A series of novel 2,6-bis(imino)pyridyl iron catalysts: Synthesis, characterization and ethylene oligomerization

Article abstract of DOI:10.1016/j.molcata.2004.11.028

A series of novel 2,6-bis(imino)pyridyl iron complexes {2,6-(2-X-4-Y-5- ZC₆H₂N=CCH₃)₂C₅H ₃N}FeCl₂ (X = Cl, Y = CH₃, Z = H (2); X = Br, Y = CH₃, Z = H (3); X = F, Y = H, Z = CH₃ (4); X = Cl, Y = H, Z = CH₃ (5); X = Cl, Y = F (7)) have been synthesized and characterized with elemental analysis and IR. These iron coordinative complexes, activated with methylaluminoxane (MAO), lead to highly active ethylene oligomerization (>10⁷ g/mol Fe h) and the products are mostly linear α-olefins (>90%). The catalytic activities and product properties depend on the substituents on aryl rings and the reaction conditions. As reaction temperature increases, the catalytic activities decrease rapidly and more low-molar-mass products are produced. The product distributions are almost independent of the Al/Fe molar ratio, but the catalytic activities change in different trends when the ortho-substituents on the aryl rings are different. The other three complexes have also been synthesized for comparison to investigate the steric hindrance and electronic effect on the properties of complexes. The complex with adaptable steric hindrance and electronic properties exhibits the highest catalytic activities.

Full text of DOI:10.1016/j.molcata.2004.11.028

Journal of Molecular Catalysis A: Chemical 230 (2005) 1–8
A series of novel 2,6-bis(imino)pyridyl iron catalysts: synthesis,
characterization and ethylene oligomerization
Zhicheng Zhanga,b, Shangtao Chena,b, Xiaofan Zhanga,b, Huayi Lia,b,
Yucai Kea, Yingying Lua, Youliang Hua,∗
a
Joint Laboratory of Polymer Science and Materials, Key Laboratory of Engineering Plastics, Institute of Chemistry,
Chinese Academy of Sciences, Beijing 100080, PR China
Graduate School of the Chinese Academy of Sciences, Beijing 100039, PR China
b
Received 24 August 2004; received in revised form 18 November 2004; accepted 23 November 2004
Available online 15 January 2005
Abstract
A series of novel 2,6-bis(imino)pyridyl iron complexes {2,6-(2-X-4-Y-5-ZC6H2N=CCH3)2C5H3N}FeCl2 (X = Cl, Y = CH3, Z = H (2);
X = Br, Y = CH3, Z = H (3); X = F, Y = H, Z = CH3 (4); X = Cl, Y = H, Z = CH3 (5); X = Cl, Y = F (7)) have been synthesized and characterized
with elemental analysis and IR. These iron coordinative complexes, activated with methylaluminoxane (MAO), lead to highly active ethylene
oligomerization (>107 g/mol Fe h) and the products are mostly linear ␣-olefins (>90%). The catalytic activities and product properties depend
on the substituents on aryl rings and the reaction conditions. As reaction temperature increases, the catalytic activities decrease rapidly and
more low-molar-mass products are produced. The product distributions are almost independent of the Al/Fe molar ratio, but the catalytic
activities change in different trends when the ortho-substituents on the aryl rings are different. The other three complexes have also been
synthesized for comparison to investigate the steric hindrance and electronic effect on the properties of complexes. The complex with adaptable
steric hindrance and electronic properties exhibits the highest catalytic activities.
© 2004 Elsevier B.V. All rights reserved.
Keywords: 2,6-Bis(imino)pyridyl iron complex; Oligomerization; ␣-Olefin; Distribution
1. Introduction
which are activated by MAO, are effective catalysts for the
oligomerization of ethylene to oligomers with high catalytic
activity (>107 g/mol Fe h) and high selectivity for linear
␣-olefins (>95%). The size and regio-chemistry of the sub-
stituents in the imino-aryl groups are of crucial importance
in controlling the oligomerization of ethylene. But when the
ortho-substituents on ligands are alkyls, the carbon number
distributions of oligomers obtained are very wide [1–9].
Bluhm et al. [10] reported on some complexes without ortho-
substituents on aryl rings. The distribution of oligomers
obtained is much narrower (mainly C4–C10), but both
catalytic activities (TOF < 2.5 × 104/h) and selectivity for
␣-olefins (<88%) are low because of small steric hindrance.
Compared to the little difference in electronic effect and
large steric hindrance originated from changing the alkyl
groups, the electronic effect and steric bulk from varying
Oligomerization of ethylene presents one of the major
processes for the production of linear ␣-olefins in the range
C4–C20. Such linear oligomers are extensively used for
preparation of detergents, plasticizers and as co-monomers
for the preparation of linear low-density polyethylene
(LLDPE). Since Ziegler’s original work on AlR3 catalysis
of ethylene oligomerization, there has been considerable
sustained interest in developing new oligomerization cata-
lysts. In 1998, Small and Brookhart [1–3] and Gibson and
co-workers [4,5] independently discovered that mono-alkyl-
substituted [2,6-bis(imino)pyridyl]iron and -cobalt dihalides,
∗
Corresponding author. Tel.: +86 10 6256 2815; fax: +86 10 6255 4061.
