Fluidized bed reactor for polyethylene production. The influence of polyethylene prepolymerization

Fluidized bed reactor for polyethylene production. The influence of polyethylene prepolymerization

F.A.N. Fernandes and L.M.F. Lona

Departamento de Processos Químicos, Faculdade de Engenharia Química,

Universidade Estadual de Campinas, 13081-970 Campinas - SP, Brazil

E-mail: [e-mail]

E-mail: [e-mail]

(Received: April 19, 1999 ; Accepted: January 13, 2000)

Abstract - This work addresses the influence of the prepolymerization of the catalyst particle on the fluidized bed reactor for polyethylene production. The influence of prepolymerization on the temperature and concentration gradients throughout the reactor was studied. The results obtained through simulations confirm industrial observations of the advantages of prepolymerization and extend these observation, showing the viable operational conditions for different superficial gas velocities and gas feeding temperatures as a function of the degree of prepolymerization.

Keywords: polymerization, fluidized bed reactor, prepolymerization, modeling and simulation.

INTRODUCTION

Great progress has been made in ethylene polymerization kinetics (Galvan and Tirrel, 1986; deCarvalho et al., 1989; McAuley et al., 1990), the resistance of polymer particles to mass and heat transfer (Galvan and Tirrel, 1986; Floyd et al., 1986a, b, 1987; Hutchinson and Ray, 1987) and prediction of polymer particle growth (Hutchinson et al., 1992).

In the area of fluidized bed reactor modeling, two key articles have presented a steady-state model for the reactor: Choi and Ray (1985) and McAuley et al. (1994). These two models assume temperature and concentration gradients in the bubble phase and the interaction of separate emulsion and bubble phases within the reactor bed. Both studies assume that the emulsion phase behaves as a continuous stirred tank reactor (fully mixed).

More recently, the fluidized bed polymerization reactor model has been futher developed to consider two phases (bubble and emulsion), flowing in plug flow at different velocities with polymer particles segregated, according to size and weight within the reactor (Fernandes and Lona Batista, 1998).

This modeling technique has allowed a more complete understanding of the fluidized bed polymerization reactor, thus permitting the study of reactor behavior (polymer yield and concentration and temperature gradients for both bubble and emulsion phases) and the prediction of the physicochemical properties of the polymer.

A thorough survey of the potentialities and limitations of the fluidized bed reactor can be carried out with better precision using new computer-aid techniques prior to lab and pilot-plant scale tests saving thereby time and money.

In this work we have studied the influence of the degree of prepolymerization on the behavior of the fluidized bed reactor. The basic operational conditions were simulated in order to study the limitations of the system.

FLUIDIZED BED REACTOR FOR POLYMER PRODUCTION

The fluidized bed reactor has an unique physical design (Figure 1). Gas and polymer particles flow in opposite directions. The gas is fed at the base of the reactor and splits into two phases: bubbles and emulsion.

Figure 1:
Fluidized bed reactor for polyethylene production.

The catalyst is fed in near the top of the reactor. In this way, as the catalyst moves downwards, the ethylene polymerization reaction takes place. The polymer grows on the catalyst, increasing its weight and size. Particle segregation occurs in the reactor according to particle weight. Polymer particles are removed from the reactor at the base. Nonreacted gases leave the reactor after passing through the disengagement zone.

Prepolymerization is used in fluidized bed systems for two main reasons: to produce specific polyethylene grades and to prevent the formation of hot spots inside the reactor.

Prepolymers are generally made in CSTRs located before the fluidized bed reactor. Catalyst particles are fed into the CSTR along with ethylene and comonomers to yield prepolymer. Afterwards, these prepolymerized catalyst particles are fed into the fluidized bed reactor to complete the ethylene polymerization. An industrial diagram for the fluidized bed reactor system using prepolymers is shown in Figure 2.

Figure 2:
Industrial polyethylene production diagram (BP Chemical Technology).

METHODOLOGY

The study was carried out using the phenomenological model developed by Fernandes (1999). This model describes a fluidized bed reactor with two phases (bubble and emulsion) flowing in plug-flow regime and with segregation of polymer particles within the reactor (Fernandes and Lona Batista, 1998).

Since the bubbles are considered to be non interactive spheres, a single bubble is used to infer the behavior of the entire bubble phase. Mass transfer between bubble and emulsion phases occurs by diffusion through the bubble clouds. Energy transfer between phases occurs due to the temperature gradient between phases and also by diffusing gases.

(1)

(2)

where i refers to the different gases fed into the reactor.

