Membrane bioreactors (MBRs) offer many advantages for the treatment of municipal and industrial wastewater compared to conventional wastewater treatment processes. It is expected that the demand for MBRs will further increase, with more than double-digit growth annually over the next decade (Santos et al., 2011). However, one of the main drawbacks of the widespread use of MBRs is membrane fouling, which can result in severe flux decline or a rapid trans-membrane pressure (TMP) increase that subsequently increases the net operation cost due to frequent membrane replacements and/or cleaning (Le-Clech et al., 2006). The problems associated with membrane fouling must be addressed before large-scale practical application of MBRs can occur. Thus, methods to prevent or reduce membrane fouling have been the subject of intense research.

In this study, we proposed an electrochemical oxidation (EO) process used to generate highly reactive oxidants as a new cleaning method for the mitigation of membrane fouling in a MBR. The motivation of this study is an initial attempt to evaluate the potential use of integrating the EO process with the membrane filtration process in one reactor, with a focus on membrane fouling reduction. The EO process is a promising and attractive technique for the effective oxidation of wastewater containing organic compounds. During electrolysis, organic compounds are destroyed by either direct or indirect oxidation processes via the main oxidizing agents, such as reactive oxygen species (i.e., ・OH, H2O2 and O3) and free chlorine species (mainly ClO- or HClO) produced from chloride ions (Comninellis, 1994; Chen, 2004). In particular, the indirect oxidation of organic compounds mediated by free chlorine species that are electrochemically generated on dimensional stable (DSA) anodes (e.g., RuO2, IrO2 and Pt) has proven effective for the degradation of organic compounds (Chen, 2004; Costa et al., 2008; Khelifa et al., 2009). The extensive use of free chlorine species in the EO process is due to the ubiquitous presence of chloride in wastewater and to its effectiveness and long lifetime.

The aforementioned characteristics of the EO process enabled us to hypothesize that organic membrane foulants accumulating on the membrane surface or clogging the inside of the membrane pores during filtration could be reduced through the electro-generated oxidants in the region close to the anode surface and/or bulk solution. The principle objective of this study was to evaluate the potential use of integrating the EO process with a micro-filtration process, with a focus on the role of the EO process in mitigating membrane fouling in the MBR. To achieve above objective, the newly designed MBRs, called the EO-MBR and modified EO-MBR, were developed by applying a direct current between perforated Ti/IrO2 anodes and Ti/Pt cathodes around a membrane module and its performance was investigated. The main empirical findings derived from this study can be summarized as follows.

First, batch cell electrolysis experiments revealed, as evidenced by linear sweep voltammetry measurements and the quantities of reactive oxygen species and free chlorine species generated, that organic membrane foulants on the Ti/IrO2 anode can be degraded by direct and indirect oxidation by free chlorine species, while other electro-generated oxidants (e.g., ・OH, H2O2 and O3) play a less important role. Furthermore, the Ti/IrO2 anode is energetically more efficient than other anodes (e.g., Ti/RuO2, Ti/IrO2+RuO2, Ti/Pt and boron-doped diamond (BDD)), which are widely applied in the EO process, in presence of chloride ion because it leads to COD removal of organic membrane foulants with lowest energy consumption due to the higher electrocatalytic activity for free chlorine species evolution, although good results in terms of COD removal are obtained with the BDD anode.

Second, batch cell electrolysis experiments with the Ti/IrO2 anode carried out using chloride as supporting electrolyte resulted in more efficient degradations by reaction with the electro-generated free chlorine species, while phosphate, sulphate and carbonate did not influence on the performance of the EO process. Furthermore, the current density plays a key role on degradation kinetics and lower current density was considered as the suitable operating condition in terms of the lower energy consumption. Additionally, the Ti/IrO2 anode proved to be the efficient anode material as an advanced treatment of various organic membrane foulants (i.e., humic acid, bovine serum albumin, starch and alginic acid) and ammonia, which might be additional advantage of the EO process integrated with activated sludge process.

