Epoxy coated reinforcement
Epoxy coated rebar is used in the USA to provide additional protection in chloride aggressive environments. It is usually prepared by a powder coating method. The cleaned bars are pre-heated and then sprayed with a fine powder mixture of epoxy pre-polymer (a diglycidyl ether of bisphenol A) and a curing agent (amine or acid anhydride) and passed through a heated chamber where the curing reaction takes place to produce a tough polymer that is largely impermeable to chloride ions. The coating should be sufficiently thick to act as a barrier to the ingress of chloride ions but not so thick as to inhibit the curing process.
A prime requirement of the coating is that it is sufficiently flexible to withstand the bending of the bar without cracking and that bare metal such as cut ends is adequately coated. ASTM A775/A775M-07b (2014) covers the manufacture and performance of the bars. This standard requires that the thickness after curing should be 175-300 pm and should be sufficiently flexible to be bent around a mandrel (the diameter of which is specified for each class of bar) without cracking.
An intact epoxy coating delays the initiation of chloride corrosion but the major problem with this system is that any defects of the coating process or, more likely, damage caused by handling, or installation can shorten the life considerably and lead to serious failures. The Appendix to ASTM А775/ A775M-07b (2014) describes in some detail the handling and storage procedures that should be adopted when using this material.
Stainless steel reinforcement
Although stainless steel is often avoided by design engineers because of the perceived cost disadvantages, an actual total lifetime costing may indicate that in particularly corrosive environments or when the cost occasioned by the necessity of closing the structure in order to effect repairs is large, the use of stainless steel may be the economic solution to the design. Selective use of stainless steel in more-corrosive locations, critical parts, outer section of the structure (skin reinforcement) is also possible.
The commonly used stainless steels described in the UK Concrete Society Technical Report No 51 (1998) are:
- • Type 304 18.5 Cr 9.0 Ni 0.06 C
- • Type 316 17 Cr 12 Ni 2.25 Mo 0.06 C
- • Type 329 26 Cr 4.5 Ni 1.5 Mo 0.08 C
Type 304 is the most common austenitic stainless steel which is used in a wide variety of applications. Type 316 is the molybdenum modified version of type 304 and is more resistant to pitting attack by chloride ions. Type 329 is more commonly known as a ‘duplex’ stainless steel because it has a two phase (ferritic and austenitic) structure and is more resistant again to pitting attack by chloride ions.
In terms of types and grades of stainless steel rebars that are used in standards, Sussex (2017) makes two initial points about standards. When talking to materials engineers, grade usually means chemical composition. For design engineers, grade means a specific design strength which, mainly for reasons of continuity, has been aligned around the grade strengths of carbon steel bar. This possible confusion is generally avoided by referring to the different compositions as type or by providing the material designation.
Secondly Sussex (2017) makes the point that while usually the switch from Euronorm to ASTM/SAE designations is routine, a comparison of the composition limits shows that the BS 6744 (2016) is aiming for greater strength from a nominal material, e.g. it lists a high nitrogen (0.12-0.22%) version of 304 compared to the maximum of 0.10% in ASTM A955 (2014b) for 304 and the unspecified norm of about 0.05%. The UK standard also requires the 2.5-3% version of 316/316L whereas ASTM A995 (2014b) allows molybdenum to be as low as 2.00%. However, while both standards list alloys commonly used for reinforcement, they both allow other alloys subject to verification of corrosion resistance - and both provide normative block test methods which are protracted but realistic (Sussex, 2017).
The common use table in BS 6477 (2016) has six austenitic alloys ranging from ‘304LN’ to a 6% Mo super austenitic and four duplex alloys ranging from utility 2101 to super duplex that are available as stainless steel rebars for concrete. These duplex alloys all require tests to confirm the absence of deleterious phases. There are no ferritic grades nor FeCrMn austenitics available (Sussex, 2017).
The common use table in ASTM A955/A955M (2018) has an austenitic alloy (XM-28) with more than 11% manganese and low levels of nickel. Because manganese is less effective in maintaining the ductile austenitic structure, the chromium content is lower than for 304 but the corrosion resistance is slightly enhanced by high nitrogen levels. Their inclusion appears to be due to requests from State Transportation bodies in the United States who had commissioned test work and were concerned about material costs during the period of high nickel prices (Sussex, 2017). The list of materials in common use also shows austenitics 304, 316L and 316LN and duplex (2205 and 2304). There are no ferritic alloys listed although there has been corrosion test work on 12% chromium alloys (Sussex, 2017).
