An efficient green ionic liquid for the corrosion inhibition of reinforcement steel in neutral and alkaline highly saline simulated concrete pore solutions

The effect of the green ionic liquid compound, Quaternium-32 (Q-32), on the corrosion inhibition performance of reinforcement steel, in a simulated concrete pore solution, was investigated at different temperatures and pH values, using electrochemical impedance spectroscopy (EIS). The inhibition efficiency was improved as the concentration of Q-32 and pH values were increased. However, it decreased as the temperature was raised. A Q-32 concentration of 20 µmol L–1 exhibited a 94% inhibition efficiency at 20 °C. The adsorption isotherm was evaluated using EIS measurements, and it was found to obey the Langmuir isotherm. The surface topography was examined using an atomic force microscope and scanning electron microscope. The effect of the Q-32 concentration with the highest corrosion efficiency on the mechanical properties of the mortars was also explained by flexure and compression techniques.

Electrochemical measurements. Electrochemical measurements were performed in a three-electrode double-jacketed cell. Carbon steel (C-steel) with an exposed area of 0.5 cm 2 and a graphite rod was used as working and auxiliary electrodes, respectively. An Ag/AgCl electrode was employed as a reference electrode. The reference electrode is coupled with a Luggin capillary to minimize the potential drop between the electrodes. A Julabo F12 thermostat (GmbH, Seelbach, Germany) was utilized to control the temperature of the solutions, as the electrochemical tests were carried out at various temperatures (30, 40, and 50 °C) in the presence and absence of the inhibitor to evaluate the inhibitor efficiency at elevated temperatures. The C-steel was immersed in the test solution for 30 min before each electrochemical test to achieve a steady-state condition. The EIS analyses were performed under an open circuit potential (OCP) condition at a frequency range of 1 × 10 -2 to 1 × 10 5 Hz, with an AC amplitude of 10 mV, using a GAMRY 3000 potentiostat (Gamry, Warminster, PA, USA) [41][42][43] . Echem Analyst Software version 7.8 from Gamry was used to plot and analyze the electrochemical data. For other plotting data, OriginPro 2018 (64-bit) SR1 b9. 51  www.nature.com/scientificreports/ L −1 were tested in three different SCPS solutions with different pH values (7, 9, and 12) to investigate its effectiveness in preventing the corrosion of reinforcement steel under different conditions that may develop with time.
Surface morphological studies. Freshly polished steel coupons were immersed in the different pH SCPS solutions in the presence of 20 μmol L −1 of Q-32 for 24 h at a temperature of 20 °C. After that, the coupons were removed, rinsed with deionized water, air dried, then characterized using a high field emission scanning electron microscopy, coupled with an energy dispersive x-ray unit (FEI NOVASEM 450, Hillsboro, OR, USA) in addition to atomic force microscopy (AFM) analysis using an MFP-3D (Asylum Research, Santa Barbara, CA, USA) piece of equipment.
Mechanical characterization. The effect of using Q-32 on the cured mortar quality was investigated after different exposure times by measuring the changes in compression and flexural strengths, before and after the addition of 20 μmol L −1 of the corrosion inhibitor. The compressive experiments were carried out using a 300 KN Tecnotest 3 compression testing machine (Modena, Italy). The flexural strength experiments were utilized by a Lloyd LR 50 K universal testing machine (Ametec Inc. USA). Each experiment was repeated three times, and the results were averaged. The mortar was prepared by mechanical mixing in a stainless steel mixer, with ingredients containing one part mass of cement and one and a half part mass of standard sand, with a water-cement ratio of 0.45 44 . A 20 μmol L −1 of the tested corrosion inhibitor was dissolved in the mixing water before applying it into the mortar components. Then, the mold was filled with the mixture under vibration to release air bubbles, and it was stored in a moist atmosphere for 24 h. After that, the demolding of the prepared specimens was conducted, and the specimens were stored under tap water over the test period 45 . The specimens were removed from the water and placed in a drying oven at 60 °C for 24 h before the strength test to avoid the influence of the hydration of the mortar and to increase the strength of the measured specimens 46 .

