The chemical composition (in wt.%) of the commercial studied cast iron is: 3.70% C, 1.98% Si, 0.25% Mn, 0.019% P, 0.017% S, 0.093% Cr, 0.01% Al, 0.021% Ni, 0.007% Cu, 0.059% Mg, 0.01% Ti and balance Fe. A round rod with an exposed area of 1.76 cm² was assembled as the working electrode. Before introducing it in the electrochemical cell, the exposed face of the electrode was ground mechanically with silicon carbide papers of graded grit ranging from 600 to 1200 grit, then rinsed with bidistilled water and degreased with acetone.

The corrosive 1 M HCl solution was prepared using analytical reagent HCl in double distilled water. Phenanthroline (Fig. 1) was added to the acid solution at different concentrations. Phenanthroline and its derivatives have been employed as a ligand in coordination chemistry; forming stable complexes with most metal ions [31].

Electrochemical experiments were performed in a conventional three-electrode cell. The cast iron sample was used as the working electrode (WE); a saturated Ag/AgCl electrode and a platinum foil (surface area of 1.0 cm²) were used as reference and counter electrode, respectively. Electrochemical impedance spectroscopy measurements were achieved at the open circuit potential for the frequency range of 100 KHz to 10 mHz, using a peak-to-peak voltage excitation of 10 mV. Potentiodynamic polarization curves were plotted from −300 to +300 mV versus open circuit potential (E_cor) at the scan rate of 1 mV/s.

All electrochemical measurements were carried out using a Potentiostat-Galvanostat (Princeton Applied Research Versastat 3). Corr-View and Z-view software’s were used for Tafel extrapolations and the fit of EIS diagrams, respectively. Before performing EIS diagrams and potentiodynamic polarization curves, the sample has been immersed in 1 M HCl solution with and without addition of inhibitor for 30 minutes until a steady state open circuit potential (E_cor) is reached.

Theoretical calculations were performed using density functional theory (DFT) in Spartan 08 program package, at B3LYP/6-31 G (d, p) level of theory. Optimization of the molecules structure was accomplished in a water phase [32]. It is believed that this level optimizes accurately and provides right electronic properties for a wide range of organic molecules. Some parameters such as the energy of the highest occupied molecular orbital (E_HOMO), the energy of the lowest unoccupied molecular orbital (E_LUMO) and dipole moments (μ) were obtained. Absolute electronegativity (χ), hardness (η), global electrophilicity (ω) and global nucleophilicity (ε) were calculated from the following equations:

Where ionization potential I = −E_HOMO and electron affinity A = −E_LUMO.

Global electrophilicity index (ω) is as follow [33]:

Nucleophilicity (ε) is the inverse of electrophilicity (1/ω).

To get more insight about ability of inhibitor molecules to interact with iron atoms, and to predict the most favorable adsorbed inhibitor, Pearson [34] has shown that the fraction of electrons transferred (ΔN) between the inhibitor molecules and the iron atom could be calculated by the following equation:

Where χ_Fe, χ_inh, η_Fe, η_inh are electronegativities and hardness values of iron and inhibitor molecule, respectively. Recently, Kokalj [35] has shown that the use of work function (Φ) of the iron surface instead of χ_Fe is a more appropriate value of iron electronegativity. In this study, we have considered Fe(110) because it is the most stable plane and it presents a dense surface package, hence derived theoretical values of iron (110) have been considered: Φ_{Fe (110)} = 4.82 eV and η_Fe = 0 eV [35].

The Fukui indices are calculated by taking the finite difference approximations from Mulliken population’s analysis (MPA) of atoms for Phen and Phen-H⁺ as follows:

Where q_k(N), q_k(N + 1) and q_k(N − 1) are the atomic charges of the systems with N, N+1, and N−1 electrons, respectively.

The cast iron specimens were in first polished with a series of emery paper to obtain a surface mirror, rinsed with double distilled water and thereafter degreased with acetone. The cleaned specimens were immersed in 1 M HCl solution with and without the optimum concentration of Phen (1.4 mM). After 04 hours of exposure to the tested solutions, the cast iron specimens were taken out, cleaned with double distilled water and finally dried with a cold air blower. Surface characterization was carried out using a ZEISS EVO MA 25 scanning electron microscope at an accelerating voltage of 20 kV.

The topology measurements (on the samples immersed for 03 hours in the above testing solutions) were performed using a Bruker Dimension Icon AFM atomic force microscope. Before starting the tests, the samples were cleaned and dried thoroughly with a cold air blower to eliminate all corrosion products.

Infrared spectra of the pure Phen and the scraped layer collected from specimen immersed during 6 hours in HCl solution containing 1.4 mM of inhibitor (Phen) were recorded via FTIR spectrometer (Agilent Cary 630) measured in the range of 4000-600 cm⁻¹.

