This series of articles is intended to highlight industry-wide engineering experience, guidance, and focussed advice to practising technologists. It is written by ICorr Fellows who have made significant contributions to the field of corrosion management. This issue features The Corrosion Resistant Properties of novel cements, concretes, and reinforcement by John Broomfield of Corrosion Engineering Solutions Ltd.
The Corrosion Resistant Properties of Novel Cements, Concretes, and Reinforcement
We use more concrete than any other material other than water. However, each tonne of concrete has a significant carbon footprint and the technical press is full of articles on new ways of reducing that carbon footprint by modifying its constituents. There are many ways of doing this, most of which involve changes in the internal chemistry of the finished product.
Concrete is made from cement, fine and coarse aggregates, and water. To this may be added other materials or admixtures, such as superplasticisers to increase workability during placement on site, set retarders or accelerants for hot or cold climate work, or even corrosion inhibitors for aggressive conditions where the steel reinforcement needs extra protection. The cement itself can be Portland cement or, more commonly these days, a blend of Portland cement and other cementitious materials such as pulverised fuel ash (pfa), ground granulated blast furnace slag, microsilica or metakaolin. The perceptive reader will note that some of these materials are going through their own changes and availability issues in response to the need to reduce carbon emissions. There are reports in the press of applications to mine deposits of pfa from ash pits as the supply of material direct from operating coal fires plants is diminishing.
The methods of achieving the required performance and durability of blended cements are covered in UK and European standards BS EN 206 parts 1 and 2, and BS 8500 parts 1 and 2. The latter gives tables of exposure conditions along with the necessary mix design, and minimum cover requirements to achieve a required minimum design life for a given exposure condition. There is also BS EN 197-1:2011 Cement Composition, specifications and conformity criteria for common cements, which gives the specifications of 27 distinct common cements, 7 sulphate resisting common cements as well as 3 distinct low early strength blast furnace cements and 2 sulphate resisting low early strength blast furnace cements and their constituents.
Portland cement based concretes provide corrosion protection to reinforcement in two ways, as a semi permeable coating and as a corrosion inhibitor. Concrete is permeable to water and oxygen. The reason that reinforcing steel embedded in concrete does not corrode is that the pores in concrete contain an excess of calcium hydroxide and other hydroxides, maintaining a pH of 12 to 14 [1]. The upper level is carefully controlled because some aggregates are at risk of alkali aggregate reaction which leads to the creation of an expansive gel which will damage the concrete.
However, in order to prevent corrosion, the pH must stay above 11, which is the threshold for corrosion. There are two mechanisms for inducing corrosion in reinforced concrete structures without physical damage to the concrete cover to the steel. One is carbonation; the process where atmospheric carbon dioxide reacts with pore water to form carbonic acid and neutralise the excess calcium hydroxide in the pores. This leads to a carbonation front which progresses through the cover. Once it reaches the steel, the pH drops below 11 and corrosion can occur. The other process is chloride attack. The chloride ions in solution (usually from sea water or deicing salts) diffuse through the concrete cover to the steel. Once the concentration at the surface exceeds a threshold level, the passive layer on the steel breaks down and corrosion ensues.
Over the years since the 1970s, much work has been done to understand these mechanisms in Portland cement based concretes. We have a simple equation to estimate the rate of carbonation of the form, x = kt½ here x is the carbonation depth, t is time and k is a constant which can be measured or estimated on a structure by structure and microclimate basis.
For chloride ingress the equation is more complicated, as Fick’s second law of diffusion applies and the threshold concentration must be estimated. However, there is much guidance now available on the parameters for predicting chloride diffusion rates into Portland cement and blended concretes [1].
However, this type of guidance has not been developed for non-Portland cement based concretes. While civil engineers have great expertise and applicable useful standards and test methods for determining the physical characteristics of new concrete, when it comes to the more chemically based properties, such as corrosion resistance and durability, there is far less guidance and often less in-house or externally available expertise.
When blended cements were used with high levels of alterative cementitious materials, there was concern that the alkali reserves would be depleted, increasing the rate of carbonation and possibly also chloride diffusion. However, since these blends were less porous than ordinary Portland cement concrete, testing and experience soon showed that there was no loss of durability. In fact, durability increased. However, there are new alternatives to Portland cement itself such as alkali activated cementitious materials (AACM). Because AACM has no Portland cement, it is outside EN 197-1, EN 206 and BS 8500. To prove a candidate new AACM is durable, EN 197-2 and EN 206 permit users to demonstrate equivalence and show the material has the required structural and durability properties.
As might be expected, AACM initially has a pH that is higher than conventional Portland cement and concrete (14 or so), but the alkali reacts during the setting process. There is therefore a concern that this cement free, lower carbon concrete does not have the same reserve of solid Ca(OH)2 to buffer carbonation, so it would carbonate faster than an equivalent convention Portland or blended cement concrete mix.
Recent testing suggests something else is going on. It seems the initial carbonation rate is higher, but a pore blocking process then seems to start, blocking CO2 ingress, and the rate of carbonation slows right down. The steel therefore remains in a dense, high pH matrix. However, this good performance may be product and mix specific. This suggests that the carbonation rate equation above is not applicable to AACM concretes, or that the constant k may be harder to determine over the long term.