E-mail address: huyl@iccas.ac.cn (Y. Hu).
1381-1169/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2004.11.028

2
Z. Zhang et al. / Journal of Molecular Catalysis A: Chemical 230 (2005) 1–8
halogens are much more significant due to the big differ-
ences in their electronegativities and covalent radius (A).
grade ethylene was obtained from Yanshan Petrochemical
Company, Sinopec, China. Methylaluminoxane (MAO) so-
lution in toluene (1.4 mol/L) was purchased from Albemarle
Corp. Toluene, THF and ether (Et2O) were distilled from
solidum/benophenone and degassed before use. All other
chemicals were obtained commercially and used without fur-
ther purification.
Qian and co-workers [11] reported that silica–alumina cat-
alyst support was an effective catalyst for the preparation of
2,6-bis(imino)pyrdyl ligand containing halogen substituents
on aryl rings. Recently, we found that MAO supported on
SiO2 (MAO/SiO2) was another effective catalyst for the syn-
thesis of ligands of this kind. MAO/SiO2 was prepared in the
following procedure: in a 50 mL dried Schlenk flask with a
stirring bar, 4 g dried SiO2 was purged with dry nitrogen two
to three times and then 16 mL MAO were injected. The reac-
tion system was magnetically stirred at 50 ◦C for 12 h. The
precipitate was filtered and washed with toluene for three
times and then dried under reduced pressure to give a white
solid. Finally, the resulting product obtained was preserved
in nitrogen atmosphere.
˚
When there are only halogen substituents at aryl rings of
ligands, detailed work has been carried out by Qian and co-
workers[11,12]. Theirresultsshowthatstericbulkofhalogen
substituents and the positive charge of central metal are two
important factors that influence the activities of complexes
greatly. The results obtained by Gibson and co-workers [5]
show that the introduction of alkyl group at the para- or
meta-position of aryl rings could improve catalytic activities
markedly. The introduction of bromine in the para-position
of aryl rings has also been reported to improve the catalytic
activity in ethylene polymerization [13]. Combining these
two kinds of substituents in the same catalyst are expected to
give a series of novel oligomerization catalysts. Recently, a
complex with a fluoro-substituent at the ortho-position and
a methyl substituent at the para-position of aryl rings was
reported by our group [14], which exhibited high activity
and selectivity for ␣-olefins. In this paper, other five novel
bis(imino)pyridyl iron complexes bearing halogen and alkyl
substituents are synthesized and used for ethylene oligomer-
ization. Their structures and synthetic route are presented
in Scheme 1. Meanwhile, the relationship between the struc-
tures of the complexes and the catalytic activities and product
properties are discussed in detail.
2.2. Synthesis and characterization of ligands and
complexes
The structures of ligands and complexes 1–5 are shown in
Scheme 1.
All the ligands were synthesized by the method reported in
[11], and the silica–aluminum catalyst support was displaced
with MAO/SiO2.
2. Experimental
2.1. Materials
2,6-Diacetylpyridine and all the anilines with different
substituents were purchased from Acros. Polymerization-
Ligand 1 (L1, C23H17F2N3): a solution of 2,6-
diacetylpyridine (0.49 g, 3 mmol), 2-fluoro-4-methylaniline
Scheme 1.

Z. Zhang et al. / Journal of Molecular Catalysis A: Chemical 230 (2005) 1–8
3
˚
(0.84 g, 7 mmol), MAO/SiO2 (0.3 g), and molecular sieve 4 A
(2.0 g) in toluene (10 mL) was stirred at 30–40 ◦C for 24 h.
Then the reaction mixture was filtered, and the molecular
sieve was washed with toluene several times. The toluene of
the combined filtrates was removed under vacuum. Anhy-
drous methanol was added to the residue. A yellow solid was
filtered off to give ligand 1 in 55% yield. 1H NMR (CDCl3)
δ: 8.26 (d, 2H, Py-m-H), 8.05 (t, 1H, Py-p-H), 6.88 (d, 2H,
aryl), 6.85 (d, 2H, aryl), 6.80 (s, 2H, aryl), 2.82 (d, 6H, aryl-
CH3), 2.29 (s, 6H, N=C CH3). EI mass spectrum: m/z 377
[M+]. Elemental analysis: calcd. (%): C, 73.19; H, 5.61; N,
11.13; found (%): C, 73.37; H, 5.66; N, 10.89. IR (KBr):
1645 (νC=N), 1570, 1499, 1423, 1364, 1324, 1270, 1242,
2.81 (s, 6H, N=C CH3), 2.25 (s, 6H, aryl-CH3). EI mass
spectrum: m/z 409 [M+]. Elemental analysis (C23H17Cl2N3):
calcd. (%): C, 67.32; H, 5.16; N, 10.24; found (%): C, 66.92;
H, 5.20; N, 9.95. IR (KBr): 1643 (νC=N), 1571, 1475, 1453,
1365, 1320, 1237, 1121, 1103, 1080, 995, 873, 821, 764 and
738 cm−1
Ligand
.
