No abstract

Particle size of nebulized aerosols can be measured directly using laser diffraction or by evaluating aerodynamic properties by cascade impaction. As of today, there are no generally accepted standards for measuring particle size distribution from nebulizers. Laser diffraction has been questioned because of potential evaporative losses of the small particles at the edge of the plume, causing an apparent shift in the particle size distribution and thus a larger mass median diameter (MMD). When particle-sizing wet aerosols, cascade impaction may give rise to an apparent shift in the distribution, resulting in a smaller mass median aerodynamic diameter (MMAD) due to evaporative losses of aerosol droplets as they enter the impactor at ambient temperature. The modified low-flow Marple 296 Personal Cascade Impactor (MPCI) is currently being proposed as the European standard for wet aerosol analysis to minimize evaporative losses during sampling. The present study compared the particle size distribution of salbutamol and sodium cromoglycate aerosols nebulized by the Pari LC Star, using laser diffraction (Malvern Mastersizer X; MMX) and cascade impaction (Andersen Cascade Impactor [ACI] and the commercially available MPCI), which was either at ambient temperature or cooled to the nebulized aerosol temperature (10 degrees C). MMDs obtained with the MMX were virtually identical to the MMADs measured with both impactors when cooled with no significant differences in geometric standard deviation (sigma(g)). When the impactors were operated at ambient temperature, MMADs were smaller (18 to 30%) with a significantly larger sigma(g) (p < 0.05) compared to the MMX. These findings suggest that droplet distribution data for wet aerosol where evaporation process has not been minimized must be viewed with caution. There was no evidence suggesting a significant evaporative loss of small droplets from the edge of the plume during laser particle sizing. The MPCI does not minimize evaporative losses of aerosol particles during sampling.