Chloride-induced Corrosion Of Steel Research Articles

Chloride-induced corrosion of steel in cracked concrete—Part II: Corrosion rate prediction models

Chloride-induced corrosion rate (icorr) prediction models for RC structures in the marine tidal zone that incorporate the influence of crack width (wcr), cover (c) and concrete quality are proposed. Parallel corrosion experiments were carried out for 2¼years by exposing one half of 210 beam specimens (120×130×375mm long) to accelerated laboratory corrosion (cyclic wetting and drying) while the other half underwent natural corrosion in the tidal zone. Experimental variables were wcr (0, incipient crack, 0.4, 0.7mm), c (20, 40mm), binder type (PC, PC/GGBS, PC/FA) and w/b ratio (0.40, 0.55). The two proposed models (one each for accelerated and natural icorr) can aid not only in quantifying the propagation phase, but also provide a novel way to select c, wcr and concrete quality.

Experimental and numerical determination of the chloride penetration in cracked mortar specimens

The presence of cracks in reinforced concrete structures added to the chloride-induced steel corrosion is worldwide one of the major causes of deterioration of reinforced concrete structures, producing loss of structural strength and stability. This study deals with the influence of artificial cracks on chloride penetration in mortar specimens by presenting both the experimental and the numerical results. The experimental results were obtained using chloride penetration tests on mortar specimens of different sizes, with and without artificial cracks, subjected to chloride exposure. Numerical results were obtained using transient finite element analysis by simulating chloride mass diffusion. Comparisons between the results on chloride penetration concluded that the numerical ones obtained using the mass transfer modulus agrees fairly well with the experimental ones. In addition, the influence of certain crack parameters such as crack length, crack depth and also time exposure on chloride ingress was considered.

Spatial distribution of crystalline corrosion products formed during corrosion of stainless steel in concrete

The mineralogy and spatial distribution of nano-crystalline corrosion products that form in the steel/concrete interface were characterized using synchrotron X-ray micro-diffraction (μ-XRD). Two types of low-nickel high-chromium reinforcing steels embedded into mortar and exposed to NaCl solution were investigated. Corrosion in the samples was confirmed by electrochemical impedance spectroscopy (EIS). μ-XRD revealed that goethite (α-FeOOH) and akaganeite (β-FeOOH) are the main iron oxide–hydroxides formed during the chloride-induced corrosion of stainless steel in concrete. Goethite is formed closer to the surface of the steel due to the presence of chromium in the steel, while akaganeite is formed further away from the surface due to the presence of chloride ions. Detailed microstructural analysis is shown and discussed on one sample of each type of steel.

Influence of Cracks on Chloride Penetration in Mortar Specimens Subjected to Cyclic Treatment

Chloride-induced steel corrosion is worldwide one of the major causes of deterioration of reinforced concrete structures, producing loss of structural strength and stability. One of the most aggressive exposure conditions for concrete is marine environment together with the drying and wetting conditions by tidal and splash action. In such an aggressive environment the presence of cracks has a great influence on chloride ingress.In this paper both the experimental and the numerical results of the influence of artificial cracks on chloride penetration in mortar specimens are presented. The experimental results were obtained using chloride penetration tests on mortar specimens of different sizes, with and without artificial cracks, subjected to a cyclic treatment. Numerical results were obtained using transient finite element analysis by simulating chloride mass diffusionwith COSMOS/FFE realized with the heat transfer modulus T, which was replaced by the total concentration C, and the thermal conductivity Kx, replaced by diffusion coefficient D. Comparisons between the results on chloride penetration concluded that the numerical ones obtained using the mass transfer modulus agree fairly well with the experimental ones. In addition the influence of crack length on chloride ingress was presented.

セメント硬化体中の塩化物イオンの拡散におけるアルカリイオン濃度とC-A-S-Hの影響

塩化物イオン拡散メカニズムの解明はセメント内塩分浸透抑制を行う上で重要である。しかしながら、塩化物イオンの拡散過程において細孔中のアルカリイオンによって形成したC-A-S-Hが与える影響について定量的な結論は得られていない。そこでC-S-HおよびC-A-S-Hの2つの吸着サイトを考慮した塩化物イオン拡散モデルを構築し、塩化物イオンの拡散実験との検証を行った。しかし、アルカリイオン濃度が増加した試料での拡散予測結果では実験結果と乖離を生じた。そこで、C-A-S-Hの構造変化が原因であると推定し、C-A-S-Hの増加に伴う構造の変化を考慮した屈曲度をモデルに組み込み、モデルの構築を行った。

Use of a linear continuous reference electrode to monitor the chloride-induced corrosion of steel in prestressed concrete

Failures of pre-stressed steel caused by hydrogen embrittlement as a consequence of chloride-induced corrosion are well known in the civil engineering field. In 1995, a wire-type reference electrode was proposed to monitor pre-stressed tendons corrosion potential, claiming that, once a pit occurs in any position on the steel, the potential measured by the linear reference electrode – regardless of its length – drops to values typical of active range, giving the evidence of pitting occurrence. From the practical point of view, the use of such an electrode is very appealing because of the simplicity, nevertheless, concerns on the true meaning of the potential reading arose both from theoretical point of view and because of lack of similar use in electrochemistry. Laboratory tests were performed on some linear reference electrodes simulating chloride-induced corrosion in concrete, in order to understand the meaning of the potential reading and the ability to detect the initiation of localised corrosion.