6
(L6, C21H17Cl2N3): 2,6-diacetylpyridine
(0.49 g, 3 mmol), 2-chloroaniline (0.89 g, 7 mmol),
MAO/SiO2 (0.15 g), and molecular sieve 4 A (1.0 g)
˚
were used. L6 was obtained as a yellow crystal in 56%
1
yield. H NMR (CDCl3) δ: 8.44 (d, 2H, PyrH3), 7.94 (t,
1H, PyrH4), 7.43 (dd, 2H, ArH), 7.28 (pseudo t, 2H, ArH),
7.07 (pseudo t, 2H, ArH), 6.86 (dd, 2H, ArH), 2.38 (s, 6H,
N=C CH3). EI-MS: m/z 381 [M+]. Elemental analysis
(C21H17Cl2N3) (382.3): calcd. (%): C 65.97, H 4.48, N
10.99; found (%): C 65.64, H 4.69, N 10.89. IR (KBr): 1635
(νC=N), 1582, 1465, 1436, 1362, 1252, 1222, 1120, 1054,
1212, 1119, 1080, 940, 872, 826, 778 and 724 cm−1
Ligand (L2, C23H17Cl2N3): 2,6-diacetylpyridine
.
2
(0.49 g, 3 mmol), 2-chloro-4-methylaniline (0.92 g, 7 mmol),
˚
MAO/SiO2 (0.3 g) and molecular sieve 4 A (2.0 g) were used.
L2 was obtained as a yellow powder in 62% yield. 1H NMR
(CDCl3) δ: 8.23 (d, 2H, PyrH3), 7.99 (t, 1H, PyrH4), 7.07 (s,
2H, ArH), 6.88 (d, 2H, ArH), 6.69 (d, 2H, ArH), 2.80 (s, 6H,
N=C CH3), 2.22 (s, 6H, aryl-CH3). EI mass spectrum: m/z
409 [M+]. Elemental analysis (C23H17Cl2N3): calcd. (%): C,
67.32; H, 5.16; N, 10.24; found (%): C, 67.21; H, 5.21; N,
9.97. IR (KBr): 1643 (νC=N), 1574, 1486, 1452, 1419, 1362,
1321, 1257, 1225, 1120, 1100, 1081, 970, 880, 824, 772 and
1032, 816, 772, 756, 743, 735 and 687 cm−1
.
Ligand 7 (L7, C21H15Cl2F2N3): 2,6-diacetylpyridine
(0.49 g, 3 mmol), 2-chloro-4-fluoroaniline (1.02 g, 7 mmol),
˚
MAO/SiO2 (0.3 g), andmolecularsieve4 A(2.0 g)wereused.
L7 was obtained as a yellow powder in 61% yield. 1H NMR
(CDCl3) δ: 8.22 (d, 2H, PyrH3), 7.94 (t, 1H, PyrH4), 7.25
(s, 2H, ArH), 7.00 (d, 2H, ArH), 6.72 (d, 2H, ArH), 2.79 (s,
6H, N=C CH3). EI-MS: m/z 417 [M+]. Elemental analysis
(C21H15Cl2F2N3) (418.3): calcd. (%): C 60.30, H 3.61, N
10.05; found (%): C 60.54, H 3.68, N 10.19. IR (KBr): 1645
(νC=N), 1575, 1482, 1426, 1368, 1322, 1254, 1191, 1124,
739 cm−1
.
Ligand 3 (L3, C23H17Br2N3): 2,6-diacetylpyridine
(0.49 g, 3 mmol), 2-bromo-4-methylaniline (1.31 g, 7 mmol),
MAO/SiO2 (0.3 g) and molecular sieve 4 A (2.0 g) were used.
1102, 1075, 972, 861, 822, 769 and 741 cm−1
.
˚
L3 was obtained as a yellow crystal in 63% yield. 1H NMR
(CDCl3) δ: 8.23 (d, 2H, PyrH3), 7.99 (t, 1H, PyrH4), 7.24 (s,
2H, ArH), 6.92 (d, 2H, ArH), 6.70 (d, 2H, ArH), 2.80 (s, 6H,
N=C CH3), 2.22 (s, 6H, aryl-CH3). EI mass spectrum: m/z
497 [M+]. Elemental analysis (C23H17Br2N3): calcd. (%): C,
55.33; H, 4.24; N, 8.42; found (%): C, 55.33; H, 4.31; N,
8.27. IR (KBr): 1638 (νC=N), 1570, 1482, 1450, 1422, 1365,
1321, 1256, 1223, 1121, 1100, 1075, 964, 885, 827, 768 and
Ligand 8 (L8, C21H15F4N3): 2,6-diacetylpyridine (0.49 g,
3 mmol), 2,4-difluoroaniline (0.91 g, 7 mmol), MAO/SiO2
˚
(0.15 g), and molecular sieve 4 A (3.0 g) were used. L8 was
obtained as a yellow powder in 65% yield. 1H NMR (CDCl3)
δ:8.38(d, 2H, PyrH3), 7.90(t, 1H, PyrH4), 6.93(m, 6H, ArH),
2.41 (s, 6H, N=C CH3). EI-MS: m/z 385 [M+]. Elemental
analysis (C21H15F4N3) (385.4): calcd. (%): C 65.45, H 3.92,
N 10.90; found (%): C 65.76, H 4.13, N 10.79. IR (KBr):
1639 (νC=N), 1596, 1575, 1497, 1426, 1366, 1322, 1261,
741 cm−1
Ligand
.
