(726e) Using Coupled CFD/DEM Simulations to Predict the Effects of Formulation and Device Resistance on Emitted and Fine Particle Doses of Dry Powder Inhalers

Dry powder inhalers (DPI) are a commonly used device type to deliver medications to the lungs. The use of DPI devices is relatively straightforward. Typically, a pre-filled capsule containing the active pharmaceutical ingredient (API), normally a formulated blend of powder, is placed in a molded cavity within the device. Both ends of the capsule ends are then pierced using self-retracting needles that are built into the device, leaving two small openings therein. As the patient subsequently performs an inhalation through the device mouthpiece, the powder blend is dispersed out of the capsule and the DPI. The performance of the formulation and the device is often quantified in terms of the emitted dose (ED), i.e. the amount of API that makes it out of the device, and the fine particle dose (FPD), i.e. the amount of API that is expected to be delivered to the patient’s lung. While it is generally accepted that both of these quantities are dependent upon the device ‘resistance’—a measure of the air flowrate through the device at a given pressure drop that is primarily a function of the device inlet dimensions—the mechanisms of why that is so is not well understood. This work concerns the development of a coupled computational fluid dynamics (CFD) and discrete element method (DEM) model that provides a mechanistic insight into factors that influence ED and FPD. Commercial software ANSYS Fluent and Rocky DEM were used to resolve the air flow and solid particles/powder momentum equations, respectively. Model results highlight the relationship between device design, flowrate, and the powder particle size distribution that in turn influences ED and FPD. The coupled model is also able to explain how device design and formulation factors affect the amount of powder retained in the capsule and the device after each device use. The predicted location and approximate quantity of in-capsule powder retention were found to be in a reasonable agreement with experimental data. Powder flow dynamics during its discharge from the capsule was also found to qualitatively agree with experimental high-speed video captures. The approach presented herein provides an example of how computational modeling can be used to inform device selection given the properties of the formulation in order to maximize ED and FPD.