Electrochemical Kinetics of Corrosion and Passivity

Download as pdf or txt
Download as pdf or txt
You are on page 1of 36

Electrochemical Kinetics of Corrosion and Passivity The basis of a rate expression for an electrochemical process is Faradays law:

m=

Ita nF

Where m is the mass reacted, I is the measured current in ampere, t is the time, a is the atomic weight, n the number of electrons transferred and F is the Faraday constant (96500 Cmol-1). Dividing Faradays law by the surface area A and the time t leads to an expression for the corrosion rate r:

r=

m ia = tA nF

With the current density i defined as i = I/A. Exchange current density: We consider the reaction for the oxidation/reduction of hydrogen: 2H+ + 2erf rr H2

This reaction is in the equilibrium state at the standard half cell potential e0(H+/H2). This means that the forward reaction rate rf and the reverse reaction rate rr have the same magnitude. This can be written as:
i0 a nF

r f = rr =

In this case is i0 the exchange current density equivalent to the reversible rate at equilibrium. In other words, while the standard half cell potential e0 is the universal thermodynamic parameter, i0 is the fundamental kinetic parameter of an electrochemical reaction. The exchange current density cannot be calculated. It has to be measured for each system. The following figure shows that the exchange current density for the hydrogen reaction depends strongly on the electrode material, whereas the standard half cell potential remains the same. 1

Electrochemical Polarization: Polarization is the change in the standard half cell potential e caused by a net surface reaction rate. Cathodic polarization c means that electrons are supplied to the surface and they build up a negative potential in the metal. Therefore c is negative by definition. Anodic polarization a is the opposite process. There are two different kinds of polarization (sometimes also known as overpotential) namely activation and concentration polarization. Activation Polarization: In this case a step in the half cell reaction controls the rate of electron (charge) flow. For example the hydrogen evolution reaction 2H+ + 2e- H2 proceeds at a metal surface in three major steps. In the first step H+ reacts with an electron from the metal to form an adsorbed hydrogen atom at the surface. H+ + e- Hads In the second steps the reaction of two adsorbed hydrogen atoms forms a hydrogen molecule. Hads + Hads H2ads In the third step enough adsorbed hydrogen molecules combine and nucleate a hydrogen bubble on the surface. Any of these steps can be the rate limiting step and therefore cause the activation polarization. The activation polarization can be calculated using the current densities, which represent the reaction rates. For cathodic c and anodic a polarization one can write:
ic i0

c = c log

a = a log

ia i0

For zero both equations reduce to i = i0. For polarization potentials below the equilibrium half cell potential the reduction or forward reaction is favored: 2H+ + 2e- H2 while above the half cell potential the reverse reaction is favored. H2 2H+ + 2eIn the figure the so called Tafel plot is shown. It depicts the change in the polarization potential with the current density.

The activation polarization is related to the activation energy of the rate limiting step. Therefore a thermodynamical derivation of is possible. Concentration Polarization: If the reaction on the metal surface is fast it can lead to a depletion of the adjacent solution of the dissolved species that reacts on the surface. This is shown for the H+ concentration in the next figure. cB is the H+ concentration in the bulk of the solution.

From the Nernst equation it is clear that the half cell potential is dependent on the H+ concentration, in the form that the potential decreases with decreasing concentration.

eH + / H = e
2

0 H + / H2

2.3RT H+ + log nF pH2

( )

The decrease in the half cell potential is the concentration polarization conc, which can be written in terms of current density using the Nernst equation as:
i 2.3RT log 1 c nF iL

conc =

A plot of this equation can be seen in the next figure. One can see that conc is low until a limiting current density iL is reached. This current density is the measure of a maximum reaction rate that cannot be exceeded because of a limited diffusion rate of H+ within the solution. The limiting current density can be calculated as follows:
DZ nFC B

iL =

DZ is the diffusivity of the reacting species Z, CB is the concentration in the bulk of the solution and is the thickness of the depleted zone. iL is increased with increasing concentration, with increasing diffusivity (temperature) and higher solution agitation, which
4

decreases .

