (ANOVA). The analysis
included the F test, related probability values, coefficient
of determination R2 which measures the goodness of fit of
regression model. The experimental and predicted developed
results were compared in order to approve the validity of
the model.
2.6.1 Amine‑SNS‑Butyl/CrCl3 (14)
The catalyst (14) was blue-green solid with 58% yield. Ele-
mental analysis: calculated for C12H27Cl3CrNS2 (found): C
35.34 (35.02), H 6.67 (7.0), N 3.43 (3.75), S 15.72(15.40),
Cl 26.08 (−), and Cr 12.75 (−).
2.6.2 Amine‑SNS‑Octyl/CrCl3 (15)
The catalyst (15) was green solid with 54% yield. Elemen-
tal analysis: calculated for C20H43Cl3CrNS2 (found): C
46.19(45.25), H 8.33(8.08), N 2.69(2.84), S 12.33(13.38),
Cl 20.45 (−), and Cr 10.00 (−).
2.6.3 Amine‑SNS‑Cyclopentyl/CrCl3 (16)
3 Results and Discussion
The catalyst (16) was green solid with 64% yield. Elemen-
tal analysis: calculated for C14H27Cl3CrNS2 (found): C
38.94(38.83), H 6.30(7.02), N 3.24(3.45), S 14.85(14.30),
Cl 24.63(-), and Cr 12.04(-).
3.1 FT‑IR Analysis of Pyridine‑Based Ligands
and Their Corresponding Catalysts
Table 1 shows a summary of the most important bands in
the FT-IR spectra which correspond to the –C=N stretch-
ing (1580–1600 cm−1), the C=C stretching aromatic
(1500–1600 cm−1), C–S stretching (600–800 cm−1) and C–H
bending vibrations. Significant shifts in vibration frequen-
cies, such as increasing of the C=N stretching frequency of
6 from 1589 to 1603 cm−1 and C–S stretching frequency of
8 from 802 to 815 cm−1, are interpreted as signs of success-
ful complexation of the pyridine-based SNS ligands to the
chromium center. There was no clear difference between
the vibration frequencies of the other bands such as C=C
stretching and C–H bending for ligands and their corre-
sponding catalysts.
2.7 Catalytic Ethylene Oligomerization
The catalytic ethylene oligomerization reactions were per-
formed in a stainless steel research reactor. The high pres-
sure reactor was temperature and pressure controlled. Before
each run, the reactor was dried in an oven at 130 °C for
3 h, vacuumed for 30 min and purged with three cycles of
argon/vacuum. The reactor was then preheated, charged with
toluene and the desired amount of MMAO, pressurized with
ethylene, and stirred at the considered reaction temperature.
After 10 min, equilibrium has been reached. At that time,
a specific volume of the catalyst solution in toluene was
1 3

Switching from Ethylene Trimerization to Ethylene Polymerization by Chromium Catalysts Bearing…
Table 1 Comparison between the important vibration frequencies of
3.2 UV–Visible Spectra of Pyridine‑Based Catalysts
the ligands and their corresponding catalysts
The UV–Visible absorption spectra of the catalysts displayed
meridional coordination of the pyridine-based SNS ligands
(Figs. 1, 2) [33]. As shown in Fig. 1, absorptions in the ultra-
violet region are related to π to π* and ligand to metal transi-
tions. In visible region, octahedral pyridine-based complexes
with d3 electron configuration show three spin-allowed tran-
sitions (Fig. 2). According to the Russell–Saunders cou-
pling, in octahedral fields, F and P terms split into crystal
field components (A2g(F), T2g(F), T1g(F) and T1g(P)) hav-
ing different energies. The pyridine-based catalysts consist
of a regular octahedron CrNS2Cl3 containing CrIII, which
gives rise to two distinct d–d transitions at ~ 450 nm (υ1:
Ligand/complex
FT-IR νmax/cm−1
C=N C=C C–S C–H
Pyridine-SNS-butyl (1)
Pyridine-SNS-butyl/CrCl3 (6)
Pyridine-SNS-octyl (3)
Pyridine-SNS-octyl/CrCl3 (8)
Pyridine-SNS-cyclopenthyl (4)
Pyridine-SNS-cyclopenthyl/CrCl3 (9) 1602 1569 733 1453
1589 1573 748 1454
1603 1577 794 1459
1584 1584 802 1465
1601 1569 815 1463
1590 1572 703 1451
2.5
Pyridine-SNS-Butyl/CrCl3 (6)
4
4A2g(F)→4T1g(P)) and ~ 650 nm (υ2: A2g(F)→4T1g(F)).
Pyridine-SNS-Pentyl/CrCl3 (7)
2
4
The third allowed d–d transition (υ3: A2g(F)→4T2g(F))
Pyridine-SNS-Octyl/CrCl3 (8)
CrCl3(THF)3
almost is placed in visible region (Table 2).