4
(L4, C23H17F2N3): 2,6-diacetylpyridine
1199, 1140, 1094, 1080, 960, 848, 825, 778 and 726 cm−1
.
(0.49 g, 3 mmol), 2-fluoro-5-methylaniline (0.84 g, 7 mmol),
MAO/SiO2 (0.3 g) and molecular sieve 4 A (2.0 g) were used.
All the complexes were synthesized as reported in liter-
ature [11]. Ligand was added to a solution of FeCl2·4H2O
in THF at room temperature with rapid stirring. After be-
ing stirred for 12 h, the reaction mixture was filtered and the
product was washed with Et2O and dried in a vacuum.
Complex 1 (C1, {2,6-(2-F-4-CH3C6H3N=CCH3)2C5
H3N}FeCl2·H2O): ligand L1 (190 mg, 0.50 mmol),
FeCl2·4H2O (110 mg, 0.55 mmol) and THF (12 mL)
were used. The desired product (210 mg) was obtained
as a purple powder in 83% yield. Elemental analysis
(C23H17N3F2FeCl2·H2O): calcd. (%): C 52.90, H 4.44, N
8.05; found (%): C 53.41, H 4.38, N 7.57. IR (KBr): 1623
(νC=N), 1584, 1502, 1424, 1377, 1323, 1267, 1223, 1113,
˚
L4 was obtained as a light yellow powder in 64% yield. 1H
NMR (CDCl3) δ: 8.23 (d, 2H, PyrH3), 7.93 (t, 1H, PyrH4),
7.02 (d, 2H, ArH), 6.85 (s, 2H, ArH), 6.73 (d, 2H, ArH),
2.80 (s, 6H, N=C CH3), 2.23 (s, 6H, aryl-CH3). EI mass
spectrum: m/z 377 [M+]. Elemental analysis (C23H17F2N3):
calcd. (%): C, 73.19; H, 5.61; N, 11.13; found (%): C, 72.92;
H, 5.63; N, 10.84. IR (KBr): 1642 (νC=N), 1572, 1501, 1452,
1424, 1366, 1322, 1258, 1209, 1120, 1098, 1077, 980, 876,
820, 777 and 742 cm−1
Ligand (L5, C23H17Cl2N3): 2,6-diacetylpyridine
.
5
(0.49 g, 3 mmol), 2-chloro-5-methylaniline (0.92 g, 7 mmol),
MAO/SiO2 (0.3 g), andmolecularsieve4 A(2.0 g)wereused.
1031, 943, 842, 808 and 729 cm−1
.
˚
1
L5 was obtained as a light yellow crystal in 54% yield. H
Complex 2 (C2, {2,6-(2-Cl-4-CH3C6H3N=CCH3)2C5
H3N}FeCl2·THF): ligand L2 (410 mg, 1.0 mmol),
FeCl2·4H2O (219 mg, 1.1 mmol) and THF (12 mL) were
NMR (CDCl3) δ: 8.23 (d, 2H, PyrH3), 8.00 (t, 1H, PyrH4),
7.27 (d, 2H, ArH), 6.63 (s, 2H, ArH), 6.55 (d, 2H, ArH),

4
Z. Zhang et al. / Journal of Molecular Catalysis A: Chemical 230 (2005) 1–8
used. C2 (462 mg) was obtained as a light blue powder in
86% yield. Elemental analysis (C23H17N3Cl2FeCl2·THF):
calcd. (%): C 53.32, H 4.64, N 6.91; found (%): C 53.43,
H 4.78, N 6.72. IR (KBr): 1624 (νC=N), 1588, 1490, 1428,
1374, 1320, 1270, 1233, 1151, 1060, 889, 840, 814 and
2.3. Oligomerization of ethylene at atmospheric
pressure
A 250 mL dried three-necked flask with a stirring bar was
purged with dry nitrogen two to three times and then ethy-
lene once. Then 50 mL of toluene and a prescribed amount
of MAO were injected into it and the mixture was magneti-
cally stirred at different temperatures. The ethylene monomer
was continuously fed in and its pressure was maintained
at 0.1 MPa by an electromagnetic valve and 2 min later,
oligomerization was started by adding 1 ␮mol catalyst sus-
pension in toluene. The reaction was terminated by the addi-
tion of 3 mL water after 30 min and catalytic activities were
calculated by pressure change in the buffer storage. The prod-
uct was stocked in the sealed reactor for more than 2 h at 0 ◦C
before subjected to GC–MS analysis, which could prevent the
evaporation of C4 and C6.
