Efficiency of generic and proprietary inhibitors in mitigating Corrosion of Carbon Steel in Chloride-Sulfate Environments

The efficiency of generic and proprietary corrosion inhibitors (based on nitrite, amine carboxylate or amino alcohol) in corrosion mitigation of carbon steel, which is exposed to concrete solutions with different amounts of chloride as well as sulfate, was studied. The corrosion protection provided by the selected corrosion inhibitors was investigated by performing a potentiodynamic polarization study. In addition, the surface morphological properties of carbon steel samples exposed to the electrolyte mixed with or without inhibitors was also evaluated by scanning electron microscopy. The potentiodynamic polarization measurements showed that the evaluated inhibitors decreased the corrosion current density by 1.6 to 6.7 times depending on the type of inhibitor and the level of sulfate concentration in the electrolyte. The performance of inhibitors based on nitrite was better than that of inhibitors based on amine carboxylate or amino alcohol. The possible mechanisms of the inhibition in the chloride plus sulfate environments are also elucidated.

In an earlier study, the corrosion rate of reinforcing steel was measured in solutions simulating electrolytic chloride environments in the presence of NaNO 2 23 . It was reported that the presence of NaNO 2 significantly decreased the corrosion rate at low chloride concentrations, although its efficiency decreased as the pH was decreased 23 . A polarization investigation was performed to evaluate the corrosion of mild steel in SCPS prepared with varying types of water 24 . It was reported that the corrosion resistance increased in the following order: Rainwater > Well water > Seawater. The effect of benzotriazole and four other benzotriazole derivatives on the corrosion resistance of the steel in SCPS was evaluated 25 . It was reported that the pitting potential decreased due to the selected protection systems and the selected inhibitors that provided a good level of protection to the steel in SCPS 25 .
A literature review showed that the efficiency of chemical inhibitors has been mainly evaluated for chloride-contaminated concrete 26 . However, their efficiency in environments contaminated with both sulfate and chloride salts and under high exposure temperature has not been adequately evaluated 26 . Accordingly, the reported work was performed to evaluate the efficacy of selected inhibitors, i.e. generic and proprietary, in the presence of sulfate and chloride ions at elevated temperature. The mechanisms of inhibition have also been proposed.

Experimental
Corrosion inhibitors. The corrosion inhibitors were selected based on the functional group (nitrite, amine carboxylate and amino alcohol) to assess their efficiency in chloride and sulfate environments. Five (generic and proprietary) corrosion inhibitors were evaluated. A complete presentation of the investigated inhibitors is listed in Table 1, which highlights the designation of the inhibitors properties such as dosage, colour, density and pH 27,28 . In this study, the inhibitors performance was evaluated under the simultaneous presence of sulfate and chloride concentrations at elevated temperature.

Preparation of Carbon Steel Specimens.
Carbon steel conforming to ASTM A706 M was used in the preparation of specimens for the reported electrochemical studies. These type of low-alloy steel deformed bars (Grade 420) are specified for applications where extensive welding of reinforcement or controlled ductility for reinforced concrete structures is required. The specimens were prepared from carbon steel bars obtained from a local steel plant (Saudi Iron and Steel Company). The chemical composition of the carbon steel is shown in Table 2. Test specimens of 16 mm in diameter and 28 mm high were prepared for the electrochemical and morphological studies. The carbon steel samples were cleaned with sandpapers followed by cleaning by acetone. Epoxy coating was used to coat the ends of the specimens to obtain an exposed surface area of 14.6 cm 2 . Figure 1 is a scheme representing the steel sample. A 6 mm diameter hole was drilled in the top surface of the carbon steel sample in order to fit a thin stainless steel rod to hold the specimen in the corrosion cell.
The SCPS was prepared based on the chemical composition of the concrete pore solution 26 by adding 14.0 g of potassium hydroxide, KOH, 10.0 g of sodium hydroxide, NaOH, and 2.0 g of calcium hydroxide, Ca(OH) 2 , to 0.974 L of distilled water. Reagent grade chemicals and deionized water were utilized in the preparation of the SCPS. Also, the pH of the prepared SCPS was maintained at more than 13.4. Potentiodynamic polarization. The carbon steel specimens were placed in a corrosion cell containing SCPS maintained at 40 °C (the selected temperature simulates the average ambient summer temperature in the hot regions of the world) and incorporating the selected concentration of chloride and sulfate ions and with  or without the corrosion inhibitor. The steel specimen was immersed in SCPS, with or without the addition of the corrosion inhibitor, for 30 minutes before each experiment. Thereafter, potentiodynamic polarization (PDP) measurements were conducted using a corrosion measurement system ( Fig. 2) that consisted of a computerized Potentiostat/Galvanostat (ACM equipment), magnetic stirrer, corrosion cell consisting of electrolyte, working electrode (steel specimen), counter electrode (platinum rod with a 36 cm 2 surface area) and a reference electrode [saturated calomel electrode (SCE), Hg/Hg 2 Cl 2 , 3.0 M KCl], and a digital thermometer for measuring the temperature of the electrolyte. The potentiodynamic polarization (PDP) is commonly used technique to assess the mechanistic and kinetic information on corrosion of metals 29 . It measures the variation in current for the potential sweep in the cathode side as well as anode side of the potential of corrosion 29 . The PDP plots were developed by varying the electrode potential between −900 and +900 mV SCE at 15 mV/min scan rate 29,30 . At this scan rate, it takes about two hours to conduct a potentiodynamic polarization scan.
Morphology. The steel specimens were positioned in the testing set up and potentiodynamic polarization scan was conducted. Thereafter, the surface of the samples was assessed by a Joel JSM-5800LV scanning electron microscopy.

