以α-矽鉬黃間歇式流動分析法測定自然水體中矽酸鹽之研究

Abstract

本研究探討以鉬酸銨測定水中矽酸鹽方法中黃色矽鉬複合物之反應動力學，本文改變了矽鉬黃複合物呈色條件，測定靈敏度較低的α-矽鉬黃複合物為主。其優點為：α-矽鉬黃複合物呈色穩定且受溫、鹽度影響與磷酸鹽干擾較小。但在應用至自動分析前，則必須改善其反應在常溫下較慢的缺點。當要求靈敏度時可採用較長的光徑或在較低的波長測定。因不使用漂白劑，此法所測出之訊號應為矽酸鹽與磷酸鹽其合併吸光值： 〖Abs〗_(Si+P)^λ ={[P]×ε_p^λ +[Si]×ε_Si^λ }×b×Q_s/Q_f ×10-6+RB 其中ε_Si^λ 及ε_p^λ 為矽鉬黃及磷鉬黃複合物在波長400 nm下的摩爾吸光係數，單位為(M-1cm-1)；b為光徑，單位為(cm)；[Si]及[P]分別為樣品的矽酸鹽與磷酸鹽濃度，單位為(M)；Qs 、Qf分別為原樣與加藥後的體積比，RB為試劑空白值，因此可算出矽酸鹽濃度，單位為(μM)： [Si](μM)=(〖Abs〗_(Si+P)^λ-[P]×ε_p^λ×b×Q_s/Q_f )/(ε_Si^λ×b×Q_s/Q_f )×106 本研究發現溫度(T,oC)、鹽度(S)對於其摩爾吸光係數有些許影響，在波長400 nm測定，可用下列經驗公式修正之： ε_Si^400(T,S)(M-1cm-1) = 1166+T×(4.4-0.1532×S+0.00154×S2) 溫度(T)適用範圍20-40oC、鹽度(S)適用範圍0-35 至於磷鉬黃的摩爾吸光係數值在上述條件下，在波長400 nm測定為： ε_p^400(M-1cm-1) = 390 本研究使用醋酸鈉緩衝溶液將以上反應控制在pH=4.2-4.3之間，在常溫之下反應大約需4分鐘以上，加溫至40oC情況之下可縮短至2分鐘，其速率足以因應自動分析之要求。本文比較了兩種自動分析系統，分別為傳統連續式測定及間歇式上樣測定，後者的原理為將樣水加入試劑後，送入恆溫的流動光槽中停止幫浦運作使樣水停滯在光槽中靜待反應完成。結果發現間歇式流動分析系統較為精密準確。使用一公分長光徑測量矽酸鹽的最小解析度為1 μM，線性範圍為0-200 μM，在100 μM的濃度下的相對誤差在0.5%以下，即使在船上顛簸海況下也可操作，不但可測定外洋水柱同批不同溫度的樣品，也可適用於在河口測定同批不同溫度鹽度的樣品，分析速率可達每小時20個樣品。

This study focuses on the reaction kinetics of the yellow α-silicomolybdenum complex for the determination of silicate in natural waters. The less sensitive yellow α-silicomolybdenum reaction has several advantages: the α-silicomolybdenum color is stable; it has less salt effect and phosphate interference. However, it is still unfavorable to be employed on automated analysis due to the slow reaction rate. To solve this problem the pH of the reaction is adjusted to 4.2-4.4 by a buffer system so as to enhance the reaction completeness to less than 4 minutes under room temperature, and can be further shortened to less than 2 min by moderate heating. Since the adding of bleaching reagent is avoided, the final absorbance may contain some signals attributed to phosphate: 〖Abs〗_(Si+P)^λ ={[P]×ε_p^λ +[Si]×ε_Si^λ }×b×Q_s/Q_f ×10-6+RB Where ε_Si^λ and ε_p^λ are molar extinction coefficients of silicomolybdic acid and phosphomolybdic acid(M-1cm-1), respectively, at a wavelength of 400(nm), b the path length(cm), [Si] and [P] the concentrations for silicate and phosphate(M),Qs and Qf represent the sample and final volume ratios and RB the reagent blank. The silicate concentration should be corrected for the phosphate interference for which the phosphate concentration has been identified by a separate channel: [Si](μM)=(〖Abs〗_(Si+P)^λ-[P]×ε_p^λ×b×Q_s/Q_f )/(ε_Si^λ×b×Q_s/Q_f )×106 The molar extinction coefficient for silicate(ε_Si^λ) was formed to be functions of temperature and salinity, but that for phosphate(ε_p^λ) is not. At the wavelength of 400 nm: ε_Si^400(T,S)(M-1cm-1)= 1166+T×(4.4-0.1532×S+0.00154×S2) ε_p^400(T,S) (M-1cm-1) = 390 20 oC < T <40oC, 0< S <35 A new automated analysis system for the determination of silicate has been proposed and it has been compared with the traditional flow injection analysis. The system uptakes samples and add reagents in a quick flow manner to fill a thermostat cuvette, then the flow is stopped so that the sample is allowed to stay in the cuvette for complete reaction. Since the determination is static and the color is fully developed, it has a higher sensitivity and better precision than do conventional methods. The minimum resolution (Abs=0.001) is 1 μM Si with 1 cm cuvette and can be lowered to 0.2 μM Si if a 5 cm long cuvette is used. The linear range is up to 200 μM Si. The precision at 100 μM Si is less than 0.5%.