As a major stainless steel producer, India (Sussex, 2017) has a stainless steel reinforcement standard which lists austenitics (304, 304 LN, and 316Mo2.5-3.0%) plus duplexes LDX2101, 2304, and 2205 as well as a 11-13.5% chromium ferritic (410 L) (IS 16651, 2017). There are four strength grades based on 0.2% proof stress, i.e. 500 MPa, 550 MPa, 600 MPa, and 650 MPa (IS 16651, 2017).
Other, non-standard but authoritative documents include the fib Bulletin 49 (2009) which has a thorough review of stainless steel reinforcement including details of international standards at the date of publication and summaries of the data and discussion of 14 different investigations of chloride resistance.
In terms of the corrosion resistance of stainless steels in concrete, all types of stainless steel reinforcement are passive in carbonated concrete (Bertolini et al., 2013).
In chloride contaminated concrete, stainless steel reinforcement can suffer pitting corrosion like carbon steel reinforcement. For stainless steels, like for carbon steel, the susceptibility to pitting attack can be expressed in terms of both the pitting potential (i.e. for a given chloride content, the potential value above which pitting can initiate) or the critical chloride content (i.e.
for a given potential, the threshold level of chloride content for the onset of pitting). These two parameters are interrelated and depend on the chemical composition and microstructure of the steel, on the surface condition of the bars and on the properties of the concrete. Because of the higher stability of the passive film of stainless steel compared with carbon steel, their resistance to pitting corrosion is much higher. This is due to the formation of chromium and nickel oxyhydroxides in contact with the alkaline pore solution of concrete (Bertolini et al., 2013).
As a result, stainless steel bars have a much higher chloride threshold level compared to carbon steel bars. The chloride threshold for stainless steel reinforcement may be 5-10 times higher than that of conventional carbon steel reinforcement dependent on stainless steel type (Concrete Institute of Australia, 2015).
However, Bertolini et al. (2013) notes that even small variations in the chemical composition, thermomechanical treatment or the surface condition may significantly affect the corrosion resistance of stainless steel bars in chloride bearing concrete. Therefore, the chloride threshold should be measured for any specific type of stainless steel and its variability should also be evaluated. Unfortunately, the evaluation of chloride threshold for steel in concrete is rather difficult and there are no standardised or generally accepted methods for its evaluation (Bertolini et al., 2013), refer also to Section 4.7.6.
In order to give an indicative picture of the order of magnitude of the chloride content at which some stainless steel bar types can resist corrosion initiation, Figure 11.11 depicts fields of applicability in chloride contaminated concrete exposed to temperatures of 20°C or 40°C (Bertolini et al., 2013).
Bertolini et al. (2013) points out that these values are indicative only, since the critical chloride content depends on the potential of the steel, and thus it can vary when oxygen access to the reinforcement is restricted as well as when macrocells or stray currents are present. For instance, they indicate that the domains of applicability are enlarged when the free corrosion potential is reduced, such as in saturated concrete. Furthermore, the values of the critical chloride level for stainless steel rebars with surface finishing other than pickling can be lower.
Bertolini et al. (2013) further comment that since the pH for noncarbon- ated concrete is around 13, while in carbonated concrete it is near 9, the right-hand side of the graphs of Figure 11.11 is representative of alkaline concrete and the left-hand side of carbonated concrete. In alkaline concrete, austenitic 304L (1.4307) can safely be used in concrete up to 5% chloride by mass of cement (i.e. -0.7% chloride by mass of concrete), and 316L (1.4404) and 2205 (1.4462) duplex stainless steel even higher than 5%, that is, for chloride contents that are rarely ever reached in the vicinity of the steel surface.
Figure ll.ll Schematic representation of fields of applicability of different stainless steel bars (pickled) in chloride bearing environments for 20°C and 40°C. The threshold levels are indicative only. They can decrease if oxides produced at high temperature, for example, welding or during manufacturing, are not completely removed, or the potential due to anodic polarisation increases (e.g., due to stray current) or the concrete is heavily cracked. Conversely, they can increase when there is a lack of oxygen or cathodic polarisation. (Courtesy of Bertolini et al„ 2013, p.272)
In the presence of a welding scale (heat tint, heat affected zone/HAZ) on the surface of the reinforcement the critical chloride content is lowered perhaps to approximately 3.5% by mass of cement (i.e. ~0.5% chloride by mass of concrete for 304L) and the same reduction takes place if the surface is covered by the black scale formed at high temperature during thermomechanical treatments (Bertolini et al., 2013). Pickling of rebars is then more efficient in pushing up the critical chloride content than say sand blasting, which does completely free the surface from the oxide scale (Bertolini et al., 2013).
