Patch Repairs (concrete)
	Cathodic Protection (concrete)
	Choride Removal (Desalination) (concrete)
	Realkalisation (concrete)
	Corrosion Inhibitors (concrete)

Patch Repairs

Before approaching concrete repairs, consideration must first be given to the cause of the problem. This is fundamental to the success or failure of the repair, and a lack of adequate attention at this point can jeopardise the whole job.

If the problem has been diagnosed as being due to carbonation induced or chloride induced corrosion of the reinforcement, patch repairs may be used, although with the latter, there are precautions needed to ensure a successful repair. In these notes we shall deal with cracking and spalling due to reinforcement corrosion only. If the problem has been diagnosed as due to ASR or any of the other mechanisms of failure, the repair may need to be specifically designed for that particular contract and structure, and it is impossible to generalise about the approach which might be taken.

A poorly executed patch repair.  This was cut out to show that the contractor did not cut behind the steel, or remove corrosion from the bar.  The life of such a repair is very short indeed.

Patch Repairing Carbonation Induced Corrosion

It must firstly be understood that carbonation is a variable process and the carbonation front will probably not be uniform over the structure. Regular testing to determine the cover to the steel reinforcement and the penetration of carbonation, by the phenolphthalein spray method, will be required.

Criteria we have employed in the past have been to establish which areas of concrete fall into the following categories

1. Visibly spalled concrete with exposed steel reinforcement.
2. Areas which sound hollow when tapped lightly with a club hammer (these may often be found around actual spalled areas showing the problem is worse than is visually apparent).
3. Areas where the cover to the steel reinforcement is less than 10mm.
4. Areas where the carbonation front has encroached to within 5 mm of the reinforcement. (this necessitates carbonation testing at least every 2 metres or so).
5. Areas of honeycombed concrete.
6. Areas where the half cell potential values numerically exceed -200 mV (copper/copper sulfate). Using this criterion, however, caution must be exercised as half cell potentials only show areas of active corrosion. In the summer, whole areas of corrosion may shut down as the concrete dries out. Detection of carbonation induced corrosion is not as reliable as for chloride induced corrosion using the half cell method. If the above criteria have been followed, half cell potential testing is not strictly necessary but can be a useful additional aid to diagnosis.

The concrete surface will usually be water jetted prior to cutting out or lightly grit blasted to remove surface grime and deposits and this will also often reveal blowholes in the concrete surface.

All of the areas identified above will require breaking out behind the steel reinforcement, sufficient to get one's fingers behind the steel, ensuring that the edge of the breakout is cut square and not "feather edged" which inevitably causes failure of the repairs.

At all locations, the steel must be at very least wire brushed to remove all loose deposits or preferably grit blasted or water jetted back to bright metal. If chlorides are present this is essential and will be dealt with in the next section. If the loss of section of the steel exceeds a critical amount, it may be necessary to splice in additional or replacement reinforcement. This decision will be taken by the Engineer.

Once the area to be repaired has been cut out and cleaned thoroughly of all debris and dust, the repair can begin. Usually, purpose designed repair mortars will be used supplied by one of the leading manufacturers. These usually contain one or more of acrylic or other polymers, shrinkage compensating additives, silica fume, ggbfs or a variety of other chemical modifiers to improve workability, permeability, or ease of application. Simple sand/cement mortars alone should never be used, although gauging sand and cement with, for example, an SBR polymer emulsion has been used and can work provided the mortar is correctly batched and gauged with SBR according to manufacturer's instructions.

It is likely that, for pre batched proprietary repair mortars, some kind of bonding agent will be specified. These must be applied strictly in accordance with the manufacturer's instructions, which will often call for the repair to be applied while the bonding agent is still wet. If used incorrectly, bonding agents can make very efficient debonding agents!

If no bonding agent is used, the broken out surface of the concrete must be thoroughly dampened to kill the suction, ensuring that the surface is saturated but surface dry at the time of application of the repair. We have been involved in the investigation of many failures where the substrate was not adequately wetted and water has been sucked from the interface of the repair and backing concrete, causing poor hydration of the cement and failure of the repair.

