Journal of Molecular Catalysis A: Chemical 214 (2004) 207–211
Linear low density polyethylene (LLDPE) from ethylene using
TpMsNiCl (TpMs = hydridotris(3-mesitylpyrazol-1-yl)) and
Cp2 ZrCl2 as a tandem catalyst system
Luciano G. Furlan a , Fábio A. Kunrath b , Raquel S. Mauler a ,
Roberto F. de Souza a , Osvaldo L. Casagrande, Jr. a,∗
a Laboratory of Molecular Catalysis, Instituto de Quı́mica, UFRGS, Av. Bento Gonçalves, 9500 Porto Alegre, RS, 91509-900, Brazil
b Instituto de Quı́mica de Araraquara, UNESP, C.P. 355, Araraquara, SP, 14800-900, Brazil
Received 1 October 2003; received in revised form 15 December 2003; accepted 7 January 2004
Abstract
Linear low density polyethylene (LLDPE) with different branching contents were prepared from ethylene, without the addition of ␣-olefin
comonomer, using a combination of catalyst precursors {TpMs }NiCl (1) (TpMs = hydridotris(3-mesitylpyrazol-1-yl)) and Cp2 ZrCl2 (2)
activated with MAO/TMA (1:1) in toluene at 0 ◦ C and by varying the nickel loading mole fraction (xNi ). The polymerization results showed
that the turnover frequencies are strongly dependent on the xNi varying from 6.6 × 103 to 32.1 × 103 mol[C2 H4 ]/mol[Zr] h. The 13 C NMR
spectra of the copolymers showed that the branch contents of the polymers increase as the xNi increase in the medium promoting the production
of polymers with a wide range of melting point (Tm ).
© 2004 Elsevier B.V. All rights reserved.
Keywords: LLDPE; Ethylene copolymerization; Nickel and zirconocenes catalysts; Tandem catalyst system
1. Introduction F5 )3 )O-2 -P,O]Ni(3 -CH2 C6 H5 ). More recently, Marks
and co-workers introduced a novel binary catalyst system
The synthesis of linear low density polyethylene (LLDPE) comprising [(5 -3-ethylindenyl)Me2 Si(1 -NCMe3 )]ZrMe2
is a matter of both technical and academic interest [1]. Typ- and [(5 -C5 Me4 )Me2 Si(1 -NCMe3 )]TiMe2 activated by a
ically, LLDPE is produced when ethylene is copolymerized binuclear activator [5].
with an ␣-olefin such as 1-butene or 1-hexene utilizing In the recent years, we have been interested in the devel-
conventional heterogeneous Ziegler-Natta and metallocene opment of the homogeneous binary catalyst systems aiming
catalysts [2]. One of the most important limitations of at produce polymeric materials with interesting properties
the current processes for LLDPE production is the cost [6]. Our initial studies have revealed that the turnover fre-
of the ␣-olefin used as comonomer. However, an alterna- quency (TOF) as well as the properties of the polymer are
tive method that has recently received attention consists strongly influenced by the polymerization parameters such
in the use of a binary system where one catalyst produces as metal mole fraction, solvent, cocatalyst, and temperature
1-alkenes in situ and the another one polymerizes ethylene of polymerization. Herein, we report the combination of
and incorporates the 1-alkene into the growing chain [3]. TpMs NiCl [TpMs = HB(3-mesityl-pyrazolyl)3 − ], which is
This process has the main advantage of requiring only ethy- highly selective ethylene oligomerization catalyst [7], with
lene as a monomer. Early studies in this field developed by standard metallocene Cp2 ZrCl2 , that copolymerizes ethy-
Komon and Bazan [4] involves the use of homogeneous lene with an ␣-olefin, in order to produce linear polyethy-
binary catalyst system comprising {[(5 -C5 Me4 )SiMe2 (1 - lene with short chain branches. The branching content on
NCMe3 )]TiMe}{MeB(C6 F5 )3 }/[(C6 H5 ) 2 -PC6 H4 C(OB(C6 the copolymer varies with the mole fraction of nickel (xNi )1
in the catalyst mixture.