Concentration polarization is only significant for cathodic reduction processes, as for anodic oxidation one usually has an unlimited supply of metal atoms at the interface. Only at very high corrosion rates or during intentional anodic dissolution by an impressed current one can find concentration polarization due to limited rates of transport of soluble oxidation products away from the surface. Mixed Potential Theory: Charge conservation is required to apply the equations derived for polarization potentials. This means that the sum of anodic oxidation currents must equal the sum of cathodic reduction currents. For anodic oxidation there is one reaction of the general form: M Mn+ + neThere are basically three different kinds of cathodic reactions: 1. Evolution of Hydrogen from acid or neutral solutions: 2H+ + 2e- H2 (acid solutions) 2H2O + 2e- H2 + 2OH- (neutral and alkaline solutions) 2. Reduction of dissolved oxygen in acid or neutral solutions: O2 + 4H+ + 4e- 2H2O (acid solutions) O2 + 2H2O + 4e- 4OH- (neutral and alkaline solutions) 3. Reduction of a dissolved oxidizer in a redox reaction (e.g.): 5

Fe3+ + e- Fe2+ Corrosion Potential Ecorr: When two half cell reactions occur simultaneously on a metal surface, both of them will change the potential to an intermediate value called the corrosion potential. At the corrosion potential all the rates (current densities) are equal: ic = ia = icorr. This is schematically shown in the figure.

Effect of Exchange Current Density: The exchange current density often outweighs the thermodynamic driving force in determining the rate of reaction. This is shown for the example of Fe and Zn. The half cell potential for the anodic oxidation of Fe is about 0.44V, while for Zn it is about 0.76V. Therefore Zn should corrode before Fe. Figure 3.8. shows the polarization diagrams for Zn (dashed lines) and Fe (solid lines). One can see that the current density at the corrosion potential for Fe is larger than the current density for Zn at the respective potential. Therefore Zn corrodes slower than Fe, although it has the more negative Ecorr. Both icorr and Ecorr are used in corrosion protection of iron with a Zn coating, as the Zn starts to corrode first (Ecorr), and the corrosion is much slower than that of Fe (icorr).

Effect of added oxidzer: The driving force of corrosion is increased by the addition of a stronger oxidizer. In other words, when a redox system with a much more noble half cell potential is added to the system, the corrosion will increase. For example we consider the addition of Fe3+/Fe2+ to an acid solution. Figure 3.9 shows the determination of the resulting mixed potential Ecorr for the corroding metal M. The horizontal lines mark the half cell potentials for the three reactions that occur in the system. The dashed lines represent the total reduction rate and the total oxidation rate. Again Ecorr is found where the total reduction rate equals the total oxidation rate. In figure 3.10 the determination of the anodic oxidation rate (icorr) is shown for the same system as in figure 3.9. As the oxidation of M is the only oxidation reaction, icorr equals the sum of the two cathodic reduction reactions i(Fe3+ Fe2+) and i(H+ H2). Figure 3.9 shows that the added oxidizer only has an effect, when its exchange current density i0 is large. With a low i0 as shown in figure 3.11 there is no increase in the oxidation due to the added oxidizer, as the total rate of reduction is only due to i(H+ H2) in this case.

Cathodic and Anodic Polarization: When an excess of electron flow is applied to a system with a current density of iappl, c, the electrode potential shifts negatively from Ecorr to E, which can be written as c = E Ecorr. This leads to a decrease in ia and an increase in ic. As we still need an equal charge flow we can calculate iappl, c = ic ia. This is shown in figure 3.14a. A similar approach is valid for anodic polarization.

In figure 3.14 b the data points represent iappl, c and iappl, a, for cathodic and anodic polarization, respectively. With high c and a iappl, c and iappl, a are large and the points coincide with the tafel plots, whereas with lower c and a the measured points start to deviate from the 9

theoretical behaviour. As can be seen in figure 3.14b one can get icorr and i0 for the two half cell reactions by extrabolation of the linear part, where Tafel behaviour is observed. While this usually works good for the cathodic part, one usually does not get Tafel behaviour from measurements of the anodic part, as shown in figure 3.15. But one can calculate the anodic part from the cathodic measurements.

An additional problem in interpreting measured data is the effect of concentration polarization. This brings significant additional deviation from Tafel behaviour as shown in figure 3.16. Only at the lowest pH, which corresponds to the highest H+ concentration, a nice linear part is observed in figure 3.16. With increasing pH (decreasing H+ concentration) the experiment is mainly determined by the concentration polarization. At higher pH hydrogen evolution by direct reduction of water occurs 2H2O + 2e- H2 + 2OH-, which again is linear following Tafel behaviour, as there is of course a very high concentration of water present.

10

11

Passivity

Passivity is defined as corrosion resistance due to formation of thin surface films under oxidizing conditions with high anodic polarization. For example Fe is passivated in concentrated nitric acid. But in diluted nitric acid Fe is dissolved, or when the passive film is scratched, Fe starts to dissolve, too. This is shown in the following figure.