The repair mortar is then mixed up again following fully any special instructions from the manufacturer. Repairs are then usually hand placed packing the mortar on in layers no thicker than those recommended and using the fingers to pack mortar behind the bar. Wearing of suitable protective clothing is essential as cementitious repair mortars are very alkaline and can cause burns or dermatitis. Similarly, dust masks must be worn when mixing the mortar. The Health and Safety at Work act and the more recent CDM regulations place very specific responsibilities on both site operatives and their supervisors and employers to work safely on site. Once the repair has been filled, a wooden rule is used to level the repair, and the repair is then floated to a smooth finish with a steel float. The repair must then be cured in accordance with the manufacturer's recommendations. This may involve the application of damp hessian or more usually a spray applied curing membrane.

Once finished and cured, there may be surface irregularities and blow holes and there will almost certainly be colour variation between the patch and the surrounding parent concrete. It is usual therefore to apply a thin skim coat of a proprietary material, usually known as a "fairing coat" to fill in any surface blemishes and to mask the patches themselves. This is followed by the application of a final coating of anti-carbonation paint which has the dual purpose of stopping further carbonation in unrepaired areas and providing a pleasing and even colour finish.

The coating will require periodic renewal at perhaps 10-15 year intervals.

Patch Repairing Chloride Induced Corrosion.

The problem with chloride induced corrosion is that the corrosion mechanism is often not fully understood by Engineers and Clients. Whereas carbonation induced corrosion causes relatively large areas of the reinforcement to corrode, chloride contamination usually causes very localised areas of the steel to corrode, which is known as pitting corrosion. This can result in deep "pits" in the steel surface and considerable localised loss of section. Often the corrosion deposits are black and hard and there appears to be little loss of section at first, until the deposits are either dug out with a penknife, or water jetted clear, when the full extent of damage can be seen.

Repairing chloride contaminated concrete is similar to that previously written for carbonation induced corrosion except for several very important points:-

	All corrosion deposits must be removed by grit blasting or water jetting to bright steel. Failure to do so leaves chloride contamination behind in the corrosion deposits and failure will ensue.
	All of the chloride contaminated concrete must be removed, not just the areas which have spalled or delaminated. See comments below on the "incipient anode effect."
	A criterion often employed is to repair all areas where chloride exceeds, say, 0.4% or 0.5% of cement. Strictly, for chloride added at the time of mixing, the criterion should be 0.4% or for chloride ingressing post hardening, 0.2%. For pre-stressed, post tensioned or heat cured concrete, the maximum tolerable chloride level is only 0.1% by mass of cement.

The incipient anode effect

When an area of steel is corroding under the influence of chloride contamination, steel is dissolving causing the formation of iron "ions", tiny charged particles of iron. Simultaneously, electrons are released which flow along the bar and react at some point remote from the corrosion with both air and oxygen. Cathodic protection, which will be discussed in the next section involves providing a small electric current to the steel which prevents further corrosion. The corroding areas are supplying electrons to surrounding areas of steel effectively providing localised cathodic protection to the adjacent steel.

If you now break out a corroding area and apply a patch repair, without dealing with chloride contamination in adjacent areas, the cathodic protection system has been removed!!

New corrosion cells will rapidly spring up on either side of the repair and early failure will often ensue. One manufacturer uses a sacrificial anode tied to the rebar on either side of the patch repair.  We have recommended this system on several occasions and it is, for example, installed at a car park in Colchester.  We have published a paper on this project, copies of which are available free of charge.

The sacrificial anode system can only deal with limited levels of chloride, up to about 1-1.5%.  It can cope with higher levels, but the lifetime would be reduced.  In such cases, chloride removal (desalination) or impressed current cathodic protection may be viable alternatives.

These repair recommendations are only a guide and apply only to conventionally reinforced structures. Repair of pre-stressed or post tensioned structures with tendon damage is a specialist task and beyond the scope of these notes.

