Introduction

Conventional natural resource stocks have been on a downward curve to depletion in the past 10 years, especially with huge increases in demand, meaning that alternatives have had to be found. Shale gas has been proposed as one alternative that may be a dependable source in the future and answers the challenge of meeting energy needs. Hydraulic fracturing (HF) technology has been the reason why shale has become a resource with great potential. In the HF process, a pressurized fluid is injected into the wellbore to induce artificial cracks within the reservoir and then the fluid is pumped out allowing the gas to flow out (Ref 1,2,3). Despite its poor corrosion resistance, carbon steel is broadly used as pipeline material in the shale gas industry infrastructures due mainly to its high strength, durability, ductility and low cost (Ref 4,5,6,7,8). One of the most common problems found in the exploitation of the shale gas reservoirs is the use of corrosive fluids with high concentrations of chlorides containing carbon dioxide (Ref 9, 10). The fluid usually used in the HF process is mainly a neutral water-based chloride solution (up to 4% of potassium chloride) with different additives (i.e., corrosion inhibitors, thickening agents, inhibitors of scaling, etc. (Ref 5). CO2 could result from the microbial degradation of hydrocarbon sources present in the environment (Ref 11). The presence of high chloride concentrations in the fluid can lead to a severe corrosion attack to the steel structure, which may be general or localized in nature. Moreover, CO2 gas dissolves in the fluid to form carbonic acid (H2CO3), which further accelerates the corrosion of the steel. This type of corrosion is known as sweet corrosion. The literature reports that the corrosion rate in chlorides containing fluids increases exponentially in the presence of CO2 (Ref 9). The use of corrosion inhibitors is one of the most cost-effective and practical methods for controlling CO2 corrosion in the oil and gas industry. The typical commercial inhibitors so far used in the oil/gas filed to inhibit CO2 corrosion occurring in internal carbon steel pipelines are nitrogen-based compounds (Ref 12,13,14,15,16). The literature reports that these inhibitors exhibit excellent anti-corrosion properties (Ref 12,13,14,15,16). However, some of these inhibitors were recognized from the Environment European Commission (EEC) as toxic and potentially dangerous to both the environment and human health. Due to these serious threats, the EEC, with the directive 76/464/EEC, has therefore limited the use of such inhibitors. These concerns have led many researchers to focus their studies toward the development of more environmentally friendly corrosion inhibitors for CO2 corrosion of the steel. Over the last decade, many natural and low-risk potential corrosion inhibitors have been investigated for carbon steel and steel alloy in different corrosive environments. Naturally occurring polymers such as guar gum (Ref 8, 17,18,19,20), Raphia Hookeri (Ref 21), xanthan gum (Ref 22), Plantago (Ref 23), sodium carboxymethyl cellulose (Ref 24) were found to be effective corrosion inhibitors in acid solutions. The results reveal that the corrosion inhibition mechanism was attributed to the strong adsorption of the inhibitor molecules on the active sites of the metal surface, thereby hindering the cathodic and/or anodic reactions on the metal surface (Ref 25,26,27,28,29,30,31). It is known that polymers are adsorbed stronger than their monomer analogs. Unlike small molecules, polymers can cover larger surface areas on the metal surface due to their multiple adsorption sites for bonding with the metal surface (Ref 20). Therefore, it is expected that polymers will be better corrosion inhibitors than their monomer analogs (Ref 32). In this regard, researchers seem to have focused their studies more on the use of plant extracts to prevent sweet corrosion (Ref 33,34,35). Hence, the use of natural polymers to prevent sweet corrosion for the steel in saline media is scanty (Ref 36). Umoren et al. (Ref 36) assessed the inhibition effect of natural polysaccharides as chitosan and carboxymethyl cellulose (CMC) for pipeline steel in CO2 saturated saline solution. The results from the electrochemical and surface analysis revealed that these inhibitors reduced the corrosion of the steel to some extent by the formation of a protective layer on the metal surface. Gum arabic (GA) is a water-soluble complex mixture of carbohydrates and glycoprotein, with a widespread application in the food, pharmaceutical and cosmetics industries (Ref 25,26,27,28). GA is also broadly used in the fracturing fluid as a thickening agent (Ref 2). Over the last years, as witnessed by an increasing number of publications, GA is also emerging as a promising candidate as an environmentally friendly corrosion inhibitor of the metal infrastructures (Ref 25,26,27,28,29, 31). In addition, GA being a biodegradable, non-toxic organic compound provides more environmental benefits than the common toxic inorganic inhibitors, with great benefits for the environment (Ref 37, 38). The studies concerning the use of GA as a corrosion inhibitor have been performed, so far, only in concentrated acidic solutions (Ref 25,26,27,28). These studies report that GA proved to be an effective corrosion inhibitor on mild steel and aluminum in both HCl and H2SO4 solutions (Ref 24). After an extended literature survey, we concluded that no work on the inhibition capability of GA in chlorides containing carbon dioxide solution for pipeline carbon steel has been reported. Therefore, the purpose of the present paper is to study the effectiveness of GA as a corrosion inhibitor to mitigate sweet corrosion for pipeline carbon steel in a saline solution. This paper aims to show that GA not only can be used as a thickening agent, but it could also be used as an active component in corrosion inhibitor in the shale gas industry. To this end, the study was carried out using weight loss and electrochemical techniques, e.g., electrochemical impedance spectroscopy (EIS) and potentiodynamic polarization (PDP). To assess the maximum possible inhibition efficiency as a function of the concentration of GA, the experiments were performed in uninhibited and inhibited solutions, at the presence of different inhibitor concentrations (i.e., form 0.2 up to 1.0 g L−1). Moreover, the effect of time (i.e., up to 168 h) and temperature on the inhibition efficiency of GA has also been taken into account. This study also aims to explain the mode of the adoption of GA on the metal; thus, the adsorption isotherm and the thermodynamic parameters (i.e., Ea and Qads) were calculated and discussed. FT-IR analysis was used to gain useful information on the mode of adoption of GA on the metal surface. SEM–EDS measurements were also employed to support the gravimetric and electrochemical results.

