以多孔性泡綿作為慣性衝擊器收集板的可行性研究

Abstract

摘要 慣性衝擊器(Inertial Impactor)根據微粒的慣性大小來分離不同氣動直徑的微粒，在微粒的分徑採樣上有很重要的應用。由文獻可知，固體微粒於衝擊板上的反彈現象往往造成微粒的損失，並且影響微粒粒徑分布量測的準確性。本研究欲以多孔性泡綿取代傳統的平板型，以解決固體微粒反彈之問題。本研究使用Tsai and Lin（1999）所設計的三種不同型式的衝擊器，將收集微粒用的多孔性泡綿下方襯以玻璃纖維濾紙置於原衝擊板上方，進行液體油酸及固體AF（Ammonium Fluorescein）微粒之實驗。使用的泡綿有710及330 ppi（pores per inches），厚3mm，直徑為25mm。 固體微粒的實驗結果顯示，對於和傳統慣性衝擊器類似的衝擊器設計一而言，使用較緻密710 ppi的多孔性泡綿，效率隨著 之增加而增加，在S/W=1時 ，當 在0.84以上時固體仍可維持在95﹪左右的收集效率（S：噴嘴到收集板的距離，W：噴嘴的直徑）；而在S/W=4時，則 在0.78以上時固體微粒的收集效率仍可維持在83﹪~ 96﹪之間，顯示使用衝擊板設計一可以使固體微粒反彈現象獲得了有效的控制。但是使用較鬆的330 ppi的多孔性泡綿則收集效率會較差，在S/W=1時， =0.52時收集效率可達最高值約為72﹪，當 在0.52以上時，收集效率維持在60 ~ 65﹪左右。在S/W=4時，收集效率比較好一些， 在0.78以上時仍可維持在70 ~ 74﹪左右的收集效率。 對於在收集板上方有個孔口板的設計二而言，使用710 ppi的多孔性泡綿，在S/W=1時， 在0.84以上時收集效率可維持在85﹪以上；在S/W=4時，則 在0.76以上時收集效率在45﹪~ 75﹪之間。使用330ppi的多孔性泡綿，在S/W=1時， 在0.86以上時，收集效率會隨著 增加而增加， 在1.34時可達74﹪左右；而在S/W=4時，則 在0.76以上時收集效率在降至37﹪~ 64﹪之間。 對於設計三的衝擊杯型（impaction cup）的慣性衝擊器而言，實驗結果顯示，收集效率較設計一及設計二低很多，無論是否使用多孔性泡綿，在 為0.41 ~ 1.34收集效率均維持在50﹪以下，多孔性泡綿並無法改善和提高固體微粒的收集效率。 綜上所述，在適當的慣性衝擊器設計下（如設計一）多孔性泡綿用於慣性衝擊器之收集板時，可以解決固體微粒反彈的問題，主要是因為部份流體會穿入泡綿中，使打在泡綿上之微粒不易反彈離開泡綿收集板，加上泡綿上有許多曲折的孔洞，固體微粒因慣性力進入孔洞內被過濾機制捕捉之機會會增加，因此 大時，收集效率仍可維持很高。本研究也發現，由於泡綿的過濾作用，同一微粒的收集效率會高於傳統的收集板，此過濾機制也使得收集效率與 關係圖上的收集效率曲線向左偏移。

Abstract Inertial impactors is used to separate particles of different aerodynamic sizes based on particle inertia. It has very important applications in particle size-selective sampling. According to literature, losses of particles due to solid particle bounce from the impaction plate influences the accuracy of particle size distribution measurement. This research intends to use the porous foam to replace the traditional solid impaction plate to eliminate particle bounce problems. The study used the three impactor designs by Tsai and Lin (1999). The porous foam was backed by a glass fiber filter and placed on top of the original impactor plate. Both liquid oleic acid and solid AF (Ammonium Fluorescien) particles were used for the experiment. Two different porous foams: 710 and 330 ppi (pores per inch squared), 3 mm in thickness, 25 mm in diameter 25, were used in the test. Results of solid particle test show that in the design No. 1, which is similar to a traditional impactor, the collection efficiency of the 710 ppi porous foam (a more closely packed foam) increases with an increasing . It remains as high as 95 % when is greater than 0.84 at S/W=1.0 (S: jet to plate distance, W: nozzle diameter); and remains as high as 83﹪~ 97﹪when is greater than 0.78 at at S/W=4.0. That is, solid particle bounce is controlled effectively in this design. However, for the 330 ppi porous foam, the collection efficiency for solid particles is not as good. At S/W=1, the collection efficiency peaks at about 72 % when =0.52, then it stays from 60-65 % when is greater than 0.52. At S/W=4, the collection efficiency is slightly better, it is from 70-74 % when is greater than 0.78。 For the design No. 2, use of the 710 ppi porous foam results in a solid particle collection efficiency greater than 85 % when is greater 0.84 at S/W=1. However, the collection efficiency of the design No.2 is not as good as the design No.1. When S/W=4, the collection efficiency increases from 44 ﹪to 74 ﹪when increases from 0.76 to 1.34. While when the 330 ppi porous foam was used, the collection efficiency increases with an increasing and peaks at 74 % when =1.34. At S/W=4, the collection efficiency is between 37 ﹪and 64 ﹪when is increased from 0.76 to 1.34. For the particle-trap like design No. 3, the experiment data show that the solid particle collection efficiency is much lower than both the design No. 1 and 2 whether or not the porous foam was used. Although the collection efficiency also increases with an increasing , it is less than 50 % when is increased from 0.41 to 1.34. That is, the porous form can't improve the solid particle collection efficiency for this design. In summary, for a proper design of an inertia impactor (such as the design No. 1), use of the porous foam as the impaction plate resolves solid particle problems because of partial penetration of air flow into the foam. As a result, the particles impacting on the foam do not rebound easily. In addition, the porous structure of the foam helps capture particles by filtration mechanism as particles enter the foam by inertia. Therefore, the collection efficiency remains high for a large . This study has also found that the collection efficiency for the impactor with the porous foam is higher than that with the traditional impaction plate given that the particle size is the same. As a result, the collection efficiency curve shifts to the left of the traditional Marple's collection efficiency curve when is the abscissa.