*3.1. E*ff*ect of Pressure and Temperature*

#### 3.1.1. Gravimetric Experiments

Figure 2 and Table S1 show the corrosion rate and the variation of the inhibition efficiency obtained at different concentrations of GA and CO2 partial pressures. It follows from the data that *CR* increases with an increase of CO2 partial pressure, going from 1.28 to 10.95 mm y−<sup>1</sup> at PCO2 = 1 bar and PCO2 = 40 bar at 25 ◦C, respectively.

**Figure 2.** Corrosion inhibitor efficiency at different concentrations of gum arabic (GA) and CO2 partial pressures after 24 h of immersion.

The solubility of CO2 in water increases sharply with increasing the pressure of the system [31]. The high corrosion rate observed at higher CO2 partial pressures can be explained with the increase of

the acidity of the solution. In fact, in the presence of CO2, the weak carbonic acid is formed, which in turn dissociates in HCO− <sup>3</sup> and in CO2<sup>−</sup> <sup>3</sup> , according to the following reactions:

$$\rm{CO}\_2 + \rm{H}\_2\rm{O} \leftrightarrow \rm{H}\_2\rm{CO}\_3\tag{5}$$

$$\rm H\_2CO\_3 \leftrightarrow H^+ + \rm HCO\_3^- \tag{6}$$

$$\rm{HCO}\_3^- \leftrightarrow \rm{H}^+ + \rm{CO}\_3^{2-} \tag{7}$$

The corrosion process in a CO2 containing solution is controlled by the anodic reaction (Equation (8)) and the three cathodic reactions (Equation (9)–(11)) [4,5]:

$$\text{Fe} \rightarrow \text{Fe}^{2+} + 2\text{e}^- \tag{8}$$

$$2\text{H}\_2\text{CO}\_3 + 2\text{e}^- \rightarrow \text{H}\_2 + 2\text{HCO}\_3^- \tag{9}$$

$$2\text{HCO}\_3^- + 2\text{e}^- \rightarrow \text{H}\_2 + 2\text{CO}\_3^{2-} \tag{10}$$

$$\text{H}^+ + 2\text{e}^- \rightarrow \text{H}\_2\tag{11}$$

The pH of the solution plays an important role in determining the corrosion rate of carbon steel in a CO2 environment. As the CO2 partial pressure increases, its solubility also increases, resulting in an increase of the carbonic acid concentration in the solution (Equation (5)). Nesic' predicted that the concentrations of H2CO3 in the solution would increase of about 40 times with changing the pressure from PCO2 = 1 bar to PCO2 = 40 bar [31]. Increasing the concentration of carbonic acid leads to an increase in the rate of reduction of carbonic acid and bicarbonate ions (Equations (9) and (10)), and ultimately the anodic dissolution of the steel (Equation (8)) as reported by several studies [4,5,31].

After the addition of the inhibitor, it can be seen that the corrosion rate of the metal is greatly reduced going from 1.28 to 0.37 mm y<sup>−</sup>1, with a maximum corrosion inhibition efficiency found to be 71.09% at PCO2 = 1 bar, after 24 h of immersion. The data shows that in contrast to the uninhibited solution, an increase in CO2 partial pressure has a favorable effect on the corrosion rate of the metal in the presence of the inhibitor. It follows from Figure 2 that *IE*, which varies inversely with *CR,* significantly increased after the addition of GA and with CO2 partial pressure, with a maximum corrosion inhibition efficiency of 78.77% and 84.53% at PCO2 = 20 bar and PCO2 = 40 bar, after 24 h of immersion, respectively [4].

The literature reports that GA [7,21], and in general polysaccharides-like inhibitors [15,20], is mainly adsorbed on the metal surface in acidic condition by weak electrostatic interaction between the protonated inhibitor molecules and the chloride ions adsorbed on the metal surface. In a weak acid solution GA molecules are in equilibrium with their protonated molecules according to the following reaction (see also Section 3.5.1) [7]:

$$\rm{GA} + \rm{xH}^+ \leftrightarrow \rm{[GAH}\_x]^{\chi+}\_{\rm{(sol)}} \tag{12}$$

