**3. Overview of Electrochemical Corrosion**

The corrosion is a natural process that occurs when metallic materials are exposed to aqueous environments [67,68]. When a metal is immersed in aqueous solution, its cations spontaneously evolve on the metal–electrolyte interface and pass into the solution. During reaction, the material microscopically dissolves. Along with the cations, electrons are also released, and an electrical double layer is created at the metal–electrolyte interface [67]. The release of electrons causes the metal to become electrically charged. As a result of reaction, an electrode potential is established on the metal–electrolyte interface. After some time of immersion, an equilibrium is restored at the electrolyte–metal interface.

An overview of the metal corrosion is presented in Figure 6. The corrosion leads to an oxidation of metal and transfer of metal cations to the electrolyte. The oxidation occurs on a metal surface at a specific site known as an anode (anodic reaction site, [67]). Electrons that are released by metal are subsequently consumed by either dissolved oxygen or hydrogen cations in the electrolyte. The reduction reaction takes place at cathode (cathodic reaction site). The relative sizes and locations of cathodic and anodic sites are important variables influencing the overall corrosion rate. The sizes of cathodic and anodic areas may vary greatly; from atomic scales to macroscopically large dimensions.

**Figure 6.** An elementary electrochemical corrosion cell on metal surface.

The metal oxidation is given by the following reaction

$$M \to M^{z+} + ze^- \tag{2}$$

The Gibbs free energy change (Δ*Gr*) of the reaction is given as

$$
\Delta G\_{\rm r} = -zFE \tag{3}
$$

In this equation, *z* is the number of electrons involved in the reaction, *F* is a Faraday's constant (96 481 C mol−1), and *E* is the electrode potential. At standard conditions (T = 298.15 K, *p* = 101 325 Pa), the standard Gibbs free energy of the reaction (Δ*G*<sup>0</sup> *<sup>r</sup>* ) is related to the standard electrode potential (*E*0)

$$
\Delta G\_r^0 = -zFE^0 \tag{4}
$$

Standard electrode potentials of metals are compared in Table 2. Since the Gibbs energy is related to electrode potential (Equation (3)), the tendency of a metal to corrode in given environment may be evaluated using E–pH plots. The diagrams have been calculated for most metals by Pourbaix and are available in ref. [69]. Figure 7 displays the E–pH diagram for Al–H2O system [69,70]. The plot indicates the stability regions of different phases in aqueous solutions. The E–pH diagram shows four different regions where metallic aluminum, aluminum cations (Al3+), aluminum hydroxide and complex anion [Al(OH)4] − are stable. The region, where the metallic Al is stable, is labelled as immunity region. The areas with aluminum cations and anions as stable species are marked as corrosive regions. In these areas the corrosion occurs. The passivity region is where the solid hydroxide exists. In this region, Al is protected by a passive layer. The E–pH diagram demonstrates that corrosion takes place in both alkaline and acidic environments. The protective layer is formed at pH 4–9 [71]. The diagram also shows that the equilibrium electrode potential between [Al(OH)4] − and Al, shifts to less noble values with increasing pH.


**Table 2.** Standard potentials, E0, for metal electrodes, compiled from reference [68].

Although E–pH plots are useful in determining the metal's tendency to corrode in the given environment, they do not provide a kinetic information. The rate of corrosion therefore needs to be determined separately by experimental methods. Corrosion rates are obtained either by weight loss measurements or electrochemical methods [67,68]. The weight loss measurement is a simple experiment to determine corrosion rates. In the experiment, a clean weighed piece of material with well-defined dimensions is exposed to the corrosive environment for a sufficient period. The corrosion rate (*vcorr*) is then calculated based on the recorded weight loss according to the following equation [19,20]

$$v\_{corr} = \frac{\Delta w}{St} \tag{5}$$

In this equation, Δ*w* is the weight loss, *t* is the reaction time and *S* is the exposed surface area.