In carbonated concrete, or in the case where the concrete is extensively cracked, the critical chloride contents are remarkably lower. The more highly alloyed stainless steels should be preferred in these more aggressive conditions (Bertolini et al., 2013).
The critical chloride content also decreases as the temperature increases; for instance Figure 11.11 shows the expected variations between 20°C and 40°C. Thus, the more highly alloyed stainless steel bars should be preferred in hot climates (Bertolini et al., 2013).
Often, the use of stainless steel reinforcement is limited to the outer part (skin reinforcement) or to its most critical parts for economic reasons. Furthermore, when stainless steel bars are used in the rehabilitation of corroding structures, they are usually connected to the original carbon steel rebars (Bertolini et al., 2013).
In terms of examples of selective use of stainless steel reinforcement in structures, the concrete of (the 100 year design life) McGee Bridge in Tasmania, Australia (Figure 11.12) is mainly reinforced with carbon steel to save costs but in the tidal zone it uses 316 stainless steel for the near surface reinforcement. The submerged reinforcement in the piles is carbon steel because the saturated concrete has very low oxygen content. In the superstructure, low penetrability concrete controls chloride access (Sussex, 2017).
The Gateway Bridge (300 year design life) in Brisbane, Queensland, Australia generally has pickled LDX 2101 stainless steel reinforcement only near the surface where saltwater contact is possible but the deeper bars have more than sufficient cover to the carbon steel. The only exception is the ship fenders around the base of the piers (Figure 11.13) because they are in the intertidal zone with the possibility of surface concentration of chlorides - and impact damage from ships - so all their reinforcement is pickled LDX 2101 stainless steel (Sussex, 2017).
Figure 11.12 McGee Bridge (100-year design life) over tidal river, Tasmania, Australia. (Courtesy of Sussex, 2017)
Figure 11.13 Gateway Bridge (300-year design life), Brisbane, Queensland,Australia, stainless steel rebar for intertidal zone elements. (Courtesy of Sussex, 2017)
Connal and Berndt (2009) describe in more detail the durability approaches adopted for the 300 year design life Second Gateway Bridge in Brisbane, Queensland, Australia. The project scope and technical requirements (PSTR) for the bridge specified that durability assurance be applied diligently and continuously throughout the process of design, construction and throughout the maintenance period, and that the Second Gateway Bridge have a design life of 300 years, with some replaceable sub-items having design lives ranging from 20 years (wearing course) to 100 years (bearings). Design life was defined as the period assumed in design for which the structure or structural element is required to perform its intended purpose without replacement or major structural repairs. The philosophy adopted in meeting the extended design life was based on ‘building in’ the required durability at the outset, where feasible, and minimising the need to take measures later in the life of the bridge to achieve a 300 year service life. Integral with this philosophy was the appropriate selection of high quality materials chosen to address the particular durability issues that are posed by the range of exposure conditions.
The reinforced concrete pile caps of the bridge were in a tidal/splash exposure zone whereby the Brisbane River was determined to have a chloride concentration up to 18,000 ppm which is similar to that of seawater. The concrete mix proposed for the pile caps was a 50 MPa grade ternary blend consisting of 30% fly ash and 21% blast furnace slag. The total cementitious content was 560 kg/m! and the maximum water/cementitious material ratio was 0.32. The proposed pile cap design was to have 150 mm minimum cover to black steel reinforcement and 75 mm minimum cover to stainless steel reinforcement. A period of 280 years was then selected as the required time to corrosion initiation, following which corrosion, and spalling may take place over a subsequent 20 year period, resulting in a 300 year service life before major repairs (Connal & Berndt, 2009)
Predicted chloride ingress profiles (deterministic chloride diffusion modelling) indicated that 150 mm cover to carbon steel and 75 mm cover to LDX
2101 or 316LN stainless were likely to provide adequate protection and prevent initiation of corrosion within 280 years. The LDX 2101 had cost savings compared with 316LN and was therefore the favoured stainless steel (Connal & Berndt, 2009).
