Ramdhane Dhib

Simulation of Free Radical Bulk/Solution Homopolymerization Using Mono- and Bi-functional Initiators

ABSTRACT Monofunctional and bifunctional peroxides are extensively utilized in the production of polystyrene. However, their decomposition kinetics is still an important area to investigate. A computer package developed previously by Gao and Penlidis (1996), which was based on mono-fimctionally initiated polymerization, is extended in this study to cover homopolymerization of styrene using bifunctional initiators. A database of a wide variety of bifimctional peroxides used in polystyrene production is developed along with a database for twelve new monofimctional initiators. The package handles several different scenarios of styrene polymerization, whether it is in bulk / solution or isothermal / non-isothermal medium. The model is tested with a wide range of conversion and average molecular weight data either from the literature or available from industry. Simulation results agree with the data trends, which makes the simulator a very powerful tool in aid of initiator selection and evaluation. At the end of the paper, results on homopolymerizations of acrylates and methacrylates are also discussed.

CFD modeling of the mixing of water in oil emulsions

Abstract A computational fluid dynamics (CFD) model was developed for the mixing of water in oil emulsion in a lab-scale mixing tank equipped with a Rushton turbine impeller. Multiple reference frames (MRF) technique, k – ɛ model, and Eulerian–Eulerian approach were employed to model the impeller rotation, turbulence, and multiphase flow, respectively. The droplet size distribution within the mixing tank was estimated by means of the population balance approach, which employs the discrete method to describe coalescence and breakage of water droplets. To validate the CFD model, the cumulative probability size distribution computed using the model was compared with the experimentally determined values reported in the literature. This validated CFD model was then utilized to explore the effects of the impeller speed, oil type, and volume fraction of water on the cumulative probability droplet size distribution, number density, and distribution of local volume fraction of the dispersed phase.

CFD analysis of mixing in thermal polymerization of styrene

Abstract Thermal polymerization of styrene in a lab-scale CSTR equipped with a pitched blade turbine impeller was simulated using a computational fluid dynamics (CFD) software package. The rotation of the reactor impeller was modeled using the multiple reference frames (MRF) technique. The path lines of the particles, released at the reactor inlet, were also generated to analyze the reaction progress throughout the reactor vessel. The effects of the impeller speed, the input–output stream locations and the residence time were investigated. The simulation showed the formation of a well-mixed region around the impeller and stagnant or slow moving fluids elsewhere in the reactor due to high viscosity of the polymer mass. The monomer conversion computed using the CFD model was in good agreement with that obtained from the CSTR model at low residence time. The input–output locations have a significant impact on the monomer conversion and the system homogeneity in the CSTR.

Infrared Drying: From Process Modeling to Advanced Process Control

The article surveys the drying of solids materials and polymer solutions when infrared radiation (IR) is employed as the main heating source. The study reviews the current research trends of IR drying of specific applications. A case study similar to an industrial setting is presented to illustrate a model development and control scheme of an IR drying unit. The design and online implementation of an internal model controller (IMC) is discussed. The study demonstrates the controller capabilities to suppress random variations of the moisture content in the material entering the dryer. Simulation results also showed the success of model predictive control (MPC) multivariable controller ability, while handling process interactions and process constraints, to track setpoint changes in the humidity and temperature of the material exiting the dryer and to reject unmeasured stochastic disturbances in the inlet humidity stream.

Model predictive control of an infrared-convective dryer

Abstract For product quality purposes, the drying of some delicate materials is not easily feasible without the realization of a tight, well-designed controller. This study presents the results of a model predictive control for the drying of a thin fiber sheet moving through an electric infrared dryer, whose process dynamics are described by a minimum realization, previously developed. The simulation results indicate that the MPC multivariable controller is able to handle process interactions and produced good performance for tracking set point changes in the humidity and temperature of the material exiting the dryer and also for rejecting measured stochastic disturbances in the inlet humidity stream.

