LA EDUCACIÓN, UN ARMA CONTRA LA DESIGUALDAD

When corrosion is controlled by diffusion of oxygen, the corrosion rate at a given oxygen concentration approximately doubles for every 30 ° C (55 ° F) rise in tem-perature [11] . In an open vessel, allowing dissolved oxygen to escape, the rate increases with temperature to about 80 ° C (175 ° F) and then falls to a very low value at the boiling point (Fig. 7.2 ). The lower corrosion rate above 80 ° C is related to a marked decrease of oxygen solubility in water as the temperature is raised, and this effect eventually overshadows the accelerating effect of tempera-ture alone. In a closed system, on the other hand, oxygen cannot escape, and the corrosion rate continues to increase with temperature until all the oxygen is consumed.

When corrosion is attended by hydrogen evolution, the rate increase is more than double for every 30 ° C rise in temperature. The rate for iron corroding in hydrochloric acid, for example, approximately doubles for every 10 ° C rise in temperature.

7.2.3 Effect of p H

The effect of pH of an aerated pure, or soft, water on corrosion of iron at room temperature is shown in Fig. 7.3 [12] ; however, the effect of pH may be different in a hard water, in which a protective fi lm of CaCO 3 forms on the metal surface,

AQUEOUS ENVIRONMENTS 121

Figure 7.2. Effect of temperature on corrosion of iron in water containing dissolved oxygen.

(Reprinted with permission from F. N. Speller, Corrosion, Causes and Prevention , 3rd edition, McGraw - Hill, New York, 1951, p. 168.)

Figure 7.3. Effect of pH on corrosion of iron in aerated soft water, room temperature [12] . [ Reprinted with permission from W. Whitman, R. Russell, and V. Altieri, Ind. Eng. Chem. 16 , 665 (1924). Copyright 1924, American Chemical Society ].

as will be discussed in Section 7.2.6.1 . In obtaining the data shown in Fig. 7.3 , sodium hydroxide and hydrochloric acid were used to adjust pH in the alkaline and acid ranges.

Within the range of about pH 4 – 10, the corrosion rate is independent of pH and depends only on how rapidly oxygen diffuses to the metal surface. The major diffusion barrier of hydrous ferrous oxide is continuously renewed by the corro-sion process. Regardless of the observed pH of water within this range, the

surface of iron is always in contact with an alkaline solution of saturated hydrous ferrous oxide, the observed pH of which is about 9.5. *

Within the acid region, pH < 4, the ferrous oxide fi lm is dissolved, the surface pH falls, and iron is more or less in direct contact with the aqueous environment.

The increased rate of reaction is then the sum of both an appreciable rate of hydrogen evolution and oxygen depolarization.

Above pH 10, an increase in alkalinity of the environment raises the pH of the iron surface. The corrosion rate correspondingly decreases because iron becomes increasingly passive in the presence of alkalies and dissolved oxygen, as explained in Section 7.2.1.2 . Confi rming the occurrence of passivity by Defi ni-tion 1 in Secni-tion 6.1 , the potential of iron changes from an active value of − 0.4 to − 0.5 in water of pH < 10, to a noble value of 0.1 V in 1 N NaOH, with an accompanying decrease in the corrosion rate. If the alkalinity is markedly increased, for example, to 16 N NaOH (43%), passivity is disrupted, and the potential achieves the very active value of − 0.9 V. The corrosion rate correspond-ingly increases slightly to 0.003 – 0.1 mm/y (0.05 – 2.0 gmd), but that is still a rela-tively low rate. In this region, iron corrodes with formation of soluble sodium ferrite (NaFeO 2 ). In the absence of dissolved oxygen, the reaction proceeds with hydrogen evolution forming sodium hypoferrite, Na 2 FeO 2 [14] . The fact that Fe ²⁺

is complexed by OH ⁻ in strong alkalies to form FeO₂⁻ with accompanying reduc-tion in activity of Fe ²⁺ accounts for the observed active potential of iron. Although the rate of formation of FeO₂⁻ in concentrated alkalies at room temperature is low, caused by marked polarization of probably both anodic and cathodic areas, the rate becomes excessively high at boiler temperatures.

