Introduction

Highly electrical and thermal conductivities, as well as good mechanical workability of copper and its alloys enable them to have a diverse range of applications in pipelines for domestic and industrial water systems, shipbuilding, seawater desalination and heat exchanger [1,2,3,4,5]. However, copper and its alloys suffer serious corrosion in chloride environments, causing huge economy losses [6,7,8]. Thus, a considerable amount of efforts have been made to improve the corrosion resistance properties of copper using effective organic inhibitors with aromatic rings, and electronegative functional groups involving the heteroatoms of sulfur, nitrogen and oxygen, which may adsorb at the copper surface to form inhibitive coatings [9,10,11,12,13,14,15].

Unfortunately, among them, some are toxic and expensive. In recent years, various types of nontoxic organic compounds have been investigated to meet the recommendation using eco-friendly inhibitors as substitutes for restricted toxic inhibitors [16,17,18,19,20,21]. Some hydroxamic acids and their derivatives with biological activities, such as anti-inflammatory and anti-asthmatic, are used as pesticides and plant growth promoters [22]. In addition, hydroxamic acids could chelate with metal ions to form complexes [23, 24]. Thus, hydroxamic acid derivatives have already been reported as effective corrosion inhibitors for carbon steel and copper corrosion [25,26,27]. AHA as a potential corrosion inhibitor could form a protection film on the copper surface due to multiple adsorption centers of a nitrogen atom and two oxygen atoms in its structure (see Scheme 1). So far, the adsorption behavior of AHA on copper surface as well as its corrosion inhibition efficiency has not been observed in detail.

Scheme 1
scheme 1

Optimized structure of acetohydroxamic acid

Full size image

In this work, AHA was adsorbed on copper surface by self-assembled monolayers (SAMs) technique and the efficiency against corrosion in 3 wt% NaCl solution was estimated by electrochemical impedance spectroscopy (EIS) and polarization methods. SERS technique as a powerful tool to provide molecular fingerprint information was used to elucidate formation mechanism of AHA coating on the copper surface as well as Langmuir adsorption isotherm measurement. Furthermore, quantum chemical studies were used to predict the feasible adsorption centers and confirm the relationship between the molecular structures of AHA and its inhibition property.

Experimental

Materials and chemicals

Acetohydroxamic acid (AHA) was purchased from Sigma-Aldrich. Analytical grade NaCl was dissolved in ultrapure water (18 MΩ cm) to prepare 3 wt% NaCl corrosion media. Sulfuric acid and ethanol were analytical reagents, purchased from Sinopharm Chemical Reagents Company.

Apparatus

Raman spectroscopic measurement was conducted using LabRam II confocal Laser Raman system (Dilor, France). A 1024 × 800 pixels charge-coupled device detector cooled by semiconductor was used, and the excitation source was a He–Ne laser at 632.8 nm with power of ca. 5 mW. The slit and pinhole were controlled at 100 and 1000 μm, respectively. The laser was focused onto the copper surface through a long-working-length of 50× objective. Each Raman spectrum was taken with 8 s integration time and 3× repeats. The line of silicon positioned at 519 cm−1 was used for spectral calibration.

The electrochemical measurements were carried out using CHI 750C electrochemistry workstation (CH Instruments, Inc.).

Pretreatment for electrode

The copper electrode was constructed from polycrystalline copper (99.999%) rod inside of a Teflon sheath, and the exposure area of surface was 0.0314 cm2. Before the Raman spectroscopic and electrochemical measurements, the copper electrode was sequentially abraded with 500- and 1000-grit papers, followed by 0.3 μm alumina powders to get a shiny mirror-like electrode surface, and then cleaned with Milli-Q water and pure ethanol in an ultrasonic bath to remove any remaining alumina particles and possible rust. For SERS detection, to obtain the necessary roughness of the copper surface, copper surface was first treated in 2 M H2SO4 solution using an oxidation–reduction cycle (ORC) process in the potential range from −0.55 to 0.45 V (vs. SCE) with scan rate at 0.02 V/s and 10 sweep segments and final potential was applied at −0.55 V (vs. SCE) for 60 s [28]. In a conventional three-electrode cell, the AHA-modified copper electrode (or bare copper specimen) and a platinum electrode were used as working electrode and the counter electrode, respectively. All potentials referred to in this paper are reported relative to the saturated calomel electrode (SCE), which was used as reference electrode.

