C. Kiparissides

Characterization of the LCST behaviour of aqueous poly(N-isopropylacrylamide) solutions by thermal and cloud point techniques

Abstract A systematic investigation of the phase separation behaviour of aqueous low molecular weight PNIPA solutions was carried out in order to quantitatively investigate the effect of measurement conditions on the reported phase separation temperatures in relation to three techniques, namely, differential scanning calorimetry, optical cloud point and u.v. turbidimetry. The PNIPA concentration was varied in the range 0.5–22 wt%. All three techniques yielded comparable phase separation temperatures, independently of the transition kinetics, provided that the time scale of the experiments is large enough to ensure close to equilibrium conditions.

Polymerization reactor modeling: A review of recent developments and future directions

Abstract Synthetic polymers are produced via a multitude of reaction mechanisms and processes, including addition (e.g. free-radical, ionic, group-transfer, Ziegler-Natta coordination) and step-growth polymerizations. A major objective of polymerization reaction engineering is to understand how the reaction mechanism, the physical transport processes (e.g. mass and heat transfer, mixing), reactor configuration and reactor operating conditions affect the macromolecular architecture (e.g. molar mass, molecular weight distribution, copolymer composition distribution, branching distribution, stereoregularity, etc.) as well as the morphological properties of the polymer product (e.g. particle size distribution, porosity, etc.). As the polymer industry becomes more competitive, polymer manufacturers face increasing pressures for production cost reductions and more stringent “polymer quality” requirements. To achieve these goals one needs to develop comprehensive mathematical models capable of predicting the molecular and morphological properties in terms of reactor configuration and operating conditions. These mathematical representations can be classified into microscale kinetic models, mesoscale physical, transport and thermodynamic models and dynamic reactor ones. The present paper provides an overview of the different polymerization processes and mathematical modeling approaches. It is also addresses the problems related with the computer-aided design, monitoring, optimization and control of polymerization reactors.

Inferential Estimation of Polymer Quality Using Stacked Neural Networks

The robust inferential estimation of polymer properties using stacked neural networks is presented. Data for building non-linear models is re-sampled using bootstrap techniques to form a number of sets of training and test data. For each data set, a neural network model is developed which are then aggregated through principal component regression. Model robustness is shown to be significantly improved as a direct consequence of using multiple neural network representations. Confidence bands for the neural network model predictions also result directly from the application of the bootstrap technique. The approach has been successfully applied to the building of software sensors for a batch polymerisation reactor.

A comprehensive model for the calculation of molecular weight–long-chain branching distribution in free-radical polymerizations

Abstract A new method for the calculation of the joint molecular weight–long-chain branching distribution in free-radical highly branched polymerizations is developed. The method is based on the numerical fractionation of the total polymer population into a series of ‘classes’, each one representing a population of polymer chains with the same long chain branching content (e.g., linear chains, chains with one long–chain branch, etc.). Accordingly, dynamic molar balance equations are derived for the leading moments of the molecular weight distribution (MWD) of each polymer class as well as for the moments of the overall ‘live’ and ‘dead’ polymer chain distributions. A two-parameter Wesslau distribution is employed to reconstruct the MWD of each class in terms of its leading moments. The overall distribution is then calculated by the weighted sum of all class distributions. Simulation results are presented for a high-pressure ethylene continuous stirred tank reactor (CSTR) and a series of two CSTRs with or without a recycle stream. The effect of process parameters (e.g. temperature, chain transfer agent concentration and reactor residence time) on the MWD of low-density polyethylene (LDPE) is analyzed. It is shown that under typical operating conditions the calculated MWD exhibits a bimodal character in agreement with experimental measurements on MWD of LDPE produced in industrial autoclaves.

In vivo evidence of oral vaccination with PLGA nanoparticles containing the immunostimulant monophosphoryl lipid A.

Although oral vaccination has numerous advantages over the commonly used parenteral route, degradation of vaccine and its low uptake in the lymphoid tissue of the gastrointestinal (GI) tract still impede their development. In this study, the model antigen ovalbumin (OVA) and the immunostimulant monophosphoryl lipid A (MPLA) were incorporated in polymeric nanoparticles based on poly(D,L-lactide-co-glycolide) (PLGA). These polymeric carriers were orally administered to BALB/c mice (Bagg albino, inbred strain of mouse) and the resulting time-dependent systemic and mucosal immune responses towards OVA were assessed by measuring the OVA-specific IgG and IgA titers using an enzyme-linked immunosorbent assay (ELISA). PLGA nanoparticles were spherical in shape, around 320 nm in size, negatively charged (around -20 mV) and had an OVA and MPLA payload of 9.6% and 0.86%, respectively. A single immunization with formulation containing (OVA + MPLA) incorporated in PLGA nanoparticles induced a stronger IgG immune response than that induced by OVA in PBS solution or OVA incorporated into PLGA nanoparticles. Moreover, significantly higher IgA titers were generated by administration of (OVA + MPLA)/PLGA nanoparticles compared to IgA stimulated by control formulations, proving the capability of inducing a mucosal immunity. These findings demonstrate that co-delivery of OVA and MPLA in PLGA nanoparticles promotes both systemic and mucosal immune responses and represents therefore a suitable strategy for oral vaccination.

