Direct measurement of the adsorption kinetics of 2-Mercaptobenzothiazole on a microcrystalline copper surface

Jorge A. Ramírez-Cano^*, Lucien Veleva

Research and Advanced Studies Centre of the National Polytechnic Institute of México, (CINVESTAV-Unidad Mérida), Department of Applied Physics, Antigua Carretera a Progreso km. 6, Cordemex, 97310 Mérida-Yucatán, México

*Corresponding author: iqramirez59@gmail.com, ramirezj@mda.cinvestav.mx

1. INTRODUCTIONTOP

Corrosion is a serious problem in many fields of industry, destroying the metal components in structures, electronic devices, vehicles, etc., as a consequence has a great impact on the economy of any country. Thus, the protection of metals and alloys is of paramount importance for large industrial conglomerates and the search for effective means to prevent corrosion is a very active area.

Copper and its alloys are extensively used in industrial applications, ranging from construction to electronics, this metal is present in almost every aspect of modern life (Davis, 2001). Copper is a relatively resistant metal in atmospheric neutral environments, in the absence of chlorides and SO₂, protected by the formed copper patina (Cu₂O). However, it is susceptible to corrosion in aggressive media, when the patina is converted to other corrosion products (Allam et al., 2009). There are several methods to prevent metal corrosion and one of them is the use of inhibitors, organic or inorganic (Aballe et al., 1997).

Organic inhibitors are compounds that, applied in very small concentrations, are capable of effectively reduce the corrosion rate. Many of them are heterocyclic compounds consisting of a π-system and/or containing O, N, or S heteroatoms (Schmitt, 1984; Quraishi and Jamal, 2001; Godínez et al., 2003; Antonijevic and Petrovic, 2008; Khaled et al., 2009; Behpour and Mohammadi, 2012; Pan et al., 2013; Finšgar and Merl, 2014; Winkler et al., 2014; Jakeria et al., 2015). Among these compounds those of the azole group, containing nitrogen and/or sulphur, are considered as effective corrosion inhibitors for copper (Marconato et al., 1998; Subramaninan and Lakshminarayanan, 2002; Finšgar and Merl, 2014; Winkler et al., 2014). Despite the fact that organic inhibitors have been extensively investigated in the last decades, their inhibition mechanism is usually unknown (Trabanelli and Carassiti, 1970; Chadwick and Hashemi, 1979; Zhang et al., 2007; Morales-Gil et al., 2014). It is generally accepted that inhibition is achieved by the formation of a protective film formed by adsorbed inhibitor molecules on the metal surface (Marconato et al., 1998; Subramaninan and Lakshminarayanan, 2002; Godínez et al., 2003; Solmaz et al., 2011; Kazansky et al., 2012), which acts as a barrier between the aggressive media and the metal surface.

However, the kinetics of the formation process of this protective film remains unresolved. There are several techniques that can be used to study the adsorption process of organic molecules on metal substrates.

The most notable techniques are: measurement of mass change, using the electrochemical quartz crystal micro-balance (EQCM), measurement of contact angle, ellipsometry, reflectance absorption infrared spectroscopy (RA-IRS), surface plasmon resonance (SPR), electrochemical impedance spectroscopy (EIS) and atomic force microscopy (AFM)(Subramanian and Lakshminarayanan, 2000).

The EQCM is a valuable tool that enables the acquisition of relevant data about the adsorption of chemical species (molecules or ions) on metallic substrates in the microscale and characterization of their adsorption kinetics (Eickes et al., 2000; Telegdi et al., 2000). It consists of a quartz resonator, sandwiched between two metal electrodes, these electrodes produce an oscillating electric field, causing vibrational motion of the crystal at its fundamental frequency. The frequency is sensitive to the mass change of the resonator and when adsorption or desorption of mass occurs, a frequency shift is detected. Thus, adsorption will be registered as a frequency decrease, while desorption will produce a frequency increase.

The aim of this work is to enlarge the information about the adsorption of 2-Mercaptobenzothiazole (2-MBT, C₇H₅-NS₂) see Fig. 1, a member of the azole’s group, used as corrosion inhibitor for copper (Winkler et al., 2014). The research was carried out employing the EQCM to obtain the frequency response over time, the adsorption isotherm, the fraction of surface coverage and the free energy of 2-MBT adsorption.

