Effect of ion-pair strength on ethylene oligomerization by divalent nickel complexes

Article abstract of DOI:10.1007/s10562-013-1021-7

Detailed ethylene oligomerizations were performed over highly active nickel complexes ligated by N-((4-butyl-6-phenylpyridin-2-yl)methylene)-2,6- diisopropylaniline derivatives. The activity trend according to the structure of the complex was attempted to understand using the concept of ion-pair formation during activation process. The active species were characterized by comparing the oligomerization results with the UV-Vis spectroscopic and the cyclic voltammetry experiments.

Full text of DOI:10.1007/s10562-013-1021-7

Catal Lett (2013) 143:717–722
DOI 10.1007/s10562-013-1021-7
Effect of Ion-Pair Strength on Ethylene Oligomerization
by Divalent Nickel Complexes
Deepak Chandran
•
Seong Jin Byeon
•
Hongsuk Suh Il Kim
•
Received: 21 March 2013 / Accepted: 2 May 2013 / Published online: 15 May 2013
Ó Springer Science+Business Media New York 2013
Abstract Detailed ethylene oligomerizations were per-
formed over highly active nickel complexes ligated by
N-((4-butyl-6-phenylpyridin-2-yl)methylene)-2,6-diiso-
propylaniline derivatives. The activity trend according to
the structure of the complex was attempted to understand
using the concept of ion-pair formation during activation
process. The active species were characterized by com-
paring the oligomerization results with the UV–Vis spec-
troscopic and the cyclic voltammetry experiments.
ion-pairs are important in understanding and optimizing the
performances of various types of complexes in catalysis.
Most of such studies reported so far are based on early
transition metal complexes [1–5]. Even though a large
number of complexes of N^N and N^N^N ligands bearing
various late transition metals such as Ni, Pd, Co, Fe and Cr
are widely employed catalysts in ethylene polymerization/
oligomerization [6, 9–13] attempts to study their perfor-
mance based on the ion-pair strength are hardly made.
We have recently reported the synthesis of a series of
Ni-diimine catalysts based on N-((4-butyl-6-phenylpyridine-
2-yl)methylene)-2,6-diisopropylaniline derivatives [14]. Apart
from the interesting observation and information obtained
during their synthesis, these complexes were observed with
high catalytic potential in ethylene oligomerization (EO)
which serves them suitable candidates to study the ion-pair
formation and strength at various reaction conditions. Three
different complexes having the same core structure has been
adopted in order to investigate this effect in this study. We
found that, as in the case of early transition metal complexes,
the ion-pair formation and its stability are equally influential
in the cationic late transitions metal complexes. Here,
attempts have been made to study EO by using the three
Ni(II) complexes based on the ion-pair concept.
Keywords Catalyst Á Ethylene Á Oligomerization Á
Nickel Á Transition metal chemistry
1
Introduction
Alkyl or hydride anion abstraction reactions using strong
Brønsted or Lewis acids from transition metal complexes
are the pivotal activation step in homogeneous ‘single site’
Ziegler–Natta catalysis [1–7]. This abstraction process
mediated by Lewis acidic reagents and reaction conditions
yields highly electrophilic and mononuclear ion pairs,
exhibiting unprecedented olefin polymerization activity
and selectivity [8]. A detailed investigation on the forma-
tion and strength at various reaction conditions for these
2
Experimental
D. Chandran Á S. J. Byeon Á I. Kim (&)
The WCU Center for Synthetic Polymer Bioconjugate Hybrid
Materials, Department of Polymer Science and Engineering,
Pusan National University, Busan 609-735, Korea
e-mail: ilkim@pusan.ac.kr
2
.1 General Methods and Materials
All reactions and operations were performed under a
purified nitrogen atmosphere using the standard glove box
and Schlenk techniques. Polymerization grade ethylene
H. Suh
Department of Chemistry and Chemistry Institute
for Functional Materials, Pusan National University,
Busan 609-735, Korea
(SK Co., Korea) was purified by passing it through col-
umns of Fisher RIDOXTM catalyst and molecular sieve
123

