学位授与年月日: 2007-07-23 ; 学位の種類: 新制・課程博士 ; 学位記番号: 工博第2837号 / Perfluorochemicals (PFCs) were produced by industries and consumed “safely” as surfactants, repellents, additives, fire-fighting foams, polymer emulsifiers and insecticides for almost fifty years. However they are now considered as persistent, bioaccumulated and toxic (PBT) chemicals, and ubiquitously distributed in waster, air, human body and biota. Although some efforts were contributed to reduce PFCs in environment, such as development of alternatives and recycling processes, huge amount of persisted PFCs have already been discharged in environment and accumulated in biota including humans. In some industrialized areas, such as Yodo river basin in Japan, water environment and human blood were polluted by some PFCs, and thus reduction and control of PFCs were urgently required for the purpose of environmental safety and human health in these areas. Unfortunately, some studies implied that current water and wastewater treatment processes seemed ineffective to remove PFCs in trace levels. Therefore, this study will try to develop some proper technologies to treat trace level of PFCs in wastewater. In order to achieve this main objective, several works have been accomplished as follows. 　Current available literature has been reviewed to obtain a solid background for this study. Basic information of PFCs was summarized in physiochemical properties, PBT properties, productions and applications, regulations and etc.. Analytical methods for PFCs, especially of LC-ESI-MS/MS, were reviewed including pretreatment processes in diverse matrices, which derived objectives of chapter III. Distributions and behavior of PFCs were briefly discussed in water environments, biota sphere and human bloods. Available control strategies were shown in detail about alternatives, industrial recycling processes, and newly developed treatment processes. Current wastewater treatment processes showed inefficient removal for some PFCs, deriving objectives of chapter IV on the PFC behavior in treatment process. Newly developed treatment technologies seemed able to decompose PFCs completely but unsuitable for application in WWTP. Therefore, granular activated carbon (GAC) adsorption and ultra violet (UV) photolysis were developed in chapter V and VI as removal and degradation processes respectively. 　Fifteen kinds of PFCs were included in this study, consisting of twelve kinds of perfluorocarboxylic acids (PFCAs) with 4~18 carbons and three kinds of perfluoroalkyl sulfonates (PFASs) with 4~8 carbons. An integral procedure was developed in chapter III to pretreat wastewater samples. LC-ESI-MS/MS was applied to quantify all PFCs in trace level. Pretreatment methods were optimized between C18 and WAX-SPE processes for aqueous samples, and between IPE, AD-WAX and ASE-WAX processes for particulate samples. Standard spiking experiments were regularly conducted for each wastewater sample to calculate recovery rate and control analytical quality. As the result, WAX-SPE showed better performance on samples with very high organics concentrations, and C18-SPE performed better for long-chained PFCs. ASE-WAX was proposed as the optimum method to pretreat particulate samples because of the simple and time saving operations. 9H-PFNA was used as internal standard to estimate matrix effect in wastewater. 　Behavior of PFCs in a municipal WWTP has been studied in chapter IV by periodical surveys for six times in half a year. All PFCs used in this study were detected in WWTP influent and effluent. According to their carbon chain lengths, all PFCs can be classified into “Medium”, “Long” and “Short” patterns to simplify behavior analysis. PFCs in same pattern showed similar properties and behavior in wastewater treatment facilities. Very high concentrations of PFCs existed in WWTP influent, indicating some point sources of industrial discharge in this area. “Medium” PFCs, such as PFOA(8), PFNA(9) and PFOS(8), were primary contaminants in the WWTP and poorly removed by overall process. Performances of individual facilities were estimated for removal of each PFC. Primary clarification and secondary clarification were helpful to remove all PFCs in both aqueous phase and particulate phase. “Medium” PFCs in aqueous phase were increased after activated sludge process, but other PFCs can be effectively removed. Ozone seemed ineffective to decompose PFCs because of the strong stability of PFC molecules. Sand filtration and biological activated carbon (BAC) filtration in this WWTP can not remove PFCs effectively too, which required further studies. Performances of combined processes were estimated by integrating individual facilities along the wastewater flow. Activated sludge process coupled with clarifiers showed satisfied removal of most PFCs in the investigated WWTP except “Medium” PFCs. 　