Model-based design of scalable continuous mixed-mode crystallisation

Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2024
Thesis identifier
  • T16954
Person Identifier (Local)
  • 201888403
Qualification Level
Qualification Name
Department, School or Faculty
  • Crystallisation is a complex process that consists of multiple mechanisms that can compete for solute particles in solution. Subsequently, modelling the crystallisation process accurately is challenging. Within this work, two different types of crystallisation, two different solute systems and two methods of parameter estimation have been modelled using population balance modelling software. The ability to model a system of interest allows the model to then be considered as a digital representative of the process. The mechanistic model can then be used to test the process under new conditions and ultimately allow for optimisation of the process for given process outputs. The main component system of interest for this work has been lactose and water. A mechanistic model of a seeded cooling lactose crystallisation has been achieved within this work. The characteristic slow growth and nucleation kinetics of lactose were seen experimentally and as such evaporative crystallisation was then focused upon. The development of a vacuum-induced evaporative crystalliser was attempted to circumvent the slow kinetics seen from the cooling lactose crystallisation work. The kinetics of this system were far faster and a representative model was built for this process. Despite a noisy and difficult process platform a sufficient model was built describing the primary nucleation, growth and agglomeration kinetics of the system. An innovative approach to process design was endeavoured to combine the two models and develop a multi-mode platform for the recovery of lactose from water. This allowed for the fast nucleation rate of an evaporative crystalliser to be utilised while combining with a multi-stage cooling crystallisation to improve yield recovery. The ultimate goal of this work is to show the usefulness of developing mechanistic models. The models developed within this work allowed for the optimisation of given crystallisation processes for focused optimisation objectives. Optimisation of crystallisation without further experimentation allows for well-defined and designed processes to be built while performing minimal experiments and producing less waste. Finally, within this work, the combination of mechanistic models has been attempted to develop a multi-mode crystallisation platform. This approach allowed for an entirely simulation-based design of a continuous platform made up of both evaporative and cooling crystallisation stages. From this work, the ability to model effectively and utilise mechanistic models to reduce experimentation and waste has been clearly demonstrated.
Advisor / supervisor
  • Brown, Cameron J.
  • Florence, Alastair
Resource Type
Date Created
  • 2023