G. Weickert

Modeling of free radical polymerization up to high conversion. II. Development of a mathematical model

In free radical polymerization diffusion-controlled processes take place simultaneously to the normal chemical reactions. Despite extensive efforts to model such processes a consistent model for the design of a polymerization reactor has not yet been established. In this article a semiempirical model describing the conversion, polymerization degree, and molecular weight distribution (MWD) for the free radical polymerization is developed for the entire course of the reaction. The model includes the change of termination, propagation, transfer, and initiation rate. By simultaneous parameter estimation from the conversion and degrees of polymerization data the model parameters have been determined for isothermal polymerizations of methyl methacrylate (MMA) and styrene (ST). The simulation results for the conversion, degrees of polymerization, and MWD are in good accordance with experimental data for suspension and bulk polymerization of MMA and ST up to very high conversions. The influence of diffusion on the propagation rate in case of polymerization of MMA is negligible compared to the influence of the cage effect on the radical efficiency; in case of ST polymerization both effects must be included in the kinetic model. The model presented is also tested for polymerizations conducted in the presence of solvent and/or chain transfer agents.

Modeling of free radical polymerization up to high conversion. I. A method for the selection of models by simultaneous parameter estimation

A systematic quantitative method for the selection of models for the high-conversion free radical polymerization exhibiting gel and glass effects has been developed. Four representative models were selected from the literature and were compared on the basis of the same experimental data. All models describe the isothermal time-conversion data over the entire conversion range for a single type and loading of initiator well. Models that are not considering the effect of molecular weight of the polymers on the diffusion of macro radicals fail to describe the time-conversion data if the concentration of the initiator varies at the same time. By simultaneous fitting of the conversion and polymerization degree data it was shown that the Marten-Hamielec model and its extended form (Panke-Stickler-Hamielec model) were not able to describe the number average polymerization degree Pn at the final conversion, where the glass effect occurs. This occurred because both models neglect the change of the radical efficiency f in this region, which has more effect on Pn than the change of the propagation rate coefficient (see part II of this series).

Polymerization of liquid propylene with a 4th generation Ziegler-Natta catalyst-influence of temperature, hydrogen and monomer concentration and prepolymerization method on polymerization kinetics

In a batch-wise operated autoclave reactor, liquid propylene was polymerized using a 4th generation, TiCl4/MgCl2/phthalate ester-AlEt3-R2Si(OMe)2, Ziegler-Natta catalyst system. By using a calorimetric principle it was possible to measure full reaction rate versus time curves for obtaining data on polymerization kinetics, under industrially relevant conditions. The influence of polymerization temperature, the hydrogen and monomer concentration, and the prepolymerization method on reaction kinetics were investigated. A new method for prepolymerization, the so-called non-isothermal prepolymerization, is described. In this short prepolymerization procedure featuring an increasing polymerization temperature the thermal runaway on particle scale was avoided. It was shown that this prepolymerization method can relatively easily be applied to an industrial process, with the introduction of a continuous plug flow reactor, giving a narrow residence time distribution, acceptable yield-in-prepolymerization and a method for monitoring catalyst activity. Using different methods for calculating the monomer concentration at the active site of the catalyst, the influence of polymerization temperature was determined. It was shown that at high polymerization temperatures, the reaction rate is barely influenced by polymerization temperature, when no prepolymerization is used. This is ascribed to thermal runaway on particle scale of a fraction of the catalyst particles. When a prepolymerization is used, this effect disappears and thermal runaway is avoided. When systematically reducing the monomer concentration (Cm,bulk) in the bulk by replacing part of the liquid propylene by hexane, the reaction rate proved to be remarkably independent of the monomer concentration. With reducing Cm,bulk, reaction rate decreased very slowly until Cm,bulk=150 g/l. When further decreasing Cm,bulk, reaction rate dropped rapidly. The hydrogen concentration was varied over a wide range at 60°C and 70°C. For both temperatures it was shown that reaction rates increased rapidly with increasing hydrogen concentration at low hydrogen concentrations. At higher hydrogen amounts, this effect disappeared and a maximum reaction rate was found.

