This paper describes the findings of a study to develop a “Disinfectant Demand Sensor”. The aim was to determine the 3- or 7-day demand of a water sample in a shorter period of time, such as a few hours. Three approaches were taken to achieve this: (i) assessment of water quality surrogate parameters, (ii) “short” disinfection demand assessment, and (iii) temperature elevation approaches. All these methods can be used for the prediction of chlorine, but only temperature elevation and short demand assessment can be used for the prediction of chloramine with good accuracy.
Water quality data (prior to disinfection) was used to identify surrogate water quality parameters for disinfectant demand. The study found that the surrogate parameters of major interest were UV254, DOC, and colour with R2 values of 0.94, 0.91 and 0.83, respectively for chlorine demand prediction.
A rapid fractionation technique using ion-exchange resins to separate the organic component into various fractions demonstrated that the VHA (Very Hydrophobic Acid/humic acid) fraction showed the best relationship with chlorine demand and was also identified as a possible surrogate parameter with an R2 value of 0.88. Chloramine demand did not behave similarly. Other factors influencing chloramine decay make it more difficult to predict than chlorine decay.
The possibility of using the demand of a shorter time span to predict demands of a longer time was also investigated. This technique was proved to work well for both chlorine and chloramine and results have shown that a 3-h chlorine demand correlated well with the 3-d demand (R2=0.97). Also, the 1-d chloramine demand correlates with the 3-d demand (R2=0.92).
Thirdly, the temperature investigation used the concept of raising the temperature of a sample thereby reducing the time taken to reach the 3-d demand equivalent. This method was found to be suitable for both chlorine and chloramine demand prediction. For example the 3-d chlorine/chloramine demand at 20°C equivalents of Myponga water can be determined after just 5.3/5.5 h respectively when the sample is incubated at 40°C.
- demand prediction
- water quality
- Received May 2, 2005.
- Accepted June 21, 2006.
- © IWA Publishing 2006