A low-symmetry nickel(II) α-diimine complex for homopolymerization of ethylene: Study of interactive effects of polymerization parameters

Article abstract of DOI:10.1080/00958972.2015.1057130

In this contribution, the synthesis and characterization of a non-symmetric nickel(II) α-diimine complex [(2,4,6-trimethylphenyl)imino)-((2,6-diisopropylphenyl)imino)-acenaphthene nickel(II) dibromide] have been reported as a catalyst precursor for ethyle

Full text of DOI:10.1080/00958972.2015.1057130

This article was downloaded by: [New York University]
On: 05 June 2015, At: 04:28
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Coordination Chemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/gcoo20
A low-symmetry nickel(II) α-diimine complex
for homopolymerization of ethylene: Study of
interactive effects of polymerization parameters
Nona Ghasemi Hamedania, Hassan Arabia & Francis S. Mairb
a Department of Polymerization Engineering, Iran Polymer and Petrochemical Institute,
P.O. Box: 14965/115, Tehran, Iran
b School of Chemistry, The University of Manchester, Brunswick Street, Manchester M13
9PL, UK
Accepted author version posted online: 02 Jun 2015.
Click for updates
To cite this article: Nona Ghasemi Hamedani, Hassan Arabi & Francis S. Mair (2015): A low-symmetry nickel(II) α-diimine
complex for homopolymerization of ethylene: Study of interactive effects of polymerization parameters, Journal of
Coordination Chemistry, DOI: 10.1080/00958972.2015.1057130
To link to this article: http://dx.doi.org/10.1080/00958972.2015.1057130
Disclaimer: This is a version of an unedited manuscript that has been accepted for publication. As a service
to authors and researchers we are providing this version of the accepted manuscript (AM). Copyediting,
typesetting, and review of the resulting proof will be undertaken on this manuscript before final publication
of the Version of Record (VoR). During production and pre-press, errors may be discovered which could
affect the content, and all legal disclaimers that apply to the journal relate to this version also.
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Publisher: Taylor & Francis
Journal: Journal of Coordination Chemistry
DOI: http://dx.doi.org/10.1080/00958972.2015.1057130
A low-symmetry nickel(II) α-diimine complex for homopolymerization of
ethylene: Study of interactive effects of polymerization parameters
NONA GHASEMI HAMEDANI†, HASSAN ARABI*† and FRANCIS S. MAIR*‡
†Department of Polymerization Engineering, Iran Polymer and Petrochemical Institute, P.O. Box: 14965/115,
Tehran, Iran
‡School of Chemistry, The University of Manchester, Brunswick Street, Manchester M13 9PL, UK
A low-symmetry nickel(II) α-diimine complex [(2,4,6-trimethylphenyl)imino)-((2,6-
diisopropylphenyl)imino)-acenaphthene nickel(II) dibromide] (C1) was synthesized and
characterized crystallographically. Response surface method (RSM) was employed to study the
interactive effects of critical factors on ethylene polymerization. The maximum activity of C1,
when activated by methylaluminoxane (MAO) co-catalyst {4960 kgPE [(mol Ni)-1 h-1]} was
achieved at 10 °C, 7 bar and CC (co-catalyst to catalyst ratio) equal to 1550, which is higher than
most similar counterparts. Considering surface plots of MW, it is concluded that at high levels
(>1000) of CC, there is competition between chain transfer to aluminium and reinsertion of
macromonomer to newly formed metal-methyl bonds.
Keywords: Low symmetry ligand; Nickel(II) α–diimine complex; Polyethylene (PE); Response
surface method; Polymer properties
Introduction
The development of late transition metal complexes as catalysts for oligomerization and
polymerization of α-olefins is of interest, mainly because of their air stability, polymerization
characteristics and tolerance to polar functional groups. A major advancement in this area was
the discovery of homogeneous nickel and palladium α–diimine complexes in 1995 [1, 2]. Unique
to these systems is the ability to yield a high degree of branching in the polyethylene produced
without the use of α-olefin comonomers, a feature attributed to a catalytic mechanism referred to
*Corresponding authors. Email: h.arabi@ippi.ac.ir (H. Arabi); francis.s.mair@manchester.ac.uk (F.S. Mair)
1

as "chain-walking" [3]. This mechanism is under the control of catalyst structure and
polymerization conditions in such a way that properties of the polymer produced cover a wide
range of grades from highly branched completely amorphous to semi crystalline materials [3-5].
Tremendous efforts have been spent in both industrial and academic laboratories on improving
late transition metal catalyst systems. Many structural modifications, including changing the
ligand backbone structure or chelating heteroatoms, have been pursued on α-diimine systems
[3, 6], since control of steric hindrance and ligand electronic structure are critical to attainment of
the desired properties in the catalytic system, and hence in the polymers produced from it. These
parameters have been studied via modification of the ligand frameworks in two ways: (i) via
variation of the substituents on the arylimino groups [4-15] and (ii) via changes in the backbone
substituents [16-18]. As for (i), it has often been shown that the steric hindrances of ortho
substituents affect the catalytic activity [11-13], where sufficient bulk retards the rate of
associative olefin displacement, affording polymers with higher molecular weights, while
electronic effects are also found for aryl substituents [9, 10, 12-15], whereas for (ii) it has also
been shown that incorporation of electron-donating or -withdrawing substituents onto α-diimine
ligand backbones affects their σ-donating ability to metal centers. This in turn affects the Lewis
acidity of the metal [15-18], leading to effects on activities and selectivities of polymerization.
An increase in polar olefin incorporation has been reported for electron-donating ligands [13-15],
whereas para-chloro-substituents gave higher activities [9] than their hydro-equivalents. The
effect of p-Me groups on aryls has been mixed, with reports of both higher [7] and lower [8]
activities for p-Me vs p-H on aryl groups, dependent on organoaluminium and catalyst used.
These reports employed a means to impart finer control on the degrees of bulk at the metal center
through lowered symmetry [19, 20]. Catalyst C1, for example, lacks one of the molecular planes
of symmetry present in benchmark C2v-symmetry Brookhart catalyst C2 [1], since it has different
aryl groups on either side of the metal (scheme 1). Comparisons of activity for such a pairing
have already been made in the literature [7], as have comparisons of the effect of p-Me groups
and other remote electronic effects [7, 9, 10, 12-15], and of variations in the nature of the
lowering of symmetry, e.g. in comparisons of C4, which retains a C2 axis in the conformation
shown, but which also possesses an alternative conformation where the plane of symmetry
containing the C2N2Ni metallacycle is removed, whereas the plane of symmetry bisecting this
plane, the one absent in C1 and C3, is recovered (scheme 1) [19]. Interestingly, C4, with
2