Use of a Novel Tetrafunctional Initiator in the Free Radical Homo‐ and Copolymerization of Styrene, Methyl Methacylate and α‐Methyl Styrene

An investigation into the effect of initiator functionality on the free radical polymerization of various monomer systems including styrene, methyl methacrylate (MMA), styrene–MMA and α‐methyl styrene–MMA with a tetrafunctional peroxide initiator (JWEB50) was completed. The performance of the tetrafunctional initiator was compared to a monofunctional counterpart (TBEC). Kinetic results showed that regardless of the monomer system studied, JWEB50 produced a faster rate of polymerization compared to TBEC at an equivalent concentration. Molecular weights, radii of gyration and intrinsic viscosities were obtained from two size exclusion chromatography setups: one equipped with a multi‐angle laser light scattering detector and the other with low‐angle laser light scattering and viscosity detectors. For the homopolymerization of styrene, JWEB50 produced polymer molecular weights similar to TBEC at the same concentration, while experiments with MMA indicated that JWEB50 produced molecular weights closer to those obtained with TBEC at a concentration four times as great. Runs with a feed mixture of styrene and MMA gave results that were a combination of the observations made for the individual homopolymerizations. The results for the feed mixture of MMA and α‐methyl styrene were similar to the findings with pure styrene. Plots of radius of gyration, intrinsic viscosity and their corresponding branching factors provided evidence of branching for all cases except for the homopolymerization of MMA.

Modelling of free radical polymerisation of ethylene using difunctional initiators

Abstract Low-density polyethylene (LDPE), commonly produced in high-pressure free radical polymerisation processes, is very important for the manufacture of engineering and commodity plastics. However, the thermodynamic conditions of the process hinder ethylene from going to full conversion. Other than recycling the product, one way of improving the monomer conversion is to initiate the polymerisation with difunctional organic peroxides. But, the kinetic decomposition of multi-/difunctional peroxides is still a controversial issue. The present study proposes a kinetic model based on a postulated reaction mechanism for free radical ethylene polymerisation initiated by difunctional initiators. The model describes the rates of initiation, propagation and the population balance equations in an isothermal autoclave reactor operated at a constant pressure of 1700 bars and a temperature range of 110–300°C. The simulation confirmed the trends of experimental data collected from literature for one monofunctional and two difunctional initiators. Due to the dual functionality of difunctional peroxide, the ethylene conversion obtained was about twice as much as that obtained with monofunctional peroxide, for only a half amount of the initial initiator concentration. The model also gives good prediction of the number average molecular weights data of the polymer and predicts reasonable values of the polydispersity index.

NEURAL NETWORK IDENTIFICATION OF STYRENE FREE RADICAL POLYMERIZATION

In this study, a one-hidden-layer Artificial Neural Network (ANN) using a back-propagation structure has been trained on experimental data for the identification of styrene conversion and polymer average molecular weight produced in a bulk polymerization initiated by bifunctional peroxides. Two identification schemes, a leave-N-out and black-box with constant and adaptive step-size, are presented. Both schemes gave good predictions of monomer conversion and molecular weight averages. Neural networks for nonlinear identification of monomer conversion and average molecular weights can be used along with other approaches to improve modeling of polymerization reactions, especially when the kinetic mechanism is not well established.

Model predictive control of an infrared-dryer

For product quality purposes, the drying of some delicate materials is not easily realizable and requires a tight well-designed control scheme. This study presents control results for the drying of a thin fiber sheet moving through an electric infrared dryer. The drying process dynamics is described by a minimum realization linear model that is used to develop a model predictive controller (MPC) and test its performance. The results obtained indicate that the MPC multivariable controller is able to handle process interactions and produced good results for setpoint tracking of the humidity and temperature of the material exiting the dryer and for disturbance rejection of measured and unmeasured disturbances.

Effects of hydrogen peroxide feeding strategies on the photochemical degradation of polyvinyl alcohol

ABSTRACT The performance of batch and fed-batch photoreactors with that of continuous photoreactor for the treatment of aqueous polyvinyl alcohol (PVA) solutions is compared. Hydrogen peroxide feeding strategies, residence time, and [H2O2]/[PVA] mass ratio are examined for their impacts on the molecular weight distribution (MWD) of PVA and the total organic carbon (TOC) removal. The results prove that a continuous addition of H2O2 during the degradation reaction ensures the utilization of the produced radicals to minimize the oxidant consumption and maximize the TOC removal and the PVA degradation in a short irradiation time. Also, the MWD of PVA is found to be bimodal and shifted towards lower molecular weights with small shoulder peak indicating a progressive disappearance of the higher molecular weight fractions that is in accordance with the random chains scission mechanism. Besides, the hydrogen peroxide feeding strategies are found to have a great effect on the reduction in H2O2 residuals in the effluent.