In the region of pH 4 – 10, the corrosion rate depends only on the rate of dif-fusion of oxygen to the available cathodic surface. The extent of the cathodic surface is apparently not important. This was established in an experiment of Whitman and Russell [15] , who exposed steel specimens that were copper - plated over three - quarters of the surface to tap water of Cambridge, Massachusetts. The total weight loss of these specimens compared to control specimens not plated was found to be the same. All oxygen reaching the copper surface, acting as cathode, was reduced in accord with the reaction In addition, all oxygen reaching cathodic areas of the iron surface itself produced

† This equivalence is valid in waters of low conductivity only if anode and cathode areas are in close proximity. In seawater, anodes and cathodes may be several feet apart.

* On the other hand, some investigators, using different natural waters and different chemicals to control pH, have observed that the corrosion rate does change with pH in the pH range 6 – 9. This behavior has been attributed to reduced buffer capacity of HCO3− (an inhibitor) as pH increases, causing a pH at local sites that is lower than the pH that would otherwise exist in a solution of satu-rated hydrous ferrous oxide. The reader may wish to consult the review by Matsushima on this matter [13] .

AQUEOUS ENVIRONMENTS 123

an equivalent amount of Fe ²⁺ . Thus, the total amount of iron corroding was the same regardless of whether copper was plated over part of the specimens.

However, penetration of iron in the case of the plated specimens was four times that of the bare specimens.

Therefore, it follows that so long as oxygen diffusion through the oxide layer is controlling, which is the case within pH 4 – 10, any small variation in composi-tion of a steel and its heat treatment, or whether it is cold worked or annealed, has no bearing on corrosion behavior, provided that the diffusion - barrier layer remains essentially unchanged. Oxygen concentration, temperature, and veloc-ity of the water alone determine the reaction rate. These facts are important because almost all natural waters fall within the pH range 4 – 10. This means that whether a high - or low - carbon steel, low - alloy steel (e.g., 1 – 2% Ni, Mn, Mo, etc.), wrought iron, cast iron, or cold - rolled mild steel is exposed to fresh water or seawater, all the observed corrosion rates in a given environment are essen-tially the same. Many laboratory and service data obtained with a variety of irons and steels support the validity of this conclusion [16] . A few typical data are summarized in Table 7.1 . These observations answer the once vociferous argument that wrought iron, for example, is supposedly more corrosion resistant than steel.

In the acid range, pH < 4, oxygen is not controlling, and the corrosion reac-tion is established, in part, by the rate of hydrogen evolureac-tion. The latter, in turn, depends on the hydrogen overpotential of various impurities or phases present in specifi c steels or irons. The rate becomes suffi ciently high in this pH range to make anodic polarization a possible contributing factor (i.e., mixed control).

Because cementite, Fe 3 C, is a phase of low hydrogen overpotential, a low - carbon steel corrodes in acids at a lower rate than does a high - carbon steel. Similarly, heat treatment affecting the presence and size of cementite particles can appre-ciably alter the corrosion rate. Furthermore, cold - rolled steel corrodes more rapidly in acids than does an annealed or stress - relieved steel because cold working produces fi nely dispersed low - overpotential areas originating largely from interstitial nitrogen and carbon.

Since iron is not commonly used in strongly acid environments, the factors governing its corrosion in media of low pH are less important than those in natural waters and soils. Nevertheless, there are certain applications where such factors must be considered — for example, in steam - return lines containing car-bonic acid, as well as in food cans where fruit and vegetable acids corrode the container with accompanying hydrogen evolution.

Less is known about the effect of impurities and metallurgical factors on the corrosion rate in the very alkaline region (pH ∼ 14) where corrosion is again accompanied by hydrogen evolution. In the passive region (pH ∼ 10 – 13), the effect on passivity by impurities in their usual concentrations, as well as the effect of metallurgical factors, is not expected to be pronounced. In general, any condition producing a large cathode - to - anode area ratio facilitates achieve-ment of passivity and increases the stability of the passive state once it is achieved.

7.2.3.1 Corrosion of Iron in Acids. We have seen that, in strong acids, such as hydrochoric and sulfuric acids, the diffusion - barrier oxide fi lm on the surface of iron is dissolved below pH 4. In weaker acids, such as acetic or carbonic acids, dissolution of the oxide occurs at a higher pH; hence, the corrosion rate of iron increases accompanied by hydrogen evolution at pH 5 or 6. This difference is explained [12] by the higher total acidity or neutralizing capacity of a partially dissociated acid compared with a totally dissociated acid at a given pH. In other words, at a given pH, there is more available H ⁺ to react with and dissolve the barrier oxide fi lm using a weak acid compared to a strong acid.