1.5
1
0.5
0
3.3 Evaluation of Thermal Stability
of Pyridine‑Based Catalysts
The thermogravimetric analysis (TGA) curves in nitrogen
atmosphere showed that the pyridine-based complexes dis-
play high thermal stability (Fig. S1). The decomposition
temperatures (Tdec) for 5% weight loss were observed in the
range of 210 °C for compound 7 [34]. During the heating
process, there is a multi-stage weight loss without formation
of steady interfaces. Due to the lower binding energy of C-S
bond, the possibility of the catalyst cleavage is increased
from this position. Table 3 shows the comparison between
experimental and theoretical weight loss percent for various
fragments of catalyst (7). The initial weight loss (~3.9%) is
related to the residual organic solvent and water. The pres-
ence of water molecules has been already seen in the FT-IR
spectra of the catalyst (7). The second and the third weight
loss (~ 43.77%) in the temperature range of 120–380 °C,
presumably due to loss of the SC5H11 fragment of the SNS
ligand. The fourth weight loss (6%) in the temperature range
of 480–480 °C is related to the dissociation of the methyl-
ene. The final weight loss (5.86%) in the temperature range
of 640–800 °C is attributed to the degradation of the C3H3
fragment of the pyridine ring. The residual mass above
200
250
300
350
400
Wavelength (nm)
Fig. 1 Comparison between UV–Vis absorption spectra of
CrCl3(THF)3 and pyridine-SNS-alkyl/CrCl3 catalysts (UV region)
2
Pyridine-SNS-Butyl/CrCl3 (6)
1.75
1.5
1.25
1
Pyridine-SNS-Pentyl/CrCl3 (7)
Pyridine-SNS-Octyl/CrCl3 (8)
CrCl3(THF)3
0.75
0.5
0.25
0
400
450
500
550
600
650
700
750
800
Wavelength (nm)
Fig. 2 Comparison between UV–Vis absorption spectra of
CrCl3(THF)3 with the pyridine-SNS-alkyl/CrCl3 catalysts (visible
region)
Table 2 Experimental values
of the d–d transitions for
the ethylene trimerization
chromium site
Compounds name
Chromophore
4A2g
4A2g
4A2g
(F)→4T1g (P) (F)→4T1g(F)
(F)→4T2g(F)
(υ3)
(υ1)
(υ2)
nm
CrCl3 (THF)3
CrO3Cl3
–
–
–
–
–
–
Pyridine-SNS-butyl/CrCl3 (6)
Pyridine-SNS-pentyl/CrCl3 (7)
Pyridine-SNS-octyl/CrCl3 (8)
CrNS2Cl3
CrNS2Cl3
CrNS2Cl3
450
449
449
670
659
652
1 3

M. Soheili et al.
Table 3 Comparison between
experimental and theoretical
weight loss percent for various
fragments of catalyst (7)
Segments
Molar masses of the
fragments (g/mol)
Temperature
(ºC)
Weight loss %
(Theoretical)
Weight loss %
(Experimental)
2×(SC5H11)
2×(CH2)
C3H3 (pyridine)
CrCl3NC2
206
28
39
120–380
380–460
640–800
Above 800
44
5.98
7.7
43.8
6
5.9
40.5
195
41.8
800 °C can be assigned to the sum of the residual pyridine
fragment of the ligand with CrCl3.
cocatalyst fragment [RX−] and a cationic metal fragment
[LnM+], which in combination represents the active cata-
lytic system as an ion pair represented by [LnM+][RX−]
[25, 35].
3.4 Catalytic Ethylene Oligomerization Using
Pyridine‑SNS‑Pentyl/CrCl3 Catalyst (7)
According to the preliminary tests, our focus is on 1-C6
production with catalyst (7) only and the various factors
which could affect the catalytic activity including Al/Cr
ratio, catalyst concentration (µmol), ethylene pressure (bar),
reaction temperature, were investigated using the surface
response methodology (RSM) based on the CCD design
in order to reach the maximum 1-C6 production (Table 4).