Experimental Procedures

Material and Solution

The study was carried out on carbon steel (N80) with composition of (wt.%): C 0.39%, Si 0.26%, Mn 1.80%, Cr 0.04%, V 0.19%, Ni 0.04%, Mo 0.003%, Al 0.03%, Cu 0.26%, Co 0.002%, Sn 0.004%, P 0.001%, S 0.001%, and the remainder Fe. The microstructure of the N80 pipeline steel is composed by cementite (Fe3C) and ferrite (α-Fe), where the ferritic phase accounts for circa 41% as shown in Fig. 1. The samples used in this study were obtained from pipeline carbon steel (N80) and embedded in a PTFE cylinder block, with the flat surface (0.50 cm2). Each time prior to a test, the exposed surface was ground with silicon carbide abrasive paper up to 1200 grit and then was ultrasonically washed in acetone, rinsed with distilled water and dried. All experiments were carried out in 0.5 M of KCl. The aggressive solution was prepared from reagent grade material potassium chloride (Sigma-Aldrich) and pure deionized water with an electrical resistivity of 0.055 µS cm−1 at T = 25 °C. Gum arabic was purchased from Sigma-Aldrich. The concentrations of the inhibitor solution prepared and used for the study ranged from 0.2 to 1.0 g L−1. Prior to each experiment, CO2 was purging in the test solution for 1 h and then the purging was maintained throughout the test to minimize the ingression of air. After saturation, the conductivity and pH of the blank solution were 60.30 mS cm−1 and 4.5, respectively. The addition of the inhibitor had a negligible effect on the conductivity and pH.

Fig. 1
figure 1

Microstructure of the N80 pipeline steel sample

Full size image

Gravimetric Measurement

The gravimetric experiments were carried out suspending the coupons in an unstirred solution (150 mL) with and without of different concentrations of inhibitor (i.e., 0.2 up to 1.0 g L−1) at 25 and 45 °C. The weight loss was determined by retrieving the coupons after 24 h of immersion, by means of an analytical balance with the accuracy of ± 0.1 mg. The effect of time was carried out in the absence and at the concentration where GA exhibits a maximum in the concentration–efficiency curve (i.e., 0.5 g L−1), by retrieving the coupons after 1, 12, 24, 120 and 168 h of immersion. The corrosion product was removed from the metal surface, and the residual corrosion product was then removed according to the ASTM G1-90 (Ref 39). The specimen was ultrasonically washed in acetone, rinsed with distilled water, dried and reweighed. In each case, the experiments were conducted thrice and the corrosion rate (CR) was calculated (Ref 26, 28, 29, 31):

$${\text{CR}}\, ({\text{mg }}\,{\text{cm}}^{ - 2} \,{\text{h}}^{ - 1} ) = \frac{{\Delta W}}{At}$$
(1)

where ΔW is the weight loss (mg), A is the area of the specimen (cm2) and t is the exposure time (h). The inhibition efficiency (IE%) was evaluated using the following equation (Ref 26, 28, 29, 31):

$${\text{IE}}\% = \frac{{{\text{CR}} - {\text{CR}}^{\text{inh}} }}{\text{CR}} \times 100$$
(2)

where CRinh and CR are the corrosion rates of the steel with and without the inhibitor, respectively.

Electrochemical Experiments

The electrochemical experiments were performed in a three-electrode cell consisting of a platinum foil as a counter electrode (CE), a saturated calomel electrode (SCE) as a reference electrode and N80 carbon steel used as a working electrode. The measurements were performed on a Gamry reference 600 potentiostat/galvanostat electrochemical system by means of different electrochemical techniques such as impedance spectroscopy (EIS) and potentiodynamic polarization (PDP). The EIS tests were recorded after 24 h of immersion time in the presence and absence of various concentrations of the inhibitor (i.e., 0.2 up to 1.0 g L−1), over the frequency range of 10 kHz to 10 mHz and an amplitude of 10 mV at open circuit potential. To assess the effect of time, the EIS experiments were carried out in the absence and at the optimum GA concentration (i.e., 0.5 g L−1) at a prefixed immersion, i.e., 1, 12, 24, 120 and 168 h and at the same frequency range and amplitude described above. The data were then fitted by means of Echem Analyst 5.21 software using the opportune equivalent circuit. The IE% was calculated from the charge transfer resistance (Rct) determined from the fitting process using the following equation (Ref 25, 30, 31):

$${\text{IE}}\% = \frac{{R_{\text{ct}}^{\text{inh}} - R_{\text{ct}} }}{{R_{\text{ct}}^{\text{inh}} }} \times 100$$
(3)

where R inhct and Rct are the charge transfer resistance values in the presence and absence of the inhibitor, respectively. The PDP measurements were performed by sweeping the potential from − 1.0 to − 0.3 V with a scan rate of 1 mV s−1 after holding the specimen at open circuit potential for 24 h (i.e., to assess the effect of GA concentration) and at 168 h (i.e., to assess the effect of time) at open circuit potential. The corrosion current density (jcorr), the anodic (βa) and cathodic (βc) Tafel constants were determined by means of Echem Analyst 5.21 software. The IE% was calculated from the measured Icorr values using the relationship (Ref 25, 30, 31):

$${\text{IE}}\% = \frac{{j_{\text{corr}} - j_{\text{corr}}^{\text{inh}} }}{{j_{\text{corr}} }} \times 100$$
(4)

where jcorr and j inhcorr represent the corrosion current densities values without and with the inhibitor, respectively.

Surface Analysis

Fourier Transform Infrared (FT-IR) Spectroscopy

The FT-IR measurements were carried out using Thermo Scientific Nicolet 6700 spectrophotometer equipped with an attenuated total reflectance (ATR) accessory. The FT-IR spectra of the native inhibitor and the adsorbed inhibitor on the metal surface were recorded at a resolution of 4 cm−1, with 128 co-added scans over the range from 3700 to 700 cm−1.

Scanning Electron Microscopy and Energy-Dispersive Spectroscopy (SEM–EDS)

The steel specimens, prepared as described above, were investigated after being exposed for 24 and 168 h in the absence and presence of the optimum concentration of the inhibitor (i.e., 0.5 g L−1). After the immersion time had elapsed, the specimens were removed and rinsed with deionized water. The investigation was carried out by using a JEOL scanning electron microscope with the magnification of 2000 ×, while the chemical compositions of the corrosion product layer were analyzed by an energy-dispersive spectroscopy.

