Edited by: David M. Bastidas, University of Akron, United States
Reviewed by: Zehra Canan Girgin, Yıldız Technical University, Turkey; Jian-Guo Dai, Hong Kong Polytechnic University, Hong Kong; Jose Maria Bastidas, Centro Nacional De Investigaciones Metalurgicas, Spain
This article was submitted to Structural Materials, a section of the journal Frontiers in Materials
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
It is commonly mentioned that corrosion of steel in presstressed concrete can induce brittle failure. It is also understood that in order to estimate the presstressed structures service life is necessary to study the pitting factor that compares the maximum corrosion pit depth with the section loss by homogeneous corrosion. Results are presented on corrosion tests to measure the corrosion depth in tendons in contact with water. In present paper, a statistical analysis has been carried out in order to obtain the pitting factor for different cross sections submerged in water. The results indicate that the pitting factor value varies between 1.20 and 1.85 depending on the confidence interval considered, which can significantly reduce the service life of the presstressed tendons as long as the exposure conditions are those set out in this study.
Corrosion is the principal reason of durability reduction of reinforced concrete structures (Andrade et al.,
The work presented in this study attempts to quantify statistically the corrosion depth in tendons whose ducts have not been grouted with mortar. Therefore, tendons have been exposed to the weather or partially submerged in water in order to simulate similar conditions of tendons that are within the sheaths in direct contact with rainwater or other sources and which have been accumulated during the time they have not been injected.
With the aim of obtaining the pitting factor (González et al.,
The bars used as test materials were of high strength steel, cold drawn, 15.24 mm diameter and 25 cm length. Each tendon is made up of seven steel wires and has a diameter of 5 mm. Twelve tendons have been tested in total.
Before and after testing, the bars were cleaned in hydrochloric acid diluted with water and degreased in acetone.
The water used in the tests has the composition given in
Water composition.
6.9 | 79 μS/cm | 0.06 mg/l | 0.02 mg/l | 0.05 mg/l |
The methodology take into account the following steps: (i) submerged the bars and tendons in water and measure the corrosion rate during 30 days; (ii) measure the corrosion depth at different sections with metallographic microscope.
The following experimental methods have been carried out in order to quantify the corrosion depth in tendons.
Tendon and bars deterioration over time due to corrosion has been studied through electrochemical techniques for measuring potential and corrosion rate. For this purpose, Polarization Resistance Method according ASTM G 59-91 standard has been used with an ACM potentiostat (Stern and Geary,
In the present work potentiodynamic measurements were taken by polarizating cathodically from the Ecorr up to −10 mV, and then, anodically up to +10 mV (Andrade and González,
Two test cells have been studied with three bars in each cell. The tests had duration of 30 days (
- Reference electrode: Ag/AgCl has been used as reference electrode.
- Working electrode: Tendons were cut and divided in bars. Each test cell is formed by three steel bars object of study acting as working electrode.
- Counter electrode: a graphite bar has been used as auxiliary or counter electrode.
Summary of samples and tests performed.
B1-1 | Bar | 30 days | Electrochemical and weight loss |
B2-1 | Bar | 30 days | Electrochemical and weight loss |
B3-1 | Bar | 30 days | Electrochemical and weight loss |
B4-2 | Bar | 30 days | Electrochemical and weight loss |
B5-2 | Bar | 30 days | Electrochemical and weight loss |
B6-2 | Bar | 30 days | Electrochemical and weight loss |
T1-1 | Tendon | 30 days | Metallographic microscope |
T2-2 | Tendon | 19 days | Metallographic microscope |
T3-2 | Tendon | 19 days | Metallographic microscope |
T4-2 | Tendon | 30 days | Metallographic microscope |
T5-2 | Tendon | 30 days | Metallographic microscope |
T6-3 | Tendon | 30 days | Metallographic microscope |
T7-3 | Tendon | 19 days | Metallographic microscope |
T8-3 | Tendon | 30 days | Metallographic microscope |
T9-3 | Tendon | 19 days | Metallographic microscope |
The total weight loss is obtained from the area integration of the Icorr-time curve (Sanchez et al.,
where P
On the other hand, gravimetric weight loss has been obtained as the difference between initial and final weight of steel bars at the end of the test. Both cases, steel bars were cleaned in hydrochloric acid diluted with water and methenamine, afterwards it degreased in acetone.
Corrosion depth in tendons has been studied with metallographic microscope at x500 magnification. This technique allows, through further image processing, measuring the pit depth due to corrosion on the surface of the steel.
