Model-Based Design of Pharmaceutical Crystallization Processes
Abstract
Developments in the technology are followed by the methods and frameworks in industry to intensify the production of information, tools, goods, and services. This trend has been followed by the pharmaceutical industry which is highly regulated by administrations to meet the quality requirements of the drugs and produce more for cheaper in a shorter time. As the knowledge and understanding of crystallization systems have been increasing with the development of process analytical technology (PAT) tools, it is inevitable to use the experience and data coming with it to develop data-driven, better processes. In the light of these developments, Industry 4.0 has started becoming the new paradigm in pharmaceutical industry pushing the data-driven design of pharmaceutical processes. This thesis demonstrates the development and usage of a framework for data-driven, model-based design of pharmaceutical processes that fall in line with this latest paradigm shift. The proposed framework can be summarized in four levels showing the benefits of the collected data and model development. These levels are data collection from specially designed experiments, model writing, using the data collected from the first step to train the model, validation of the model to call it ‘Digital Twin’ of the process, using the digital twin for process design via in-silico design of experiments (DoE) or process optimization. The chapters in this thesis are different case studies that follow these steps for model-based process design. The systems studied are batch cooling crystallization with temperature cycling to produce a drug compound, batch cooling crystallization with integrated milling and temperature cycling for the shape optimization of the same drug compound from previous step, and hot melt extrusion for amorphization of another drug compound. In addition to demonstrating the development of the whole framework and its possible benefits in each chapter, Chapter 4 is the solely experimental proof of concept of a previous, more general model-based design of a mill integrated crystallization work.
Degree
Ph.D.
Advisors
Nagy, Purdue University.
Subject Area
Analytical chemistry|Chemistry|Computer Engineering|Optics|Pharmaceutical sciences
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.