Fe only passivates in extremely oxidizing environments. The reason is the potential an pH range where iron oxides are thermodynamically stable. This can be seen in the Pourbaix diagram (see figure). Chromium (dashed lines and shaded area) passivates under less oxidizing conditions. But Cr has bad mechanical properties, therefore it is only used as an alloying element in Fe to make stainless steel (min 12% Cr). Ni (> 8%) further enhances corrosion resistance and improves mechanical properties by stabilizing the fcc phase of Fe.

With increasing potential and anodic polarization passive metals show a distinct behaviour. At low potentials corrosion rates measured by anodic current density are high and increase with potential in the active state. Above the primary passive potential Epp the passive film becomes stable and the corrosion rate decreases significantly. ipass can be up to 106 times lower than icrit, as indicated in figure 4.3. 12

At very high potentials the passive to transpassive transition occurs and the corrosion rate increases again. For stainless steel this potential is near the oxygen evolution potential, where the Cr rich passive film is unstable, as can be seen in the Pourbaix diagram. Going to higher H+ concentrations (higher acidity) and to higher temperatures leads to a decrease in the extend of the passive region, as shown in figure 4.4.

Oxidizer Concentration Effects: An increase in the oxidizer concentration increases the potential of the redox half cell reaction according to the Nernst equation. Taking mixed potential theory into account one can depict the effect of oxidizer concentration as shown in figure 4.5. The corrosion rate changes with 13

increasing oxidizer concentration from 1 to 8. The corresponding corrosion rates (current densities) are labeled A through F.

From figure 4.5 one can make another plot with corrosion rate versus oxidizer concentration. In this case (figure 4.6) one can clearly see a hysteresis effect. This borderline passivity is an instable state, where any surface disturbance (scratching, stress, etc) will destabilize the passive film.

Solution Velocity Effects: In a dilute salt solution (e.g. seawater) there is under deaerated conditions only one cathodic 14

reduction reaction available: 2H2O + 2e- H2 + 2OHand therefore the metal corrodes in the active state. But the rate of hydrogen evolution is small and therefore the corrosion rate is small, too. If the solution becomes aerated, reduction of dissolved oxygen O2 + 2H2O + 4e- 4OHstarts to predominate as shown by line 1 in figure 4.7. In this case one has borderline passivity.

But agitation of the solution leads to an increase in the limiting diffusion current from 1 to 6, and the system gets in the passive state. In figure 4.8 a the corrosion rate from figure 4.7 is plotted against the solution velocity (agitation). For comparison figure 4.8 b shows the same plot for a metal that does not form a passive film. There the corrosion only shifts from concentration to activation control.

15

Experimental Apparatus and Procedures: Potentiostatic and potentiodynamic procedures are used to measure the anodic polarization behaviour of metals. A potentiostat automatically adjusts the applied polarizing current to control potential between the working electrode (WE) and the reference electrode (REF). A schematic of such a system is shown in figure 4.10.

The current I polarizes the WE to the prescribed potential with respect to REF. In this setup no current passes through the REF, which has a constant potential. The result of such a measurement is shown in figure 4.11.

In the measurment the potential is increased from Ecorr in the active state in steps. After a step 16

the current is recorded and so on. In a potentiodynamic setup the potential is continually increased, while the current is continually measured at the same time. The result is the same as in the potentistatic experiment. An example of a real system is shown in figure 4.12. The potentiostatic polarizatzion cures of Fe and Fe-10.5Cr are shown.

Anodic polarization curves are affected by the rate of polarization (scan rate in Vs-1) (see figure 4.13) and by differences in the measured samples (see figure 4.14).

17

Potentiodynamic anodic polarization curves are used to judge corrosion resistance of alloys and the corrosivity of solutions. The experiments only need a couple of minutes. Potentiostatic measurements need a few hours, which is still short in comparison to the weeks or months needed for conventional testing. For example one can use the data to choose an alloy for different environments. The plots are shown in figure 4.16.

18

There are 5 parameters that may be used to judge an alloy: 1. In the active state, corrosion rate is proportional to the anodic current density whether or not the alloy is of the active-passive type. 2. The current density (rate) of reduction must exceed the critical current density for passivation to ensure low corrosion rate in the passive state. 3. Borderline passivity should be avoided in which either the active or the passive state may be stable. 4. Breakdown of the passive film in oxidizing conditions due to transpassivity or initiation of localized corrosion should be avoided. 5. The passive state in oxidizing conditions is essential for corrosion resistance, but reasonably small variations in the passive current density may not be significant. A real world example is shown in figure 4.17. Hastealloy B is Ni-25Mo, C is Ni-15Cr-15Mo5Fe and C-276 has low Si and C impurities. One can see that the latter has the best passive region. For use in the active region one would use alloy B as it does not contin expansive Cr. Apparently is the anodic polarization very sensitive to impurities.