alternative conformations, gave a broader molecular weight distribution than its structural isomer
C3, as well as higher molecular weight and lower branching density, demonstrating that even an
identical set of ortho substituents, but differently arrayed, can generate different polymerization
behaviors [19]. There are reports of polymerization activity for each of C1-C4 already present in
the literature [1, 7, 19], the area has been reviewed [21, 22] and a fundamental theoretical basis
for the observations, which differ little from those of more symmetric cases, has been laid down
[23-25]. However, reports of exhaustive and systematic study of the interplay of experimental
conditions (temperature, pressure, co-catalyst ratio, etc.) with observed output parameters
(branching rate, crystallinity, MW, etc.), the efficacy of which has recently been demonstrated
for Ni(II)-catalyzed ethylene polymerizations [26], is absent for all diimine catalysts except C2,
using AliBu3 co-catalyst [27]. In this contribution we apply response surface method to C1, its
first application in this area of lowered-symmetry Ni(II) α-diimine catalysts.
C1, bearing 2,6-diisopropylaniline on one side and 2,4,6-trimethylaniline on the other
side, has been previously reported [7], though it was not structurally characterized. Its
performance in ethylene polymerization was reported for a 30 min run at 10 bar ethylene
pressure with CC 500, diethylaluminium chloride co-catalyst and 50 °C as 5.43 kg PE
(mmol Ni)-1 h-1, i.e. 0.543 kg PE (mmol Ni)-1 h-1 bar-1, whereas switching to CC 2000 and MAO
co-catalyst gave a value of 0.715 kg PE mmol-1 Ni h-1 bar-1 [7]. These were the only two sets of
conditions in that report. The performance of this catalyst in ethylene polymerization with MAO
over a relatively wide range of experimental factors, especially CC, was investigated. The effect
of factor variations and also their interactions on responses (activity, MW and crystallinity) were
studied and visualized via RSM. Furthermore, a slightly different synthesis of the complex is
reported, as is its crystallization and full structural characterization by single-crystal X-ray
diffraction.
Experimental
Materials
All manipulations of air or moisture sensitive compounds were carried out under an inert
atmosphere using standard Schlenk line technique. Polymerization grade ethylene with high
purity was obtained from Bandar Imam Petrochemical Corporation (BIPC, Iran) and was further
purified by passage through an oxygen/moisture trap. Toluene for polymerization was obtained
3

from Bandar Imam Petrochemical Corporation (BIPC, Iran) and was refluxed in the presence of
calcium hydride and distilled prior to use (water content was lower than 5 ppm by Karl–Fischer
titration). Tetrahydrofuran (THF) and diethylether were purchased from Merck (Germany) and
distilled from sodium/benzophenone prior to use. Dichloromethane (DCM) was purchased from
Fluka (Switzerland) and dried over CaH2. MAO was purchased from Aldrich (USA) as a 10 wt%
solution in toluene and used as received. Acenaphthenequinone and ethylene glycol dimethyl
ether nickel(II) bromide complex ((DME)NiBr2) were purchased from Aldrich and used without
purification. 2,6-Diisopropylaniline and 2,4,6-trimethylaniline were purchased from Aldrich and
refluxed over calcium hydride prior to use.
Synthesis
Synthesis of ((2,6-diisopropylphenyl)imino)-acenaphthenone (A1). In modification of the
literature procedures [7, 19] an ethanol (80 mL) solution of acenaphthenequinone (4 g, 0.02 mol)
was treated with 1 mL of formic acid, followed by slow, dropwise addition (over approx. 8 h) of
a solution of 2,6-diisopropylaniline (3.76 mL, 0.02 mol) in 50 mL of ethanol. The resulting
mixture was stirred at 60 °C (over 8 h), cooled to room temperature and filtered to remove
unreacted acenaphthenequinone. The filtrate was cooled to −10 °C overnight. The red-orange
solid that deposited was filtered, washed with ether and dried to yield 3.6 g (45%) of product.
1H NMR (250 MHz, CDCl3, δ) 8.18-8.22 (dd, 2H, acenaphthene protons), 8.0 (d, 1H,
acenaphthene protons), 7.84 (t, 1H, acenaphthene protons), 7.42 (t, 1H, acenaphthene protons),
7.27 (m, 3H, aromatic protons), 6.61 (d, 1H, acenaphthene protons), 2.38 (hep, 2H, CH(CH3)2),
1.15, 0.82 (2d, 12H, CH(CH3)2). 13C NMR (100 MHz, CDCl3, δ selected resonances) 189.87
(C=O), 160.52 (C=N), 28.31 (CH(CH3)2), 23.39, 23.12 (CH(CH3)2). IR (KBr): 1425.41, 1652.33
cm-1 (C=N), 1724.47 cm-1 (C=O). Its spectroscopic data matched that given in the literature [7].
Synthesis of ((2,6-diisopropylpheny)imino)-((2,4,6-trimethylphenyl)imino)-acenaphthene
(A2). In modification of a literature procedure [7], a solution of A1 (1 g, 3.10 mol), 2,4,6-
trimethylaniline (0.54 mL, 3.36 mol) and a catalytic amount of p-toluenesulfonic acid in THF
(60 mL) was refluxed for 17 h in the presence of molecular sieves (4 Å). The mixture was
filtered, concentrated in vacuo and the resulting oily compound was dissolved in
dichloromethane (40 mL). This solution was washed twice with 1 M NaOH (20 mL) and three
4