Results and discussion
Electrochemical impedance measurements. EIS technique has been employed to describe the electrode/electrolyte interfaces quantitatively. It is considered as a robust method that can explain the corrosion behavior and calculate their rates 47,48 . Figure 2 shows two proposed equivalent circuits (ECs) that are utilized to analyze and fit the collected experimental data. Figure 2A exhibits the one-time constant equivalent circuit, which is commonly used for analyzing electrodes undergoing uniform corrosion. Moreover, the two-time constant equivalent circuit, which is primarily used for electrodes with coatings or adsorbed layers on top [49][50][51] , is displayed in Fig. 2B. The parameters of the electrochemical reactions occurring at the metal/solution interface are listed in Table1. They are measured and calculated from the EIS Nyquist and Bode plots. From these plots the electrolyte resistance (R s ), pore resistance (R po ), charge transfer resistance (R ct ), constant phase element for the time constant associated with the pore resistance (CPE po ), constant phase elements for the time constant associated with the charge transfer resistance (CPE ct ), and the deviation parameters (n 1 and n 2 ) from the double-layer capacitance (C dl ) are listed. It is worthy of mentioning that the tests were repeated three times to ensure reproducibility, and the obtained results are the mean value. Additionally, a standard deviation has been tabulated for the crucial parameter such as R ct and C dl . The constant phase element is in place of a pure capacitor as it is composed of the capacitance and deviation parameter to avoid the imperfectness behavior of the ideal double layer, which may occur because of a non-uniform thickness of the corrosion inhibitor layer, non-uniform corrosion reaction on the surface, or non-uniform current distribution and surface roughness 52,53 . The capacitance behavior is mainly attributed to the dielectric nature of the surface film (corrosion product and/or inhibitor film) which affects the corrosion rate of the metal, and it can be expressed by Eq. (1) 54,55 : where Z CPE is the impedance of CPE (Ω cm −2 ), Y o is a proportional factor in s n Ω −1 cm −2 , j = (−1) 1/2 , ω is the angular frequency in rad s −1 , and n is the deviation parameter, and its value is between 0 and 1. When n = 1, the CPE becomes equivalent to an ideal capacitor, and when n = 0, the CPE becomes equivalent to a resistor. Figures 3 and 4 represent the EIS Nyquist and Bode plots , respectively for the reinforcement steel immersed in different pH solutions (7, 9 and 12) at OCP, in the presence of 5, 10, 15 and 20 μmol L −1 of the Q-32 corrosion inhibitor at room temperature, within a frequency range of 1 × 10 −2 to 1 × 10 5 Hz and at an AC amplitude of  www.nature.com/scientificreports/ 10 mV. The measured data are represented by symbols, while the fitted data, using the equivalent circuits shown in Fig. 2, are represented by the solid lines. It is evident in Nyquist plots that the higher the pH, the wider the diameter of the semicircle. Besides, a noticeable depression at intermediate frequencies (not perfect semicircles) due to the electrode surface heterogeneity resulting from the roughness of the surface 56,57 . Furthermore, the width of the phase angle degree curve in the intermediate frequency region of the Bode plot reveals more capacitive response as the capacitive loop diameter increases with the increase in the inhibitor concentration, which means a lower corrosion rate [58][59][60] . Consequently, the charge transfer resistance (R ct ) increases proportionally as the corrosion inhibitor concentration increase which is attributed to increasing the surface coverage θ and the inhibition efficiency IE eis % of the Q-32 corrosion inhibitor as is introduced in Table 1. The corrosion inhibition is noticeable at the low frequency, and its efficiency increased as the Q-32 inhibitor concentration was increased. This is attributed to the adsorption of more Q-32 molecules at the reinforcement steel surface, which enhances the thickness of the protecting layer on the metal/solution interface [61][62][63] . It is worth mentioning that the R ct values diminish as the pH value lessens as the R ct values for the reinforcement steel in the presence of 20 mol L −1 Q-32 are alleviating from 18.2 kΩ cm −2 at pH 12 to 2.1 and 1.2 kΩ cm −2 at pH 9 and 7, respectively. The passivity loss of reinforcement steel could be justified for different reasons. It could be attributed to the increase of the Clions adsorption with attenuating the pH, which initiates a passive layer breakdown, thus leading to a localized attack on the metal surface. Additionally, the Clions would penetrate the protective oxide layer leading to form chloride-contaminated oxides. Moreover, the adsorbed Clions would induce the de-passivation potential of the passive film to a value higher than the critical one 64,65 .
