Emulsion polymerisation is a process of considerable technological and industrial significance. The process presents many challenges in respect to design, optimisation and multi-objective distribution control. The quality of latex is determined by the final product properties, Le. viscosity, mechanical strength and film-forming ability, which are in turn a function of the latex attributes of PSD and MWD. This motivates an inferential control scheme utilising these distributions. This research addresses model development and controllability analysis towards model-based control. A population balance model for PSD and MWD is developed. The PSD information is incorporated via a one-dimensional population distribution of the polymer particles with respect to their size, in conjunction with a population distribution of the total live polymer radicals/particle. MWD information is incorporated via one-dimensional population distributions of the live radicals and dead polymer chains with respect to their length (in different sized particles). The model solution is facilitated by a number of algorithmic developments, including a decomposition algorithm coupled with a multi-level discretisation for PSD and the application of the method of moments for MWD. This model is compared to extensive experimental data for its validation. Improvements in the form of a twodimensional version of this model enable better prediction of compartmentalisation and hence the growth rates, thereby improving model match with experiments. This work presents a study into the simultaneous controllability of PSD and MWD, assessed through an experimental sensitivity analysis on the main process manipulations: initiator, eTA, monomer and surfactant. The work analyses the practical limitations on the attainability of PSD and MWD. The range of experiments carried out clearly indicates the operation of individual mechanisms in the simultaneous formation of PSD and MWD.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:486551 |
| Date | January 2008 |
| Creators | Sweetman, Stephen John |
| Publisher | Imperial College London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/10044/1/7430 |
Page generated in 0.0013 seconds