times with saturated NaCl solution (20 mL). The solution was dried on MgSO4, filtered and
concentrated, yielding a red powder. Yield: 51%. The crude product was dried and purified by
column chromatography (ethyl acetate:hexane = 1:4, including 2% triethylamine, silica gel) to
afford 0.5 g of red solid (m.p.: 258-260 °C, yield = 39%).
1H NMR (250 MHz, CDCl3, δ) 7.86-7.91 (dd, 2H, acenaphthene protons), 7.33-7.44 (2pst,
2H, acenaphthene protons), 7.26 (m, 3H, aromatic protons), 6.98 (s, 2H, aromatic protons), 6.75
(d, 1H, acenaphthene protons), 6.62 (d, 1H, acenaphthene protons), 3.0 (hep, 2H, CH(CH3)2),
2.38 (s, 3H, para-CH3), 2.21 (s, 6H, meta-CH3), 1.21, 0.96 (2d, 12H, CH(CH3)2. 13C NMR
(100 MHz, CDCl3, selected resonances δ) 161.50, 160.90 (C=N), 28.41 (CH(CH3)2), 23.27,
22.95 (CH(CH3)2), 20.81, 17.53 (CH3). IR (KBr): 1425.41, 1652.33 cm-1 (C=N). Spectroscopic
data matched that reported [7].
Synthesis of [2,6-diisopropylpheny)imino)-((2,4,6-trimethylphenyl)imino)-acenaphthene
nickel(II) dibromide (C1). In a slight modification of the literature procedure, which was
recently reported [7], (DME)NiBr2 (0.12 g, 0.40 mmol), A2 (0.20 g, 0.44 mmol) and CH2Cl2 (12
mL) were placed in a Schlenk flask under nitrogen and the reaction mixture was stirred at room
temperature for 24 h. The solvent was evaporated in vacuo. The solid was washed three times
with Et2O (10 mL) and dried in vacuo to afford red-brown powder (yield: 80%). Anal. calcd for
C33H34N2Br2Ni: C, 58.52; H, 5.02; N, 4.14. Found: C, 58.37; H, 4.86; N, 4.09. A crystalline
sample of C1 suitable for X-ray diffraction studies was obtained by vapor diffusion of hexane
into a DCM solution under non anhydrous conditions.
Crystallographic data
Crystallographic data for the structure of C1 has been deposited at the Cambridge
Crystallographic Data Centre, CCDC 1023990, deposition@ccdc.cam.ac.uk,
http://www.ccdc.cam.ac.uk/deposit, Telephone: (44) 01223 762910, Facsimile: (44) 01223
336033, Postal Address: CCDC, 12 Union Road, Cambridge CB2 1EZ, UK.
Ethylene polymerization procedure
The polymerizations were carried out in a 300 mL stainless steel reactor containing systems for
full control of temperature and reaction pressure. The reactor was repeatedly evacuated and
5

refilled with argon and finally filled with ethylene gas (1 bar). It was then charged with toluene
(150 mL) and MAO and its temperature was set. The catalyst was dissolved in 3 mL of toluene
under a dry nitrogen atmosphere and transferred into the reactor. Immediately the reactor was
pressurized and then the solution stirred for 1 h. The polymerization was terminated by venting
unreacted monomer and adding 10 vol. % HCl/methanol solution (80 CC). The polymer was
washed with an excess of methanol and dried in vacuum at 40 °C overnight.
Response surface experimental design
The ethylene polymerization performance of C1 was statistically analyzed through the response
surface method, based on Box-Behnken design. The factors or independent variables were
temperature (xT: levels 10, 30, 50 °C), ethylene pressure (xP: levels 3, 5, 7 bar) and co-catalyst to
catalyst ratio (CC) (xC: levels 100, 1550, 3000), whereas the response or independent variable
were catalyst activity, MW and crystallinity. The temperature and pressure were selected based
on previous reported results with nickel(II) systems [7, 19, 26-32]. The levels of CC were
selected based on a group of polymerization runs, detailed in the ethylene polymerization section.
The experimental plan generated using Minitab®15 software involved 13 runs. Also, to provide
an estimate of the experimental error in the process and achieve more precise estimates of the
factor effects, the center point, that is, the midpoint between the high and low levels, was
replicated twice. Therefore, in total, the software suggested 15 test runs, as detailed in the
ethylene polymerization section.
Characterization
The 1H and 13C NMR data of the organic compounds were obtained on a Bruker (Germany)
Avance 250 MHz spectrometer at ambient temperature and CDCl3 as a solvent.
IR spectra were recorded on a Perkin-Elmer (USA) SpectrumRX1 Fourier transform
infrared spectrometer using Nujol mulls between KBr plates. Elemental analysis measurements
were performed on a CHNSO Elementar Analyzer (Germany) Vario ELIII. Single crystals were
mounted in perfluoropolyether oil into an Oxford Instruments Cryostream (UK) 700. Diffraction
measurements were performed on an Oxford Diffraction X-Calibur 2 diffractometer using
graphite-monochromated Mo-Kα radiation, and the data were collected and processed by
CrysAlis PRO and CrysAlis RED (Oxford Diffraction Ltd., Abingdon, UK, 2007). The structure
6

was solved using SHELXS86 and refined with SHELXL [33]. The DSC measurements were
performed on a Mettler-Toledo (Switzerland) model 822e instrument, interfaced to a digital
computer equipped with Star E 9.01 software (Sencor FRS5) at a heating/cooling rate of 10 °C
min−1. Samples were heated from room temperature to 170 °C and held there for 2 min, followed
by cooling to -120 °C. Finally they were reheated to 170 °C and the melting point and
crystallinity were determined according to the results obtained from the final step. The molecular
weights and the molecular weight distributions of the polymer samples were determined at
145 °C with a Varian (USA) PL-GPC120 high temperature chromatograph. Trichlorobenzene
was employed as the solvent and calibration was made by polystyrene standard.
Results and discussion
Synthesis and structure of the complex
The low symmetry diimine compound A2 was obtained in two steps as shown in scheme 2. The
mono aryl imino acenaphthene A1 was synthesized by condensation of the acenaphthenequinone
with 2,6-diisopropylaniline in ethanol. The resulting product was further reacted with 2,4,6-
trimethylaniline in THF in the presence of a catalytic amount of p-toluenesulfonic acid and
molecular sieves (4 Å) to remove the water formed during the condensation reaction. The crude
product was purified using column chromatography, followed by recrystallization to obtain bis
aryl imino acenaphthene A2. Its synthesis has also recently been independently reported by
Redshaw, Sun and co-workers [7], although solvent choice was different. We chose lower-
boiling solvents in order to lower the reaction temperature, and hence amounts of
acenaphthenequinone substituted with two identical anilines. This could have occurred in the
first step via over-reaction, or in the second step via statistical scrambling of the two anilines,
since the imine formation is in equilibrium. However, where there is a large steric mismatch
between the two anilines, this is not thought to be a serious problem, as reflected in the fact that
the other reported synthesis is in fact higher-yielding and did not require chromatography [7].
The synthesis of C1 by addition of A2 to (DME)NiBr2 is depicted in scheme 2.
Crystallization from DCM allowed isolation of the product as red-brown crystals in 80% yield.
Single crystals of C1 suitable for X-ray diffraction studies were obtained by vapor diffusion of
hexane into a DCM solution under non anhydrous conditions. The resulting molecular structure
is shown in figure 1.
7