Coating the copper surface with AHA

The cleaned copper electrodes were immersed immediately into the deoxygenated AHA solutions with various concentrations. The assembly time effect on the formation of AHA film at the Cu surface was also considered. Before spectroscopic and electrochemical experiments, the AHA solution was removed and the electrode surface with AHA film was rinsed using Milli-Q water, and then dried by flowing nitrogen gas.

Electrochemical measurements

The impedance spectra were performed in a three-electrode cell starting from open circuit potential (OCP) with the AC voltage amplitude of 5 mV (vs. SCE) in the frequency range from 0.01 Hz to 100 kHz. The impedance results were simulated with a compatible electronic equivalent circuit fitting. The electrochemical polarization curves were obtained from −0.1 V (vs. SCE) to −0.3 V (vs. SCE) with a scan rate of 1 mV s−1.

Theory computation details

The geometry optimization of AHA molecule and its Raman spectral calculation were performed at the density functional theory (DFT) based on B3LYP/6-31G(d) level using Gaussian 03 package and the vibrations were observed using GaussView 5.0. The SERS bands of AHA molecule were tentatively assigned according to the computation result obtained using B3LYP/Lanl2DZ method on the basis of AHA-Cu4 mode. Additionally, the calculations of the frequencies have been scaled by a factor of 0.9940 for B3LYP/6-31G(d) level and a uniform scaling factor of 0.9762 for B3LYP/LanL2DZ method.

Results and discussion

EIS spectra

Electrochemical impedance spectroscopy for studying the inhibitor against corrosion does not result in any destruction of the film state on the electrode surface [29].

Nyquist plots along with bode and phase angle plots for the copper electrodes with AHA films formed in various AHA concentration solutions and under different assembly time corroded in 3 wt% NaCl solutions are shown in Figs. 1 and 2. The shape of the Nyquist plot for bare copper electrode is different from those of AHA-modified electrodes. The latter ones are composed of a high-frequency imperfect semicircle and a straight line seen as Warburg impedance in the low frequency [30].

Fig. 1
figure 1

Nyquist (a), bode (b) and phase angle (c) plots of copper electrodes with acetohydroxamic acid films formed in different concentrations of acetohydroxamic acid solutions, acquired in 3 wt% NaCl solution

Full size image
Fig. 2
figure 2

Nyquist (a), Bode (b) and phase angle (c) plots of copper electrodes in the absence and presence of film formed in 10 mM acetohydroxamic acid solution for different time, recorded in 3 wt% NaCl solution

Full size image

Clearly, the Warburg impedance for copper is connected with the diffusion of many oxide species due to lacking protection film [31]. In Fig. 1a, compared with that of bare copper electrode, semicircle diameters of the electrodes with AHA increase visibly and it reaches a maximum when the film was formed in 10 mM AHA solution. It means that formed in an optimized assembly concentration of AHA solution, molecular adsorption layer on the copper surface is compact for enhancing corrosion protection [32]. Similarly, the Z mod value of the bode plot (Fig. 1b) and phase angle value from phase angle plot (Fig. 1c) for the film-modified electrodes formed in 10 mM AHA solution are highest, indicating a superior protection performance. In Fig. 2a, semicircle diameter of the electrode with AHA formed in 10 mM AHA solution for 3 h assembly time comes to the greatest value; the same results have been observed in the bode plots (Fig. 2b) and phase angle plots (Fig. 2c). If the assembly time is less than 3 h, the AHA layer adsorbed on the electrode might not be dense enough while more than 3 h, accumulation of AHA molecules occurs on the surface, which also affects the compact structure of film and hinders inhibition of copper corrosion.

