負載型FeOOH流體化床的開發：可行性、反應動力及最適化的研究

Abstract

高級氧化程序是利用氫氧自由基的強氧化力來分解水中的有機物，在各種高級氧化程序中，Fenton試劑（H2O2/Fe2+）已被證實是有效且簡單的氧化劑，但其缺點是在反應過程中會產生大量的氫氧化鐵污泥，還需進一步作固液分離及污泥處置。為解決此問題，以H2O2為氧化劑，以氧化鐵作為異相觸媒的氧化程序值得進一步研究。 本研究開發了一種負載式FeOOH觸媒流體化床的新技術，流體化床因其具有高質傳效率及容易固液分離等優點而較適用。在本論文中，包括負載式FeOOH觸媒（supported FeOOH）的製備、利用批式反應槽進行氧化苯甲酸 (benzoic acid, BA) 的可行性研究、以及利用流體化床進行分解H2O2及氧化苯甲酸的動力學研究都被詳細探討。最後並將此新技術應用來處理一種實際的染整廢水，並尋求其最適操作條件。 實驗中所使用的新式觸媒是在流體化床結晶槽中，利用H2O2氧化Fe2+製備而成的負載式FeOOH，觸媒表面的主要成分經分析證實為非結晶型FeOOH及□-FeOOH。為尋求流體化床結晶槽較適當的反應物濃度範圍，首先以批式實驗界定介穩區的範圍以避免初成核的形成，負載式FeOOH的表面特性也一一被研究，包括利用電子顯微鏡及梅思堡光譜所得的表面結構、總鐵量、草酸可溶解的鐵量、以及比表面積等。 本研究共合成了三種負載式FeOOH：FeOOH I是在pH 3.5時製備，FeOOH II是將FeOOH I置於pH 13溶液中所得到的老化氧化鐵，FeOOH III則是在pH 5.0時製備而成。FeOOH I用於後續所有的實驗中， FeOOH II和FeOOH III則與FeOOH I的催化性能進行比較，結果顯示氧化鐵的晶體特性對其催化性能的影響很大。 在負載式FeOOH的製備條件方面，pH、上流速度、鐵的比表面負荷及H2O2的進流濃度對進流水Fe2+的結晶比例等效應都被詳細研究。結果發現除上流速度影響較小外， pH對Fe2+的結晶比例及所合成的氧化鐵型態影響均相當大，其餘二者僅會影響Fe2+的結晶比例。 在分解H2O2及氧化苯甲酸的可行性研究方面，此部分的實驗是在批式反應槽中進行。首先，在未添加苯甲酸的情形下探討H2O2的分解動力行為，然後研究pH、H2O2加藥量及觸媒濃度對氧化苯甲酸的影響。實驗顯示在初始pH (pHi) 3.2時，苯甲酸的總處理效率較pHi 6.0及pHi 10.0時好，此乃因FeOOH在酸性條件下會還原溶解 (reductive dissolution) 出Fe2+所致；但初期Fe2+溶出量尚低時，應屬於異相催化反應主導，此時pHi 3.2有較高的反應速率則是由於有較多未解離的苯甲酸被FeOOH吸附所致。因此我們依據Fe2+溶出量加入等量的Fe2+來進行控制實驗，以瞭解異相催化所扮演的角色，由此得知苯甲酸的氧化大部分發生在觸媒的表面上，少部分則因溶出的Fe2+而發生在溶液中。 接著在連續式流體化床中探討未添加苯甲酸時分解H2O2的動力學，包括pH、H2O2濃度及觸媒濃度對H2O2分解的影響。我們假設迴流式流體化床的反應行為與連續攪拌式反應槽（CSTR）類似，在低H2O2濃度時，H2O2的異相催化分解速率和H2O2濃度及觸媒濃度成正比，但在高濃度H2O2條件下，催化活性會受抑制，此可用基質抑制模式來模擬。另外，H2O2的分解速率亦隨pH的提高而增加，此可以觸媒表面不同的物種與H2O2有不同的反應速率來解釋。 在連續式流體化床中氧化苯甲酸的部份，除了探討pH、H2O2濃度及苯甲酸濃度對其影響外，另一個目的是要釐清同相催化與異相催化所佔的比例。苯甲酸的氧化速率與苯甲酸及H2O2濃度成正比；在pH較低時，苯甲酸的去除率較高，此亦可用觸媒表面不同的物種有不同的異相催化反應速率來解釋，且經數學模式的推導，算出□FeIIIOH2+的異相催化反應速率遠較□FeIIIOH為高。由此模式可計算出不同pH下異相催化和同相催化所佔的反應比例：在pH 4.4 ~ 7.0時，苯甲酸的氧化均是異相催化的效應，pH < 4.4以後同相催化所佔的重要性愈來愈高。 另外，Fe2+對負載式FeOOH觸媒流體化床的效應亦被探討，我們並將外加Fe2+的程序稱為流體化床-Fenton法 (FBR-Fenton method)，其中Fe2+作為同相觸媒，FeOOH作為異相觸媒來催化氧化苯甲酸。經由實驗顯示：在適當的操作條件下，苯甲酸的礦化和Fe(III)的結晶可同時在流體化床中進行，且二者均與pH及進流Fe2+濃度密切相關。我們並推測了流體化床-Fenton法可能的反應機構，包括FeOOH的還原溶解及合成，以及苯甲酸的氧化三部分。 最後，將流體化床-Fenton法用來處理染整廢水，並利用實驗設計法尋求其最適操作條件，在此用到的實驗設計法有部分因子實驗法、中心組合設計法及應答區面法。由部分因子實驗法找出了對TOC、Fe及色度去除率影響最大的三個因素，再應用中心組合設計法及應答區面法，依據此三個因素求出TOC及Fe去除率的迴歸模式。由於最適化牽涉到不同操作參數及效應（即TOC及Fe去除率），而各效應的最適條件不盡相同，因此以五個指標 (pH、H2O2加藥量/COD、Fe2+/H2O2、 TOC及Fe去除率) 決定望想函數 (Derringer’s desirability function)，來尋求最適的操作條件，且經由確認實驗證實迴歸分析與實驗的結果相當吻合。研究中並證明流體化床-Fenton法比傳統Fenton法的TOC去除率為高，且有相當好的Fe去除率，亦即可解決傳統Fenton法為人詬病的污泥問題。

Hydroxyl radical is very reactive, underlying the chemistry of advanced oxidation processes (AOPs) for degrading organic compounds in water. Among various AOPs, Fenton’s reagent (H2O2/Fe2+) has been known to be an effective and simple oxidant. The major drawback of Fenton’s reaction is the production of substantial amount of Fe(OH)3 sludge that requires further separation and disposal. To solve this problem, the application of iron oxide as the heterogeneous catalyst in oxidizing organic contaminants deserves an in-depth investigation. In this study, a novel supported-FeOOH catalytic fluidized-bed reactor (FBR) was developed. The FBR may be better suited for the application in practice because of its high