Third, scale deposition on the cathode surface caused gradually an increase of voltage and average bubble size, whereas scale deposition did not influence on the performance of the EO process in terms of the quantity of free chlorine species production and bubble generation efficiency. The SEM-EDS analysis characterized scale deposition as a cluster of CaCO3, Mg(OH)2 and phosphorus compounds. However, current switching after scale growth proved to be an efficient control strategy of scale deposition.

Forth, batch cell electrolysis experiments revealed that organic compounds in the mixed liquor, such as protein, humic acid and fulvic acid, were effectively reduced by the EO process. Moreover, the zeta-potential of the mixed liquor was decreased, increasing the size distribution of the sludge floc. These are some of the factors that may be responsible for the EO process via direct or indirect oxidation mediated by free chlorine species. In the microbial activity experiments, a current density of 0.4 mA/cm2 was determined to be adequate for the EO process; this current does not inhibit the activities of the heterotrophic and autotrophic biomasses. Further increases in the current density, however, resulted in a significant decrease in biomass activity.

Fifth, the continuous experiments confirmed that compared to the increase in the TMP observed in the control MBR, incorporating the EO process into the EO-MBR system effectively suppressed the increase in the TMP with a current density of 0.4 mA/cm2. The cleaning cycle of the EO-MBR was almost twice as long as that of the control MBR. According to the hydraulic and chemical analyses of the cake sludge layer, the reduction of fouling observed in the EO-MBR is largely due to the electro-generated free chlorine species on the Ti/IrO2 anode, which reduced the physically irremovable membrane fouling and was effective at degrading organic membrane foulants, such as SMP, proteins, and humic acid, in the cake sludge layer that accumulate on the membrane. In terms of effluent quality, the DOC concentration of effluent from the EO-MBR did not show a significant difference compared with the control MBR, implying that most of the microorganisms do not lose their activity with the EO process. However, the variation in the microbial community observed in the EO-MBR was significantly influenced by the EO process compared to control MBR. The energy consumption for the EO process was 0.029 - 0.034 kWh/m3 (permeate), which was much lower than the overall operation costs of the conventional MBRs (0.7 - 1.2 kWh/m3 kWh/m3 (permeate)), supporting the economic feasibility of using the EO process as a new control method for fouling reduction in MBRs.

Sixth, the modified EO-MBR system, which was comprised of placing the membrane module combined with perforated anodes inside non-woven fabric filter (as a pre-filter), was found to effectively reduce the increased TMP at intermittent current density of 5 mA/cm2 ("30 s on, 2 min 30 s off" mode) and simultaneously the wastewater treatment efficiency of the MBR in terms of ammonia removal was improved. Moreover, the energy consumption of the modified EO-MBR was approximately 0.028 kWh/m3 (permeate), which was identical to that of the EO-MBR in which the applied current density was 0.4 mA/cm2.

Keywords: membrane bioreactor; membrane fouling; electrochemical oxidation process; free chlorine species.

本論文は「Mitigation of membrane fouling by applying the electrochemical oxidation process in a membrane bioreactor(メンブレンバイオリアクターにおける電気化学的酸化法による膜ファウリング制御に関する研究)」と題し、メンブレンバイオリアクター(MBR)における新たな膜表面近傍に焦点を当てた電気化学的ファウリング制御法を提案し、その有用性とファウリング抑制の機構を明らかにした独創的研究である。

第1章は「序論」である。研究の背景、目的と位置づけ、及び論文構成等を述べている。

第2章は「文献レビュー」である。MBRの歴史、ファウリング、電気化学的プロセスによる下水処理等に関する既往の知見をまとめている。

第3章は「実験方法」である。以下の各章に詳細な実験方法の記述があるが、ここでは本研究全体を通して共通する実験材料や分析方法についてまとめている。

第4章は「有機性ファウリング物質の電気化学的酸化に及ぼす陰極材料の影響」である。本章で用いた実験装置や材料、方法を記述したのち、各種電極の有機物分解メカニズムを確認し、使用した陰極材料の中で、Ti/IrO2がエネルギー消費や分解効果を総合的に勘案して、実用的に優れていると評価している。今後の浸漬型MBRに適用する電気化学的ファウリング制御に関する基礎的情報を与えるものである。