More detailed recommendations, together with specimen bills of quantities can be found in a Concrete Society Technical Report.

Cathodic Protection

Basic Electrochemistry

A galvanic cell consists of two different metals (electrodes) connected through a conducting solution (an electrolyte) and also connected externally completing a circuit. In such a situation, one of the metals (the more reactive) will tend to dissolve in the electrolyte while the other will tend to have new metal deposited on it. In the process of dissolving, the more reactive metal will liberate electrons which flow via the external connection (as an electric current) to be used in metal deposition at the other electrode. The dissolution and deposition reactions are called half cell reactions; the former is the anode reaction and the latter is the cathode reaction.

Corrosion

Corrosion is a similar process but with electrode sites at adjacent locations on the same metal surface. For the corrosion of steel in concrete, the two half cell reactions are as follows:

1. Iron dissolves forming positively charged iron "ions" and releasing electrons (e^-)

Anodic reaction: Fe ® Fe²⁺ + 2e^- Equation 1

2. The released electrons migrate along the steel and react with oxygen and moisture to form hydroxyl ions (OH^-)

Cathodic reaction: H₂O + ½O₂ + 2e^- ® 2OH^- Equation 2

The rate of the corrosion process depends upon the availability of the reactants which are iron (Fe), water (H₂O), oxygen (O₂) and electrons (e^-). Thus, for example, if the system is completely dry or starved of oxygen, the steel may still be active but the corrosion rate will be negligible. The relative availability of electrons is equivalent to the corrosion current which depends upon the electrical resistance of the system.

Reactivity

Whether or not the corrosion process can occur at all depends on the pH of the solution adjacent to the steel which, together with other conditions such as the availability of reactants, determines the electrical potential of the steel. Generally, a metal exists in one of three possible reaction states; passivity, corrosion or immunity depending on the pH and on the electrical potential of the steel.

Passivity and corrosion are conditions whereby the above reactions are occurring initially but the Fe²⁺ (ferrous) ion is unstable in the presence of oxygen and it will react further forming ferric oxide. The pH and electrical potential will determine the physical nature of the ferric oxide formed and thus whether the iron is corroding or passive.

The corrosion form of iron oxide is expansive, non-adherent and tends to flake leaving the iron surface exposed to further oxidation. At more positive potentials, a passive oxide film is formed on the metal surface. This film is adherent, self repairing and impermeable, limiting further oxygen access to the metal and effectively halting the corrosion process.

Immunity occurs at more negative potentials than corrosion and is a condition whereby the metal has a bright surface and cannot corrode. So called 'noble metals' like gold and platinum exist in this condition under normal circumstances but for iron, the electrical potential needs to be so highly negative that it will cause water to break down and bubbles of hydrogen gas will be seen on the surface. This is not the normal condition for steel and will generally only occur where an external potential is applied to it.

The presence of chloride ions significantly modifies the corrosion states of metals including iron. For iron the range of pH and potential for immunity remains the same but the range for general corrosion is considerably extended. In addition, a different type of corrosion called pitting corrosion can occur which is intense and localised. Consequently, the conditions for passivity become severely limited.

Cathodic Protection

The idea of cathodic protection is to artificially shift the potential of a metal so that it becomes either immune or passive. In natural soils and waters it is normal to shift the potential of steel to the immune region whereas for steel in concrete it is preferable to re-establish passivity.

In sacrificial anode cathodic protection, a galvanic cell is set up by connecting the steel to a more reactive metal, usually zinc. The zinc then undergoes the anodic reaction and corrodes whilst the steel is rendered entirely unreactive because the whole surface undergoes the cathodic reaction mentioned above and the iron no longer dissolves. This may also be thought of as the anodic sites on the steel being shifted to the zinc.