Results and Discussion

Gravimetric Measurement

The corrosion rate and inhibition efficiency for carbon steel (N80) pipeline calculated after 24 h of immersion time in CO2-saturated chloride (0.5 M of KCl) solution (blank solution) and with the addition of different concentrations of GA (i.e., from 0.2 up to 1.0 g L−1) are presented in Table 1. It follows from the data that the corrosion rate decreases and the inhibition efficiency increases with an increase in GA concentration, with the maximum inhibitor efficiency found to be 68.78% at 0.5 g L−1 of GA. The results are in line with the findings reported in the literature by other researchers. Umoren et al. (Ref 26, 28), in the study concerning the inhibition effect of GA for mild steel in 0.1 M H2SO4 solution, in the weight loss section, reported that the addition of GA, to the free acid solution, reduced the dissolution of the metal and that the extent of its protection varied with the inhibitor concentration. The authors justified the findings based on the presence of heteroatoms such as oxygen and nitrogen in the GA structure that work as adsorptive centers. The authors reported that the inhibition of the steel was ascribed to the adsorption of GA molecules through these adsorptive centers. In this study, the increase in the IE% with increasing the concentration of GA can be ascribed to a gradual increase in the surface coverage due to its adsorption on the metal surface thus reducing the contact between the aggressive solution and the metal surface. It is worth mentioning that the authors, at the same experimental conditions (i.e., after 24 h of immersion and at the same temperature), reported a maximum inhibition efficiency of 21.9% at 0.5 g L−1 of GA, which is circa three times lower than that reported in this study. The higher value of the IE% shown by GA in this study can be attributed to the different nature of the anions involved in the adsorption process. Electronegative ions may adsorb on the positive metal surface changing the surface of the metal from positive to negative. These negative charges will favor the adsorption of the protonated inhibitor. Chloride ions have a smaller degree of hydration; therefore, the number of chloride ions absorbed on the metal surface is thought to be higher than that of the sulfate ions (Ref 40, 41). Gum inhibitors are reported to be adsorbed on the metal surface mainly via electrostatic interaction established between the adsorbed anions and the inhibitor molecules (Ref 21, 25, 26, 29, 42). Therefore, the higher value of the IE% observed in this study may be due to the fact that the number of protonated inhibitor molecules adsorbed on the metal surface will be higher in the chlorides-containing solution than that of the sulfate solution. Moreover, Table 1 also reveals that the value of the IE% decreases upon increasing the temperature of the tested solution, with the maximum value found to be 38.0% at 0.5 g L−1 of GA. It has been reported (Ref 22, 23, 26) that the inhibitor molecules are continuously adsorbed and desorbed on the metal surface and that this equilibrium is temperature dependent. It seems that for GA, upon increasing the temperature, the desorption of the inhibitor molecules is more favored than that of their adsorption, leading to a less stable adsorbed layer. According to the literature, the decrease in the IE% at a higher temperature is an indication of a physical adsorption mechanism (Ref 22, 23, 26).

Table 1 Corrosion rate and inhibition efficiency obtained from weight loss measurements for carbon steel (N80) at various concentrations of GA after 24 h of immersion time at 25-45 °C
Full size table

Adsorption Study and Standard Adsorption Free Energy

In the present study, the nature of the adsorption of GA on the corroding N80 carbon steel surface in the tested solution was assessed by several adsorption isotherms, namely Langmuir, Frumkin, Temkin, Flory–Huggins, Freundlich and El-Awady adsorption istotherm (Ref 25, 27). Of all the adsorption isotherms tested, the Temkin’s adsorption isotherm was found to give the best description of the adsorption behavior of the studied inhibitor. The Temkin’s adsorption isotherm is defined by the following equations:

$$\theta = \frac{{ - 2.303\log K_{\text{ads}} }}{2a} - \frac{ 2.303\log C}{2a}$$
(5)

where θ is the surface coverage degree (θ = IE%/100), C is the inhibitor concentration, Kads the adsorption–desorption equilibrium constant, a is the molecules interaction parameter. Positive values of a imply attractive forces between the inhibitor molecules, while negative values indicate repulsive forces between them. The thermodynamic parameters obtained from the Temkin’s isotherm for the studied inhibitor, calculated from the weight loss at different temperatures, are given in Table 2 and displayed in Fig. 2. The regression coefficient ranges between 0.995 and 0.994 at 25 and 45 °C, respectively. It follows from the table that the values of a are negative in both temperatures, which is an indication of the repulsive forces between the adsorbed molecules.

Table 2 Temkin’s adsorption isotherm parameters calculated from weight loss measurements after 24 h of immersion time
Full size table
Fig. 2
figure 2

Adsorption Temkin’s isotherm carbon steel (N80) pipeline steel in CO2-saturated chloride (0.5 M KCl) at 25-45 °C

Full size image

The equilibrium constant, Kads, for the adsorption process of the inhibitor is related to the free energy of adsorption (ΔGads˚) with the following relationship:

$${\Delta }G_{\text{ads}}^{\circ } = - RT{ \ln }\left( {1\times 10^{3} K_{\text{ads}} } \right)$$
(6)

where R is the gas constant (8.314 J K−1 mol), T is the absolute temperature (K) and the value 1 × 103 is the concentration of water molecules in the solution expressed in g L−1. It should be noted that for values of ΔGads˚ higher than − 20 kJ mol−1, the adsorption process is regarded as a physisorption process while lower than − 40 kJ mol−1 indicates a chemisorption process (Ref 25, 26). In the present study, the values of ΔGads˚ presented in Table 2 are − 24.90 and − 23.36 kJ mol−1 at 25 and 45 °C, respectively. The negative values of ΔGads˚ indicate that the adsorption of GA on the carbon steel surface in the tested solution is a spontaneous process (Ref 25, 26). Moreover, the results indicate that the adsorption mechanism is a mixed-type adsorption (i.e., physical and chemical adsorption), with the physical adsorption being more dominant. The findings observed in this study corroborate the earlier reports found in the literature (Ref 18, 19, 25, 26).

Activation Parameters

The values of the activation energy were calculated with the help of the linearized Arrhenius equation:

$${ \log }\frac{{{\text{CR}}_{2} }}{{{\text{CR}}_{1} }} = \frac{{E_{\text{a}} }}{2.303R}\left( {\frac{1}{{T_{1} }} - \frac{1}{{T_{2} }}} \right)$$
(7)

where CR1 and CR2 are the values of the corrosion rates at temperature T1 (25 °C) and T2 (45 °C), respectively. Ea is the apparent activation energy, R is the gas constant (8.314 J mol−1 K−1).