Application of cathodic protection for controlling macrocell corrosion in chloride contaminated RC structures

This paper examines the effect of macrocell corrosion on the current distribution in cathodic protection. The electrochemical corrosion process involved in typical chloride-induced steel corrosion is numerically simulated by employing the finite element method. The modeling results show that the protection current tends to flow into the more positive cathodes (passive zone) and reduces the galvanic macrocell current between the active and passive steel. Macrocell corrosion may also increase the non-uniformity of cathodic protection around the steel–concrete interface. Patch repairs containing zinc sacrificial anodes have been found to be rather effective in inhibiting the ring-anode macrocell corrosion effect around the substrate–patch interface.

Effect of electrochemical treatment at fresh condition in enhancing the corrosion resistance in concrete

The present study concerns the electrochemical treatment to increase the resistance against chloride-induced corrosion of steel in concrete, which was ensured by a microscopic examination and corrosion tests. The current density (DC) applied to the embedded steel immediately after casting of concrete ranged 125, 250 and 500 mA/m2 for two weeks. As a result, an increase in the current density resulted in an increase in the precipitated calcium hydroxide at the steel-concrete interface within 100 μm from the steel surface, which would enhance the buffering capacity to the corrosion reaction. However, an application of the electrochemical treatment did not always impose the benefit in raising the corrosion resistance: the surface chloride content, when the concrete specimens containing the steel embedment was exposed to a salt solution, was increased, presumably because of modification of the chemistry at the right surface of concrete. Notwithstanding, the critical chloride ranged 0.88, 1.21 and 1.76% by weight of cement for the current density of 125, 250 and 500 mA/m2 respectively, while untreated specimens indicated only 0.62%. Consequently, the electrochemical treatment increased the corrosion-free life by enhancing the resistance to corrosion, resulting from the formation of precipitated calcium hydroxide at the steel-concrete interface.

Modeling of chloride-induced corrosion in reinforced concrete structures

A numerical procedure is presented in this paper for the prediction of chloride induced steel corrosion in reinforced concrete structures. Finite element analysis is introduced for the mechanical analysis of crack initiation and propagation due to the accumulation of corrosion products around the reinforcement, while the alternating direction implicit method is used to solve the transport equations of temperature, humidity, chloride ions and oxygen in concrete. Based on the assumption of a uniform distribution of corrosion products, a self-adaptation process for the variation of boundary conditions is proposed through a series of diffusion analyses together with crack propagation in concrete. Therefore, the interaction between the corrosion rate and the propagation of cracks in concrete is taken into account. Furthermore, a numerical program is developed and a case study involving bridge deck exposed to a marine environment in Hong Kong is investigated. The results show that interactive behavior has a significant effect on the corrosion rate of the reinforcement, and the non-cracking model significantly overestimates the service life of structures.

Semi-quantitative Prediction of the Corrosion Risk of Steel in Concrete Using XRD Analysis

The present study concerns a technique for determining the chloride binding capacity, using the XRD curve containing the peaks for Friedel' s salt. The influence of chloride binding on the corrosion behaviour was also evaluated. The binding capacity of chloride ions in cement paste was determined by the water extraction method. As a result, it was found that the chloride binding capacity is strongly dependent on the W/C, binder type, curing age and total chloride concentration in the paste. It is notable that chloride binding has a crucial influence on corrosion propagation: an increase in the binding capacity resulted in a decrease in the corrosion rate at a given chloride concentration in mortar specimens. However, the impact of chloride binding on the onset of corrosion was marginal, presumably due to a release of bound chlorides into free chlorides at the stage of corrosion, accompanying a pH fall in the vicinity of the steel. A strong relation between the peak intensity for Friedel' s salt and the concentration of bound chloride was observed, which can be used to determine the bound chlorides and thus the risk of chloride-induced corrosion of steel in concrete.