With impressed current cathodic protection, the steel is connected to the negative terminal of an electrical power supply forcing it to undergo a cathodic reaction. If the potential of the steel is made negative enough to make it immune, the cathodic reaction becomes one whereby water is broken down and hydrogen is liberated as follows:

2H₂O +2e^- ® H₂ + 2OH^- Equation 3

This situation would normally be avoided in concrete since the pH is high and it is possible to re-establish passivity by applying a somewhat less negative potential. This consumes considerably less current and so reduces the cost. The cathode reaction is then as given in reaction (2) above and the OH^- generated helps to maintain the conditions necessary for passivity.

The anode, connected to the positive terminal of the power supply, is usually chosen to be a relatively non-reactive conductor such as carbon or titanium so that its corrosion rate is low. The anode reaction then generates oxygen and acid (H⁺) as follows:

H₂O ® O₂ + 4H⁺ + 4e^- Equation 4

The current densities normally encountered in CP. systems are sufficiently low for the amount of acid generated to be safely taken up by the normal alkalinity of the concrete.

Practical Anode Systems

Techniques for the cathodic protection of steel in concrete vary according to the anode system used. The longest established system uses cast iron primary anodes over a bridge deck with a conductive asphalt secondary anode laid over them and an asphaltic concrete wearing course over that. The system operates at relatively low current densities with an expected life of 10 to 20 years. It is a low cost, low technology system which has proved to be highly reliable. Its disadvantage is its limited application being only suitable for tops of decks.

Conductive coating systems typically contain graphite in a resin binder and are applied at a thickness of 250 to 400 microns as a secondary anode. The primary anode consists of thin wires or strips of titanium laid into the surface beneath the coating. The coating itself is black but may painted over. The systems operate at medium current densities with an expected life of 5 to 15 years. An advantage of this type of system is the range of different surface orientations and shapes that it can be applied to.

Expanded titanium mesh may be applied as both primary and secondary anode. This is overlaid often with sprayed concrete, paving concrete or cast superplasticised mortar. Depending on the grade of mesh used, the range of current densities may be very large and design lives of 25 to 100 years or more have been claimed.

The major factor in deciding anode life is the presence of water due to leakage or permeation through the concrete. Attention to detail regarding waterproofing of joints and treated concretes can extend anode life considerably.

Sacrificial flame or arc sprayed zinc has also been used successfully used in marine locations. These systems are relatively cheap, require little maintenance and have achieved lifetimes of 8 years or more. An advantage can be that minimal repairs need to be carried out to the deteriorated concrete and, in fact, zinc sprayed over exposed steel is a convenient way of achieving the required electrical contact.

Electrochemical Chloride Extraction (Desalination) and Realkalisation.

Introduction

This section is concerned with two recently developed processes for dealing with corrosion of steel in concrete. Above we have looked at the mechanisms of corrosion of steel in concrete, and other methods of dealing with it. Although chloride removal (CR, also known as chloride extraction and desalination) and realkalisation (ReA) are relatively new techniques, their use will undoubtedly expand with time.

We will first see how CR and ReA stop corrosion. We will then review the advantages and disadvantages of these processes versus other ways of stopping corrosion such as concrete removal, sealers and cathodic protection. We will also look at some case histories of the processes.

The two processes are closely linked, but to avoid confusion we will deal with the generalities of corrosion, and electrochemical techniques and then will deal with CR and ReA separately.

The Mechanisms of Corrosion of Steel in Concrete

We already know that steel is normally passive when embedded in concrete, due to the alkalinity of the pore water. There are two ways in which this passivity can he destroyed.

        Carbonation

       Chloride attack

Electrochemical Processes

Chloride Removal

We have already determined that the chloride ion is a catalyst to corrosion. As it is negatively charged we can use the electrochemical process to repel the chloride ion from the steel surface and move it towards an external anode.

Anode types

The most popular anode is the same coated titanium mesh used for CP. Instead of embedding it permanently in a cementitious overlay, a temporary anode system is used. This is placed inside a cassette shutter, as in the diagram below. Where the shape of the member is especially difficult, a sprayed papier maché system can be applied over the anode, soaked with the electrolyte.