The equivalent circuit models for analyzing impedance characteristics of electrodes with and without AHA are displayed in Fig. 3. Such equivalent circuit models were selected, considering the possible adsorption fashion and the film structure of AHA at the copper surface as well as evaluating the least error for each parameter routinely less than 10% and the Chi-square values (χ 2) less than 1 × 10−3.

Fig. 3
figure 3

The electrical equivalent circuit models for impedance data of copper electrode without film (a) with AHA film (b)

Full size image

In Fig. 3a, the equivalent circuit model of R(Q(RW))(QR) is fitting the Nyquist plot of the bare copper while R(C(R(Q(RW)))) equivalent circuit in Fig. 3b is for the AHA-modified copper.

R s is the solution resistance, and R ct, the charge transfer resistance is attributed to the corrosion reaction at the electrode/solution interface [33]. R f represents the resistance of the film formed on the copper surface, and W indicates the Warburg impedance. Q dl and Q f are defined as constant phase elements (CPE), representing film capacitance and a modified double-layer capacitance [34], described by the following equation [35]:

$$ Z_{\text{CPE}} = Y_{0}^{ - 1} \left( {j\omega } \right)^{ - n} $$
(1)

where Y 0 is the modulus, j is the imaginary root, ω is the angular frequency and n is the phase (−1 ≤ n ≤ + 1). Relying on the different values of n, CPE may be inductance (n = −1, Q = L), resistance (n = 0, Q = R), Warburg impedance (n = 0.5, Q = W) or capacitance (n = 1, Q = C).

Electrochemical parameters calculated from EIS measurements for copper electrodes without and with AHA films are listed in Table 1 (bare copper), Table 2 (with films formed in different AHA concentrations solutions) and Table 3 (with films under different assembly time).

Table 1 Electrochemical parameters calculated from EIS measurements for copper electrodes without the AHA films, in 3 wt% NaCl solution
Full size table
Table 2 Electrochemical parameters calculated from EIS measurements for copper electrodes with the AHA films formed in different concentrations of AHA solutions, in 3 wt% NaCl solution
Full size table
Table 3 Electrochemical impedance parameters for copper electrodes in 3 wt% NaCl solution, in the absence and presence of film formed in 10 mM AHA solution for different time
Full size table

All the n values are over 0.5, indicating the relatively low corrosion of the electrode [36]. The inhibition efficiency (η%) is described in the following equation [37]:

$$ \eta {(\% ) = }\frac{{R_{\text{p}} \text{ - }R_{\text{p}}^{ 0} }}{{R_{\text{P}} }}{ \times 100} $$
(2)

\( R_{\text{p}}^{ 0} \) refers the polarization resistance of the bare copper, and R p is the polarization resistance of the AHA-modified electrode (R p is the sum of R ct and R f) [38].

Observation from the three Tables reveals that the R p values for the AHA-modified copper electrodes increase visibly and the W values tend to decrease obviously compared to the bare copper electrode. Under the above-mentioned optimized condition, R p reaches the maximum value, W reaches the minimum value and the maximum inhibition efficiency reaches 93.5%. The RSD (relative standard deviation) of the inhibition efficiencies calculated from electrochemical impedance parameters for three copper electrodes with film is 0.13%. In all, impedance data suggest that the AHA monolayer has a remarkable protection behavior for copper.

