Polymerization control and fast characterization of the stereo-defect distribution of heterogeneous Ziegler–Natta isotactic polypropylene

Abstract

Adjustment of the stereo-defects distribution of isotactic polypropylene (iPP) is important for obtaining better properties of the resins in Ziegler–Natta polymerization. In this study, in order to adjust the stereo-defect distribution of iPP prepared by the Spheripol process (Basell), the cocatalyst/external donor (triethylaluminium/dicyclopentyldimethoxysilane, Al/Si) mole ratio was gradually changed from 30 to 36, accompanied with hydrogen adjustment. And we obtained a series of iPP samples (A–E). The characterization of ¹³C NMR, xylene soluble fraction (XS) showed the average isotacticity of the obtained samples are nearly same, but the results of high resolution ¹³C NMR, temperature rising elution fractionation (TREF) and xylene solvent fractionation indicated that, as the Al/Si ratio increases, the amount of high isotacticity in iPP gradually decreases, the amount of medium and low isotacticity gradually increased, the distribution of stereo-defects becomes more uniform.

A relative time-saving method, successive self-nucleation and annealing (SSA) fractionation was applied to evaluate the stereo-defect distribution of iPP. A calibration which can be used to convert SSA final melting curves to the average meso sequence length curves was constructed. A good agreement between ¹³C NMR, XS, TREF results and SSA fractionation results indicated that it is feasible by using SSA to provide quick characterize of the stereo-defect distribution of iPP.

Introduction

Since the heterogeneous Ziegler–Natta catalysts contain multiple active centers (usually including highly isospecific centers, centers of moderate isospecificity, and nearly completely aspecific centers), the prepared isotactic polypropylene (iPP) resins usually have varying distribution of stereo-defects, which not only reflects the information of catalyst and polymerization mechanism, but also determines for most part of the processability and mechanical properties. Therefore, it is important to intentionally control the distribution of stereo-defects.

One direct method for this achievement is to modify the polymerization conditions. In the past decades, great efforts have been made to understand the basic principles and action mechanisms of polymerization conditions (such as the constitution of the catalyst system, monomer concentration, the concentration of hydrogen, pressure, temperature and polymerization equipment, etc.) on the molecular structure of the product, great achievement were obtained. A large number of novel catalyst systems and polymerization technologies with high performance and high stereospecificity were developed by the researchers and companies involved in iPP [1], [2], [3], [4], [5], [6], [7], [8]. However, due to proprietary reasons, and the complex influences on the molecular structure and the complexity of interactions of polymerization conditions, the basic principle of polymerization conditions on the stereo-defect distribution of iPP is still not clear enough, and is a subject open to debate.

In heterogeneous Ziegler–Natta polymerization, catalysts comprising MgCl₂, TiCl₄ and an electron donor are widely utilized. The cocatalyst most commonly used is a trialkylauminium such as AlEt₃. For high stereospecificity the ester-containing catalysts, it normally require the presence of an additional Lewis base in polymerization, termed external donor, which is typically a second aromatic ester for catalysts containing ethyl benzoate, and an alkoxysilane for catalysts containing a phthalate ester. The performance of MgCl₂-supported catalysts in propene polymerization is strongly dependent on the cocatalyst, donors and their mole ratio present in the catalyst system. According to literatures, type and concentration of the co-catalyst and external donor greatly influence the insertion and transition manner of the monomer, the region- and stereo-selectivity of the catalyst system, and therefore determine the isotacticity, molecular weight and its distribution, production yield, sensitivity to hydrogen, etc. [1], [2], [3], [4], [5], [6], [7], [8]. This gives the researchers a light to adjust the stereo-defect distribution by changing the mole ratio of the cocatalyst and external donor, which is of great value from both economic and academic point of view. However, the feasibility of which is still unknown and needs to be investigated.

On the other hand, currently the quantitative methods for determining the distribution of tacticity comprise high resolution ¹³C NMR, solvent fractionation [9], [10], [11], [12], [13], [14] (usually an extremely time consuming and tedious work), temperature rising elution fractionation (TREF), crystallization analysis fractionation (CRYSTAF) [9], [15], [16], [17], [18], crystallization elution fractionation (CEF) and calorimetric methods [19].

TREF involves two consecutive steps, crystallization of the sample slowly on a column from a dilute solution and elution from it when raising temperature, while in CRYSTAF the elution step is skipped by measuring the change in concentration of the solution directly during the cooling phase. In both TREF and CRYSTAF, sample is fractionated on the basis of chain crystallizabilities in a dilute solution, and can give information about the inter-chain distribution of the longest crystallizable sequence (denoted as LCS) [20]. Moreover, by combination the analysis of the separate fractions, preparation (P-type) TREF and CRYSTAF can provide more molecular information of the polymer. However, both typical TREF and CRYSTAF are time-consuming, which demand solvents and specific installation. Recently, Monrabal et al. [21], [22] proposed a promising technique, CEF. They believed that using CEF, results of chemical composition distribution (CCD) of the polymer can be obtained in much shorter time. However, CEF still require expensive specific installation and is not widely applied. TREF, CRYSTAF and CEF can hardly provide quick analysis of tacticity distribution with common instrument, which is in demand greatly during the manufacturing process, especially for the large-scale production of polyolefin.

Because of this practical consideration, other less time consuming methods such as thermal fractionation technique have been developed [19]. The thermal fractionation technique mainly comprises stepwise isothermal crystallization (SIC) [23], [24], [25], segregation fractionation technique (SFT) [26], [27], [28], [29], successive self-nucleation and annealing (SSA) [19], [30], [31], etc. Among which, the SSA fractionation applies self-nucleation and annealing steps sequentially to a polymer sample to promote the potential molecular fractionation occurred during crystallization, and to encourage the unmolten crystals to anneal at each stage of the process, so that small differences in molecular structure can be magnified. The main advantages of SSA include enhanced resolution, effective molecular segregation, relatively short measurement times, free of solvent and additional molecular structural information [19]. Because of these advantages, SSA is widely used in the characterization of the microstructure of polymers, mainly for polyethylene and ethylene/α-olefin copolymers [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], aging and degradation of polyethylene [44], [45], crystallization behavior of the nano-composites [46], [47], [48] and evaluation of the compatibility of polymer blends [49], [50]. In the case of iPP, however, only few literatures had investigated its tacticity distribution using DSC-based fractionation, because iPP possesses no branched structures and is not easy to be fractionated thermally. Virkkunen, Sundholm et al. [51], [52] investigated the tacticity distribution of iPP fractions with varying isotacticity using a combination of solvent fractionation and SSA, the results obtained were compared with ¹³C NMR and TREF. They classified the fractions into three main groups according to the lamellar structure generated in SSA, and found good correspondence between the SSA melting curves and the TREF fractograms.

The aim of this study is to intentionally modify the cocatalyst/external donor (triethylaluminium/dicyclopentyldimethoxysilane, Al/Si) mole ratio in the heterogeneous Ziegler–Natta polymerization process, and to investigate the variation in stereo-defect distribution and other molecular structure of the obtained resin using a relative time-saving manner, SSA. High resolution ¹³C NMR, TREF, solvent fractionation combined with DSC and GPC, and SSA fractionation were applied. Quantitative information about the distribution of tacticity was obtained, the results of each method were compared, and the action mechanism of Al/Si ratio on the stereo-defect distribution of the iPP resin was discussed.