where [GAHx] x+ (sol) is the protonated inhibitor in the solution. As mentioned before, an increase in CO2 partial pressure leads to an increase in the acidity of the solution [32]. The higher value of *IE* observed at higher CO2 partial pressures can be ascribed to the higher concentration of H<sup>+</sup> ions present in the solution, which in turn leads to an increase in the number of protonated inhibitor molecules that can be adsorbed on the metal surface. Moreover, Figure 2 also reveals that *IE* varies with the concentration of the inhibitor until the system reached a state (e.g., 1.0 g L−<sup>1</sup> of GA), in which it can be said that the inhibitor molecules are in equilibrium with their protonated counterpart. For further increase in GA concentration, *IE* remains almost stable. The results clearly demonstrate that GA has greatly reduced the *CR* of the metal in the tested environment, and the high corrosion inhibition activity of GA was influenced by both its concentration and CO2 partial pressure. The lower values of *CR* observed in the

presence of the inhibitor can be ascribed to its adsorption on the metal surface, covering the metal surface and thereby, blocking the active corrosion sites on its surface [4,7,28]. The gravimetric results are also supported by the SEM analysis presented from Figures 7–9, where it can be seen that the surface coverage increases and the protective layer becomes more compact in the presence of GA and with increasing CO2 partial pressure.

As the temperature rises, *IE* slightly decreased. This decrease may be due to the combination of two different reasons. For instance, the solubility of CO2 decreases with increasing the temperature of the solution [31], which can lead to a less acid environment. The pH of the solution increases slightly and therefore shifting the equilibrium reaction Equation (12) to the left. At higher pH, the concentration of H<sup>+</sup> ions in the solution is smaller, which would result in the formation of fewer protonated inhibitor molecules available for the absorption process. Another possible reason may be due to the fact that these types of inhibitors get absorbed via electrostatic interactions (e.g., van der Waal forces) onto the surface of the metal, and it is known that this types of interaction generally grow weaker with an increase in temperature due to larger thermal motion [3,20]. Consequently, an increase in temperature will increase the metal surface kinetic energy, which has a detrimental effect on the adsorption process and encourages desorption processes [15,20].

Table S2 lists the inhibition efficiency of various corrosion inhibitors used to mitigate sweet corrosion obtained at different immersion times and temperatures. It is worth mentioning that most of these inhibitors are labeled either as toxic or are expensive to synthesize. Umoren et al. [15] reported the corrosion inhibition efficiency of a commercial inhibitor to be 87 and 88% at 25 and 60 ◦C, respectively after 24 h of immersion. The table shows that GA, compared to other studied corrosion inhibitors, and the commercial corrosion inhibitor, can be considered a good environmentally friendly corrosion inhibitor for carbon steel in a CO2-saturated saline solution. Moreover, since GA is already used as a thickening agent in the make-up of the fracturing fluid, can also work as an active component in corrosion inhibitor in the shale gas industry.

#### 3.1.2. Electrochemical Experiments

The electrochemical experiments such as electrochemical impedance spectroscopy (EIS) and potentiodynamic polarization (PDP) were also employed as a means to support the gravimetric findings. These experiments were carried out at 1.0 g L−<sup>1</sup> of GA at different CO2 partial pressures after 24 of immersion. 1.0 g L−<sup>1</sup> is the concentration in which the tested inhibitor exhibited a maximum in the concentration-efficiency curve.

The EIS measurements were used to evaluate the resistance of the protective layer from the electrochemical angle and are presented in Figure 3. It can be seen from the Bode (Figure 3b) and phase angle plots (Figure 3c) that the system is characterized by two-time constants at low (LF) and high frequencies (HF). The presence of these two-time constants suggests that the electrochemical reaction process of the N80 carbon steel in a CO2 saturated saline solution is affected by two state variables i.e., the corrosion products layer and/or the protective adsorptive layer, and electric double-layer, as also reported by Dong et al. [33]. For this reason, the EIS plots presented in Figure 3 were fitted with the help of the equivalent circuit (EC) presented in Figure 3d, consisting of the following elements: *R*s is the electrolyte resistance. *CPE*<sup>l</sup> and *R*<sup>l</sup> are the constant phase element and the resistance of the layer formed on the metals surface, respectively. *CPE*dl and *R*ct are the constant phase element representing the double-charge layer capacitance and the charge transfer resistance, respectively. The EIS parameters are listed in Table 1 and from the small values of χ<sup>2</sup> (i.e., the goodness of fit) it can be said that the EC used to fit the system under investigation was the most appropriate one.

