Solar disinfection of water by photocatalytic processes

Chlorine, widely used as a disinfectant for drinking water production, is known to interact with organic matter to yield disinfection by-products (DBPs). Some of these compounds, as the trihalomethanes, are highly toxic for human health. Alternative disinfection methods are thus currently being developed, especially in industrialized and wealthy countries. Methods such as ozone and ultraviolet light are equally effective and less toxic. However, these techniques are expensive and often technically difficult to apply in less favorable conditions. In this context, solar disinfection could become promising to potabilize waters in developing countries with high availability of solar irradiation. The main objective of this research is to evaluate the solar photocatalytic treatment of water as: a) an alternative to bacterial inactivation by chlorination; b) a complement by assessing the influence of the process on the disinfection by-products precursors (DBPPs). In the proposed system a physical, chemical or biological technology can be used as pretreatment, preceding the solar photocatalytic. This thesis is organized in 6 chapters in which the link is the solar treatment applied to disinfect and decontaminate water. In the first chapter a review of the disinfection of water by solar and photocatalytic treatments is presented. Particular enfaces is made on the mode of action of TiO2 upon bacteria. Chapter 2 focuses on the study of physico-chemical and catalytical aspects of water disinfection by photocatalysis. Physico-chemical parameters related either to the operative system or the intrinsic properties or phototreated water influences the processes. The increase of some parameters such as light intensity, extend of continuous irradiation and catalyst concentration has a positive effect on disinfection. TiO2 immobilized on Nafion membranes inactivates E. coli K12 with efficiencies close to those observed for bacterial suspension containing the same concentration of suspended TiO2. TiO2 immobilized on glass flask, showed lower effectiveness than suspended catalyst. Chapter 3 focuses on the influence of biological aspects on the effectiveness of photocatalytic disinfection. To illustrate these aspects, synthetic and real waters were used as a model of study. In the first case, deionized water was contaminated by E. coli; in the second case, wastewater was taken at the outlet of a biological treatment plant located in Switzerland. The influence of initial bacterial concentration, physiological state of bacteria, and bacterial transfert in the culture are presented using pure culture of E. coli. The response of different Gram (+) and Gram (-) bacteria such as Enterococcus sp, and Fecal coliform, present in the effluents of urban treatment plants is discussed. The behavior of the bacterial suspension during the subsequent dark period was extensively discussed in order to estimate the potential of using the photocatalytic treatment process in a real water disinfection situation. The effective disinfection time (EDT) was defined as the time required for total inactivation of bacteria without re-growth in a subsequent dark period referenced at 24 or 48 h). Variability in the response to photocatalytic and photolytic (without TiO2) treatment for waters treated in Switzerland at different date is studied. E. coli was the most sensitive group to photocatalytic disinfection in all Swiss wastewaters. No E. coli recovery was observed after photocatalytic treatment. Chapter 4 focuses on the influence of chemical aspects on water disinfection by photocatalysis. To illustrate these aspects, deionized water was contaminated by E. coli and used as a model of study. This chapter is organized around the effect of five major topics: (i) pH, (ii) common anions such as, H2PO4-/HPO4-2, HCO3-, SO4-2, Cl-, NO3-; (iii) cations; Na+ and K+, (iv) electron acceptors O2 and H2O2, (v) mixture of anions and organics. The influence of chemical matrix constituted by a mixing of unknown organic and inorganic substances is also studied. Other experiments demonstrated the simultaneous effect of inorganic ions and organic matter in the photodisinfection. To assess this effect, we used two types of water to suspend bacteria during the phototreatment such as 1) tap water and 2) Milli-Q water containing nutrient broth + resorcinol (a DBPP) as well as phosphate + resorcinol. Chapter 5 illustrates the influence of selected organic substances on photocatalytic disinfection. Different interactions are studied. The evolution of dihydroxibenzenes (a group of DBPPs) degradation under light in the presence of both oxygen and H2O2 as electron acceptor is discussed. These structural isomers of di-hydroxibenzenes had negative influences on the E. coli inactivation by solar light with or without TiO2 addition. The extent of bacterial inactivation and photodegradation of these DBPPs were highly dependent on chemical structure of the substance, and therefore of chemical and physical propierties as acidity, absorption coefficient and water solubility. The influence of the DBPPs on E. coli was to protect the bacteria from: (a) solar, (b) photocatalytic inactivation and (c) adsorption on TiO2. These three kinds of protections are a consequence of: (a) the concurrent absorption of the light by the bacteria and the compounds, (b) photocatalytic degradation of the compounds and (c) DBPPs adsorption onto TiO2 surface. Direct correlation between photocatalytic bacterial inactivation and photocatalytic degradation of DBPPs was observed. Finally in chapter 6, field experiments under direct solar irradiation using CPC reactor are presented. Water from the Leman Lake contaminated by E. coli K 12 was exposed to sunlight in different seasons. The obtained results indicate that the presence of TiO2 accelerates the inactivating action of sunlight light. Total photocatalytic disinfection was obtained in two periods of year and no bacterial recovery was observed during 24 h after stopping sunlight exposure. In the absence of TiO2, total disinfection was not always reached; and bacterial recovery was observed, especially when inactivation was not complete. During E. coli inactivation in the absence of TiO2 the EDT was not reached. Residence time of water in the illuminated part of the system, light intensity and the period of the day selected for the treatment (morning, afternoon) strongly influence the EDT. It was also demonstrated that solar UV dose is not a pertinent parameter to standardize solar disinfection. Indeed, the relative UV and Visible wavelengths intensities, characteristics of each season and day period significantly affect the solar photoinactivation, and photoreactivation as well as the bacterial behavior in the subsequent dark period. The influence of the following topics on solar water disinfection is also studied in this chapter: i) UV and total solar spectra characteristics, (ii) volume of phototreated water, and (iii) the post irradiation events. A remarkable advantage of this kind of technology is that it can be transferred to countries with high sun irradiation levels as the tropical developing countries.