1. INTRODUCTION

The main problem for which a concrete structure needs repair is the corrosion of steel, which negatively impacts the durability of the constructions, putting at risk their functionality and safety (^{Andrade C., Feliu S., 1989}). In industrial and urban environments, the big concentrations of CO₂ represent a problem for the reinforced concrete due to the carbonation of the same, which causes a reduction in alkalinity and generates the depassivation and corrosion of steel. For the development of carbonation, the humidity and temperature are factors that modify its advance rate.

After having detected the corrosion in a structure, it is necessary to take actions to do repairs and prolong its useful life. The repair work can be divided into the following steps: elimination of the concrete, cleaning the exposed steel, and the application of a repair material (^{Fernández Cánovas, 1989}).

The selection of the repair material is the most important criterion to consider in the repair work; normally, the resistance to compression is taken into consideration to select a repair mortar, this is completely insufficient if other parameters that are equally important, such as the compatibility of the mortar with the concrete substrate, are not taken into consideration. The topic of compatibility implies the study of the difference between the properties of the concrete substrate and the repair mortar, as said difference could cause negative effects on the repair, causing cracking and thus decreasing its durability (^{Decter, 1997}).

Several studies have been conducted to evaluate the different types of repair materials that are available on the market. Unfortunately, the properties that should be considered for the evaluation and selection of the repair materials are not specified (^{Cabrera, 1997}). Some authors (Emmons, 1994) present the considerations that affect the compatibility for a new selection of the repair materials. From these considerations, the most important is probably the capacity to support changes in volume, without loss of adherence or cracks; this is known as “Dimensional compatibility”. In addition to the volume changes, a repair should also have a protector effect and to ensure this, the chemical, electro-chemical, and permeability of the repair material should be considered.

The problems related to the properties to be evaluated and the selection of the repair materials are attributed to the lack of regulation in relation to the repair activities, maintenance, and renovation (^{Kay, 1987}; ^{Treadaway, 1987}), in addition to the lack of information that the manufacturers of the repair materials provide. For this reason, it is necessary to carry out researches on the repair materials that are adequate to the environment and the service conditions of each structure. In turn, the use of alternative cement materials such as fly ash, will allow the use of repair materials of low environmental impact, due to the decrease in the use of ordinary Portland cement (OPC), and that comply with the necessary characteristics for a lasting repair.

The objective of this work is to evaluate the durability and potential performance of three mortars prepared in the laboratory, two with substitution of FA (20 and 50%) for OPC and three mortars of commercial repair; in particular, their mechanical behavior, dimensional stability, adherence, and their capacity to protect steel from corrosion by carbonation in an industrial environment.

2. EXPERIMENTAL PROCEDURE

Table 1 shows the proportion of the mortars elaborated in the laboratory. The MR mortar is used as reference, mortars M1 and M2 were elaborated with the same water/cement and cement-sand ratio as the reference one, but with FA additions of 20 and 50% in substitution with regard to the weight of MR cement. The cement used is an OPC 40 and the FA is type F-it was obtained from the Carboeléctrica of the city of Nava Coahuila, Mexico. The chemical composition of the OPC and FA can be observed in Table 2, obtained by X-ray fluorescence (XRF).

The commercial repair mortars are identified as C1, C2, and C3, these are pre-packaged mortars, cement based and of only one component; C1 is a reinforced mortar with fiber, C2 is fast-setting; C1 and C3 are modified with polymers and with additions of micro-silica in proportions that are known by the manufacturer. The mixture and the amounts of water used were the ones specified in the data sheet of each mortar.

Different types of specimens were manufactured in accordance with the test to be carried out; after their manufacture, the specimens were kept in a curing chamber at 100% Relative Humidity (RH) and at 21±2 ºC, for 14 days. After curing, the specimens were removed from the chamber until they reached 28 days of age.

For the corrosion potentials test, specimens with two steel bars of 6 mm in diameter were manufactured to leave a coating of 7 mm; additionally, a graphite column of the same diameter was placed to be used as a counter electrode.

To evaluate the capacity to resist the entry of aggressive agents, the specimens were exposed to carbonation in an industrial environment.

In the industry where the specimens were placed, industrial processes of sodium carbonate and bicarbonate are carried out, which creates an aggressive and damaging environment for concrete.

2.5 Corrosion potentials and rates

The corrosion potentials (E_corr) and rates (i_corr) are determined through the resistance to polarization test (ASTM G 59) with a sweep feed of 0.06 mV/s; only in the cathodic zone does it go from 0 to -10 mV. This methodology is applicable for the study of the protection that the repair mortars provide to reinforced steel.

The steel used was prepared and cleaned following the procedure established in the ASTM G1 (ASTM G 1) standard, which completely removes the corrosion products that it could have before starting the test. After the cleaning of the steel, it was weighted and carefully wrapped to define the area of study (50 mm).

Figure 1. Specimen to evaluate the durability of the repair mortars (units in mm) 

3. RESULTS AND DISCUSSION

Table 3 shows the mechanical behavior of each repair mortar and the results of the tests at 28 days. It can be observed that the C1 commercial repair mortar shows resistance at the highest compression, which is due to the content of fibers that decrease deformation in the face of a load and their role as reinforcement for the mortar.

The M1 mortar follows in the order of resistance, its value being higher than the MR mortar and the commercial C1 and C3 mortars. As per flexural strength, the C1 mortar shows the highest resistance due to the addition of fibers. The commercial C2 mortar is the lowest in regard to its mechanical behavior-its low resistance can be associated to the high porosity which, as is mentioned in figure 2, is not visible due to it being more permeable and it does not appear because it is outside the scale ranges. According to various investigations (^{Bjegovic 1990}; ^{Jiang, 2009}), the lower resistance of the C2 and C3 mortars is probably the result of their additives.