∗ Corresponding author. Tel.: +55-51-3316-7220;
fax: +55-51-3316-7304. 1 x
Ni is defined as the nickel mole fraction calculated from [Ni]/([Ni]
E-mail address: osvaldo@[Link] (O.L. Casagrande Jr.). + [Zr]).
1381-1169/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/[Link].2004.01.002
208 L.G. Furlan et al. / Journal of Molecular Catalysis A: Chemical 214 (2004) 207–211
2. Experimental of 4.0 s. Samples solutions of the polymer were prepared
in o-dichlorobenzene and benzene-d6 (20 vol.%) in a 5 mm
2.1. Materials NMR tube.
TpMs NiCl (1) was synthesized following procedures de-
scribed in the literature [7]. Cp2 ZrCl2 (2) was purchased 3. Results and discussion
from Aldrich and used as received. Ethylene, provided by
White Martins Company, and argon were deoxygenated and The ethylene polymerization (or oligomerization) reac-
dried through columns of BTS (BASF) and activated molec- tions were carried out using TpMs NiCl (1) and Cp2 ZrCl2 (2)
ular sieves (3 Å) prior to use. Methylaluminoxane (MAO) (Plate 1) in toluene at 0 ◦ C using a mixture of MAO/TMA
(Witco, 5.21 wt.% Al, in toluene, average molar mass (1:1) as cocatalyst. Table 1 shows the results of polymeriza-
900 g mol−1 ) and trimethyl aluminum (TMA) (Aldrich, so- tion runs by varying the nickel loading mole fraction (xNi ).
lution 10 wt.% in hexane) were used as received. Toluene Previously studies have demonstrated that 1 shows high se-
was refluxed and distilled over sodium diphenylketyl com- lectivity in 1-butene (dimers: 86%; 1-butene: 78%) with
plex prior to use. turnover frequency of 5.4 × 103 h−1 for ethylene oligomer-
ization reactions performed in toluene at 0 ◦ C in the pres-
2.2. Polymerization procedures ence of a mixture of MAO/TMA (1:1). Under identical
polymerization conditions 2 showed turnover frequency of
All polymerization reactions were performed in a double- 7.9 × 103 mol[C2 H4 ]/mol[Zr] h, and produced linear high
walled glass reactor (120 ml) equipped with magnetic stir- density polyethylene.
ring, thermocouple temperature control, and a continuous In order to produce branched polyethylene, we carried
ethylene feed (1.1 atm). Under an argon atmosphere the out the polymerization of ethylene using the combination
proper amounts of toluene and MAO/TMA solution were of 1 and 2 under identical polymerization conditions stud-
introduced sequentially, and then the system was saturated ied for the systems employing the catalysts separately.
with ethylene. After complete saturation with ethylene at at- Table 1 shows the data concerning to the performance of
mospheric pressure and thermal equilibration of the system, this catalytic system. Polymerization runs carried out vary-
the polymerization reactions were started by adding solu- ing xNi showed that the turnover frequencies are strongly
tions of 1 and/or 2 to the reactor. The total volume of the dependent on this parameter, varying from 6.6 × 103 to
reaction mixtures was 60 ml for all polymerizations. The 32.1 × 103 mol[C2 H4 ]/mol[Zr] h. Scheme 1 shows a plau-
polymerizations are quenched with acetone, filtered and the sible tandem catalytic mechanism using a combination of 1
polymer washed with acidic ethanol, then water and ethanol, and 2. The mechanism is composed by two complementary
and dried in a vacuum oven at 40 ◦ C for 12 h. On the basis catalytic cycles characterized by ethylene dimerization, and
of the results of multiple runs, we estimate the accuracy of the copolymerization process. In the first cycle, we assumed
the productivities to ±8%. that the active species is a neutral TpMs NiH complex that
promotes the production of 1-butene via -hydrogen elim-
2.3. Polymer characterization ination in a similar way described by the nickel oligomer-
ization catalysts [8]. Once the 1-alkene has been generated
Melting temperatures were determined by means of dif- the copolymerization reaction takes place by insertion of
ferential scanning calorimetry (DSC) with a thermal analy- 1-butene into the growing polyethylene chain (P) at the
sis instruments DSC-2010 using a heating rate of 10 ◦ C/min Cp2 Zr(P)+ catalytic specie.