The values of the heat of adsorption were calculated using the following equation

$$Q_{\text{ads}} = 2,303R\left[ {{ \log }\left( {\frac{{\theta_{2} }}{{1 - \theta_{2} }}} \right) - { \log }\left( {\frac{{\theta_{1} }}{{1 - \theta_{1} }}} \right)} \right] \times \left( {\frac{{T_{1} \times T_{2} }}{{T_{2} - T_{1} }}} \right)$$
(8)

where θ1 and θ2 are the values of the surface coverage degree at temperatures T1 (25 °C) and T2 (45 °C), respectively. Table 3 lists the values of Ea and of Qads. It follows from the data that: (1) the values of Ea are higher in the presence of GA than for the blank solution; (2) Ea increases with increasing the concentration of the inhibitor; (3) the values of Qads are negative for all the studied range of GA concentration.

Table 3 Thermodynamic parameters for carbon steel (N80) pipeline steel in CO2-saturated chloride (0.5 M KCl) solution in the absence and presence of GA from weight loss measurements
Full size table

It has been reported that the increase in Ea after the addition of the inhibitor indicates either a physical adsorption process or the formation of weak chemical bondings between the inhibitor molecules and the metal surface (Ref 24). These results, combined with the trend of the decreased in inhibition efficiency with an increase in temperature and the negative values of Qads, further support the proposed physical adsorption nature of the process (Ref 22, 24).

EIS Measurement

Figure 3 shows the EIS measurements carried out on carbon steel (N80) in CO2-saturated chloride (0.5 M of KCl) solution with different concentrations of GA. Figure 3(a) reveals that the shape of the Nyquist plots does not change after the addition of GA into the tested solution, suggesting that its addition does not alter other aspects of the corrosion mechanism. The EIS parameters calculated from the analysis of the impedance spectra fitted with the help of the equivalent circuit shown in Fig. 4 are listed in Table 4. Due to the imperfection of the metal surface, the double layer capacitance (Cdl) is simulated using a constant phase element (CPE) (Ref 18, 24). The impedance of CPE is described by the following equation:

$$Z_{\text{CPE}} = \frac{1}{{Q\left( {j\omega } \right)^{n} }}$$
(9)

where Q stands for CPE constant, n is the exponent, j is the imaginary number, and ω is the angular frequency at which Z reaches its maximum value. The parameter n quantifies the imperfections of the surface. For n = 1, CPE is a pure capacitor, and for n = −1, CPE is an inductor. The values of the double layer capacity were calculated from the following equations (Ref 25):

$$C_{\text{dl}} = Q\left( {2\pi f_{\rm max} } \right)^{n - 1}$$
(10)

where fmax represents the frequency at which Z reaches its maximum value on the Nyquist plot. As shown from Fig. 3, the fitted results present similar shape with those obtained experimentally with the values of the “goodness of fit” chi-squared (χ2) found to be very low (Table 4). This clearly indicates that the equivalent circuit used to simulate the system under investigation is the most appropriate one. From the plots, it is clear that the impedance of the steel in the blank solution changes significantly after the addition of 0.2 g L−1 of GA to the corrosive solution, reflecting an increase in Rct values from 98.91 to 168.70 Ω cm2 for the blank and inhibited solution, respectively. Moreover, it can be seen that the impedance of the metal increases with an increase in GA concentration. This behavior is likely due to the continuous adsorption of the inhibitor and the continuous growth of a protective layer on the metal surface (Ref 24,25,26,27,28,29, 31), as also confirmed by the SEM–EDS analyses (Fig. 9c). Inspection of Table 4 also reveals that Rct and Cdl have an opposite trend in the whole concentration range. The value of Rct increases while the value of Cdl decreases upon increasing the concentration of the inhibitor. The greatest effect was observed at 0.5 g L−1 of GA. A decrease in Cdl, compared to the blank solution, could be due to either an increase in the thickness of the electrical double layer and/or a decrease in the local dielectric constant, according to the Helmholtz model (Cdl = εεoA/δ) (Ref 43). Where ε is the dielectric constant of the medium, εo the vacuum permittivity, A is the electrode area, and δ is the thickness of the protective layer. This behavior is often ascribed to the replacement of water molecules on the metal/solution interface due to the adsorption of GA molecules on the metal surface (Ref 29, 40, 44). The Bode modulus plots (Fig. 3b) agree with the Nyquist spectra in which it can be seen that the impedance modulus was observed to increase with an increase in GA concentration for the entire frequency range, which indicates that GA reduced the corrosion rate in the inhibited solution. Moreover, the effect of the inhibitor can also be seen from the Bode phase angle plots shown in Fig. 3(c), where the phase angle becomes broader after the addition of the inhibitor. The broadening is likely due to the formation of a porous inhibitor film on the electrode surface.

Fig. 3
figure 3

Impedance plots recorded at OCP after 24 h of immersion for carbon steel (N80) pipeline steel in CO2-saturated chloride (0.5 M KCl) solution in the absence and presence of various concentrations of GA. (a) Nyquist, (b) Bode and (c) phase angle representations at 25 °C

Full size image
Fig. 4
figure 4

Equivalent circuit used to fit the EIS plots. Here, Rs is the electrolyte resistance, Qdl is the constant phase element representing the double charge layer capacitance, Rct is the charge transfer resistance, Rf is the layer resistance and Qf is the constant phase element of the layer

Full size image
Table 4 Electrochemical parameters and corrosion inhibition efficiency in the absence and presence of different concentrations of GA after 24 h of immersion time
Full size table

The inhibition efficiencies calculated from Eq 2 are listed in Table 4. It follows from the data that the IE% gradually increases with an increase in GA concentration up to 0.5 g L−1, with the maximum value found to be 70.27%. This behavior suggests that the adsorption of GA retards the dissolution of the metal due to the formation of a protective layer on its surface. Moreover, up to 0.5 g L−1 of GA, the protective layer is strengthened by the further addition of GA, after which it remains almost constant.