Multivariate data analysis applied to low-density polyethylene reactors

Abstract Skagerberg, B., MacGregor, J.F. and Kiparisides, C., 1992. Multivariate data analysis applied to low density polyethylene reactors, Chemometrics and Intelligent Laboratory Systems , 14: 341–356. In this paper we discuss how partial least squares regression (PLS) can be applied to the analysis of complex process data. PLS models are here used to: (i) accomplish a better understanding of the underlying relations of the process; (ii) monitor the performance of the process by means of multivariate control charts; and (iii) build predictive models for inferential control. The strategies for applying PLS to process data are described in detail and illustrated by an example in which low-density polyethylene production is simulated.

RECENT DEVELOPMENTS IN MODELING GAS-PHASE CATALYZED OLEFIN POLYMERIZATION FLUIDIZED-BED REACTORS: THE EFFECT OF BUBBLE SIZE VARIATION ON THE REACTOR'S PERFORMANCE

In the present study recent developments in modeling gas-phase catalyzed olefin polymerization fluidized-bed reactors (FBR) are critically reviewed. A new FBR model is developed to account for the effect of varying bubble size with the bed height on the reactor dynamics and the molecular properties of the polymer product. A comprehensive kinetic model for ethylene copolymerization in the presence of a multisite Ziegler–Natta catalyst is considered to describe the molecular weight developments in the FBR. The bubble-growth model developed in the present study is subsequently compared with two well-known FBR models, namely, the well-mixed and the constant bubble size model. The effect of important reactor parameters such as superficial gas velocity, maximum stable bubble size, mean particle size, catalyst injection rate, monomer/comonomer feed ratio and temperature of the feed stream, on the dynamic and steady-state behavior of the FBR is investigated. It is shown that the maximum stable bubble size, a critical parameter in the constant bubble size model, turns out to be less important when the bubble size is allowed to vary with the bed height. For typical industrial operating conditions, the constant bubble size model consistently overpredicts the emulsion phase temperature and monomer conversion, while the well-mixed model underestimates them. On the other hand, the bubble-growth model shows an intermediate behavior since it represents a more realistic description of the gas phase in the bed.

Recent Developments in Hardware Sensors For the On-Line Monitoring of Polymerization Reactions

2. ON-LINE CONVERSION AND COPOLYMER COMPOSITION MONITORING 61 2.1. Densimetry 63 2.2. Refractive Index Measurements 68 2.3. Gas Chromatography 70 2.4. Calorimetry/Reactor Energy Balances 75 2.5. Ultrasound Measurements 81 2.6. Fluorescence Spectroscopy 85 2.7. Ultraviolet Reflection Spectroscopy 91 2.8. Near-Infrared Spectroscopy 92 2.9. Midrange Infrared Spectroscopy 97 2.10. Raman Spectroscopy 98

Prediction of polymer quality in batch polymerisation reactors using robust neural networks

Abstract A technique for predicting polymer quality in batch polymerisation reactors using robust neural networks is proposed in this paper. Robust neural networks are used to learn the relationship between batch recipes and the trajectories of polymer quality variables in batch polymerisation reactors. The robust neural networks are obtained by stacking multiple nonperfect neural networks which are developed based on the bootstrap re-samples of the original training data. Neural network generalisation capability can be improved by combining several neural networks and neural network prediction confidence bounds can also be calculated based on the bootstrap technique. A main factor affecting prediction accuracy is reactive impurities which commonly exist in industrial polymerisation reactors. The amount of reactive impurities is estimated on-line during the initial stage of polymerisation using another neural network. From the estimated amount of reactive impurities, the effective batch initial condition can be worked out. Accurate predictions of polymer quality variables can then be obtained from the effective batch initial conditions. The technique can be used to design optimal batch recipes and to monitor polymerisation processes. The proposed techniques are applied to the simulation studies of a batch methylmethacrylate polymerisation reactor.

CFD analysis of turbulence non-homogeneity in mixing vessels A two-compartment model

Abstract A two-compartment model has been developed for calculating the droplet/particle size distribution in suspension polymerization reactors by taking into account the large spatial variations of the turbulent kinetic energy and its dissipation rate in the vessel. The two-compartment model comprised two mixing zones, namely an impeller zone of high local energy dissipation rates and a circulation zone of low kinetic energy. Computational fluid dynamics (CFD) was employed for generating the spatial distribution of energy dissipation rates within an unbaffled mixing vessel agitated by a flat two-blade impeller. A general methodology was developed for extracting, from the results of the CFD simulations, the volume ratio of the impeller over the circulation zone, the ratio of the average turbulent dissipation rates in the two zones, and the exchange flow rate between the two compartments. The effect of agitation rate, continuous phase viscosity, impeller diameter, and mixing vessel scale on the two-compartment model parameters was elucidated. The two-compartment model was then applied to a non-homogeneous liquid–liquid dispersion process to calculate the time evolution of the droplet size distribution in the mixing vessel. An excellent agreement was obtained between theoretical and experimental results on droplet size distributions obtained from a laboratory-scale reactor operated over a wide range of experimental conditions.