Abstract
In order to promote strategies for reducing the fine particulate matters (PM10/PM25) in the atmosphere in Japan, it is essential to assess the emission characteristics of PM10/PM2.5 at various stationary souroes, and it is also necessary to develop accurate measurement methods for PM10/PM25 emissions, and to prepare an early standardization of these methods. In this paper, the impaction method designated for low concentration measurement according to ISO 23210, which is still in the draft stage, is described, and several issues related to the measurement of PM10/PM2.5, as well as their countermeasures are investigated experimentally using test particles.
The main conclusions obtained from a set of experiments and considerations are as follows;
(1) Though a straight nozzle is a prerequisite for the two-stage impactor in ISO 23210, it is difficult to insert the impactor with the straight nozzle into a small measuring hole, and therefore a two-stage impactor with a 90° bend nozzle was adopted in the study. In the case of a flow velocity of 11.5m/s in the test duct, and a dust concentration of about 22mg/m3N, the amount of particulates adhered inside the 90° bend nozzle with an internal diameter of 6mm was approximately 30%. In the case these adhered particulates are neglected, the measured dust concentration decreases considerably, while it has almost no effect on the results of the particle size distribution (average diameters and the distribution curves). On the other hand, the PM10/PM2.5 measured concentrations are approximately 20-23% lower than those obtained from straight nozzle measurements.
(2) When it is necessary to correct the results for adhered particulates, the particulates may be collected after dismounting and weighed with the whole nozzle, and the measurements of the particle size distribution are corrected. Thus, it is possible to compensate the measured values for dust concentration, particle size distribution and PM10/PM2.5 due to the wall loss. In the case of such dry particles or particulates with low adhesive properties as used in this experiment particulates are easily and effectively collected using the pulse air purge onto particulates adhered inside the 90° bend nozzle.
(3) In general, an increase in the separation stages decreases re-entraining of particulates in the impaction method, as the collected particles are further divided at each stage. In this experiment, a large amount of particulates entrains from the PM10-2.5 collector in the two-stage impactor of the ISO method. Howevei re-entraining of particulates was noticeably suppressed in the three-stage impactor, in which the PM10-5.0 collector was added. A filter paper made of quartz fiber is most suitable for the collection plate because of its rough surface, and it could be also used for high temperature gas measurements and chemical analyses.
(4) Since the particle weight limit for the collection plate depends on physical characteristics of PM10-2.5 (especially, size distribution width, number concentration and adhesive property) and property of the flue gas, it is difficult to establish a certain limit value as used in the ISO method. Therefore, it is required to establish strategies for suppression of particle re-entraining through an increase in the number of separation stages.