Figure 2. Evolution of the permeability of the mortar (Kt) over time. 

C2 mortar, being a fast-setting mortar, could contain magnesium sulfate or higher amounts of calcium sulfate, thus increasing its setting speed but decreasing its final resistance; for its part, C3 mortar contains polymers and microsilica in addition to having a higher content of air in order to improve its resistance to freezing and melting, this is not evident on the permeability graph, due to the fact that the polymers allow creating a greater amount of pores, but in an isolated and well-distributed manner.

The adherence was evaluated by the type of failure that the prisms present; Table 4 shows the resistance to adherence of each mortar at 28 days as well as their type of failure. If there is good adherence, the sample malfunctions monolithically as one piece, presenting cracks that spread from the repaired concrete to the repair mortar, instead of faltering throughout the union in one plane at 45°.

The other type of failure presented is on the interface, where the adherence failure between the concrete and the mortar happens before either of the two materials fails. The results of the resistance to adherence show that the mortars with the best behavior are C1 and C3 due to the failure presenting itself along with the concrete (monolithic); even though the monolithic failure is desirable, it can be observed that the M1 mortar presented a failure in the interface and yet showed the highest resistance for adherence (^{Cabrera, 1997}). This advantage is possibly due to the cement used for the manufacture of repair mortars being the same as the one used for the manufacture of the concrete samples on which the repair was done.

The C2 mortar, for its part, is the mortar with the least mechanical properties; in the adherence test, the failure immediately showed at the beginning of the test, not allowing the team to record anything. It is worth noting that the compatibility of the deformations is important to do a good repair, in materials with lower elasticity modulus (usually related to lower resistance) more deformations for the same level of load will manifest, originating the failure in the material with more deformations.

Figure 2 shows the relation of air permeability in time for the different mortars. It can be observed how permeability increases in almost all the mortars, with C3 mortar being the one that presents the lowest permeability values and without any increases in relation to time. The C2 mortar is not shown in the ranges on the scale because it is too permeable. While the commercial C1 mortar was not tested due to it having a rough surface and the presence of larger sized aggregates compared to the rest of the mortars, but as can be observed in carbonation, it is not possible to measure its advance as it does not present any reaction to phenolphthalein (Figure 3).

Figure 3. Carbonation Depth. 

Therefore, it can be said that the M1 mortar shows a good resistance to the influx of CO₂, as this test has been shown to have a good correlation with other durability tests as good as the C3 commercial mortar. MR and M1 presented values classified as low permeability (0.01-0.1) while M2 presented moderate permeability coefficients (0.1-1.0), the rest of the classifications are shown in table 5. The results obtained through the quick permeability method have shown a good correlation with the permeability tests to water, chlorides, and carbonation (^{Ebensperger, 2010}).

The carbonation depth gives an indication of how the advance of CO₂ reaches the steel and causes a significant decrease in the pH of the mortar; with the decrease of pH, depassivation of the layer begins, which starts the corrosion of the steel. The coating for the steel is of 7 mm, the behavior of the repair mortars versus the advancement of carbonation can be seen in Figure 3. The mortars that showed more advancement in carbonation are M2 and C2, whereas M1, C1, and C3 mortars are the ones that show a lesser depth of carbonation; these results are related to the permeability of each mortar, their CaO content, and with the presence of some polymers which are frequently present in commercial products. The CaO content can be observed in Table 2, said content is lower for FA and its content in the mixtures in the mortars decreases with the increase of the substitution. It is important to mention that for the C1 mortar it was not possible to obtain the measurements as there was no reaction with the phenolphthalein indicator. The carbonation of M2 was higher due to the ingress being higher in relation to the short face of the specimens, thus influencing the final average.

Figure 4 shows the corrosion potentials (Ecorr) in regard to the Saturated Salomel Electrode (SCE). Mortars M2, C1, and C2 show a high probability that there could be corrosion on the steel starting from 98 days, as per what is presented in Table 6, and in accordance with the ASTM C 876 standard. The mortars that present a lower probability of corrosion are MR, M1, and C3 for which the behavior was very similar.

Figure 4. Ecorr evolution during the time of exposure. 

The evolution of i_corr during the carbonation process is shown in Figure 5. From this, it is possible to say that, at the start, the values of i_corr at first age show passivity in the reinforced steel (lower than 0.1 μA/cm²) in all mortars. However, as Ecorr indicates, for the steel within mortars M2, C1, and C2 depassivation occurs after 42 days.

Figure 5. Corrosion rate (i_corr) in reinforcement Steel for each repair mortar. 

The behavior of M2 and C2 mortars can be confirmed due to its low mechanical behavior with greater porosity and a greater carbonation depth. While for mortar C1, despite having excellent mechanical properties due to the use of fibers, it is very permeable and allows access of CO₂ to the steel, which creates high corrosion values.

As previously mentioned, mortar C1 did not present a reaction with the phenolphthalein indicator that allowed for a correct read of the carbonation.

4. CONCLUSIONS

In terms of performance and durability, the mortars that showed the best behavior are M1 and C3. The M1 mortar with a substitution of FA of 20%p/p surpasses the behavior of the reference mortar (MR) with regard to the attacks of CO₂, even showing a better behavior than the commercial mortars C1 and C2.

Regarding corrosion, only commercial mortar C3 shows good durability properties. If the use of traditional mortars with good durability properties is implemented, then it is not necessary to use a commercial mortar with a higher cost due to the additions and modifications that could be unnecessary for some repairs.

The durability tests carried out determine, with greater certainty, the protection given to a repair in terms of the corrosion on the reinforcement steel. These tests, in addition to the tests for resistance to compression, flexure and adherence, leads us to think that a traditional mortar (cement-sand with mineral substitutions) could behave as an effective and durable repair.