RILEM (The International Union of Laboratories and Experts in Construction Materials, Systems and Structures) have produced a State of the Art Report on Alkali Activated Materials [2], which is a comprehensive review of these materials, their formulations, chemistry and performance. It discusses the durability of the material and of structures made with it, as well as the testing of the materials during production and casting and also field testing of existing structures. It found considerable variability in performance and test results, particularly with regard to transport properties, corrosion performance and durability. The lack of specifically applicable performance standards was a concern.
Another major contribution to the field is “The Field Performance of Geopolymer Concrete Structures report” by the Cooperative Research Centre (CRC) for Low Carbon Living Ltd in Australia [3]. The report describes in-situ testing and core sampling of geopolymer (alkali activated) concrete at four sites across Australia and long-term performance monitoring of two geopolymer concrete structures. The field test results found extremely variable resistance to carbonation and to chloride ingress and the authors stated that this confirms the necessity of developing performance based specifications for Geopolymer concretes
Like the RILEM report, the CRC report concludes that suitable testing methods are required to assess the performance of concrete in order to assist engineers to specify Geopolymer concrete conservatively and confidently, particularly in more aggressive environments.
Unlike the physical testing of concrete for properties such as cube strength, the durability testing is more complex. We need to understand the chemistry of the material and how it changes over long periods of time if it is to be used with conventional steel reinforcement. We already have examples of unsuitable use of specialised concretes most recently reinforced autoclaved aerated concrete leading to failures and expensive repairs or rebuilding.
The obvious corrosion related tests for novel concretes are for the alkali reserves in concrete to be measured and its resistance to chloride and carbonation. There are tests for these under BS EN 205 and BS EN 1504. However, we know that these properties can be very variable for alkali activated materials and change over the long term so a full understanding of the chemistry and its changes over time are required before we can develop prescriptive tests.
As one of the concluding statements of the RILEM Report states. “These issues are by no means limited to the area of AAMs – these points are relevant across many areas of non-traditional cement and concrete development and commercialisation”. The replacement of Ordinary Portland cement based concretes with new materials expected to perform and be durable for many decades is a challenge for engineers in the construction industry.
Of course, one solution to the corrosion risks of reinforced concrete is to use alternative reinforcement. There have been corrosion resistant reinforcement materials available for many decades. For ferrous materials these include galvanised steel reinforcement, fusion bonded epoxy coated reinforcement, a range of stainless steels and low alloy steels. These are reviewed in the AMPP/NACE State of the Art Report on Corrosion Resistant Reinforcement [4]. Some of the ferrous materials can on the whole be treated by conventional testing both in the laboratory and on site, the coated ferrous reinforcing bars, less so.
However, there is now a range of non-metallic reinforcements, including glass and carbon fibre reinforced polymers and more recently basalt. These are polymers with fibre reinforcement. While we can detect ferrous reinforcement in a structure with electro-magnetic based cover meters, we have no simple way of detecting these materials. Will we be using more ground penetrating radar to detect non-metallic reinforcement? Has any work been done on detecting these alternative reinforcement materials once they are cast into a structure?
Do we understand the deterioration mechanism for these products? The failures of fusion bonded epoxy coated reinforcing bars in Florida were partly due to the softening and debonding of the epoxy in a saturated environment, particularly where the coating was stressed on the outer radius of bends. These failures happened over a few years. Could similar degradation of polymer-based reinforcement occur in concrete over the decades? If so, how will we detect it? We have NDT techniques such as reference electrodes, resistivity meters and linear polarisation to provide information of the corrosion condition of ferrous reinforcement. We have no equivalent tests for polymer-based reinforcement. If we start putting alternative reinforcement materials in alternative concretes will we understand their interactions in the short and long term?
It is obviously important to innovate to reduce the carbon footprint of the construction industry and improve the durability of the built environment. However, we need a deep understanding of the materials we are using as well as suitable standards and test methods and equipment to ensure the products we use can perform in the environment they are exposed to and can be assessed and maintained throughout their lives which will last for many decades. Corrosion engineers have a major role to play in the understanding, testing and developing performance-based specifications and tests for the new wave of low carbon and corrosion resistant materials in the construction industry.
The author would like to acknowledge the help of Professor Peter Robery for his advice on alkali activated concretes and the RILEM and CRC reports.
References
1. Broomfield, J.P. Corrosion of Steel in Concrete, 3rd Ed. Publ. CRC Press London, 2023.
2. John L. Provis, Jannie S.J. van Deventer Editors, Alkali Activated Materials State-of-the-Art Report, RILEM TC 224-AAM Publ. Spinger, 2014.
3. Stephen Foster et al. The Field Performance of Geopolymer Concrete Structures report RP 1020 by Cooperative Research Centre (CRC) for Low Carbon Living Ltd in Australia, 2018.
4. NACE Publication 21429-2018-SG – State of the Art Report on Corrosion-Resistant Reinforcement Publ. AMPP, Houston TX.
CAPTIONS:
Hilti Ferroscan, a scanning cover meter, can identify the size and depth of steel reinforcement. New techniques will need to be developed for detecting and locating non ferrous reinforcement.