The increased corrosion rate of iron as pH decreases is not caused by increased hydrogen evolution alone; in fact, greater accessibility of oxygen to the metal surface on dissolution of the surface oxide favors oxygen depolarization, which is often the more important reason. The sensitivity of the corrosion rate of iron or steel in nonoxidizng acids to dissolved oxygen concentration is shown T A B L E 7.1. Corrosion Rates of Various Steels when Oxygen Diffusion Is Controlling

% Carbon Treatment Environment Temperature

AQUEOUS ENVIRONMENTS 125

by data of Table 7.2 . In 6% acetic acid at room temperature, the ratio of corro-sion rate with oxygen present to corrocorro-sion rate with oxygen absent is 87. In oxi-dizing acids, such as nitric acid, which act as depolarizers and for which the corrosion rate is, therefore, independent of dissolved oxygen, the ratio is almost unity. In general, the ratios are larger the more dilute the acid. In more concen-trated acids, the rate of hydrogen evolution is so pronounced that oxygen has diffi culty in reaching the metal surface. Hence, depolarization in more concen-trated acids contributes less to the overall corrosion rate than in dilute acids, in which diffusion of oxygen is impeded to a lesser extent.

Traces of oxygen in dilute H ₂ SO ₄ — or substantial amounts in more concen-trated acid, in which the corrosion rate is higher — inhibit the corrosion reaction.

For zone - refi ned iron, the average corrosion rate in aerated 1.0 N H ₂ SO ₄ at 25 ° C was found to be 41.5 gmd, whereas in hydrogen - saturated acid the rate was 68.0 gmd [18] . Similar effects are shown by pure 9.2% Co – Fe alloy in 1.0 N H ₂ SO ₄ , both aerated and deaerated, for which the corrosion rates are high and the diffu-sion of oxygen is impeded by rapid hydrogen evolution (Fig. 7.4 ) [18] . Potential and polarization measurements indicate that oxygen in small concentrations at the metal surface increases cathodic polarization, thereby decreasing corrosion;

in higher concentrations, oxygen acts mainly as a depolarizer, increasing the rate.

The important depolarizing action of dissolved oxygen suggests that the velocity of an acid should markedly affect the corrosion rate. This effect is observed, particularly with dilute acids, for reasons previously stated (Fig. 7.5 ).

In addition, the inhibiting effect of dissolved oxygen is observed within a critical velocity range, with the critical velocity becoming higher the more rapid the inherent reaction rate of steel with the acid. Relative motion of acid with respect to metal sweeps away hydrogen bubbles and reduces the thickness of the stag nant liquid layer at the metal surface, allowing more oxygen to reach the metal surface. Accordingly, the corrosion rate of steel in the presence of air in 0.0043 N H ₂ SO ₄ at a velocity of 3.7 m/s (12 ft/s) is the same as that in 5 N acid at the same velocity. At rest, the ratio of corrosion rates is about 12 [19] . In the absence of dissolved oxygen, only hydrogen evolution occurs at cathodic areas, and an effect of velocity is no longer observed (Fig. 7.5 ) [20] . This result is expected because hydrogen overpotential (activation polarization) is insensitive to velocity of the electrolyte.

Figure 7.5. Effect of velocity on corrosion of mild steel (0.12% C) in 0.33 N sulfuric acid under air and oxygen [19] , 23 ± 2 ° C, 45 - min test, rotating spec., 18 mm diameter, 56 mm long; and 0.009% C steel under nitrogen [20] .

Figure 7.4. Corrosion of 9.2% Co – Fe alloy in 1 N H 2 SO 4 , showing an inhibiting effect of dis-solved oxygen, 25 ° C [18] .

AQUEOUS ENVIRONMENTS 127

For aerated acid, the minimum rate occurs at higher velocities the more concentrated the acid because the rate of hydrogen evolution is more pronounced, thereby impeding oxygen diffusion to the metal surface. Similarly, the minimum shifts to higher velocities the higher the carbon content of the steel because the corrosion rate and the accompanying hydrogen evolution both increase with carbon content.