The Extent of Chloride Corrosion on Rebar

An Experimental Study

Word Count: 3766

Table of Contents

Introduction        2

Literature Review        4

Methodology        6

Errors and Limitations        10

Results and Discussion        12

Conclusion        18

References        19

Introduction

The corrosion of rebar in bridges is a growing international problem and is the most costly cause of concrete failures. Annually, the United States experiences an estimated $8.3 billion worth of damage due to bridge failures that are caused by rebar corrosion (“A Global Need”, n.d.). Rebar is necessary in concrete bridges because it provides the tensile strength that concrete lacks and is also capable of resisting compression forces, which allows it to be used for structural purposes. While the concrete does provide some protection for rebar from corrosion, different mixtures and ratios within the concrete oftentimes leave the rebar susceptible to corrosion (McCormac & Brown, 2009, p. 9). The most common form of corrosion is caused by chlorides, which occurs when deicing salts diffuse into the concrete (Akindahunsi, Falade, Afolayan, & Oke, 2010). Corrosion begins when the level of chlorides around the rebar reaches a threshold and begins to break down the protective passive film on the surface. This chloride threshold is dependent on many outside factors such as the cement type, water to cement ratio, temperature, humidity and the source of chloride ions, etc (Jiang, Wang, Chu, et al., 2017). Based on the possibility for variation, specifying a value for the corrosion threshold is misleading and inaccurate. Once the corrosion threshold is passed, the volume of the rebar will increase because  the resulting oxides will occupy a larger volume, causing more pressure to be exerted on the concrete around it as the rebar continues to grow in size (Akindahunsi, Falade, Afolayan, & Oke, 2010). This extra pressure may cause cracking or spalling of the concrete, both of which will accelerate corrosion (McCormac & Brown, 2009, p. 26).

Researchers have found ways to improve standard rebar by putting different protective coatings on it, resulting in galvanized and epoxy-coated rebar. It is also possible to use completely different materials that are more corrosion-resistant, which is the basis of stainless steel rebar. One benefit of galvanized rebar is that the zinc coating on the rebar will act as a barrier that protects the steel from corrosive elements (International Zinc Association [IZA], 2018). During the galvanizing process, a metallurgical reaction takes place between the steel and zinc, which extracts the metals and makes refined alloys that cover the steel. The top layer of coating is made up of an iron/zinc alloy layer and an outer layer of pure zinc. Zinc is able to protect the steel from corrosion because an oxidation reaction will occur on the zinc layer, making it the anode. Electrons will flow from the anode, so the zinc will corrode before the steel will corrode (IZA, 2018). Unlike galvanized rebar, epoxy-coated rebar has a thin, powder coating on it that is applied at temperatures of 400 °F - 450 °F. The powder particles are charged as they leave the spray nozzle, which causes them to be attracted to the grounded steel surface of the rebar (Epoxy Interest Group, 2020). This will form a thermosetting cross-linked polymer on the surface of the rebar, which is what makes it resistant to corrosion (Concrete Reinforcing Steel Institute [CRSI], 2015). However, this coating can easily be broken off when the rebar is not handled with extreme care, leaving areas exposed to corrosion and making the remaining coating useless. Epoxy-coated rebar is still used in projects though because it is very cost effective when compared to both galvanized and stainless steel rebar, while still offering some corrosion resistance. Like galvanized rebar, stainless steel rebar has the advantage of being more user friendly than epoxy-coated rebar, but it is much more expensive (McCormac & Brown, 2009, p. 26). It is completely different from the other two types of rebar because it is made of an entirely different material, rather than just placing a coating on top of regular steel. Originally, stainless steel rebar was an iron alloy that contained about 12% chromium, but as new problems are surfacing more variations of it have been manufactured and there are currently hundreds of types of stainless steel rebar (Anupoju, 2014). Galvanized rebar, epoxy-coated rebar, and stainless steel rebar have revolutionized reinforced concrete. These revolutions have made it possible to build bridges that are longer and higher than previously imagined. However, corrosion is still a large issue in bridge failures today.

