Date of Award


Degree Type


Degree Name

Master of Science in Aeronautics and Astronautics


Aeronautics and Astronautics

Committee Chair

Alina Alexeenko

Committee Member 1

Elizabeth Topp

Committee Member 2

John Sullivan

Committee Member 3

Steven Nail


Freeze drying is a time consuming, expensive process to ensure drug stability and increase shelf life of drugs that are not stable in solution. The process can be optimized for production scale freeze dryers but this requires scale up from lab scale process design. The methods currently in place to characterize freeze dryers are important. This work presents a Computational Approach as a tool to reduce the number of experiments required and study the heat transfer in freeze dryers. In freeze-drying processes, the equipment capability curve is a critical part of the design space. Understanding of the equipment capability limits is important for process design, transfer of cycles from one manufacturing site to another and scale up. Two experimental methods are presented to determine equipment capability, the Choked Point Flow and Minimum Controllable Pressure Test. A computational model to generate the equipment capability curve is also presented and compared with experimental vapor flow rate data obtained using the tunable diode laser absorption spectroscopy. Experiments on Lyostar 3 show that the Minimum Controllable Pressure test is 66% faster than the Choked Point Flow method. Experiments performed on a laboratoryscale Lyostar2 and Lyostar3 freeze dryers are used for validation of the computational modeling. The simulations for Lyostar 2 are on average within 4.8% from the experimental data. From computations, it is evident that the isolation valve is a critical component of the system and influences the minimum controllable chamber pressures by as much as 23.7%. The data analyzed in this paper points to the utility of Computational Fluid Dynamics in establishing equipment capability limits. Monitoring the sublimation rate in lab scale freeze drying process is important to help optimize the cycle. The current methods to determine sublimation rate include gravimetric measurements and TDLAS. The gravimetric methods do not provide real time sublimation rate data. The current work discusses two methods to determine the vapor flow rate: using heat flux sensors and through a numerical model. The method based on heat flux sensors measures the total amount of heat to the product through conduction from the shelf. Additional heat input through radiation needs to be calculated. Calibration tests and minimum controllable pressure tests are conducted and compared with a CFD model. The second approach is a soft sensor which uses the chamber and condenser pressure data from experiments to determine the sublimation rate via a CFD model. Both approaches have the potential to be real time sublimation rate monitoring methods. There has been some debate in the literature on the contribution of convection in the total heat transfer to a product. The non-uniformity in drying vials on a shelf, where the edge vials dry faster can be attributed to additional heat transfer to these vials. This work presents an overview of the vapor velocity magnitudes in a lab scale freeze dryer for different sublimation rates and shelf gap heights. Using previous experimental data from literature, a CFD model is built to calculate the convective component of heat transfer using empirical relations for heat transfer. The contribution of convection to the heat transfer of product vials has been found to be anywhere between 20 to 25%.