Gradual Addition of a Comonomer to Continuous Copolymerization
Mechanistic study on comonomer effect in ethylene/1-hexene copolymerization with TiCl4/MgCl2 model Ziegler-Natta catalysts
Abstract
Two MgCl2-supported Ziegler-Natta model catalysts were prepared by contacting activated MgCl2 with deficient or excess amount of TiCl4. Ethylene/1-hexene copolymerization with the catalysts was conducted under different 1-hexene feed, and active center concentration of the reaction system was determined by quench-labeling method using thiophene-2-carbonyl chloride as the quencher. The catalytic activity was only moderately enhanced (increment <50%) by adding 1-hexene in ethylene polymerization with the model catalyst (Cat-1) of 0.1 wt% Ti content, and the active center ratio ([C*]/[Ti]) was only slightly increased (from 64 to 72 mol%). The comonomer activation effect in Cat-1 catalyzed copolymerization was mainly attributed to increase of apparent propagation rate constant by the comonomer, which can be explained by reduction of mass transfer barrier in the polymer/catalyst particles because of larger monomer diffusion coefficients in the copolymer phase than in the more crystalline polyethylene phase. In contrast, the activity was nearly tripled by adding 1-hexene when the catalyst of 1.47% Ti content (Cat-2) was used, meanwhile its [C*]/[Ti] was also tripled (from 15 to 49 mol%). The strong comonomer activation effect in Cat-2 catalyzed copolymerization was caused by marked increase of active center concentration. In this case, fragmentation of the polymer/catalyst particles was intensified by adding the comonomer, and more active centers that were originally inaccessible to the cocatalyst and monomer were exposed through the particle fragmentation. According to the results of this work, the comonomer effect in conventional supported Ziegler-Natta catalysts can be explained by the combination of physical factor (increase in diffusion coefficient) and chemical factor (exposure of inaccessible active center precursors through particle fragmentation).
Introduction
In ethylene/α-olefin copolymerization with Ziegler-Natta catalysts, metallocene catalysts or Cr-based Phillips catalysts, enhancement of reaction activity by the α-olefin comonomer (so-called "comonomer effect") has been reported and studied since the 1970's [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27]. The extent of rate enhancement by the comonomer can be as large as an order of magnitude when heterogeneous Ziegler-Natta catalysts were used [14], [20]. The rate profiles of polymerization were often changed from build-up type in ethylene homopolymerization to decay type in copolymerization [14], [17]. Besides the kinetic changes, addition of an α-olefin comonomer in ethylene polymerization with Ziegler-Natta catalysts also causes strong broadening and shifting of the polymer molecular weight distribution (MWD) [20], [26], [27]. For the great importance of polymerization kinetics and polymer chain structure in industrial polyolefin production and end uses of the polymer, in the past decades extensive efforts have been devoted to fundamental studies aiming at disclosing and clarifying the mechanism of comonomer effect. Various explanations have been proposed in the studies on comonomer effect, which can be roughly classified into the following three categories:
- (1)
-
The comonomer effect was attributed to formation of new active centers or activation of dormant centers by the comonomer. Kissin et al. proposed that Ti−C2H5 species formed by β-hydrogen shifting or chain transfer with ethylene are dormant sites because of stabilization by β-agostic interaction between the methyl hydrogen and the titanium center, and the addition of the α-olefin comonomer can reactivate these dormant sites through bypassing their formation [17], [28]. The comonomer effect observed in Cr/silica (Phillips) catalysts has been explained by activation of the active centers by coordination of α-olefin on their vacancies that will not be used in ethylene homopolymerization [23], [24], [25]. Similar model has been proposed to explain the comonomer effect in Ti and V based catalysts [9]. These models are largely based on pure chemical factors.
- (2)
-
Incorporation of the comonomer in polymer chains leads to reduced crystallinity and enhanced diffusion coefficients of the monomers in polymer/catalyst particles. This factor has been considered as the main reason for the rate enhancement in some early literatures [4], [5], [6]. This kind of explanation is based purely on physical factors.