A "sacrificial" anode can be used instead of coated titanium. Copper was tried in the early 1970s trials, but copper (or its salts) may accelerate corrosion if its gets into concrete. Steel mesh has been used more recently but has fallen out of favour as it may be completely consumed in some areas before the treatment process is over.

Electrolytes

Calcium hydroxide is the usual electrolyte. This has the advantage over water of being alkaline and stops chlorine gas evolution, although additional chemicals may be added to increase alkalinity. Lithium salts have also been used to stop ASR (see below).

End Point Determination

End point determination can be by several means:

Point of diminishing returns - resistance goes up, amount of chloride removed goes down, when the current is small and the amount of chloride removed is small, switch off. Switching off for about a week will bring the system resistance down. But how much more chloride is removed by allowing "rest" periods is not known.

Direct measurement - take samples from the concrete and when an agreed level is reached, stop. This assumes that good sampling is possible and that samples are representative.

Indirect measurement - sample the anode system and or electrolyte. When chloride level is either at a plateau or an agreed level, stop. This assumes good sampling.

Experience of charge density needed - measure charge passed (amp hours per square metre) , and when an agreed limit is reached, switch off.

In practice a combination of systems is used. The experienced contractor will know the charge density needed and a trial may give a definite value for a given structure (or element within the structure). Sampling directly and indirectly will show that the system is responding in the expected manner. The point of diminishing returns should be reached soon after the other thresholds.

It is of course impossible to remove all the chlorides from the concrete. The area immediately around the rebar is almost chloride free, but further away there is less effect. This is particularly true behind the steel. Chloride removal will deplete the amount of chloride immediately in contact with the steel, and will replenish the passive layer. Field data shows that this is effective for at least six years, but for how much longer is uncertain. The results suggest about 10 years, but only real experience will show that.

One implication of the amount of chloride which can be removed is that if large amounts of chloride have penetrated beyond the steel, or were cast uniformly into the concrete, then chloride removal will only affect the chloride level in the "covercrete". The chlorides in the bulk of the concrete will then diffuse back around the steel and the corrosion may eventually be re-established. When the process is carried out, the steel is polarised for some time, perhaps a year, making half cell potential testing difficult to interpret. The charge on the steel means, however, that chloride ions will be repelled by the charge during this time. Thereafter, there will still be a reservoir of negatively charged hydroxyl ions around the steel, and work is currently being undertaken to define how long this additional protection remains in place.

Possible Effects

Passing large amounts of electricity through concrete can have effects upon its chemistry and therefore its physical condition. Brown staining around the rebar has been observed on specimens when high currents and voltages are used (in excess of about 2 or 3 A.m²) This is an effect on the concrete, not the steel. Current levels are therefore carefully controlled using an expert management system, with data logging equipment built into the power supply.

There are two known side effects of CR. The first is the acceleration of alkali silica reactivity (ASR), and the other is reduction in bond at the steel concrete interface.

Alkali Silica Reactivity

Research at Aston University here in England and by Eltech Research in the USA under the SHRP program shows that there can be excess ASR induced by the cathodic reaction (C) that generates excess alkali at the steel surface. This is exacerbated by the movement of alkali metal ions (Na⁺ and K⁺) to the steel surface under the influence of its negative charge. Some researchers in Japan have suggested that the pH can be so high that the silica gel dissolves, stopping the expansive process.

Eltech are presently undertaking a field trial to see if that can be controlled by the application of lithium ions in the electrolyte. Lithium is known to reduce or stop ASR, and proved effective in lab tests. If a corroding structure is made with aggregates susceptible to ASR, a detailed investigation of its likely reaction to CR will be required.

Bond Strength

The effect of current on bond strength of steel in concrete has been a subject for discussion in the literature for many years, usually with reference to cathodic protection. In most practical applications, the major part of the bond is supplied by the ribbing on the bars, so the details of the performance of the steel/concrete interface is irrelevant.