The crystallographic parameters, crystal data and refinement details are summarized in
tables 1 and 2, respectively. The molecular structure features a distorted tetrahedral geometry,
similar to other published examples, notably C3, which differs from C1 only in the absence of
the p-Me group from one aryl. Consequently, C1 and C3 are very nearly isotypic and
isostructural [7]. This comparison is complicated somewhat by the fact that C1 was solved and
refined in P21/c, the standard setting of space group No. 14, whereas C3 was set in the alternative
P21/n. However, transposing the published cell dimensions of C3 to P21/c gives the values
a: 10.642, b: 20.891 and c: 15.788 Å, β = 124.70, remarkably close to those of C1. The cell
lengths a and b are essentially identical, as is the angle β, but there is a lengthening of
approximately 1 Å along c caused by the presence of the extra methyl group in C1, as is
expected given the orientation of the molecules in the cell approximately parallel to the c axis.
The aryl rings of the α-diimine lie nearly perpendicular to the plane formed by the metal
and coordinated nitrogens, the diisopropyl-substituted ring being closer to perpendicular (89°)
than the trimethyl one (79°), as for C3 [7]. This slight skewing of the trimethyl-substituted ring
is accompanied by a twisting of the Br2Ni plane relative to the N2Ni plane. The bond distances in
the X-ray structure of A2 reveal that the Ni(1)-N(1) is longer than Ni(1)-N(2) by 0.063(9) Å, an
amount which is on the edge of what may be considered chemically significant, most likely due
in part to the greater steric repulsion from the more heavily substituted aromatic ring. This
difference is greater, however, than that recorded for C3 (0.010(4) Å), by a statistically
significant amount [7]. Thus, it seems that there is a just-perceptible electronic structural effect in
that the electron-pushing p-Me group makes N2 a marginally better σ-donor than N1. These
marginal structural changes result in detectable variations in polymerization activity and
branching rate, however (vide infra). To summarize, catalyst C1 shares a pseudo-tetrahedral
structure [7] with C3 and C4 and several other NiBr2 complexes of diimines, but not all: even
restricting comparisons to other cases bearing an acenaphthene backbone, there are several cases
where dimeric five-coordinate structures with bridging bromides are found, even among bulkier
cases such as C2 [34].
Ethylene polymerization
Selection of CC levels
8

The catalytic behavior of C1 was evaluated in the presence of MAO, as for most previously
reported nickel(II) α-diimine based catalytic systems. It has been reported that low-symmetry
catalysts such as C3 and C4, just as for other nickel(II) α–diimine complexes such as C2,
performed better (higher activity, lower branching) at lower temperature and higher pressure [3].
Most previous reports include an assessment of the effect of CC on activity via trials of limited
ranges [4, 6-9]. For example, Guo, Sun and Glaser report an optimum of CC equal to 3000 from
the range 2000-3500 [9]. In other similar cases [7] a range of 1500 to 2500 produced an optimum
of 2000. Much lower ratios have occasionally been found possible, especially with non-MAO
activators [28]. Therefore, it is considered that an important parameter to vary in assessing any
polyethylene precatalysts is CC. For C1 the activity was evaluated at constant levels of
temperature and pressure (30 °C and 7 bar) while varying CC from 100 to 3000. Figure 2
illustrates the ethylene polymerization activity of C1 versus CC. It shows that although variation
of activity with CC is not especially dramatic at lower levels of CC (1000-3000), it is at higher
levels. Hence, a wide range of CC was selected for evaluation of C1 in ethylene polymerization.
Estimated regression models
The results of the performance evaluation of C1 according to the experimental plan are shown in
table 3. The behavior of the system was explained by the following second degree polynomial
equation (Eq. 1):
ꢅ
ꢅ
ꢅ
ꢀ = ꢁꢂ + ꢃ ꢁꢄ ꢈꢄ + ꢃ ꢁꢄꢄꢈꢉ + ꢃ ꢁꢄꢊꢈꢄꢈ
(1)
ꢊ
ꢄ
ꢄꢆꢇ
ꢄꢆꢇ
ꢄꢋꢊ
where R denotes the predicted response of the process, xi refers to the coded factors (temperature,
xT, ethylene pressure, xP and co-catalyst to catalyst ratio, xC) and b0, bi, bii, bij are regression
coefficients. Reducing Equation (1) led to three partial quadratic models for the responses of
activity (RA), weight-average molecular weight (RMW) and crystallinity (%) (RXtl), as shown in
Eqs. 2–4, respectively.
9