Results and Discussion
Assessment of Potentiodynamic Polarization. The effect of chloride and/or sulfate ions on the corrosion of carbon steel specimens positioned in SCPS with the presence or absence of inhibitors is discussed in the next subsections.   the anodic dissolution of carbon steel samples exposed to 0 and 500 ppm sulfate could be observed. However, the PDP for the carbon steel specimen in SCPS with 2000 ppm sulfate indicates an anodic behaviour compared to the specimens exposed to 0 and 500 ppm sulfate concentration. This behaviour indicates that sulfate ions influence the mechanisms of chloride-induced corrosion of carbon steel. The polarization results for the samples exposed into SCPS with no use of any inhibitor is presented in Table 3. There was an increase in corrosion current density from 1.09 to 1.55 μA/cm 2 with the change of sulfate concentration from 0 to 2000 ppm. Figure 4 presents the PDP for carbon steel samples exposed to SCPS with the incorporation of inhibitor I and contaminated with 1000 ppm Cl plus 0, 500 or 2000 ppm SO 4 . Almost similar corrosion was observed in the samples. The corrosion potential decreased from −366 to −555 mV SCE with an increase in the sulfate concentration from 0 to 2000 ppm. The decrease in the corrosion potential indicates an enhancement in the inhibition performance. The anodic dissolution of the carbon steel specimen exposed to 2000 ppm sulfate, which was more anodic in the absence of an inhibitor, became less anodic when inhibitor I was used, thereby indicating that this inhibitor was able to mitigate the corrosion of the carbon steel sample exposed to both chloride and chloride plus sulfate solutions.   Table 3. Potentiodynamic polarization results for carbon steel inside SCPS; no use of corrosion inhibitor (control). The potentiodynamic data for the carbon steel sample inside SCPS with inhibitor I are listed in Table 4. There was an increment in the corrosion current density between 0.14 to 0.31 µA/cm 2 due to the change in the sulfate level from 0 to 2000 ppm.