2. MATERIALS AND METHODSTOP

Commercially available reactive grade (96%) Aldrich 2-Mercaptobenzothiazole, was used without further purification to prepare solutions with distilled water. The concentration range of 2-MBT was 1×10⁻⁶ to 1×10⁻¹ M (pH=6.5).

The EQCM working electrodes are manufactured by Gamry Instruments and they consist of AT-cut quartz crystals coated with Cu, having an active area of 0.205 cm² and a sensitivity factor C_f of 226 Hz cm² μg⁻¹, having a fundamental frequency of 10 MHz. All experiments were carried out in a Gamry Teflon Cell, specifically designed to be used with the EQCM working electrodes.

2.1. Frequency response over timeTOP

During 2-MBT adsorption process, conducted at 22 °C, EQCM 10M Gamry frequency monitor was used to record the frequency response during 3600 seconds, inside a Gamry Vista Shield Faraday Cage.

2.2. Adsorption isothermTOP

Frequency variations (Δf) detected by the EQCM are due to several contributions:

Δf=Δf_d+Δf_m+Δf_a+Δf_x    (1)

where Δf_d is the portion of the frequency shift related to viscous damping; Δf_m is caused by adsorbed mass; Δf_a is surface stress and Δf_x is caused by energy dissipation.

As can be seen the frequency shift (Eq. 1) is the result of a very complex process and it does not depend of only the mass change. Therefore, obtaining surface coverage information from frequency variation data is at least problematic. However, the time dependence of this frequency response is the result of the formation of an adsorbed monolayer, therefore the time domain measurements are sensitive to adsorption kinetics of the monolayer and thus, surface coverage information can be extracted from frequency response over time data (Karpovich and Blanchard, 1994). The obtained frequency shift data were fitted to the Langmuir isotherm (Langmuir, 1918; Karpovich and Blanchard, 1994; Subramanian and Lakshminarayanan, 2000; Marczewski, 2010; Haerifar and Azizian, 2013):

θ(t−t₀)=K’[1−e^(k_m)(t−t0)]    (2)

where K’=C[C+ (kd)(ka⁻¹)]⁻¹; θ is the surface coverage; t is time; t₀ is the diffusion time and k_m is the rate constant of adsorption and it is related to the adsorption and desorption constants by the following relation:

k_m=k_aC+k_d    (3)

being k_a the adsorption constant, k_d the desorption constant and C the inhibitor concentration. A linear fit of k_m vs C gives k_a as the slope of the line and k_d as the intercept.

2.3. Fraction of surface coverage and free energy of adsorptionTOP

Surface coverage (θ) was calculated using the following equation:

Θ=C[C+K_eq⁻¹]⁻¹ | K_eq=(k_a)(k_d)⁻¹    (4)

where C is the inhibitor concentration and K_eq is the equilibrium constant (Karpovich and Blanchard, 1994). The free energy of adsorption ΔG_ads was found directly from equilibrium constant data, using the thermodynamic relation (Karpovich and Blanchard, 1994):

ΔG_ads=−RTlnK_eq    (5)

where R is the ideal gas constant and T is temperature.

3. RESULTS AND DISCUSSIONSTOP

3.1. Frequency response over timeTOP

The frequency response for the adsorption of 2-MBT on copper is shown in Figs. 2–7. The curves were obtained inverting the Y-axis (−Δf vs t). The observed frequency shift follows a well-defined exponential law, when 2-MBT concentration was in the range of 1×10⁻¹ to 1×10⁻² M (Fig. 2 and Fig. 3). The exponential behavior starts decaying at 2-MBT concentration 1×10⁻³ M (Fig. 4).

At 1×10⁻² to 1×10⁻⁶ M 2-MBT concentrations (Fig. 3 to Fig. 7), an initial desorption process was observed. This desorption process was attributed to the substitution of water molecules by 2-MBT molecules on the copper surface, according to the model described by Kern and Landolt (2001).

Desorption region increases as the concentration of 2-MBT decreases. It is clearly visible that a three step process takes place. In the first step 2-MBT molecules reach the metallic surface and force water molecules to desorb (initial desorption).