PDP Measurement

The potentiodynamic polarization curves for carbon steel (N80) pipeline in CO2-saturated chloride (0.5 M of KCl) solution in the absence and presence of GA are displayed in Fig. 5. The corrosion kinetic parameters calculated from the polarization plots are given in Table 5. The data clearly shows that the addition of GA prominently reduces the values of jcorr with the maximum inhibition efficiency of 67.04% achieved at 0.5 g L−1 of GA. Beyond this critical value, the performance of the inhibitor remains constant for a further increase in GA. Moreover, Fig. 5 also shows that the addition of GA shifts both the cathodic and anodic branches of the polarization curves toward lower current densities compared to the blank solution. It is apparent from the figure that the cathodic current density is markedly reduced even at lower inhibitor concentrations, and it does not change much for further addition of GA. This behavior indicates that the addition of GA does not change the mechanism of the corrosion process and that the hydrogen evolution reaction is in activation-controlled (Ref 25, 45). This suggests that GA forms an adsorptive layer on the metal surface that hinders the hydrogen evolution reaction (Ref 25, 45), as discussed in more detail in section 3.8.2. In contrast, the shape of the anodic curves changed greatly after the addition of GA, suggesting that the dissolution reaction is predominantly retarded. The observations indicate that the addition of the inhibitor hinders both partial reactions (i.e., hydrogen evolution and anodic dissolution) by either covering part of the metal surface and/or blocking the active corrosion sites on the steel surface. Moreover, a closer inspection of the polarization plots shows that in the presence of the inhibitor, a clear inflection is quite visible in the anodic branches (i.e., for potential around – 600 to − 500 mV). Other authors also reported a similar behavior (Ref 45,46,47). They defined this potential as the desorption potential. The authors also suggested that, for potentials lower than this desorption potential, the inhibitor hinders the anodic reaction while, beyond this critical potential, the dissolution of the metal occurs. The dissolution may be attributed to the desorption of the adsorbed inhibitor layer from the metal surface (Ref 45,46,47). It should be noted that in the case of the uninhibited solution, the anodic polarization curve does not show the typical Tafel behavior; therefore, the corrosion current density was calculated from the extrapolation of the cathodic Tafel region according to Amin et al. (Ref 45). Table 5 shows that the Ecorr shifts toward both the cathodic and anodic regions with no definite trend after the addition of various concentrations of the inhibitor, indicating that GA is a mixed-type inhibitor (Ref 25, 31). The literature reports (Ref 18, 25) that, if the shift of the Ecorr in the presence of the inhibitor was bigger than ± 85 mV, compared to the Ecorr of the blank solution, the inhibitor is regarded either as a cathodic or an anodic type. In this study, the shift is less than 85 mV, which further support the mixed-type nature of the tested inhibitor.

Fig. 5
figure 5

Potentiodynamic polarization parameters in the absence and presence of different concentrations of GA after 24 h of immersion time

Full size image
Table 5 Potentiodynamic polarization parameters in the absence and presence of different concentrations of GA after 24 h of immersion time
Full size table

Effect of Time

Immersion time plays an important role in assessing the stability of the inhibitive behavior of a given corrosion inhibitor. The effect of immersion time on the corrosion inhibition of N80 carbon steel in a CO2-saturated chloride (0.5 M of KCl) solution by gum arabic was determined using both gravimetric and electrochemical techniques at a prefixed immersion time (e.g., 1, 12, 24, 120, 168 h). The experiments were carried out at 0.5 g L−1 of GA, which is the concentration in which GA exhibits a maximum in the concentration–efficiency curve.

Gravimetric Measurement

Table S1 (supplementary information) shows the corrosion rate and the inhibition efficiency as a function of immersion time in the absence and presence of the inhibitor. It is apparent from the table that CR in the absence of the inhibitor increases with an increase in immersion time. Form Fig. 1, it can be seen that the N80 carbon steel has a ferrite–pearlite microstructure, with 41% of the ferritic phase. The ferritic phase is known to be more active than the iron carbide contained in the eutectoid mixture (Ref 10). The increase in the corrosion rate with time has been attributed to a micro-galvanic effect established between these two phases, with the ferrite phase acting as an anode and the cementite as a cathode (Ref 10, 48). Over time, the uncontrolled selective dissolution of the ferrite phase leads to an increase in Fe3C over the metal surface, which will eventually create a vicious circle (Ref 48,49,50,51,52,53,54). The growth of the cementite layer on the metal surface causes an increase in the cathodic area, which in turn, accelerates the dissolution of the steel and so on. (Ref 10, 48). By contrast, it follows from the data shown in Table S1 that upon the introduction of GA into the system, the corrosion rate of the metal decreases, which indicates that GA is an effective corrosion inhibitor, isolating the metal surface from the attack of the aggressive solution. Inspection of Table S1 also reveals that, at the optimum concentration, the inhibition efficiency increases with an increase in immersion time and attains an optimum value at 120 h (90.76%) after which, for further increase in immersion time, no appreciable change is observed. The results clearly indicate that a continuous formation of an inhibitor protective layer on the metal surface occurs during the first 120 h, after which the active sites on the surface are almost occupied by the inhibitor and therefore the corrosion process becomes stable. Table S3 (supplementary information) lists the inhibition efficiency and the time of exposure at which the maximum inhibition efficiency has been calculated of various corrosion inhibitors derived from natural products. It follows from the table that these studies were mostly performed at short immersion times (i.e., up to 6 h) and only a few of them were carried out at long immersion (i.e., 168 h). It can be seen from the table that, compared to other natural corrosion inhibitors, GA can be considered a good environmentally friendly corrosion inhibitor for carbon steel in a CO2-saturated chloride solution over prolonged exposure times.

EIS Measurement

Figure 6 presents the impedance responses observed at 1, 12, 24, 120 and 168 h of immersion in the absence and presence of 0.5 g L−1 of GA. Table S2 (supplementary information) lists parameters calculated from the analysis of the impedance plots with the use of the equivalent circuit presented in Fig. 4. It can be seen from Fig. 6(a) that the mechanism of the corrosion process in the presence of GA did not change with the immersion time, since the shape of the Nyquist plots, in its presence and absence, remains unchanged. Nyquist plots also show that immersion time plays a significant role in the corrosion resistance of the steel in the case of the investigated concentration. For instance, it follows from Table S2 that in the presence of 0.5 g L−1 of GA, Rct gradually increases with an increase in immersion time, going from 144.60 Ω cm2 (1 h) up to 417.10 Ω cm2 after 120 h of immersion, reflecting an increase in inhibition efficiency from 57.90 to 90.60%. After 120 h, the value of Rct reaches a certain value and it does not change much over time. The increase in the IE% with increasing of immersion time suggests that GA molecules are progressively adsorbed on the metal surface, leading to the formation of a thicker protective layer after 168 h of immersion, compared to the uninhibited solution (Ref 18, 24, 44). In contrast, in the absence of the inhibitor, the values of Rct gradually decrease with the immersion time. This behavior can be ascribed to the formation of a less thick and/or stable protective layer on the metal surface as confirmed form the cross section analysis (Fig. 10c and d) (Ref 25, 26, 49). The Bode phase angle plot shown in Fig. 6(c) reveals that in the case of the uninhibited solution, the size of the phase angle does not change much over time, with a phase angle maxima close to 46° after 168 h of immersion. Conversely, the size of the phase angle becomes broader and bigger with time for the inhibited solution, with a phase angle maxima close to 70° after 168 h of immersion. It should be noted that in the case of an ideal capacitor, the phase angle maxima in the intermediate frequencies is 90° (Ref 36, 55). The inhibitor system under study showed a wider capacitive behavior, compared to the blank solution, suggesting an improvement in the capacitive performance in the presence of the inhibitor with time. Therefore, the approach of the phase angle values toward 90° in the presence of GA reflects a strong molecular adsorption and thus the formation of a thicker and/or a more compact protective layer on the steel/electrolyte interface after long exposure times.