Weight loss measurements, although useful, can be time-consuming and may not provide a complete information about reaction mechanism. Electrochemical techniques are therefore widely used to study the corrosion mechanisms of metals in different electrolytes. A potentiodynamic polarization is an electrochemical technique where the progress of reaction is controlled by potentiostat [67,68,70]. It brings in a variety of parameters and provides valuable information about reaction mechanism. In the experiment, three electrodes are assembled in a corrosion cell [72]. The corrosion cell includes a working electrode (sample), counter (auxiliary) electrode and reference electrode. During the experiment, the potential of the working electrode is systematically varied with respect to reference electrode. The resulting current is measured by counter electrode. The potential of reference electrode is constant and serves as reference value. Silver chloride (Ag/AgCl) and calomel electrodes (Hg/Hg2Cl2) immersed in a saturated KCl solution are most frequently used

reference electrodes. Platinum mesh is used as counter electrode as this metal is corrosion resistant in most environments.

A schematic potential versus current density curve recorded during the polarization experiment is given in Figure 8. Several different regions can be distinguished on the curve. The first region is immune region. In this region, observed at low potentials, the metal is thermodynamically stable. In immune region, cathodic reactions prevail at the metal surface. The second region is labelled as active region. It is observed once a corrosion potential, Ecorr, has been reached. In the active region, the metal actively corrodes according to Equation (2). The active corrosion means that the anodic dissolution of the metal takes place in the studied solution. Some metals can passivate. Therefore, a passivation region can be also observed on the polarization curve. The passive region corresponds to passive layer formation on the metal surface. The passivation is reflected by rapid current density increase or stabilization at potentials higher than Ep (passivation potential) on the polarization curve. At very high potentials, the current may start to abruptly increase. The increase is a result of passive film breakdown and happens at potential higher than transpassive potential, Etr. The passive film breakdown may be initiated by aggressive halide anions and lead to localized corrosion (pitting). A given alloy system may contain either some or all regions shown in Figure 8a.

**Figure 8.** Schematic polarization curve of a passivating metal: (**a**) full curve, (**b**) Tafel extrapolation of cathodic and anodic regions.

The polarization curve provides a variety of electrochemical parameters. The corrosion potential and corrosion current density are important parameters that can be determined by Tafel extrapolation of the polarization curve. The procedure is shown in Figure 8b. In Tafel extrapolation, tangents to the polarization curve measured in immune and active regions are plotted. The intersection of the tangents determines the corrosion potential and corrosion current density. The corrosion potential reflects the tendency of the metal to corrode in given environment. The corrosion current corresponds to the rate of corrosion. In corrosion cell, a Faraday's law is valid [67,68]. The weight loss of the metal at the working electrode, Δ*w*, is calculated by the following equation

$$
\Delta w = \frac{A}{zF} I\_{corr} t \tag{6}
$$

In this equation, *A* is the atomic weight of the metal, *Icorr* is the corrosion current, *t* is the reaction time, *F* is Faraday's constant and *z* is the number of electrons involved in the electrochemical reaction. The corrosion rate, *vcorr*, is calculated from the corrosion current as

$$v\_{corr} = \frac{\Delta w}{St} = \frac{A}{zF} j\_{corr} \tag{7}$$

In this equation *S* is the sample surface area and *jcorr* is the corrosion current density (*jcorr* = *Icorr*/*S*). Equation (6) is valid for pure metals. For alloys, however, an equivalent weight, *Ew* must be introduced to account for different molar masses of constituent metals and different valence states. The following equation defines the equivalent weight of an alloy [73]

$$E\_{\varpi} = \frac{1}{\sum \frac{z\_i f\_i}{A\_i}}\tag{8}$$

In the equation, *zi* is the valence state, *fi* is the weight fraction and *Ai* is the atomic weight of metal *i* in the alloy. The corrosion rate than becomes

$$w\_{corr} = \frac{E\_{w}}{F} j\_{corr} \tag{9}$$

The assignment of valence states for TMs is often ambiguous as these elements have multiple stable valences. An independent experimental technique is therefore required, in addition to corrosion experiments, to establish the proper valence state. Another approach is to consult equilibrium Pourbaix diagrams [69]. The equilibrium E–pH diagrams can be used estimate the stable valence state of TM at the experimental conditions (electrode potential and pH of the electrolyte during corrosion test).

