Metallic materials and equipment can easily corrode during storage or daily operation in everyday environments, affecting and hindering their subsequent normal use [1,2]. The most common solved method is acid pickling treatment, which can effectively eliminate unfavorable dirt, oxides, and rusts on metal to obtain the clean metal surface [[3], [4], [5]]. For such applications, dilute sulfuric acid and hydrochloric acid are usually the choice of cleaning agent. However, metal also suffer from severe corrosion in acid pickling process at the same time.

To alleviate the side effect of these acids, corrosion inhibitors are selected as the “threshold effect” for harnessing and preventing the metallic corrosion during acid cleaning process [[6], [7], [8]]. Most of conventional inhibitors belong to inorganic salts such as chromate and nitrite and so on because of the inexpensive price [9,10]. Even so, high toxicity still limits their widespread application [11,12], and organic inhibitors begin to enter the market to replace these traditional inorganic inhibitors [13,14]. The structure of organic compounds mainly contains N, P, O, S, conjugated double bonds, aromatic rings, and heterocyclic rings, which can endow inhibitors mighty adsorption capacity on metal [[15], [16], [17], [18]]. Organic molecules can self-assemble on the metal surface in a well-organized manner to form an anchored protective film and thus corrosive attack can be effectively retarded [19].

It is worth noted that research on organic corrosion inhibitors has become a hot spot in the past decade [[20], [21], [22], [23], [24]]. However, the main means of developing new organic inhibitors is still blind filtration nowadays because of the complicated and indistinct correlation between molecular structure and inhibition efficiency. Therefore, it is great challenge to clarify the influence of active functional groups on the inhibition effectiveness of organic compounds to design new and better organic inhibitors that will be efficient under harsh corrosive conditions.

In present study, we proposed two tetrazole derivatives (Fig. 1) as copper-corrosion inhibitors in sulfuric acid solution, and investigated the influence mechanism of S-linker on the inhibition performance of tetrazole compounds. The inhibition effect of studied inhibitors was evaluated through corrosion morphological observation including scanning electron microscopy (SEM) and atomic force microscopy (AFM), and weight loss measurement followed by electrochemical techniques such as electrochemical impedance spectroscopy (EIS) and potentiodynamic polarization. The adsorption and bonding mechanism of investigated compounds was analyzed by Langmuir adsorption isotherm and X-ray photoelectronic spectroscopy (XPS). Finally, molecular dynamics (MD) simulation validated the experimental results and gave deep insights into the action mechanism at molecular level.

5-Benzyl-1H-Tetrazole (BTA) and 5-(Benzylthio)-1H-tetrazole (BTTA) were purchased from Adamas-beta corporation. Pure copper was used as the working electrode (WE) and was prepared as before [25]. 0.5 M H₂SO₄ was diluted from concentrated sulfuric acid (98%). The test solutions were 0.5 M H₂SO₄ containing different concentrations (0, 0.5, 1, 1.5, and 2 mM) of studied tetrazole compounds.

Electrochemical measurement was conducted by CHI-760E workstation with a traditional three-electrode system at 298 K. Saturated calomel and Pt sheet electrodes were utilized as reference and auxiliary electrodes, respectively. The open circuit potential (OCP) was first carried out to reach stability, then EIS was recorded at frequency range of 10⁵-10^-2 Hz with 5 mV amplitude. EIS data were fitted via Zsimpwin software. For the polarization measurement, scanning potential range is within ±250 mV around OCP and scanning rate is 1 mV s^-1. The relevant inhibition efficiency (η) values were calculated as follows [26]:

(2) ${{\eta }_{\mathbf{EIS}}}\left(\% \right)=\frac{{{R}_{\text{ct}}}-{{R}_{\text{ct,0}}}}{{{R}_{\text{ct}}}}\times 100$

where i_corr,0 and i_corr denote current densities without and with inhibitor, R_ct,0 and R_ct denote charge-transfer resistances without and with inhibitor, respectively. It is noted that a fresh solution was utilized for each test. Besides, three parallel tests were performed with the same conditions.

At 298 K, the gravimetric measurement was carried out in 0.5 M H₂SO₄ solution with different concentrations of BTA/BTTA for 24 h. Copper specimens were prepared, cleaned, and then dried with a dryer. After immersion tests, these coupons were removed, washed, and dried again. The initial and final mass were determined to obtain weight loss (ΔW). Corrosion parameters such as corrosion rate (v) and η_wl were computed by:

where A is immersion area, v₀ and v are corrosion rates of copper without and with protection of inhibitor, respectively. The weight loss measurement was performed in triplicate for each case.