716 cm−1
Complex
.
3
(C3, {2,6-(2-Br-4-CH3C6H3N=CCH3)2
C5H3N}FeCl2·THF): ligand L3 (499 mg, 1.0 mmol),
FeCl2·4H2O (219 mg, 1.1 mmol) and THF (12 mL) were
used. C3 (564 mg) was obtained as a light blue powder in
90% yield. Elemental analysis (C23H17N3Br2FeCl2·THF):
calcd. (%): C 46.52, H 4.05, N 6.03; found (%): C 46.33,
H 4.22, N 5.95. IR (KBr): 1623 (νC=N), 1588, 1485, 1427,
1373, 1321, 1270, 1231, 1152, 1049, 888, 842, 813 and
714 cm−1
Complex
.
4
(C4, {2,6-(2-F-5-CH3C6H3N=CCH3)2
C5H3N}FeCl2·THF): ligand L4 (378 mg, 1.0 mmol),
FeCl2·4H2O (219 mg, 1.1 mmol) and THF (12 mL) were
used. C4 (490 mg) was obtained as a purple powder in 90%
yield. Elemental analysis (C23H17N3F2FeCl2·THF): calcd.
(%): C 52.90, H 4.44, N 8.05; found (%): C 53.11, H 4.29,
N 7.89. IR (KBr): 1628 (νC=N), 1593, 1501, 1377, 1263,
2.4. Measurements
1H NMR spectra were recorded on a Bruker DMX
(300 MHz) spectrometer. Elemental analyses were obtained
using Carlo Erba 1106 and ST02 apparatus. IR spectra were
recorded using a Perkin-Elmer system 2000 FT-IR spectrom-
eter. EI and TOF mass spectra were carried out with GCT-MS
(Micromass UK) and BIFLFX III (Bruker) spectrometers, re-
spectively. The distribution of oligomers was determined by
GC–MS analysis using an HP-5890 apparatus with an HP-1
capillary column (30 m × 0.25 mm) and an HP-5971 mass
spectrometer. The column temperature started with 35 ◦C
(10 min), heated at 10 ◦C/min to 220 ◦C and kept at 220 ◦C
for 10 min.
1216, 1160, 1115, 815, 768 and 715 cm−1
.
Complex
5
(C5, {2,6-(2-Cl-5-CH3C6H3N=CCH3)2
C5H3N}FeCl2): ligand L5 (410 mg, 1.0 mmol), FeCl2·4H2O
(219 mg, 1.1 mmol) and THF (12 mL) were used. C5
(430 mg) was obtained as a light blue powder in 80% yield.
Elemental analysis (C23H17N3Cl2FeCl2): calcd. (%): C
51.43, H 3.94, N 7.82; found (%): C 51.15, H 4.12, N 7.69.
IR (KBr): 1624 (νC=N), 1590, 1478, 1430, 1375, 1268,
1205, 1169, 1144, 1109, 815 and 713 cm−1
.
Complex
6
(C6, {2,6-(2-ClC6H4N=CCH3)2C5H3
N}FeCl2): ligand L6 (382 mg, 1.0 mmol), FeCl2·4H2O
(219 mg, 1.1 mmol) and THF (10 mL) were used. C6 was
obtained as a light blue powder in 76% yield. Elemental
analysis (C21H17N3Cl2FeCl2) (509.0): calcd. (%): C 49.55,
H 3.37, N 8.25; found (%): C 49.76, H 3.56, N 7.98. IR
(KBr): 1623 (νC=N), 1587, 1470, 1440, 1373, 1262, 1228,
3. Results and discussions
3.1. Preparation and characterization of ligands and
complexes
1060, 1033, 820, 778, 759, 745, 730 and 688 cm−1
.
Complex
7
(C7, {2,6-(2-Cl-4-FC6H3N=CCH3)2C5
All the ligands were synthesized in good yields by con-
densation of 2,6-diacetylpyridine with the corresponding ani-
line using MAO/SiO2 as the catalyst and molecular sieve as
the water adsorbent [11] (Scheme 1). Elemental analysis, 1H
NMR, IR and mass spectrometry were used for characteri-
zation of ligands. Iron complexes were synthesized by dis-
solving the ligands in tetrahydrofuran (THF) (Scheme 1),
followed by the addition of 1.1 equiv. of FeCl2·4H2O. The
complexes were characterized with elemental analysis and
IR. The IR spectra of the free ligands show that the C=N
stretching frequencies appear at 1638–1645 cm−1. In com-
plexes 2–4, the C=N stretching vibrations shift toward lower
frequencies around 1619–1628 cm−1 and are greatly reduced
in intensity, which indicated the coordination interaction be-
tween the imino nitrogen atoms and the metal ions. Elemental
analysis results also show good accordance with correspond-
ing ligands and complexes.