The surface coverage θ and the inhibition efficiency IE eis % of the Q-32 can be assessed using the R ct value from the following relationships 66,67 : where R ct and R o ct are the charge transfer resistance, with and without the Q-32 corrosion inhibitor, respectively. It is obvious that the decrease of C dl values indicates an increase in the area or the thickness of the electrical double layer. This is attributed to the inhibitor molecule adsorbed on the metal surface, which replaces the adsorbed water molecules. Moreover, the model describes the localized breakdown mechanism of the passive www.nature.com/scientificreports/ layer, which takes place because of the competitive adsorption of the Cland OHions, which are inhibited in the presence of Q-32 68,69 . The increase in the charge transfer resistance values may be attributed to either (i) the formed passive film, which is promoted by the presence of the inhibitor molecules that block the active sites on the steel surface, according to Uhlig and Bohni 70,71 , or (ii) the increase in the adsorbed layer thickness/area of the inhibitor, which acts as a physical barrier. The double layer capacitance (C dl ) is explained according to the Helmholtz model, and can be expressed in Eq. (4) 72,73 : where ε 0 is the vacuum permittivity,ε 0 is the local dielectric constant, A is the surface area of the electrode, and d is the thickness of the protective layer. Figures 5 and 6 indicate the measured (symbols) and fitted (solid line) of Nyquist and Bode graphs, respectively, for the reinforcement steel after being immersed in saline SCPS, in the presence of 5, 10, 15 and 20 μmol L −1 of the Q-32 corrosion inhibitor under elevated temperatures within a frequency range of 1 × 10 -2 to 1 × 10 5 Hz at OCP and 10 mV of AC amplitude. Figure 2B demonstrates the two-time constant equivalent circuit, which is deployed for fitting the EIS measurements. It is noticed that for the Nyquist plots, the capacitive loop diameters increase as the inhibitor concentration increases at any temperature. It is also worth mentioning that increasing the temperature reduces the surface coverage as well as the inhibition efficiency. This is attributed to an increase in the desorption rate of the corrosion inhibitor molecules from the reinforcement steel surface, which leads to a surge of the dissolution rate of the electrode surface 19,43 . This indicates that the characteristic features of the EIS measurements do not change as the temperature is altered. A comparison of the IE eis % recorded data in Tables 1 and 2 indicates the increasing corrosion inhibition behavior of reinforcement steel with an increasing Q-32 concentration, where the highest inhibition efficiency reached 93.4% at 20 μmol L −1 at 20 °C and pH 12. This inhibition efficiency decreases as the temperature increases or the pH and/or Q-32 concentration decrease.