7
18
D. Chandran et al.
˚
A/13X. All solvents were purified according to the
5
varying steric and electronic demands. These complexes
can be divided into two groups based on the type of sub-
stituent position. Complexes 8–10 have different methy-
lenic substituents, –H, –Me and –Ph. The different
substituents on the core structure vary the steric and elec-
tronic effects on the metal center. Complexes 11–13 pos-
sess an n-butyl tail in common. This n-butyl tail does not
change the steric pattern of the metal center, but may
influence electronically. Complexes 11–13 were obtained
in high yields and showed better activity than complexes
8–10. Accordingly, complexes 11–13 were employed for
the detailed EOs. These complexes can shed light into the
effect of structural factors on the ion-pair strength.
˚
standard procedures and stored over molecular sieves (4 A)
under nitrogen condition. Methylalumoxane (MAO)
(
8.4 wt% total Al in toluene) was donated by LG Chemi-
cals, Korea was used without purification. All other
reagents including diethylaluminum chloride (DEAC),
methylaluminum dichloride (MADC), and ethylaluminum
sesquichloride (EASC) were purchased from Sigma–
Aldrich and used without further purifications.
2
.2 Characterization
UV–Vis spectra were recorded on a Shimadzu UV-1650
PC spectrometer in toluene at room temperature under
EOs have been performed at different conditions of
solvent, temperature, co-catalysts and ethylene pressure and
the results are summarized in Table 1. The complex 12 has
been chosen for general study due to its intermediate steric
demands. EO at high ethylene pressure (5.5 bar) gave a
1
13
nitrogen atmosphere. H (300 MHz) and C (75 MHz)
NMR spectra were recorded on a Varian Gemini 2000
spectrometer and chemical shifts were reported in parts per
million relative to internal (CH ) Si. Oligomerization
3
4
6
products were analyzed by a 7,890 A gas chromatograph
GC) (Agilent Tech.) with a J&W Scientific 30 m column
higher activity (42.3 9 10 g-oligomer/mol-Ni h bar),
(
keeping the oligomer distribution similar (run no. 19). Even
at 1.3 bar of ethylene pressure the activity was considerably
6
high in chlorobenzene (26.5 9 10 g-oligomer/mol-Ni
with 0.250 mm inner diameter. Cyclic voltammetry (CV)
measurements were conducted on a BAS CV-50 W vol-
tammetric analyzer with scan rates of 100 mV/s. The
electrolytic cell used was a conventional three-compart-
ment cell, in which a glass carbon working electrode, a
platinum counter electrode, and Ag/AgCl reference elec-
trode were employed. The CV measurements of the com-
plexes were performed under a nitrogen atmosphere at
room temperature in N,N-dimethylformamide (DMF) using
h bar). The complex 13 gave the highest activity among the
6
catalysts (35.7 9 10 g-oligomer/mol-Ni h bar), while 11
6
with the lowest activity (16.3 9 10 g-oligomer/mol-Ni
h bar) at the same conditions. The gradual change in
activity pattern can be explained based on the electronic and
steric parameters of the substituents.
Among the substituents employed in this study –Ph
and –Me groups are electron donating with the former group
slightly higher, and -H is the least electron donating group.
Higher electron donating –Ph group makes the metal center
more electron rich, while less electron donating –Me group
makes the metal center only moderately electron rich. As the
least electron donation group, the influence of –H group on
electron availability on the metal center will be the lowest.
This type of electronic influences has been successfully
monitored previously by using CV studies [13–15]. The CV
curves of the complexes showed reduction waves between
0.0 and –1.0 V (Fig. 1). The reduction potentials obtained
from CVs are -0.99, -0.74 and -0.46 V for complexes 11,
12 and 13, respectively, showing the electron density on the
metal center is highest for 13, lowest for 11 and intermediate
for 12. It is a direct consequence of electron donating effects
exerted by the substituents on the organic backbone of the
complexes. In general as the electron density on the metal
center becomes higher, the oxidative addition of the neutral
ligand, ethylene to the metal center becomes easier and
hence the activity in EO/ethylene polymerization is
enhanced. Thus, the CV results can be a reasonable evidence
to explain the activity trend: 13 [ 12 [ 11.
0
.10 M tetrabutylammonium perchlorate (TBAP) pur-
chased from TCI as the supporting electrolyte.
2
.3 Oligomerization
EOs were performed at 1.3 bar by using a 250 mL round
bottom flask equipped with a magnetic stir bar as a reactor
[
14]. After the given reaction time the reactor was cooled
to 0 °C and the samples for analysis were collected from
the reactor by passing a 10 mL of this cold mixture through
a silica column to remove Al species.
2
.4 Synthesis of Complexes
Synthesis of the complexes for EO study is summarized in
Scheme 1 and a detailed description of their structure has
been previously reported [14].
3
Results and Discussion
As previously reported, complexes 8–14 were synthesized
according to the Scheme 1 by using different ligands [14].
Complexes 8–13 retain the same core structure with
Irrespective of these electronic and structural factors all
the complexes give oligomer distribution in more or less
1
23

