Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Author ORCID Identifier

https://orcid.org/0000-0002-4851-6087

AccessType

Campus-Only Access for Five (5) Years

Document Type

dissertation

Degree Name

Doctor of Philosophy (PhD)

Degree Program

Chemical Engineering

Year Degree Awarded

2022

Month Degree Awarded

February

First Advisor

Dimitrios Maroudas

Second Advisor

Neil Forbes

Third Advisor

David Schmidt

Fourth Advisor

Kostas Saranteas

Subject Categories

Chemical Engineering

Abstract

The pharmaceutical downstream manufacturing process of active pharmaceutical ingredient (API) involves several unit operations including crystallization, filtration, drying, and formulation. As a solid recovery step in the purification of APIs (Active Pharmaceutical Ingredients), filtration is an essential liquid-solid separation process and an important contributor to impurity level control in downstream pharmaceutical manufacturing. Under comparable operating circumstances, depending on the properties of different materials being filtered, the filtration time could vary by days, resulting in delays in the manufacturing cycle. Poor filtering performance could sometimes quickly become the bottleneck of the downstream operations. Therefore, predictive modeling of the pressure filtration process based on material properties and operating conditions can help to increase our knowledge, as well as to drive continuous improvements in developing efficient filtration processes. We have taken different approaches, ranging from macroscopic modeling to detailed particle-based simulations, effectively linking the operation conditions, liquid properties, cake structure, and the properties of the constituent particles to the cake filterability and filtration performance.

We have evaluated the impacts of solid particle properties on fluid flow through the cake by a coupled computational fluid dynamic (CFD) and discrete element method (DEM) approach. The combined CFD-DEM model captured both particle-particle and particle-fluid interactions through fundamental mathematical descriptions. The model was validated using data collected by filtering spherical glass beads and deionized water mixtures using a dead-end cell over a range of applied pressures and for two mean particle sizes. Numerical experiments were then performed to study the effects of particle properties, liquid properties, and operating conditions on filtration performance. The model predicted that the cake resistance and filtrate flow rate could be strongly affected by the mean size and polydispersity of the spherical particles, the presence of small particles (i.e., fines) in the particle distribution, the viscosity of the liquid, and particle deformation leading to cake compression. We have further extended our investigation into non-spherical, polydisperse particles that are more representative of actual APIs. We have studied the combined effects of particle shape and size on pressure filtration performance using rigid body dynamics (RBD) simulation, which enables the formation of a realistic porous cake structure based on the shape and size distribution of the constituent particles. The predicted flow rate was estimated using a spatial averaged approach with locally varied cake properties and agreed with previous experimental measurements. By simulating cylindrical particles over a wide range of aspect ratios, we showed that particle shape can have a dramatic effect on cake porosity and filtrate flow with disk-shaped particles predicted to be particularly difficult to filter. Additionally, we have investigated the breakage effect of particles on filtration. By simulating breakage cases for rod-like particles with various properties and aspect ratios, we have shown that needle-like particles have a higher tendency to fracture, resulting in reductions in filtration rate. We also demonstrated that population balance models, which accelerated the prediction processing time, can be established based on simulation results from discrete particle simulations.

DOI

https://doi.org/10.7275/27109973

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Available for download on Wednesday, February 01, 2023

Share

COinS