Inhibitor adsorption and thermodynamic analysis. The reaction of metal active centers with the corrosion inhibitor molecules occurs via the substitutional replacement process for the electrolyte molecules at the metal/solution interface 74 . The adsorption isotherm models determine the type of reaction, whether it is spontaneously or not, and whether it is physical or chemical interaction from the value of the standard adsorption www.nature.com/scientificreports/ free energy change (ΔG o ). The relation between C inh θ and C inh at (A) different pH and B) at different temperatures, which is a straight line, i.e., it follows the Langmuir adsorption isotherm described by Eq. (5), is shown in Fig. 7 75,76 .  www.nature.com/scientificreports/ where C inh is the inhibitor concentration, K ads is the equilibrium constant of the desorption-adsorption mechanism, and θ is the inhibitor surface coverage. A straight-line relationship, with an R 2 correlation coefficient of 0.98, is shown in Fig. 7. The Langmuir isotherm proposes monolayer adsorption at specific reaction sites on the metal surfaces. Additionally, it is hypothesized that there are no lateral interactions between the adsorbed molecules, and the adsorption is identical and equivalent 75,76 . Meanwhile, the intermolecular interaction between the adsorbed inhibitor molecules, which have donor groups, and the metal surface is not considered in the Langmuir equation, which may cause a small deviation in the calculations 77 . The thermodynamic inhibition mechanism can be utilized to calculate the strength and the type of the adsorption process by calculating the K ads values from the intercepts of the plotted straight lines from Fig. 7B. The standard free energy of the adsorption reaction, G o ads , in kJ mol −1 , can be calculated from with the following equation: where R is the universal gas constant (8.31 J mol −1 K −1 ), T is the absolute temperature, and C is the concentration of water molecules (55.5), expressed in molarity units (M) 78,79 .
The values of K ads and ΔG°a ds are given in Moreover, the Van't Hoff equation can be utilized to calculate the standard heat enthalpy change ∆H o ads by plotting a straight-line graph between ln K ads versus T −1 , as shown in Fig. 8 59 .
where ∆H o ads and ∆S o ads are the standard enthalpy and entropy changes, respectively. It is worth mentioning that the standard entropy changes of the adsorption S 0 ads can be obtained from the intercept of Fig. 8. However, H 0 ads can also be calculated using Eq. (8), with the help of the calculated standard  Thermodynamic activation parameters and inhibition mechanism. To estimate the activation energy (E a ) for the corrosion of reinforcement steel in saline SCP, without and with the presence of different concentrations of the Q-32 corrosion inhibitor, under an elevated temperature of 20 °C to 50 °C, the relation between log (i corr ) and the reciprocal of the temperature (1/T) is plotted to obtain a straight line, as shown in Fig. 9, according to the Arrhenius equation 82,83 : where CR is the corrosion rate of the reinforcement steel, which is expressed in the (i corr ). A is the Arrhenius constant, which varies with the metal type and the electrolyte 84 . The E a values are calculated from the slopes of the plotted lines for the relation between log i corr and 1/T, as shown in Fig. 9, which have a high regression coefficient close to unity, as listed in Table 4. It is shown that the addition of the Q-32 corrosion inhibitor increases the activation energy value, indicating a strong adsorption mechanism on the reinforcement steel surface 85,86 . However, the adsorption of Q-32 molecules on the reinforcement steel surface occurs through both simultaneous chemi/physisorption, as the activation energy parameter refers mainly to the chemical adsorption. This is  www.nature.com/scientificreports/ due to the high competition between the Q-32 molecules and the Claggressive ions and/or the OHions for adsorption on the metal surface 87 . According to the transition state equation 88 , the values of the apparent enthalpy of activation, ΔH a , and entropy of activation, ΔS a , for the reinforcement steel corrosion in SCPS can be calculated from the corrosion rate values (corrosion current density), at different temperatures and in the absence and presence of different concentrations of the Q-32.