Moreover, the effect of different R groups on the 1-C6 pro-
ductivity was examined. 1-C6 as a main product and other
by-products such as 1-octane and 1-decene in the liquid
fraction were measured by gas chromatography (GC). In
Table 5, the factors and their levels are shown to achieve the
optimum conditions for the catalyst activity based on the
An ethylene trimerization reaction using the pyridine-
based catalysts activated by MMAO co-catalyst, afforded
99% 1-C6. Transition metal-based oligomerization catalysts
generally not able to produce alpha-olefin alone and must be
activated by appropriate co-catalysts. Among the important
co-catalysts, perfluoroaryl boranes, fluoroarylalanes, trityl,
aluminum alkyls and methylaluminoxane (MAO) are men-
tioned. It is generally accepted that the cocatalyst, among
other functions such as removal the water and oxygen con-
taminant and easily reduction of chromium, facilitates alkyl
abstraction from the catalyst precursor to yield an anionic
Table 4 Experimental
Run
A: Al/Cr
B: Catalyst
(µmol)
C: Ethylene pres- D: Tempera-
1-C6 productivity (g PE (g)
1-C6/g Cr h)
design layout for ethylene
oligomerization and the
obtained results
sure (bar)
ture (°C)
1
2
3
4
5
6
7
8
700
700
700
700
700
900
900
900
500
900
500
500
500
500
900
700
700
700
700
700
700
6
6
6
6
6
9
9
3
9
3
3
9
3
6
6
3
9
6
6
6
6
21
21
21
21
21
27
15
27
15
15
27
27
15
27
21
21
21
15
27
21
21
52.5
52.5
52.5
52.5
52.5
25
25
80
80
80
25
80
25
52.5
52.5
52.5
52.5
52.5
52.5
25
2090
1537
1537
1537
2070
1200
130
0.15
0.13
0.13
0.13
0.13
–
–
7.9
350
9
169
6420
817
0.25
0.021
0.1
10
11
12
13
14
15
16
17
18
19
20
21
934
311
0.14
0.11
0.14
0.1
0.12
0.08
0.07
0.1
4318
4084
1167
4020
3404
5330
14
0.1
80
5135
0.08
1 3

Switching from Ethylene Trimerization to Ethylene Polymerization by Chromium Catalysts Bearing…
Table 5 Factors and levels according to response surface model
Factor
Name
Units
Low actual
High actual
Low coded
High coded
Mean
Std. dev.
A
B
C
D
Al/Cr
–
500
3
15
25
900
9
27
80
−1
−1
−1
−1
1
1
1
1
700
6
21
138.013
2.070
4.140
Catalyst
Pressure
Temperature
µmol
Bar
°C
52.5
18.977
Response
Name
Units
Analysis
Minimum
14
Maximum
Mean
Std. dev.
Y
1-C6 productivity g 1-C6/g Cr h
Polynominal
6420
2217.81
1903.88
RSM. The variables A, B, C and D, respectively, indicate the
Al/Cr ratio, the catalyst concentration, the ethylene pressure,
and the reaction temperature.
Ln(productivity 1-C6) = +7.65 − 0.028A + 0.62B + 0.24C
+ 2.95D + 2.68AB − 0.42AC
+ 1.01AD + 0.73BC − 0.30BD
Some factors in the analysis of variance (ANOVA) table
such as coefficient of determination (R2), adjusted R2, lack
of fitness and P-value are important for selection of adequate
models. The significance of the lack of fitness for a model
indicates that the points are not well positioned around the
model and that the model cannot be used to predict the val-
ues of the dependent variables. So the data was fitted with
various models and their subsequent ANOVA showed that
1-C6 productivity was most suitably defined by quadratic
polynomial model. The final model to predict the 1-C6 pro-
ductivity of catalyst 7 is shown in Eq. (1):
− 0.55CD + 0.54A2 − 0.12B2
+ 0.56C2 − 2.21D2
(1)
As shown in ANOVA Table 6, the lack of fit numbers
for all parameters, which measure the fitness of the model,
were not significant (< 0.05) and high model F-value
(39.52) further confirmed the reliability of the models
within the considered range of process conditions. The
R2 value (R-squared) close to 1, shows the greater ability
of the model to predict a trend. Normally, a regression
model with an R2 higher than 0.90, is considered to have
Table 6 Analysis of variance
Source
Sum of squares
df
Mean square
F value
p-value
Remark
(ANOVA) according to
Prob>F
response surface quadratic
model for the response of 1-C6
productivity using catalyst (7)
Model
R2: 0.99
Adj R2: 0.96
A: Al/Cr
B: Catalyst
C: Pressure
D: Temperature
AB
45.261
0.001
0.765
0.594
17.433
11.479
1.413
1.636
4.315
0.143
2.407
0.750
0.037
0.790
12.461
0.491
0.381
0.110
45.752
14
1
1
1
1
1
1
1
1
1
1
1
1
1
1
6
2
3.233
0.001
0.765
0.594
17.433
11.479
1.413
1.636
4.315
0.143
2.407
0.750
0.037
0.790
12.461
0.082
0.190
0.027
39.526
0.019
9.351
<0.0001
0.8949
0.0223
0.0358
<0.0001
<0.0001
0.0060
0.0042
0.0003
0.2344
0.0016
0.0232
0.5268
0.0209
<0.0001
Significant
Significant
Significant
Significant
Significant
Significant
Significant
Significant
7.262
213.136
140.336
17.275
20.008
52.751
1.747
29.426
9.163
0.451
AC
AD
BC
BD
CD
A^2
B^2
C^2
Significant
Significant
9.651
152.352
Significant
Significant
D^2
Residual
Lack of fit
Pure error
Correlation total
6.933
0.0501
Not significant
4
20
1 3

M. Soheili et al.
a very high association [36]. The R2 of 0.99 is in reason-
able agreement with the “adj R2” of 0.96. The “adj R2”
value corrects the R2 value for the 1-C6 productivity and
for the number of terms in the model. A high value of the
adjusted R-squared (adj R2 = 0.9642) promoters for a high
significance of the model.
increased. The activity decreased with a further increase in
temperature due to the deactivation of the catalyst. Further
experiments at elevated temperatures such as 90, 100 and
120 °C, showed a great reduction toward 1-C6 production.
