While earlier studies have mostly concentrated on the application of TFFBR systems for chemical degradation, TiO2-based photocatalysis has proved its ability to enhance the rate of inactivation of microbes in contaminated drinking waters and waste waters, enabling such waters to be disinfected [20, 21]. The present study has clearly shown that A. hydrophila ATCC 35654 can be effectively inactivated in spring water using the TFFBR under sunlight conditions of > 600 W m-2, demonstrating its potential for applications in aquaculture, especially in tropical and sub-tropical developing countries where sunlight is abundant and the resources for alternative forms of disinfection are scarce.
The efficiency of the TFFBR was also investigated in this study by flowing (at 4.8 L h-1) contaminated spring water sample under high sunlight intensities and by using same sized glass with and without TiO2 under the same reactor conditions. The findings of this study confirm the results of two previous studies [7, 21]. The presence of TiO2 showed a clear enhancement in solar photocatalysis . The current study clearly shows that solar energy alone is unsufficient to inactivate A. hydrophila and that a photocatalyst such as, TiO2 is required for effective reduction in counts.
Microbial disinfection by solar photocatalysis is a complex and challenging process . The extent of inactivation observed in A. hydrophila ATCC 35654 under high sunlight intensity was also found to be similar to that reported for other microbes in early studies [8, 16]. Thus one investigation showed that when the UV irradiance was 20-43 W m-2, the inactivation of the fungus Fusarium sp. was faster than than at lower irradiances (cloudy weather condition), using a CPC reactor . Similar effects of solar irradiation on inactivation were observed in the present study, under different sunlight condition. For example, at lower sunlight conditions (total sunlight intensity = 300-600 W m-2 or UV irradiance = 20-40 W m-2) inactivation was considerably less than was observed at the highest sunlight conditions (> 1100 W m-2 and > 65 W m-2) at 4.8 L h-1. Solar photocatalytic activity was also demonstrated for various pathogens in drinking water in a batch culture reactor using simulated sunlight , in contrast to the TFFBR system tested under natural sunlight used in the present study. Similarly, recent studies have succeeded in photocatalysis but they required a long UV exposure times to achieve a log inactivation of 6-fold for E.coli K12 using a CPC pilot plant solar reactor [7, 21]. Such inactivation is far greater than that observed in the present study, where the log inactivation was around 1.38 with an average initial count of 1.36 × 105 CFU mL-1 and average final count of 5.10 × 103 CFU mL-1, at the highest sunlight intensities--this is most likely due to the rapid transfer of contaminated liquid across the TFFBR plate, which is around 2.5 min at 4.8 L h-1flow rate, in the present study. As most previous studies have used an artificial UV light source for exposure, it is difficult to make direct comparisons to the present study, where natural sunlight has been used. Additionally, different type of reactors will have different dynamics of inactivation and flow, as well as dissimilar kinetics of change with light intensity.
Counts of A. hydrophila ATCC 35654 exposed to the TFFBR system at low sunlight (< 600 W m-2) under ROS-neutralised conditions were substantially higher than those obtained from standard aerobic plate counts, which validates the finding from previous studies of E. coli and other bacteria [22–24]. This indicates that the antioxidant system of many cells of A. hydrophila ATCC 35654 was damaged by solar photocatalysis at low sunlight intensities, resulting in their sensitivity towards their own respiratory by-products. Such cells were only able to form colonies when sodium pyruvate (a scavenger of hydrogen peroxide) is added, coupled with growth under anaerobic conditions, which will enable the bacteria to use fermentative pathways, rather than aerobic respiration, for energy generation. The findings of this present study unequivocally demonstrate that at all three different flow rates tested, at low sunlight intensities (< 600 W m-2) there was a substantial difference between the log inactivation results based on ROS-neutralised and conventional aerobic counts (Figures 3 and 4). At 4.8 L h-1, there was close to 1 log difference between the ROS-neutralised and aerobic log inactivation results, suggesting that the aerobic data provide an apparent inactivation that overestimates the true value. For other two flow rates (8.4 and 16.8 L h-1) the difference between the two sets of data were around 0.9 and 0.5 (with similar initial inoculam of 1.33 × 105 CFU mL -1 and final count of 9.40 × 103 and 1.75 × 104 CFU mL-1) respectively, indicating a reduction in the amount of sub-lethal injury at higher flow rates that is also coupled with a lower overall inactivation (Table 1). While previous studies of solar disinfection have demonstrated sub-lethal injury and ROS-sensitivity in batch culture with uncalatysed reactors, this is the first study to do so for the TFFBR continuous flow photocatalytic system. On the other hand, at higher sunlight intensities (> 600 W m-2), the differences between the results based on aerobic counts and ROS-neutralised counts were negligible for all flow rate conditions, demonstrating the strength of high sunlight to provide powerful inactivation, with no sign of sub-lethal injury.