A feasibility study is presently underway for the application of CR to a long bridge structure in the North of England. In this case, extensive use was made of smooth rebars, due to steel shortages at the time of construction. Some tests have shown that CR reduces bond strength by as much as 50%. However, careful review of the laboratory data shows that a large amount of charge was passed to achieve this drop. About five times as much as is used normally in a CR treatment. A very high current density was also used.

Work has been carried out by Imperial college that showed reduction in bond could be a problem with smooth bars, although the effects were small and in no case did the bond drop below that of the control, uncorroded, sample. It was considered likely that the lowering of bond was due to passivation of the corrosion deposits. In any case, spalled concrete, which is the other alternative is not noted for its bond to steel reinforcement! Ribbed bars showed no reduction in bond strength.

Why choose Chloride Removal?

One of the major issues facing any consultant or owner of a structure suffering from chloride induced corrosion is what form of repair to undertake. There are coatings and sealants, specialised patch repair materials, options for total or partial replacement, cathodic protection and now chloride removal.

Since the pores in concrete contain significant amounts of water and air, sealing it to stop corrosion is unlikely to be effective once chlorides have penetrated the concrete. This comparatively cheap solution is not effective once corrosion has started.

Patch repairs are only effective if chloride ingress is local and the chlorides can all be removed. Just patching up damaged areas is a very short term palliative (owing to the incipient anode effect), not a long term rehabilitation. Contaminated concrete must be removed from all around the steel, and support may be needed during the repair process.

Cathodic protection has been described as the only solution to chloride induced corrosion that can stop (or effectively stop) the corrosion process. It has been applied to concrete bridges in the USA since 1973, and in this country to buildings and other structures since 1988. The problems with cathodic protection are its requirement for a permanent power supply, regular monitoring and maintenance.

Chloride removal has the advantage that, like a patch repair, it is a one off treatment. A generator can be brought in for the duration of the treatment, so mains power is not needed. There is no long term maintenance need, but the system does treat the whole structure.

The disadvantage is its unknown duration of effectiveness. We cannot remove all the chlorides from the concrete. If we can stop further chloride ingress then the system may be effective for many years (10 to 20). If chlorides are still impinging on the structure then it will be shorter (or may require a coating or sealant).

Chloride removal cannot be applied to pre-stressed structures due to the risk of hydrogen embrittlement. Work is progressing on its application to structures suffering from ASR, but this is in the early development stage at the moment. As stated earlier there must be electrical continuity within the reinforcement network for either of these techniques to be applied.

Conclusions about Chloride Removal

The origins of chloride removal lay in trials carried out in the 1970's. It is now being vigorously pursued by the developers of the Norwegian system, based on the spayed cellulose anode and a major US research program. Ontario Ministry of transportation now has three trial systems (two of the Norwegian system, one of the US SHRP researchers) applied. There is a trial and ongoing monitoring underway in the UK. Results are all encouraging.

We do not know how long the treatment process will last, but a range of five to twenty years is likely, depending upon conditions.

Realkalisation

In equations 1 and 2 we saw how carbonic acid reacts with calcium hydroxide to form calcium carbonate. This removes the hydroxyl ions from solution, and the pH drops, so that the passive layer is no longer maintained and corrosion can be initiated.

The cathodic reaction (C) showed that by applying electrons to the steel, we can generate new hydroxyl ions at the steel surface, regenerating the alkalinity, and pushing the pH back.

The realkalisation process has been patented, and uses the same cassette shutter or sprayed cellulose system developed to apply chloride removal. In addition to generating hydroxyl ions, the developers claim that by using a sodium carbonate electrolyte they make the treatment more resistant to further carbonation.

The patent claims that sodium carbonate will move into the concrete under electro-osmotic pressure. A certain amount will then react with further incoming carbon dioxide. The equilibrium is at 12.2% of l M sodium carbonate under atmospheric conditions.

