On the prediction of the phase distribution of bubbly flow in a horizontal pipe.

Horizontal bubbly flow is widely encountered in various industrial systems because of its ability to provide large interfacial areas for heat and mass transfer. Nonetheless, this particular flow orientation has received less attention when compared to vertical bubbly flow. Owing to the strong influence due to buoyancy, the migration of dispersed bubbles towards the top wall of the horizontal pipe generally causes a highly asymmetrical internal phase distributions, which are not experienced in vertical bubbly flow. In this study, the internal phase distribution of air-water bubbly flow in a long horizontal pipe with an inner diameter of 50.3 mm has been predicted using the population balance model based on direct quadrature method of moments (DQMOM) and multiple-size group (MUSIG) model. The predicted local radial distributions of gas void fraction, liquid velocity and interfacial area concentration have been validated against the experimental data of Kocamustafaogullari and Huang (1994). In general, satisfactory agreements between predicted and measured results were achieved. The numerical results indicated that the gas void fraction and interfacial area concentration have a unique internal structure with a prevailing maximum peak near the top wall of the pipe due to buoyancy effect.

Comparison of micron- and nanoparticle deposition patterns in a realistic human nasal cavity.

Knowledge regarding particle deposition processes in the nasal cavity is important in aerosol therapy and inhalation toxicology applications. This paper presents a comparative study of the deposition of micron and submicron particles under different steady laminar flow rates using a Lagrangian approach. A computational model of a nasal cavity geometry was developed from CT scans and the simulation of the fluid and particle flow within the airway was performed using the commercial software GAMBIT and FLUENT. The air flow patterns in the nasal cavities and the detailed local deposition patterns of micron and submicron particles were presented and discussed. It was found that the majority of micron particles are deposited near the nasal valve region and some micron particles are deposited on the septum wall in the turbinate region. The deposition patterns of micron particles in the left cavity are different compared with that in the right one especially in the turbinate regions. In contrast, the deposition for nanoparticles shows a moderately even distribution of particles throughout the airway. Furthermore the particles releasing position obviously influences the local deposition patterns. The influence of the particle releasing position is mainly shown near the nasal valve region for micron particle deposition, while for submicron particles deposition, both the nasal valve and turbinate region are influenced. The results of the paper are valuable in aerosol therapy and inhalation toxicology.

Optimising nasal spray parameters for efficient drug delivery using computational fluid dynamics.

Experimental images from particle/droplet image analyser (PDIA) and particle image velocimetry (PIV) imaging techniques of particle formation from a nasal spray device were taken to determine critical parameters for the study and design of effective nasal drug delivery devices. The critical parameters found were particle size, diameter of spray cone at a break-up length and a spray cone angle. A range of values for each of the parameters were ascertained through imaging analysis which were then transposed into initial particle boundary conditions for particle flow simulation within the nasal cavity by using Computational Fluid Dynamics software. An Eulerian-Lagrangian scheme was utilised to track mono-dispersed particles (10 and 20 microm) at a breathing rate of 10 L/min. The results from this qualitative study aim to assist the pharmaceutical industry to improve and help guide the design of nasal spray devices.

Deposition of inhaled wood dust in the nasal cavity.

Detailed deposition patterns of inhaled wood dust in an anatomically accurate nasal cavity were investigated using computational fluid dynamics (CFD) techniques. Three wood dusts, pine dust, heavy oak dust, and light oak dust, with a particle size distribution generated by machining (Chung et al., 2000), were simulated at an inhalation flow rate of 10 L/min. It was found that the major particle deposition sites were the nasal valve region and anterior section of the middle turbinate. Wood dust depositing in these regions is physiologically removed much more slowly than in other regions. This leads to the surrounding layer of soft tissues being damaged by the deposited particles during continuous exposure to wood dust. Additionally, it was found that pine dust had a higher deposition efficiency in the nasal cavity than the two oak dusts, due to the fact that it comprises a higher proportion of larger sized particles. Therefore, this indicates that dusts with a large amount of fine particles, such as those generated by sanding, may penetrate the nasal cavity and travel further into the lung.

Flow and particle deposition patterns in a realistic human double bifurcation airway model.

Velocity profiles, local deposition efficiencies (DE), and deposition patterns of aerosol particles in the first three generations (i.e., double bifurcations) of an airway model have been simulated numerically, in which the airway model was constructed from computed tomography (CT) scan data of real human tracheobronchial airways. Three steady inhalation conditions, 15, 30, and 60 L/min, were simulated and a range of micrometer particle sizes (1-20 mum diameter) were injected into the model. Results were then compared with experimental and other numerical results which had employed either similar model geometry or test conditions. The effects of inhalation conditions on velocity profiles and particle deposition were studied. The data indicated that the local deposition efficiencies in the first bifurcation increased with a rise in the Stokes number (St) within St range from 0.0004 to 0.7. Within the same St range, DE in the second bifurcations (both left and right) was dropped dramatically after St increased to 0.17. Also, the second bifurcation in the right side (B2.1, closer to first bifurcation than left side, B2.2) was found to show a much higher (almost double) DE than the left side. This may be due to the fact that the left main bronchus is longer and has greater angulation than the right main bronchus. Generally, the present simulation using a computational fluid dynamic (CFD) technique obtained concurrent results with subtle differences compared to other works. However, due to omission of larynx in the model, which is known to significantly modify airflow and hence particle deposition, the present model may only serve as the "stepping stone" to simulating and analyzing dose-response or inhalation risk assessment visually for clinical researchers.