Plumbing Africa August 2017 | Page 27

Health and sanitation << Continued from page 23 Although the treatment wing had a smaller number of positive cultures (3 of 12) than the control wing (8 of 12), the researchers could not reach a conclusion on the role of ozone in the inactivation of L. pneumophila. The study indicated that when ozonation was stopped, L. pneumophila regrew and reached levels close to the pre-test conditions at the end of the stagnation phase. Moreover, the authors pointed out one important factor for continual dosing of ozone, namely that residual ozone at the faucet or shower head led to the release of gaseous ozone into the air (an issue discussed in Section 2.3.6.4). Several laboratory studies have reported rapid and effective inactivation of legionella with ozone (Jacangelo et al., 2002; Domingue et al., 1988; Muraca et al., 1987). • Loret et al. (2005) used a simulated distribution system consisting of pipe loops to compare the effectiveness of several disinfectants to control legionella in biofilms in premise plumbing. The study concluded that ozone was effective to control planktonic and biofilm-associated populations within the pipe loops, but was ineffective within dead end sections. • Jacangelo et al. (2002) conducted laboratory studies to evaluate multiple disinfectants, including ozone, for inactivation of waterborne emerging pathogens including legionella. The ozone dosage rate was 1.0mg/L. The model-predicted CT values for 2-log (99-percent) inactivation of legionella at pH 7 were 2.5, 0.16 and 1.1 min-mg/L at 5, 15 and 25 degrees C (41, 59 and 77 degrees F), respectively. • Domingue et al. (1988) conducted laboratory experiments to compare the bactericidal effects of ozone, hydrogen peroxide and free chlorine on “free” L. pneumophila cells. Ozone was the most potent of the three disinfectants, with a greater than 2-log (99-percent) inactivation in L. pneumophila occurring during a 5-minute exposure to 0.10–0.3mg/L ozone. The researchers reported little to no effect of pH and temperature on ozone inactivation of L. pneumophila. The pH ranged from 7.2 to 8.9. Experiments were conducted at 25, 35 and 45 degrees C (77, 95 and 113 degrees F). • Muraca et al. (1987) compared the efficacy of chlorine, heat, ozone, and UV light for inactivating L. pneumophila in a bench-scale model plumbing system. L. pneumophila was added to the system and allowed to circulate. Continuous ozonation for five hours at a concentration of 1 to 2mg/L achieved a 5-log inactivation of L. pneumophila at 25 and 43 degrees C (77 and 109.4 degrees F, respectively). Neither turbidity nor the higher temperature (43 degrees C, or 109.4 degrees F) was reported to affect the efficacy of ozone. www.plumbingafrica.co.za 25 Potential water quality issues Ozone decomposes in water relatively rapidly. The half-life of ozone in finished drinking water depends on temperature, pH and alkalinity, and can vary from minutes to hours. This time-scale is short relative to chlorine-based disinfectants, and as such, ozone is not generally considered to produce a disinfectant residual. Therefore, water treated with ozone may, in some cases, be susceptible to contamination at downstream points. For this reason, more than one type of treatment or control measure may be necessary to protect the treated water. Disinfection by-products formed from ozone disinfection include bromoform, monobromoacetic acid, dibromoacetic acid, dibromoacetone, cyanogen bromide, chlorate, iodate, bromate,hydrogen peroxide, hypobromous acid, epoxides, ozonates, aldehydes, ketoacids, ketones and carboxylic acids (WHO, 2011b). Ozonation of water containing inorganic bromide can produce bromate, a regulated DBP with an MCL of 10µg/L. The disinfection process of a PWS will likely have transformed any bromide in water to organically bound bromine or inorganic bromamines. In either case, these forms of bromine are less likely to contribute to bromate formation via an ozonation process in a premise plumbing system. As such, bromate formation may not be as relevant as in the water treatment plant. Other ozonation by-products such as aldehydes and organic acids are more readily biodegradable and may contribute to assimilable organic carbon (AOC) and hence biological growth in the distribution system. In addition, these ozonation by-products are more likely to form some types of DBPs upon chlorination or chloramination (Carlson and Amy, 2001; Shah and Mitch, 2012). However, these general concepts regarding ozonation pertain to treatment of water at the plant. Ozonation of water that has already undergone treatment, including exposure to a chlorine or chloramine residual in the distribution system en route to the building (for example, hospital) has not been studied to a great extent. Therefore, impacts of ozonation on AOC or DBP formation in a premise plumbing system are still unclear. Loret et al. (2005) observed corrosion marks on mild and galvanised steel coupons installed in pipe loops for ozone treatment that were similar to corrosion effects caused by other disinfectants (chlorine, chloramine, chlorine dioxide and CSI), except that the coupons exposed to CSI also had copper deposits. PA As a primary disinfectant, ozone is more effective than chlorine. August 2017 Volume 23 I Number 6