High-fidelity simulations of multi-component pressurised metered-dose inhaler sprays

Accurate predictions of drug delivery from pressurised-metered dose inhalers (pMDI) require a detailed understanding of the formulation’s state when it leaves the nozzle. This is difficult to experimentally determine, but is important as it sets the initial conditions for the formation of inhaled droplets and particles. The aim of this study is to develop an improved computational model to predict the initial state of the droplets produced by pMDIs. The flow inside a pMDI is complex and unsteady. Most existing models ignore the unsteady turbulent flow inside the device and the effects of the internal geometry. Most models also assume that the propellant state is at thermodynamic equilibrium or in a fixed state, while in reality an intermediate state is likely. We present a novel computational framework for multicomponent pMDI formulations that simultaneously addresses all these phenomena. A homogeneous relaxation model allows the formulation to relax toward thermodynamic equilibrium over an empirically determined time scale. Our model assumes that the liquid and vapour are well mixed within each computational cell, and a population balance model is used to simulate droplet formation. Interaction of the formulation with the ambient air is also considered. Results for placebo P-134a and P-134a-ethanol solutions are shown and compared favourably with optical, X-ray and Malvern measurements of sprays from a conventional Bespak actuator. The model allows us to determine droplet properties (initial D32, temperature and density) and predict the likelihood of water adsorption into the primary droplets due to entrainment of the warmer, humid ambient air.