where h is the Planck's constant, N A , is the Avogadro's number, and R is the universal gas constant. The plotting of log ( i corr /T ) against 1/T gives a straight-line relation, as shown in Fig. 10. The values of ΔH a and ΔS a are calculated from the slope of the plotted lines and their intercept with the y-axis, respectively, and are tabulated in Table 4. The endothermic nature of the reinforcement steel dissolution reaction is inferred from the positive sign of ΔH a . Increasing the ΔH a values by adding inhibitors means that the dissolution of reinforcement steel becomes more difficult in the presence of the tested inhibitors 89,90 . The positive trend of ΔS a values, referring to the activated complex, is the rate-determining step, which represents an association rather than dissociation, meaning a decrease in the disordering of an activated complex, which is due to reactants 91,92 . Scanning electron microscope analysis. The images in Fig. 11 show the surface morphological examination using the SEM micrograph for the reinforcement steel specimens when they are exposed to different pH media in the presence and absence of 20 µmol L −1 of the Q-32 corrosion inhibitor, after 24 h of immersion at 20 °C. The steel surface, immersed in 3.5% NaCl, without and with the presence of Q-32, is shown in Fig. 11A,D, respectively. The surface of the specimen is severely damaged and roughened due to highly aggressive media. In the presence of the Q-32 molecules, the corrosion features lessen. The effect of adding 20 µmol L −1 of Q-32 to the carbonated SCPS at pH 9 is shown in Fig. 11B,E. The tested sample shows less corroded areas, and a smoother surface can be noticed. Meanwhile, the effect of Q-32 in saline SCPS is shown in Fig. 11C,F, and the protective effect of Q-32 in pH 12, compared to another pH, is notable. The damage of the surface is significantly decreased due to the additional protection obtained by increasing the pH to 12, since the steel surface is passivated at this high pH value.  www.nature.com/scientificreports/ AFM analysis. AFM is considered a robust technique in the morphological investigation of metal surfaces at the nano-to micro-level and mainly in three-dimensions (3-D). Thus, it can be used to evaluate the activity of the corrosion inhibitor via the surface roughness calculation 79 . The 3-D images of the reinforcement steel surface over an area of 25 μm 2 are presented in Fig. 12, where the mean roughness factor (R a ) is measured. The surface topography of reinforcement steel, in the absence and the presence of 20 µmol L −1 of Q-32 at pH7, is shown in Fig. 12A,D. The R a values show that the roughness decreases from 690 to 215 nm in the presence of Q-32 molecules. Meanwhile, the reinforcement steel substrate at pH 9, in the absence and presence of 20 µmol L −1 of Q-32, is shown in Fig. 12B,E. The R a value at pH 9 is within the range of 430 nm, and with the addition of 20 µmol L −1 of Q-32, the R a is reduced to 129 nm. Moreover, at pH 12, the R a value is 223 nm for the reinforcement steel in the absence of Q-32, which is decreased to 52 nm in the presence of 20 µmol L −1 of the Q-32, as shown in Fig. 12C,F.   Table 5, for the flexural and compressive strength of the cured mortar, after 2, 7 14, 21, and 28 days of exposure in tap water, are obtained according to EN-197-1 93,94 . Cured mortar samples with 20 μmol L −1 of Q-32 have a higher flexural and compressive strength, compared to the inhibitor-free ones, for the first 14 days, until it reaches maximum bending stress and breaks at 8.5 MP, after 14 days of the exposure test. Then, the strength remains constant, with no difference due to the presence or the absence of 20 μmol L −1 of Q-32. The SEM images for the fractured cured mortar samples surface in the absence and the presence of 20 µmol L −1 after 28 days of age are presented in Fig. 13. The cured mortar surface topography seems to be similar before and after adding the Q-32 corrosion inhibitor to the mixing components, which could be evidence of un reacting the Q-32 corrosion inhibitor with the cement components. It is worthy of mentioning that the cementitious hydration products are represented by the rough surfaces, and the sand particles give a smooth morphology ends 93,94 .

Conclusions
In this work, the electrochemical results showed that the green ionic liquid, Quaternium-32, is utilized as a corrosion inhibitor for the reinforcement steel in saline simulated concrete pore solution at different pH and temperatures. It is found that the inhibition efficiency IE% increases with the increase of the pH and reaches the optimum condition in saline SCPS (pH 12), with 20 μmol L −1 of Q-32 at 20 °C. It is confirmed that Q-32 is a chemi/physisorbed process, according to the calculated free energy change values, G o ads . SEM and AFM micrographs for the steel samples immersed in inhibitor-free SCPS depict a high surface roughness, compared to those in the inhibited SCPS. Based on flexural and compressive measurements, Q-32 is suitable for addition to concrete mixtures, with no noticeable effect on the mechanical properties on them after curing.