The thermal analysis results also confirmed the decomposi-
tion of the catalyst at high temperatures.
The influence and importance of the variable can be
shown by the sign and their coefficients. Therefore, the
negative sign of variable A (Al/Cr) indicates its opposite
effect on the 1-C6 productivity parameter. This is while
the positive sign of the coefficient of variable B (catalyst
concentration), C (gas pressure ethylene) and D (tempera-
ture) indicate their direct correlation with the response.
The value of the coefficient D (2.95) is greater than the
coefficients A, C and B indicating the greater importance
of temperature on the catalyst productivity in ethylene tri-
merization, while Al/Cr with a coefficient of − 0.028 has
the least effect on activity.
According to the Fig. 3d, with increasing the catalyst
concentration the 1-hexene productivity increases. Figure 3e
further confirms that temperature plays a key role in the
catalyst productivity. It should also be noted that increas-
ing the catalyst concentration at a constant temperature of
25 °C does not have an impact on productivity. Increasing
the temperature resulted in a significant increase in 1-C6
productivity (for catalyst concentration from 3 to 9 µm).
As presented in Fig. 3f, increasing the ethylene pressure
at a constant temperature of 25 °C does not show any posi-
tive influence on the productivity. Our studies have identi-
fied low temperatures which could be limiting the ethylene
reaction and hence negatively affect the yields.
The significance of each coefficient in Eq. 1 was
checked using F-test. Values of ‘‘Prob > F’’ less than 0.05
indicates that the model terms are significant. In this case
B, C, D, AB, AC, AD, BC, CD, A2, C2 and D2 are the
model significant terms. In order to improve the model,
the non-significant coefficients were excluded and the final
model was developed as Eq. (2):
Figure 4 also indicates the 1-C6 productivity (the
response) versus those of the empirical model. The predicted
data of the response from the empirical model agreed well
with the observed ones in the range of the operating vari-
ables. The model predicted the optimal values of the four
variables; Al/Cr: 841, catalyst concentration: 8.7 µmol, tem-
perature: 58.2 °C, ethylene pressure: 19.5 bar corresponding
to Ln(productivity 1-C6)=10.6 g 1-C6/g Cr h. The statistical
optimization used in this research showed the higher 1-C6
productivity comparing with other results on the ethylene
trimerization using pyridine-based catalyst [29].
Ln(productivity 1-C6) = +7.65 + 0.62B + 0.24C + 2.95D
+ 2.68AB − 0.42AC + 1.01AD
+ 0.73BC − 0.55CD + 0.54A2
+ 0.56C2 − 2.21D2
(2)
A generalized mechanism for activation by co-catalyst
and ethylene oligomerization via metallacycle intermedi-
ates is shown in Scheme 2. Among the several transition
elements in different oxidation states that show catalytic
activity in terms of ethylene oligomerization, trivalent chro-
mium appears to be the most versatile species. Trivalent
chromium has produced the best performing trimerization
and tetramerization catalysts in terms of both activity and
selectivity and accounts for over 90% of all the existing oli-
gomerization catalysts. The oxidation state of chromium(III)
was previously confirmed by using UV–Vis spectra for all
synthesized catalysts. The process begins with oxidative
(with respect to the metal) coupling of two ethylene units
to produce a metallacyclopentane complex. From here,
insertion of further ethylene into the metallacyclopentane
produces larger ring metallacycles, while breakdown of the
metallacycle at any point can produce linear alpha-olefins.
Thus the selectivity of the process is controlled by the rela-
tive stability of the different sized metallacycles, in particu-
lar their tendency to either decompose or grow via ethylene
insertion. For selective ethylene trimerization, release of
1-hexene is fast inhibiting further ring growth and thus for-
mation of higher α-olefins [9, 37].
The interaction between factors also was investigated by
using the response surface methodology (RSM). Accord-
ing to Fig. 3a, when the Al/Cr ratio is proportional to the
catalyst concentration, the productivity of 1-C6 is high
(e.g., Al/Cr = 900 and 9 µm catalyst or Al/Cr = 500 and
3 µm catalyst) which indicates the significant influence
of the co-catalyst on catalytic activity in transition metal-
catalysed ethylene oligomerization. Enhancing the Al/Cr
ratio up to 900, led to decreasing the productivity because
MMAO acts as poison at higher aluminium alkyl concen-
trations (e.g., Al/Cr = 900 and 3 µm catalyst).
As shown in Fig. 3b, increasing the Al/Cr ratio up to 900
(at a constant ethylene pressure of 15 bar), leads to enhanced
1-C6 productivity as well as transition metal activation. Fur-
thermore, at higher ethylene pressure of 27 bar (for Al/Cr
ratio between 500 and 900), the catalyst (7) showed rela-
tively high activity due to the greater catalyst stability and
higher ethylene solubility under higher ethylene pressures.