after twice previous heating to 190 ◦ C and cooling to 40 ◦ C
at 10 ◦ C/min. The molecular weight (Mw ) was evaluated
by gel permeation chromatography (GPC) with the wa-
ters 150CV system equipped with three columns Styragel
HT3, HT4, and HT6 (103 ,104 , and 106 Å, respectively)
and a refractive index detector. Analyses were undertaken
using 1,2,4-trichlorobenzene as solvent (with 0.5 g/l of
Irganox 10/10 as antioxidant) at 140 ◦ C and the Mw were
calculated using a universal calibration curve built with
polystyrene standards (American Polymer Standard Corpo-
ration) and checked with polyethylene and polypropylene
known samples. The ethyl and butyl branch content in the
LLDPE samples were determined by NMR spectroscopy.
The 13 C NMR spectra were obtained with a Varian Inova
300 spectrometer operating at 75 MHz and at 90 ◦ C, with
a 71.7◦ flip angle, an acquisition time of 1.5 s and delay Plate 1.
L.G. Furlan et al. / Journal of Molecular Catalysis A: Chemical 214 (2004) 207–211 209
Table 1
Results of ethylene polymerization using a combination of TpMs NiCl (1) and Cp2 ZrCl2 (2)a
Entry xNi b Polymer TOF2 Tm (◦ C)c χ (%) Mw (×10−3 ) Mw /Mn Branchingd
yield (g) (×10−3 )c
Ethyl Butyl Total
1 0.00 1.10 7.9 138 37 570 2.2 – – –
2 0.20 0.74 6.6 132 29 535 5.5 – – –
3 0.50 0.70 10.0 121 22 408 4.1 12.5 0 12.5
4 0.80 0.37 13.2 113 17 226 4.5 17.7 2.1 19.8
5 0.90 0.30 21.4 100/109 20 312 4.3 20.0 2.6 22.6
6 0.98 0.09 32.1 98/118/128129 7 237 4.2 23.3 3.5 26.8
a Polymerization conditions: Glass-reactor (120 ml), [M] = 5 mols, toluene = 60 ml, polymerization time = 1 h, [Al]/[M] = 200, P
C2 H4 = 1.1 atm,
MAO/TMA (1:1) as activator.
Ni = [Ni]/([Ni] + [Zr]).
b x
c mol[C H ]/mol[Zr] h.
2 4
d Number of branches by 1000 carbon atoms of the backbone.
Cp2ZrCl2
poly(ethylene-co-1-butene)
activation/MAO
Cp2ZrH
TpMsNi
H
TpMsNi Cp2Zr Cp2Zr
TpMsNi
n
H H
TpMsNi TpMsNi Cp2Zr
Cp2Zr
n
n
TpMsNiH Cp2Zr
n
activation/MAO
TpMsNiCl
Dimerization process Copolymerization process
Scheme 1. Plausible tandem catalytic mechanism to produce copolymer from ethylene using a combination of 1 and 2.
