Page 313 - Failure Analysis Case Studies II

298
of the pipe after 7 years. Localized attack of this sort would have been encouraged by the wet
crevices which probably formed between the polystyrene and the outside of the pipe.

9. DESIGN IMPLICATIONS

Many central heating installations function perfectly well without corrosion inhibitors. But what
must be done to avoid corrosion when inhibitors are not used? The first requirement is to make
sure that a uniform, stable film of oxide forms on the metal to act as a barrier to corrosion. The pH
of the water must be in the range where a passive layer is thermodynamically possible. pH ranges
can be estimated from the Pourbaix diagrams (see Appendix A): they are 7-1 2.5 and 9-14 for copper
and iron, respectively. Hard water (pH%8) is less aggressive to copper and steel than soft water
(pH~6.5). The initial condition of the system has an important influence on the state of the
protective films. It is bad to test a system with water and then leave it drained down. Corrosion will
occur in the aerated pool of water at the bottom of each radiator, and the film may be penetrated
by corrosion pits. It is best to put the system into working order straight away, and to keep it filled
with water. Because chloride and sulphate ions attack the oxide films, the water should be as free
from them as possible. Copper pipes are sometimes attacked by residues of soldering flux or graphite,
and these, too, should be avoided.
Once a stable film has formed, the rate of corrosion depends on the cathodic reaction. The
oxygen-reduction reaction can be prevented by keeping the oxygen content of the water as low as
possible. Pumps should be placed so they do not pump water through an open expansion tank or
draw air into the system. The water in an open expansion tank should be kept cold to minimize
evaporation. It is preferable to use a sealed air cushion expansion tank instead of an open expansion
tank. Obviously, the system should not be drained if this can be avoided. The hydrogen-reduction
reaction can be minimized by keeping the pH high, and the temperature low.
Because many systems depart from these ideals, it is common to add inhibitors. But these
must be used and specified correctly. They are generally recommended for systems which contain
aluminium because its oxide film breaks down under mildly alkaline conditions. To get the best
results, the internal surfaces should first be cleaned by circulating a chemical descaling agent through
the system. The system is then rinsed, and the inhibitor is added immediately afterwards. The
inhibitor will react strongly with the bright metal surfaces, and will penetrate well into the crevices
formed at the welds.
Inhibitors are often supplied as a multicomponent “package”. Sodium nitrite can attack the lead-
tin solder in soldered joints; nitrite-based inhibitors often contain sodium nitrate to prevent this.
Sodium benzoate is often added because it resists pitting better than sodium nitrite. When the water
contains more than one inhibitor, the overall effect is usually better than the sum of the effects of
the individual inhibitors (there is usually a synergistic effect). The pH of the water is controlled by
adding a buffer such as sodium borate. Many packages contain specific copper inhibitors such as
benzotriazole; this is especially important when the system contains aluminium. In hard-water areas,
a scale inhibitor is also added to stop hardness deposits building up in the boiler. Finally, a biocide
is needed to stop bacterial corrosion. The design of inhibitors is a complex business which draws
on long experience of laboratory tests and field performance.

REFERENCES
1. Pourbaix, M., Atlas of Electrochemical Equilibria in Aqueous Solurions. Pergamon Press, Oxford, 1966.
2. Shreir, L. L., ed., Corrosion, Vol. 1: MelallEnvironment Reactions, 2nd edn. Newnes-Buttenvorths, Oxford, 1976.
3. Boffardi, B. P., in Metals Handbook. Vol. 13,9th edn: Corrosion. American Society for Metals, Metals Park, OH, 1987,
p. 487.
4. yon Fraunhofer, J. A., British Corrosion Journal, 1971,6,23.
5. Butler, G., Ison, H. C. K. and Mercer, A. D., British Corrosion Journal, 1971,6,32.
6. Fontana, M. G., Corrosion Engineering, 3rd edn. McGraw-Hill, New York, 1986.
7. Cotton, J. B. and Jacob, W. R., British Corrosion Journal, 1971,6,42.
8. Stott, J. F. D., Metals and Materials, 1988,4,224.
9. Rozenfeld, I. L., Corrosion Inhibitors. McGraw-Hill, New York, 1981.
10. Scully, J. C., The Fundamentals ojCorrosion, 2nd edn. Pergamon Press, Oxford, 1975.