The influence of the solution is shown in figure 4.18. In this case one can see that the addition of chloride to the solution has a profound effect on the anodic polarization of steel, whereas hastealloy C is less affected. But in chloride free solutions the steel performs better. 19

There are of course limitations to the data achieved by potentiodynamic anodic polarization. Under plant conditions additional processes can occur that influence the corrosion resistance of a material. Passive current density and corrosion rates can change over time. Therfore only qualitative predictions can be made. Then more extensive in-plant exposure tests are necessary. Anodic Protection: If an active passive metal is maintained in the passive region with a potentiostat its corrosion rate will be low at ipass. With the electrical parts scaled up to plant size this is called anodic protection. The metal needs to have a broad passive region, as a wrong potential outside of the passive region would lead to increased corrosion. The method so far is only used for storage tanks for very corrosive acids (e.g. sulfuric acid (steel), phosphoric acid (stainless steel) and nitric acid (nickel)). Properties of Passive Films: The properties of passive films are unfortunately rather elusive. While potentiodynamic anodic polarization contains a lot of information, it is not possible to delineate the mechanism of the transition to the transpassive state. Passivity has been investigated for about 140 years, 20

but still not much is known about the structure of the films. It is clear that the films contain some water and most films are more likely hydroxides and not oxides. Surface analysis of the films is a problem, as the morphology of the film can change, when it is transferred into the vacuum. Water bound in the structure can be lost and this may lead to or change crystallization. Due to the fact that passive films are very thin most of them show semiconducting behaviour. For some cases it has been shown that the passive film is amorphous (e.g. Fe, Al). It has been shown recently [Mamun A, Schennach R, Parga JR, Mollah MYA, Hossain MA and Cocke DL Electrochimica Acta, 46 (22) 3343 (2001)] that oxygen evolution can initiate crystallization of the passive film on zirconium, which coincides with the passive transpassive transition.

21

Galvanic and Concentration Cell Corrosion

A galvanic cell is formed, when two different metals are connected electrically, while both are immersed in an electrolyte. In this case one of the two is corroded preferentially by galvanic corrosion. When different solution concentrations are present on the same metal surface, a concentration cell is formed, which leads to preferential corrosion, too. When attack on the junction of two dissimilar alloys is limited to one of the two, one usually has galvanic corrosion. The preferentially corroded alloy is the anode in the galvanic couple and the unattacked alloy is the cathode. The cathode may be damaged by hydrogen or by cathodic polarization out of the passive into the active zone. Any metal or alloy has a unique corrosion potential Ecorr immersed in a corrosive electrolyte. When two metals are coupled, the one with the more negative Ecorr will lose electrons, which leads to an increased corrosion rate. The anodic dissolution, or corrosion reaction is M Mn+ + nefor the active metal. The rate for the more noble metal N is decreased due to the excess electrons from M. N Nm+ + meN is the cathode of the galvanic cell and M is the anode. The corrosion rate decrease in the cathode is the basis of cathodic protection by a sacrificial anode. The corrosion potentials of different metals and alloys in a specified electrolyte give a galvanic series. Such a series gives a qualitative prediction, which alloy or metal will corrode first in a galvanic couple. But the quantitative corrosion rate can not be predicted from this series, as the influences of solution velocity and temperature are neglected. And the series is not necessarily the same for a different electrolyte, although people sometimes try to draw conclusions from the seawater series for different electrolytes. Area Effects: A large ratio of the cathode to anode surface area should be avoided. For small anodes the galvanic attack is concentrated and penetration of the attack through the metal is faster. A larger anode surface will give a more uniform attack and will slow down the rate of penetration. A classic example are storage tanks with new stainless steel bottoms and coated sides made of normal steels. The galvanic corrosion penetrated the sides quickly, due to small 22

defects in the protective coating which are unavoidable. The new tanks started to leak already after a few months. The problem is shown in figure 6.2.

Prevention: The best way to prevent galvanic corrosion is to eliminate the galvanic couple by design, whenever possible. Using materials with similar Ecorr reduces the driving force for galvanic corrosion drastically. Insulated flanges can be used to avoid a galvanic couple (figure 6.3).

