氧化鋁擔持的金屬氧化物吸附劑在常溫下的矽甲烷去除研究

Abstract

半導體廠CVD及光電廠MOVCD的製程中，常會使用到SiH4氣體，當SiH4濃度達1 % (V/V)時在常溫下會自燃，若處理不當會威脅到廠區人員的安全。為了去除SiH4，高科技產業皆會在製程後端裝設有局部洗滌器。若SiH4和AsH3或PH3一起排放時，SEMI(Semiconductor Equipment Manufacturer Institute)不建議使用濕式洗滌法和高溫氧化法來去除SiH4。所以本研究利用共沉澱法製備氧化鋁擔持CuO、ZnO及CeO2等不同金屬氧化物之吸附劑，並依不同比例混合，來探討常溫常壓下對SiH4的吸附效果。 本研究的實驗系統分為氣體供應裝置、測試腔和氣體分析系統等三部份。實驗所使用之SiH4氣體流量由MFC控制在0.19~0.25 L/min。測試腔採用不鏽鋼製，內徑為2.45 cm，吸附床長度為3.1~10.6 cm，接觸時間4.9~9.5秒，吸附床底部有一40 mesh的不鏽鋼篩網。本研究使用FTIR量測吸附床前後之SiH4濃度，決定吸附效率及吸附容量，最後在系統的後端加裝一個商用的乾式吸附系統來處理廢氣。為了解吸附機制和吸附前後的物化特性，以比表面積分析儀、SEM附設EDS、XRPD及ESCA等分析儀器作進一步分析。 結果顯示，除了CeO2/Al2O3之外，本研究所製備出的所有吸附劑，在常溫下對於SiH4在破出之前的去除效率皆大於99.9%。混合活性成分吸附劑的吸附容量較單一活性成分的高，吸附容量依序為CuO- CeO2/Al2O3 > CuO- ZnO /Al2O3 > CuO/Al2O3 >ZnO/Al2O3 >>CeO2/Al2O3。當CuO和ZnO或CuO和CeO2的重量百分比為2:1時(活性成分總含量為40 wt%)，吸附容量最好，分別為0.23 mole SiH4/mole M和0.27 mole SiH4/mole M (M為所有活性金屬氧化物)。吸附劑和氣體的接觸時間在4.9秒到9.5秒之間，對吸附容量並不會有影響。XRPD圖中，CuO-CeO2/Al2O3吸附劑無峰值的出現，表示CuO和CeO2可以在氧化鋁擔體上散佈的很好。由ESCA的分析結果，可知活性成分CuO、ZnO及CeO2在吸附後會被SiH4還原，以Cu(Cu2O)、Zn及Ce等元素態的形式存在於吸附劑上，而SH4反應後的產物為Si和SiO2。

Silane (SiH4) is frequently used in some processes of semiconductor and optoelectronics factories. And it is a pyrophoric gas which ignites automatically when its concentration reaches 1% (V/V) at ambient temperature, causing safety hazard to personal in the factory. Therefore, high-tech industries usually use local scrubbers to abate the emission of SiH4 right after the process equipment. However, the wet scrubbing and thermal oxidation method are not suggested by SEMI (Semiconductor Equipment Manufacturer Institute) when the flue gas contains AsH3, PH3, and SiH4 simultaneously. In this study the co-precipitation method were used to synthesize alumina supported metal oxide adsorbents including CuO, ZnO, and CeO2. The adsorption performance of synthesized alumina supported adsorbents containing different mixture of metal oxide was investigated at ambient temperature. Apparatus used in the present study consisted of gas supply device, test chamber, and gas analysis system. The flow rate of SiH4 was tuned to 0.19 L/min ~ 0.25 L/min by a mass flow controller. The test chamber is made of stainless steel with the inside diameter of 2.45mm. The length of the adsorption bed is 3.1~10.6 cm. The contact time of the adsorbent and gas is 4.9~9.5 seconds. And the Fourier transform infrared spectrometer (FTIR) was used to measure the upstream and downstream concentration of the adsorption bed, and determine the adsorption efficiency and the adsorption capacity. A commercial dry adsorbent system was installed at the end experimental system to treat the exhaust gas. Analysis instruments such as BET, SEM equipped with EDS, XPRD and ESCA, were used to characterize the mechanism of the adsorption and the adsorbent properties before and after adsorption. Results show that the SiH4 removal efficiency of all synthesized adsorbents is higher than 99.9 % before breakthrough except the CeO2/Al2O3 adsorbent. The adsorption capacity of the mixed metal oxides is better than single active component, and the adsorptive capacity was found to be in the following order: CuO-CeO2/Al2O3 > CuO-ZnO/Al2O3 > CuO/Al2O3 > ZnO/Al2O3 >> CeO2/Al2O3. When the weight ratio of CuO/ZnO and CuO/CeO2 is 2:1, the adsorptive capacity reaches a maximum value of 0.23 mole SiH4/mole M and 0.27 mole SiH4/mole M (M is all of the active metal oxide), respectively. Moreover, it was found that there is no influence on the adsorptive capacity for contact time between 4.9 to 9.5 seconds. In the XRPD experiments, no significant diffraction peak was found in the CuO-CeO2/Al2O3 adsorbent. It suggests that the binary mixture of CuO and CeO2 disperses well on the support. Based on the findings of ESCA, the active component CuO, ZnO, and CeO2 as the form of Cu (or Cu2O), Zn, and Ce existing on the adsorbent were reduced by SiH4 after adsorption. And the SiH4 were oxidized into Si and SiO2.