Chloride-Induced Corrosion of Steel in Cracked OPC and PPC Concretes: Experimental Study

The corrosion behavior of steel in cracked ordinary portland cement (OPC) and portland pozzolana cement (PPC) concretes when exposed to a chloride-rich environment is presented here. Three water-to-cement (w/c) ratios, and precrack widths of 0.2 and 0.4 mm were studied. A simple U-shaped specimen with precracks under flexural load was used in the study; the specimen has a deformed steel bar of 12 mm diameter placed at the bottom with 20 mm clear cover. Measurements made include chloride ion penetrability, sorptivity, half-cell potential, resistivity, total charge passed and gravimetric weight loss. The chloride ion penetrability of PPC was found to be nearly three times lower than that of OPC. The reduction in the gravimetric weight loss of rebars in PPC with respect to those in OPC is in the range of about 1.5 to 2.0 in the case of precracked concretes. The reduction in diameter (i.e., caused by pitting) is higher for OPCs. Studies showed that PPC is significantly superior to the corresponding uncracked ...

Condition Survey of Older West Virginia Bridge Decks Constructed with Epoxy-Coated Reinforcing Bars

Chloride-induced corrosion of steel reinforcing bars is a major cause of deterioration in concrete bridge decks in northern climates. Corrosion-resistant epoxy-coated reinforcing (ECR) bars were introduced in the 1970s. In 1993, the West Virginia Department of Transportation (DOT) performed surveys of bridge decks treated with ECR and uncoated bars built in the mid-1970s. While deterioration was observed on uncoated bar decks, deterioration of decks built with ECR was found to be limited. The current investigation consisted of a questionnaire and review of the condition reports of the decks examined by the West Virginia DOT in 1993, followed by field condition surveys of six decks built with ECR to determine how the ECR was performing. The six bridge decks inspected were in good to excellent condition, with the exception of two spans of Bridge 2930, which were reinforced with uncoated black bars. The decks with ECR exhibited less than 0.15% corrosion-induced deterioration. Any deterioration observed on ECR decks was concentrated at cracks or construction joints. All actively corroding ECR bars had an average coating thickness less than 7 mil (0.007 in.), the current minimum specified thickness. Decks reinforced with uncoated black bars included in the 1993 study had an initial service life of 18 to 21 years, and an overlay was applied to all of these decks to address corrosion-related damage. The decks constructed with ECR are now 33 to 35 years old, and none have required rehabilitation to address corrosion-related deterioration.

Probabilistic assessment to predict the time to corrosion of steel in reinforced concrete tunnel box exposed to sea water

This study concerned the service life prediction, in terms of chloride-induced steel corrosion, of a concrete tunnel-box structure placed on seabed. To calculate the time to steel corrosion, the rate of chloride transport in the identical concrete mix to the structure was tested, and the chloride threshold level for corrosion was assumed by a literature review from 65 published data. Then, the Monte Carlo Simulation was used to calculate the probability of steel corrosion and its corresponding service life, assuming that the corrosion initiates at 10% of the probability of corrosion. As a result, the service life depended on the time dependency of chloride transport. The service life was equated to 31, 51, 85, and 147 years, at 0.1, 0.2, 0.3 and 0.4 of the age factors, respectively, while the time independent model indicated only 27 years of the service life. Finally, methods to raise the service life were discussed.

Initiation of chloride-induced corrosion of steel in concrete: role of the interfacial zone

This paper provides a brief review of research aimed at characterising the steel–concrete interfacial zone (SCIZ) and its influence on the susceptibility of the metal to pitting corrosion when concrete is exposed to environments that cause ingress of chloride ions accompanied by leaching of hydroxyl ions. For reinforced concrete made from Portland cements, exposed to aqueous solutions of sodium chloride, the buffering effect of solid calcium hydroxide (portlandite) at pH ∼12.6 has been shown to restrain the gradual decline in the hydroxyl ion concentration of the concrete pore solution phase at depths corresponding to the embedded steel. When the concrete is produced under laboratory conditions that are carefully controlled to exclude macroscopic defects from the SCIZ and the steel is cleaned before being embedded, this can lead to observed chloride threshold levels being consistently greater than 1% chloride by mass of cement. The buffering action of cement hydration products formed in the SCIZ is believed to be partly responsible for this high tolerance to chloride-induced corrosion because it counters the generation of ‘anodic acidity’ that is a necessary condition for stable growth of pits to occur. Translating this behaviour of laboratory specimens to the performance of full-scale reinforced concrete structures has often proved difficult in the past and there is a need for further research in this area, particularly in relation to the role of non-traditional cements.

Potential pitfalls in assessing chloride-induced corrosion of steel in concrete

Electrochemical assessment of the condition of embedded reinforcing bar (rebar) in concrete is being carried out increasingly routinely, both in the laboratory and in the field. However, because of the perceived need to produce results very rapidly, the results may not, in fact, be representative of the actual behaviour of the rebar. This paper describes some of the pitfalls the authors have encountered in their own work and have, therefore, analysed and quantified, together with others which have appeared in the literature.