Fig. 6
figure 6

Impedance plots recorded for carbon steel (N80) pipeline steel in CO2-saturated chloride (0.5 M KCl) solution in the absence and presence of 0.5 g L−1 of GA as function of time. (a) Nyquist, (b) Bode and (c) phase angle representations at 25 °C

Full size image

It is worth mentioning that Bentrah et al. (Ref 25) observed, after 1 h of exposure, a maximum inhibition efficiency of 91% at 1.5 g L−1 of GA. Beyond this critical value, the IE% was constant. Moreover, at 0.5 g L−1 of GA Bentrah reported an inhibition efficiency of 79%, after 1 h of immersion, against an inhibition efficiency of 57.72% reported in this study, at the same concentration of GA and exposure times. The higher value of the IE% observed by Bentrah and co-authors can be ascribed to the higher concentration of chloride ions and lower pH (i.e., a higher concentration of H+ ions in the solution). As mentioned before, Gum inhibitors are mainly adsorbed on the metal surface via electrostatic interaction between the charged metal surface and the charged inhibitor molecules. Therefore, an increase in concentration of both Cl− and H+ ions increases the number of inhibitor molecules adsorbed, hence the inhibition efficiency. A similar result was also reported by Abu-Dalo et al. (Ref 31).

PDP Measurement

Figure 7 shows the corrosion potentiodynamic measurement without and with the addition of 0.5 of GA after 168 h of immersion. The potentiodynamic parameters calculated from the curves are given in Table S4 (supplementary information). The data show that the corrosion current density decreases in the presence of GA, compared to the blank solution. Moreover, both the cathodic and anodic curves shift to lower corrosion current densities. This behavior indicates that after 168 h of immersion, GA still has a significant effect on both partial reactions of the corrosion process. Thus, GA reduced the mechanism of the hydrogen evolution and the anodic dissolution of the steel (Ref 10, 56). In addition, Fig. 7 shows that the shape of the cathodic portion does not change with the addition of GA. However, the shape of the anodic domain changed greatly in the presence of 0.5 g L−1 of GA compared to the blank solution. At the optimum concentration of GA, and over a long immersion time, the metal exhibits a pseudo-passive region, only, extends to a small range of potential (i.e., − 650 to − 550 mV). This change in the shape of the anodic branch can be explained by the fact that after 168 h of immersion, a thicker adsorptive layer is formed on the metal surface (Ref 25, 26, 49), as also confirmed by the cross section analyses presented in Fig. 10(d). An increase in thickness could prevent the diffusion of the aggressive species from the bulk solution to the metal surface and therefore, limiting the dissolution of the metal. However, at a potential higher than circa − 550 mV (e.g., the desorption potential), the current increases again with the potential, which means that the dissolution of the metal occurs likely due to the desorption of the adsorbed inhibitor layer from the surface. On the other hand, the absence of this pseudo-passive region in the absence of GA points to the formation of a less thick and stable adsorbed layer (Ref 25, 26, 49) (Fig. 10c).

Fig. 7
figure 7

Polarization curves for carbon steel (N80) pipeline steel in CO2-saturated chloride (0.5 M KCl) solution in the absence and presence of 0.5 of GA as function of time (up to 168 h)

Full size image

Surface Analysis

FT-IR Measurement

The FT-IR spectra of the pure GA and the adsorbed GA on the metal surface are shown in Fig. 8. Differences between the two spectra can be used to gain useful information regarding the adsorption mechanism of GA to the metal surface. The spectrum of the native GA shows a peak at 3302 cm−1 ascribed to the extensive intramolecular hydrogen bonds and stretching vibration of the –OH groups (Ref 17, 20). The peak at 2919 cm−1 is attributable to the vibration mode of the C–H bonds. The peaks at 1597 cm−1 and at 1413 cm−1 are attributable to the asymmetric and symmetric stretching vibration of the carboxylic groups (COOH) of GA (Ref 57, 58). From the spectrum of the native inhibitor, a broad peak between 1400 and 1200 cm−1 can be seen likely due to the band assigned to the O–H in-plane bending vibration and the band of –CH2– scissoring vibration that overlaps each other (Ref 17). The bands at 1143, 1050, 1015 cm−1 are due to the symmetrical and asymmetrical ring breathing vibrations of C–C–O and C–O–C (Ref 57). The (Ref 20) bands at 1050 and 1015 cm−1 are due to the stretching vibration of C–O (Ref 57). The peaks at 912, 837, 772 cm−1 can be associated with the glucoside (1-4) linkage of the galactose unit and the (1-6) linkage of the mannose unit (Ref 18, 59). The characteristic bands related to the primary (3400-3500 cm−1) and secondary (3310-3350 cm−1) amine groups of the protein fraction (i.e., glycoprotein) are not visible due to swamped of these peaks by the adsorption of the O–H groups of the saccharide units (Ref 31, 60). Figure 8 shows that the spectrum of the adsorbed GA presents all the characteristic peaks shown by the native GA although with a lower intensity. Some authors (Ref 17, 20, 31) reported that the decrease in intensity of the peak observed at 3302 cm−1 may be due to the binding of some of the hydroxyl groups of GA with the metal surface via the H bonds formation (Ref 17, 18, 20). The decrease in intensity of the characteristic peaks of gum arabic at 1597 and 1413 cm−1 may also be attributed to the formation of hydrogen bonds between the carboxylic groups of GA and the oxidized metal surface, as also reported by other researchers (Ref 31, 45, 61). Some authors have attributed the decrease in intensity of the peaks between 1143 and 772 cm−1 to a possible interaction between the endocyclic oxygen atom of the galactose unit of GA and C6–O of the monosaccharide unit with the surface (Ref 17, 18). The results are in agreement with the previous works regarding the adsorption of gum inhibitors on the steel surface (Ref 17, 18, 20, 31).

