High pressure fluidization

Abstract

Polymers are often produced in pressurized fluidized beds. Large surface area and good mixing properties are key advantages of a fluidized bed. Despite decades of research, fluidization is still not completely understood. Especially since most academic research on fluidized beds is performed at atmospheric conditions. The objective of this work is to gain knowledge on fluidization of polymeric particles at elevated operating pressure, employing a combined modelling and experimental approach. The discrete particle model (DPM) and the two-fluid model (TFM) are used to gain detailed information of porosity distribution, bubble properties and solids mixing. Electrical capacitance tomography (ECT) was used to measure porosity distributions in a 30 cm diameter gas-fluidized bed. ECT is a relatively cheap and fast technique based on the difference in permittivity of air and polymeric particles. ECT requires a sophisticated reconstruction technique, for which the \cite{landweber1951} iteration method was used in this work. Since the permittivity and porosity are not linearly correlated, a concentration model is needed. In this work, an inverted Maxwell model is used for this purpose, since it represents the bubble emulsion structure best. Since opening and emptying the pressure vessel requires about 2 days, an advanced calibration method was developed to prevent frequent opening of the vessel. In this approach the permittivity of a packed bed is measured at the beginning and at the end of each measurement. If the calibration has changed during the measurement, the measurement is not used. Solids mixing is key in industrial reactors, since it prevents hot spots, it prevents undesired clustering and it ensures mixed product removal. Solids mixing is investigated using the DPM and TFM. A new method to quantify the degree of mixing based on the distance between particles and their initial neighbour was developed. The initial neighbour method performed better than existing methods since it is independent of the computational grid and the particle colouring, it can be used in all directions and it is highly reproducible. With increasing pressure five observations were made, which are explained below Emulsion phase becomes more porous. The emulsion phase becomes more porous with increasing operating pressure. At atmospheric operating pressure the porosity of the emulsion phase is similar to the porosity of a randomly packed bed (0.4), while at 20 bar the porosity of the emulsion phase rises to 0.5. Bubble-emulsion structure becomes less distinct. In both simulations and experiments it is observed that the clear distinction between bubbles and the emulsion phase gradually disappears with increasing pressure. At atmospheric pressure the emulsion phase is dense and the bubbles are clear voids containing little particles. At high pressure it is no longer possible to observe separate bubbles, although dense and porous regions in the bed still prevail, intermediate porosities occur just as frequent. Fluidization is more vigorous and bubbles behave more chaotic. From animations of simulations results (pressure drop fluctuations and bubble properties) it was observed that the fluidization is more vigorous at elevated pressure. Bubbles move chaotically through the bed and bubbles coalescence and break-up takes place frequently, although it is hard to distinct individual bubbles. (Micro) mixing is improved via increased granular temperature only caused by increased porosity. From DPM and TFM simulations it is observed that solids mixing is improved with increasing operating pressure. Based on DPM simulation results is found that this effect is caused by increased granular temperature. Granular temperature is not directly increased by the elevated operated pressure, but rather via the increased porosity of the emulsion phase, which creates more space for the neighbouring particles to attain different velocities. Bed expansion limits macro mixing. Micro mixing is mixing at the scale of individual bubbles, while macro mixing is at the scale of the entire bed. The micro mixing rate is increased with pressure because of the increased granular temperature. For pressures below 8 bar, macro mixing is enhanced with increasing operating pressure. At higher pressures, the bed expands, which decreases the mixing rate, since particles have to travel larger distances before they can become fully mixed.