efficiency in mass transfer and easy solid/liquid separation. The crystallization of supported FeOOH, feasibility of oxidizing benzoic acid (BA), kinetics of decomposing H2O2 and oxidizing BA were investigated. Eventually, this reactor was used to treat a real dyeing/finishing wastewater, and the optimization of operating conditions was performed. The innovative catalyst (supported FeOOH) was prepared via the oxidation of Fe2+ by H2O2 in the acidic condition using a fluidized-bed crystallization reactor (crystallization-FBR). The major components coated on the surface were identified as amorphous FeOOH and *-FeOOH. The metastable region was found first to decide the rough inlet reagent concentration of crystallization-FBR to prevent the formation of primary nucleation. The characteristics of supported FeOOH including the particle morphology, Mossbauer spectrum, total and oxalate-soluble Fe contents, and the specific surface area were determined. Three kinds of supported FeOOH grains were synthesized: FeOOH I was prepared at pH 3.5, FeOOH II was formed by aging FeOOH I at pH 13, and FeOOH III was prepared at pH 5.0. FeOOH I was applied in following experiments including feasibility and kinetic studies of oxidizing BA, and treatment of the dyeing/finishing wastewater. FeOOH II and FeOOH III were used to compare the catalytic performance with FeOOH I. The crystalline property was found to obviously influence the performance of catalytic oxidation. In the aspect about the crystallization conditions of supported FeOOH, some parameters including the operational pH, superficial velocity, specific iron loading, and influent H2O2 concentration were investigated their effects on the crystallization efficiency. All these parameters were found to be the factors influencing the crystallization efficiency, but only the operational pH affected iron oxide type of grown crystal. The feasibility study about the H2O2 decomposition and BA oxidation using the supported-FeOOH catalyst was studied in the batch reactors. The kinetics of H2O2 decomposition in the absence of BA was conducted first. Oxidation of BA by H2O2 was performed so as to understand the effects of pH, H2O2 dosage, and the catalyst concentration. The treatment efficiency of BA at an initial pH of 3.2 was higher than at initial pHs of 6.0 and 10.0; this can be explained by the reductive dissolution of FeOOH in acidic environment. At the initial stage of reaction, however, the oxidation of BA was mostly contributed by heterogeneous catalysis because of trace amount of dissolved Fe2+. The higher initial oxidation rate of BA by heterogeneous catalysis at pHi 3.2 was due to its higher amount of undissociated BA adsorbed on the FeOOH surface. Therefore, the extent of heterogeneous catalysis was evaluated by comparing with an equivalent concentration of soluble ferrous ion in a parallel experiment. It is concluded that this catalytic oxidation mostly occurred on the catalyst surface, with some occurred in the aqueous solution due to the iron dissolution of the catalyst. In the absence of