Stress corrosion cracking mechanism of prestressing steels in bicarbonate solutions

doi:10.1016/j.corsci.2007.05.025

Corrosion Science

Volume 49, Issue 11, November 2007, Pages 4069-4080

https://doi.org/10.1016/j.corsci.2007.05.025 Get rights and content

Abstract

Prestressing steels occasionally fail by a process named “stress corrosion cracking”. This process has not been fully elucidated and several theories exists in order to explain the cases in which real structures have collapsed. This paper briefly mentions the different theories and identifies the progress in understanding whether it is necessary to use a testing method, which is able to separate the different steps and mechanisms contributing to the failures.

This paper presents the methodology used for inducing controlled localized attack to study the susceptibility of the high strength steels resistance to stress corrosion cracking (SCC). The method is designed to study the growth of cracks initiated from a mechanical notch; the crack is not produced by fatigue.

It consists of several stages: coating of the bar with epoxy resin, generation of a small notch, constant load and controlled potential test in the media, mechanical test in air and fractographic study. It allows us to calculate the crack propagation rate and the fracture toughness in the same test.

Finally, it has been possible to apply the surface mobility mechanism (SMM) in order to identify the SCC mechanism that operates.

Introduction

Steel reinforcement in concrete is protected from corrosion by passivation due to the high alkalinity produced by the hydration of the cement. This protection can be maintained indefinitely until an aggressive element in sufficient concentration reaches the bar. The most common causes of corrosion are the carbonatation of the concrete cover, which produces a reduction in the pH of the pore solution, and the penetration of chlorides, which induces pitting corrosion.

A particular case of corrosion of the steel embedded in concrete is the stress corrosion cracking (SCC), which can appear in prestressed structures. The SCC is produced by the simultaneous action or synergy of a mechanical tension and a corrosive medium. Nucleation at the steel surface results in the appearance of microscopic cracks that penetrate and induce the brittle failure of the wire due to a triaxial stress condition.

The failures due to SCC in prestressed structures are seldom occurring. However due to the brittle failure, the accidents can give rise to collapses and catastrophes as that of the Point Pleasant bridge [1], of the cover of the Conference hall of Berlin [2], or of the pressure pipelines [3].

Up to the present, the mechanism of SCC has not been explained satisfactorily. Numerous mechanisms have been proposed to explain the brittle failure of metal failure under stress, but only some of them, three especially, are considered to be relevant:

1.
Mechanism of anodic dissolution; developed by Parkins [4].
2.
Film-induced cleavage model; whose theoretical aspects have been developed by Newman [5].
3.
Surface mobility SCC mechanism; developed by Galvele [6].

The mechanism of anodic dissolution (AD) considers the electrochemical anodic dissolution at the tip of the crack as the fundamental kinetic parameter. The film-induced cleavage model (FC) accepts the anodic dissolution at the crack tip but it places the emphasis to the mechanical properties and the effect of micro-notchs that are the cause of microscopic cracks. The surface mobility mechanism (SMM) proposes a new perspective in which the crack advances, not due to anodic dissolution but due to diffusion of atomic vacancies created in the lips of the crack towards its tip. SMM is the only mechanism that proposes equations enabling the prediction of crack propagation rate and that incorporates the effect of the hydrogen produced during the process, achieving the formulation of an extension of the theory on SCC to hydrogen embrittlement [6].

Coherent with the lack of agreement in the type of mechanism that operates, an agreed testing method does not exist for the study of the susceptibility to SCC or to hydrogen embrittlement either [7]. In the case of high strength steel wires for prestressed concrete, there is a standard with a test type where the aggressiveness is increased to accelerate the process. This test enables one to detect the susceptibility of steel to hydrogen embrittlement and serves as a quality control test to detect defects [3], [8].

Other authors suggest a test closer to concrete performance, which is based on the application of the theory of anodic dissolution [9], [10], [11] to specimens in alkaline solutions containing chlorides or of sodium bicarbonate [12], [13] or in the use of pre-cracked specimens induced by fatigue [10], [11], [14], [15] in whose studies the fracture toughness is calculated by fracture mechanics.

None of these tests can be generalized and give conclusive results. It has thus seemed, necessary to try to develop a more suitable testing method. The main measurements would be: It should be practical, able to be used for control of production and for making predictions of long term performance.

In the present work, some results of a more realistic testing methodology than the present ones are given. It allows us to calculate the fracture toughness of a wire and to predict the speed of the advancement of the crack. It replaces the generation of fatigue cracks by the creation of a natural crack made to grow from one notch, electrochemically by treating in a solution of sodium bicarbonate with the wire under tension. When the crack reaches a certain depth, it is tested in air. The fracture toughness is calculated by Fracture Mechanics (FM). Besides, the theory of the SMM is applied to the fractographic results.

Section snippets

Materials and equipment

A steel of eutectic composition has been tested in two conditions: cold drawn steel (1510 MPa yield strength) and the modified parent pearlitic steel (1300 MPa yield strength). The chemical composition of both is therefore the same and it is as shown in Table 1.

