Use this URL to cite or link to this record in EThOS:
Title: Tablet mechanical and dissolution properties : a comparison of melt granulation techniques
Author: Wang, Zhiyu
ISNI:       0000 0004 7223 4981
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Granulation is a process of size enlargement of primary particles which aims to improve their properties such as flowability and compressibility. For decades of development, the granulation has been widely used in many industries such as food and pharmaceutical. Melt granulation has received more attention from industries due to its unique advantages in recent years. In this research, the melt granulation experiments were performed on high shear mixer (HSM) and fluidized bed granulator (FBG) to produce granules and then processed into tablets. The experiments were carried out with different combinations of process and formulation variables, such as impeller speed, liquid to solid ratio in HSM; and air velocity and liquid to solid ratio in FBG. The product properties such as granule size, strength, dissolution, and tablet strength, hardness and dissolution were intensively studied, along with the granule and tablet surface studies with scanning electron microscope (SEM). It was found that the properties of HSM granules were mainly determined by the impeller speed, while the air velocity also plays an important role in affecting FBG granule properties. In terms of tablet properties such as strength, porosity and dissolution time, it was noticed that during tabletting, the compression force was the most dominant variable, as the effects of other variables could be minimized when the tablets were compressed with high compression force. After the experimental data was analysed, the data were used to build models to predict both granule (strength, dissolution and porosity) and tablet properties (strength, dissolution, hardness and porosity) of both equipment by design of experiment (DoE) and artificial neural networks (ANN). The input variables for modelling were corresponded to different product properties. Then the predicted values of both models were compared to the actual data, in order to find which model was more suitable to provide accurate predictions. The models will give an opportunity to predict product properties without doing a large number of experiments which saves time.
Supervisor: Salman, Agba Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available