A probabilistic approach to modelling of chloride ingress was also adopted and this determined that the target reliability index ((3 = 1.28) is met over 300 years for 150 mm cover to black steel reinforcement steel with S50 ternary blend concrete even if the coefficient of variation for depth of cover is 20% (however, it was recommended that tight construction quality control be implemented to reduce the coefficient of variation). The reliability index calculations also predicted that the S50 ternary blend concrete with 75 mm cover to pickled LDX 2101 stainless steel would achieve the required life and that it was also noted that the predicted corrosion rate of stainless steel will be significantly lower than that for black steel and this further enhances design for durability (Connal &c Berndt, 2009).
Selective use of stainless steel reinforcement begs the question as to whether there is a risk of galvanic corrosion of carbon steel reinforcement induced by coupling with stainless steel reinforcement. Experimental studies clearly show that the use of stainless steel in conjunction with carbon steel does not increase the risk of corrosion of the carbon steel. When both carbon steel and stainless steel rebars are passive and embedded in aerated concrete, macrocell action does not produce appreciable effects, since the two types of steel have almost the same corrosion potential. Indeed, in this environment, carbon steel is even slightly nobler than stainless steel. In any case, both carbon steel and stainless steel remain passive even after connection (Bertolini et al., 2013).
Only when the carbon steel corrodes does the macrocell current become significant. However, stainless steel is a poor cathode. Figure 11.14 shows that the consequences of coupling corroding carbon steel reinforcement with stainless steel reinforcement are generally modest, and they are negligible with respect to those coupling with passive carbon steel that always surround the corroding area. Consequently, the increase in corrosion rate on carbon steel reinforcement embedded in chloride contaminated concrete due to galvanic coupling with stainless steel reinforcement is significantly lower than the increase brought about by coupling with passive carbon steel reinforcement (Bertolini et ah, 2013).
Bertolini et al (2103) confirm this behaviour by examination of the cathodic polarisation curves in saturated calcium hydroxide (Ca(OH)2) solution (pH 12.6) of 316L stainless steel compared with carbon steel, refer Figure 11.15. Stainless steel is seen to be a less efficient cathode and a higher overvoltage for the cathodic reaction of oxygen reduction is necessary for stainless steel with respect to carbon steel.
Bertolini et ah (2013) note, however, that stainless steel with welding oxide (heat tint, HAZ) or with the black scale formed at high temperature is a better cathode and increased galvanic corrosion can occur in the presence
Figure 11.14 Macrocell current density exchanged between a corroding bar of carbon steel in 3% chloride contaminated concrete and a (parallel) passive bar of: carbon steel in chloride-free concrete, 3I6L stainless steel in chloride-free concrete or in 3% chloride contaminated concrete (20°C, 95% RH). Also results of stainless steel bars with oxide scale produced at 700°C (simulating welding scale) are reported. (Courtesy of Bertolini et al., 2013, p.274)
of these types of scale in stainless steel bars. However, in evaluating the effect of galvanic coupling, at least in the case of welds, it has to be considered that the area covered by the scale will normally be small compared to the total rebar area. The presence of these types of scale increases the macrocell current density generated by stainless steels, to the same order of magnitude or even greater than that produced by coupling with carbon steel (Figure 11.14), as a consequence of a change in the cathodic behaviour (Figure 11.15).
Sussex (2017) notes that cutting, bending, welding, cleaning, and packing of stainless steel reinforcement should ideally be carried out off site in controlled conditions. The stainless steel must be clean and dry before wrapping and transport requires non-ferrous materials including the truck tray and tie down cables. If the stainless steel is not to be used immediately, it should be stored in dry conditions, preferably out of the sun and protected from grime and damage until needed.
He goes on further to advise that there is a regrettable tendency for construction sites to consider stainless steel as invulnerable and, unless there is strong supervision, allow it to get dirty or greasy, walked on, sprayed with carbon steel grindings (Figure 11.16), smeared with carbon steel by using the same tools as for galvanised steel, splashed with acids used to remove excess mortar or concrete, overheated during cutting or grinding ... and the list goes on. Stainless steel is an expensive commodity with a carefully prepared protective surface which is very robust but can be degraded by ignorance whether wilful or through lack of training (Sussex, 2017).