In order to consider axial temperature and concentration gradients and variations in the physicochemical properties within the bed, the mass and energy balances of the emulsion phase were expressed as differential equations that can be evaluated for any point in the reactor bed. The mass balance of the emulsion phase assumes consumption of gas by the reaction and mass transfer between bubble and emulsion phases by diffusion.

(3)

The energy balance for the emulsion phase comprehends energy transfer due to temperature gradient and for the diffusing gases, the heat of reaction and the heat loss to the surroundings through the reactor wall.

(4)

The mass balances for polyethylene and catalyst were combined since the final polymer particle grows on the catalyst particles and their masses cannot be separated. This mass balance is given by the variable c, which is the accumulated polymer mass produced by the mass of catalyst fed into the reactor. Strictly speaking c is the accumulated mass of the particle (polymer + catalyst) produced, but as the mass of polymer in the particle is much greater than the mass of catalyst, then c can be taken as the accumulated mass of polymer produced by the mass of catalyst fed into the reactor.

(5)

where

(6)

This variable allows the calculation of the total rate of polymer produced in the reactor and the prediction of particle size all along the reactor height.

In the rate expression, the gas concentration at the catalyst sites is assumed to be proportional to the gas concentration in the emulsion phase, since the gas dissolved in the polymer phase is in equilibrium with the gas in the emulsion phase (Floyd et al., 1986a, b, 1987).

Several simulations were carried out and the influence of prepolymerization was analyzed, paying special attention to the temperature of the bubble and emulsion phases, observing whether or not they would rise above the polymer melting temperature (temperature at which polymer melts, clogging the feeding and removal points of the reactor and resulting in the shutdown of the reactor).

The data and operational conditions used in the simulations are summarized in Table 1 (McAuley et al., 1990).

RESULTS

The Ziegler-Natta catalyst has a higher activity in the propagation of young polymer particles. This higher rate for young particles can lead to polymer overheating and melting in the reactor catalyst feeding zone (top portion of the reactor) owing to the exothermic characteristic of the polymerization reaction.

When fed into the reactor, prepolymerized catalyst particles when fed into the reactor help to prevent reactor overheating, since the catalyst activity is less intense on prepolymerized particles. The rise in temperature in the bubbles and emulsion phases, at the top portion of the reactor, is inversely proportional to the extent of prepolymerization of the fed catalyst particles, as shown in Figure 3.

In the case presented in Figure 3, the increase in emulsion temperature for the nonprepolymerized particles is of 62 K and for the catalyst with a 3000g_pol/g_cat degree of prepolymerization it is of 32 K, 48% lower than that occurring with no prepolymerization.

Lower increases in temperature can be useful to control the temperature inside the reactor, which favors the use of prepolymerized particles.

The superficial gas velocity of the gas entering the reactor is an operational condition that requires a lot of attention due to the risk of polymer melting. In industrial reactors, the superficial gas velocity is set to 3 to 6 times the minimum fluidizing velocity. When the superficial gas velocity is low, it directly affects the gas residence time in the reactor, lowering the heat transfer rate between bubble and emulsion phases. Figure 4 shows the emulsion temperature in the catalyst feeding zone as a function of the catalyst feed rate for three different degrees of prepolymerization and for two different superficial velocities. Observe that when the superficial velocity is low, not enough heat is removed from the system to prevent polymer melting. Higher degrees of prepolymerization help to prevent this risk since the reactor temperature does not rise above the polymer melting temperature.

A screening of the operational conditions was conducted to identify those which were viable, which resulted in ranges of operational conditions for different degrees of prepolymerization, as shown in Figure 5. The figure indicates the ranges where operational conditions do not increase the emulsion and bubble temperature above the polymer melting temperature, consequently risking a system shutdown.

When a polyethylene grade can only be produced under operational conditions that may cause polymer melting, then prepolymerization must be employed and a study of the minimum degree of prepolymerization for the catalyst needs to be carried out.

Gas feed temperature is the second critical operational condition affecting the possibility of polymer melting. The gas feed temperature increases the polymerization reaction rate, thereby increasing even more the emulsion and bubble temperature throughout the reactor (Figure 6).

Screening the operational conditions resulted in viable gas feed temperatures for the fluidized bed reactor. As can be seen in Figure 7, the gas feed temperature has limitations in the region over 360 K and 0.4 g_cat/s, where even where using high degrees of prepolymerization, the temperature inside the reactor reaches the polyethylene melting temperature, making operation at these points inviable.

As shown in Figures 5 and 7, the use of prepolymerized particles can broaden the range of polymer grades produced in the fluidized bed polymerization reactor, without compromising either polymer quality or the reactor itself.

The prepolymerization reactor (CSTR) does not exhibit problems of overheating or polymer melting. The residence time inside the prepolymerization reactor and the amount of polymer processed are small enough so as not to cause damage to the reactor or the quality of the polymer.

