傅立葉轉換紅外光譜儀之單束光譜定量方法與污染源定位方法研究

Abstract

本論文探討兩個關於開放光徑傅立葉轉換紅外光譜儀 (Open path Fourier transform infrared spectroscopy，OP-FTIR) 的議題，包括單束光譜定量和污染源定位之方法。 使用 OP-FTIR 在現場所獲得的單束光譜可以直接計算污染物質的濃度，將已知濃度的標準穿透光譜加入樣品單束光譜中，可以將分析物的光學吸收特徵圖形逐漸消除，稱為「滴定」。定義平方微分合成光譜積和 (sum of squared differential synthetic spectrum，SSDSS) 作為滴定過程所使用的指示劑，可以確認滴定過程到達終點。讓樣品光譜達到終點時所需添加的標準穿透光譜量，可以進行污染物濃度的計算。使用可追蹤含有六種已知成分濃度的標準氣體進行濃度計算驗證，以氣密樣品槽進行紅外光譜數據擷取，並比較滴定方法與古典最小平方法 (classical least square，CLS) 方法的準確度。對工業區煙囪進行 FTIR 連續監測的分析結果顯示 CLS 和滴定法的濃度趨勢是一致的。由於滴定方法在進行定量計算時不需要使用背景單束光譜，因此適合在現場開放光徑量測時使用。在環境背景中經常存在的物質，如NH3、CH4 及 CO 等，可以用單束光譜滴定法順利完成濃度計算的工作，而這些數據在傳統的 CLS 方法中，未取得理想的背景光譜時是不容易做到的。在滴定終點所產生的合成光譜是不含污染物的單束光譜，其他的條件和現場參數是完全一致，應用在紅外光譜儀中是很理想的背景光譜。 利用兩組污染玫瑰圖計算污染源機率分布圖以標定污染逸散源的位置。在一個半導體廠陽台架設兩組開放光徑紅外光譜儀進行一週左右的連續量測，解析污染物濃度並且計算各成分的污染玫瑰圖。量測現場同時架設氣象站以紀錄風速與風向。量測期間可以偵測到 CF4、C2F6、CH3OH、NH3、NO2 及 SF6，其濃度在 ppb 範圍。使用各物種前 20 % 高濃度的資料計算污染玫瑰圖。每個物種都利用兩組污染玫瑰圖計算污染源機率分布圖以標示污染源。SF6、CF4、NO2、及 C2F6 等物種的污染高點可以指到工廠的煙囪區，但是 CH3OH 和 NH3 指向不同的來源。本研究提供一個簡單而快速評估污染源的方法There are two topics for open-path Fourier transform infrared spectroscopy (OP-FTIR) discussed in this thesis, including the method of single-beam spectrum quantification and the method of pollutant emission source locating. The concentration of analytes can be directly determined from a single-beam spectrum of OP-FTIR. The peak shapes of the analytes in a single-beam spectrum were gradually cancelled (i.e., “titrated”) by dividing an aliquot of a standard transmittance spectrum with a known concentration, and the sum of the squared differential synthetic spectrum was calculated as an indicator of the end point for this titration. The quantity of a standard transmittance spectrum that is needed to reach the end point can be used to calculate the concentrations of the analytes. A traceable gas standard containing six known compounds was used to compare the quantitative accuracy of both this titration method and that of a classic least square (CLS) using a closed-cell FTIR spectrum. The continuous FTIR analysis of industrial exhausting stack showed that concentration trends were consistent between the CLS and titration methods. The titration method allowed the quantification to be performed without the need of a clean single-beam background spectrum, which was beneficial for the field measurement of OP-FTIR. Persistent constituents of the atmosphere, such as NH3, CH4 and CO, were successfully quantified using the single-beam titration method with OP-FTIR data that is normally inaccurate when using the CLS method due to the lack of a suitable background spectrum. Also, the synthetic spectrum at the titration end point contained virtually no peaks of analytes, but it did contain the remaining information needed to provide an alternative means of obtaining an ideal single-beam background for OP-FTIR. A new approach employing two pollution rose plots to locate the sources of multiple hazardous gas emissions was proposed and tested in an industrial area. The data used for constructing the pollution rose plots were obtained from two side-by-side measurements of OP-FTIR spectrometers during one week of continuous analysis on the rooftop of a semiconductor plant. Hazardous gases such as CF4, C2F6, CH3OH, NH3, NO2, and SF6 were found and quantified at the ppb level by both OP-FTIR measurement sites. The data of the top 20 % highest concentrations and associated wind directions were used to construct the pollution rose plots. Pollution source probability contours for each compound were constructed using the probability-product of directional probability from two pollution rose plots. Hot spots for SF6, CF4, NO2, and C2F6 pointed to the stack area of the plant, but the sources of CH3OH and NH3 were found outside of this plant. The influences of parameters for this approach such as the variation in wind direction, lower limit concentration threshold and the nearby buildings were discussed.