Figure 3c shows the interaction of temperature with the
Al/Cr ratio. In general, the suitable active sites were formed
at elevated temperature up to 80 °C and 1-C6 productivity
1 3

Switching from Ethylene Trimerization to Ethylene Polymerization by Chromium Catalysts Bearing…
Fig. 3 The interaction effect among various effective parameters on 1-C6 productivity
1 3

M. Soheili et al.
Yang et al. theoretically studied a mechanism for ethyl-
ene trimerization catalyzed by MeS(CH2)2N(H)(CH2)2SMe-
Cr(I) catalyst and calculated the Gibbs free energy for
different intermediates. Their results confirmed that metal-
lacycloheptane are important intermediate responsible for
ethylene trimerization into 1-hexene [38].
As shown in Table 7, the quantity of polyethylene pro-
duced in the ethylene reaction catalyzed by pyridine-based
catalyst was measured for several runs. Unexpectedly a
switching from ethylene trimerization to polymerization
occurred by exchanging of the ethylene pressure (Table 7,
Runs 8 and 10) that has not been reported yet. The above
propose mechanism (Scheme 2) can justify this phenom-
enon. For this specific reaction condition, low release of
1-C6 and further ring growth lead to the formation of poly-
ethylene according to the extended metallacycle mecha-
nism at higher ethylene pressure [9]. Moreover, for the
formation of polymer during the of ethylene oligomeriza-
tion process, a mechanism similar to the mechanism for
Phillips catalyst polymerization can be imagined. Three
Fig. 4 The predicted values by the model against the actual values
Scheme 2 Proposed metallacy-
R
R
R
cle mechanism for production
S
S
S
Cl
Me
Me
H3C
of 1-hexene, 1-Octene, higher
MMAO
LAOs and polyethylene using
Cl
+
2
Al
O
MAO
+2 MMAO
N
N
Cr
Cr
N
Cr
Cl
CH2CH2
chromium catalysts bearing
pyridine based SNS ligands
I
Cl
Al
Cl
Cl
CH3
S
S
S
R
R
R
O
H3C
(MAO)
n= 3, ...
n
n =2 (1- Octene)
n =1( 1- Hexene)
R
R
S
S
1
-
n
II
Cr
I
III
2
N
N
Cr
Cl
CH3
O
S
Cl
CH3
O
Al
S
Al
R
R
H3C
MAO
H3C
MAO
Table 7 Comparison between
the quantities of polyethylene
produced in the ethylene
reaction catalyzed by pyridine-
based catalyst (7) for several
runs
Run
A: Al/Cr
B: Catalyst C: Ethylene
D: Tempera- Weight of PE (g)
ture (°C)
Productiv-
ity (g PE/g
Cr h)
(µmol)
pressure (bar)
1
8
10
19
700
900
900
700
6
3
3
6
21
27
15
27
52.5
80
0.15
7.90
0.02
0.10
961
101282a
262
80
52.5
614
aMaximum PE productivity using catalyst (7) obtained by change the ethylene pressure
1 3

Switching from Ethylene Trimerization to Ethylene Polymerization by Chromium Catalysts Bearing…
kinds of typical initiation mechanisms have been proposed
in the literatures for the Phillips chromium catalyst: Cossee
mechanism, carbine mechanism, and metallacycle mecha-
nism. However, these mechanisms do not definitely justify
the polymer formation during the Phillips polymerization
process [39, 40].
by-product in Run 10 had significantly lower Tm (134 °C)
and melting enthalpy (ΔHm) values may be due to the deg-
radation and decreasing the size of crystals. The polymer
sample (Run 10) contains chains with a wide distribution
of chain lengths and shows the low crystallinity (%). Oli-
gomers with low molecular weight may be the main reason
of the lower Xc (%) observed for the obtained polyethylene
(Table S1, Run 10).
It is, highly desirable to find the reasons that are respon-
sible for the formation of polymeric materials. So DSC was
used to study the thermal properties of the resulting poly-
ethylenes (Table 7, Runs 8 and 10, Fig. 6). The samples
were heated from 25 to 180 °C, held for 1 min to erase ther-
mal history effects and cooled to 25 °C and then heated to
180 °C again for the second scan. The temperature scan was
performed with a heating and cooling rate of 10 °C/min.
As shown in Fig. S2 the DSC curve of the polymer (Run 8)
show an exothermic peak at 135 °C (ΔH = 170 J/g) which
is a typical DSC endotherms of polyethylene with differ-
ent molecular weight. Also, no difference was observed in
the cooling and second heating cycles (Table S1). The first
heating scan provides information about the thermal his-
tory of the sample (processing or aging). The cooling and
second heating cycles are then performed at known thermal
history. The similar melting point of the sample (Run 8)
in second heating cycle shows that the processing condi-
tions did not have any effect on the thermal properties of
the polyethylene. But the polyethylene sample produced as
Selected ethylene trimerization results of different pyr-
idine-based catalysts were also investigated. According to
Fig. 5, it is clear that the pyridine-SNS-alkyl-CrCl3 systems
are capable to production of 1-C6 with more than 90% selec-
tivity, which highlights them among other available catalytic
systems. The substituents on the sulphur atom have great
influences on the catalytic performance. Increasing the size
of substituents rises the steric bulk around the S (steric hin-
drance). So further insertion of ethylene into the chromacy-
clopentane ring significantly restricts.