ꢀꢌ = 3182 − 338.8ꢈꢍ − 460.8 ꢈꢎ − 197.1ꢈꢏ + 690.6 ꢈꢍꢉ − 672.8ꢈꢏꢉ − 249.2ꢈꢍꢈꢎ − 137.5ꢈꢎꢈꢏ
ꢀꢐꢑ = 189456 − 202451ꢈꢍ + 33216 ꢈꢎ − 36033ꢈꢏ + 46421 ꢈꢍꢉ + 58040ꢈꢎꢉ + 26641ꢈꢏꢉ − 78017ꢈꢎꢈꢏ (3)
ꢀꢒꢓꢔ = 27.27 − 13.15ꢈꢍ + 3.94 ꢈꢎ − 4.96ꢈꢏ + 1.73 ꢈꢍꢉ + 4.90ꢈꢏꢉ − 2.77ꢈꢍꢈꢎ − 4.65ꢈꢎꢈꢏ
(2)
(4)
Details of the statistical basis for exclusion of non significant terms (p-values higher than 0.05)
and evaluation of the goodness-of-fit of the reduced regression models (F-value and analysis of
variance (ANOVA) table) is detailed in Supplementary Material.
The estimated regression coefficients for the above models are presented in table 4.
Table 5 shows the ANOVA table at a 95% confidence level. In the activity regression model
2
(Eq. 2), the significant coefficients are xt, xp, xt2 and xc because their p-values are less than 0.05
(table 4). The suggested model for the activity variations was deemed adequate because not only
was the F value for the regression model greater than Ftab but also the F value for the lack-of-fit
was less than Ftab (table 4). Probability plots of residual distribution for the regression models are
presented in figure 3. According to figure 3a, the p-value for the Anderson-Darling test is higher
than the chosen significance level (usually 0.05), which supports the normal distribution of
residuals and adequacy of the activity fitted model [35-38].
2
In the MW regression model (Eq. 3), all of the coefficients are significant except xc and
xt2. The fitted model was adequate because the F value for the regression model was more than
Ftab and F value for the lack-of-fit was less than Ftab (table 4). Figure 3b indicates the model has
an adequate goodness-of-fit.
In the crystallinity regression model (Eq. 4), significant coefficient-related terms include
2
xt, xcc, xp, xc and xpxc. The fitted model was adequate according to F value data in table 4. Also
according to figure 3c, the residuals follow a normal distribution. The above models could be
shown as three-dimensional surface plots, representing the trend of response variations with
factors. Figure 4 shows the surface and contour plots of activity variation of MAO activated
catalyst C1 according to Eq. 2. Figure 4(a,c) shows that there is a second order variation of
activity with temperature at all levels of CC and pressure. It depicts that activity passes through a
minimum at 30 °C but the maximum activity appears at 10 °C. This is surprising, since in other
Ni(II) acenaphthenediimine-based polyethylene catalysts, though at higher temperatures activity
is impeded by reduced ethylene solubility and catalyst deactivation, in the low temperature
regime normal Arrhenius rate increase with temperature has been observed, a fact attributed to
10

lower activation energies for monomer propagation than for catalyst deactivation [24, 25],
leading to optima commonly in the 20-30 °C temperature range. The fact that the optimum
temperature is as low as 10 °C for C1 suggests that it is more susceptible than some catalysts to
deactivation processes. If those processes were reductive in nature, then the electron-pushing
p-Me group on one aryl may account for promotion of such a reductive deactivation, through
increased donation to Ni(II). This may explain the lower-temperature optimum than that found
for otherwise identical C3 (20 °C) [19]. Conversely, lower optimum temperatures have been
found for phenanthrene-based ligands, where the larger polyaromatic core facilitates ligand-
reduction processes [16]. Furthermore, a catalyst bearing a less-reducible -triimine ligand, also
with trimethyl-phenyl substituents, yielded a higher optimal temperature of 30 °C [26]. The idea
that deactivation processes are reductive in nature also fits well with the known reducing
character of Me3Al present in MAO, and thereby also goes some way to explaining the negative
effect of CC being too high (vide infra). It is also apparent from figure 4(b,c) that pressure has a
linear effect on activity at all temperatures, with the steepest gradient at lower temperatures due
to the lower solubility of gases at higher temperatures. At this range of monomer concentration,
the rate of polymerization is first order with respect to monomer pressure. Hence, it is instructive
to use units normalized to 1 bar pressure when comparing activities between groups or within a
set where pressure is a variable: the highest activity found under those terms was in fact that of
run 1, reaching 3.4 kg PE mmol-1 Ni h-1 at 3 bar, i.e. 1.13 kg PE mmol-1 Ni h-1 bar-1 collected at
CC equal to 1550 and 10 °C. This value compares with 0.715 kg PE (mmol Ni)-1 h-1 bar-1 for C3
at 10 bar, CC equal to 2000 and 20 °C, though that run was for 30 min rather than the full 1 h
runs used here [7]. Furthermore C1 under those conditions gave a similar value, 0.715 kg PE
mmol-1 Ni h-1 bar-1 for C3 at 10 bar, 30 min run [7]. This value would appear to be in broad
agreement with the models shown, when corrected for the shorter run time. An even higher
pressure-corrected value has been reported for C3 at 15 min runtime, 3.15 kg PE (mmol Ni)-1 h-1
bar-1 at 1.25 bar [19], underlining the fact that though activity does vary linearly with pressure,
the gradient is less than unity. However, the main point to be deduced from these literature
comparisons must be that though there are small effects on activity, MW and branching rate with
the p-Me variation of C1 vs C3 and in other such cases [7, 28-30], the effects of run-time and
pressure on such values are much greater [3].
11

According to figure 4(a,b), there is a non-linear variation of activity with CC; it passes
through a maximum at CC equal to 1550. This implies that below a threshold of molar ratio of
co-catalyst to catalyst (CC < 1550), there might be a hindrance of efficient activation of catalyst,
due to slow alkylation reactions or to insufficient Al-alkyl scavenging effects. But also,
exceeding this ratio facilitates dialkylation of the nickel centers. Such nickel-dialkyl
intermediates deactivate rapidly by reductive elimination to nickel(0). Another explanation for
decreasing activity of catalyst in high levels of CC is deactivation by reaction of the imine ligand
fragments with Me3Al [39].
The maximum activity (absolute, not corrected for pressure) of C1 (4960 kgPE
(mol Ni)-1 h-1) with full 1-h run appears at 10 °C after raising ethylene pressure and CC to 7 bar
and 1550, respectively. Higher pressure runs give higher maxima [40, 41].
Figure 5 shows the surface and contour plots of MW variation for polymers prepared
with MAO activated C1. The MW decreases sharply as the temperature increases (figure 5(a,c)).
It is likely that the rate of chain transfer increases with increasing temperature [24, 39] passing
through a minimum at 5 bar. These observations probably suggest that at low levels of CC, chain
transfer both to monomer and to aluminium are slow, whereas at high levels of CC chain transfer
to Al dominates. That it should show such dependence on pressure is surprising, perhaps increase
in pressure causes an increase in rate of the associative reaction needed to facilitate chain transfer
to aluminium, until such point that the concentration of ethene inhibits coordination of
organoaluminium.
The importance of chain transfer to aluminium depends on the concentration of Me3Al,
which acts as a chain transfer agent. At low levels of CC, because of the low concentration of
Me3Al, the rate of chain transfer to aluminium is negligible but at higher levels, there is a
competition between chain transfer to aluminium and insertion of previously formed oligomer or
macromonomer [39]. The possible chain transfer and propagation pathways for this system are
shown in scheme 3 [41, 42]. Maximum MW (666342 g mol-1) is achieved at 10 °C, 7 bar and CC
equal to 100 (figure 5b). This observation is in accord with previous reports that decreasing steric
bulk in the axial sites lead to an increase in the rate of chain termination relative to the rate of
propagation [25]. It is significant to note that molecular weight distribution (PDI) of polymers
produced at 50 °C is narrower than that found at 10 °C (table 3). This could be explained by the
12