**Figure 3.** EIS plot recorded in the presence and absence of1gL−<sup>1</sup> of GA after 24 h of immersion at different CO2-partial pressures. (**a**) Nyquist; (**b**) Bode; (**c**) phase angle; (**d**) equivalent circuit

**Table 1.** Electrochemical impedance parameters with and without the presence of 1.0 g L−<sup>1</sup> concentrations of GA after 24 h of immersion.


The presence of a time constant at HF is reported in several studies [34,35] and it is often observed in a Fe/water system. This time constant may be due to the capacity of a porous thin layer formed onto the metal surface. In this study, and without the inhibitor, the presence of this time constant at HF is due to the formation of a thin layer of Fe3C onto the metal surface. As mentioned before, the microstructure of the tested carbon steel is composed of circa 41% of a ferritic phase and the remaining of a perlitic phase (Figure 1). The ferritic phase is more active than the Fe3C contained in the perlitic phase [7], in this case, the former phase will act as an anode and the latter one as a cathode. This will generate a micro-galvanic effect, which will eventually lead to the formation of a thin layer of Fe3C onto the metal surface. However, it follows from the data that both the values of *R*<sup>f</sup> and *R*ct greatly increased in the presence of the inhibitor, which indicated that the GA molecules were adsorbed onto the metal surface leading to the formation of a protective layer that covers the surface, as confirmed also from the morphological analysis (e.g., SEM-EDS and XPS). Moreover, the difference between these two values obtained in the absence and the presence of GA increased even more with increasing CO2 partial pressure, suggesting that this protective layer becomes more stable and compact, with

the corrosion inhibition efficiency going from 69.83% up to 87.44% at PCO2 = 1 bar and PCO2 = 40 bar, respectively. The increase in *IE* observed with an increase in CO2 partial pressure agrees with the results obtained with the gravimetric measurements and is in agreement with the ones reported in the literature [4]. It is evident that the addition of GA had a remarkable effect on the corrosion process of the metal and that its inhibition not only depends on the concentration of GA but also from CO2 partial pressure. The results show that the coverage and thickness of the formed protective layer increased with CO2 partial pressure, acting both as a barrier against the charge and the mass transfer processes that occur onto the metal surface owing to the corrosive attack of the aggressive electrolyte.

Figure 4 and Table 2 show the potentiodynamic polarization measurements and the corrosion kinetic parameters obtained from the polarization plots in the presence and absence of GA at different CO2 partial pressures, respectively. As can be seen from Figure 4, the anodic polarization curve of the blank solution does not show the typical Tafel behavior consequently, the corrosion current densities were calculated from the extrapolation of the cathodic Tafel region.

**Figure 4.** Potentiodynamic polarization parameters obtained in the absence and presence of 1.0 g L−<sup>1</sup> of GA at different CO2-partial pressures, after 24 h of immersion. (**a**) PCO2 = 1 bar, (**b**) PCO2 = 20 bar and (**c**) PCO2 = 40 bar.

**Table 2.** Potentiodynamic polarization parameters obtained after 24 h of immersion without and with 1.0 g L−<sup>1</sup> of GA.


The data shows that in absence of GA, the corrosion current density of the steel increased with an increase in CO2 partial pressure, which is linked with the increased acidity of the solution, in agreement with the gravimetric experiments. However, it is evident from the data that the corrosion current density of the steel was prominently reduced after the addition of GA to the solution. Furthermore, both the cathodic and anodic curves of the polarization curves were shifted towards lower current densities after the addition of GA. The result suggests that the inhibitor impeded both the rate of the anodic dissolution (Equation (8)) and the cathodic reactions (Equations (9)–(11)), by either covering part of the metal surface and/or blocking the active corrosion sites on the steel surface. The dominant cathodic reaction depends on the pH value of the solution. At lower pH (e.g., less than 4) the reduction of H<sup>+</sup> ions would be the dominant cathodic reaction (Equation (11)). At pH > 4 the dominant cathodic reaction will be the reduction of HCO− <sup>3</sup> ions and H2CO3 (Equations (9) and (10)). At higher values of CO2 partial pressure, GA suppresses the Equation (11) (e.g., the pH of the solution is circa 3.5 at PCO2 = 40 bar), through the formation of H-bonding between the hydroxyl groups of the inhibitor units and the H<sup>+</sup> ions, adsorbed onto the steel surface, as discussed in more detail in Section 3.5.2.

Moreover, after the addition of GA, the *E*corr can be seen to shift with no definite trend toward both the anodic and cathodic regions. This result suggested that GA behaves as a mixed type inhibitor as also reported by other studies for this inhibitor [7,21,27,28].
