Technical Topics - Corrosion in Concrete

This month again I thought I'd draw on personal experience (although a little more recent than my nuclear adventures!) and talk about corrosion of rebar in concrete. This is a big field - and one involving large sums of money - a fairly recent report [1] on highway bridges on the USA concluded that the direct costs of repair of "corrosion" were between $6.4 and $10.7 Billion with indirect costs up to 10x that.


Concrete is a highly complex and "not as well controlled as one might like" type environment with variability both within the solid phases and within the pore water. Although it is not "natural" for plain carbon steel to corrode in concrete (the pore water is normally alkaline [pH of 12.5 or above] and in such environments steel should be passive) things can go wrong! Carbonation of the concrete can occur removing the alkali. Or the concrete can be inherently more porous than it should be allowing chloride ions to penetrate particularly in sea water environments or as a result of road salts. These similarly destroy the passivity. Or cover can be inadequate.

Interestingly all of the four methods commonly used to prevent corrosion have been tried to mitigate the effect particularly of poor quality concrete leading to access by chloride. All of them have their advocates and also their detractors (one reason why getting reproducible results in this area is difficult may be because of the afore mentioned complexity arising from the use of natural products of relatively uncontrolled composition)

For unbiased good quality reports I would recommend the work of the BRE. What I write below though is very much my own view. So one way is to add inhibitors to the concrete. This is a big and somewhat controversial area. Certainly calcium nitrite can be effective but it doesn't seem the best approach for general application as much of the inhibitor will be sitting in the concrete not doing much except maybe making the concrete more hydrophilic.Then there is the approach of changing the material to say stainless steel. Again this seems to be hammer to crack a nut stuff. Excellent material that it is, stainless steel is expensive and also not exempt from corrosion in chloride or in low oxygen (crevice) conditions.

So what about electrical methods like impressed current cathodic protection? (I am less aware of sacrificial anodes being used for this application). CP is attractive because it can be applied retrospectively ie when the structure is already corroding (apart from lowering the metals to around its protection potential it can be combined with chloride removal and realkalisation). But there is the challenge of good control of the CP /possible overprotection and the like.

The fourth approach of use of coatings is the area that I have had most to do with. At one time epoxy coated rebar was all the rage and while working at the PRA in the late 80s we did an evaluation of this. In principle it looked promising but there had been some problems in service. The most famous of these was at the Florida Keys where epoxy coated rebar had been tried and failed probably due to being joined electrically to non-epoxy coated rebar. There was also the fact that breaks in the coating become potential galvanic cells (small anode/large cathode) and the mechanical properties (pull out strength etc) were not necessarily as good as using uncoated steel (note the steel normally goes in a bit rusty anyway).

So how about zinc coatings? Well these have their uses (galvanic protection at breaks etc). However rather than passivate in alkali like iron, zinc may tend to dissolve and this needs to be kept in mind! There is also the possibility of impregnating the outside of the concrete with a coating (often silicone based) to reduce its permeability to water and ions. This has proved successful with road decks.

Other work I have been involved with in this area has been developing methods of monitoring. It is on the face of it rather similar to the challenges associated with assessing the state of a metal under an organic coating. Hence, it seemed reasonable to try some of the methods that we use successfully with coatings like LPR, EIS and particularly Electrochemical Noise (single substrate) Method (ENM), to rebar in concrete. A few papers comparing ENM with LPR were presented at conferences (work done with Aston University) in the early "noughties" .There is still work to do but I believe ENM is promising particularly for continuous monitoring with probes built in at the start of life.

That's about it on concrete. There is one other related point that I was asked about recently. What is the difference between the corrosion of pure iron and the corrosion of 0.5% carbon steel and which (if either) would you use for a sacrificial anode? (assume in sea water). Any thoughts on that or anything else technical please contact me: Douglas@Harrbridge.freeserve.co.uk
1) Corrosion of highway bridges: Economic impact and control methodologies M Yunovich and N G Thompson , Concr. Int. 2003, 25, nol,pp. 52-57