Literature Review

Chloride diffusion is the main cause of corrosion in rebar and the diffusion coefficient of chloride diffusion in solution can be found using Fick’s Second Law of Diffusion,, where D is the diffusion coefficient and C is the chloride content (Tu, Di, Pang, & Xu, 2018). The corrosion rate is one parameter that is often used to make reasonable service life predictions when building bridges (Otieno, Beushausen, & Alexander, 2011). It can be found using Faraday’s Law, , where pd(t) is the time-dependent corrosion penetration depth and icorr is the corrosion  rate (Rodriguez, Ortega, Casal, & Diaz, 1996). The penetration depth can then be estimated by substituting in the DuraCrete (1998) model and integrating (Muthulingam & Rao, 2015). These equations are beneficial for quantifying the corrosion process, but are not based on testing the actual concrete and rebar.

One way to test for chloride corrosion, as opposed to just calculating it, is finding the corrosion potential. Numerous studies have been done to test for corrosion potential of rebar, as well as to test how accurate the measurements from these tests are. Scientists have found other possible ways to test for corrosion in rebar, but half-cell potential is the most accurate indicator of corrosion initiation because it is not dependent on steel type, cement type, or water to cement ratio (Pradhan & Bhattacharjee, 2009). Specifics for how to run an experiment that will give accurate half-cell potential measurements are stated by the American Society for Testing and Materials (1991). This allows for credible research to be done across the world in multiple papers and standardizes this research process. These procedures are being continuously tested and improved, so results will only get more accurate. Using this procedure, researchers have found that dissolved salts from deicing salts or salt water will have a significant effect on rebar by increasing the corrosion potential (Akindahunsi, Falade, Afolayan, & Oke, 2010). This has a significant effect on the failure of bridges in salt water environments and places where deicing salts are frequently needed.

 Researchers have tried to test various additives in concrete that will help protect the rebar from chloride diffusion. It has been found that with an increase in rubber aggregates in the concrete surrounding the rebar, there will be a smaller corrosion area. This is due to the fact that the addition of the aggregates improves capillary pore saturation (Liang, Zhu, Chen, et al., 2019; McCormac & Brown, 2009, p. 17). Another research group found that the addition of amorphous silica additives to the concrete increases the final density of the cured concrete. The increased density leads to a slower diffusion of chlorides throughout the concrete, which in turn  slows the corrosion rate (Kline, Pesic, Raja, & Ehrsam, 2018). Various additives to concrete and rebar treatments have been tested as a possible solution to rebar corrosion. Different research groups advocate that their respective form of treatment is best for preventing rebar corrosion, such as the Epoxy Interest Group and International Zinc Association, but not many studies have been done to test solely based on rebar treatment. This paper will address this gap by testing each of the rebar treatments to find out how effective each of these treatments are at preventing chloride corrosion. Then, each rebar treatment will be compared to one another to find out which is best suited for prevention of chloride corrosion in bridges.

Methodology

For this paper, an experimental research method was chosen to test differing rebar treatments and how they affected the rate of corrosion. The results of this experiment would show the effectiveness of each treatment in preventing corrosion. This would allow for the treatments to be compared to each other to find the most corrosion resistant. The rebar would be tested by having it both directly exposed to the salt water solution and in concrete blocks. The concrete blocks would more accurately replicate the conditions the rebar would normally be in, surrounded by concrete in the bridge’s deck beams where deicing salts would be applied.

The pieces of rebar were cut to 10 inches long and had a diameter of ⅜ inches. All of the rebar pieces had rolled on deformations, which allows the rebar to better bond with the concrete and is more commonly used than plain bars. Rebar with rolled on deformations have ribbed projections rolled and stuck on to their surfaces (McCormac & Brown, 2009, p. 22).  The untreated rebar was a common steel rebar that can be found in a local hardware store.  The epoxy coated rebar was donated by the Epoxy Interest Group, who use a pre-blast cleaned and single coating method on their rebar. This method is when the steel is cleaned with abrasive grit, which removes contaminants, mill scale, and rust and then coated with only one layer of powdered epoxy coating. The galvanized rebar was donated by South Atlantic Reinforcing and was treated to ASTM A1094 specifications. Four different stainless steel pieces of rebar were donated by Salit Specialty Rebar and used in this experiment. They are EnduraMet  316LN Stainless, EnduraMet  32 Stainless, EnduraMet 2205 Stainless, and EnduraMet 2304 Stainless. Table 1 shows their differing percent compositions.  

