Distribution and reticulation systems for potable water represent a large proportion of

the capital invested by water suppliers. These installations are prone to damage by corrosive water that leads to inconvenience to the consumer when water supply is disrupted and financial losses in the repair and maintenance thereof. In many incidents, unaccounted water losses can be related directly to leaking pipelines as a result of corrosion.

Corrosion in long pipelines and in reticulation systems can be managed by controlling the calcium carbonate precipitation potential (CCPP) of the purified water by adjustment of the pH and alkalinity levels. This method is not consistently successful as the adjustment of the desired parameters is not always practically possible due to shortcomings in plant design or a lack of technical knowledge.

There is evidence that monochloramine and sodium silicates can effectively be administered to inhibit corrosion. However, it is seldom utilized for that purpose in the production and distribution of potable water.

The use of monochloramine could serve a two-fold purpose. It is primarily used to preserve microbiological quality of water in a distribution system and as a bacteriostat to prevent biofouling. Secondly, it could be used as corrosion inhibitor. The advantage of chloramination is that trihalomethanes, additional to those formed during the primary disinfection with breakpoint chlorination are not formed, nor are total organic halogens or mutagenicity levels increased.

Sodium silicate is easy to administer and do not pose a health threat since it is non- toxic in the concentrations normally used. It is inexpensive, tasteless, colourless and odourless in water and has no detrimental environmental effects.

The objectives of the research program were as follows:

1. To determine the extent by which corrosion in steel pipes can be inhibited or reduced by the use of monochloramine or sodium silicates;
2. To determine whether these two chemicals can be used in combination with any type of coagulant;
3. To determine the optimal dosages requirements of the chemicals and the conditions necessary for its effective use.

Two distinct experiments were undertaken, namely a trial based on weight loss of mild steel coupons exposed to water treated in nine different ways and a second trial with mild steel electrodes exposed to water treated in the same way in which the corrosion rates were measured with electro-chemical techniques.

The aim of the two experiments was to measure both short term and long-term effects on the corrosivity of mild steel with different treatments.

The results show clearly that the initial corrosion rates are very high and do not reflect the method of treatment. From this it can be concluded that the long-term performances of a corrosion inhibitor should not be judged on observations made during the first few weeks of exposure of coupons. The time before equilibrium was reached for the experiments described in this document was seven weeks.

In the weight-loss experiments, lower corrosion rates were measured with the addition of monochloramine, even when it was dosed in combination with sodium silicate. Lower corrosion rates were measured with incremental higher dosages.

The addition of only sodium silicate had no effect on the measured corrosion rate as, compared with the control sample without any additives. Although the analysis of the corrosion products indicated the presence of silicate, it can be concluded that it had no effect on the corrosion rate. However, there is overwhelming evidence in the literature that support the claim that sodium silicate is an effective corrosion-inhibitor. It is possible that the duration of the experiments was not long enough to observe its effect since sodium silicate is only effective when the tertiary corrosion products, which provide better protection against corrosion, are formed.

The rapid scanning technique demonstrated completely different results. The I_corr values with the monochloramine treatments demonstrated higher corrosion rates compared with the control and the treatments with only sodium silicate. An analysis of the data, where the I_corr values are plotted in relation with the E_corr values; demonstrates that there is a narrow region of E_corr values where corrosion is limited. At a surface potential below or above these values, increased corrosion rates are measured. The results clearly show that, at higher monochloramine concentrations, higher corrosion rates are measured. At lower dosages, corrosion is also stimulated.

It appears then that there is a narrow margin within which the monochloramine concentration should be maintained to protect the mild steel pipelines. It also appears that it is in the same level of 1-2 mg/l as required for bacteriostatic activity.

This has a significant practical implication since monochloramine can be used to passivate the surface of corroding metal, provided that concentrations are controlled within the required limits.

It is recommended that in order to optimise the use of monochloramine, field tests should be conducted at several points in a distribution network in order to find a correlation between the surface potential of the pipeline, the monochloramine concentration in the bulk of the water, the bacteriological quality of the water and the corrosion rate.

It is recommended that: -

• a study into the mechanism of corrosion inhibition by monochloramine and silicate be conducted
• an investigation into the application of monochloramine and silicate as corrosion inhibitors in water with chemical composition typical to the soft coastal waters compared to harder inland waters be conducted.
• an investigation into the use of monochloramine and silicate as corrosion inhibitors for metals other than mild steel, such as copper tubing, be conducted.