Modelling agglomeration and breakage during agitated vacuum thermal drying

Lead Research Organisation: University of Strathclyde
Department Name: Chemical and Process Engineering


Much emphasis has been placed on the role API particle attributes play in both drug product performance and processability. Significant progress has been made in linking crystal structure with crystallization process design. However, the API isolation steps, filtration, washing and especially drying have not been subject to such intense investigation. This PhD will deepen our understanding of the creation and subsequent breakage of granular APIs. The research objective is to understand and model the agglomeration processes which occur during drying and how these influence the evolution of: agglomerate size; size distribution; and agglomerate strength during drying. The ultimate objective being to correlate measured attributes and process parameters with outcomes.
The PhD program will start with an evaluation of alternative approaches to describe the build-up of material at individual points of contact between crystals establishing necks of deposited material as the solvent evaporates. In parallel to this two particle approach the student will develop a model for contact strength based on particle size, size distribution and crystal habit. Combining these two components the student will extend the model to multi-particle agglomerates / granules to investigate strength and breakage mode. Existing models from other fields of research will also be considered, in particular rheological insights from soil science will be reviewed and relevant approaches will be incorporated into the project.
Concerning the formation of granules, both static and agitated drying will be analysed. The modelling programme involves a number of key aspects:
- The drying kinetics of contact necks, and the dependence on particle size as well as solvent properties;
- The packing of particles in a static bed to obtain a network of particle contacts;
- The evolution of agglomerates as the drying proceeds, yielding time-dependent granule size distributions and shapes;
- The mechanical strength of the various agglomerates, with contact necks providing weak points for breakage;
- The development of stochastic models for agglomeration in agitated dryers, where the granules form and break due to collisions throughout the drying process.
The work will utilise a range of analysis and software tools, leading to the development of statistical models with predictive capabilities. An explicit model of the wetted neck between crystals and the capillary forces driving both neck formation and solute transport during drying will be built and used to predict the strength of the resulting solid bridge holding particles together. The software to be employed includes: EDEM for particle packing; Finite Element tools for mechanical strength of the agglomerates; and bespoke Monte Carlo codes for agglomerate growth and fragmentation. Finally, scaling analyses will be performed to understand how the system behaviour changes over time and over process scale, i.e. when moving from the desk-top systems to production units.
The experimental programme is designed to complement the modelling one, providing key data and characterisation that feeds into the modelling enabling a refinement of the methodology. We will first focus on static drying at specific solvent contents in a series of highly constrained experiments, using our bespoke filter cake drying apparatus. The role of temperature, solubility in the selected solvent, solvent properties, boiling point, vapour pressure, viscosity, interfacial tension and wettability with respect to the chosen API will be investigated. Drying kinetics will also be determined as a function of operating temperature, pressure and gas flow rate. In order to track the physical attributes of the API during drying a semi batch approach will be taken in which the same input material is dried to different end points then characterised. In this way, the work makes direct comparison to the predictions from the modelling programme.


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/T517665/1 01/10/2019 30/09/2024
2267867 Studentship EP/T517665/1 01/10/2019 30/09/2023 William Luke Eales