Fig. 8
figure 8

FT-IR spectra of native gum arabic and surface-adsorbed gum arabic

Full size image

SEM–EDS Morphology

SEM–EDS measurements were also employed in order to support the electrochemical and gravimetric measurements reported in this study. A SEM image of the freshly polished surface of the steel is shown in Fig. 9(a). The image clearly shows the presence of parallel scratches on the steel surface due to the polishing process; however, no noticeable defects can be seen on the metal surface. Figure 9 also shows the surface morphological examinations of the steel surface in the absence (b) and presence (c) of 0.5 g L−1 of GA after 24 h of immersion at 25 °C. The microstructure of the sample is partially revealed after the corrosion process in the absence of GA. It can be seen from the image that the metal surface is severely corroded due to the selective dissolution of the ferritic phase caused by the presence of carbonic acid in the solution. After the preferential dissolution of the ferritic phase, the presence of lamellar zones, corresponding to the non-dissolved cementite contained in the pearlitic structure, can be seen in the image. In contrast, the surface of the carbon steel sample was greatly improved after the addition of the inhibitor, showing a much smoother surface (Fig. 9c). This, as described earlier, is due to the formation of a protective adsorptive inhibitor layer on the metal surface that shields the metal sample from the aggressive solution.

Fig. 9
figure 9

SEM images of the carbon steel surface after 24 h of immersion time: (a) polished surface; (b) blank solution; (c) 0.5 g L−1 of GA

Full size image

The protective film formed on the carbon steel surface was analyzed using an energy-dispersive spectroscopy analysis technique (EDS). The EDS spectrum and the chemical composition of all three samples are shown in Figure S1 and Table S5 (supplementary information), respectively. It follows from the data that, in the blank solution, the surface of the sample is covered by a corrosion product layer mainly composed of carbon (6.55 wt.%) and iron (91.57 wt.%) and other alloying elements of the steel with only 1.03 wt.% of oxygen. These results seem to confirm the presence of Fe3C on the sample surface with no iron carbonate layer formed on it. These findings are very well in agreement with the previous results. The literature reports (Ref 48,49,50,51,52,53,54) that in the presence of CO2, at a temperature below 40 °C, the surface is mainly covered by a porous layer of Fe3C and some alloying elements of the steel with little traces of FeCO3. These previous works pointed out that, at this temperature, FeCO3 precipitates only at elevated pH and/or at a high concentration of Fe2+ ions.

The EDS analysis reveals that in the presence of GA, the sample surface is covered by a layer with a higher concentration of carbon (14.89 wt.%) and oxygen (7.30 wt.%) elements. It should be noted that carbon and oxygen are also part of the chemical composition of the inhibitor; thus, the higher concentration of these two elements in this protective layer is likely ascribed to the adsorption of GA on the metal surface. Moreover, from the data, it is also evident that the percentage of iron decreases in the presence of GA (i.e., 91.31% for the uninhibited solution and 75.28% in the presence of GA). This decrease is likely to occur due to the overlying inhibitor film.

Over time, the selective dissolution of the ferrite phase on the surface sample leads to an increase in the concentration of Fe2+ ions in the tested solution. In this condition, the precipitation of FeCO3 on the metal surface will be more favorable (Ref 48,49,50,51,52,53,54). Figure 10(a) and (b) shows the top-view SEM analysis of the metal surface after 168 h of immersion, in the absence and presence of GA, respectively. The figures show that both surfaces are totally covered with a porous layer. The EDS measurements reveal that both layers are rich in carbon, oxygen and iron elements, suggesting that the layers are mainly composed of Fe3C with some traces of FeCO3 (Ref 48). However, as in the previous case, the EDS analysis also shows that the concentration of the above-mentioned elements is higher when the metal is exposed to the inhibited solution (i.e., C:12.3%, O:6.6% and C:22.2%, O:11.0%, in the blank and the inhibited solution respectively). Furthermore, as can be seen from the cross section view of the samples, after the addition of GA, a thicker layer of approximately 50 µm (Fig. 10d) is formed on the metal surface against that one observed in the blank solution of approximately 10 µm (Fig. 10c). Other authors also reported a similar result (Ref 62, 63). These results clearly corroborate the electrochemical and gravimetrical results presented in this study. The SEM and EDS analysis indicate that GA is gradually adsorbed on the metal surface, thus leading to the formation of a thicker and stable protective layer that hinders the diffusion of the aggressive species to the metal surface.

Fig. 10
figure 10

Top view and cross section images of the carbon steel surface after 168 h of immersion time: (a) and (c) for the blank solution; (b) and (d) for 0.5 g L−1 of GA

Full size image

Mechanism of Inhibition

Gum arabic (GA) is a water-soluble and a complex mixture of arabinogalactan (AG), arabinogalactan-protein (AGP) and glycoprotein (GP) fraction comprising of 88.4, 10.4, 1.2% of the total volume, respectively, with 0, 50, 25%, of the protein content, respectively (Ref 64) (Figure S3 in supplementary information). It is well known that the mode of adsorption, either physical and/or chemical, of a given inhibitor depends on the presence of certain physicochemical properties on the inhibitor molecule such as functional groups, charge density, size and the way its molecules interact with the metal. Due to its complex chemical structure, it is not possible to determine the exact mechanism involved, since some components may be physically and others chemically adsorbed. Inspection of Table 2 reveals that the calculated values of ΔGads indicate that the adsorption mechanism of GA on the metal surface occurs through a mixed-type interaction (i.e., physical and chemical adsorption), with the physical adsorption being more dominate. Moreover, based on the analysis of the FT-IR spectra performed on the native and adsorbed inhibitor and from the electrochemical findings, it is possible to speculate that the following four types of adsorption mechanisms or a combination of them may take place in the inhibiting phenomena involving GA on the steel surface.