greater degree of branching and lower molecular weights of polymers produced at 50 °C, which
lead to more homogeneous systems [21].
Figure 6 shows the surface and contour plots of crystallinity variation for polymers
prepared with MAO activated C1. It is clear that a higher level of crystallinity is obtained at
lower temperatures (figure 6(a,c)), implying that the chain-walking process which results in
short-chain branches is suppressed relative to linear propagation at lower temperatures [33, 35],
in line with its higher activation energy. There is a linear variation of crystallinity with pressure
[42, 43]. In other words, the polyethylenes obtained under higher pressures are characterized by
a more linear and less branched structure due to an increased rate of trapping and insertion
relative to the rate of chain walking [32].
Within the broad range of CC studied, the variation of crystallinity is more significant at
lower levels of CC (figure 6b). For example at 10 °C and CC equal to 100, the crystallinity
increases from 40 to 63% with increasing pressure from 3 to 7 bar, but at CC equal to 3000 it is
just from 40 to 44%. This suggests that higher levels of aluminium may somewhat impede access
of ethylene to the active site, thereby favoring the branching reaction over linear monomer
enchainment.
The maximum crystallinity level (about 63%) implies that it should prove possible to
reach linear polymers (<5 branch per thousand carbons) with melt transition between 125 and
135 °C at 10 °C, 7 bar and CC equal to 100. In comparison to literature, these results show that
the electronic nature of the para substituent does not have a major effect on the branching degree
of the resulting polymers [21, 22].
Conclusion
The low-symmetry nickel(II) α-diimine C1 was structurally characterized by single-crystal X-ray
diffraction then used as catalyst precursor on ethylene polymerization in the presence of MAO.
RSM was applied to analyze the effect of polymerization conditions (temperature, monomer
pressure and CC) and their interactions on activity, MW and crystallinity of polymer. The
maximum activity of C1 was achieved at 10 °C, although there was an increase of activity from
30 to 50 °C. It revealed that for every single catalytic system there is a competition between
propagation and deactivation rates that lead to maximum productivity at a specific temperature.
For example, greater degrees of ortho-steric encumbrance, via dibenzhydryl [44] or cyclophane
13

[45] substituents, can increase optimum temperature as high as 90 °C, though in the
dibenzhydryl case the steric encumbrance is so great that activity suffers (only 0.08 kg (mmol
Ni)-1 h-1 bar-1, at 6.9 bar over a 15 min run) [44], while for the cyclophane case claims are made
of levels of activity which challenge those of the most active of early transition metal catalysts
[45, 46], 3.04 kg (mmol Ni)-1 h-1 bar-1, though this value relates to a 30 °C, 13.8 bar run
terminated after the most active first 5 minutes, rather than the full 1-hour runs reported here
which give rise to values approximately one third of this. It is likely that overall activities are
broadly comparable. Returning to the results specific to C1, the obtained surface and contour
plots showed that CC had a pronounced effect on responses. At high levels of CC, excessive
complexation of MAO with active centers led to the decrease of activity. Also there was an
increase of MW with pressure at high levels of CC, but it showed second order variation at lower
levels.
A linear increase of polymer crystallinity with pressure was observed, as noted before
[3, 26], but more intriguing is the fact that the effect was more significant at lower levels of CC.
This suggests that MAO may impede access of monomer at high CC. Moreover, the para
substituent on the arylimino ring did not majorly affect the branching content of final polymers,
in contrast to findings in other cases [30].
It can be concluded that polymer properties cannot be optimized without considering the
interaction of polymerization factors, which are critical to the ultimate catalyst performance. It
was also found that properties of final product could be controlled by tuning the reaction
conditions, a conclusion which has for nickel been known qualitatively since early work [1-3],
but which can best be visualized and modelled quantitatively using RSM [26, 27]. Further
studies on nickel-catalyzed polymerizations may benefit from similar treatment of all relevant
variables using RSM.
Supporting information
Additional supporting information may be found in the online version of this article.
Acknowledgement
14

We are grateful to Dr. Omar Alsayari for advice on synthesis of C1, growth of crystals of C1 and
helpful discussions during his stay in Manchester University. We are also grateful to Dr. James
Raftery for collection, solution and refinement of the X-ray data for C1.
References
[1]
[2]
L.K. Johnson, C.M. Killian, M. Brookhart. J. Am. Chem. Soc., 117, 6414 (1995).
M. Brookhart, L.K. Johnson, C.M. Killian, S. Mecking, D.J. Tempel. Polym Prep., 37,
254 (1996).
[3]
[4]
S.D. Ittel, L.K. Johnson, M. Brookhart. Chem. Rev., 100, 1169 (2000).
H. Liu, W. Zhao, X. Hao, C. Redshaw, W. Huang, W.H. Sun. Organometallics, 30, 2418
(2011).
[5]
[6]
[7]
[8]
[9]
M. Schmid, R. Eberhardt, M. Klinga, L. Leskelӓ, B. Rieger. Organometallics, 20, 2321
(2001).
H. Liu, W. Zhao, J. Yu, W. Yang, X. Hao, C. Redshaw, L. Chen, W.H. Sun. Catal. Sci.
Technol., 2, 415 (2012).
S. Kong, K. Song, T. Liang, C.Y. Guo, W.H. Sun, C. Redshaw. Dalton Trans., 42, 9176
(2013).
D. Jia, W. Zhang, W. Liu, L. Wang, C. Redshaw, W.H. Sun. Catal. Sci. Technol., 23,
2737 (2013).
S. Kong, C.Y. Guo, W. Yang, L. Wang, W.H. Sun, R. Glaser. J. Organomet. Chem., 725,
37 (2013).
[10] M. Helldorfer, J. Backhus, W. Milius, H.G. Alt. J. Mol. Catal. A: Chem., 193, 59 (2003).
[11] F.S. Liu, H.B. Hu, Y. Xu, L.H. Guo, S.B. Zai, K.M. Song, H.Y. Gao, L. Zhang, F.M.
Zhu, Q. Wu. Macromolecules, 42, 7789 (2009).
[12] C.S. Popeney, Z. Guan. Organometallics, 24, 1145 (2005).
[13] C.S. Popeney, Z. Guan. Macromolecules, 43, 4091 (2010).
[14] J. Liu, Y. Li, N. Hu. J. Appl. Polym. Sci., 109, 700 (2008).
[15] L.K. Johnson, S. Mecking, M. Brookhart. J. Am. Chem. Soc., 118, 267 (1996).
[16] L. Li, M. Jeon, S.Y. Kim. J. Mol. Catal. A: Chem., 303, 110 (2009).
[17] B.C. Peoples, G.D. Vega, C. Valdebenito, R. Quijada, A. Ibañez, M. Valderrama, R.
Rojas. J. Organomet. Chem., 700, 147 (2012).
15