CONCLUSIONS

This study showed the influence of prepolymerization on the behavior of the fluidized bed reactor, related mainly to the temperature and concentration gradients throughout the reactor. As was demonstrated, the use of prepolymerized catalyst particles in the reactor decreases the reaction activity in the catalyst feeding region, leading to milder temperatures and lower temperature gradients throughout the reactor. These conditions avoid the problem of hot spots and polymer melting and consequently reactor shutdown.

These results confirm industrial observations of the advantages of prepolymerization in the fluidized bed reactor, increasing the range of operability of the reactor to lower superficial gas velocities and higher gas feed temperatures, without compromising the operation of the prepolymerization reactor.

This work will be followed by new simulations to determine the influence of prepolymerization and of the operational conditions of the reactor on the physicochemical properties of the polymer.

NOMENCLATURE

A area Cⁱ_j concentration of gas i in phase j c_p heat capacity H_m heat exchange coefficient K_m mass exchange coefficient kp polymer propagation rate M_W molecular weight q_cat catalyst mass flow Rp’ reaction rate T_i temperature of phase i U velocity U_h wall heat coefficient z height position [m] d bubble volumetric fraction DH heat of reaction e porosity r density c accumulated mass of polymer produced per mass of catalyst fed into the reactor

Subscripts

0 initial value b bubble e emulsion g gas i monomer mf minimum fluidizing condition ref reference s solids ¥ ambient

ACKNOWLEDGEMENT

The authors would like to thank FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo) for its financial support of this work.

• Choi, K.Y. and Ray, W.H., The Dynamic Behaviour of Fluidized Bed Reactors for Solid Catalyzed Gas Phase Olefin Polymerization. Chem.Eng.Sci., 40, 2261 (1985).
• deCarvalho, A.B., Gloor, P.E. and Hamielec, A.E., A Kinetic Mathematical Model for Heterogeneous Ziegler-Natta Copolymerization, Polymer, 30, 280 (1989).
• Fernandes, F.A.N., Modelagem e Simulaçăo de Reatores de Polimerizaçăo, Masters Thesis, State University of Campinas (1999).
• Fernandes, F.A.N. and Lona Batista, L.M.F., Reactor Modeling and Physical-Chemical Properties Characterization in Polyethylene Fluidized-Bed Reactor, ACS-Preprints, 43, 621 (1998).
• Floyd, S., Choi, K.Y., Taylor, T.W. and Ray, W.H., Polymerization of Olefins through Heterogeneous Catalists III. Polymer Particle Modelling with an Analysis of Intraparticle Heat and Mass Transfer Effects. J.Appl.Polym.Sci., 32, 2935 (1986a).
• Floyd, S., Hutchinson, R.A. and Ray, W.H., Polymerization of Olefins through Heterogeneous Catalists V. Gas-Liquid Mass Transfer Limitations in Liquid Slurry Reactors. J.Appl.Polym.Sci., 32, 5451 (1986b).
• Floyd, S., Heiskanen, T., Taylor, T.W., Mann, G.E. and Ray, W.H., Polymerization of Olefins through Heterogeneous Catalists VI. Effect of Particle Heat and Mass Transfer on Polymerization Behaviour and Polymer Properties. J.Appl.Polym.Sci., 33, 1021 (1987).
• Galvan, R. and Tirrel, M., Molecular Weight Distribution Predictions for Heterogeneous Ziegler-Natta Polymerization Using a Two-Site Model.Chem.Eng.Sci., 41, 2385 (1986).
• Hutchinson, R.A. and Ray, W.H., Polymerization of Olefins Through Heterogeneous Catalysis. VI. Particle Ignition and Extinction Phenomena. J.Appl.Polym.Sci., 34, 657 (1987).
• Hutchinson, R.A., Chen, C.M. and Ray, W.H., Polymerization of Olefins Through Heterogeneous Catalysis. X. Modeling of Particle Growth and Morphology. J.Appl.Polym.Sci., 44, 1389 (1992).
• McAuley, K.B., MacGregor, J.F. and Hamielec, A.E., A Kinetic Model for Industrial Gas-Phase Ethylene Copolymerization. A.I.Ch.E.J., 36, 837 (1990).
• McAuley, K.B., Talbot, J.P. and Harris, T.J., A Comparison of Two-phase and Well-mixed Models for Fluidized-Bed Polyethylene Reactors. Chem.Eng.Sci., 49, 2035 (1994).

Publication Dates

• Publication in this collection
  06 July 2000
• Date of issue
  June 2000

History

• Accepted
  13 Jan 2000
• Received
  19 Apr 1999