Effect of Ion-Pair Strength on Ethylene Oligomerization
719
Scheme 1 Synthesis of
complexes 8–14
similar pattern. All the complexes gave exclusively butenes
and minor amount of hexenes (Table 1). A maximum
amount of butenes (93 %) was obtained when 12 is com-
bined with EASC ([Al]/[Ni] = 250) at 30 °C. In all con-
ditions a-olefins dominate both in hexenes and butenes. At
activity of the catalyst [16, 17]. Two possible structures
have been identified as active species. Firstly, a bimetallic
Al–metal bridged species can be the active species. Sec-
ondly, an ion-pair model containing cationic metal species
formed through the abstraction of halide anions by the Al
containing co-catalyst can act as the active species. Here
the Al co-catalyst with halogen can further transform to a
halogen-Al counterion. Possibility of bridged structure
alone being the active species was ruled out by the
spectroscopic investigation [18] and also by the kinetic
studies [19]. There are growing experimental evidences
that, in catalyst co-catalyst architecture, the co-catalyst
counterion’s structural matching and fitting with cation
play significant role in the structure and energetics of the
ion pairing and hence catalytic activities and selectivity of
the system [19].
-
10 °C 12 gave 98 % total a-olefins. There is no con-
spicuous trend according to experimental parameters with
slight change in the amounts of the individual oligomers.
As the temperature increases, the amount of internal olefins
increases; and with the increasing alkylaluminum (AA)
content overall hexene content decreases.
The activity of the catalysts has also been investigated
combined with different types of co-catalyst such as
MAO, EASC, MADC and DEAC. It is generally recog-
nized that AAs have different abilities to reduce organo-
metallic complex in
a higher valence state. The
replacement of one or two alkyl groups by halogens
bearing heteroatoms usually dramatically alters the
In the present study, DEAC gave no activity while MAO a
negligible activity (Table 1) due to the drastic deactivation.
123

7
20
D. Chandran et al.
Table 1 Ethylene oligomerization results over Ni(II) complexes in combination with various organoaluminum co-catalysts at different reaction
environments
a
b
-6
c
Run no.
Cat.
Cocat.
[Cocat.]/[Ni]
PC2H4 (bar)
T (°C)
Solvent
R
avg x 10
Selectivity
=
4
=
C
6
C
=
d
=e
RC4i
=d
a
=e
RC6i
a
1
2
3
4
5
6
7
8
9
11
12
13
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
EASC
EASC
EASC
MAO
150
150
150
150
150
150
50
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
5.5
30
30
30
30
30
30
30
30
30
30
50
70
0
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
PhCl
Toluene
Hexane
16.3
26.5
35.7
0.3
23.4
–
86
84
81
78
83
–
5
2
7
6
3
–
4
3
4
7
7
7
2
1
–
4
–
6
3
5
9
4
5
4
6
4
–
3
5
4
3
5
8
2
1
–
3
–
3
3
8
10
10
–
MADC
DEAC
EASC
EASC
EASC
EASC
EASC
EASC
EASC
EASC
EASC
EASC
EASC
EASC
EASC
17.3
19.7
8.6
2.1
0.5
0.1
1.8
1.4
–
83
84
85
86
80
76
85
86
–
10
8
100
200
250
150
150
150
150
150
150
150
150
150
7
1
1
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
7
8
9
4
8
9
11
12
–
-10
-20
30
30
30
30
2.0
–
80
–
13
–
CH
2
Cl
2
3.6
42.3
82
80
10
14
PhCl
Oligomerization conditions: catalyst = 2.5 lmol, solvent = 80 mL, and time = 30 min
a
Organoaluminum co-catalysts: EASC ethylaluminum sesquichloride, MAO methylaluminoxane, MADC methylaluminum dichloride, and
DEAC diethylaluminum chloride
b
Average rate of oligomerization over a period of reaction in g-oligomer/mol-Ni h bar
c
Determined by GC
d
a-Olefin
e
Sum of internal olefins other than a-olefin
Side reactions such as alkyl exchange and H-exchange
reactions or the reduction of metal ion to its lower oxidation
state can cause deactivation. In the case of MAO the major
deactivation process is a-hydrogen transfer which will lead
to the production of methane [16]. Combining with EASC,
activation process the metal complexes, LNiBr2
(L = ligand), react with AA co-catalysts producing an
?
alkylated cationic complex, LNi–R , which may again
condense with the AA to form inactive LNi–R–NiL and/or
LNi–R–Al species. Such inactive species can be reacti-
vated with excess amount of AAs [16]. Thus, the lower
activity at lower co-catalyst concentration may be due to
the formation of these inactive species that are not further
reactivated. In the presence of suitably excess amount of
co-catalyst, the inactive species might be dissembled to
6
complex 12 showed the highest activity (26.5 9 10 g-oli-
gomer/mol-Ni h bar). MADC also gave a considerably high
6
activity with relatively slow deactivation (23.4 9 10 g-oli-
gomer/mol-Ni h bar). As described above the difference in
activity is mainly caused by the nature of ion-pair formed
from the catalyst co-catalyst assembly, resulting from the
difference in structure, electronic demands and the strength
of counterion formed from these Al co-catalysts.
?
form active LM–R again. The increase of the co-catalyst
concentration to a large excess results in a situation where
the population of highly coordinating Lewis acidic species
is increased. This can nullify the effective ion-pair sepa-
ration resulting in a decrease in activity.
The effect of co-catalyst concentration has also been
studied. As [EASC]/[Ni] ratio increases from 50 to 150, the
activity gradually increases but a further increase in EASC
amount results in the decrease of activity. During the
Interestingly all the complexes gave higher activity in
chlorinated solvents than in nonpolar solvents. While the
1
23