Figure 11.15 Cathodic polarisation curves in saturated Ca(OH)2 solution (pH 12.6) of 3I6L (1.4404) stainless steel compared with curve of carbon steel. Results of stainless steel with simulated welding scale are also shown. (Courtesy of Bertolini et al„ 2013, p.275)
Figure 11.16 Carbon steel grindings on site will degrade corrosion resistance. (Courtesy of Sussex, 20 I 7)
The care and maintenance of stainless steel on site starts with incoming inspection and Figure 11.17 shows the requirements of the Ontario Ministry of Transportation, for example, to verify that the bar has been correctly pickled (Sussex, 2017). They regard A and В as acceptable but C has iron tint probably from rinse water and D appears to have been overpickled (Sussex, 2017).
Overpickling roughens the surface and has been associated with low chloride threshold (critical chloride level) results in some early lean duplex measurements (Sussex, 2017).
Figure 11.17 Site inspection of pickled rebar with iron tint (C) and overpickling (D). (Courtesy of Sussex, 201 7)
Sussex (2017) advises any on sire welding or grinding will have degraded the surface and to obtain the expected corrosion performance, these areas must be repickled. This will remove:
- • Potential carbon steel contamination.
- • Iron rich unprotective heat tint.
- • Chromium depleted layer underneath the heat tint.
- • In areas that have simply been ground, dissolve any exposed manganese sulphide inclusions that are the inevitable part of all steels including stainless steels (Standard bar stock tends to have higher sulphur content than sheet or plate although still well within standard material specifications).
Each of these four features can and will act as corrosion initiating points if exposed to the chlorides somewhat below the chloride threshold level, i.e. the stainless steel will locally act as if it is a lesser alloy (Sussex, 2017).
Metallic clad reinforcement
Stainless steel clad rebar (SCR) with a carbon steel core is commercially available. SCR types include 316L stainless steel, 316LN stainless steel, and 2205, and 2101 duplex grades. Investigation of the corrosion resistance of SCR dates back to at least the late 1990s/early 2000s (e.g. Hurley et al., 2001).
Copper-clad reinforcing was discussed as a corrosion resistant reinforcing bar as early as 1979. To date, it is understood that no significant structures have been constructed using these materials; however, several long term laboratory tests have been conducted showing that copper-clad bars may be a viable option for corrosion protection (McDonald et al., 1996).
Hot dipped aluminium as a coating for reinforcing bars has been investigated (Saremi et al., 2003) but the in-field success and performance is not known.
Non-metallic reinforcing bars, such as fibre-reinforced polymer (FRP) bars, which are manufactured from composite materials consisting of fibres embedded in a polymer matrix (resin) are commercially available. Glass (GFRP), aramid (AFRP) and carbon (CFRP) are the three most commonly used fibres (Cement & Concrete Association of New Zealand, 2009).
The advantages and disadvantages of FRP reinforcement are listed in American Concrete Institute (ACI) 440.1R-06 (2006) which is reproduced as Table 11.1 below.
FRP bars have been used in concrete close to sensitive electronic equipment (such as hospital MRI rooms), in sea walls, foundations, chemical
Table I l.l Advantages and disadvantages of FRP reinforcement
Advantages of FRP reinforcement
Disadvantages of FRP reinforcement
High longitudinal tensile strength (varies with sign and direction of loading relative to fibres)
No yielding before brittle rupture
Corrosion resistance (not dependent on a coating)
Low transverse strength (varies with sign and direction of loading relative to fibres)
Low modulus of elasticity (varies with type of reinforcing fibre)
High fatigue endurance (varies with type of reinforcing fibre)
Susceptibility of damage to polymeric resins and fibres under ultraviolet radiation exposure
Lightweight (about l/5'h to V* the density of steel)
Low durability of glass fibres in a moist environment
Low thermal and electric conductivity (for glass and aramid fibres)
Low durability of some glass and aramid fibres in an alkaline environment
High coefficient of thermal expansion perpendicular to the fibres, relative to concrete
May be susceptible to fire depending on matrix type and concrete cover thickness
Source: ACI (2006) plants, reinforced shotcrete lined tunnels and masonry walls in corrosive environments (Cement & Concrete Association of New Zealand, 2009).
Although the use of FRP reinforcing bars has been steady since the mid- 1980s, their wider adoption has been limited, most probably due to the absence of standard design codes. There have also been some concerns about the lack of ductility and fire resistance of the bars (Cement & Concrete Association of New Zealand, 2009).
However, these issues have been addressed by ACI Committee 440 (2006), whose mission it is too develop and report information on FRP reinforcement of concrete. As a result, FRP bars could become an increasingly common method of enhancing durability of concrete structures (Cement & Concrete Association of New Zealand, 2009).