The influence of cracks on chloride-induced corrosion of steel in ordinary Portland cement and high performance concretes subjected to different loading conditions

Many reinforced concrete structures are subjected to variable loads, but little information is available on the effect of such loading, and the accompanying opening and closing of cracks, on the initiation and propagation of chloride-induced rebar corrosion. The purpose of this paper is to provide that information for reinforced ordinary Portland cement concrete and high performance concrete exposed to static or cyclic three-point bending and exposed to salt solution. It was observed that: (i) corrosion occurred only at intersections of the rebar with cracks in the concrete; (ii) high performance concrete was more protective than ordinary Portland cement concrete and (iii) the type of loading had less impact on corrosion than the type of concrete and exposure conditions.

The effect of chloride-induced steel corrosion on service life of reinforced concrete structures

Chloride-induced corrosion is one of the most important deterioration mechanisms in reinforced concrete structures. This study presents detailed theoretical models for predicting service life of reinforced concrete structures exposed to chloride attack. Three stages are considered in the corrosion process: the diffusion period, the rust accumulation period and the crack propagation period. For the diffusion period, the fully coupled moisture diffusion and chloride penetration in concrete are considered as the main driving forces of the deterioration process. Two-way coupled diffusion equations are used to characterize the length of this period. For the rust accumulation period, the chemo-mechanical coupling between the formation of the rust and the development of the interface pressure is considered as the driving force. For the crack propagation period, the interface pressure induced by the rust expansion and the fracture resistance of the cracked concrete determine the rate of crack propagation. The theoretical models developed in the present study are capable of simulating the entire process of the chloride-induced corrosion damage, and predicting the service life of reinforced concrete structures.

Chloride-induced corrosion of steel embedded in mortars containing fly ash and spent cracking catalyst

Stationary and non-stationary chloride diffusion coefficients of mortars prepared with ordinary Portland cement partially substituted by spent cracking catalyst (FC3R) and fly ash (FA) have been determined. An accelerated test of chloride migration through the mortar has been performed in order to determine these coefficients. The results show that the presence of FC3R improves chloride binding properties of mortars and the addition of FA reduces the total porosity, both factors leading to a higher resistance to chloride ingress. Corrosion rate of steels embedded in these ternary-binder mortars do not show significant differences compared to reference mortars (with plain cement), so a combined cement substitution as high as 45% by FC3R and FA do not lead to a higher risk of chloride-induced corrosion of steel when tested at an early age.

The importance of chloride content at the concrete surface in assessing the time to corrosion of steel in concrete structures

A literature review in the present study has shown that the chloride ions on the concrete surface exposed to a marine environment including tidal, splash and aerated zones can be accumulated then to increase with time. In order to take this effect into account in a model prediction of the rate of chloride ingresses in concrete, the surface chloride was expressed as a function of time such that (1) constant, (2) linear build-up, (3) square root build-up and (4) square root build-up with an initial set of the surface chloride (a refined model), at a given diffusion coefficient (2 × 10 −12 m 2/s). As a result, the constant surface chloride model produced the greatest level of chloride penetration, as an initial set of the surface chloride content was overestimated. The rate of chloride ingresses derived from the highest surface chloride led to the greatest risk of chloride-induced corrosion of steel in concrete, whereas the linear and square root build-up models indicated 2–3 times longer duration of corrosion-free service life. The refinement of a surface chloride build-up suggested a more realistic prediction of chloride profile and corrosion risk, thereby imposing the rate of chloride ingresses between the constant surface chloride, and linear or square root build-up.

The effect of calcium nitrite on the chloride-induced corrosion of steel in concrete

Laboratory tests were undertaken to assess the effect of calcium nitrite-based corrosion inhibitors in raising the chloride threshold level (CTL) for the corrosion of steel embedded in concrete and hence the time to corrosion initiation. Specimens with a centrally located steel rebar were cast with 0, 1·23, 2·5 and 5·0% nitrite by weight of cement and were cured for 4 weeks. Then they were immersed in 4M sodium chloride solution and the galvanic current between the embedded steel and an external cathode was monitored. The CTL of nitrite-free specimens was typically doubled and trebled by 2·5% nitrite and 5% nitrite, respectively. With and without nitrite, the CTL was reduced with an increase in the air void content at the steel–concrete interface. The time to corrosion depended on the cement content. Use of low cement content (282 kg/m3) increased the CTL as the dosage of nitrite in concrete increased, but did not extend the time to corrosion because it accelerated chloride penetration. For a richer mix (350 kg/m3) the time to corrosion increased with the dosage of calcium nitrite. After corrosion initiation, the corrosion rate for specimens containing calcium nitrite was typically two to three times higher than for nitrite-free specimens.