Corrosion of carbon steel in SCPS and calcium nitrite-based proprietary inhibitor I.
The inhibition effectiveness η E (%)of the investigated inhibitors was computed as: where, i is the corrosion current density without the use of inhibitor; and i o is the corrosion current density when an inhibitor was used.
Corrosion of carbon steel exposed to SCPS with Calcium nitrite-based inhibitor II. The PDP plots for carbon steel samples in SCPS with inhibitor II are presented in Fig. 5. Almost similar corrosion was shown in the samples exposed to both systems, i.e. chloride or chloride plus sulfate ions. The corrosion potential decreased from −419 to −555 mV SCE with an increase in the SO 4 concentration from 0 to 2000 ppm. Further, the current required for the transition from the anodic to the cathodic regions was less in the specimens that were exposed to 2000 ppm SO 4 than those exposed to 500 ppm SO 4 (0.061 and 0.069 µA/cm 2 , respectively). A lower transition current is indicative of increased corrosion activity due to an increase in the sulfate concentration.
The PDP results for carbon steel samples exposed to SCPS with inhibitor II are listed in Table 5. There was an increment in the current density of corrosion from 0.17 to 0.44 µA/cm 2 with increasing the sulfate from 0 to 2000 ppm. The efficiency of this inhibitor decreased between 84% to 72% with changing the concentration of sulfate between 0 and 2000 ppm.
Corrosion of carbon steel exposed to SCPS with Proprietary amine carboxylate-based inhibitor III.
The PDP plots for carbon steel samples immersed in SCPS with the incorporation of inhibitor III (Proprietary Amine Carboxylate-based) are shown in Fig. 6. Again, almost the same corrosion was observed in the samples. The corrosion potential decreased from −334 to −402 mV SCE as the sulfate concentration increased from 0 to 2000 ppm. Further, the current density required for the transition from the cathodic to the anodic region varied from 0.079 to 0.043 μA/cm 2 due to changing the SO 4 from 500 to 2000 ppm. The polarization measurements for this set of specimens are listed in Table 6. There was an increase in the current density between 0.17 to 0.47 µA/cm 2 with the increase the amount of sulfate from 0 to 2000 ppm. The effectiveness of inhibitor III reduced from 85% to 70% as the sulfate concentration increased from 0 to 2000 ppm. It is apparent that the efficiency of this inhibitor decreases due to an increase in the sulfate concentration; although it should be noted that this inhibitor is effective in reducing the current density of corrosion.
Corrosion of carbon steel exposed to SCPS and proprietary modified amino alcohol-based inhibitor IV. Figure 7 depicts the PDPs for carbon steel samples exposed to SCPS with the incorporation of inhibitor IV and contaminated with 1000 ppm Cl and 0, 500 and 2000 ppm SO 4 . As shown in Table 7, there was an increment in the corrosion current density from 0.25 to 1.12 μA/cm 2 with the change in sulfate concentration between 0 to 2000 ppm. Further, the inhibitor efficiency decreased sharply from 78% to 28% as the sulfate concentration changed from 0 to 2000 ppm. Despite its superior performance in the chloride environment, this inhibitor does not perform well in the chloride plus sulfate environment.   Corrosion of carbon steel exposed to SCPS and inhibitor V (Proprietary calcium nitrite-based inhibitor). Figure Table 8. There was a change in I corr from 0.16 to 0.33 μA/cm 2 with changing the level of SO 4 from 0 to 2000 ppm. The efficiency of this inhibitor decreased marginally from 86% to 79% due to an increase in the sulfate concentration. It is apparent from the PDP data that all the investigated inhibitors were efficient in reducing the I corr on the carbon steel samples in SCPS with 1000 ppm Cl − by around four to eight times. However, when the sulfate ions were added to the SCPS, inhibitors I, II, III, and V were effective in decreasing the I corr by 3.3 to 6.7 times (depending on the sulfate concentration). In the case of inhibitor IV, the I corr decreased by only 1.4 to 1.6 times (again depending on the sulfate concentration). This indicates that this inhibitor was marginally effective in mitigating corrosion in the combined presence of sulfate and chloride ions.
Generally, an increase in the sulfate concentration from 0 to 2000 ppm significantly increased the anodic dissolution of carbon steel (see Fig. 3 through 8); indicating that the sulfate or chloride tends to significantly accelerate the rate of corrosion. There was an increment in the current density of the steel in SCPS from 1.3 to 3.5 times with increment in the sulfate level from 0 to 2000 ppm (see Tables 4 through 8). However, the incorporation of the selected inhibitors (organic and inorganic) decreased the corrosion rate of carbon steel. Figure 9 depicts the PDPs for carbon steel samples exposed to SCPS with and without corrosion inhibitors and contaminated with 1000 ppm Cl and 2000 ppm SO 4 . Same corrosion was observed in the tested samples. There was a significant and uniform rise in the anodic dissolution for the control specimens (carbon steel samples exposed to SCPS without corrosion inhibitor and with 1000 ppm Cl and 2000 ppm SO 4 ). However, the PDP for the carbon steel specimen exposed to SCPS, incorporating corrosion inhibitors and contaminated with 1000 ppm Cl and 2000 ppm SO 4, was less anodic than on the control specimens. Further, the current required for the transition from the anodic to the cathodic regions was less in the control sample than on the samples that were    Table 8. Potentiodynamic polarization results for carbon steel immersed in SCPS with incorporation of calcium nitrite based inhibitor V (Proprietary liquid concrete mixture; Dosage of 15 L/m 3 ).
exposed to SCPS incorporating corrosion inhibitors. Also, a lower transition current is indicative of increased corrosion activity. The results of the PDP measurements, as well as the comparison plots depicted in Fig. 9, revealed that the effectiveness of corrosion inhibition of the studied inhibitors was in the order: inhibitor -I > inhibitor -V > inhibitor -III > inhibitor -II > inhibitor -IV.   Morphology of carbon steel specimens immersed in SPCS with chloride and sulfate. Figure 10 shows the morphology of the carbon steel specimen immersed in SCPS with no corrosion inhibitor and with 1000 ppm Cl plus 2000 ppm sulfate. Uniform corrosion was observed of the specimen. In the presence of inhibitors,   marginal localized corrosion and a good protective film were noted on the surface of the specimens (Figs 11-15). The morphology of the steel specimens indicates that the investigated inhibitors were effective in reducing the corrosion of carbon steel. The surface of the carbon steel samples in SCPS with the incorporation of the selected inhibitors was more adherent and thinner than that on the control specimen, i.e. without an inhibitor. Further, the corrosion product on the control specimen, i.e., without an inhibitor, was loose and less adherent.

