IN-SITU LIGHT-SCATTERING MEASUREMENTS OF MAINSTREAM AND SIDESTREAM CIGARETTE-SMOKE

This paper presents in situ and continuous size measurements for submi cron cigarette smoke particles. The method, which can be applied to an y ensemble of small particles narrowly distributed in size, uses the l ight scattered from the particles at angles of 60 degrees and 120 degr ees and the ratio of polarization components of the scattered light at 55 degrees to determine the mean particle size. Polystyrene latex sph eres diluted in distilled water are used to calibrate the system. Good agreement exists between the experimental measurements and the theore tical calculations for the calibration particles. Based on the assumpt ion that cigarette smoke particles cluster into locally uniform parcel s during their formation as condensate, the light scattering system is employed to measure the mean size and size fluctuations of mainstream smoke and sidestream smoke from research cigarettes (1R3 and 1R3f). S imilar light scattering techniques have been employed by other researc hers, but this paper describes in situ measurements of smoke without a ging or dilution. These in situ measurements characterize the spatial inhomogeneity of cigarette smoke, and show that sidestream smoke is ap proximately 0.27 mu m in count median diameter (CMD) with a geometric standard deviation (GSD) of 1.60 and that mainstream smoke has a CMD o f approximately 0.15 mu m with a GSD of 2.0, The corresponding mass me dian diameters are approximately 0.50 and 0.70 mu m, respectively. Int erestingly, there is little difference in the mean size of sidestream smoke fresh from the tip of a smoldering cigarette and sidestream smok e captured in a vessel. There are differences, however, in the cluster ing of the smoke and in its nonuniformity. Fresh sidestream smoke exhi bits more size fluctuation, but less fluctuation in local particle num ber density when compared to captured sidestream smoke. The measuremen ts also indicate little difference between mainstream smoke from cigar ettes with and without filters. Based on the assumption that the smoke partitions into small parcels with relatively narrow size distributio ns within parcels, the calculated relative number density of fresh sid estream smoke particles is fairly uniform, while the number density fl uctuates significantly for mainstream and captured sidestream smoke.