[18] F.S. Liu, H.B. Hu, Y. Xu, L.H. Guo, S.B. Zai, K.M. Song, H.Y. Gao, L. Zhang, F.M.
Zhu, Q. Wu. Macromolecules, 42, 7789 (2009).
[19] M. Jeon, S.Y. Kim. Polym. J., 40, 409 (2008).
[20] M.M. Wegner, A.K. Ott, B. Rieger. Macromolecules, 43, 3624 (2010).
[21] G.J.P. Britovsek, V.C. Gibson, D.F. Wass. Angew. Chem., 111, 448 (1999).
[22] G. Mao, Y. Jiang, N. Li, Q. Wang, D. Zheng, M. Li, Y. Ning. Chin. Sci. Bull., 59, 2505
(2014).
[23] S. Mecking, L.K. Johnson, L. Wang, M. Brookhart. J. Am. Chem. Soc., 120, 888 (1998).
[24] L. Deng, P. Margl, T. Ziegler. J. Am. Chem. Soc., 119, 1094 (1997).
[25] L. Deng, T.K. Woo, L. Cavallo, P. Margl, T. Ziegler. J. Am. Chem. Soc., 119, 6177
(1998).
[26] N. Ghasemi-Hamedani, H. Arabi, G.H. Zohuri, F.S. Mair, A. Jolleys. J. Polym. Sci.
Part A.: Polym. Chem., 51, 1520 (2013).
[27] H. Arabi, M.S. Beheshti, M. Yousefi, N. Ghasemi Hamedani, M. Ghafelebashi. Polym.
Bull., 70, 2765 (2013).
[28] H. Liu, W. Zhao, X. Hao, C. Redshaw, W. Huang, W.H. Sun. Organometallics, 30, 2418
(2011).
[29] M. Schmid, R. Eberhardt, J. Kukral, B. Rieger. Z. Naturforsch B: Chem. Sci., 57b, 1141
(2002).
[30] J. Yuang, F. Song, J. Li, Z. Jia, F. Wang, B. Yuan. Inorg. Chim. Acta, 400, 99 (2013).
[31] S. Wenhua, W. Zhang, H. Lih, 2012, WO2012122854.
[32] O. Alsayari, Group 4 and group 10 post metallocene ethylene polymerization catalysts:
catalyst structure-polymer properties relationship. PhD Thesis, The University of
Manchester (2010).
[33] G. Sheldrick. Acta Crystallogr., A64, 112 (2008).
[34] J.O. Liimatta, B. Lofgren, M. Miettinen, M. Ahlgren, M. Haukka, T.T. Pakannen. J.
Polym. Sci. Part A.: Polym. Chem., 39, 1426 (2001); R.J. Maldanis, J.S. Wood, A.
Chandrasekaran, M.D. Rausch, J.C.W. Chien. J. Organomet. Chem., 645, 158 (2002);
H.R. Liu, P.T. Gomes, S.I. Costa, M.T. Duarte, R. Branquinho, A.C. Fernandez, J.C.W.
Chien, R.P. Singh, M.M. Marques. J. Organomet. Chem., 690, 1314 (2005); F. Wang, J.
Yuan, Q. Li, R. Tanaka, Y. Nakayama, T. Shiono. Appl. Organomet. Chem., 28, 477
16

(2014); C. Wen, S. Yuan, Q. Shi, E. Yue, D. Liu, W.–H. Sun. Organometallics, 33, 7223
(2014).
[35] I. Xiarchos, A. Jaworska, G. Trznadel. J. Membr. Sci., 321, 222 (2008).
[36] M. Ahmadi, R. Jamjah, M. Nekoomanesh, G. Zohuri, H. Arabi. Iran Polym. J., 16, 133
(2007).
[37] L.C. Ferreira Jr., P.A. Melo Jr., G.L. Crossetti, G.B. Galland, M. Nele, J.C. Pinto, Polym.
Eng. Sci., 50, 1797 (2010).
[38] M. Vatankhah, S. Pourmahdian, F. Afshar. Iran. Polym. J., 20, 897 (2011).
[39] L.C. Simon, C.P. Williams, J.B.P. Soares, R.F. De Souza. J. Mol. Catal. A: Chem., 165,
55 (2001).
[40] J.D. Azoulay, Z.A. Koretz, G. Wu, G.C. Bazan. Angew. Chem., Int. Ed., 49, 7890 (2010).
[41] D.P. Gates, S.A. Svejda, E. Oñate, C.M. Killian, L.K. Johnson, P.S. White, M. Brookhart.
Macromolecules, 33, 2320 (2000).
[42] P.H.M. Budzelaar. Comput. Mol. Sci., 2, 221 (2012).
[43] F. AlObaidi, Z. Ye, S. Zhu. Polymer, 45, 6823 (2004).
[44] J.L. Rhinehart, N.E. Mitchell, B.K. Long. ACS Catal., 4, 2501 (2014).
[45] D.H. Camacho, E.V. Salo, J.W. Ziller, Z. Guan. Angew. Chem., Int. Ed., 43, 1821 (2004).
[46] H.G. Alt, A. Koeppl. Chem. Rev., 100, 1205 (2000).
17

