Chlorine Residual Boosting in Distribution Water: Problems with Chlorine Application and Disinfection Byproducts - Part 1

March 28, 2003

About the author: Shin-ichi Tokuno retired as a microbiologist in December 2002 from the Water Utilities Laboratory at the City of Corpus Christi, Texas.

Previous research on the boosting of chlorine residual1 included how to increase low levels of chlorine disinfectants (free and combined chlorine) in the distribution system. Simple bench tests using a pocket photometer showed that there are no problems in boosting the low level of chlorine residual when boosting the same disinfectant to the water (e.g., free chlorine to free chlorine, or chloramine to chloramine). In the boosted chlorine residual, there is no significant instability in decay or dissipation during the time needed (72 hours) after boosting for the small utility distributors.

Some utilities (Ingleside, Port Aransas, Texas) inject ammonia and chlorine to the Corpus Christi (C.C.) water to boost the residual. However, most utilities (30-40 miles from the plant) boost the low chlorine residual only by chlorine gas. This means that the free chlorine interacts with the preexisting chloramine of C.C. water to form the breakpoint (BP) chlorination. As a result, the mode of chlorine changes from the combined chlorine to the free chlorine or, if the total ammonia (free and combined) is completely depleted prior to boosting, the free chlorine develops without BP formation after chlorination. Because this boosting forms free chlorine, it produces more disinfectant action against bacteria and viruses than chloramines. However, a few stability studies have made it clear that the newly formed free chlorine becomes significantly unstable in the boosted water, probably with an increase in the disinfection byproducts (DBPs).

This type of boosting is similar to a situation where chlorinated water is mixing with chloraminated water somewhere in the distribution system from two separate plants that are discharging the treated water with two different disinfectants.

BP Chlorination Results in the Lab and Field

Several different distribution waters within the normal range of chloramines (3.0-3.5 mg/L) changed into low levels (0.3-1.6) after being stored in the refrigerator (5° C) for four days. However, the water still maintained the same level of total ammonia (ammonia value after thiosulfate dechlorination). These samples then were boosted with BP chlorination all to the same level (3.0-4.0 mg/L). However, a rapid decline occurred (approximately 50 percent reduction at 25° C in 24 hours) with the all-boosted distribution water after BP chlorination.1 Such a reduction was not detected when BP chlorination was made with a phosphate buffer (Table 1).

Over the last several years, C.C. wholesale purchasers had occasional difficulties in maintaining the chlorine residual at the utilities or in the field after applying the chlorine gas for boosting. In order to prevent this situation, a suggestion was made to remove the biofilm layers inside the storage tank. This removal minimizes problems in nitrification or other biofilms involved in the dissipation during the storage of boosted water. Another suggestion was to measure the nitrite and nitrate increase due to the nitrification (C.C. discharge water contains nitrate 0.2-0.7 mg/L, and nitrite <0.02). While this might be a partial solution, it was found that significant dissipation occurred even without nitrification after the breakpoint chlorination.

A rapid decline of free chlorine after boosting was observed occurring in the C.C. distribution water in the laboratory. However, it may be developing on a large scale in the storage tank. The same mechanisms might cause this rapid decline in both the laboratory and storage tank. This phenomenon shows the difficulty of chlorine dosing and residual maintenance by utilities after boosting with chlorine gas.

BP chlorination tests were conducted not only with the C.C. distribution water, but also with the plant discharge and field waters at the utility's intake. All water originally was processed through the C.C. treatment plant.

Instability of BP Free Chlorine

When the same disinfectant is used to boost the C.C. water, there is no problem with the dissipation of the chlorine residual after boosting. However, a rapid decline of the chlorine residual occurs during the first 24 hours (at a rate between 50 and 75 percent). This same decline occurs in the sampled water from the plant discharge, the water from utility intakes (30-40 miles from the plant) or the distribution water inside the city (Table 2a, b, c and d). This rapid decline after BP chlorination also was observed in the neighboring town of Robstown's plant discharge water (monochloramine) that shares the same source water (Nueces River).

Most likely, some of the stable, unreacted free chlorine demand from the source water had not been neutralized at the plant due to the concurrent injection with chlorine and ammonia to form the chloramine. In addition, the residual demand moves toward the distant sites to react with the newly formed free chlorine. This type of demand is so stable that it is active even after three weeks in the refrigerator (Table 2b2). The monochloramine is formed in the C.C. plant by injecting free chlorine and ammonia sulfate concurrently at two different sites, at the front (before rapid mixing) and in the clear well to form the monochloramine.

Therefore, some of the free chlorine demand would not react with the free chlorine in the plant to slip away into the distribution system. The TOX (total organic halide), formed by chloramination is only 10 to 20 percent of the TOX formed by chlorination with source water.2

The residual free chlorine demand can survive the long haul from the plant to the intake to affect the chlorine disinfectant even in the summer. Studies were made on chlorine boosting by breakpoint chlorination using two utilities' waters both receiving very low chloramines (0.1-0.5 mg/L) and needing boosting, as well as distribution water with a normal chloramine level (2.5-3.5 mg/L) inside the city. A comparison also was made in order to find out differences between the laboratory and the field, especially in respect to the stability of residual chlorine after boosting chlorination.

A low residual (total <0.5 mg/L) is common when sampled 30 to 40 miles from the plant. The dissipation of monochloramine1 is caused by pH, temperature, longer retention time in the biofilmed pipes (e.g., nitrification) to interact with chloramines and the operator's setting of the ratio of Cl:NH3.

Stability of Chloramine and Cl:NH3

Studies were made with a phosphate buffer (Table 1) in which the monochloramone is formed using different ratios of ammonia to free chlorine without any chlorine demand. The results were found to be different depending on the ratio of Cl:NH3-N. A monochloramine formed at a low ratio, (e.g., 1:1, 2:1 or 3:1) is very stable at 25° C for several days, but at a higher ratio (4:1), the residual is stable for the first 24 hours, but slowly becomes unstable (Table 1a and b).

A ratio as high as 10:1 to 14:1 enters into breakpoint chlorination. Free chlorine formed after BP is stable except for the first 24 hours (20 to 35 percent reduction, Table 1b). This probably is due to residual chlorine-chloramine interaction and the resultant reduction of chlorine residual. The formed free chlorine is very stable after 24 hours because of the absence of any chlorine demand in the phosphate buffer.

Part two of this article covers chloramine levels in the water as well as the THM rule and boosting safety.

About the Author

Shin-ichi Tokuno