Before measuring the corrosion depth, tendons have been pre-corroded partially submerging them into a container with water. Four tendons have been pre-corroded for 19 days and five tendons for 30 days
From each pre-corroded tendon three cross sections have been made at 3 different heights: air/water interface area (section A), submerged intermediate area (section B), and bottom tendon area (section C).
Each section is embedded in resin and polished to obtain samples as shown in
Embedded and polished sample after corrosion tests.
x500 magnification photo and measures of pit depth in pre-corroded bars.
Corrosion rate over time measurements in submerged steel bars are shown in
Corrosion rate (
Corrosion potential (
Corrosion rate stabilizes during the first 24 h and after that, it remains constant throughout the whole test, being around 10 μA/cm2, a value above the limit considered for the steel depasivation threshold, located at 0.2 μA/cm2 (Andrade et al.,
All values, both in the corrosion rate and in the corrosion potential measurements, are very similar to each other, which allow deducing a representative value with little dispersion.
Gravimetric weight losses in steel bars submerged in water.
As previously mentioned, a total of 4,500 corrosion depth measurements have been carried out by metallographic microscopy at 500 magnifications. 3 areas have been statistically studied in 9 tendons: (i) air/water interface area (section A), (ii) submerged intermediate area (section B), and (iii) bottom tendon area (section C).
Corrosion depth distribution at air/water interface (section A).
Corrosion depth distribution at submerged intermediate area (section B).
Corrosion depth distribution at bottom tendon area (section C).
A summary of the corrosion depth experimental distribution values by cross section are shown in
Corrosion depth experimental distribution values and main parameters of probability density functions fitted to a lognormal density function.
No. of measures | 1,439 | 1,550 | 1,550 |
Mean (μm/year) | 29.7 | 27.0 | 28.9 |
St. deviation (μm/year) | 19.8 | 14.9 | 18.0 |
Percentile 25% (μm/year) | 18.1 | 18.1 | 18.1 |
Median (μm/year) | 25.4 | 25.4 | 25.4 |
Percentile 75% (μm/year) | 35.6 | 32.7 | 33.1 |
Percentile 90% (μm/year) | 47.1 | 43.4 | 47.1 |
Percentile 95% (μm/year) | 64.0 | 52.0 | 61.7 |
Maximum (μm/year) | 177.7 | 189.4 | 223.8 |
Log (mean) (μm/year) | 3.22 | 3.1 | 3.2 |
Log (St. deviation) (μm/year) | 0.6 | 0.7 | 0.5 |
Corrosion depth for the 90% |
58.2 | 53.6 | 49.7 |
Corrosion depth for the 95% |
74.0 | 68.1 | 60.3 |
Corrosion depth for the 99% |
116.0 | 106.7 | 86.8 |
With the data obtained from both gravimetric and electrochemical weight loss, and since there is a low dispersion in the results obtained in the different bars studied, it is possible to generalize these values and extrapolate them to other different ages. Therefore,
Since the corrosion depth was measured by microscope when the tendons were pre-corroded, in order to calculate the pitting factor, α, that relates the homogeneous corrosion with the maximum pitting penetration (González et al.,
Outline of pitting factor
Cross section loss by bar.
1 | 9.9 | 0.114 |
2 | 10.7 | 0.125 |
3 | 9.6 | 0.111 |
4 | 10.1 | 0.117 |
5 | 10.2 | 0.118 |
6 | 10.2 | 0.118 |
Mean | 0.117 | |
St. Deviation | 0.004 |
Cross section loss by section.
Air/water interface | 142.3 | 171.1 | 217.8 | 1.21 | 1.46 | 1.86 |
Submerged | 140.3 | 170.9 | 185.4 | 1.20 | 1.46 | 1.91 |
Bottom | 142.3 | 167.0 | 204.0 | 1.21 | 1.42 | 1.74 |
Mean | 141.6 | 169.6 | 202.4 | 1.21 | 1.45 | 1.84 |
Durability predictions.
According to
The main conclusions that may be pointed out are:
Tendons studied in contact with water have homogeneous corrosion. A maximum pitting factor value varies between 1.20 for a confidence interval of the 50% percentile and 1.85 for a confidence interval of the 99% percentile.
Durability predictions show that the service life of the tendons can be significantly reduced for a service stress of 70% of the bar ultimate load if maximum pitting factor is considered as long as the exposure conditions are those set out in this report.
There is no evidence that the corrosion depth depends on the submerged tendon cross section, since both their corrosion depth experimental distributions, and their probability density functions fitted are very similar to each other.
JM carry out the test. JT and NR carry out metallographic analysis. JS are the advisor of the research group.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.