H3N}FeCl2·THF): ligand L7 (418 mg, 1.0 mmol),
FeCl2·4H2O (219 mg, 1.1 mmol) and THF (10 mL)
were used. C7 was obtained as a light blue powder in 70%
yield. Elemental analysis (C21H15N3Cl2F2FeCl2·THF)
(616.1): calcd. (%): C 48.66, H 3.76, N 6.81; found (%):
C 48.76, H 3.68, N 6.55. IR (KBr): 1625 (νC=N), 1588,
1485, 1429, 1375, 1262, 1205, 1106, 1049, 908, 813 and
713 cm−1
Complex
.
8
(C8, {2,6-(2,4-F2C6H3N=CCH3)2C5
H3N}FeCl2·THF): ligand L8 (231 mg, 0.60 mmol),
FeCl2·4H2O (132 mg, 0.66 mmol) and THF (6 mL) were
used. C8 was obtained as a light blue powder in 66% yield.
Elemental analysis (C21H15F4N3FeCl2·THF) (584.2): calcd.
(%): C 51.40, H 3.97, N 7.19; found (%): C 51.36, H 4.16,
N 7.27. IR (KBr): 1619 (νC=N), 1587, 1502, 1429, 1374,
1266, 1219, 1145, 1093, 966, 875, 806 and 715 cm−1
.

Z. Zhang et al. / Journal of Molecular Catalysis A: Chemical 230 (2005) 1–8
5
Table 1
central metal positive charge. Alternatively, complexes with
both large steric bulk and high electron-withdrawing effect
will have the highest activities. Halogens are a family of inter-
esting substituents. The electronegativities of fluorine, chlo-
rine and bromine atom are 4.0, 3.0 and 2.8 while the covalent
Results of ethylene oligomerizations by iron complexes with different ortho-
substituents
Run
Complex
Yield (g)
Activity
α
␣-Olefin (%)
(106 g/mol Fe h)
1
2
3
4
5
1 [13]
9.15
12.10
7.75
11.30
14.45
18.30
24.20
15.50
22.60
28.90
0.42
0.55
0.61
0.29
0.52
>93
>94
>95
>91
>95
˚
radius (A) increases in this order: 0.72, 0.99 and 1.14, respec-
2
3
4
5
tively. This could explain why the complex 2 has the highest
activities, and complexes 1 and 3 have the lower activities be-
cause of smaller steric bulk and lower electron-withdrawing
ability.
Reaction conditions: reaction temperature = 30 ◦C; Al/Fe = 1400.
The distributions of products obtained from complexes
with different ortho-substituents are illustrated in Fig. 1(b).
The distribution of oligomers obtained follows Schulz–Flory
rules, which could be characterized by a constant α, where
α represents the probability of chain propagation [α = rate
of propagation/(rate of propagation + rate of chain trans-
fer) = mol of Cn + 2/mol of Cn]. The α value can be deter-
mined by the molar ratio of C12 and C14. The percent of
low-molar-mass products decreases and α value (as shown
in Table 1) increases in the order of fluoro-, chloro- and
bromo-substituents, which is in the same order with the cova-
3.2. Catalysis of ethylene oligomerization
Ethylene oligomerizations catalyzed by these iron coordi-
native complexes using MAO as co-catalyst have been sys-
tematically investigated. The aim of this research is to gain a
preliminary insight into the effects of the ortho-substituents,
electronic effect, the Al/Fe molar ratios and the reaction tem-
peratures on ethylene oligomerization.
3.2.1. Effect of ortho-substituent on the catalytic activity
and oligomer distribution
˚
lent radius (A). Meanwhile, complexes with the same ortho-
substituents (such as complex 1 corresponding to complex 4
and complex 2 to complex 5) have almost the same oligomer
distributions. With the increasing in steric bulk, the selectivity
for linear α-olefin increases as shown in Table 1. The result
indicates that steric bulk is the main factor that influences the
product distribution and reducing the steric bulk will result in
more low-molar-mass products, which agrees well with the
results obtained in literatures [5].
The steric bulk and electronic effect of ortho-substituents
play important roles in the complex activities and product
properties. The activities of complexes are influenced greatly
by the steric bulk and electronegativities of ortho-substituents
as shown in Table 1. The activities of complex 2 (the ortho-
substituent is chloro) at different reaction temperatures are
the highest, and complex 3 has the lowest activities and the
activities of complex 1 are between those of complexes 2 and
3 as shown in Fig. 1(a). The activities of complex 4 are lower
than those of complex 5. The results obtained by Gibson
and co-workers [5] show that larger steric bulk of the ortho-
substituent at aryl rings could improve the activities of com-
plexes. In addition, the results of Qian and co-workers [12]
show that electron-withdrawing groups at the ortho-position
of aryl rings could lead to high activities by increasing the
3.2.2. Influence of electronic effect on the catalytic
activity and oligomer distribution
Complexes 2, 6 and 7, and complexes 1 and 8 have the
same ortho-substituents but different para- ones in the aryl
rings, respectively. The change of para-substituents from F,
H to CH3 is not expected to obtain significant change in
Fig. 1. Catalytic activities: (a) (Al/Fe = 1400) and distribution (b) of oligomers (reaction temperature = 60 ◦C; Al/Fe = 1400) obtained by complexes with
different ortho-substituents.

