Technical Topics - "Cost of Corrosion"
by the Technical Secretary, Douglas Mills
The last CM has only just hit the mats as I write this so not had much feedback yet on corrosion of aluminium. This month I want to talk mainly about cost of corrosion. I decided this might make an interesting theme even before I picked up my daughter's copy of New Civil Engineer a couple of weeks ago and read about the Forth Road Bridge. It appears that the multistranded steel cables holding the bridge up were inadequately protected and are corroding. The estimates for cost of repair vary between £91 and £122 million and will take between 7 and 9 years to complete. The cost of associated travel delays will be between £238 million and £335 million. So already we have nearly 30% of the annual figure which the Hoar report arrived at in 1972 (around £1365 million) for the total cost of corrosion. And that on just one structure! Ok there has been inflation and the cost for repair may spread over several years. But it is still frighteningly high.

The cost of corrosion is a hard figure to pin down. Part of the figure can be arrived at by costing corrosion prevention methods. Like changing the metal/material (we could stop corrosion if everything was made of gold ; but gold is scarce, unacceptably costly and its mechanical properties leave something to be desired). Changing to an expensive metal is done when it makes economic sense or is obligatory eg using condenser tubes for condensers subjected to aggressive impingement conditions and extensive use of stainless steel in the food industry. Certainly if corrosion dictates the change such a cost is reasonable to include at least the first time it is done.

More common than changing the metal is to protect a base metal (often carbon steel) with a coating (CP, Inhibitors also have part to play and have an attendant cost). Thus in the car industry the extra cost of corrosion protection (mainly the body shell coating system) was estimated at about £90 per car in 2000 (£120 a car today?). Note that the extra cost at the design stage of a good protection system is probably small compared with the cost of subsequent remedial work when the system fails (certainly £120 is a small figure compared with the costs of putting right car paintwork which has been scratched!) Unexpected failures may additionally lead to highly expensive collateral damage eg when the bridge collapses because the rebar has corroded, the train falls off the line due to a corroded rail or the pressure vessel breaks or the pipeline fails. That figure is even more difficult to pin down (I guess most analyses prior to today would not have included the cost of extra travel mentioned above).

There is also the common situation of throwing away the whole component because of corrosion of a small piece as I had to do with my son's bike lights where the circuit board contacts were poorly protected (this is true of most electronic components - anything involving a circuit board is not worth repairing). As an interesting aside there is an organic coating called parylene applied by a PVD process to protect vulnerable electronic circuitry eg those used in aerospace and medical applications - if this was applied more universally maybe the need to discard before their time of items containing electronic parts would be much reduced. When it became unacceptable to have to scrap cars after say just ten years because of rot, car manufacturers eventually improved their protection systems.

Some seven or eight years ago the Government decided the Hoar report should be updated and commissioned the PRA (Nigel Whitehouse(NW) and Helma Jotischky(HJ) and IOM (Paul Mclntyre) to investigate cost of corrosion in five key UK industry sectors : Automotive (HJ), Constructional Steelwork(NW), Offshore (PM) , Chemical and Petrochemicals (PM) and Food and Drink (NW). These reports were issued from October to December 2001. The figures for 1998-2000 were respectively (in fMillion): 160, 2500, 250, 1730 and 5 (one company).

Compared with 1972 it would appear that automotive and food and drink sectors had improved considerably, the other areas less so. Thus in the two areas above that PM looked at corrosion under insulation is a big problem. The full reports make fascinating reading and are available (at a cost-the original Hoar report was public domain) from PRA. Studies have also been done in USA (see figure: courtesy of the US study) and Japan with similar results. Overall it appears metallic corrosion still costs the economy something around 3% of GDP. Note this figure is using corrosion in its narrower (metallic) sense. If we use it in the broader sense to include the deleterious interaction of any material with the environment - well maybe 10% is not too fanciful! The work to confirm that awaits doing. As usual any comments please send to douglas@harrbridge.freeserve.co.uk
Note: thanks are due to Paul Mclntyre for assistance with this month's article.