A B S T R A C TWith a view to evaluating the effects and interactions of three influential factors, turbidity, pH and dissolved oxygen (DO), on the efficiency of solar water disinfection process, response surface methodology (RSM) based on Box-Behnken design was employed. Total coliform (TC) removal and heterotrophic plate count (HPC) removal were used as response factors. The RSM models were developed based on the experimental results of four-hour solar exposure in polyethylene terephthalate bottles. The measured and predicted removals were close to each other. Though the HPC removal and TC removal showed similar trends, the HPC removal was much lower (40-66%) compared to TC removal (68-97%) for the conditions studied. The results showed that turbidity up to certain levels (10-20 NTU) did not adversely impact the bacterial removal. In the range studied (6.0-9.5), pH showed little influence on the removals. While bacterial inactivation showed significant reduction at low DO levels, increasing DO beyond a certain level (5-6 mg/L) did not show any beneficial effect on bacterial inactivation. The study thus demonstrated the usefulness of RSM in modelling and analysing solar disinfection process.

Solar water disinfection is a proven method for household purification of drinking water. Compound parabolic collectors (CPCs) have been used to enhance the inactivation rate of microorganisms. In the present study, a modified CPC with a higher concentration ratio of 1.1 was employed to further speed up the process. A large number of tests were conducted with water samples collected from different naturally contaminated water sources of varying characteristics under natural sunlight. Results indicated substantial reduction in exposure time for complete inactivation of indicator organisms, total coliforms and Escherichia coli, in modified CPCs compared with that in PET bottles at different solar intensity conditions. Even with weak solar intensity conditions (average values <400 W/m2), complete inactivation of test organisms could be achieved in all the test waters including those with high initial concentrations (total coliforms – 1.1 × 106 MPN/100 mL and E. coli - 1.25 × 105CFU/mL) within 6 h of exposure when modified CPC was employed. No regrowth of microorganisms was observed in the modified CPC up to 48 h of storage following exposure. The study thus suggests that the use of modified solar concentrators can result in significant reduction in the exposure time required for complete inactivation of naturally occurring microorganisms.

Water samples from four different sources of varying physico-chemical and microbial quality with their naturally occurring microorganisms were exposed to sunlight in polyethylene terephthalate bottles under similar conditions. Up to 3-log10 reduction of total coliforms (TC) was observed during a 6-h exposure period under weak/moderate radiation conditions (<600 W/m2). Complete inactivation of TC was not achieved in 6 h of exposure for waters with larger initial TC such as river water (1 × 103 most probable number [MPN]/100 mL) and treated municipal wastewater (2 × 105 MPN/100 mL) under these conditions. Heterotrophic bacteria showed lower inactivation rates than did TC. The inactivation rate for spiked Escherichia coli was faster than for naturally occurring coliforms. Further tests with compound parabolic collectors showed that complete inactivation of naturally occurring TC could be achieved within 6 h of exposure for all the natural waters tested. The results of the study thus indicate the need to use naturally occurring organisms in testing the effectiveness of solar disinfection, and the importance of source quality on the inactivation rates of microorganisms.