The influence of xNi and temperature on copolymer prop- 0.80 (entry 4, 226 × 103 g mol−1 ). Furthermore, no signifi-
erties have been evaluated by means of differential scanning cant changes on Mw were observed for higher nickel mole
calorimetry, gel permeation chromatography, and 13 C NMR fraction (xNi ≥ 0.90) as can be seen comparing entrys 4–6.
spectroscopy. The results are summarized in Table 1. The polyethylene produced using exclusively 2 showed
The GPC results show that the molecular weight (Mw ) of melting temperature (Tm ) of 138 ◦ C and crystallinity of 37%.
the polymers is dependent on xNi . In all cases the GPC curves Conversely, the use of xNi as high as 0.98 (entry 6) produced
displayed monomodal molecular weight distributions with copolymer with Tm of 98 ◦ C. It should be pointed out that
polydispersities varying from 2.2 to 5.5. The homopolymer the presence of xNi up to 0.20 yielded copolymers with broad
made by 2 showed molecular weight of 570 × 103 g mol−1 . melting transitions suggesting the formation of copolymers
On the other hand, as the xNi increases in the medium, the with different fractions of comonomer/short chain branches
Mw is reduced gradually reaching a lowest value for xNi = (Fig. 1). In the specific case of xNi = 0.98, it was observed
the formation of copolymer with three melting peaks.
2 Turnover frequency has been expressed in terms of [Zr] assuming The melting temperatures of the copolymers obtained with
that it determines the production of branched polyethylene. increasing the nickel content decrease almost linearly as the
210 L.G. Furlan et al. / Journal of Molecular Catalysis A: Chemical 214 (2004) 207–211
-2
-4
heat flow (w/g)
-6
xNi = 0.00
-8 xNi = 0.20
xNi = 0.50 xNi = 0.90
xNi = 0.80
xNi = 0.90
-10 xNi = 0.98 xNi = 0.80
60 80 100 120 140 160
xNi = 0.50
0
Temperature ( C) 40 35 30 25 10
20 15
Fig. 3. 13 C {1 H} NMR spectra of copolymers obtained varying xNi .
Fig. 1. DSC curves of the polymers varying xNi . Polymerization reactions
performed at 0 ◦ C, [Al]/[M] = 200, and using a mixture MAO/TMA (1:1)
as activator. branches in the polymer to be determined and as shown
in Fig. 2, the percent of 1-butene/1-hexene in the polymer
correlates well against the xNi . For instance, the number
1-butene and 1-hexene incorporation increases as displayed
of branches per thousand backbone carbon atoms increases
in Fig. 2. It is well-known that due to short chain branches,
from 12.5 to 26.8 as the nickel mole fraction increases from
copolymers have a lower melting point than the ethylene
0.50 to 0.98.
homopolymer [9]. When the amount of short chain branches
In summary, we have development a novel homogeneous
increases, the melting point decreases due to the decrease
binary catalyst system able to produce linear low density
of CH2 crystallizable segments. Crystallization of such a
polyethylene using only ethylene as feedstock. The turnover
polymer is determined to a major extent, if not completely,
frequencies as well as the polymer properties are depend
by the comonomer distribution along and among the chains.
on xNi . Higher activities were reached using high nickel
The 13 C NMR spectra of the copolymers (Fig. 3) show
mole fraction. The melting points of the copolymers and the
the presence of branches due to the 1-butene and 1-hexene
branching contents show a linear correlation with respect to
incorporation in the polymer chain [4a,10]. In these sam-
xNi . Further investigations into the reactivity of this catalytic
ples, it was observed one peak at 30.0 ppm, which is charac-
system will be the subject of a future report.
teristic for the linear chain (CH2 )n , and peaks at 11.1, 26.5,
30.5, 33.8, and 39.4 ppm, attributed to the presence of ethyl
branches, besides the incidence of much smaller signals cor-
Acknowledgements
responding to the butyl branches. The simplicity of poly-
mer structure allows for the percentage of ethyl and butyl
We are grateful to the CNPq for the fellowships,
CTPETRO-CNPq, CAPES, and FAPERGS for financial
support of this research.
2.4
130
References
1.8
% Branching
120
[1] (a) A.K. Kulshrestha, S. Talapatra, in: C. Vasile (Ed.), Handbook of
Tm ( C)
1.2 Polyolefins, Marcel Dekker, New York, 2000, pp. 1–70;
0
(b) D.E. James, in: H.F. Marks, N.M. Bikales, C.G. Overberger, G.