Polarization measurements

Polarization curves of the electrodes with different AHA concentrations and different assembly time recorded in 3 wt% NaCl solutions are given in Figs. 4 and 5, respectively. The reaction of the cathodic oxygen reduction is depicted as follows:

$$ {\text{O}}_{2} + 4{\text{e}}^{ - } + 2{\text{H}}_{2} {\text{O}} \to 4{\text{OH}}^{ - } $$
(3)
Fig. 4
figure 4

Anodic and cathodic polarization curves of the copper without and with AHA film formed at different concentrations for 3 h: a bare, b 1 mM, c 5 mM, d 10 mM, e 50 mM, in 3 wt% NaCl solution

Full size image
Fig. 5
figure 5

Anodic and cathodic polarization curves for copper electrodes without and with AHA films formed in 10 mM AHA solution for different time: a bare, b 1 h, c 3 h, d 5 h, e 8 h in 3 wt% NaCl solution

Full size image

Also, in the presence of Cl−, the anodic dissolution process of copper consists of the following steps (the ionization of copper with Cl− and the diffusion of CuCl2 − to the bulk solution) [39]:

$$ {\text{Cu}} + {\text{Cl}}^{ - } \to {\text{CuCl}} + {\text{e}}^{ - } $$
(4)
$$ {\text{CuCl}} + {\text{Cl}}^{ - } \to {\text{CuCl}}_{2}^{ - } $$
(5)
$$ {\text{CuCl}}_{2}^{ - } \to {\text{Cu}}^{2 + } + 2{\text{Cl}}^{ - } + {\text{e}}^{ - } $$
(6)

The related electrochemical parameters obtained from the polarization curves, such as cathodic and anodic Tafel slopes (β c and β a), corrosion potential (E corr), corrosion current density (i corr) and the inhibition efficiency (η%), are given in Tables 4 and 5. E corr and i corr were estimated by the method of Tafel extrapolation.

Table 4 Corrosion parameters obtained from potentiodynamic polarization curves for copper surface without and with AHA film formed in different concentrations of AHA solutions for 3 h assembly time, in 3 wt% NaCl solutions
Full size table
Table 5 Polarization parameters for the copper without and with AHA film formed in 10 mM AHA solution for different time, in 3 wt% NaCl solution
Full size table

It can be seen from Figs. 4 and 5 that the Tafel slopes greatly change after the addition of AHA. Besides, current densities shift to lower values obviously at the same potential, compared with the bare electrode. Additionally, corrosion potential shifts to the positive direction significantly, indicating that AHA has more pronounced influence in the anodic dissolution process of copper with respect to the cathodic oxygen reduction. It might point out that in Fig. 4 the plot c (5 mM) looks tricky in comparison with others. We repeated the each experiment more than 5× and the results were quite similar. A possible reason was that the best uniform and compact film of AHA at the copper surface just formed under optimal concentration and assembly time.

The comparison of this work with other inhibitors, such as 4-amino-antipyrine (AAP) [30], as well as 2,5-dimercapto-1,3,4-thiadiazole (DMTD) [40], bis-(1-benzotriazolymethylene)-(2,5-thiadiazoly)-disulfide (BBTD) [5], purine (PU) [41], 5-Mercapto-3-phenyl-1,3,4-thiadiazole-2-thione potassium (MPTT) [42], is listed in Table 6. By contrast, it can be seen that the AHA is an effective corrosion inhibitor for copper in 3 wt% NaCl.

Table 6 Comparison of the inhibition efficiencies of different copper inhibitors
Full size table

Microscopic analysis

Figure 6 shows the surface morphologies of AHA-modified copper and bare copper electrode corroded in 3 wt% NaCl solutions for 5 h. In Fig. 6d, after 5 h immersion in high-salt media, the porous structure and rough surface of the bare copper indicate serious corrosion occurrence while the corrosion behavior of the AHA-modified copper surface is less severe under the same condition (see Fig. 6b). The above observation confirms that AHA is an effective inhibitor for copper corrosion, acquired from electrochemical measurement.

Fig. 6
figure 6

Micrographs of copper surfaces with AHA film formed in 10 mM AHA solution for 3 h assembly time (a), and with AHA film formed in same condition as a and exposed to 3 wt% NaCl for 5 h (corrosion time) at 298 K (b). Bare copper before (c) and after (d) exposed to 3 wt% NaCl for 5 h (corrosion time) at 298 K

Full size image

Adsorption isotherm

For further shedding insight on adsorption mechanism of inhibitor interaction with the copper surface, many adsorption modes such as the Langmuir, Temkin and Frumkin isotherms could be investigated [43]. Therefore, the surface coverage (θ) for different concentrations of AHA modification film was calculated from EIS parameters in Table 1 \( \left( {\eta \left( {\% } \right)\text{ = }100{ \times }\theta } \right) \).