But soluble corrosion products can be deposited on active metals downstream and form in situ galvanic couples. If the galvanic couple cannot be avoided a large anode area and a large anode thickness should be used. The anode can also be designed foe easy replacement. Coating is usually not advisable (example with the tanks). If coating is applied it should be put on the cathode to prevent the cathodic reaction. In this case small defects in the coating do not have such severe effects as on the anode.

23

Fundamental Aspects: Corroding Metal Inert Metal Couple: When zinc is coupled to platinum in dilute acid solutions experiments show that, 1. the corrosion potential of Zn is shifted to a more noble value 2. the corrosion rate of Zn is increased and 3. the hydrogen evolution on Zn is reduced. These effects are due to cathode and anode polarization within the couple, as shown in figure 6.4.

The electrons from the anodic reaction (Zn Zn2+ + 2e-) flow to Pt until both electrodes are at the same potential Ecouple, which is a similar effect as an impressed current iapp, shown in figure 3.14. The exchange current density of the hydrogen evolution reaction also affects the galvanic cell as shown in figure 6.5.

24

Increasing the surface area of the cathode leads to an increase in the corrosion rate of the anode. As the current from the reduction reaction increases, the anodic oxidation reaction needs to increase too, to maintain a constant charge. The effect is schematically shown in figure 6.6.

Corroding Metal Corroding Metal Couple: Two metals M and N are corroding metals. M is the cathode metal and N is the anode metal. Again the coupled potential is found where the total reduction is equal to the total oxidation, as shown in figure 6.7.

The potential of the couple always falls between the corrosion potentials of the two active metals. The rate of the metal with the more active Ecorr is increased and the rate of the metal with the more noble Ecorr is decreased. At breaks in a Zn coated steel, the Zn is galvanically corroded to protect the exposed steel cathodically. 25

Experimental Measurements: A schematic of an experimental result is shown in figure 6.8. The dashed lines represent the half cell reactions defining the uncoupled corrosion potentials for anode and cathode. The solid lines represent data from a galvanic couple, where the couple potential is established when cathode and anode are polarized to equal potentials by the same current Icouple.

For experimental measurements one uses a high surface area ratio cathode to anode (e.g. 12:1 in figure 6.9) and due to the large cathode area one does not need the auxiliary electrode, from polarization measurements.

26

Results from the cell shown in figure 6.9 are depicted in figure 6.11. Different area ratios shown in the figure were simulated by moving the experimental cathodic curve to higher or lower values with respect to the anodic curve. The intersections of the cathodic and the anodic curves show the couple potentials for the different surface area ratios.

The couple potentials graphically determined from figure 6.11 are shown as a function of the surface area ratio in the solid line in figure 6.12. The points in this figure are from actual experiments with the corresponding surface area ratio. One can see a rather good fit between experiment and theory.

In principle one should be able to measure the coupled current by simply putting an ampere meter between the two electrodes. But due to the resistance of the ampere meter one gets a wrong current, as shown in figure 6.13. This problem can be avoided by using zero resistance ampere meters, which can be realized using a special electronic setup. 27

The zero resistance ampere meter can be realized with the usual electrode setup combined with a potentiostat. In this case one can measure Icouple continuously and automatically. The measurement principle is shown in figure 6.14.

The potentiostat senses a difference between REF and WE and controls the difference at a preset value by automatically varying the current between WE and AUX. When WE and REF are shorted as shown in the figure the potentiostat will control the potential between cathode and anode at any specified value. If this value is zero the circuit will continuously and automatically read the Icouple at short circuit on the ampere meter. The result of such a measurement of short circuit current with time is shown in figure 6.15. In addition the effect of a dichromate inhibitor is shown in this figure.

28

Determining Rates of Galvanic Corrosion: The rate of galvanic attack in a two metal couple is measured by the galvanic current density at the anode. The anodic current density on any point on the anode depends on the polarization of both, anode and cathode, on the conductivity of the solution and the physical geometry of the couple. Anodic and cathodic polarization curves are used to visualize the polarization within a galvanic couple. The two curves can be obtained in two separate experiments using a potentiostat. A composite polarization diagram for different metals and alloys in aerated 3% NaCl is shown in figure 6.17. The intersection of an anodic curve with a cathodic curve of a more noble metal gives the couple current and potential. The dashed vertical lines in figure 6.17 represent the spread of values for the limiting current for reduction of dissolved oxygen, which is essentially the same for all alloys listed. A prediction of the surface area ratio (figure 6.17 corresponds to 1 cm2 surfaces for both electrodes) can again be made by graphically increasing the cathodic current. This is shown in more detail in figure 6.19. The new galvanic current has risen by a factor of 10. But the anodic surface area is still 1 cm2, the anodic current density is now 700Acm-2. However, the surface of the cathode has increased leading to a 29

cathodic current density of only 70Acm-2.

30

Potentiostatic polarization curves again have to be used with caution, for predicting corrosion behavior, as real environments tend to differ from controlled experiments. The influence of the solution can be seen when figure 6.17 for NaCl is compared to figure 6.18 for a deaerated 1N surfuric acid solution.

Galvanic Passivation of Titanium: When Ti is coupled to Pt, Pd, Rh or Ir the couple current is dominated by the hydrogen reaction at a potential within the passive range of Ti. This is used to give Ti corrosion resistance to hot acidic solutions can therefore be obtained by alloying these elements to Ti. A 31

similar effect can be reached using Ni. The polarization curves for Ti and Pt (Ni) are shown in figure 6.20.

Current and Potential Distributions: Until now uniform current distribution on the surface was assumed. In practice this assumption does not hold. The current and potential distributions between two coupled metals depend on the conductivity of the electrolyte, the physical geometry and on the polarization of anode and cathode. Solution conductivity leads to an ohmic drop R, between any pair of points on anode an cathode. This leads to a reduced current I<Icouple. The potential distribution within the solution under a steel and Ni couple is shown by lines of constant potential in figure 6.21.

32

The Ni cathode is obviously highly polarized in contrast to the steel anode. This leads to increased corrosion near the contact point on the steel and decreased corrosion near the contact point in the Ni. Concentration Cells: Corrosion is often affected by concentration calls. Differential aeration with different concentrations of dissolved oxygen on a single metal surface or on electrically connected surfaces are common. Such cells are especially important for Fe and carbon steels. Acid chloride concentration cells are important for the initiation and growth of pitting and crevice corrosion in stainless steel. A laboratory setup of a differential aeration cell is shown in figure 6.23. Two iron electrodes are immersed in 0.1 N NaCl on both sides of a porous diaphragm, allowing charge transfer, but not sufficient mass transfer to affect the dissolved oxygen content.

For uncoupled electrodes the anodic and cathodic reactions on the nitrogen side are Fe Fe2+ + 2e2H2O + 2e- 2OH- + H2. As there is hardly any dissolved oxygen, reduction of dissolved oxygen as the cathodic reaction is impossible. On the aerated side the anodic reaction is still the oxidation of Fe, but the cathodic reaction is the reduction of the dissolved oxygen by O2 + 2H2O + 4e- 4OH33

The corrosion rate of the uncoupled cathode is controlled by concentration polarization and diffusion of dissolved oxygen at IL,C. The uncoupled corrosion potential on the aerated electrode is Ecorr, C as shown in figure 6.24.

When the two sides are coupled one finds the couple potential between the two initial potentials. The potential distributions on the surface again lead to increased attack near the junction. On the aerated side the reduction reaction leads to an increase in pH, whereas at the deaerated side the Fe2+ ions react with water leading to a decrease in pH by the reaction Fe2+ + 2H2O Fe(OH)2 + 2H+ Increased alkalinity passivates the cathode and increased acidity increases the activity of the anode. This addiditonal effect is shown schematically in figure 6.25.

34

This effect can be clearly seen at waterline corrosion on steel and zinc. Corrosion depletes the water from dissolved oxygen, which is easiest replaced near the surface of the water in an unstirred solution. Depleted oxygen creates the anode at greater depth and the cathode formed at the waterline by reduction of the excess dissolved oxygen. This is schematically shown in figure 6.26.

The surface at the waterline is passivated by formation of local alkalinity from reduction of dissolved oxygen. But the metal is strongly attacked nearby, where dissolved oxygen is less accessible, but solution resistance is minimized between adjacent surfaces. This process can be shown by putting a drop of saltwater on an abraded steel surface. In the water small amounts of phenolphthalein and ferricyanide indicators show cathodic areas pink (OHformation) and anodic areas blue (liberation of Fe2+). The initial and final surface distributions are shown in figure 6.27. Similar concentration cells can also form when the metal is fully immersed. One has found concentration cell corrosion in both agitated and quiescent solutions. But in full immersed experiments the formation of the concentration cells takes longer times up to several days. So anodic and cathodic areas can form at a surface even if the solution flows, which one would think should lead to a homogeneous aeration of the solution.

35

36

You might also like