The corrosion morphology of copper specimen was observed by both SEM (JEOL-JSM-7800 F, Japan) at 10 kV and AFM (MFP-3D-BIO, America) at tapping mode. The composition of adsorption film and bonding information were determined by XPS (ESCALAB 250Xi, America).

In order to provide insightful information on interfacial interaction and inhibition mechanism, the interaction between inhibitor compounds and copper substrate was explored by MD simulation using Forcite module from Accelrys Inc. Four investigated organic molecules in neutral and protonated forms (BTA, BTTA, BTAH⁺, BTTAH⁺) were first built. Accordingly, the densely packed Cu (111) surface model was chosen as representative because it is the most stable low Miller index copper surface and then the most abundant. Then Cu(111) plane was cleaved from pure Cu crystal and expanded to a (12 × 12) supercell. The simulation box containing one inhibitor, 500 water molecules along with 30 Å vacuum slab was established, which was allowed to freely interact with Cu(111) surface. Simulation process followed COMPASS force field, NVT canonical ensemble, and periodic boundary condition with time step of 1 fs and simulation time of 500 ps. Binding energy (E_binding) values between Cu(111) surface and organic molecules were calculated [27] and discussed.

The SEM morphologies of copper are shown in Fig. 2. After 24 h dipping in 0.5 M H₂SO₄ medium, many corrosion products are detected on blank copper (Fig. 2(a)) owing to aggressive acid attack, while copper protected by BTA (Fig. 2(b)) shows very uniform surface. In addition, the smoothest surface is found in the presence of BTTA (Fig. 2(c)), indicating the slightest corrosion. Further, Fig. 3 shows 2D and 3D AFM graphs of copper after immersion in different test solutions. Clearly, the bare copper (Fig. 3(a, d)) is badly corroded, and high bumps and low grooves can be observed. Meanwhile, introducing two inhibitors respectively remarkably retards copper corrosion and fairly even surfaces are displayed owing to the fabrication of inhibitor film. Compared with BTA addition (Fig. 3(b, e)), the copper protected by 2 mM BTTA (Fig. 3(c, f)) is flattest. This is also revealed by relevant height profiles in Fig. 4. The copper surfaces covered by BTA/BTTA have a slight fluctuation within 200 nm, whereas the surface fluctuation of bare copper is about 700 nm. Besides, the average roughness value decreases from 91.5 nm (bare) to 26.8 nm (BTA) and 23.5 nm (BTTA) respectively, revealing the formation of denser and more well-ordered BTTA-adsorption film on copper surface than that of BTA. Therefore, BTTA can provide superior inhibition ability to BTA toward copper corrosion in harsh sulfuric environment.

To investigate protective ability and mechanism of studied tetrazole compounds, EIS results such as Nyquist and Bode diagrams of copper in different testing media are illustrated in Fig. 5, Fig. 6. As illustrated in Nyquist graph, a capacitive loop (concerning charge transfer and electrical double layer) at high-frequency region and a diffusion (Warburg) impedance at low-frequency region are found in all spectra. At the meantime, the diameter of semi-circle remarkably enhances in the presence and increasement concentration of two studied compounds, which indicates the formation of inhibitor-adsorption film. The same shape of all arcs is also noted, demonstrating that introducing these inhibitors only retard the charge-transfer rate but doesn’t affect relevant corrosion mechanism. Particularly, the depressed capacitive semicircles can be attributed to the frequency dispersion phenomenon, which is associated with the roughness and inhomogeneity of copper electrode in test solutions [28].

For Bode images in Fig. 6, the phase angle values significantly increase with increasing concentration of two tetrazole compounds, that is to say, an efficient adsorption inhibitor layer is formed at copper/solution interface. In addition, the impedance modulus at lowest frequency (0.01 Hz) exhibits an upward trend as the concentration of BTA or BTTA enhances. Compared with the bare copper, the log|Z| value is improved 1.5 orders of magnitude by the maximum concentration of BTA, and even more than 2 orders of magnitude can be obtained for 2 mM BTTA, manifesting the more compact adsorption layer and thus stronger inhibition performance of BTTA than BTA for copper corrosion in acid medium.

The spectrum of copper in blank solution was fitted by the circuit presented in Fig. 7(a), while the other spectra in presence of tetrazole inhibitor were fitted by the circuit in Fig. 7(b). These circuits contain several elements such as R_s (solution resistance), R_ct, W (Warburg impedance), R_f (film resistance), CPE_dl and CPE_f (constant-phase angle elements), respectively [29]. Double electric layer (C_dl) [30] and film capacitances (C_f) [31] can be calculated by:

(6) ${{C}_{\text{f}}}={{Y}_{0}}{{\left( 2\text{ }\!\!\pi\!\!\text{ }{{f}_{\max }} \right)}^{n-1}}$

where Y₀ represents the modulus of CPE, the exponent n (-1 ≤ n ≤ 1) is associated with CPE distribution, f_max denotes the frequency value at peak position in imaginary impedance.