6
Z. Zhang et al. / Journal of Molecular Catalysis A: Chemical 230 (2005) 1–8
Fig. 2. The catalytic activities of complexes with the same ortho-substituents. Reaction conditions: Al/Fe = 1400.
steric bulk. For these complexes we can consider that the
electronic effect is more important than the steric effect. As
reported before, the para-substituents dramatically influence
the catalytic activities by changing the positive charge of
center metal, which could accelerate or decelerate the ethy-
lene coordination and chain transfer [12]. The catalytic ac-
tivities of these complexes at different reaction temperatures
are given in Fig. 2. The catalytic activities are in the follow-
ing orders: complex 2 > complex 6 > complex 7 and complex
1 > complex 8, which is in the contrary order with the elec-
tronegativities of the para-substituents on the aryl rings. This
means the introduction of electron-withdrawing groups will
decrease the catalytic activities and vice versa. It has also
been reported that the introduction of electronic-withdrawing
bromo group in the para-position of the aryl rings increases
the catalytic activity if the ortho-substituents are electron-
donating groups [12]. All the results indicate that the com-
bination of substituents with contrary electronic character
favors the increase of catalytic activity.
is interesting to find that the α values increase as the para-
substituents change from CH3, H to F, which indicates the
increase in chain-transfer rate. This could be also proved by
the distribution of oligomers obtained by these complexes as
shown in Fig. 3. The oligomers shift to the lower mass weight
part as the electron-donating ability of the para-substituents
increases. All results mean that the para-substituents influ-
ence the chain transfer by changing the electronic character
of the aryl rings and then the metallic center.
In addition, if methyl substituent is changed from the
fourth position to the fifth position (such as complex 1
corresponding to complex 4 and complex 2 corresponding
to complex 5), the catalytic activities increase and the α
values decrease more or less as shown in Table 2. When
the ortho-substituent is fluoro, the activities of complexes
are higher than when the ortho-substituent is chloro. As
an electron-donating group, methyl could provide differ-
ent positive charge of center metal when depending on
the positions of aryl rings, which is one of the impor-
tant factors in the activation of complexes as discussed
above.
In order to study the effect of different para-substituents
onthepropertiesofproducts, theresultsofethyleneoligomer-
ization obtained by these complexes at the same reaction con-
ditions are summarized in Table 2. The para-substituents do
not exhibit obvious effect on the selectivity of ␣-olefins. It
3.2.3. Effect of reaction temperature on the catalytic
activity and oligomer distribution
As other complexes reported before [1–10], the catalytic
activities of these complexes are influenced strongly by re-
action temperature as shown in Fig. 1(a). The catalytic ac-
tivities of all the complexes decrease by almost one order
of magnitude when reaction temperature increases from 30
to 70 ◦C. Generally, an increase in temperature is expected
to result in overall enhanced propagation rate and there-
fore increased productivities. However, a decrease in ethy-
lene solubility and increased rates of catalyst deactivation at
higher temperatures may result in a decline in the catalytic
activity.
Table 2
Results of ethylene oligomerizations by iron complexes with different para-
substituents
Run
Complex
Yield (g)
Activity
α
␣-Olefin (%)
(106 g/mol Fe h)
1
2
3
4
5
6
7
2
6
7
12.10
8.20
6.20
9.15
7.70
24.20
16.40
12.40
18.30
15.40
22.60
28.90
0.55
0.62
0.65
0.42
0.52
0.29
0.52
>94
>94
>95
>93
>92
>91
>95
1 [13]
8
4
5
11.30
14.45
The product distributions of complex 2 at different reac-
tion temperatures are given in Fig. 4. As reaction temperature
Reaction conditions: reaction temperature = 30 ◦C; Al/Fe = 1400.

Z. Zhang et al. / Journal of Molecular Catalysis A: Chemical 230 (2005) 1–8
7
Fig. 3. The distributions of the oligomers obtained by complexes 1–5 with different para-substituents. Reaction conditions: Al/Fe = 1400; reaction tempera-
ture = 30 ◦C.
3.2.4. Effect of Al/Fe molar ratio on the catalytic activity
and oligomer distribution
The effect of Al/Fe ratio on activities of complexes is stud-
ied and the results are given in Fig. 5. It is very interesting
to find that as the Al/Fe ratio increases the activities of com-
plexes change in the same trend when the ortho-substituent is
the same. The activities of complexes 1 and 4, which have a
fluorine atom at the ortho-position of aryl rings, increase and
reach the maximum at Al/Fe ratio of 1000 and 420, respec-
tively, then decrease rapidly. But the activities of complexes
2 and 5, which bear a chlorine atom at the ortho-position of
aryl rings, decrease monotonously. The activity of complex 3,
whose ortho-substituent is displaced with bromine, increases
gradually and reaches the maximum at about 3000 Al/Fe. In
addition, the product distribution is almost unchanged when
the Al/Fe ratio is altered as shown in Fig. 6, which is similar
with the results we reported before [14]. More active species
are formed when more MAO is introduced to the reaction
system, which results in the increase in activities, but too
Fig. 4. Effect of reaction temperature on product distribution (obtained by
complex 2, reaction conditions: Al/Fe = 1400).
increases from 30 to 70 ◦C, the products shift to lower mass
weight part and the α value decreases as shown in Table 3.