- (3)
-
Mechanistic model taking consideration of the changes of polymer/catalyst particles during the polymerization process have been proposed, some with quite convincing experimental evidences. According to this model, changes in the polymer properties (diffusion coefficients of monomers) and morphology (porosity, pore size and pore size distribution) caused by the comonomer can lead to accelerated fragmentation of the catalyst particles, resulting in exposure of more active centers that are inaccessible to cocatalyst and monomer in the homopolymerization system [11], [14], [21], [26]. Tait et al. observed evident increased in the concentration of active sites ([C*]) with the addition of 4-methyl-1-pentene in ethylene polymerization with Ziegler-Natta catalyst [14]. Our previous study on the changes of [C*] with the comonomer concentration in ethylene-1-hexene copolymerization with a TiCl4/MgCl2 catalyst showed similar phenomenon [26]. This kind of model can be thought as based on both the physical and chemical factors.
In fact, many of the above explanations on comonomer effects have been supported by rather clear experimental evidences. Enhancement of olefin diffusion coefficient in ethylene copolymer with the decrease of the polymer's crystallinity has been confirmed in many literatures [29], [30]. This factor must be considered when the (co)polymerization rate is either moderately or severely limited by the rate of monomer diffusion in the polymer layer surrounding the active centers. As mentioned above, increase of [C*] by comonomer has been experimentally determined [14], [26]. Fragmentation of heterogeneous Ziegler-Natta catalysts particles with the increase of polymer yield has been observed by different researchers, and the observed multilevel particle structure was considered an important reason for the particle fragmentation [31], [32], [33], [34], [35]. Our previous studies on polymer particle morphology and change of [C*] in the initial stage of olefin polymerization have showed close correlations between fragmentation of the catalyst particles and increase of active center concentration [36], [37], [38]. It is likely that fragmentation of the particles of Ziegler-Natta catalysts cannot be avoided under conventional polymerization conditions, which can lead to increase of [C*] in the polymerization process. Reduction of polymer crystallinity by the comonomer may affect the mode and extent of particle fragmentation, and cause a larger extent of [C*] enhancement. However, if possible activation of dormant sites by the comonomer occurs concurrently in the same system, its effects on [C*] will be overlapped by the [C*] alteration caused by particle fragmentation. In fact, activation of active centers by coordination of α-olefin has been experimentally observed in Phillips catalyst [25]. In such cases, it will be very hard to differentiate more than one source of comonomer enhancement effects by studying a conventional heterogeneous catalyst with multilevel particle structure.
Terano et al. studied the effects of adding small amount of ethylene (the comonomer) in propylene polymerization with supported Ziegler-Natta catalysts [39]. By reducing the polymerization duration to as short as 0.2 s using the stopped-flow technique, fragmentation of the catalyst particles was largely prevented, because the polymer volume is too small to expand the cracks in the particle. Meanwhile, monomer diffusion rate in the particles will not be evidently influenced by incorporation of the comonomer, because the amount of polymer is too small to form sufficient barriers for mass transfer inside the particle. In this way the possible comonomer effects originated from physical sources were largely depressed. The stopped-flow polymerization results showed that adding ethylene comonomer caused enhancement in intrinsic reactivity (k p) of the active centers, meanwhile the concentration of active centers was not changed. This forms a solid evidence to support the mechanistic model that ethylene insertion in Ti–polymeryl bond with a 2,1-inserted last propylene unit will reactivate it from dormant state [40], [41], [42], [43]. A remaining problem with this method is the markedly lower concentration of dormant sites in such short polymerization durations (0.1–0.2 s), as it takes more than 1 s to complete accumulation of the dormant sites in similar systems [44]. This may cause under evaluation of the chemical activation effect from the comonomer. On the other hand, comonomer effects in ethylene copolymerization with small amount of α-olefin have not been studied using the stopped-flow method.
In our previous work, kinetics of ethylene polymerization with a model Ziegler-Natta catalyst of very low Ti loading (Ti content = 0.1%) was studied [45]. It was found that nearly 90% of the titanium species that were firmly adsorbed on the catalyst particles can be activated to active centers after a few minutes of polymerization. This model catalyst is thus suitable for identifying the possible chemical activation effect of α-olefin on active centers that catalyze ethylene polymerization with low reactivity, because in this system there will be only a small fraction of potential active centers to be released through particle fragmentation facilitated by the comonomer. If strong enhancement in active center's reactivity by the comonomer is observed in this system, the effect can be confidently explained by a chemical activation mechanism, because the interferences from changes in active center concentration by the comonomer will be largely depressed.