The calculated electrochemical parameters are summarized in Table 1. It shows that both R_f and R_ct values enhance, whereas C_dl and C_f values decline in the presence of two tetrazole inhibitors, and this trend continues with the augment of investigated inhibitor concentration. That is to say, these compounds effectively protect copper against acid attack and the protection ability enhances at higher compounds concentration. It reveals that the self-assembly fabrication of anchored film on copper surface due to the gradual displacement of water molecules by BTA/BTTA molecules at copper/solution interface [32,33]. Especially, R_ct value of bare copper (365 Ω cm²) is significantly improved to 9346 Ω cm² for BTA and 67,621 Ω cm² for BTTA respectively at maximum concentration. Meanwhile, η_EIS values also enhance with increasing inhibitor concentration, reaching up to 96.1% and 99.5% for BTA and BTTA respectively. It demonstrates greater protection ability of BTTA layer than that of BTA toward copper corrosion, which may correspond to stronger multiple anchor adsorption of BTTA molecules by the S-linker than BTA.

Fig. 8 represent polarization curves of copper in different test solutions. The polarization parameters including i_corr, η_Taf, corrosion potential (E_corr), anodic and cathodic slope (β_a, β_c) are listed in Table 2. These results indicate that compared with blank, both anodic and cathodic reactions shift to lower current densities while the anodic reaction is mainly restrained in presence of BTA/BTTA compound. The shift of E_corr values is within 85 mV, thus both tetrazole compounds behave as efficient mixed-type inhibitors [34]. Besides, the changes in β_a and β_c values confirms the inhibition effect of two studied compounds on both anodic and cathodic reactions [35].

It is easy to found that the i_corr value remarkably reduces after addition of both two compounds and then shows a downtrend as the concentration of two tetrazole inhibitors increases, which indicates that the corrosion reaction is effectively suppressed. Meanwhile, η_Taf values also increase with enhancing inhibitor concentration and reach up to 96.3% (BTA) and 99.8% (BTTA), respectively at 2 mM. This implies that BTTA protected copper owns superior anti-corrosion ability to that of BTA because compared with BTA film, more active sites on copper are blocked by the BTTA adsorption layer.

In order to provide information on the average corrosion rate for a long exposure time, weight loss method was adopted at 298 K and the obtained results can be seen in Fig. 9. It is clear that increasing concentration of two inhibitors reduces corrosion rate of copper, and enhances η_wl value meanwhile. At the maximum concentration of both compounds, the corrosion rate of copper in sulfuric acid decreases more than one order of magnitude, manifesting the effective inhibition effect of these tetrazole inhibitors. Moreover, compared with BTA (Fig. 9(a)), BTTA (Fig. 9(c)) exhibits higher η_wl values at same concentration.

Langmuir isotherm equation can be expressed as follows [36]:

where θ (η_wl/100) is surface coverage ratio, K_ads is adsorption equilibrium constant. Through fitting gravimetric data, the linear relation between C and C/θ is illustrated in Fig. 9. As shown that R² values for BTA and BTTA were 0.975 and 0.995 respectively, which reveals that the adsorption behavior of these inhibitors on copper follows Langmuir model. The adsorption free energy ΔGads0 was calculated by [36]:

(8) $\text{ }\!\!\Delta\!\!\text{ }G_{\text{ads}}^{0}=-RT\ln \left( 55.5{{K}_{\text{ads}}} \right)$

The calculated parameters are also shown in Fig. 9. Clearly, the higher K_ads and lower ΔGads0 values of BTTA represent its stronger adsorption action and higher inhibition effect than BTA [37]. Furthermore, the negative ΔGads0 value implies a spontaneous adsorption process of these compounds on copper surface. It is generally accepted that ΔGads0 ≤-40 kJ mol^-1 indicates chemisorption process (covalent bonds by transfer or sharing of electrons) while the ΔGads0 ≥-20 kJ mol^-1 is associated with a physical adsorption process (electrostatic interaction) [38]. As seen that the adsorption free energy of BTA and BTTA are about -28.3 and -29.1 kJ/mol, respectively, inferring that the tetrazole inhibitor layer at copper/solution interface can be fabricated by mixed adsorption involving both chemical and physical adsorption.