The machine of slow strain rate tests (SSRT) used in the study has a speed range between 5000 μm/min and 0.10 μm/min. The load limit is 40 kN and the displacement is 10,000 μm. The specimen is fixed to gags by a thread.

The SCC cell has been

Selection of testing conditions to localize the pit and generate a crack

Due to the application of a fixed potential in the range where corrosion develops, oxides were visible in the mouths of the notch. Having verified this performance, the bars were mechanically tested in air notice how this slight corrosion can affect the strain–stress curve. Both steel types tested show lower values of strain at the failure load and lower reduction in area in the bars coated with the epoxy resin.

The uncoated notched specimens that were not covered with epoxy resin (b) (Fig. 2)

Discussion

Brittle failures were obtained, thanks to the localization of the crack in the manner described above and by the application of an external potential. For the analysis of the results, the SMM mechanism proposed by Galvele [6] and the theory of fracture mechanics applied to cylindrical wires proposed by Valiente and coworkers [15] has been used.

From the results obtained, two parameters were calculated:

1.
The crack propagation rate obtained from the potentiostatic test.
2.
The fracture toughness

Conclusions

In the present study, we have attempted to explore experimental conditions to induce SCC phenomena closer to some real conditions (carbonated) instead of using pre-crack specimens generated by the fatigue test and additionally to calculate the crack propagation rate by the SMM and the metal toughness by fracture mechanics. The first objective has been only partially achieved, because the crack initiation phase has not been naturally obtained. It was necessary to produce a mechanical notch on

Acknowledgments

The authors thank the Ministerio de Fomento for the funding and the accomplishment of the project “Not destructive methods and strategies for the control of the corrosion in pretested steels”, the Department of Science and Technology (MAP2003-03912), to the CSIC for the scholarship of investigation I3P and, specially, Prof. Gustavo Guinea (UPM). They are also grateful to Dr. J.R. Galvele for his useful explanation of the SCC theory.

References (16)

J.R. Galvele
A stress corrosion cracking mechanism based on surface mobility
Corros. Sci.
(1987)
A. Valiente et al.
Premature failure of prestressed steel bars
Eng. Fail. Anal.
(1998)
S.T. Rolfe et al.
Fracture and Fatigue Control in Structures
(1977)
U. Nurnberger, Análisis and evaluation of failures in prestressed steel, in: Proceedings of the Third FIP Symposium,...
M. Elices, Problemas de Corrosión Bajo Tensión en el Hormigón Pretensado, I Congreso Iberoamericano de Corrosión y...
R.N. Parkins
K. Sieradzke et al.
Stress corrosion cracking
J. Phys. Chem. Solids
(1987)
J. Fullea, Experimental methods to evaluate the susceptibility of prestressing steels to stress corrosion cracking, in:...

There are more references available in the full text version of this article.

Cited by (56)