Mechanisms of inhibition.
Thermo Scientific Nicolet iS10 Smart iTR Fourier transform infrared spectroscopy was utilized to obtain IR spectra (Fig. 16) of the specimens to observe the iron oxide formed on the corroded steel specimens. Formation of iron oxide was observed on the metal surface on the samples exposed to SCPS with no inhibitor and with inhibitor II and inhibitor IV. The high intensity of the peaks assigned to Fe-O bond, between 450 to 650 cm −1 in the spectra of the control specimen and with inhibitor IV, indicates the increased formation of iron oxides compared to the samples exposed to SCPS incorporating inhibitors I or V. The FTIR results (Fig. 16) support the PDP and SEM observations. The data on inhibition efficiency indicate that the nitrite-based inhibitors performed better than the inhibitor based on amine carboxylate which performed better than the amino alcohol based inhibitor. The observed trend can be chemically explained as illustrated in Fig. 17. The nitrite-based inhibitors form an inhibition layer surrounding carbon steel 28 . As illustrated in Fig. 17(i), nitrogen (N) may interact with the carbon steel while the oxygen (O) forms a negative layer on the outer surface which at the end leads to form a protective film. To have homogeneous monolayer protective film formation on the steel surface acting as a barrier coming from mass and charge transfer, the adsorption relies on the interaction between particle and electrode which occurs from the specimen (B) by the action of the unshared electrons pair of the nitrogen atom. Another particle-electrode interaction should be considered coming from the strong negative charge of oxygen [on the specimens (B) and (C)] as the more electronegative constituent of the inorganic compound (calcium nitrite). Thus, the negative surface can act in the repulsion of the chloride and sulfate ions because of their negative charge.
On the other hand, it can be hypothesized that the amine carboxylate based inhibitors, Fig. 17(ii), could form a film on the steel surface by interacting nitrogen with iron atoms while an outer negative surface can be formed by carboxylates. Thus, building up a repulsion force between both the chloride in the environment or media and the carboxylate. While this force is strong at low concentrations, the repulsion force became weak at high amounts of chloride and sulfate. Thus, the decrease in efficiency of the film can be attributed to the possible penetration of chloride or sulfate to the steel.
On the other side, the lower efficiency of the amino alcohol based inhibitor can be explained by having only one hydroxyl group which forms a lower negative charge on the carbon steel, thus, forming a weak film on the steel, as schematically presented in Fig. 17(iii), compared with the nitrite-and carboxylate-based inhibitors.

Conclusions
The effectiveness of all the investigated corrosion inhibitors decreased marginally with an increase in the sulfate concentration from 0 to 2000 ppm. While the effectiveness of amino alcohol-based inhibitor IV, in chloride plus sulfate, decreased sharply with changing the concentration of sulfate from 0 to 2000 ppm. Thus, this inhibitor IV does not perform well in the chloride plus sulfate environment. It can be concluded that the inhibiting effect of the nitrite-based inhibitors, under chloride and sulfate environments, was superior to that of the amine carboxylateand amino alcohol-based inhibitors, which can be ascribed to the varying chemical structure of the functional groups. The nitrite-based inhibitors form a protective layer surrounding the carbon steel which promotes the repulsion of the chloride and sulfate. This film is stronger than the film formed in the case of organic inhibitors based on amine carboxylate and amino alcohol due to the higher negative charge. A loose and non-adherent corrosion product was observed on the samples in SCPS only with no inhibitor, while a thin layer of a well adherent corrosion product was noted on the steel specimens which incorporated the investigated inhibitors. Thus, the incorporation of the inhibitors not only decreases the rate of corrosion, but it also produces a more adherent corrosion product that is beneficial in inhibiting further corrosion.