Investigation Of Mixing Characteristics In LDPE Autoclave Reactor

The mixing condition in the LDPE (low density polyethylene) autoclave reactor is not only closely related to safe and stable production status of the reactor, but also directly affects the properties of polymer. It is significance to deeply understanding of technology characteristics and operation optimization of the autoclave reactor by building up the correspondence between mixing characteristics and LDPE polymer properties on the basis of researching into the macromixing and micromixing characteristics of the autoclave reactor and combining reactor mixing model and free-radical polymerization kinetics. Focused on this subject, this thesis has carried out three parts of research as follows.Firstly, cold model experimental apparatus for LDPE autoclave reactor was built, and the macromixing characteristics were obtained by the way of investigating (macro-)mixing time (tM) and residence time distribution (RTD) systematically. The results showed that the bearing house in the middle of the reactor played the role of radial baffle, and could divide the reactor into two well mixed zones. Multiple inlets method not only ensured the well-mixed condition in the whole area of the reactor, but also made overall RTD less influenced by the agitation speed. For flow system with multiple inlets, the variance analysis would lead to conflicting conclusion with real situation while information entropy mixing degree (M) which was based on the information entropy theory could reflect the flow and mixing characteristics accurately.Secondly, the micromixing characteristics of the cold model experimental apparatus were investigated by the iodide-iodate reaction method, as well as the micromixing condition in the industrial reactor was analyzed. The results of experiments showed that the micromixing is enhanced along with the increase of agitation speed and the decrease of fluid viscosity, and the influence of viscosity on micromixing was greater than that of agitation speed. The micromixing time (tm) calculated by incorporation model was about0.003～0.05s under the experimental conditions. Based on the assumption that the relationship between segregation index and fluid viscosity of industrial reactor is the same as the relationship of cold model apparatus, the micromixing time (tm) of the industrial reactor was about0.03～0.05s at260℃. The characteristic reaction time (tr) for initiator decomposition was about0.9s at260℃. Thus tm was far less than tr, which illustrated that the micromixing had little effect on polymerization reaction. However, when temperature reached290℃, tm (0.013s) got close to tr (0.1s), then the effect of micromixing on reaction should not be ignored.Thirdly, the reactor mixing model (PFR-CSTR-PFR composition model) was established which can reflect the imperfect mixing characteristics in LDPE autoclave reactor, and free radical polymerization kinetics were used for Aspen simulation, finally the influences of different variables on polymer properties had also been investigated. The model parameters, CSTR ratios and backflow ratios, were obtained and validated by the RTD experimental data. The backflow ratios of industrial reactor were determined by the relationship curve between Reynolds number and backflow ratios, and they were used for Aspen simulation of industrial reactor. The results of simulation showed that monomer conversion was increased along with the raising of temperature, pressure or initiator flow-rate; temperature and pressure had greater influence on molecular weight of LDPE; flow-rate of ethane (chain transfer agent) had little effect on molecular weight. Furthermore, the influences of macromixing (backflow ratios) and micromixing (initiator efficiency) on LDPE production, results showed that macromixing had little effect on polymer properties; worse in local micromixing condition led to decrease in efficiency of initiator, thus, monomer conversion was decreased, number-average molecular weight and weight-average molecular weight were decreased, polydispersity index was increased and frequency of long/short chain branching were decreased.