Table 1. Crystallographic data for unsymmetrical catalyst C1.
Empirical formula
Formula weight
Temperature
C33H34Br2N2Ni
677.15
100(2) K
Wavelength
0.71 Å
Crystal system
Space group
Monoclinic
P21/c, Z=4
Unit cell dimensions
a = 10.624(7) Å
b = 20.543(13) Å
c = 16.627(12) Å
α = 90.00°
β = 123.461(13)°
γ = 90.00°
Absorption coefficient
F(000)
3.305 mm-1
1376
Theta range for data collection
Index ranges
1.77 to 28.38°
-13≤ h ≤13, -27≤ k ≤27, -22≤ l ≤22
23971
Reflections collected
Completeness to theta
93.9%
(to theta=28.38°)
Full-matrix least-squares on F2
7137 / 24 / 350
0.611
Refinement method
Data / restraints / parameters
Goodness-of-fit on F2
Final R indices [I>2sigma(I)]
R indices (all data)
R1 = 0.0648, wR2 = 0.1076
R1 = 0.3660, wR2 = 0.1446
0.567 and -0.755 e.Å–3
Largest diff. peak and hole
18

Table 2. Selected bond distances and angles (Å, °) for C1.
Br1—Ni1
2.340(2)
2.333(3)
82.0(4)
Ni1—N2
1.989(9)
2.052(9)
104.0(3)
109.8(3)
125.23(10)
Br2—Ni1
Ni1—N1
N2—Ni1—N1
N2—Ni1—Br2
N1—Ni1—Br2
N2—Ni1—Br1
N1—Ni1—Br1
Br2—Ni1—Br1
117.4(3)
110.0(3)
19

Table 3. Performance of C1/MAO in ethylene polymerizationa.
Independent variables (factors)
Activity
MW
Experimental run
Crystallinity (%)
(kg/molNi.h)
(g/mol)
xP (bar)
xCC
PDI
xT (°C)
1
10
50
10
50
10
50
10
50
30
30
30
30
30
30
30
3
3
7
7
5
5
5
5
3
7
3
7
5
5
5
1550
1550
1550
1550
100
3400
520800
81173
2.72
2.09
2.92
2.18
2.72
2.12
2.73
2.08
3.30
2.77
2.93
3.58
2.78
3.07
2.69
35.97
16
2
3333.3
5000
.3
4
505507
68189
46.29
17.22
54.34
22.39
43.85
17.59
28.47
46.81
25.59
25.30
25.10
29.36
29.89
3936.6
3966.6
3156.5
3133.3
2363.3
2033.3
3050
5
497079
74619
6
100
7
3000
3000
100
399324
79090
8
9
171374
474409
229904
220873
190000
206182
172186
10
11
12
13
14
15
100
3000
3000
1550
1550
1550
2333.3
2800
2833.3
3066.6
3466.6
aPolymerization condition: time of polymerization=1 h, [Ni]= 3 µmol
20

Table 4. Estimated regression coefficients for various regression models.
Activity
MW
Crystallinity
Terms of the
models
Coefficient p-value
Coefficient p-value
Coefficient p-value
xt
-338.8
460.8
-197.1
690.6
-
0.018
0.004
0.117
0.004
-
-202451
33216
-36033
46421
58040
26644
-
0.000
0.006
0.002
0.282
0.013
0.094
-
-13.156
3.949
-4.960
1.732
-
0.000
0.006
0.002
0.282
-
xp
xc
xt2
2
xp
xc2
-672.8
-249.2
-
0.004
0.154
-
4.905
-2.775
-
0.013
0.094
-
xtxp
xtxc
xpxc
-
-
-137.5
0.408
-78017
0.014
-4.658
0.014
21

Table 5. ANOVA table for the regression models.
Activity
Source
MW
p-value DF
Crystallinity
p-value DF
DF
F
Ftab.
F
Ftab.
F
Ftab.
p-value
0.0
Regression
Residual error
Lack-of-fit
Pure error
Total
7
10.22 3.78
0.003
7
19.87 3.78
0.0
7
33.49 3.78
7
7
7
5
0.93
19.29 0.591
5
13.24 19.29 0.072
5
1.26
19.29 0.49
2
2
2
14
14
14
22

Figure captions
Figure 1. ORTEP representation of the crystal and molecular structure of C1 at 60% probability.
Hydrogens have been omitted for clarity.
Figure 2. Effect of co-catalyst to catalyst ratio on the activity of MAO-activated C1.
Polymerization at T=30 °C, P=7 bar, t=1 h.
Figure 3. Probability plots of residual distribution for the regression models of a) activity,
b) MW and c) crystallinity.
Figure 4. Surface and contour plots of activity variation for MAO-activated C1 as a function of
a) temperature and CC, b) pressure and CC, and c) temperature and pressure.
Figure 5. Surface and contour plots of MW variation for polymers prepared with MAO-activated
C1 as a function of a) temperature and CC, b) pressure and CC, and c) temperature and pressure.
Figure 6. Surface and contour plots of crystallinity variation for polymer produced with MAO-
activated C1 as a function of a) temperature and CC, b) pressure and CC, and c) temperature and
pressure.
23

GRAPHICAL ABSTRACT
Nona Ghasemi Hamedani, Hassan Arabi, Francis S. Mair
A low-symmetry nickel (II) α-diimine complex for homopolymerizzation of
ethylene: Study of interactive effects of polymerization parameteers
In this contribution the synthesis and characterization of a non symmetric nickel(II) α-diimine commplex [(2,4,6-
trimethylphenyl)imino)-((2,6-diisopropylphenyl)imino)-acenaphthene nickel(II) dibromide] has bbeen reported as a
catalyst precursor for ethylene polymerization. Response surface method (RSM) has been applied successfully for
investigation of operational conditions; it provides predictive models for activity, weight-average molecular weight
(Mw) and crystallinity with adequate goodness-of-fit. The resultant activities challenge the best achieved with other
unsymmetrical nickel (II) α-diimine systems. Considering the most critical factors and also their iinteractions on
catalyst performance has been provided possibility to design final polymer characteristics. Furtheer studies on nickel-
catalyzed polymerizations may benefit from similar treatment of all relevant variables using RSM
.
24

25

26

27

28