| Summary: | 碩士 === 國立交通大學 === 環境工程所 === 87 === Inertial impactors are widely used in aerosol sampling devices to determine the size distribution of particles. However, for solid particles collected by the inertial impactor, particle bounce occurs and will severely affect the measured size distribution, especially when the particle concentration is high. In this study, liquid oleic acid particles were used to investigate the influence of the parameters, such as different S/W ratios and collection surfaces (S: distance from the circular nozzle to the impaction plate; W: the diameter of circular nozzle), on the collection efficiency of the inertial impactor. The impactor operates at a flow rate of 5.0 lpm and the designed cutpoint is 2.0 mm. To study the bounce of solid particles, different types of collection plates were designed in this work study. A porous metal disc was used as the support for the filter substrate. A minor flow drawn under the porous metal disc was fixed at the 10 % of the total flow rate to improve the collection efficiency while minimizing the contamination of fine particles.
One of the designed collection plates (Design No. 1) is an open cylindrical cavity of 2 mm in depth and 16 mm in diameter. The experiment data of liquid droplets show that the collection efficiency and 50 % cutpoint are not significantly changed by different S/W ratios. ThesStk50^0.5 for the PC filter is 0.47 but is 0.40 for the glass fiber filter. Therefore, filter substrate is an important factor, which affects the sampling results.
The experiment data of solid particles show that the minor flow can reduce the bounce of particles effectively. The proper minor flow rate is 0.5 lpm and the diameter of the suction area is 5 mm . For the Design No. 1, the collection efficiency reaches a value of 80 % at theStk^0.5 of 0.70 while S/W equals 1. But the collection efficiency decreases gradually when the Stk^0.5 is above 0.70 and S/W equals 1. Another design of collection plates (Design No. 2) has an enclosed cylindrical cavity of 3.6 mm in depth , 18 mm in diameter, and has an orifice of 5 mm in diameter at the top. Under the same minor-flow suction condition as the Design No. 1, the collection efficiency of the Design No. 2 is 65 % at the of 0.70 while S/W equals 1. When the Stk^0.5 is above 0.70, the collection efficiency of the Design No. 1 and 2 both maintain at 80 % and 70 % while the S/W ratio equals 4, but both decrease while the S/W ratio equals 1.
The wall loss is mainly found on the outer surface of the nozzle when the Stk^0.5 is less than 0.70 and gradually increase as the increases. For the Design No. 1, the loss is under 2 % as the Stk^0.5 is less than 0.70 and reaches a value of 8 % as the is 1.0, regardless of the S/W ratio. The wall loss of the Design No. 2 is higher than that of the Design No. 1 and reaches a value of 12 % as the Stk^0.5 is 1.0. The increase of the S/W ratio from 1 to 4 can significantly reduce the wall loss for the Design No. 2. When the Stk^0.5 is less than 0.90, the percentage of the wall loss for S/W=1 is twice as much as S/W=4.
The Design No. 3 is an enclosed cylindrical cavity of 14 mm in depth, 18 mm in diameter, and has an orifice of 5 mm in diameter at the top. The experiment data show that the collection efficiency for the Design No. 3 is lower than the Design No. 1 and 2 and maintain at 25-30 % as the Stk^0.5 is above 0.55. The wall loss for the Design No. 3 is severer than that of the Design No. 1 and 2. Therefore, the Design No. 1 is the better design for further study and application.
|
|---|