Influence of strand rupture on flexural behavior of reduced-scale prestressed concrete bridge girders with different prestressing levels
2024, Engineering Structures
In prestressed concrete (PC) bridges, high-strength prestressing steel plays a key role in the flexural response of girders. In both pre-stressed and post-tensioned PC bridge girders, damage induced by corrosion or impact of under-passing high vehicles may led to rupture of one or more strands, significantly affecting the load-carrying capacity against dead and traffic loads. This paper presents an experimental study on the response of PC girders affected by strand rupture at different locations. Four reduced-scale post-tensioned concrete girders with different levels of initial prestressing force were tested under four-point bending. Two specimens with a different level of prestress were initially tested in undamaged configuration to provide a basis for the assessment of damage effects. The other pair of specimens were intentionally damaged - having the same initial prestressing level of their undamaged counterparts - in the lower tendon to investigate their limited flexural response under different performance levels including serviceability and collapse. Damage turned into a different critical cross section outside the loading region, i.e. both cracking and collapse were attained 1.1 m away from midspan. In the last section of the paper, a cross sectional analysis of the girder was developed at cracking and ultimate conditions for the prediction of damage impact on failure mechanisms, thus validating the experimental findings.
Future trends in research on reinforcement corrosion
2023, Corrosion of Steel in Concrete Structures
The concept of service life has made to evolve the science in reinforcement corrosion from a more basic description of phenomena to a more engineering perspective in which quantification of the processes is of vital importance. The chapter treats the following subjects: (1) Causes of reinforcement corrosion: carbonation and Chloride ingress, Stress corrosion Cracking (2) Corrosion onset and Chloride threshold, (3) Corrosion propagation. Period, (4) modeling of service life, (5) Additional Preventive methods, (6) Repair materials and techniques, and (7) Corrosion measurement techniques. The paper does not develop the known aspects of these different subjects but identifies those that still need further research work. In the exercise, it is identified that the modeling of processes and their incorporation into the structural and engineering calculations as the most important subject to be improved and further developed in the near future.
Durability of a reinforced concrete structure exposed to marine environment at the Málaga dock
2022, Case Studies in Construction Materials
Citation Excerpt :
The main causes of corrosion in reinforced concrete structures are concrete carbonation and/or chloride diffusion, mainly in marine environments. On the other hand, the consequences of corrosion such as cracking or even spalling of the concrete cover, loss of bonding and loss of cross-section, which can lead to brittle failure of the structures, are well known [12–20]. Is it a design issue, an on-site placing problem, or is it because a bad material selection?
The durability of structures in aggressive environments does not often reach the service life for which they were designed. This paper shows the methodology and results obtained in the evaluation of a 22-year-old breakwater’s crow wall. The results reveal that there are no degradation processes or expansive reactions in the concrete, nor carbonation. However, there is high chloride contamination, which has led to depassivation of the reinforcement of the entire structure, which in turn has led to bursting and delamination of the cover in some areas. Monte Carlo analysis shows that the chloride concentration exceeds the critical concentration given by the regulations up to a depth of 150 mm. It is advisable immediate repair and protection of the structure given the condition of the structure.
Influence of microstructure on stress corrosion cracking of X100 pipeline steel in carbonate/bicarbonate solution
2022, Journal of Materials Research and Technology
The stress corrosion cracking (SCC) behaviors of X100 pipeline steels with different heat-treated microstructures in carbonate/bicarbonate solution at various potentials were investigated in this study. Several analyses were performed using electrochemical measurements, slow strain rate tensile (SSRT) tests, scanning electron microscopy (SEM), and electron back-scattered diffraction (EBSD). The results showed that the SCC susceptibilities of heat-treated microstructures increase as the cathodic potential shifts from −850 to −1200 mV_vsSCE. At a weak cathodic potential, the different microstructures show clear SCC sensitivity, showing that hydrogen cannot be ignored in this circumstance when the alkaline environment is not considered seriously. At a mild cathodic potential, hydrogen sensitivity controls SCC behavior, but at a larger negative potential, both electrochemical effect and hydrogen sensitivity influence SCC behavior. The SCC cracks propagate along the high angle grain boundary (HAGB) at open circuit potential (OCP) and as the applied potential is adjusted negatively, the cracks would either follow low angle grain boundary (LAGB), HAGB or random orientation propagation paths. The synergistic influence of dislocation, grain boundary, grain size, martensite/austenite (M/A) island, and carbide controls the SCC process of the X100 pipeline. SCC susceptibility was shown to be high in microstructures with coarse grain and high dislocation density, whereas grain boundaries such as the martensite/austenite (M/A) island and carbide offer sites for SCC initiation.
Mechanical behaviour of corroded strands under chloride attack: A new constitutive law
2022, Construction and Building Materials
The paper presents an experimental and analytical study on the tensile behaviour of naturally corroded prestressing strands. A total of twenty-four strands were extracted from full-scale beams which were exposed during 10 years to chlorides from sea water. Firstly, a database has been created on the dependency of the maximum pit depth on the strand mass loss or on the section loss of the most corroded wire. Secondly, the morphology of corrosion pits and the maximum cross-sectional loss of each external wire have been measured. Finally, based on conducted tensile tests, a new constitutive law (CPS-model) is proposed to predict the mechanical behaviour of prestressing strands induced by pitting corrosion.
Full incremental elastic field induced by a mode-I crack in a prestressed orthotropic material
2020, Engineering Fracture Mechanics
Prestressed structures are frequently encountered in civil engineering. This paper gives a theoretical analysis of a cracked orthotropic elastic plane with initial stress. The problem of an orthotropic composite with a crack parallel to a material principal axis under initial stress along the crack line reduces to a mixed boundary value problem. It is solved using the Fourier transform. Application of boundary and continuity conditions leads to dual integral equations. An analytical solution can be obtained. Full incremental elastic field in an infinite prestressed cracked orthotropic composite is derived explicitly for linearly distributed loading on the crack faces. The fracture parameters such as incremental stress intensity factor (SIF), crack opening displacement (COD), and the energy release rate (ERR) are all obtained. The results for a prestressed isotropic elastic medium with a crack can be reduced from the present. When prestress is removed, the explicit expressions of the elastic field induced by a crack embedded in an orthotropic composite are recovered. Finally, the influence of initial stress on the elastic stress field, incremental SIF, COD and ERR for a cracked orthotropic elastic plane are presented graphically.

View all citing articles on Scopus

View full text

Stress corrosion cracking mechanism of prestressing steels in bicarbonate solutions

Abstract

Introduction

Section snippets

Materials and equipment

Selection of testing conditions to localize the pit and generate a crack

Discussion

Conclusions

Acknowledgments

Corros. Sci.

Eng. Fail. Anal.

Fracture and Fatigue Control in Structures

Stress corrosion cracking

J. Phys. Chem. Solids