BA, the kinetics about the decomposition of H2O2 by the supported-FeOOH catalyst was performed in a continuous FBR. In this part, we attempted to study the effects of pH, H2O2 concentration, and catalyst concentration on the decomposition of H2O2. At the lower H2O2 concentration, the decomposing rate of H2O2 was found to be proportional to both H2O2 and catalyst concentrations. At the higher H2O2 concentration, however, it decreased with the increasing H2O2 concentration. This phenomenon was well fit with the substrate inhibition model. The significant difference of the observed first-order rate constant under different pH values was also modeled. In the part of oxidizing BA by the supported-FeOOH catalytic FBR, the effects of pH, H2O2 concentration, and BA concentration on the oxidation of BA were studied. The another aim of oxidizing BA using the continuous FBR was to realize the proportions of homogeneous catalysis and heterogeneous catalysis. It was found that the oxidation rate of BA was dependent on both H2O2 and BA concentrations. The change in rate constant of heterogeneous catalysis by pH was described in terms of the ionization fractions of surface hydroxyl group. From the mathematical deduction, we suggest that the reaction rate associated with □FeIIIOH2+ is much higher than with □FeIIIOH. Conclusively, the oxidation of BA at pH 4.4 ~ 7.0 is contributed by heterogeneous catalysis alone, but the homogeneous catalysis is of increasing importance below pH 4.4. The effect of Fe2+ on the catalytic oxidation in the FBR was also examined. The process with Fe2+ addition is named FBR-Fenton method, in which Fe2+ is the homogeneous catalyst and FeOOH is heterogeneous catalyst to oxidize BA. Both mineralization of organics and crystallization of Fe(III) were simultaneously well performed under the adequate condition. The efficiencies of BA mineralization and Fe(III) crystallization both closely related to the pH and inlet Fe2+ concentration. The reaction mechanism of FBR-Fenton method, including the reductive dissolution of FeOOH, synthesis of FeOOH, and oxidation of BA, was proposed based on experimental results. Finally, the FBR-Fenton method was applied to treat a dyeing/finishing wastewater, and the optimal operating conditions were found by the experimental design analysis. The methods of experimental design used here included the fractional factorial design, central composite design (CCD), and response surface methodology (RSM). Three most influential variables were decided by the fractional factorial design. Using these three variables, the regression models for the removal efficiencies of TOC (TOCr) and Fe (Fer) were determined by CCD and RSM. To select the optimal operating conditions, five conflicting criteria including pH, H2O2 dosage/COD, Fe2+/H2O2, TOCr and Fer were considered on the basis of Derringer’s desirability function. Furthermore, the FBR-Fenton method was proven superior to the conventional Fenton’s reaction due to high removal of TOC, effective Fe removal, and little sludge production by FBR-Fenton method.