PARTICLE LOSSES IN GLASS HONEYCOMB DENUDER SAMPLERS

Particle loss tests have been performed on glass honeycomb denuders fo r particles in the size range 0.05-2 mum. The tests were conducted at three different How rates (1.5, 4, and 10 L/min). Fluorescent monodisp erse latex particles were used as a test aerosol. Experimental results were compared to those predicted by the theory that combines diffusio n, impaction, and interception. It was demonstrated that particle loss es reach a minimum for particles in the range 0.5-0.75 mum and become higher for smaller or larger particles. Although for every flow rate t ested the particle losses appear to be higher than the predictions of theory, they exhibit similar trends with the theoretical predictions. In all cases the particle losses on the glass honeycomb denuder were f ound to be < 5%. The effect of relative humidity on particle loss was examined at 1.5 L/min sampling flow rate. The results did not demonstr ate a dependence of the loss on the humidity. It was expected that a l ower relative humidity (10% to 20%) would accentuate losses due to ele ctrostatic charges. Particle loss tests were conducted for uncoated de nuders as well as denuders coated with citric acid. Although in both c ases the losses are small, they appear to be smaller on coated denuder s, as expected. Finally, collocated fine particle mass sampling was co nducted using the Harvard-Marple impactor and an impactor/denuder samp ler using two glass honeycombs. The mass collected on the honeycomb de nuder sampler was approximately 5% lower than on the Harvard-Marple im pactor.