Na₂CO₃ + CO₂ + H₂0 ® 2NaHCO₃

In laboratory tests they have shown that it is very difficult, if not impossible for a treated specimen to carbonate again. Over 80 realkalisation treatments (50,000m²) have been undertaken on structures around Europe over the past few years. The treatment is faster than chloride removal, only requiring a few days of treatment. Work by Banfill at Heriot Watt University showed that realkalisation actually improved the properties of concrete, reducing porosity and permeability, increasing strength and modulus and not apparently causing any difficulties.

Anode types

Anode types are the same as for chloride removal. Cassette shutters or sprayed cellulose are used by the owners of the patented system, with a steel or coated titanium mesh. The steel is more likely to be used here as the treatment time is shorter and the steel is less likely to be completely consumed.

Electrolytes

As stated above, sodium carbonate solution is the preferred electrolyte to give long lasting protection against further CO₂ ingress. However, introducing sodium ions can accelerate ASR so in some cases just water are used. Also, some problems have been found with adhesion of coatings when sodium carbonate is used, so the tendency is now to use a very low dose of sodium carbonate, or just to use water.

Operating Conditions

In one case a current density of 0.3-0.5A per m² was applied (at 12V) to 2000m² of a building in Norway, with a treatment time of 3-5 days. In another case 10-22V was applied to give a current density of 0.4-l.5A/m² in 12 days on 300m² of a bridge control tower in Belgium. A further section of 140m² was treated in 9 days with a current of l-2A/m². All figures are for concrete surface area. The steel to concrete surface ratio was not given.

End Point Determination

This is easy for carbonation. A simple measurement of carbonation depth will show when it has been reduced to zero. Measurement of current flow also gives guidance and a rapid on site method for measuring sodium content of dust samples has been developed, where sodium carbonate is used.

Possible Effects

As there is a smaller charge density applied, the risks of damage are lower than for chloride removal. As mentioned above, ASR is a risk if sodium carbonate is used as the electrolyte.

Sodium carbonate can also cause short term efflorescence, and the high alkalinity after treatment can attack some coatings. Sodium carbonate will attack oil based paints, varnishes and natural wood finishes.

Case Histories

The patent owners claim that over 80 structures have been treated with realkalisation around Europe. Two cases are summarised above. A recent application was completed on 1500m² roof area of Walthamstow Magistrates Court. Carbonation depths ranged from 5-25 mm. After treatment a polymer modified mortar was applied to the surface and an elastomeric decorative finish applied. The advantage to the Client in this case was the lack of noise, so the Court could remain in session.

Why choose Realkalisation?

Realkalisation is a simpler treatment than chloride removal. However the alternative of patching and coating with an anti-carbonation coating is much more effective than patching and coating for chloride attack. The extent to which carbonation has reached the rebar, and the requirements for patch repairing to restore alkalinity will determine whether realkalisation is preferred either because it is more economic, or in other cases to avoid the noise, dust and vibration required for extensive patch repairing.

The technique is becoming popular in Europe and the Middle East (in North America very little attention is paid to carbonation). If the treatment is as effective as claimed, then the choice between realkalisation and patching and coating is a question of convenience and cost, together with a realistic appraisal of the effectiveness of anti-carbonation coatings.

The patentees and the licensees of the system claim that:

	It is financially competitive with the alternatives.
	There is greatly reduced vibration and noise.
	All of the surface is treated.
	Guarantees are offered.

Conclusions

There is no doubt that both these techniques will become established in the concrete repair field. They have some unique advantages which will make them the system of choice for some applications. For other applications they will need to compete financially with other systems.

Corrosion Inhibitors

These materials have seen a dramatic increase in promotion via manufacturers such as Sika, Balvac, Grace and others. They vary in mode of use. There are two main types:-

Penetrating liquid or vapour phase corrosion inhibitors (Sika Ferroguard, Balvac MFP)

Concrete additives such as calcium nitrite, designed to be added to the concrete at the time of placing, perhaps in conjunction with ggbs, pfa or silica fume.