110
Menges (Eds.), Encyclopedia of Polymer Science and Engineering,
0.6 Wiley-Interscience, vol. 6, New York, 1985, pp. 429–454.
[2] (a) A.L. McKnight, R.M. Waymouth, Chem. Rev. 98 (1998) 2587;
100 (b) O. Olabisi, M. Atiqullah, W. Kaminsky, Rev. Macromol. Chem.
0.0
Phys. C37 (1997) 519;
0.0 0.2 0.4 0.6 0.8 1.0 (c) K. Soga, T. Shiono, Prog. Polym. Sci. 22 (1997) 1503;
xNi (d) B.A. Krentsel, Y.V. Kissin, V.I. Kleiner, L.L. Stotskaya, Polymers
and Copolymers of Higher ␣-Olefins, Hanser/Gardner Publications
Fig. 2. Melting point and branching content as a function of xNi of the Inc., Cincinnati, 1997, p. 243;
copolymers. Polymerization reactions performed at 0 ◦ C, [Al]/[M] = 200, (e) R. Quijada, G.B. Galland, R.S. Mauler, Macromol. Chem. Phys.
and using a mixture MAO/TMA (1:1) as activator. 197 (1996) 3091;
L.G. Furlan et al. / Journal of Molecular Catalysis A: Chemical 214 (2004) 207–211 211
(f) P. Lehmus, O. Härkki, R. Leino, H.J.G. Luttikhedde, K.H. Näs- [6] (a) F.A. Kunrath, R.F. de Souza, O.L. Casagrande Jr., Macromol.
man, J.V. Seppälä, Macromol. Chem. Phys. 199 (1998) 1965; Rapid. Commun. 121 (2000) 277;
(g) J. Koivumäki, G. Fink, J.V. Seppälä, Macromolecules 27 (1994) (b) F.F. Mota, R.F. de Souza, R.S. Mauler, O.L. Casagrande Jr.,
6254. Macromol. Chem Phys. 202 (2001) 1016;
[3] (a) Z.J.A. Komon, G.C. Bazan, Macromol. Rapid Commun. 22 (2001) (c) F.A. Kunrath, R.S. Mauler, R.F. de Souza, O.L. Casagrande Jr.,
467; Macromol. Chem Phys. 203 (2002) 2058;
(b) R.F. de Souza, O.L. Casagrande Jr., Macromol. Rapid Commun. (d) F.F. Mota, R.F. de Souza, R.S. Mauler, O.L. Casagrande Jr.,
22 (2001) 1293; Polymer 44 (2003) 4127.
(c) O. Sperber, W. Kaminsky. Macromolecules 36 (2003) 9014; [7] F.A. Kunrath, R.F. de Souza, R.S. Mauler, O.L. Casagrande,
(d) K. Musikabhumma, T.P. Spaniol, J. Okuda, J. Polym. Sci. Part Organometallics 22 (2003) 4739.
A: Polym. Chem. 41 (2003) 528. [8] G. W. Parshall, S. D. Ittel, Homogeneous Catalysis, second ed.,
[4] (a) Z.J.A. Komon, X. Bu, G.C. Bazan, J. Am. Chem. Soc. 122 (2000) Wiley-Interscience, New York, 1992, p. 68.
1830; [9] W. Kaminsky, M. Arndt, Adv. Polym. Sci. 127 (1997) 143;
(b) Z.J.A. Komon, X. Bu, G.C. Bazan, J. Am. Chem. Soc. 122 H.H. Brintzinger, D. Fischer, R.. Mülhaupt, B. Rieger, R.M. Way-
(2000) 12379. mouth, Angew. Chem., Int. Ed. Engl. 34 (1995) 1143.
[5] G.P. Abramo, L. Li, T.J. Marks, J. Am. Chem. Soc. 124 (2002) [10] G.B. Galland, R.F. de Souza, R.S. Mauler, F.F. Nunes, Macro-
13966. molecules 32 (1999) 1620.