An approach towards a multi-scale model of the fluidized bed reactor for LLDPE production

Conference Dates

May 10-15, 2015


Fluidization is a complex process describing the transformation of solid particles into a fluid-like state through suspension in a gas or liquid [1]. This complexity is further increased in reacting systems, such as the production of linear low density polyethylene (LLDPE). Gas phase polymerization in fluidized bed reactors (FBR) is a well-recognized technology for polyolefin production. FBRs in comparison to other reactors, employing slurry or solution polymerization, have added advantages in transporting solids in and out of the reactor and fast reaction occurs due to high mass and transfer rates between the gas and particles [2]. In LLDPE production particle growth is influenced by ethylene, co-monomer and condensing agents. The phenomena occurring in commercial size FBR in reacting conditions cannot fully be explained by experiments on a small, bench scale. Although such experiments provide insight into particles behavior (like flow profiles and softening) and kinetics, but fail to provide information which is fully transferrable to industrial scale. In addition, the commercial size FBRs cannot be fully modelled or experimented on due to their large and multi scale, particles dynamic behavior, and turbulent flow. The main reason for this is that a lot of particles have to be accounted for in order to get results that are representative enough of the reactor dynamics in the bigger scale. In this work, PE softening measurements and lab scale 2D fluidization experiments were conducted on different LLDPE reactor products to determine particle properties (particle size distribution), bubble properties (bubble size distribution and bubble shape) and the process conditions influencing their behavior in FBR. This experimental work is fitted into an FBR simulation module which can be used to deepen understanding of the fluidization fundamentals. The industrial scale FBR for LLDPE production is then modelled with a computational fluid dynamics (CFD) tool. This multi scale model provides information on the fluidization characteristics and thermal behavior focusing on gas and particle distributions within the reactor. Both these approaches (bench scale experimentations and the CFD models) proved useful and showed how challenges like gas distribution in the reactor, bubble types and reactor hot spots, etc. can be traced back to particle behavior caused by softening which are shown on a small scale. In the end this information assists in optimizing our industrial FBR performance.


abstract only

This document is currently not available here.