On particle size distributions of aerosol sprays 645 lung (e.g. for use in the mining industry). An estimation of the inhalation hazard of an aerosol spray cannot be made by merely determining the particle size distribution or the respirable fraction, since the relative toxicity of the sprayed material must be con- sidered. Although particle size parameters are of use in rapid screening of the potential inhalation characteristics of aerosol products, they need to be interpreted with the guid- ance of a toxicologist and they cannot replace biological testing. METHODOLOGY FOR THE PRESENT STUDY The purpose of the present study is to show the effect of some formulation and hardware variables on the particle size of some model aerosol systems. The particle sizes were deter- mined by means of laser light scattering and optical array imaging systems (designed and supplied by Particle Measuring Systems Inc., Boulder, Colorado, USA) and cali- brated with monodisperse polystyrene latices. The principle of operation of these two systems has been published elsewhere (19). These systems are ideally suited for rapidly determining the airborne size distributions of aerosol sprays and they are illustrated in Figs 1-3. As the measurements are made on airborne particles, rather than captured particles, they are of particular value for studying cosmetic aerosols containing substan- tial amounts of volatile materials (e.g. propellants). To the authors' knowledge this is the first published measurement of this type by this method. Optical • Ss'em/ i ....... •//J},,• • A ! •/•/•,•A• Array Source • •+'•/'•""•••••••••••••••••• g '•/•• Particle • Figure 1. The optical array system. Photodetectors Optical System Figure 2. The light scattering system. Particle Laser Source

646 R. W. Pengilly and J. A. Keiner The effective range of the light scattering system was 1.7-14 gm and two measure- ment ranges of the optical array system were employed (3-3-50 and 20-300 gin). Each range was subdivided into fifteen separate size channels, from which number/size channel information was obtained. The particle size distributions were recorded by spraying the aerosol for five seconds into the appropriate probes. With the actuator orifice and the axis of the laser beam in one horizontal plane, distances were selected at 28 and 48 cm from the actuator orifice to the laser beam axis. This was because of experimental convenience and because the distributions could be expected to exhibit significant differences due to evaporation of the volatile components of the particles. The prevailing conditions of measurement were 20øC and 50 •o Relative Humidity. Two model aerosol formulations were selected for this work, based on 15/85 and 35/65 product/propellant ratios. The formulations are detailed in Table II. The valve/ actuator systems selected are listed in Table III these were selected to include a range of valve/actuator variables used in aerosol formulation studies. These variables included the size of the valve housing orifice and the presence or absence of a vapour phase tap in the valve. The actuators employed were a mechanical break-up (four-channel swirl chamber) system and a standard 0.015-inch spray orifice button. The particle sizing systems were calibrated with monodisperse polystyrene latices (0.5-2 gm diameter) and with glass beads (50-300 gm diameter). Table II. Aerosol formulations (a) 35/65 Product/Propellant System •o w/w 'Resyn' 28-2930 * 1'10 Amino-methyl propanediol '[' 0'11 Ethanol 33'79 Fluorotrichloromethane (3) :l: 42.20 Dichlorodifluoromethane (3) :l: 22.80 lOO.OO (b) 15/85 Product/Propellant System •oW/W 'Resyn' 28-2930 * 1.10 Amino-methyl propanediol •' 0.11 Ethanol 13.79 Fluorotrichloromethane :[: 55.25 Dichlorodifluoromethane/ 29.75 100.00 * National Starch and Resin Corporation Inc. '1' British Drug Houses Laboratory Reagent. :l: Imperial Chemical Industries (Mond Division).