The electrochemical measurements evidenced the Phen adsorption on the cast iron surface and therefore the inhibition of general corrosion of the metal surface, by forming a barrier for the hydrated corrosive ions. Consequently, it would be very useful to describe the adsorption process by an appropriate adsorption isotherm. The relative coverage area (θ) is proportional to the inhibition efficiency (IE %), which is determined directly as follow [47]:

Adsorption isotherms are usually used to describe the adsorption process of inhibitors. Their determination can provide important clues about the nature of metal-inhibitor interactions [53]. The surface coverage (θ) values obtained from the Tafel and EIS methods, corresponding to different Phen concentrations at 25°C, were used to deduce the best isotherm. Several attempts have been made to identify among all the adsorption isotherms; the best fit was obtained with the Langmuir adsorption isotherm modeled by the following equation [54]:

The plot of C_inh/θ vs C_inh is characterized by a straight line (Fig. 4).

The slope of the plot was close to 1, indicating that Phen molecules interact significantly with cast iron surface to form an inhibiting film that corresponds to a single layer [55]. The hypothesis of this isotherm implies that all the adsorption sites are equivalent and that the adsorption capacity of the molecules on a given site is independent for the occupation of neighboring sites [56].

In addition, with using this model (Langmuir isotherm), the Gibbs free energy (∆G^o_ads), which can be used to describe the stability of the adsorption bond between the compound and the metal, was determined using K_ads in the following equation:

It is generally accepted that the values of ΔG^o_ads revolving around −20 kJmol⁻¹, indicate that the adsorption is considered as a physisorption, the inhibition occurs via electrostatic interactions between the charged molecules and the charged metal [57], while the values around −40 kJ.mol⁻¹ or smaller (more negative) are seen as chemisorption, which is due to the charge sharing or electrons transfer from the inhibitor molecules to the metal surface [58]. The values of ΔG^o_ads deduced from the potentiodynamic polarization curves and the EIS measurement are −35.53 kJ mol⁻¹ and −36.38 kJ mol⁻¹, respectively. A large negative value of ΔG^o_ads indicates the greater adsorption capacity of Phen molecules on cast iron surface. In addition, calculated ∆G^o_ads values exceed electrostatic interactions range and are close to the chemical interactions region. This confirms the mixed mode of adsorption where physisorption and chemisorption contribute to the whole adsorption process [59].

Temperature has a great influence on the corrosion process and generally, it increases the corrosion rate [60]. For this purpose, EIS measurements were performed in the temperature range of 25-55°C for the cast iron sample immersed in 1 M HCl solution, without and with the optimal Phen concentration (Fig. 5a-b).

Examination of Nyquist plots in Fig. 5a-b reveals that the rising in temperature does not change the profile of diagrams, which means that the corrosion process is charge transfer controlled [61]. The non-ideal circular shape of these diagrams could be attributed to the frequency dispersion and heterogeneity of the cast iron surface resulting from the interfacial phenomena [61-63]. The size of these imperfect semi-circles decreases with increasing temperature; this is probably attributed to the decrease of cast iron corrosion resistance resulting from the partial desorption of the inhibitive molecules [63]. The equivalent electrical circuit used to fit the EIS data is shown in the inset of Fig. 5a-b. The deduced results are summarized in Table 3.

It can be seen from table 3 that the temperature increase reduced the value of the charge transfer resistance R_ct in the presence or absence of inhibitor and simultaneously the double layer capacitance value increased. However, it should be noted that the inhibition efficiency remains almost constant, which provides evidence for the existence of chemisorption because this adsorption type has a great adsorption heat and it is usually irreversible [64]. It follows from the above that Phen is an effective corrosion inhibitor for cast iron immersed in the acidic solution since even at 55°C its inhibition efficiency is about 95%.

It would be interesting to investigate the corrosion kinetics of cast iron in HCl solution in the presence and absence of inhibitor, using the Arrhenius equations as follows:

Where A (A cm⁻²) represents a pre-exponential factor, E_a(J mol⁻¹) is the apparent activation energy, R(8.314 J mol⁻¹K⁻¹) is the gas constant and T (K) is the absolute temperature, N is the Avogadro’s number, h is the plank’s constant, and are the enthalpy and entropy of activation, and i_cor (A cm⁻²) represents the corrosion current densities computed from the charge transfer resistance R_ct at various temperatures in the presence and absence of Phen using the following equation:

Herein; R_ct represents the charge transfer resistance, z is the valence of iron (z=2), F is the Faraday constant (F = 96485 C), R and T retain the earlier meaning in the previous equation.

The plots ln (i_cor) versus (Fig. 6a) corresponding to the blank and inhibitive solutions are linear throughout the temperature range, which indicates that the corrosion and corrosion inhibition processes follow the kinetics Arrhenius. The apparent activation energy values deduced from the slopes of the two straight lines are: 30.50 kJ mol⁻¹ and 40.78 kJ mol⁻¹ in the blank and inhibitive solution respectively. It has been reported [65-67] that inhibitors whose inhibition efficiency decreases with increasing temperature have a higher apparent activation energy value than that of the blank solution, due to physical adsorption of inhibitors on the metal surface. In the present study, the increase of the activation energy value in the presence of Phen indicates that the adsorption of Phen molecules occurs primarily through physical interactions.