3.5 Ethylene Trimerization Using Amine‑SNS‑Alkyl/
CrCl3 Catalysts
Amine-SNS-Alkyl/CrCl3 catalysts similar to pyridine-
based catalysts have also been effective for selective eth-
ylene trimerization. In addition, intensely lower cost of the
SNS ligands, small amount of expensive MMAO, improved
activity (141,000 g/g Cr h) (Table S2) retaining high 1-C6
selectivity make them superior in terms of potential tech-
nical application. Similar to the trend observed for the
pyridine-based complexes, increasing the steric hindrance
on side chain leads to lower catalytic activities, while the
opposite effect is observed as the steric hindrance decreases.
In addition to R-group steric difference as a determining
factor of catalytic activity in the prepared complexes, the
amine N-donor containing complexes (14–16) displayed a
significantly higher catalytic activity which is likely due to
the additional flexibility of the amine backbone. A flexible
two carbon spacer between a simple amine N donor and the
S-donor atoms generated more effective catalysts than a rigid
5000
4000
3000
2000
1000
0
Octyl
Cyclohexyl Cyclopentyl
Pentyl
Fig. 5 Comparison of various pyridine-based catalysts with different
substituent on sulphur atom (temperature: 52.5 °C, catalyst concentra-
tion: 6 µmol, ethylene pressure: 27 bar, Al/Cr: 500, solvent: toluene)
Fig. 6 Comparison of pyridine-
SNS-alkyl/CrCl3 and amine-
SNS-alkyl/CrCl3 catalysts in
H
ethylene trimerization
N
N
Cl
Cl
S
S
R
Cr
Cl
R
Cr
Cl
S
S
Cl
R
R
Cl
Amine-SNS/Catalyst
Selectivity: > 99%
Productivity:
141*10
Pyridine-SNS/Catalyst
Selectivity: > 99%
Productivity: 6420 (g 1-C6/gr Cr. h)
3
.
6
1 3

M. Soheili et al.
Fig. 7 TGA diagram of amine-
SNS-pentyl/CrCl3 catalyst
under nitrogen atmosphere
and heating rate of 10 °C/min
from ambient temperature up to
800 °C
pyridine N-donor linked to the S-donors via a methylene
spacer. Direct comparison of the two structurally catalysts
in Fig. 6 further confirmed the importance of the flexibility
factor influence as a key factor of catalytic activity and more
ethylene insertion as well as the metallacycloheptane ring
formation. The thermogravimetric analysis (TGA) curves
in nitrogen showed that the amine-based complexes display
lower thermal stability compared to pyridine-based com-
plexes (Fig. 7). The decomposition temperatures (Tdec) for
5% weight loss were observed in the range of 162 °C for
amine-SNS-pentyl/CrCl3. The initial weight loss (~3.23%)
is related to the residual organic solvent and water. The sec-
ond and the third weight loss (16.59%) in the temperature
range of 120–220 °C, presumably due to loss of the NHC4H8
fragment of the SNS ligand. The fourth weight loss (49%)
in the temperature range of 260–420 °C is related to the
dissociation of the SC5H11 of the ligand. The residual mass
(31%) above 800 °C can be assigned to CrCl3. The presence
of pyridine ring increases the thermal stability and degra-
dation temperature of pyridine-base catalysts compared to
amine-based catalysts.
and experiments. A switching from ethylene trimerization
to polymerization of the catalyst system can be occurred
utilizing exchanging of the reaction conditions. The polymer
produced as by-product of catalytic ethylene oligomeriza-
tion using pyridine-based catalyst contains chains with a
wide distribution of chain lengths and shows the low crys-
tallinity. Increasing the steric hindrance on side chain leads
to lower catalytic activities. Chromium catalysts based on
amine ligands show higher 1-C6 productivity. Comparison
of the two structurally catalysts confirmed the importance
of the flexibility factor influence as a key factor of catalytic
activity and more ethylene insertion as well as the metalla-
cycloheptane ring formation. The presence of pyridine ring
increases the thermal stability and degradation temperature
of pyridine-base catalysts compared to amine-based cata-
lysts. The selectivity of these systems makes them very suit-
able for use in tandem with olefin polymerization catalysts
to produce LLDPE from ethylene alone; results in this area
will be reported shortly.
Acknowledgements The authors would like to thank Mr. Abbas Biglari
for his sincere efforts to NMR analysis of the samples.
Funding The authors would like to thank the Iran National Science
Foundation (INSF) for the financial support of this study (Grant Num-
ber 95824926) and from the Institute for Advanced Studies in Basic
Sciences (IASBS) for its financial and spiritual supports.
4 Conclusions
Two types of chromium catalysts bearing pyridine and
amine based SNS ligands were synthesized. Response sur-
face methodology (RSM) based on a three level-four vari-
able central composite design, was employed to evaluate the
effect of the most important parameters on 1-C6 productivity.