Effect of Ion-Pair Strength on Ethylene Oligomerization
721
Fig. 2 UV–Vis spectra of nickel complex 12 (2.5 lmol) combined
with different amounts of EASC ([EASC]/[Ni] = 0–250) in
chlorobenzene
Fig. 1 Cyclic voltammograms of various nickel complexes, 11–13
in DMF at 25 °C with scan rate of 100 mV/s
complexes give a very high activity in methylene chloride
and chlorobenzene, they show very low activity in toluene
and no activity in n-hexane. Increasing the polarity of the
solvent can increase the ion pair separation. Activation
energy for ethylene insertion is lower in cationic systems
than that in ion-pair system [20]. Thus, the reason for
giving a lower or no activity in nonpolar solvents might be
due to an inefficient ion-pair separation, while high activ-
ities in chlorinated solvents are related with the facilitated
separation of ion-pair in polar medium.
containing 2.5 lmol of complex 12 by varying the co-
catalyst concentration from [EASC]/[Ni] = 50 to 250 for
simulating the EO conditions. Note that EOs were carried
out in this range of co-catalyst concentration (Table 1). By
adding EASC to the solution, a new absorption peak is
observed at around 850 nm (Fig. 2). Absorbance of this
peak reaches the maximum at [EASC]/[Ni] = 150 and then
starts to decrease by further increase of the EASC
concentration.
This trend of absorbance of the peak at around 850 nm
is in line with the trend of activity observed in EO. A plot
of activity and molar extinction coefficient against the
EASC concentration gives a better idea about the correla-
tion between the population density of active species and
activity (Fig. 3). Since the activity is directly related to the
ion-pair separation, a correlation between population den-
sity of active species and ion-pair separation can be
understood. According to this observation, the population
density of active species will be less at lower co-catalyst
concentration. It is interesting to note that this trend mat-
ches with the trend of activity according to EASC con-
centration, demonstrating the absorption peak observed at
around 850 nm is generated due to the formation of the
active species.
An expedition on the influence of temperature on EO
activities revealed that the maximum activity of
6
6.5 9 10 g-oligomer/mol-Ni h bar for complex 12 is
2
observed at 30 °C. At higher temperature a drastic decrease
in activity was noticed (run no. 11 and 12 in Table 1). For
a-diimine nickel complexes, Brookhart has found that the
catalyst life time is significantly short at elevated tempera-
tures [21]. Along with the low solubility of ethylene at high
temperatures, catalyst decay plays an important role in this
decreased activity. However, the catalyst showed a consid-
6
erable activity at a very low temperature (1.4 9 10 g-oli-
gomer/mol-Ni h bar at -10 °C). The inactive LNi–R–NiL
or LNi–R–Al species are predominantly reversible at lower
temperatures, which increase the population density of
active species in the system giving a remarkable activity
even at a low temperature.
Separation of ion-pair is directly related to the formation
of active species and thus to the activity. In order to
investigate the active species formation, UV–Vis spectro-
scopic technique has recently been used [17, 22]. The UV–
Vis spectra were recorded for a chlorobenzene solution
4 Conclusions
A series of N^N type bidentate nickel complexes bearing N-((4-
butyl-6-phenylpyridin-2-yl)methylene)-2,6-diisopropylaniline
derivatives as ligands have been prepared by one-pot reactions
123