Liquid‐phase polymerization of propylene with a highly active Ziegler–Natta catalyst. Influence of hydrogen, cocatalyst, and electron donor on the reaction kinetics

This paper presents an experimental kinetic study of the polymerization of propylene in liquid monomer with a high activity catalyst. The influences of the concentration of hydrogen and the molar ratios of the catalyst, cocatalyst, and electron donor on the activation period, the maximum activity, the yield, and the decay behavior have been investigated at a temperature of 42°C using a relatively simple kinetic model. On the basis of the experimental data, the reaction rate has been modeled as a function of the hydrogen concentration, the molar ratio of cocatalyst and titanium, and the molar ratio of the electron donor and the cocatalyst.

Gas-phase polymerization of propylene: Reaction kinetics and molecular weight distribution

Gas-phase polymerizations have been executed at different temperatures, pressures, and hydrogen concentrations using Me2Si[Ind]2ZrCl2 / methylaluminoxane / SiO2(Pennsylvania Quarts) as a catalyst. The reaction rate curves have been described by a kinetic model, which takes into account the initially increasing polymerization rate. The monomer concentration in the polymer has been calculated with the Flory-Huggins equation. The kinetic parameters have been determined by fitting the reaction rate curves with the model. At high temperatures, pressures, and hydrogen concentrations a runaway on particle scale may occur leading to reduced polymer yields. The molecular weight and molecular weight distribution of the polymer samples could be described by a two-site model. At constant temperature the chain-transfer probability of sites 1 and 2 depends only on the hydrogen concentration divided by the monomer concentration.

Single particle modeling of gas phase propylene polymerization: viscous modulus and time scale analysis

A single particle model for the gas phase propylene polymerization including convection due to pressure drop in the particle is used to study the mass transfer of two component system. The addition of hydrogen in small amounts increases the bulk diffusion of the propylene–hydrogen gas mixture, increasing the reaction rates (physical effects). As hydrogen also reactivates the active sites and increases the deactivation, these chemical effects along with mass transfer effects are studied with empirical correlations. The experimentally observed increase in reaction rates in the presence of hydrogen may be due to physical and chemical effects combined together. The time scales for reaction and mass transfer of reactant in propylene–nitrogen mixture is analysed using the viscous modulus (a modified Thiele modulus) which accounts for effects of gas composition on mass transfer in porous particle due to convective flow and enrichment of the non‐reacting inert. These time scales are also analysed for catalysts which show activation behaviour. #An earlier version of this paper was presented at ECOREP II, 2nd European Conference on Reaction Engineering of Polyolefins, Lyon, France, July 1–4, 2002.

Estimation of the Polymerization Rate of Liquid Propylene Using Adiabatic Reaction Calorimetry and Reaction Dilatometry

The use of pressure-drop and constant-pressure dilatometry for obtaining rate data for liquid propylene polymerization in filled batch reactors was examined. The first method uses reaction temperature and pressure as well as the compressibility of the reactor contents to calculate the polymerization rate; in the second, the polymerization rate is calculated from the monomer feed rate to the reactor. Estimated polymerization rates compare well to those obtained using the well-developed isoperibolic calorimetry technique, besides pressure-drop dilatometry provides more kinetic information during the initial stages of the polymerization than the other methods.

Influence of Porosity on the Fragmentation of Ziegler-Natta Catalyst in the Early Stages of Propylene Polymerization

Abstract Catalyst fragmentation and polymer growth in the early stages of propylene polymerization has been investigated using different Ziegler-Natta catalysts. Polymerization was carried out in slurry under mild conditions and stopped at low yield. Scanning electron microscopy (SEM) has been used to characterize the surface and cross sectional morphology of polymer particles at different stages of particle growth. Different fragmentation behavior is observed, and it is found that the way that catalyst fragments in the early stages of polymerization is greatly influenced by the porosity of the catalyst. When a less porous catalyst was used, the resulting mass diffusion limitation causes layer-bylayer fragmentation, starting at the outer surface of the catalyst particle to the center. In contrast, for highly porous catalyst, monomer can easily penetrate into the pores of the catalyst / polymer particle. The polymer growing throughout the particle results in a coarse and instantaneous fragmentation.