The temperature was found as major influential factor on
the 1-C6 productivity. The experimental results indicated
that there is reasonably good agreement between predictions
Compliance with Ethical Standards
Conflict of interest There is no conflict of interest for each contribut-
ing authors.
1 3

Switching from Ethylene Trimerization to Ethylene Polymerization by Chromium Catalysts Bearing…
20. Albahily K, Shaikh Y, Ahmed Z et al (2011) Isolation of a self-
References
activating ethylene trimerization catalyst of a Cr-SNS system.
Organometallics 30(15):4159–4164
1. Liu S, Zhang Y, Han Y et al (2017) Selective ethylene oligomer-
ization with chromium-based metal–organic framework MIL-
100 evacuated under different temperatures. Organometallics
36(3):632–638
21. Mohamadnia Z, Ahmadi E, Haghighi MN et al (2011) Syn-
thesis and optimization of ethylene trimerization using [bis-
(2-dodecylsulfanyl-ethyl)-amine] CrCl3 catalyst. Catal Lett
141(3):474–480
2. Britovsek GJ, Malinowski R, McGuinness DS et al (2015) Eth-
ylene oligomerization beyond Schulz–Flory distributions. ACS
Catal 5(11):6922–6925
22. Yang Y, Liu Z, Liu B et al (2013) Selective ethylene tri-/
tetramerization by in situ-formed chromium catalysts sta-
bilized by N,P-based ancillary ligand systems. ACS Catal
3(10):2353–2361
3. Rozhko E, Bavykina A, Osadchii D et al (2017) Covalent
organic frameworks as supports for a molecular Ni based eth-
ylene oligomerization catalyst for the synthesis of long chain
olefins. J Catal 345:270–280
23. Yang Y, Gurnham J, Liu B et al (2014) Selective ethylene
oligomerization with chromium complexes bearing pyridine–
phosphine ligands: influence of ligand structure on catalytic
behavior. Organometallics 33(20):5749–5757
4. Vilms AI, Babenko IA, Bezborodov VA et al (2017) Ethylene
oligomerization over chromium(III) complexes with pyrrole
derivatives. Petrol Chem 57(3):244–250
24. Zhang J, Braunstein P, Hor TA (2008) Highly selective chro-
mium (III) ethylene trimerization catalysts with [NON]
and [NSN] heteroscorpionate ligands. Organometallics
27(17):4277–4279
5. Mohamadnia Z, Ahmadi E, Haghighi MN et al (2015) Prepara-
tion of LLDPE through tandem ethylene polymerization using
chromium and zirconium catalysts. Iran Polym J 24(8):621–628
6. Fallahi M, Ahmadi E, Ramazani A et al (2017) Trimerization
of ethylene catalyzed by Cr-based catalyst immobilized on the
supported ionic liquid phase. J Organomet Chem 848:149–158
7. Xu Z, Chada JP, Xu L, et al. (2018) Ethylene dimerization and
oligomerization to 1-butene and higher olefins with chromium-
promoted cobalt on carbon catalyst. ACS Catal 8(3):2488–2497
8. Lappin GR, Nemec LH, Sauer JD et al (2000) Olefins, higher.
Kirk-Othmer encyclopedia of chemical technology. Wiley, New
York
25. Liu S, Pattacini R, Braunstein P (2011) Reactions between an
ethylene oligomerization chromium(III) precatalyst and alu-
minum-based activators: alkyl and cationic complexes with a
tridentate NPN ligand. Organometallics 30(13):3549–3558
26. Simler T, Braunstein P, Danopoulos AA (2016) Chromium(II)
pincer complexes with dearomatized PNP and PNC ligands: a
comparative study of their catalytic ethylene oligomerization
activity. Organometallics 35(24):4044–4049
27. Jie S, Pattacini R, Rogez G et al (2009) Synthesis, struc-
ture, magnetic and catalytic properties of new dinuclear
chromium(III) complexes with oxazoline alcoholate ligands.
Dalton Trans 1:97–105
9. McGuinness DS (2010) Olefin oligomerization via metallacy-
cles: dimerization, trimerization, tetramerization, and beyond.
Chem Rev 111(3):2321–2341
28. Ai P, Danopoulos AA, Braunstein P (2015) N-Phosphanyl and
N,N′-diphosphanyl-substituted N-heterocyclic carbene chro-
mium complexes: synthesis, structures, and catalytic ethylene
oligomerization. Organometallics 34(16):4109–4116
29. Temple CN, Gambarotta S, Korobkov I et al (2007) New insight
into the role of the metal oxidation state in controlling the selec-
tivity of the Cr-(SNS) ethylene trimerization catalyst. Organo-
metallics 26(18):4598–4603
10. Bergamo AL, Da Cas HK, Rambo RS et al (2016) Chromium
complexes bearing pyrazolyl-imine-phenoxy/pyrrolide ligands:
synthesis, characterization, and use in ethylene oligomerization.