It is found that the adsorption of AHA on the copper surface can be described by Langmuir adsorption isotherm equation [44, 45]:

$$ \frac{c}{\theta }\text{ = }\frac{1}{{K_{\text{ads}} }}\text{ + }c $$
(7)

where K ads is the equilibrium constant, θ is the degree of surface coverage on the metal surface and c is the AHA concentration. According to EIS experimental results, a plot of c / θ against c (Fig. 7a) showed a straight line (y = 1.06463x + 0.2659), with the linear correlation coefficient R 2 = 0.9999, and its nearly unit slope confirms that the adsorption of AHA on the copper surface fits Langmuir adsorption isotherm. Additionally, adsorption isotherm drawn from potentiodynamic polarization results in Table 4 is also given in Fig. 7b. A plot of c/θ against c shows a straight line (y = 1.16282x + 0.0452), with the linear correlation coefficient R 2 = 0.9999.

Fig. 7
figure 7

Langmuir adsorption isotherm plot for AHA film formed under the optimized condition at the copper surface in 3 wt% NaCl solution at 298 K according to a EIS results and b potentiodynamic polarization results

Full size image

In addition, the standard free energy of adsorption, \( \Delta G_{\text{ads}}^{0} \), is obtained from the following equation [46]:

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

where R is the general gas constant (8.314 J mol−1 K−1), the temperature in Kelvin (298.15 K) is the absolute temperature, and the value of 55.5 mol L−1 is the concentration of water in pure solution. The calculated K ads values are 3.76 × 103 M−1 and 2.21 × 104 M−1, based on EIS and potentiodynamic polarization results, respectively. The relevant \( \Delta G_{{\text{ads}}}^{\text{0}} \) is −30.4 kJ mol−1(EIS results) or −34.8 kJ mol−1(potentiodynamic polarization results). The large negative value of \( \Delta G_{{\text{ads}}}^{\text{0}} \) indicates that AHA was strongly adsorbed on the copper surface [47, 48].

Raman studies

Figure 8a, b illustrates the normal Raman spectrum of AHA powder and SERS spectrum of the AHA-modified copper electrode formed in 10 mM AHA solution for 3 h assembly time, respectively. Table 7 shows density functional theory (DFT) calculation results for Raman and SERS analysis. The corresponding assignments for Raman spectral analysis were performed on the basis of B3LYP/6-31G(d) calculation. Also, to assign SERS spectrum of AHA, DFT calculation for geometry optimization and vibration modes based on B3LYP/LanL2DZ was performed with model AHA-Cu4 [49]. The optimized geometry model AHA-Cu4 can be seen in Fig. 9.

Fig. 8
figure 8

a Normal Raman spectrum of AHA in powder, b SERS spectrum of AHA film at the copper electrode formed in 10 mM AHA solution for 3 h assembly time, inset showing the SERS spectrum (200–750 cm−1) of the oxide layers

Full size image
Table 7 Assignment for Raman vibrational modes of AHA and SERS vibrational modes of AHA-Cu4 based on DFT calculation
Full size table
Fig. 9
figure 9

The geometry model of AHA-Cu4 optimized by B3LYP/LANI2DZ method

Full size image

It should be mentioned that in inset of Fig. 8 with spectral range from 200 to 750 cm−1, the Raman bands around 528 and 618 cm−1 are from the oxide species of copper, indicating that the oxide layers were inevitably generated in ORC pretreatment process for SERS activity of copper [50].

