This CFD study was performed on behalf of company “ELPEN S.A.”, one of the larger and more research active pharmaceutical companies in Greece. The ELPENHALER device is used for drug delivery in dry powder form for the prevention and therapy of Asthma and Chronic Obstructive Pulmonary Disease and is available in two designs: (a) Single blister and (b) Double blister (Figure 1). ELPENHALER has been developed exclusively by ELPEN, aiming at design simplicity and minimization of patient interference, in order to eliminate operation errors.

The drug-delivery principle is based entirely on the appropriate aerodynamic design of the internal spaces of ELPENHALER, so that patient receives the necessary dose of medicine. As the patient breathes in from the mouthpiece, air enters the device and then splits in two streams, one of which flows upon the blister(s) which are filled with powder and carries it along (Figure 2). Then this main stream reunites with the bypass stream, before entering patient’s oral cavity.

In order to gain detailed insight of ELPENHALER, but also to certify its experimentally measured good performance, SimTec constructed a CFD model in ANSYS Workbench (ANSYS DesignModeler, ANSYS Meshing & ANSYS FLUENT) for the simulation of the airflow and powder tracks during the breathing of the patient.

The three dimensional geometry of the model (Figure 2) was supplied by ELPEN in the form of electronic CAD files. After the removal of the unnecessary (for CFD analysis) geometric regions, a computational mesh of mainly tetrahedral (due to the complexity of the shape of the device) elements was constructed comprising of 7.7 million elements (Figure 3) and employing two important properties: (a) each of the many small gaps of the device was covered with at least 12 computational elements, and (b) mesh inflation layers of at least 4 (prismatic or hexahedral) elements were constructed over each device surface, so as to accurately model the flow turbulent boundary layer, which is absolutely essential for the proper simulation of ELPENHALER operation.

The flow was simulated under steady conditions, driven by an appropriate underpressure (suction) level of 4000 [Pa] at the device mouthpiece, as provided by the directive of the European Pharmacopoeia. A Reynolds Stress Turbulence Model was employed for turbulence modeling, capable of predicting turbulent flows with swirl, separation and streamline curvature, as in the case of the airflow inside ELPENHALER. After the steady state solution of the air flow field, the calculation of the tracks and fate of the powder particles followed, in order to estimate the particle residence time inside the device, the percentage of the particles that escape from ELPENHALER, as well as other qualitative characteristics that provide an evaluation of the critical operational parameters.

Fig. 4 shows the volume percentage particle size distribution. The powder diameter ranges between 0.9 / 224.4 [μm], however the active substance is contained only in particles with diameter up to approximately 7 [μm]; the larger particles correspond to the carrier, which is necessary for the avoidance of small particle agglomeration.

Turbulent air flow with swirl inside the device is highly desirable, in order to achieve the fast, complete and with minimal inhaling effort transportation of the powder from blister to mouthpiece outlet. This is greatly accomplished by the aerodynamic design of ELPENHALER, as clearly indicated by Figs. 5 and 6, where the flow path lines and the distribution of air turbulence intensity are presented. The device-average air velocity is calculated at 11 [m/s], with its maximum value reaching as high as 65 [m/s]. The CFD prediction for the total air flow rate is 57.7 [lt/min], compared to the measured 59.5 [lt/min], resulting in a low relative error of -3.0 [%].

After the calculation of the air flow field is completed, the powder particle tracks are estimated, followed by their statistical analysis to evaluate the device performance. Fig. 7 illustrates the time-average and statistical maximum residence time of the particles, as a function of their diameter. All particles are computed to have escaped from the device in 0.2 [s], a time interval much less than the actual duration of patient’s inhale. The statistical analysis of the particle tracks predicts that only 0.3 [%] of the powder remains inside the device. The time-averaged, as well as three random tracks of the powder particles with diameter of 1.553 [μm], corresponding to active substance, are presented in Fig. 8.

The performed CFD study proved that both ELPENHALER versions achieve very high delivery performance, without the need for user operation, except the positioning of the medicine blisters in the device. The CFD study certified with the most appropriate methodology the observed high efficiency of ELPENHALER and simultaneously provided to ELPEN scientists valuable information regarding the structure and characteristics of the air flow, as well as the tracks and fate of the medicine particles.