Background. In recent years, there has been a growing interest in the development of new processes for water disinfection since the traditional processes, such as chlorination, can lead to the production of toxic disinfection by-products. Sunlight has been used as a method of water disinfection and heliophotocatalysis by titanium dioxide (TiO2) has been recently considered as a new approach to improve the conventional solar water disinfection. This paper discusses the effect of the chemical composition of water on the E. coli photo inactivation. Method of Approach. Ten types of water having a different chemical composition were contaminated by E. coli K12 and exposed to a simulated solar irradiation in the absence of TiO2 (photolysis) and in presence of TiO2 (photocatalysis). Bacteria were monitored by plate count. The durability of disinfection was assessed in terms of the effective disinfection time (EDT) in a subsequent dark period of 24 h (EDT24). Natural water from the Leman Lake (LLW), milli-Q water (MQW), MQW containing a mixture of NO3-, PO4-3, SO4-2, Cl− and HCO3-, phosphate buffered saline water, water from the outlet of a biological wastewater treatment plant (WW); MQW containing a mixture of KCl-NaCl and commercial bottled drinking water (CBW) where used to suspend E. coli at laboratory scale. Field scale experiments using solar irradiation in a compound parabolic concentrator (CPC) with E. coli suspended in LLW were also carried out. Results. The sensitivity of bacteria to the phototreatment depends on the nature of the water. Moreover, experiments systematically performed under the solar simulator showed that the order of E. coli inactivation rate and the EDT24 are different for each system. In photolytic systems, E. coli solar inactivation rate is accelerated by the presence in water of NO3- and HCO3- when compared to that observed in MQW. EDT24 was reached at 3 h of irradiation for only 3 (WLL, WW1, and CBW) of the ten studied waters. In the presence of TiO2, the rate of the solar disinfection generally increased. However, a negative effect of chemical substances present in water on the E. coli photocatalytic inactivation was observed in waters when compared to MQW. This effect was especially important in the presence of phosphate, and carbonate. EDT24 was less than 2 h for the majority of the water types. In the presence of TiO2, a “residual disinfection effect” was observed for samples even when bacterial culturability below the detection limit was not reached after photocatalytic treatment. Solar irradiation in a CPC photoreactor indicates that the presence of TiO2 accelerates the detrimental action of sunlight. The EDT24 was reached before 3 h, in photocatalytic experiments but not in those in the absence of TiO2. The influence of TiO2 surface characteristics and charge, as well as the postirradiation events observed in heliophototreated water, in an optimal growth medium are also discussed. Conclusions. The presence of NO3-, HCO3-, PO4-3, SO4-2, Cl−, and HCO3- greatly affects the photolytic and photocatalytic disinfection processes. The natural ions and organic matter affect the solar disinfection of water in the presence and absence of TiO2 and influences the post irradiation events after sunlight illumination. Antagonistic effect in several conditions or synergistic effects in others can be expected when inorganic and/or organic substances are present in phototreated water sources. EDT24 is useful tool for standardization of solar water disinfection. EDT24 values depend on parameters such as the chemical composition of water, light intensity, initial bacterial concentration, and TiO2 concentration.