Combined with the calculation results in Table 7, we can better understand the vibrational modes in Fig. 8a, b. In Fig. 8a, the Raman peaks at 967 and 989 cm−1 represent N–O–H bending in plane and C–H bending in plane. The stretching vibration modes of N–O,C–C,C=O and C–H appear at 1088, 1366, 1619 and 2998 cm−1. The asymmetric stretching vibration mode of C–H appears at 2941 cm−1. The Raman peaks at 1390 cm−1 represent OH bending and C-N bending out of plane.

In Fig. 8b, the peaks at 1011 and 1035 cm−1 represent C–H bending in plane and C–H bending out of plane. The bands with medium intensities at 1305 and 1495 cm−1 could be assigned to N–H bending and C-N stretching, while the strong bands at 1088 and 1600 cm−1 are from N–O stretching and C=O stretching. According to the surface selection rule for SERS [51, 52] and the SERS mechanism [53, 54], the enhanced bands in the SERS spectrum might correspond to either the vibrational moieties attached to the surface or the vibration direction perpendicular to the metal surface; in contrast, the intensities of vibrational modes with parallel polarized components to the surface will be decreased. Thus, the N–O and C=O groups might be perpendicular to the surface, due to their high SERS intensities. The proposed adsorption model of AHA at the copper surface is shown in Fig. 10.

Fig. 10
figure 10

The proposed adsorption model of AHA on the copper surface

Full size image

Quantum chemistry calculations

As mentioned in SERS spectrum analysis, AHA molecule may adsorb at the copper surface by transferring electrons from the N–O and C=O groups to the unfilled hybrid orbital of copper; quantum chemical calculations are used to predict the feasible adsorption centers of a free single molecule on the bare metal surface to confirm the relationship between the AHA molecular structures and its inhibition property. In addition, note that all the calculated parameters obtained in the gas phase and the solvent effects were ignored.

The geometry of the AHA is fully optimized based on the method of B3LYP/LANI2DZ. The illustration of the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) is given in Fig. 11. The quantum chemical parameters, such as E HOMO, E LUMO, dipole moment, μ, the energy gap, ΔE (ΔE = E LUMO − E HOMO), are listed in Table 8.

Fig. 11
figure 11

Molecular orbital plots for acetohydroxamic acid

Full size image
Table 8 Quantum chemical parameter comparison of some molecules in literature with the AHA molecule calculated with DFT method
Full size table

Clearly in Table 8, the high value of E HOMO (E HOMO = −427.64 kJ mol−1) indicates the strong ability of AHA molecule donating electrons to form covalent bond with unoccupied d-orbitals of metal [55], while the low value of E LUMO (ELUMO = −175.86 kJ mol−1) indicates that the AHA molecule has a tendency to accept electrons from d-orbitals of metal to form back-donating bond. The values of the high dipole moment (μ = 5.4125 D) and the low energy gap (ΔE = 251.78 kJ mol−1) facilitate adsorption to the copper surface and, therefore, enhance the inhibition efficiency [56], which is in good agreement with the experimental results.

It can be found in Fig. 11 that the HOMO is located within the region around the N–O and C=O groups, which could be regarded as the feasible sites for interaction with the copper surface. Additionally, the nitrogen atom and two oxygen atoms have large electron densities and their Mulliken atomic charges are −0.302, −0.357 and −0.363, respectively. Thus, AHA can adsorb on the copper surface by donating the electrons from the two O atoms to the d-orbitals of copper.

Conclusions

Consequently, the above investigations confirm that the application of acetohydroxamic acid as an effective corrosion inhibitor for copper in 3 wt% NaCl solution is promising. Some conclusions could be drawn:

  1. 1.

    A relatively compact AHA film could be formed on the copper surface in 10 mM AHA solution for 3 h assembly time and its inhibition efficiency in 3 wt% NaCl solution reached 93.5%, supported by surface morphologies images recorded when the copper with optimized AHA film corroded in 3 wt% NaCl solution for 5 h.

  2. 2.

    Adsorption of AHA on the copper surface fits Langmuir adsorption isotherm.

  3. 3.

    AHA molecules adopted the N–O and C=O groups to attach onto the surface for constructing the protection film.