In order to confirm the adsorption mode and get insights into the bonding information, the XPS spectra of copper surface is shown in Fig. 10. By observation in Fig. 10(a), Cu, C, and O elements can be found on copper surface after dipping in blank medium, demonstrating that copper corrosion happens in harsh acid solution. A new peak of N element can be seen in copper-BTA full spectrum, while new elements of both N and S can be detected on full spectrum of BTTA covered copper, which reveals successful adsorption of studied compounds on copper substrate. Therefore, O peaks of copper adsorbed with two inhibitors respectively decrease compared with that of bare copper. In addition, the N1s spectrum of copper-BTA (Fig. 10(b)) can be fitted into three peaks of 399.4, 400.0, 400.6 eV, which correspond to two nitrogen environments in tetrazole ring and N-Cu bond respectively [32,39,40]. Owing to the same N environments in both compounds, the de-convolution of high resolution N1s spectrum of copper-BTTA (Fig. 10(c)) exhibits fitted peaks similar to those of copper-BTA. The S2p spectrum of BTTA covered copper (Fig. 10(d)) is resolved into two peaks, which are located at 163.1 and 164.2 eV, respectively. The first peak is ascribed to the coordinate bond between sulfur in BTTA molecule and copper surface (S-Cu), while the second peak is related to the S-C bond in BTTA molecular environment [41,42]. In conclusion, the existence state of N1s and S2p on copper substrate infers that two inhibitor films are formed on copper by both chemical and physical adsorption modes, which is well consistent with above Langmuir analysis.

MD simulation can reasonably predict the adsorption equilibrium configuration of organic compounds on metal substrate. The optimized interfacial configurations of Cu(111) surface adsorbed BTA/BTTA molecules are exhibited in Fig. 11. It is evident that both inhibitor molecules in different forms adsorb onto Cu(111) surface with a nearly parallel mode and thus cover maximum surface area of copper. This phenomenon mainly ascribes the high tendency of two inhibitor compounds to donate electrons to the unoccupied Cu orbitals and also to receive the electrons from d-orbitals of copper through a back-bonding [43,44]. It can be inferred that the chemical bond formation happens and thereby strengthen the adsorption affinity of these organic compounds, which is verified by XPS analysis above. As a result, the protective layer based BTA (BTAH⁺) or BTTA (BTTAH⁺) can be formed due to an effective replacement of water molecules by these inhibitor molecules at Cu/solution interface. Both the compounds have heteroatom N, aromatic ring, along with tetrazole unit in the molecular skeleton, while BTTA also have heteroatom S, which can offer more sufficient number of electrons to the vacant copper d-orbitals than BTA molecule. Thus the higher E_binding values of BTA (127.2 kcal/mol) and BTTAH⁺ (131.0 kcal/mol) than BTA (118.7 kcal/mol) and BTAH⁺ (123.2 kcal/mol) were obtained, indicating the stronger interaction between BTTA (BTTAH⁺) and Cu(111) surface than BTA (BTAH⁺) [45,46]. These theoretical outcomes strongly corroborate with the inhibition capacity and mechanism obtained from experimental techniques.

Self-assembling anchored films based two tetrazole derivatives respectively were fabricated on copper substrate and the protecting effect for copper in harsh sulfuric acid was thoroughly investigated. The following conclusions can be drawn:

(1)By corrosion appearance observation, better protective ability of BTTA than BTA toward copper corrosion can be inferred, which is verified by EIS, Tafel, and weight loss methods. Specifically, η values increase with enhancing inhibitor concentration and achieve 96.3% (BTA) and 99.8% (BTTA), respectively at 2 mM. It may correspond to stronger multiple anchor adsorption of BTTA molecules than BTA on copper because of the existence of S-linker.

(2)Tetrazole inhibitor film at copper/solution interface obeys Langmuir model concerning mixed adsorption involving both chemical and physical adsorption, which can be confirmed by XPS analysis. Besides, the negative value of adsorption free energy reveals a spontaneous adsorption process of these tetrazole compounds on Cu.

(3)The higher E_binding values of BTTA/BTTAH⁺ than those of BTA/BTAH⁺ was obtained, indicating the stronger interaction between BTTA/BTTAH⁺ and Cu(111) surface than BTA/BTAH⁺. Therefore, it is inferred that BTTA compound can provide more sufficient number of electrons to the vacant Cu d-orbitals than BTA molecule and thus reinforce the anchor adsorption of BTTA/BTTAH⁺ layer, which is associated with the inhibition performance obtained from experimental methods.

This research was supported financially by the National Science Fund for Distinguished Young Scholars of China (No. 51825505).