第5章は「電気化学的酸化法による分解効率に及ぼす諸因子」である。4章で選定したTi/IrO2を用い、電解質や電流密度の違いが有機性ファウラント分解効率に及ぼす影響を調べたもので、本章で用いた実験装置や材料、方法を記述したのち、各種排水に含まれる塩化物イオンを利用した有効塩素生成によるファウリング制御が有効で、また電極表面へのスケール生成も、電流を間欠的に短時間逆転させる方法で十分制御可能であることを示した。

第6章は「活性汚泥混合液中の有機性膜ファウラント及び微生物活性に及ぼす電気化学的酸化法の影響」である。電気化学的酸化が汚泥の活性に及ぼす悪影響を定量化し、電流密度の最適値があることを明らかにしたもので、本章で用いた実験装置や材料、方法を記述したのち、本実験条件下では0.4 mA/cm2以下に制御することにより微生物活性の低下を防ぐことが出来るとした。生物処理プロセスとしてのMBRへの電気化学的酸化法の適用条件を明らかにした重要な成果である。

第7章は「電気化学的酸化法の膜表面近傍への適用による膜ファウリング抑制」である。前章までに得られた基礎的知見を基に、浸漬型MBR用の膜ファウリング抑制法を新たに開発したものである。新たなMBRをEO-MBR(Electrochemical Oxidation MBR)と提案し、本研究の中核をなす研究成果がまとめられている。本章で用いた実験装置や材料、方法を記述したのち、前章で求めた電流密度条件0.4 mA/cm2でMBR連続処理実験を行った結果、コントロールMBRに比して、ファウリング進行速度を約半分に落とすことに成功した。また、実験終了後の膜ファウリング物質の分析により、EO-MBRでは膜ファウラント物質のうち物理的に洗浄可能なものがコントロールMBRに比して多く、また溶解性微生物産生物質としてのタンパクや多糖類成分の分解が進んでいることが実証された。またそれらの物質の分解による膜処理水の水質の悪化は認められず、本実験条件下で、エネルギー消費が0.03kW/m3程度と現状のMBRによる実処理でのエネルギー消費に比べて十分低く、実際への適用が大いに期待できる結果を得ている。電気化学的酸化による膜ファウリング制御法の実用化へ向けての大きな一歩与える重要な成果である。

第8章は「前処理用に不織布を組み合わせたEO-MBRによる膜ファウリング制御性能」である。さらにEO-MBRを改良し、不織布と組み合わせ中空糸膜との間の密閉空間を利用して陰極を配置し、より効率的で微生物活性への低下を抑える画期的なファウリング抑制法の試みたものである。本章で用いた実験装置や材料、方法を記述したのち、通常の下水に含まれる塩化物濃度でも、間欠的に電流密度を上げて運転させた結果、エネルギー消費を前章の結果と同様に0.03kW/m3程度に抑えつつ、コントロールMBRに比してファウリング抑制効果があること、電極の配置の工夫が重要であることを示した。

第9章は「結論と今後の展望」である。

以上要するに、本論文は、ユニークな発想により、膜モジュールに付随させる電気化学的酸化膜ファウリング抑制法の開発に成功し、それを用いたこれまでにない画期的な、コンパクト膜モジュール表面近傍に付随する連続洗浄デバイスを提案し、各種排水処理への実用化への道を切り開いたものであり、同時に電気化学的酸化法のMBRへの適用における基礎的かつ学術的情報を与える独創性の高い研究であると評価できる。また、本研究で得られた知見は、都市環境工学の学術の発展に大きく貢献するものである。

よって本論文は、博士(工学)の学位請求論文として合格と認められる。