Adsorption characteristics of PFCs onto GAC have been studied by batch experiments in chapter V. Freundlich equation and homogenous surface diffusion model (HSDM) were applied to interpret experimental data. Isothermal and kinetics experiments implied that PFC adsorption on GAC was directly related with their carbon chain lengths. By ascendant carbon chain length, adsorption capacity for specific PFC was increased, and diffusion coefficient (Ds) was decreased. Ds of GAC adsorption was also decreased gradually in smaller GAC diameters. Coexisted natural organic matters (NOMs) reduced adsorption capacities by mechanism of competition and carbon fouling. Carbon fouling was found reducing adsorption capacity much more intensively than competition by organics. Acidic bulk solution was slightly helpful for adsorption of PFCs. However adsorption velocity or kinetics was not affected by NOM and pH significantly. GAC from Wako Company showed the best performance among four kinds of GACs, and Filtra 400 from Calgon Company was considered more suitable to removal all PFCs among the commercial GACs. Preliminary RSSCT and SBA results implied that background organics broke through fixed GAC bed much earlier than trace level of PFCs. Medium-chained PFCs can be effectively removed by fixed bed filtration without concerning biological processes. 　Direct photolysis process has been developed in chapter VI to decompose PFCAs in river water. Irradiation at UV254 nm and UV254+185 nm can both degrade PFCAs. Stepwise decomposition mechanism of PFCAs was confirmed by mass spectra analysis, and consecutive kinetics was proposed to simulate experimental data. PFASs can also be degraded by UV254+185 photolysis, although the products have not been identified yet. Coexisted NOMs reduced performance of UV photolysis for PFCAs by competition for UV photons. Sample volume or irradiation intensity showed significant influence on degradation of PFCAs. Local river water polluted by PFOA can be cleaned up by UV254+185 photolysis effectively. Ozone-related processes were also studied but ineffective to degrade PFC molecules. However, PFCs could be removed in aeration flow by another mechanism. / 京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第13340号 / 工博第2837号 / 新制||工||1417(附属図書館) / UT51-2007-M963 / 京都大学大学院工学研究科都市環境工学専攻 / (主査)教授 田中 宏明, 教授 藤井 滋穂, 教授 伊藤 禎彦 / 学位規則第4条第1項該当 / Doctor of Engineering / Kyoto University / DFAM

Die intratracheale Applikation von Perfluorcarbonen (PFC) in ungeschädigte Lungen, z.B. als Röntgenkontrastmittel und zur PFC-induzierten cerebralen Kühlung, wird derzeit diskutiert. Ob es aufgrund der Dichte und Sauerstofflöslichkeit der PFC zu einer Beeinflussung der cerebralen Oxygenierung und Hämodynamik durch die pulmonale PFC Applikation kommt, ist bisher unbekannt. Änderungen der Konzentration von cerebralem oxygeniertem und totalem Hb koennen mittels Near-infrared Spectroscopy gemessen werden. I.) Effekt durch die PFC-Applikation in gesunde Lungen neugeborener Ferkel: Es wurden zwei verschiedene Applikationsgeschwindigkeiten (30 ml PFC pro kg Körpergewicht als Bolusgabe versus 1,5 ml/min pro kg) und zwei Füllvolumen (30 versus 10 ml/kg) verglichen. Die Bolusgabe bewirkt einen sofortigen Abfall des PaO2 und der cerebralen Oxygenierung, bei langsamer Gabe des gleichen Volumens ist dies weniger ausgeprägt. Mit 10 ml/kg PFC fand sich nahezu keine Beeinflussung der Parameter. II.) Effekte der Änderung der FiO2 in PFC-gefüllten Lungen: Hierfür wurden Änderungen der FiO2 unter konventioneller Druckbeatmung vor PFC-Applikation und unter PFC-Füllung (30 oder 10 ml/kg PFC) verglichen. Beide PFC-Volumen (mit FiO2 1,0) bedingten PaO2-Werte wie unter CMV mit FiO2 von 0,5. Mit 30 ml/kg PFC kam es unter FiO2 von 0,5 zur Abnahme des cerebralen oxygenierten Hb. Zusammenfassend ergibt sich, dass die Applikation von 10 ml PFC/kg bevorzugt werden sollte. Bei kompletter Füllung der Lungen werden durch langsame Applikation cerebrale Nebenwirkungen minimiert. In PFC-gefüllten gesunden Lungen ist zum Erhalt der systemischen und cerebralen Oxygenierung die FiO2 zu erhöhen. / Intratracheal administration of perfluorochemicals (PFC) has been suggested for reasons other than respiratory insufficiency, such as pulmonary imaging and PFC-associated brain cooling. Due to their high density and oxygen solubility, PFC application has been described to affect systemic hemodynamics and oxygenation during liquid ventilation. Whether the PFC application into healthy lungs or changes in inspired oxygen fraction (FiO2) in PFC-filled healthy lungs affects cerebral hemodynamics is not known. Changes in the concentration of cerebral oxygenated and total Hb can be measured by near-infrared spectroscopy (NIRS). I.) Initial effects of PFC application into healthy lungs of newborn piglets: Two different filling modes (rapid versus slow) and two different filling volumes (slow filling of 30 versus 10 ml PFC/kg body weight) were compared. Rapid filling caused an immediate drop of systemic and cerebrale oxygenation, which was less prominent by filling 30 ml/kg PFC slowly. Almost no changes for all parameters were found with 10 ml kg/PFC. II.) Effects of variations in FiO2 in the PFC filled lungs of healthy piglets: Changes in FiO2 during conventional mechanical ventilation (CMV) prior to PFC-application and in the PFC-filled lungs (30 ml/kg or 10 ml/kg PFC) were compared. Both PFC-volumes (at a FiO2 of 1.0) caused PaO2-values that were similar to CMV with FiO2 of 0.5. In the 30 ml/kg PFC group a reduction of cerebral oxygenated Hb was found at FiO2 of 0.5. According to the data application of 10 ml/kg PFC should be preferred. If complete filling of the lung is needed, the slow administration of PFC minimizes cerebral side effects. In PFC- filled healthy lungs an increase in FiO2 is necessary to maintain systemic and cerebral oxygenation.