7
22
D. Chandran et al.
Acknowledgments This work was supported by Basic Science
Research Program (2012R1A1A2041315) and the Fusion Research
Program for Green Technology (2012M3C1A1054502) through the
National Research Foundation of Korea funded by the Ministry of
Education, Science and Technology.
References
1
2
3
4
. Kaminsky W, Arndt M (1997) Adv Polym Sci 127:143
. Bochmann M (1996) J Chem Soc Dalton Trans 255
. Grubbs RH, Coates GW (1996) Acc Chem Res 29:85
. Brintzinger HH, Fischer D, M u¨ lhaupt R, Rieger B, Waymouth
RM (1995) Angew Chem Int Ed 34:1143
5
. Baumann R, Davis WM, Schrock RR (1997) J Am Chem Soc
1
19:3830
6
7
8
. Ittel SD, Johnson LK, Brookhart M (2000) Chem Rev 100:1169
. Zurek E, Ziegler T (2001) Organometallics 21:83
. Deck PA, Beswick CL, Marks TJ (1998) J Am Chem Soc
1
20:1772
9. K o¨ hn RD, Haufe M, Kociok-K o¨ hn G, Grimm S, Wasserscheid P,
Keim W (2000) Angew Chem Int Ed 39:4337
10. Johnson LK, Killian CM, Brookhart M (1995) J Am Chem Soc
Fig. 3 Ethylene oligomerization activity and difference in molar
extinction coefficients from UV–Vis spectra estimated by complex 12
combined with different amounts of EASC ([EASC]/[Ni] = 0–250)
1
17:6414
1. Small BL, Carney MJ, Holman DM, O’Rourke CE, Halfen JA
2004) Macromolecules 37:4375
2. Chandran D, Kwak CH, Oh JM, Ahn IY, Ha CS, Kim I (2008)
Catal Lett 125:27
13. Chandran D, Bae C, Ahn IY, Ha CS, Kim I (2009) J Organomet
Chem 694:1254
4. Chandran D, Lee KM, Chang HC, Song GY, Lee J-E, Suh H,
Kim I (2012) J Organomet Chem 718:8
5. Gibson VC, Long NJ, Oxford PJ, White AJP, Williams DJ (2006)
Organometallics 25:1932
6. Bahuleyan BK, Son GW, Park D-W, Ha C-S, Kim I (2008)
J Polym Sci Part A: Polym Chem 46:1066
7. Appukuttan V, Zhang L, Ha C-S, Kim I (2009) Polymer 50:1150
8. Long WP (1959) J Am Chem Soc 81:5312
19. Fink G, Zoller W (1981) Makromol Chem 182:3265
1
1
(
with2,6-diisopropyl anilineandnickel bromide derivatives. All
these complexes exhibited considerably high activity toward
ethylene when activated with EASC and MADC to give
butenes and hexanes with the butenes and l-olefins as domi-
nating products. Detailed EO reactions have been carried out at
different experimental parameters such as the type and the
amount of co-catalyst, temperature and solvents. The activity
difference according to the structure of the complex was
attempted to analyze using the concept of a possible ion-pair
formation during activation process. The ion-pair formation
and its stability were equally influential in cationic late transi-
tion metal complexes as in their early metal counterparts. An
attempt to assign the active species was made by comparing the
EO results with UV–Vis spectroscopic study, successfully
identifying an absorption peak assignable to active species. The
CV experiments were also made to get an idea about the
electronic influence of the structure of ligands on metal center.
1
1
1
1
1
2
2
0. Yang SH, Huh J, Jo WH (2005) Macromolecules 38:1402
1. Gates DP, Svejda SA, O n˜ ate E, Killian CM, Johnson LK, White
PS, Brookhart M (2000) Macromolecules 33:2320
2. Chandran D, Kwak CH, Ha C-S, Kim I (2008) Catal Today
131:505
2
1
23