Electrostatic Metal–Inhibitor Interaction

The polysaccharide fraction of GA (i.e., AG) contains hydroxyl functional groups (–OH), and carboxyl functional groups (–COOH). The protein fraction (i.e., AGP and GP) contains amino acids thus rich in amino (–NH2) and carboxyl groups. In an acid solution, these functional groups are protonated such that the polymer is in equilibrium with its polycation (Ref 25, 28):

$${\text{GA }} + x{\text{H}}^{ + } \to \left[ {{\text{GAH}}_{x} } \right]_{{\left( {\text{sol}} \right)}}^{x + }$$
(11)

where (GAHx) x+(sol) is the protonated inhibitor in the solution.

Moreover, in an acid solution, chloride ions have a strong tendency to be adsorbed on the positively charged metallic surface (Ref 25, 28, 65):

$${\text{Cl}}_{{\left( {\text{sol}} \right)}}^{ - } + {\text{H}}_{2} {\text{O}}_{{\left( {\text{ads}} \right)}} \to {\text{Cl}}_{{\left( {\text{ads}} \right)}}^{ - } + {\text{H}}_{2} {\text{O}}_{{\left( {\text{sol}} \right)}}$$
(12)

The adsorptions of these anions create an excess of electrons such that the metal surface carries negative charges. The formation of these negative charges facilitates the adsorption of the inhibitor on the metal surface through electrostatic interaction between the protonated GA molecule and the mild steel surface (Ref 25, 43) (Figure S4a from supplementary information).

Adsorption via H-Bond Formation

In addition to the electrostatic adsorption, the electrochemical and morphological findings also suggest that GA might be adsorbed on the metal surface via the formation of H bonds. As can be seen from Fig. 5, the cathodic current density decreases after the addition of the inhibitor. Roy et al. (Ref 17) and Messali et al. (Ref 18) studied the corrosion inhibition effect of guar gum in acid solution and also reported a similar result. They suggested that the decrease in the cathodic current density might be due to the formation of H bonds between the hydroxyl groups of the mannose units and the H+ ions present on the cathodic sites of the metal surface. This leads to the suppression of the hydrogen evolution reaction (Figure S4b in supplementary information). The FT-IR measurements reported in this study seem to support such a hypothesis. Figure 8 shows that the intensity of the characteristic peak associated with the hydroxyl groups decreases after the adsorption of GA on the metal surface. This behavior may be likely due to the formation of hydrogen bonds established between the hydroxyl groups of the carbohydrate molecules and the metal surface up to a different extent. Moreover, it is also possible that some of the free hydroxyl and carboxylic groups can also interact with the oxide layer formed on the metal surface through the formation of hydrogen bonds. The literature reports that oxide and/or hydroxide species formed on the metal surface might promote the adsorption of the inhibitor via the formation of such bonds (Ref 45, 66). Leong et al. (Ref 62) and Williams et al. (Ref 63) studied the adsorption effect of GA macromolecules toward oxide nanoparticles surface (i.e., magnetite nanoparticles). The authors reported that GA showed a strong affinity toward such nanoparticles. They observed that the charged and/or uncharged carboxylate/hydroxyl groups of the inhibitor molecules were able to interact with the oxide nanoparticles surface, leading to the formation of a thick adsorptive layer on their surface. In the presence of CO2, FeCO3, oxide and/or hydroxide species would form on the surface (Ref 5, 53, 54). Therefore, the adsorption of GA can also be promoted via the formation of hydrogen bonds between the hydroxyl and carboxylic groups of the inhibitor and the oxidized intermediate species on the metal surface.

Chelating Action

Polysaccharides, due to lone pair electrons among their macromolecules, might act also as a ligand, forming stable metal ion–polysaccharide complexes (Ref 67, 68). The FT-IR analysis presented in Fig. 8 reveals that the peaks associated with the glucoside (1-4) linkage of the galactose unit and the (1-6) linkage of the mannose unit decrease after the adsorption of GA on the metal. It has been postulated that (Ref 17,18,19) the endocyclic “O” atom of GA molecules could interact with freshly generated Fe2+ ions present on the anodic sites to form stable metal ion–inhibitor complexes. These complexes can be adsorbed on the steel surface through van der Waals forces forming of a protective layer, preventing the aggressive species present in the solution to reach the metal surface (Figure S4d in supplementary information). Roy et al. (Ref 17) and Messali et al. (Ref 18) also reported a similar result.

Chemical Adsorption

As mentioned before, a small contribution of the chemical adsorption process is also observed. In a weak acid solution, GA can exist as neutral macromolecules in equilibrium with its polycations (Eq 11). These neutral macromolecules can be adsorbed on the metal surface via the formation of coordinate bonds with the metal surface. In this case, the inhibitor will act as a Lewis base by donating the unshared electrons of the heteroatoms (e.g., oxygen and nitrogen) to vacant d-orbitals of iron atom on steel surface (e.g., Lewis acid) (Figure S4c in supplementary information) (Ref 17,18,19). However, it should be noted that this mechanism is more likely to occur in the first hours of the experiment. In fact, as the immersion time increases, the metal surface is progressively covered by a corrosion product layer; therefore, the direct coordination of oxygen to an exposed metal atom decreases.

Conclusion

Considering the obtained results, the effect of gum arabic on the corrosion behavior of carbon steel in a CO2-saturated chloride (0.5 M of KCl) solution can be summarized in the following points:

  • The thickening agent gum arabic was found to be an efficient corrosion inhibitor for carbon steel in the tested solution.

  • The inhibition efficiency increases with an increase in inhibitor concentration up to 0.5 g L−1, but it decreases with increasing the temperature.

  • The inhibition efficiency also increases over prolonged exposure times up to 92.44% after 168 h.

  • The adsorption of GA on the carbon steel surface follows the Temkin’s adsorption isotherm model.

  • The negative free energy of adsorption ΔG˚ads indicates a strong and spontaneous adsorption of GA on the carbon steel surface. Moreover, the values of ΔG˚ads indicate that the tested inhibitor adsorbs mainly via physical adsorption with a little contribution of chemical adsorption.

  • The values of the activation energy and heat of adsorption in the presence of GA were found to be higher than those obtained in the free inhibitor solution, further supporting the physical adsorption nature of the process.

  • Polarization measurements indicate that GA acts as a mixed-type inhibitor with a higher influence on the anodic reaction.

  • The FT-IR measurements reveal that GA was strongly adsorbed on the metal surface.

  • The cross section analysis confirmed the formation of a thick protective layer of approximately 50 µm after the addition of 0.5 g L−1 of GA, while in the blank solution it was approximately 10 µm.