Metals become anodic and corrode only if their equilibrium half-cell potentials are smaller than the half-cell potential of the corresponding cathodic reaction [67,68]. When metals are combined into alloys it is no longer possible to define a unique half-cell potential. In multiphase alloys, different phases may act as local anodes and cathodes. The physical condition of the material may also be important. Constitutional variables such as the type and amount of structural defects (dislocations, grain boundaries) and crystal orientation are also important factors influencing the overall corrosion behavior.

Aluminum has a low standard electrode potential (Table 2, [68]). Therefore, aluminum and aluminum alloys are prone to corrosion. Nevertheless, the materials are also easily passivated. The passivation is related to spontaneous aluminum oxide/hydroxide film formation at the interface [74]. The passive film protects the material and impedes further reaction with the environment. Oxide layers grown on aluminum alloys at ambient temperatures are generally non-crystalline, although short-range cubic ordered structure has also been observed. In humid atmospheres, hydroxyl-oxides such as AlOOH or Al(OH)3 may also form on aluminum surface. The passive film is generally self-renewing and self-healing. Therefore, an accidental loss of the passive film due to, for example, abrasion is rapidly restored.

Aluminum and its alloys are prone to pitting corrosion [75,76]. This type of local corrosion is often observed in seawater as it is initiated by chlorides and other halide anions in the electrolyte. The process may lead to passivity breakdown. Secondary phase particles are important constitutional variables affecting the corrosion rate. They can be classified into three different groups based on their electrochemical potential [76]: particles with active elements, noble elements, and particles with both active and noble elements. Reactive particles with active metals (such as Li, Mg or Zn) have low electrode potentials. These particles behave as anodes and subsequently dissolve when embedded in aluminum matrix. Particles with more noble elements (such as Fe or Cu) have higher electrode potentials and constitute local cathodes. They initiate anodic dissolution of Al matrix. The matrix adjacent to local cathode is preferentially attacked due to galvanic microcell created at the matrix/particle interface (Figure 9, [76]).

**Figure 9.** Schematic of the de-alloying and subsequent trenching of Al2CuMg intermetallic in AA2024 aluminum alloy (**a**); microstructure image of corroded alloy surface after 1-h exposure in aerated H2O at 30 ◦C (**b**), redrawn (**a**) and reproduced (**b**) from ref. [76].

If intermetallic particles contain both noble and active elements, their electrochemical behavior changes over time. The active elements may preferentially dissolve, leaving behind the noble metals. This process is known as de–alloying. It is schematically shown in Figure 10 for Al2CuMg [76]. The galvanic interactions at the matrix-particle interface change because of de–alloying. The de–alloyed particle becomes nobler over time and may initiate an anodic dissolution of the surrounding matrix. Experimental conditions may also influence the particle dissolution behavior. For example, Al20Cu2Mn3 is a noble particle with respect to matrix at room temperature. Nevertheless, it may become anodic at temperatures higher than 50 ◦C. At 50 ◦C a dealloying behavior of Al20Cu2Mn3 has been observed, with de–alloying features much the same as Al2CuMg [77,78].

**Figure 10.** Schematic of trenching of Al7Cu2Fe intermetallic in AA2024-T3 aluminum alloy (**a**); microstructure image of corroded alloy surface after 1-h exposure in aerated H2O at 30 ◦C, redrawn (**a**) and reproduced (**b**) from ref. [76].

In this review we aim to address the following fundamental questions:

