When a monochromatic light is scattered from a molecule, most photons are elastically scattered with the same frequency (energy) as the incident photons. A small fraction of light is scattered at frequencies different from those of the incident photons, namely inelastic scattering. The elastic scattering is named Rayleigh scattering while the inelastic scattering is called Raman scattering, or Raman effect. The Raman effect was discovered in 1928 by V. C. Raman.

The Raman effect arises when incident photons interact with the electric dipole of a molecule. The photons excite the molecule from its ground state to a virtual energy state. The molecule emits a photon when it relaxes and returns to a different energy state. The energy difference between the two states results in the energy of the incident photons being shifted away from the excitation wavelength. If the final energy state of the molecule is higher than the initial state, then the emitted photon will be shifted to a lower frequency, which is designated as Stokes shift. If the final energy state is lower than the initial state, then the emitted photon will be shifted to a higher frequency, and this is designated as an anti-Stokes shift (Figure 20.1). The Stokes shift is what is usually observed in Raman spectroscopy.

The Raman effect is based on molecular deformations in the electric field (E) generated by the laser beam. Assuming that the laser is fluctuating at frequency ν, E can be written as a function of the amplitude, E₀, and the time, t:

$E=E_0\text{cos}\left(2\pi \nu t\right)$

The dipole moment (μ) of a molecule induced by this light is, therefore:

$\mu =\alpha E=\alpha E_0\text{cos}\left(2\pi \nu t\right)$

Here, the proportionality constant α is called the polarizability of the molecule. The polarizability represents the ease with which the electron cloud around a molecule can be deformed.

Similarly, if the molecule is also vibrating or rotating in sinusoidal fashion with frequency ν_m, the nuclear displacement Δr can be written as:

$\Delta r=r_0\text{cos}\left(2\pi {\nu }_{\text{m}}t\right)$

where r₀ is the vibrational amplitude.

When a molecule is at its equilibrium nuclear geometry, the polarizability of the molecule is α₀. While at some distance, the instantaneous polarization α is a linear function of the nuclear displacement, Δr:

$a=a_0+\left(\frac{\partial a}{\partial \text{r}}\right)\Delta r$

where the derivative (∂a/∂r) represents the change in polarizability with change in position.

Combining Equations (20.2)–(20.4), we have:

$\begin{array}{c}\hfill \mu =\alpha E=\alpha E_{0}cos\left(2\pi \nu t\right)\hfill \\ \hfill ={\alpha }_0{E}_{0}cos\left(2\pi \nu t\right)+\left(\frac{\partial a}{\partial \text{r}}\right)r_0{E}_{0}cos\left(2\pi {\nu }_{\text{m}}t\right)cos\left(2\pi \nu t\right)\hfill \\ \hfill ={\alpha }_0{E}_{0}cos\left(2\pi \nu t\right)+\frac{1}{2}\left(\frac{\partial a}{\partial \text{r}}\right)r_0{E}_0\left\{cos\left[2\pi t\left(\nu -{\nu }_{\text{m}}\right)\right]+cos\left[2\pi t\left(\nu +{\nu }_{\text{m}}\right)\right]\right\}\hfill \end{array}$

The first term relates to the scattered photon that has the same frequency as the incident photon, and thus explains Rayleigh scattering. The second term relates to the scattered photons with increased frequency (ν − ν_m) and decreased frequency (ν + ν_m) as compared to the incident photon, representing Stokes shift and anti-Stokes shift, respectively.

Equation (20.5) also gives a general selection rule that governs Raman scattering. Notice the derivative (∂a/∂r) in the second term. If that derivative equals zero, the entire second term will be zero, and there is no Raman scattering. Therefore, the selection rule for Raman scattering can be given as:

$\left(\frac{\partial a}{\partial \text{r}}\right)\ne 0$

that is, a molecular motion will be Raman-active only if the motion occurs with a change in polarizability. The amount of the polarizability change determines the Raman scattering intensity.

20.2.2 IR spectroscopy

Infrared radiation encompasses a section of the electromagnetic spectrum with wavenumbers ($\overline{\nu }$) from 12,800 to 10 cm^− 1 or wavelengths (λ) from 0.78 to 1000 μm. This infrared portion of the electromagnetic spectrum is usually divided into three subregions: near-IR region (12,800-4000 cm^− 1), mid-IR region (4000-200 cm^− 1), and far-IR region (200-10 cm^− 1). The mid-IR region is the most commonly used region in IR spectroscopy.

Infrared spectroscopy is based on the interaction between IR radiation and molecules that results in the absorption of IR radiation by the molecules. At temperatures above absolute zero, all the atoms in molecules are in continuous vibration with respect to each other. When the frequency of a specific vibration equals the frequency of the incident IR radiation, the molecules absorb the radiation. The absorbed radiation matches the transition energy of the bond or group that vibrates in the molecule.

A molecule can vibrate in many ways, with each way called a vibrational mode. Simple diatomic molecules have only one bond and thus one vibrational mode that is not observed if the molecule is symmetrical. Polyatomic molecules have more bonds and correspondingly more vibrational modes. The total number of fundamental vibrational modes can be determined as follows. A polyatomic molecule with n atoms has 3n total degrees of freedom because each atom has three degrees of freedom. For a nonlinear molecule, 3 degrees of freedom are rotational and 3 are translational. Therefore, there are 3n − 6 degrees of freedom that are fundamental vibrations for nonlinear molecules. For linear molecules, because only 2 degrees are rotational, the total number of fundamental vibrations is 3n − 5. However, not all the vibrations can be observed in IR spectra. For a mode to be observed, changes must occur in the permanent dipole:

$\left(\frac{\partial \mu }{\partial r}\right)\ne 0$

where μ is the dipole moment and r is the normal coordinate. In general, the larger the dipole change, the stronger the intensity (I) of the band in an IR spectrum:

$I={\left(\partial \mu /\partial r\right)}^2$

A comparison of Raman and IR spectroscopy is shown in Table 20.1. The major difference between the two techniques lies in that Raman spectroscopy determines vibrations that cause a change in the polarizability of a molecule, while IR spectroscopy determines vibrations that cause a change in the dipole moment of a molecule. However, in theory, there are some vibrations that induce changes in both polarizability and dipole moment.

There are many variations of Raman spectroscopy that have been developed to enhance the sensitivity (e.g., surface-enhanced Raman), to improve the spatial resolution (Raman microscopy), or to acquire very specific information (resonance Raman). For in situ studies in corrosion science, normal Raman spectroscopy is sufficient in most cases, but sometimes surface enhanced Raman spectroscopy (SERS) is necessary. Therefore, other techniques (e.g., resonance Raman) are not further covered here.

20.3.2 IR spectroscopy

There are two primary experimental techniques in IR spectroscopy, transmission and reflection techniques. For in situ studies of corrosion phenomenon, reflectance spectroscopy is the commonly used technique. Among the various types of reflectance techniques, specular reflection, infrared reflection-absorption spectroscopy (IRAS), and attenuated total reflection (ATR) are well suitable for in situ studies.

20.3.2.1 Cells with optical windows for IR path

Because of the intense absorption of IR light by solvent (water), IR spectroscopy cannot be used as readily as Raman spectroscopy for in situ monitoring corrosion phenomenon in aqueous environments. Therefore, specific cells have to be designed with the aim of minimizing radiation losses by solvent absorption. One approach is by depositing very thin layers of electrolyte or small droplets on metal surfaces, which represents the common configuration for atmospheric corrosion studies. Figure 20.2 shows two typical experimental setups for in situ studies of atmospheric corrosion, with the one in Figure 20.2a for specular reflectance study and the one in Figure 20.2b for IRAS. For the IRAS studies, grazing angle incidence is usually used.

For studies in bulk solutions, IR reflectance spectra can still be obtained by reducing the distance between the sample and the IR window (Figure 20.2c). The IR transparent window is pressed against the sample, creating an extremely thin solution layer in between them. One problem with this configuration is the possible depletion of solution species due to electrochemical activities (e.g., corrosion) in the confined environment. To solve this problem, a flow cell can be designed to replenish solution species as reaction proceeds.

For all three configurations in Figure 20.2, the IR windows are in direct contact with liquid electrolyte or high-humid environment. Therefore, non-water-soluble IR window materials have to be used, such as zinc selenide (ZnSe) and calcium fluoride (CaF₂).

20.3.2.2 ATR

When an IR radiation enters from optically denser ATR crystals (refractive index n₁) to the optically rarer samples (refractive index n₂) at an angle exceeding the critical angle (θ_c), total internal reflection will occur inside the crystal. The critical angle is defined as:

${\theta }_{\text{c}}={sin}^{-1}\left(\frac{n_2}{n_1}\right)$

  (20.9)

This internal reflectance creates an evanescent wave that extends beyond the surface of the crystal into the sample that is in contact with the crystal. The evanescent wave protrudes only a few microns (0.5-5 μm) beyond the crystal surface and into the sample.

Some common ATR crystals are ZnSe, thallium bromide-thallium iodide (KRS-5), germanium (Ge), or silicon (Si). Prisms (Figure 20.3a) and hemispheres (Figure 20.3b) made of ATR crystals, for example, ZnSe, can be used for ATR-FTIR studies. In these configurations, IR beam is reflected once inside the crystal. Multiple internal reflection can be achieved by using the configuration shown in Figure 20.3c. In all three configurations, samples have to be in close contact with the crystals or can be plated on the crystals as thin films.

20.4.1 Aqueous corrosion

The corrosion of metallic materials in aqueous environments is one of the most prevalent forms of corrosion and has been extensively studied over the past few decades. In situ Raman spectroscopy is well suited for monitoring the corrosion process in metal/liquid interface. A great number of in situ Raman spectroscopic studies have been conducted to characterize the anodic corrosion films or monitor the formation of corrosion products on metal surfaces in corrosive environments.

20.4.2 Atmospheric corrosion

Atmospheric corrosion proceeds under thin electrolyte layers or small droplets that form by condensation or absorption of water by airborne sea salt particles. The small amount of electrolyte facilitates the use of Raman spectroscopy to monitor the corrosion process in situ. Corrosion of zinc caused by predeposited NaCl particles with a density of ~ 0.4 μg/cm² serves as an example.³⁸ Zinc chloride (ZnCl₂) with a Raman band at 290 cm^− 1 was detected on the Zn surface 2 h after the sample was exposed to high humidity (84%). After 6 h exposure, when ZnCl₂ in the surface layer reached saturation, simonkollite (ZnCl₂(Zn[OH]₂)₄) with major Raman bands at 255 and 390 cm^− 1 was detected. The corrosion process under a single droplet was also studied for Zn³⁹ and carbon steel.^40–42 The Raman spectra from Zn under a seawater droplet showed the formation of ZnO, a sulfate-containing and a carbonate-containing compound.³⁹ The Raman spectra from carbon steel under single NaCl or Na₂SO₄ droplets indicated green rust and γ-FeOOH as the initial rust formation during sea salt droplets-induced marine atmospheric corrosion.^40–42 The spatial distribution of the corrosion reaction products was also revealed by in situ Raman spectroscopy, as shown in Figure 20.4.⁴¹ Ferrous chloride (FeCl₂) and also possibly its chloride ions (245 and 286 cm^− 1) were detected inside the corrosion initiation site or the corrosion pit, as the corrosion reaction products (Figure 20.4a). Green rust (GR1(Cl⁻)) (425 and 501 cm^− 1) was identified as the corrosion product formed in the vicinity immediately outside of the pit, while γ-FeOOH (246, 377, 525, and 645 cm^− 1) was detected in regions outside of the GR region (Figure 20.4b and d). In the transitional region from GR to γ-FeOOH, Fe₃O₄ (666 cm^− 1) also formed together with GR and γ-FeOOH (Figure 20.4c).

20.4.3 Corrosion inhibitors

Corrosion inhibitors are chemical compounds that are added to the surroundings (liquids or gases) of metallic materials to decrease their corrosion. Inhibitors work by the formation of a coating on metals, which prevents access of corrosive substance to the metal surfaces. The inhibitor molecules either attach to the metal or react with the surface to form a thin, adherent layer. In both cases, in situ Raman spectroscopy has been employed to study the adsorption mechanism and the film formation mechanism.

Benzotriazole (BTAH) has been used as an effective corrosion inhibitor for Cu and its alloys since the 1950s,⁴³ and was also extended to other metals. The high corrosion inhibition efficiency of BTAH is attributed to the formation of a compact, polymerlike metal-BTAH complex on the metal surface.^44–46 The structure and composition of this complex can be revealed by in situ SERS, which provides information on even only a few monolayers of adsorbed species. Most of the in situ SERS studies on BTAH as corrosion inhibitors have been focused on Cu and Ag due to their strong Raman-enhancing effect. The inhibition effect of BTAH on Cu has been studied in various environments, such as halide solutions,^44,46–48 sulfuric acid media,^49–52 organic solution,⁵³ and ionic liquids.⁵⁴

Earlier studies showed that, in near neutral KCl solution, the complex formed by the reaction between Cu and BTAH was [Cu(I)BTA] according to in situ Raman data. In KCl/acid solution, the Raman spectra obtained at less negative potentials resembled that of [Cu(I)BTA] complex but, at more positive potentials the spectra were similar to that of [Cu(I)-ClBTAH]₄.⁴⁴ A recent study reported that, in neutral KCl solution, the initially formed complex was [Cu^I(BTA)]_n and might have transformed to [Cu^I(BTAH)]₄ at more negative potentials. This was evidenced by the redshifts of the in-plane trigonal breathing mode from 1041 cm^− 1 at − 0.5 V to 1021 cm ^− 1 at − 1.1 V and the increase in intensity of the NH in-plane bending mode (1144 cm^− 1) at more negative potentials (Figure 20.5).⁴⁶ Therefore, it is probably necessary to take into account the effect of the presence of the [Cu^I(BTAH)]₄ or [Cu(I)-ClBTAH]₄ complexes in the corrosion inhibition of Cu in mild acidic KCl solution containing BTAH. In relatively strong acidic environment (e.g., H₂SO₄), [Cu(I)BTA]_n complex was found as the film formed on the surface of Cu and its alloys.

Similar in situ SERS results were observed on Ag in either halide media or acetonitrile solution with BTAH.^45,55 In addition, in situ SERS studies were also extended to Fe, Ni, and Co by proper electrochemical roughing procedures.^46,56

Other inhibitors that have been studied using in situ SERS include triethylstibine for Fe and Ni,⁵⁷ propargyl alcohol for Fe,⁵⁸ 2-mercaptobenzothiazole (MBT) for Cu,⁵⁹ phytic acid (IP6) for Ag,⁶⁰ volatile corrosion inhibitors for carbon steel,⁶¹ salicylate for Cu,⁶² and benzyldimethylphenylammonium chloride (BDMPAC) for low carbon steel,⁶³ methimazole (MMI) for Cu,⁶⁴ 2-amino-5-mercapto-1,3,4-thiadiazole (AMT) for Ag⁶⁵ and Co,⁶⁶ 3-amino-5-mercapto-1,2,4-triazole (AMTA) for Fe,⁶⁷ and 2-amino-5-(4-pyridinyl)-1,3,4-thiadiazole (4-APTD) for Cu.⁶⁸

20.4.4 Coatings

Coatings including both organic and inorganic coatings are an effective approach to achieve corrosion protections for metals. In situ Raman spectroscopy is a powerful tool for the study of corrosion-protection coatings. First, the formation of coatings on metallic materials could be monitored using in situ Raman spectroscopy. Second, the changes and degradation of coatings especially organic coatings upon exposure to corrosive environments can be studied using in situ Raman spectroscopy. Last, the corrosion products underneath the coatings can be characterized through in situ Raman spectroscopy provided that the coatings are thin and transparent.

20.4.4.2 Polymeric coatings

Raman spectroscopy is especially suitable for studying polymers thanks to its high sensitivity towards the structure and conformation of the polymer backbone. Therefore, in situ Raman spectroscopy could be used to follow the growth of polymeric coatings on metals and their degradation in corrosive environments. In addition, this technique enables monitoring the corrosion product formation under the coatings provided that the coatings are very thin and transparent. The thin coating also reduces the time frame for corrosion to occur to match with that of Raman acquisition. Another requirement is the absence of fluorescence caused by the epoxy coating. Finally, it is better if the Raman bands of the coating do not interfere with those from the corrosion products.⁷⁷

20.4.4.2.1 Epoxy coating

The earliest applications of in situ Raman spectroscopy in the study of polymer coatings were discussed in a series of papers on the corrosion of painted galvanized steel.^77–80 After being immersed in NaCl solutions for a certain time, the epoxy-polyamide (15-20 μm) coated galvanized steel developed two clearly different regions, with and without blisters. Two types of blisters were observed, black and large blisters and white and small blisters.⁷⁷ Raman spectra of the corrosion products inside the blisters could be obtained directly through the thin paint layer. The Raman spectra from the nonblistered regions corresponded to epoxy-polyamide resin coating with characteristic bands at 570 and 637 cm^− 1. The Raman spectra from inside the black blisters showed the presence of zinc hydroxychloride with sharp bands at 260 and 390 cm^− 1, while those from the white blisters indicated the existence of amorphous zinc oxide with a prominent band at 569 cm^− 1.^78–80 It was believed that the black blisters correspond to the cathodic zones and the white blisters are anodic. This study was also extended to the case of zinc-coated steel samples with chromate or phosphate conversion layers.⁷⁹

20.4.4.2.2 Electronically conducting polymer coating

Recently, there has been an increasing interest in the use of electronically conducting polymers (ECPs) for corrosion resistant coatings. Some of the ECPs that have been investigated for corrosion protection include polyaniline (PANI), polypyrrole (PPy), and polythiophene (PT).

The use of PANI for steel corrosion protection was first reported in the early 1980s.^81,82 Subsequently, a substantial amount of research has been devoted to the use of PANI for corrosion protection of various metals and alloys. The conventional approach to obtain the protective PANI layers was by electrochemical polymerization. In situ Raman spectroscopy has been extensively used to study the formation mechanisms of the PANI films,^83,84 normally on platinum electrodes. Even though not directly related to corrosion study, these studies laid the foundation for studying the protective PANI coating on metals using in situ Raman spectroscopy.

In situ Raman spectroscopy combined with cyclic voltammetry was employed to investigate the interaction between the passive layer on iron and PANI coating in sulfate medium.⁸⁵ The oxidation state of the PANI film on iron, which changed after voltammetric cycling, could be followed. After one redox cycle (0 to − 1.0 V), the PANI film deprotonation was verified by the depletion of the band at 1330 cm^− 1 (CN⁺ stretching). Also, the depletion of benzenoid units was evident due to the decrease of the 1620 cm^− 1 band (CC ring stretching) and the depletion of 1185 cm^− 1 band (quinine ring CH bending).⁸⁵ After further cycling, the Raman spectra evolved toward a less oxidized form of PANI with a predominance of benzenoid rings (1185 and 1620 cm^− 1). The reduced PANI on iron is no longer able to provide enough electrons to maintain iron in its passive state, and the passive layers break down. They also analyzed the PANI layer potentiostatically polymerized in phosphoric/metanilic solution as compared to the PANI layer grown in inorganic acid (HCl) by collecting Raman spectra at different polarization potentials (e.g., from 150 to − 700 mV/SSE).⁸⁶ They found that the film prepared in metanilic acid had a more oxidized and less protonated state than that grown in inorganic acid solution, as evidenced by the strong Raman bands at 1330 cm^− 1 (CN^+·) and 1500 cm^− 1 (C = N^+·) observed from the PANI film formed in metanilic acid. Therefore, the reduction of the PANI film is more difficult in the presence of metanilic acid, resulting in higher protective properties.

Another strategy for protecting metals using PANI is by forming epoxy and acrylic blends. The acrylic blend formed by PANI, poly(methyl methacrylate) (PMMA), and camphorsulfonic acid (CSA) is an example,^87–90 which had a low percolation threshold, a high conductivity, and enhanced mechanical properties as compared to PANI. Raman spectra of the PANI-PMMA-CSA blend indicated characteristic features of “secondary doped” PANI,⁹¹ with a significant increase of the band at 1333 cm^− 1 corresponding to the C = N^+· mode. The ratio of the band at 1333 cm^− 1 and that at 1599 cm^− 1 corresponding to the aromatic CC mode is equal to ~ 1, indicating that the film is in the emeraldine state. However, it diminished to ~ 0.7 after the coating (on iron) was immersed in H₂SO₄ solution for 20 days, suggesting the reduction of the polymeric coating.⁸⁷ This ratio varies for different metal substrates, which indicates that the extent of the redox reaction in the film depends on the metals that have different reducing power (Figure 20.6).⁸⁸ Raman data also showed the presence of a second layer formed by counter ions released from PANI and various metal cations.

Another ECP that has been widely used for corrosion protection coatings is PPy. The formation or the electropolymerization of the PPy films on various metals could be followed by using in situ Raman spectroscopy.^92–96 Raman bands characteristic of the oxidized form of PPy at 927, 1086, 1238, 1368, and 1605 cm^− 1 were observed during the growth of PPy film, and those typical of the reduced form of PPy at 988, 1038, 1260, 1313, and 1564 cm^− 1 were detected during corrosion in corrosive medium.⁹⁴

Take the thin PPy film on iron electrode and immersed in a 0.1 M K₂SO₄ solution (pH = 4) as an example.⁹⁷ Raman spectra were recorded during immersion and at different OCPs to monitor the redox states of the PPy and composite films. The Raman data revealed that the PPy film was in the oxidized form in the original state (E_oc = 0 V) and was protective to the iron substrate. Then, the OCP dropped to − 0.4 V during protection and finally to − 1.0 V after corrosion. During this process, the Raman bands at 1606 cm^− 1 (inter-ring C = C) and 1385 cm^− 1 (ring C = C) corresponding to the oxidized PPy both shifted towards lower wavenumbers, ~ 1570 and 1315 cm^− 1, respectively, for the reduced form of PPy (Figure 20.7a). The bands for the defect modes at 929, 1410, and 1240 cm^− 1 also decreased. The Raman bands which increased significantly during the OCP measurement are at 985 and 1050 cm^− 1, which correspond to the reduced form of PPy. As compared to the PPy film alone, the PPy-PDAN composite film showed better ability to protect iron because the Raman bands that are typical of the oxidized form of PPy (929, 1240, 1410, and 1610 cm^− 1) decreased more slowly upon reduction (Figure 20.7b).⁹⁷

20.5.1 Aqueous corrosion

It is difficult to probe corrosion phenomenon in bulk solutions using in situ FTIR due to intense absorption of IR radiation by water. However, by reducing the distance between the sample surface and the IR window, the corrosion process on the metal surface could be followed. The electrochemically controlled oxidation and reduction of lead in sulfuric acid was successfully monitored using in situ IRAS.⁹⁹ The growth and reduction of lead sulfate (PbSO₄) was followed by monitoring the band at 631 cm^− 1, the integrated area of which showed a linear relationship with the consumed charge in the potential region where PbSO₄ forms. At more anodic potentials, the formation of lead dioxide (PbO₂) was observed as evidenced by an increased absorption band at 5200 cm^− 1. Similar experiments were conducted to characterize the corrosion products formed on Ni electrodes in aqueous solutions containing pseudohalide ion series (OCN⁻, SCN⁻, and SeCN⁻).¹⁰⁰ Ni corroded in all three solutions with the formation of Ni(II) cyanate complex, Ni(II) thiocyanate complex, and Ni(II) selenocyanate complex depending on the ion species in the solution. In the Ni/OCN⁻ system, the in situ data obtained using subtractively normalized interfacial Fourier transform infrared spectroscopy (SNIFTIRS) also showed the formation of CO₂ (2343 cm^− 1), which was believed to be the ultimate product of electrolysis of the cyanate ion.

Other in situ FTIR studies of corrosion-related phenomena include anion adsorption on colloidal chromium (III) oxide hydroxide film simulating passive stainless steel surface,¹⁰¹ characterization of the anodic film on iron in neutral phosphate solution (Fe^IIIPO₄),¹⁰² and monitoring of the transport of corrosive species through the mortar layer.¹⁰³

20.5.2 Atmospheric corrosion

Given that atmospheric corrosion occurs under extremely thin electrolyte layers, there is less absorption of IR radiation by water. Therefore, a great amount of work has been conducted to study atmospheric corrosion using in situ IRAS techniques. In situ IRAS has been developed to study the corrosion of metals (e.g., Cu, Zn, Al, and Mg) exposed to humid air¹⁰⁴ or humid air mixed with corrosive gases, such as SO₂,^104–110 NO₂,^{107,109–111} O₃,¹⁰⁹ and SO₃.¹¹² The corrosive environments were generated by passing air with humidity and certain levels of corrosive gases through the sample chamber. The interaction of metals with corrosive environments and the formation of corrosion products could be followed using in situ IRAS. For example, the initial corrosion products on Cu exposed to humid air (90% RH) were identified as copper(I) oxide while that formed in humid air containing SO₂ (0.23 ppm) was sulfite.¹⁰⁴ Similarly, the corrosion products formed on Ni,¹⁰⁵ Zn,¹⁰⁵ and Al¹¹² exposed to SO₂-containing humid air were also characterized using in situ IRAS. As complementary to in situ IRAS, quartz crystal microbalance (QCM)^106,109,111 and atomic force microscopy (AFM)^108,109 were the other two in situ techniques used for the study of humid air-induced atmospheric corrosion, to obtain the mass changes and morphological information, respectively.

The influence of organic constituents such as acetic acid, acetaldehyde, and formic acid on the atmospheric corrosion of metals, especially in indoor environments, was also studied using in situ IRAS.^113–117 Zinc acetate was detected as the major corrosion products in humid environments with sub-ppm concentration of acetic acid and acetaldehyde,^113,114 while zinc formate formed on zinc exposed to environments containing formic acid.¹¹⁵

Atmospheric corrosion of metals is greatly enhanced by NaCl deposition, which can originate from either airborne sea salt particles or deicing salts. NaCl particle- induced atmospheric corrosion has been studied using in situ IRAS for Zn,^118,119 Al,¹²⁰ Fe,¹²¹ Cu,^122,123 and Mg.¹²⁴ For Zn with predeposited NaCl particles and then exposed to high humidity (> 90%), a surface film containing ZnO, Zn₅(OH)₈Cl₂·H₂O, and zinc Zn₅(OH)₆(CO₃)₂ was detected by in situ IRAS.¹¹⁸ The rate of the corrosion process could also be evaluated from the rate of surface film formation obtained from the IRAS measurements. In situ IRAS also detected carbonate ions (1390 cm^− 1) in the thin solution films in the secondary spreading region close to the original NaCl droplet.¹¹⁹ Similar secondary spreading effects were observed on Cu and studied using in situ IRAS.^122,123 Another important application of in situ IRAS in atmospheric corrosion studies is the investigation of filiform corrosion of Al¹²⁰ and Fe.¹²¹ The movement of the active filament head on coated Al surface could be followed with in situ FTIR microspectroscopy using the characteristic IR band at ~ 2500 cm^− 1 from Al(H₂O)₆^3 +, the corrosion product in the filament.¹²⁰ Notice that the coating on Al must be thin and transparent in the spectral region of interest.