The second step is the adsorption of 2-MBT molecules at the active copper sites that are now vacant, due to the water desorption. It is considered that initially 2-MBT molecule is adsorbed in the metal surface through the exo-sulfur atom (Oshawa and Süetaka, 1979; Subramaninan and Lakshminarayanan, 2002; Kazansky et al., 2012; Ramenzanzadeh et al., 2014) this fact is possible because 2-MBT molecules exhibit the thiol-thione tautomerism (Oshawa and Süetaka, 1979; Ramenzanzadeh et al., 2014), in the range of pH of 3 to 9 (Jing et al., 2013), which leads to an anionic form of 2-MBT (Winkler et al., 2014) (Fig. 8).

The final step of the adsorption is the rearranging of 2-MBT molecules on the metal surface, as proposed in (Kokalj et al., 2010), for molecules containing bencene rings. This molecule rearranging gives place to more water desorption, leaving unoccupied active sites on the metal surface, in the vicinity of the already adsorbed 2-MBT molecules.

Due to this fact, the new active sites are rapidly occupied by other inhibitor molecules available in the solution.

However, when 2-MBT concentration is too low (i.e., 1×10⁻⁴ to 1×10⁻⁶ M), there are not enough molecules to adsorb and compensate the mass loss, due to water desorption and thus, the resulting frequency response corresponds only to a desorption process (Fig. 5 to Fig. 7).

At concentration higher than 1×10⁻⁴ M 2-MBT (Fig. 2 to Fig. 4), there are enough inhibitor molecules to take the vacant sites left by water and therefore, no subsequent desorption is observed.

3.2. Adsorption isothermTOP

The frequency response of the three higher concentrations (1×10⁻¹ up to 1×10⁻³ M 2-MBT) was fitted to the Langmuir kinetic model (Eq. 2). These concentrations were chosen because 2-MBT adsorption presents a well-defined exponential behavior (Figs. 2–4).

The fitted curves are presented in Figs. 9–11 and their respective calculated parameters are shown in Table 1.

The correlation coefficients show a very good agreement between the experimental data and the Langmuir isotherm. Figure 12 presents a plot of k_m vs C and its linear regression, which yields a k_a value of 8.21×10⁻³ M⁻¹ s⁻¹ and a k_d value of 8.38×10⁻⁴ s⁻¹.

Previously reported investigation (Karpovich and Blanchard, 1994) states that the negative deviation from linearity in the relation between k_m and C (for a thiol system, concentration range from 3×10⁻⁶ to 3×10⁻³ M) does not indicate that this system is ill-described by the Langmuir adsorption model. This fact only shows the intrinsic limits to this treatment of a complex equilibrium process. In other words, for low adsorbent concentration, the system remains in close proximity to equilibrium and the expected mass change for such conditions lies beneath our limit of detection.

3.3. Fraction of surface coverage (θ) and free energy of adsorption (ΔG_ads)TOP

According to equations 4 and 5, the equilibrium constant for 2-MBT adsorption process is K_eq=9.79, and the fraction of surface coverage (θ) values, as a function of adsorbent concentration, are shown in Table 2.

The free energy of adsorption was found to be −5.59 kJ mol⁻¹. It is considered that a chemisorbed molecule must have a free energy of −40 kJ mol⁻¹ (Atkins, 1999). Thus, the ΔG_ads energy found for 2-MBT on copper surface clearly indicates that the adsorption is of physical nature, probably electrostatic.

4. CONCLUSIONSTOP

2-MBT adsorbs on copper following the Langmuir adsorption isotherm. The initial desorption process observed in the concentration range 1×10⁻³ to 1×10⁻⁶ M is probably due to water molecules being substituted by 2-MBT molecules at the metal surface. The kinetic parameters of 2-MBT adsorption process showed a very low rate in the range of 2.48×10⁻⁴ to 1.60×10⁻³ s⁻¹ and low surface coverage of 0.01 to 0.50. These facts are probably due to the lack of aggressive ions, which can deteriorate the copper patina, forming copper cations and exposing a pure metal surface to 2-MBT. Copper cations make possible the interaction with the anionic form of 2-MBT, leading to strong interactions, obtaining as a consequence higher adsorption rate and better surface coverage. Our research results showed that the calculated free energy of adsorption (−5.59 kJ mol⁻¹) corresponds to a physisorption process, probably of electrostatic nature for the interaction between 2-MBT and copper surface in aqueous solution.

ACKNOWLEDGMENTSTOP

The authors would like to express their gratitude to the Mexican Council of Science and Technology (CONACYT), grant 179110, for the financial support of this study.

REFERENCESTOP