The latter have a good track record of successful use in the United States, employ sound technology, and have been shown by trials to be effective at dealing with chloride induced corrosion for external sources of chloride, such as de-icing salt.

The former are materials such as amino-alcohols (Ferroguard), or sodium monofluorophosphate (MFP). These materials are claimed to deal with both carbonation induced corrosion, and relatively high levels of chloride. The technology on which they are based has been successfully used in the metals industry for many years. They work by inhibiting both the anodic and the cathodic corrosion reactions, by penetrating in the liquid or vapour phase down to the steel reinforcement and surrounding the steel.

Research by various workers has been carried out and it is evident that in some low risk situations, these materials can have a role in concrete repair. In our opinion, based upon current research, their effectiveness is limited to

	Concrete suffering from carbonation with shallow cover, say up to 15mm. Their ability to work with deeper cover depends on their ability to penetrate adequately to the reinforcement. This varies from concrete to concrete and cannot be guaranteed.
	Concrete containing up to 0.6% chloride as chloride ion, by mass of cement. Research has not supported their effectiveness at higher values.

Notwithstanding this, repair contracts are being carried out using a combination of patch repairs, corrosion inhibitors and coatings. In some cases an insurance backed warranty is being offered, for typically 10 years. Given the combined effectiveness of inhibitors and coatings together, with existing damage dealt with by patch repair, then if the warranty is considered acceptable, we see no reason why the Client should not accept this approach, provided caution is exercised. We would not, however, consider this approach to be likely to be successful in high chloride situations. At chloride levels above 1%, specific reassurances should be sought, in our opinion.

Acknowledgements

The Author would like to acknowledge the help of Dr John Broomfield in preparing these notes.

Data from Dr. David Manning of Ontario Ministry of Transportation is also gratefully acknowledged, as well as background data supplied through John Broomfield's consultancy work with the Strategic Highway Research Program.

These acknowledgements do not imply endorsement by any of the persons or organisations mentioned.

Recommended Reading on Chloride Removal and Realkalisation

Anderson, Gordon. Chloride extraction and realkalisation of concrete. Hong Kong Contractor pp 19-25, July-August (1992)

Bennett, J. E.; Schue, T . J. Electrochemical Chloride Removal from Concrete: A SHRP Contract Status Report. Corrosion 90. April 23-37: Paper 316 (1990).

Broomfield, John P., SHRP Structures Research. Institute of Civil Enqineers ICE/SHRP, Sharing the Benefits; 29-31 October 1990; Tara Hotel, Kensington, London. London: Imprint, Hitchin, Herts; 1990; ICE 1990: p 35-46.

Electrochemical Chloride Removal and Protection of Concrete Bridge Components: Laboratory Studies. Strategic Highway Research Program Report SHRP-S-657, National Research Council, Washington DC 1993.

Jayprakash, G. P.; Bukovatz, J. E.; Ramamurti, K. and Gilliland, W.J. Electro-osmotic techniques for removal of chloride from concrete and for emplacement of concrete sealants. Kansas DoT Final Report; August 1982; FHWA-KS-82-2. (1982)

Lankard, 0. K.; Slater, J. E.; Diegle, K. B.; Martin, C. J.; Boyd, W. K.; and Snyder, M. J. Development of Electrochemical Techniques for Removal of Chlorides from Concrete Bridge Decks. Battelle; June 1978; Revised Final Report. (1978)

Lankard, D. R.; Slater, J. E.; Diegle, R. B.; and Boyd, W. K. The Electrochemical Removal of Chlorides From Concrete: Guide to Materials and Methodology. Battelle; June (1978).

Lankard, 0. K.; Slater, J. E.; Redden, W. A.; and Neisz, D. E. Neutralisation of Chlorides in Concrete FHWA-RD 76-60 September (1975).

Miller, J. B. Chloride Removal and Corrosion Protection of Reinforced Concrete. Strategic Highway Research Program and Traffic Safety on two Continents, Gothenburg Sweden.; September 29, (1989).