Catal Commun 86:77–81
11. Kuhlmann S, Paetz C, Hägele C et al (2009) Chromium cata-
lyzed tetramerization of ethylene in a continuous tube reactor—
proof of concept and kinetic aspects. J Catal 262(1):83–91
12. Reagan WK, Freeman JW, Conroy BK et al (1994) Phillips
Petroleum Company, Bartlesville
30. McGuinness DS, Wasserscheid P, Keim W et al. (2003) First Cr
(III)—SNS complexes and their use as highly efficient catalysts
for the trimerization of ethylene to 1-hexene. J Am Chem Soc
125(18):5272–5273
13. Tanaka E, Urata H, Oshiki T et al (2000) Process for producing
alpha-olefin oligomer compositions. EP 0611743 B2. Mitsubishi
Chemical Corporation
31. Komarsamy L, Bala MD, Friedrich HB et al (2011)
2,6-Bis(tosyl-oxymeth-yl)pyridine. Acta Crystallogr E 67:302
32. Soobramoney L, Bala MD, Friedrich HB (2014) Coordination
chemistry of Co complexes containing tridentate SNS ligands
and their application as catalysts for the oxidation of n-octane.
Dalton Trans 43(42):15968–15978
14. Carter A, Cohen SA, Cooley NA et al (2002) High activity
ethylene trimerisation catalysts based on diphosphine ligands.
Chem Commun 8:858–859
15. Bollmann A, Blann K, Dixon JT et al (2004) Ethylene tetrameri-
zation: a new route to produce 1-octene in exceptionally high
selectivities. J Am Chem Soc 126(45):14712–14713
16. Hao Z, Xu B, Gao W (2015) Chromium complexes with N, N,
N-tridentate quinolinyl anilido-imine ligand: synthesis, charac-
terization, and catalysis in ethylene polymerization. Organome-
tallics 34(12):2783–2790
33. Dixon JT, Green MJ, Hess FM et al (2004) Advances in selec-
tive ethylene trimerisation—a critical overview. J Organomet
Chem 689(23):3641–3668
34. Nuñez-Dallos N, Lopez-Barbosa N, Muñoz-Castro A et al
(2017) A new copper (I) coordination polymer from 2, 6-bis
(1H-benzotriazol-1-ylmethyl)pyridine: Synthesis, characteriza-
tion, and use as additive in transparent submicron UV filters. J
Coord Chem 70(19):3363–3378
17. Suttil JA, Wasserscheid P, McGuinness DS et al (2014) A survey
of pendant donor-functionalised (N, O) phosphine ligands for
Cr-catalysed ethylene tri-and tetramerisation. Catal Sci Technol
4(8):2574–2588
35. Janse van Rensburg W, van den Berg JA, Steynberg PJ (2007)
Role of MAO in chromium-catalyzed ethylene tri-and tetrameri-
zation: a DFT study. Organometallics 26(4):1000–1013
36. García-Muñoz RA, Morales V, Linares M et al (2014) Influence
of the structural and textural properties of ordered mesoporous
materials and hierarchical zeolitic supports on the controlled
release of methylprednisolone hemisuccinate. J Mater Chem B
2(45):7996–8004
18. Alzamly A, Gambarotta S, Korobkov I (2013) Synthesis, struc-
tures, and ethylene oligomerization activity of bis (phospha-
nylamine) pyridine chromium/aluminate complexes. Organo-
metallics 32(23):7107–7115
19. Ahmadi E, Mohamadnia Z, Haghighi MN (2011) High produc-
tive ethylene trimerization catalyst based on CrCl
Catal Lett 141(8):1191
3/SNS ligands.
1 3

M. Soheili et al.
37. Elowe PR, McCann C, Pringle PG et al (2006) Nitrogen-linked
diphosphine ligands with ethers attached to nitrogen for chro-
mium-catalyzed ethylene tri-and tetramerizations. Organometal-
lics 25(22):5255–5260
39. Gandon V, Agenet N, Vollhardt KP et al (2006) Cobalt-mediated
cyclic and linear 2:1 cooligomerization of alkynes with alkenes:
a DFT study. J Am Chem Soc 128(26):8509–8520
40. Ghiotti G, Garrone E, Zecchina A (1991) An infrared study of CO/
C2H4 coadsorption and reaction on silica-supported Cr(II) ions. J
Mol Catal 65(1–2):73–83
38. Yang Y, Liu Z, Zhong L et al (2011) Spin surface crossing
between chromium (I)/sextet and chromium (III)/quartet with-
out deprotonation in SNS-Cr mediated ethylene trimerization.
Organometallics 30(19):5297–5302
Affiliations
Majid Soheili1 · Zahra Mohamadnia1 · Babak Karimi1
1
* Zahra Mohamadnia
Department of Chemistry, Institute for Advanced Studies
z.mohamadnia@iasbs.ac.ir
in Basic Sciences (IASBS), No. 444, Prof. Yousef Sobouti
Blvd., P. O. Box 45195-1159, Zanjan 45137-66731, Iran
1 3