一個控制細微粒及奈米微粒排放的電極線-平板型濕式靜電集塵器

Abstract

本研究設計並開發一個高效率電極線-平板型濕式靜電集塵器用以控制細微粒及奈米微粒污染物。本集塵器之寬為75 mm，有效微粒沈降長度及氣膠通道間隙分別為48 mm及9.0 mm。本濕式靜電集塵器之收集板表面經噴砂處理後塗敷TiO2奈米微粒，以提高表面親水性，用以取代親水性薄膜的設計；電暈放電電極由三條黃金電極線(直徑為100μm)所構成；一個氣體脈衝閥用以連續且規律的清潔附著在電極線表面之粉塵，使集塵器維持最佳的集塵效率。本濕式靜電集塵器之設計目的在於解決以下傳統乾式靜電集塵器在使用上之問題：粉塵沈降於收集板表面及放電電極上使集塵效率下降、背電暈及粉塵再揚起。最後，我們在起始乾淨條件下及經過高濃度粉塵負荷後進行濕式靜電集塵器集塵效率測試，並與乾式靜電集塵器之測試結果做比較。 實驗結果顯示當施加電壓為4.3 kV、氣膠流量為5 L/min（微粒停留時間為0.39 s）時，本濕式靜電集塵器在起始乾淨狀況下對16.8-615 nm電移動度粒徑之奈米微粒去除效率為96.9-99.7 %。在相同操作條件下經過1.2±0.06 g/plate之TiO2高粉塵濃度負荷後，濕式集塵器對於16.8-615 nm之玉米油微粒之去除效率仍可維持在94.7-99.0 %。 本研究首先建立了一個拉式數值模式來推估靜電集塵器對粒徑分佈為0.1≦dp≦10 μm之微粒的集塵效率。微粒的充電及運動方程式係利用四階Runge-Kutta法來求解，以求得微粒的帶電量及微粒移動軌跡。比對結果發現，模擬出的收集效率與Huang and Chen (2002)及Chang and Bai (1999)的實驗值相符，誤差範圍分別為0.68-14.57 %及1.49-12.46 %。但此模式僅適用於推估靜電集塵器對dp≧100 nm之微粒的集塵效率，無法有效計算出部分充電效應對奈米微粒收集效率的影響。 為了準確預估單極電極線-平板靜電集塵器對奈米微粒(dp≦100 nm)的收集效率，本研究建立了一個2維數值歐拉模式來模擬電場、離子濃度分佈，及微粒充電量。集塵器內部的流場係利用SIMPLER方法來計算之，而電場強度及離子濃度分佈則利用Poisson方程式及對流擴散方程式來求解。帶電微粒的濃度分佈及微粒去除效率分別利用對流擴散方程式及Fuch充電理論進行計算。 針對6-100 nm之奈米微粒，本研究模擬之微粒收集效率與Huang and Chen (2002)的實驗數據比對結果發現，模擬值與實驗值吻合(氣膠流量: 100 L/min, 施加電壓: -15.5~-21.5 kV)。進一步比對模擬的收集效率值與濕式靜電集塵器(Lin et al. 2010)的收集效率實驗值結果發現，模擬值與集塵器對單徑分佈、粒徑為10及50 nm的食鹽微粒及多徑分佈、粒徑分佈範圍為5.23-107.5 nm之銀微粒的去除效率實驗值相符(氣膠流量: 5 L/min, 施加電壓: +3.6~+4.3 kV)。 本單極電極線-平板型濕式靜電集塵器在高濃度粉塵負荷下可有效的去除細微粒及奈米微粒。預期本濕式靜電集塵器可作為高效率微粒去除設備及奈米微粒採樣器。本研究建立的數值模式可用於設計大型的單極平板濕式靜電集塵器，以解決傳統乾式靜電集塵器在實際操作上所面臨的問題。

In this study, an efficient parallel-plate single-stage wet electrostatic precipitator (wet ESP) with a width of 75 mm, effective precipitation length of 48 mm and gap of 9.0 mm was designed and tested to control fine and nanosized particles without the need of rapping. The collection plates are made of sand-blasted copper plates coated with TiO2 nanopowder instead of hydrophilic membranes. Three gold wires (diameter: 100 µm) were used as the discharge electrodes and a pulse jet valve was used to regularly purge the wires. The design of the present wet ESP is aimed at solving the problems of traditional dry ESPs: reduction of the collection efficiency due to particle deposition on the discharge electrodes and collection electrodes, back corona, and particle re-entrainment. The collection efficiency at initially clean and heavy particle loading conditions was tested and compared to a similar dry ESP. Experimental results showed that when the wet ESP was initially clean, the particle collection efficiency ranged from 96.9-99.7% for particles ranging from 16.8 to 615 nm in electrical mobility diameter at an aerosol flow rate of 5 L/min (residence time of 0.39 sec) and an applied voltage of 4.3 kV. After heavy loading with TiO2 nanopowder about 1.2±0.06 g/plate, the collection efficiency of the present wet ESP for corn oil particles was shown to reduce only slightly to 94.7-99.0 % for particles from 16.8 to 615 nm in diameter. In order to predict collection efficiency for particles with 0.1≦dp≦10 μm, a numerical model based on Lagrangian method was developed. The equation of particle charging and particle motion were solved by using the fourth order Runge-Kutta integration to obtain particle charges and trajectory. The simulated particle collection efficiencies based on Lagrangian method were shown to agree reasonably with the experimental data in Huang and Chen (2002) and Chang and Bai (1999) with deviation of 0.68-14.57 % and 1.49-12.46 %, respectively. This model, which can’t be used to predict partial charging effect on nanoparticle collection efficiency, is appropriate to be used to predict collection efficiency of ESPs for particles with dp≧100 nm In order to predict nanoparticle (dp≦100 nm) collection efficiency in single-stage wire-in-plate ESPs accurately, the electric field, the ion concentration distribution, and the particle charge were obtained by a 2-D numerical model based on the Eulerian method. Laminar flow field was solved by using the Semi-Implicit Method for Pressure-Linked Equation (SIMPLER Method), while electric field strength and ion concentration distribution were solved based on Poisson and convection-diffusion equations, respectively. The charged particle concentration distribution and the particle collection efficiency were calculated based on the convection-diffusion equation with particle charging calculated by Fuchs diffusion charging theory. The simulated collection efficiencies of 6-100 nm nanoparticles were compared with the experimental data of Huang and Chen (2002) for a wire-in-plate dry ESP (aerosol flow rate: 100 L/min, applied voltage: -15.5~-21.5 kV). Good agreement was obtained. The simulated particle collection efficiencies were further shown to agree with the experimental data obtained in the study for a wire-in-plate wet ESP (Lin et al. 2010) (aerosol flow rate: 5 L/min, applied voltage: +3.6~+4.3 kV) using monodisperse NaCl particles of 10 and 50 nm in diameter, and polydisperse Ag particles of 5.23-107.5 nm in diameter. The present single-stage wire-in-plate wet ESP was found to remove fine and nanosized particle effectively under heavy loading conditions. It is expected that the present wet ESP could be used as efficient particle removal equipment or a nanoparticle sampler. The present model could facilitate the design and scale-up of the single-stage wire-in-plate wet ESP to control fine and nanosized particles without the typical problems associated with dry ESPs.