In order to make the concrete blocks with rebar running through them, 7 plywood boxes that measured 3 inches x 4 inches x 8 inches were made and used as the forms for the concrete blocks. The forms had a hole that was ⅜ inch in diameter cut out of each of the 3 inch x 4 inch pieces. The holes were 1 ½ inches from the top and 2 inches from the side, so that the rebar could be suspended through the middle of the box. See Figures 1 and 2 for a diagram of the layout. Suspension through the box would ensure that the rebar was level because it would prevent the rebar from shifting while the concrete was poured in. The rebar was then inserted through the forms with 1 inch of rebar exposed on each side of the box and a ring of electrical tape applied around the part of the rebar that was closest to the outside of the box, to prevent it from sliding while the concrete was being poured. The concrete was mixed with a 0.55 water to cement ratio and poured into each of the forms to fill them up to the top. In order to get air bubbles out of the concrete, an immersion vibrator was placed into the concrete mixture for 10 seconds per box. The concrete was allowed to cure for 3 days, undisturbed and uncovered, in a shaded area on a table outside before the wooden forms were broken off.

After the forms were broken off, the blocks were moved to an air conditioned building where the experiment would be taking place. They were placed into two plastic boxes in front of a window, so that they could be exposed to temperature changes throughout the day.

 For the concrete block that contained galvanized rebar, it was originally going to be put into a separate container for testing because the electrical current that would be generated by the unpreventable de-passivation of the zinc coating on the rebar would be very hard to distinguish from the actual half-cell potential values and corrosion activity (Yeomans, 2004). The galvanic current measured would only equate to the half-cell potential if a pure anode was used that contained no microcell activity. This theoretically could only occur if an entirely oxygen-free atmosphere existed around the rebar (Andrade et al., 2005).  However, it was decided that this could only affect the measurements taken for the galvanized rebar and would not affect the half cell potential readings of the other pieces of rebar that happened to be in that same container, so they were not placed into separate containers.

The boxes were then filled with 15,000 mL of water, enough to cover the blocks completely. After the water was poured in, 525 grams of industrial grade deicing salt was then mixed in, creating a 3.5 % NaCl solution surrounding the blocks. This percentage was chosen based on an experiment that collected half cell potential values in a way very similar to what would be done here (Liang, Zhu, Chen, et al., 2019). Then, separate alligator clips were connected to each piece of rebar and to the positive end of a 9-V battery. In order to complete the circuit, a wire was connected to the negative output of the power source and placed directly into the NaCl solution.

The experiment initially was going to run for 28 days, but it became apparent after only 22 hours and 45 minutes that too much current was being run through the rebar. Resistors that would have been able to decrease the current were not available, so the power was shut off. The experiment was reset in new containers with the same specifications as previously stated, but the alligator clips and circuit were removed from the design in order to ensure that the experiment would run safely and still yield results. This was then allowed to run for the remainder of the 28 day period, with qualitative observations being taken every fourth day. Although half-cell potential values would not be able to be taken, qualitative measurements could still be recorded on the same days that observations were being taken for the bare rebar.  After the corrosion test period the concrete was broken off of the rebar pieces with a jackhammer, so the appearance of the concrete covered rebar could be observed. This was then compared to the noncovered concrete to see what effects the concrete cover had on the corrosion of the rebar.      

In order to set up the non-concrete rebar portion, two boxes were placed in the same area as the boxes that held the concrete blocks. These boxes, Sterilite Underbed storage boxes, measured 23 ½ inches x 16 ⅞ inches x 5 ⅞ inches and could hold 27 liters of water. The mass of each of the rebar pieces was measured and then they were randomly placed into the two boxes. The rebar in each box were then submerged in 4000 mL of water. Then, 140 grams of industrial grade deicing salt was added to reach a 3.5% NaCl solution to surround the rebar pieces. The boxes then sat covered and undisturbed for 28 days, only having the lid removed when observations were made. Qualitative observations were taken every fourth day on the overall appearance of the rebar and the surrounding water. After the 28 day period, the loose rebar was dried off and brushed with a steel brush to remove any rust that was present. Calculations were done to determine how much mass was lost by the loose rebar due to corrosion.

