10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Counting Efficiency Evaluation of Optical Particle Counters in Micrometer Range by Using Inkjet Aerosol Generator
KENJIRO IIDA, Hiromu Sakurai, AIST
Abstract Number: 154 Working Group: Instrumentation
Abstract It is challenging to aerosolize PSL spheres whose diameter is greater than 1 μm and use them to evaluate the counting efficiencies of optical particle counters (OPCs). Particle number concentration of PSL spheres tend to drift during nebulization of liquid suspension or strongly fluctuate during dry-dispersion. Instead of aerosolizing PSL spheres previous studies often use a monodisperse aerosol generator such as vibrating orifice aerosol generator to generate test particles whose diameter is greater than 1 μm.
Regardless of the particle material the parallel comparison method has been used to evaluate the counting efficiencies of OPCs, η. During the evaluation of η the OPC under test (hereafter the DUT-OPC) samples from a distributing box a well-mixed aerosol of test particles in parallel with a reference instrument. The η is determined as the ratio of the number concentration reported by the DUT-OPC to that reported by the reference. The parallel comparison method requires that test particles are well-mixed in a distribution box and transported to the DUT-OPC and a reference instrument. However, the particles above 1 μm may not mix well in the distribution box, and also their losses in the transfer tubes from the distribution box may be different between the tube to the DUT-OPC and that to the reference. This effect becomes more significant as the difference in the sampling flowrate between the DUT-OPC and the reference increases.
This study introduces the method to evaluate the η at particle diameter greater than 1 μm by using Inkjet aerosol generator (IAG). The IAG-based method defines the η as ratio of the particle count rate of a DUT-OPC to the particle generation rate of the IAG. This count-rate-based η can be converted to a concentration-based value if the true flowrate of the DUT-OPC is known or measured. This study demonstrates the evaluations at 5 μm and 10 μm in volume equivalent diameter. The chemical composition of the particles are either sodium chloride (SC) or lactose monohydrate (LM). The aerosol flowrate of the IAG is set at 0.3 L/min, and the aerosol is delivered to an OPC (ZN-PD50-S, OMRON, Japan) whose sampling flowrate is 6 L/min and minimum detectable size can be set at either 5 μm or 10 μm. The mismatch of the flowrates is compensated by adding particle free-air in a laminar flow chamber.
In order to simulate the sampling of uniformly mixed aerosol from a real environment the particles are delivered to different points over the inlet plane of the isokinetic probe attached to the OPC. Particle flux into the isokinetic probe is proportional to the gas-velocity into the probe; however, the true velocity distribution is usually unknown. It is assumed that the true velocity distribution is bounded by two flow models: the plug and parabolic flows. A set of delivery points are prepared to simulate the particle flux under each flow model.
Experimental results show that the choice of flow model influences the value of η at 10 μm indicating the η is potentially different from the true value since the true value can be evaluated only if the true velocity distribution is known. The potential bias in is considered as a source of systematic error in our uncertainty analysis. When the minimum detectable size of the OPC is equal to the volume equivalent diameter of the test particles the value of η strongly depends on the particle material, and the greater fraction SC particles are detected since their PSL-equivalent optical diameter is greater than LM particles.