20.5.3 Corrosion inhibitors

In situ FTIR is a powerful tool for the investigation of the mechanism of corrosion inhibitors. In combination with electrochemical techniques, SNIFTIRS was used to study the formation of BTAH on Cu(100) surface.¹²⁵ The Cu(I)BTA complex was detected on Cu surface at potentials >−0.3 V as evidenced by the negative-going bands at 1119 and 1155 cm^− 1. The bands assigned to BTAH in the solution were observed as the positive-going bands at 1014, 1217, 1248, 1268, and 1308 cm^− 1. The effects of anions (Cl⁻ and SO₄^2 −) in the supporting electrolyte on the formation and decomposition of the Cu(I)BTA film on Cu were also investigated using in situ FTIR.¹²⁶ It was found that the Cu(I)BTA complex was formed more easily in Cl⁻-containing solution than in sulfate or bisulfate solutions. Other corrosion inhibitors that have been studied using in situ FTIR include BDMPAC, glutamic acid, triethylenetetramine (TETA), sodium tartrate, and sodium benzoate.¹²⁷

20.5.4 Coatings

In situ FTIR studies of corrosion protection coating systems have been limited to polymeric coatings. Generally, two techniques can be used: FTIR-reflection spectroscopy and ATR-FTIR. While FTIR-reflection spectroscopy was employed to monitor the growth of organic coatings on metal surfaces, ATR was used mostly to study the transport of water and some ionic species through the organic coatings.

20.5.4.2 ATR-FTIR

The adhesion between the polymeric coating and the metal substrate plays an important role in the protective properties of the coatings. Water that transports through the coating to the metal/coating interfaces is a major cause of degradation and delamination of the coating. In addition, water also promotes corrosion of the metal substrate.

ATR-FTIR has been extensively used to study the transport of water and/or ionic species through polymer films to metal/polymer interfaces.^130–139 FTIR in the multiple internal reflection mode (FTIR-MIR) was used to obtain information on the water layer at the organic coating/substrate interface.^130–132 The effects of water evident in the 3000-3650 and 1625-1645 cm^− 1 regions were observed after the coating was exposed to water for a certain time. These features increased in intensity with time while those for the coating decreased. The fact that the intensity change of 3400 cm^− 1 is proportionally related to the total amount of water in the coating enables a quantitative analysis of the water layer. By comparing the intensity changes of the water OH stretching band as a function of time, the rates at which water entered the coating/substrate interfacial region for three coating systems can be determined and compared.¹³² Normal ATR-FTIR using ZnSe crystal as substrate was also employed to study the transport kinetics of both water and inhibitor anion (HPO₄^2 −).^133,134

In a series of papers, ATR-FTIR with the Kretschmann configuration was applied for in situ studies of the transport of water and ionic species (thiocyanate ions) through a polymer film to metal/polymer interfaces, such as aluminium/polymer interface and conversion coated-zinc/polymer interface. The fast increase of the stretching and bending vibrations of water at 3450 and 1650 cm^− 1 suggested a fast sorption of water in the coating and the aluminium/coating interface (Figure 20.8).¹³⁵ Another band at 950 cm^− 1 due to the vibrations of Al-O groups indicated the formation of corrosion products—aluminium oxide or hydroxide at the aluminium surface. In the thicyanate solution, a band at 2075 cm^− 1 due to the C = N vibration of the thicyanate ion was detected after prolonged exposure. As complementary to ATR-FTIR, which provided information about a specific region, electrochemical impedance spectroscopy (EIS) was employed to take into account the whole system.^137,138 The combination of ATR-FTIR and EIS was also extended to investigate the hidden interface between a conversion-coated zinc surface and a polymer coating upon exposure to an electrolyte.¹³⁹