Errors and Limitations

 Due to unforeseen events, there were some errors that occurred that could have affected the outcome of this experiment. The first issue that occured was that the circuit that was  originally set up to test the rebar in concrete blocks had too much current through it. This caused the iron to precipitate out of the rebar and form the brownish red solid that was found floating on top of the water. Furthermore, the current was strong enough to melt the alligator clips entirely, interrupting the circuit and making it useless. Rather than setting the experiment up again, the circuit aspect was removed altogether and run in a more simplified way.

Another error that occurred during this experiment was the unforeseen cancellation of school due to the Coronavirus pandemic. The rebar was originally going to be taken out of the saltwater solution on March 9th, but due to school closures the rebar was removed on March 23rd. Also, this prevented observations from being taken at the set four day increments at the end of the experiment.

Throughout this experiment there were limitations that caused the experiment to have a more simplified design. The experiment was conducted in a high school science lab. There was no access to resistors that would have been necessary to continue the circuit design and have accurate half-cell potential readings. One measurement that could have been taken for the concrete covered rebar was a four point bending test, which would show load-deflection curves and failure cracks. These tests can be carried out with a WE-30 universal material test machine, which was not available (Liang, Zhu, Chen, et al., 2019). However, the largest limitation during this experiment was time. Rebar corrosion occurs over long periods of time, depending on the conditions. This experiment attempted to accelerate the process by directly exposing the rebar to the chlorides and through the use of electricity. While this caused the rebar to corrode to some extent, if left for an extended amount of time, the experiment would have garnered much more definitive results.

Results and Discussion

        For this experiment, qualitative measurements of the overall appearance were recorded for both the bare and concrete covered rebar pieces and quantitative measurements of the mass lost because of corrosion were taken for the bare rebar. Observations were taken every four days and the mass of each piece of bare rebar was recorded before and after the experiment.

        Before the experiment was run, the mass of each piece of bare rebar was measured using an electronic balance. The percent change of the mass of each piece of rebar was then calculated using these numbers and the formula:. For example, the percent change in mass for the untreated rebar is or -1.00%. These values, along with the mass of the rebar before and after the experiment, can be found in Table 2 below.

The qualitative observations that were taken every four days can be found in Table 3, for bare, and Table 4, for concrete covered, below.

The results of this experiment found comparative qualitative rankings amongst both the concrete covered and bare rebar pieces. These rankings, along with the percent mass loss for each piece of bare rebar, will give insight into which rebar treatment is most effective. The untreated rebar was the worst performing, as expected, and experienced the most corrosion in both sections of this experiment. It had an overall percent change in mass of -1.00%, which is 0.237% more than the next highest percent change. In terms of observational results, the untreated rebar had the highest corrosion, which was apparent because of an opaque orange cloud that filled the water surrounding the rebar. The best performing rebar treatment was the EnduraMet 2205 stainless steel rebar, which had no mass lost over the experimental period. For both the bare and concrete covered pieces, this treatment had the least amount of rust visible and only had rust dissolve into the water at the last observation for the concrete covered piece. While the EnduraMet 2304 stainless steel rebar also had a 0% change in mass, there was slightly more rust noticeable in the water surrounding the rebar and at an earlier time, making it slightly less effective than the EnduraMet 2205 stainless steel rebar in this experiment. The galvanized rebar did not have a large amount of rust apparent on its surface, which was largely due to the fact that the outer coat was deteriorating as it was protecting the steel underneath. However, the galvanized rebar had the second highest change in its mass, losing 0.763%. Due to the high loss of mass, it was concluded that the galvanized rebar was not the best performing treatment in this experiment. The epoxy-coated rebar also did not have widespread corrosion, but had very concentrated spots where the powder coating had small holes, which was especially prevalent on the ends. It lost slightly less mass than the galvanized coating, 0.510%, but the extremely dense rust patches along the rebar piece make this treatment ineffective in any real life applications where knicks would be almost impossible to prevent. The EnduraMet 32 stainless steel rebar and EnduraMet  316LN stainless steel rebar had similar mass loss, only 0.003% different, but the EnduraMet 316LN stainless steel rebar appeared the worst out of the four stainless steel treatment variations for both experimental sections. While the EnduraMet 32 stainless steel rebar appeared to perform better than the EnduraMet 316LN stainless steel rebar, it was not comparable to the other two stainless steel treatments.