Sometimes, sunlight itself is not sufficient for water disinfection, due to the effectiveness of photoreactivation mechanisms in microorganisms . A recent study has demonstrated the effectiveness of immobilised TiO2 reactors in inactivating bacteria to such an extent that their photoreactivation mechanisms are not able to repair the damage , indicating that fixed-bed TiO2 reactors increase the extent of damage to bacteria from the very beginning of the process, whereas TiO2 slurry systems required longer irradiation times to cause an equivalent amount of cellular damage. In a slurry system, TiO2-related damage occurs at the cell membrane of bacteria; however, damage is distributed across the whole membrane, so membrane permeability effects are not always strong enough to cause irreversible inactivation in the early stages of the process. On the other hand, in a fixed-bed reactor, while the free radicals generated may be lower in number, the damage can be concentrated on the cell membrane area, causing inactivation . The result of the current study can be interpreted in similar approach, but with respect to sunlight intensity. Here it was observed that while low sunlight resulted in substantial sub-lethal injury, with results based on ROS-neutralised counts being far lower than for aerobic data, at higher light intensities, ROS neutralised data were similar to those based on aerobic counts. As the data at high sunlight intensities showed little evidence of sub-lethal injury, this demonstrates that the TFFBR system will be more efficient in full sunlight, where maximum inactivation is achieved.
The dynamics of flow rate in pilot-scale photocatalytic reactors have not been well studied to date. In considering treating large volumes of water, as in aquaculture systems, it is obvious that flow rate will be a crucial parameter. A pilot-scale CPC reactor using TiO2 in suspension with different flow rates has been used to study the inactivation of Fusurium sp. spores ; achieving the highest inactivation rate of Fusurium spores at a flow rate of 30.0 L min-1 with added TiO2 at 100 mg L-1 concentration. However, such systems require separation of the suspended TiO2 after treatment, which adds to the complexity, in contrast to immobilised systems such as the TFFBR. Another recent solar disinfection study also showed the importance of evaluating different parameters including: flow rate; water volume within the reactor; temperature; and solar energy . They used a CPC reactor with no added TiO2 and suggested that increasing flow rate has a substantial negative effect on the inactivation of bacteria, which is in agreement with the flow rate investigations of the present study. Here, the lowest flow rate of 4.8 L h-1 was found to be the most effective for inactivation of A. hydrophila ATCC 35654 as the residence time of 2.5 minin the 4.8 L h-1 experiment is almost twice as high as the 8.4 L h-2 experiment.(86 s) Similarly, when the total sunlight intensity is at average of 1000 W m-2, the cumulative energy, 150 KJ m-2 at 4.8 L h-1 is higher than that of 86 KJ m-2 at 8.4 L h-1 which will play a major role A. hydrophila inactivation. In this study, the water temperature in the reservoir was maintained at (22-23)°C throughout the experiments. Due to the open structure of the TFFBR, the temperature of the water on the reactor plate was not measured, though it is logical to expect that it would be positively related to sunlight intensity.