10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Aerosol Deposition in the Sampling Train of PM CEMS
Yu-Mei Kuo, Shi-Bo Wang, Chih-Wei Lin, Sheng-Hsiu Huang, Hsien-Shiow Tsai, CHIH-CHIEH CHEN, National Taiwan University
Abstract Number: 672 Working Group: Instrumentation
Abstract There are several PM CEMS currently commercially available. From the standpoint of aerosol sampling, the performance of PM CEMS might be plagued with aerosol deposition in the sampling train, this potential and significant defect of PM CEMS has never been addressed before. The aims of this study were to evaluate the particle deposition in the sampling train and to improve the sampling efficiency.
In aerosol penetration test apparatus, an ultrasonic atomizer was used to generate micro-meter-sized challenging NaCl particles. The aerosol output was then introduced into the test chamber through a radioactive source, 10 mCi Am-241, to neutralize aerosols to the Boltzmann charge equilibrium, before diluted with filtered dried air. An Aerodynamic Particle Sizer was employed to measure the aerosol size distributions and number concentrations upstream and downstream of the sampling probe. The PM CEMS sampling train could be divided into three parts: sampling inlet with an elbow or goose-neck, connecting straight tube, and elbow adaptor connecting to TEOM or beta-gauge. The experimental data were then compared with the empirical models of particle deposition in bends of circular cross section and gravitational deposition of particles from laminar flows in channels. For a particular type of PM CEMS, the inner diameter of the sampling tube was 10 mm and the sampling flow rate was 2.1 L/min. The goose-neck sampling inlet was composed of a 120° bend tube followed by a 30° bend tube. The length of the straight sampling tube was 192 cm and the elbow tube had a 90° bend. The air velocity ranging from 10 to 100 cm/sec was employed to study the velocity dependency.
The loss of particle loss in the typical sampling train was found to be significant. The elbow or goose neck design resulted in over 10% aerosol deposition of 10 μm. The 192 mm long straight tube also caused over 10% deposition loss. The connecting elbow adaptor caused 20% deposition loss because of higher velocity. The overall loss of 10 μm particles was up to 40%, while the 2.5 μm particles experienced 5% loss. The experimental data showed higher aerosol deposition loss than the modelled ones. The discrepancy between the experiment and model was mainly on the inertial impaction loss in the elbow. That means the effect of curvature-diameter ratio of the elbow on the deposition loss needs to be further studied. Other factors such as elastic properties of the particle and the elbow tube, the roughness might contribute to the bounce-off and make the deposition loss unstable and unpredictable.
The currently commercially available PM CEMS sampling trains were designed to follow the isokinetic protocol, but did not consider the aerosol deposition loss in the sampling train due to gravitational settling and inertial impaction. The aerosol size separator with cut-point of 1.0 μm is highly recommended because of deposition loss due to gravitational settling and inertial impaction is less than 1%. The deposition loss is less than 5% if the 2.5 μm pre-separator is used. If the pre-separator is adopted for 1.0 or 2.5 μm, the isokinetic sampling criteria might become not that critical.