        The best overall rebar treatment in this experiment was the stainless steel rebar because 75% of these treatments were unbeaten by any other type of treatment in any of the categories. The epoxy treatment had areas of very dense corrosion, which would make the rebar very susceptible to breaking in these sections. While the galvanized rebar did not have a large amount of rust present at the end of the experiment, it was very apparent due to the white film forming over the rebar that the chlorides were negatively impacting the coating. After a long period of time the chlorides would destroy the galvanized layer and begin to break down the actual rebar.

Conclusion

        It is reasonable to conclude from the results of this experiment that the most effective strategy to protect rebar from chloride corrosion is using stainless steel rebar. This is based on the fact that it had 0% mass lost due to corrosion and that 75% of the stainless steel treatments were unbeaten by any other type of treatment in terms of appearance after the 28 day period. This experiment provides evidence that it would be beneficial to use stainless steel rebar in the construction of bridges from this point forward. Although it is not reasonable to replace all non-stainless steel rebar currently present in bridges because that would be too costly and time consuming, an increase in bridges being built with stainless steel rebar would greatly limit the amount of bridge failures in the future. To increase information on this topic, future research should be done to investigate which composition of elements in the stainless steel rebar would provide the most chloride corrosion resistance. This would clear up the ambiguity and broadness of simply suggesting the category of stainless steel because varying compositions will have varying effectiveness and provide a more definitive answer to which type of rebar is best for preventing chloride corrosion.  

References

A Global Need (n.d.). Corrosion Instrument. Retrieved May 7, 2020, from https://corrosioninstrument.com/gn/

Akindahunsi, A. A., Falade, F. A., Afolayan, J. O., & Oke, I. A. (2010). Characterization and mathematical modeling of chloride diffusion in Lagos coastal waters. Journal of Failure Analysis & Prevention.

Andrade, C., Alonso, C., Gulikers, J., Polder, R., Cigna, R., Vennesland, Ø., Salta, M., Raharinaivo, A., & Elsener, B. (2005). Test methods for on-site corrosion rate measurement of steel reinforcement in concrete by means of the polarization resistance method. Materials and Structures, 37, 623-643.

Anupoju, S. (2014). Stainless steel reinforcing bars- Types and properties. Retrieved on November 23, 2019, from https://theconstructor.org/building/stainless-steel-reinforcing-bars-types-properties/945/

Concrete Reinforcing Steel Institute. (2015) Frequently asked questions (FAQ) about epoxy-coated reinforcing bars. Concrete Reinforcing Steel Institute.

Epoxy Interest Group. (2020) Manufacturing Process. Retrieved on December 12, 2019, from http://www.epoxyinterestgroup.org/use-applications/manufacturing-process/

International Zinc Association Web site. (2018). A. Green (Ed.).Retrieved November 20, 2019, from https://www.zinc.org/

Jiang, J., Wang, D., Chu, H., Ma, H., Liu, Y., Gao, Y., Shi, J., et al. (2017). The passive film growth mechanism of new corrosion-resistant steel rebar in simulated concrete pore solution: Nanometer structure and electrochemical study. MDPI.

Kline, J., Pesic, B., Raja, K., & Ehrsam, I. (2018). Corrosion studies of rebar in contact with saturated solution produced by leaching of portland type-II cement: Examining the effects of chloride, carbonation and an amorphous silica additive. Corrosion Engineering, Science and Technology, 53(6), 431-443.

Liang, J., Zhu, H., Chen, L., Han, X., Guo, Q., Gao, Y., & Liu, C. (2019). Rebar corrosion investigation in rubber aggregate concrete via the chloride electro-accelerated test. MDPI.

McCormac, J. C., & Brown, R. H. (2009). Design of reinforced concrete. John Wiley & Sons, Inc. 

Otieno, M., Beushausen, H., & Alexander, M. (2011). Prediction of corrosion rate in reinforced concrete structures- A critical review and preliminary results. Materials and Corrosion, 63(9), 777-190.

Pradhan, B., & Bhattacharjee, B. (2009). Half-Cell potential as an indicator of chloride-induced rebar corrosion initiation in RC. Journal of Materials in Civil Engineering.

Rodriguez, J., Ortega L. M., Casal, J., & Diez, J. M. (1996). Corrosion of reinforcement and service life of concrete structures. Durability of Building Materials and Components.

Yeomans, S. R. (2014) Galvanized steel reinforcement in concrete: An overview. Elsevier.