In this work, ethylene/1-hexene copolymerization (1-hexene as the minor comonomer) using a model Ziegler-Natta catalyst of very low Ti loading (Ti content = 0.1%) was compared with ethylene homopolymerization under the same conditions. Comparisons with ethylene (co)polymerization with a Ziegler-Natta catalyst of high Ti loading (Ti content = 1.47%) were also made, and mechanisms of the comonomer effects in the two catalysts are proposed based on experimental data of active site concentration and polymerization kinetics.
Section snippets
Experimental section
All the synthesis and polymerization reactions were performed under a dry nitrogen atmosphere using standard Schlenk techniques.
Effects of comonomer concentration on activity and polymer structure
Two supported Ziegler-Natta catalysts were prepared by reaction between a δ-MgCl2 support with different amount of TiCl4. The δ-MgCl2 was synthesized by the reaction of Mg powder with 1-chlorobutane. One of the catalysts, Cat-1 was prepared by adding controlled small amount of TiCl4 into a suspension of δ-MgCl2, thus its Ti content (0.1%) is very low. The other one (Cat-2) was prepared by reacting excess of TiCl4 with δ-MgCl2, so it has Ti content of 1.47%. As reported in our previous work [45]
Conclusions
In summary, no comonomer activation effect in ethylene/1-hexene copolymerization catalyzed by model supported Ziegler-Natta catalyst with very low titanium content (Cat-1) was observed at very short reaction time (<1 min). The moderate comonomer activation effect in this system for longer reaction time was mainly caused by increase of apparent propagation rate constant, which can be explained by reduction of mass transfer barrier in the polymer/catalyst particles because of larger monomer
Acknowledgements
This work has been supported by National Natural Science Foundation of China (Grant No. 21374094 and 51773178).
References (64)
- et al.
Catal. Commun.
(2015)
- et al.
J. Mol. Catal. A: Chem.
(2011)
- et al.
Chem. Eng. Sci.
(2001)
- et al.
Stud. Surf. Sci. Catal.
(1986)
- et al.
J. Catal.
(2018)
- et al.
Angew. Chem. Int. Ed.
(2016)
- et al.
Mol. Catal.
(2018)
- et al.
Chem. Eng. Sci.
(2001)
-
Eur. Polym. J.
(1998)
- et al.
J. Organomet. Chem.
(2015)
J. Catal.
(2014)
J. Catal.
(1991)
Eur. Polym. J.
(2006)
Rate enhancement effects in the prepolymerization and copolymerization of ethylene and α-olefins
(1994)
Polymer
(1976)
Ziegler Natta Catalyst and Polymerization
(1979)
Makromol. Chem.
(1984)
J. Appl. Polym. Sci.
(1987)
Makromol. Chem.
(1989)
Makromol. Chem. Macromol. Symp.
(1992)
Makromol. Chem., Rapid. Commun.
(1992)
J. Polym. Sci., Part A: Polym. Chem.
(1993)
J. Polym. Sci., Part A: Polym. Chem.
(1993)
Macromolecules
(1993)
Makromol. Chem.
(1993)
J. Appl. Polym. Sci.
(1993)
J. Appl. Polym. Sci.
(1993)
J. Polym. Sci. Part A: Polym. Chem.
(1994)
J. Polym. Sci. Part A: Polym. Chem.
(1996)
J. Polym. Sci. Part A: Polym. Chem.
(1999)
Polym. Sci. Ser. A
(2008)
Chin. J. Polym Sci.
(2008)
Cited by (28)
Recommended articles (6)
© 2018 Elsevier Inc. All rights reserved.
Source: https://www.sciencedirect.com/science/article/abs/pii/S0021951718304755
0 Response to "Gradual Addition of a Comonomer to Continuous Copolymerization"
Post a Comment