Increasing reaction temperature will result in overall higher
propagation and transfer rates. The dependence of α value on
temperature indicates that the rate of chain transfer increases
more than the rate of propagation, which is expected to afford
lower molecular weight products.
Table 3
Results of ethylene oligomerizations by complex 2 at different reaction
temperatures
Run
Temperature
(◦C)
Yield (g)
Activity
(106 g/mol Fe h)
α
␣-Olefin
(%)
1
2
3
4
5
30
40
50
60
70
12.10
10.05
7.80
4.50
1.45
24.2
21.0
13.6
9.00
2.90
0.55
0.55
0.54
0.53
0.51
>94
>94
>96
>96
>95
Fig. 5. Effect of Al/Fe ratio on activities of complexes 1–5 (at 60 ◦C).
Reaction conditions: Al/Fe = 1400.

8
Z. Zhang et al. / Journal of Molecular Catalysis A: Chemical 230 (2005) 1–8
ortho-position. The steric bulk is one of the most important
factors that influence the distribution of oligomers. As reac-
tion temperature increases, the activities decrease and low-
molar-mass products increase. As the Al/Fe ratio increases,
the activities change more or less, depending on the differ-
ent ortho-substituents at aryl rings, and the distribution of
products is almost unchanged.
Acknowledgments
This project was supported by the National Natural Sci-
ence Foundation (No. 20334030) and Sinopec.
References
Fig. 6. Effect of Al/Fe ratio on distribution of product (obtained from com-
plex 5 at 60 ◦C).
[1] B.L. Small, M. Brookhart, J. Am. Chem. Soc. 120 (1998)
7143–7144.
[2] B.L. Small, M. Brookhart. US Patent 6,150,482 (2000) to E.I. du
Pont de Nemours and Company.
[3] B.L.Small, M. Brookhart. US Patent 6,103,946 (2000) to E. I. du
Pont de Nemours and Company.
[4] G.J.P. Britovsek, V.C. Gibson, B.S. Kimberley, P.J. Maddox, S.J.
Mctavish, G.A. Solan, A.J.P. White, D.J. Williams, Chem. Commun.
(1998) 849–850.
[5] G.J.P. Britovsek, S. Mastroianni, G.A. Solan, S.P.D. Baugh, C. Red-
shaw, V.C. Gibson, A.J.P. White, D.J. Williams, J. Chem. Eur. J. 6
(2000) 2221–2231.
much MAO could deactivate the active species through the
reaction with trimethylaluminium (TMA) contained in com-
mercial MAO. The results indicate that the reaction between
these complexes and TMA occurs more easily in the order:
F > Cl > Br, which is in the same order with the steric bulk of
their ligands. Alternatively, the reaction occurs more readily
with complexes containing a fluoro substituent on the ortho-
position of aryl rings, which could not provide sufficient pro-
tection of central metal. This result is well in agreement with
that obtained by Gibson and co-workers [5].
[6] R. Schmidt, U. Hammon, S. Gottfried, M.B. Welch, H.G. Alt, J.
Appl. Polym. Sci. 88 (2003) 476–482.
[7] C. Bianchini, G. Mantovani, A. Meli, F. Migliacci, F. Zanobini, F.
Laschi, A. Sommazzi, Eur. J. Inorg. Chem. (2003) 1620–1631.
[8] H. Wang, W. Yan:, T. Jiang, B. Liu, W. Xu, J. Ma, Y. Hu, Chin.
Sci. Bull. 47 (2002) 1616–1618.
4. Conclusions
[9] T. Jiang, W. Yan, H. Wang, B. Liu, W. Xu, J. Ma, Y. Hu, Acta
Polym. Sinica (2002) 293–295.
[10] M. Bluhm, C. Folli, M. Do¨ring, J. Mol. Catal. A: Chem. 212 (2004)
13–18.
[11] Y. Chen, C. Qian, J. Sun, Organometallics 22 (2003) 1231–
1236.
[12] Y. Chen, R. Chen, C. Qian, X. Dong, J. Sun, Organometallics 22
(2003) 4312–4321.
[13] I.S. Paulino, U. Schuchardt, J. Mol. Catal. A: Chem. 211 (2004)
55–58.
[14] Z. Zhang, J. Zou, N. Cui, Y. Ke, Y. Hu, J. Mol. Catal. A: Chem.
219 (2004) 249–254.
In summary, all the complexes exhibit high catalytic activ-
ities for ethylene oligomerization. The iron complexes bear-
ing a 2,6-bis(imino)pyridyl ligand with a chloro substituent
at the ortho-position and a methyl one at the para- or meta-
position have higher catalytic activity than other complexes.
The steric bulk and positive charge of center metal caused by
different substituents at aryl rings are main factors that influ-
ence the activities of complexes. The percent of low-molar-
mass products decreases with increase the steric bulk at the