The representative plots of ln () versus corresponding to the acidic and inhibitive solutions are linears (Fig. 6b), with a great coefficient of correlation. The slope gives and the intercept ln gives . The calculated values are gathered in Table 4.

The positive values of reflected the endothermic nature of the cast iron dissolution process. The value of for the uninhibited solution is lower than that for the inhibited solution. This phenomenon means a decrease in ramdomness has occurred going from reactants to the activated complex. This could be the result of the adsorption of the organic inhibitor on the electrode surface. The great negative value of entropy in the inhibited solution means that the disorder on cast iron surface decreased in the presence of Phen.

Scanning electron microscope (SEM) was used in order to demonstrate that the corrosion inhibition is mainly due to the formation of an adsorptive layer on the metal surface. Fig. 8a shows the surface of the freshly polished cast iron alloy, which consists of iron matrix embedded graphite nodules; this surface is very smooth, except some defects which had arisen during polishing.

After immersion in the uninhibited solution (Fig. 8b), the cast iron surface is strongly damaged due to the uniform dissolution of the matrix. In addition, graphite nodules peeled off from the surface caused by galvanic corrosion leading to wide and deep holes, as shown in the magnified image (Fig. 8b’). However, in the presence of Phen at its optimal concentration (Fig. 8c) the cast iron surface is smooth, intact and well improved. These observations confirm that Phenanthroline forms a protective layer by adsorption, which blocks the matrix dissolution, leading to the lowering of cast iron corrosion rate.

The 3D AFM image of the freshly polished sample (Fig. 9a) shows a regular and smooth surface, his average roughness (Ra) is 17 nm. After immersing the sample in the blank solution (Fig. 9b), the average roughness increases to 275 nm , as a result of severe damage of the surface. However, in the presence of Phen (Fig. 9c) the average roughness decreases to 97 nm, due to the formation of a protective film by adsorption of Phen molecules on the metal surface.

The FTIR spectra for pure Phen and scrapped adsorbed layer on the cast iron (Fig. 10) have been recorded to determine if Phen effectively adsorbs on the metal surface. The FTIR spectrum of pure phen shows a broad water band at 3365 cm⁻¹ attributed to O-H stretching. The absorption band at 3050 cm⁻¹ is related to the aromatic C-H stretching mode, and the strong absorption bands around 1585 cm⁻¹, 1558 cm⁻¹ are assigned to C=N stretching vibrations. The peaks at 1502 and 1418 cm⁻¹ are associated with the carbocyclic ring vibrations (C=C), the presence of band at 1134 cm⁻¹ corresponds probably to the in-plane hydrogen deformation modes or ring vibrations, and other peaks at 850 and 736 cm⁻¹ correspond to the hydrogen (C-H) out of plane bending on the benzene and the heterocyclic rings. A full vibrational assignment of phenanthroline molecule has been previously published [77]. The most bands observed in the FTIR spectra of the adsorbed layer on cast iron surface closely resemble those appearing in the pure Phen spectra. A broad band around 3300 cm⁻¹ attributed to O-H stretching, which further indicates the formation of FeOOH on the adsorbed layer. the aromatic C-H stretching band has vanished from 3050 to 3029 cm⁻¹. The shifted to higher wavenumbers of the C=N and C=C bands confirm the involvements of these centers in the adsorption. in addition, the absorption bands at 850 and 738 cm⁻¹ assigned to the out of plane C-H vibrations are shifted to 835 and 715 cm⁻¹, respectively. Reference has been made to the fact that these bands (out of plane vibrations) move to lower wavenumbers on coordination [78]. All the informations indicate clearly towards adsorption of Phen molecules on the cast iron surface, similar conclusions have been made by many other researchers [79-80].

The adsorption process of organic molecules on a metal surface depends on their chemical structure, the distribution charge in the molecule, the aggressive medium and the topography of the metal surface. Therefore, on the basis of the experimental results, FTIR in particular which indicates the presence of most functional groups of Phen on the adsorbed layer, a corrosion inhibition mechanism could be proposed:

In HCl solution, Phenanthroline may exist in the protonated form in equilibrium with its molecular form, thus, two modes of adsorption could be envisaged. According to the literature, generally ferrous alloys carry an excess positive charge in acid solution, thus, the acidic anions (Cl⁻) specifically adsorbed on the metal surface create an excess negative charge which favors the adsorption of Phen-H⁺, by involving electrostatic forces.

In addition to the physical adsorption, the neutral Phen may be adsorbed on the metal surface through chemical adsorption by sharing electrons on the basis of donor-acceptor between the π-electrons of the phenanthroline ring and the vacant d-orbital of iron surface atoms. The molecular structure of Phen-H⁺ remains almost unchanged with respect to its neutral form, it can be stated that when Phen-H⁺ is adsorbed on the metal surface, the coordination bond can also be formed by partial transference of electrons from the polar Nitrogen atom to the metal surface.