It is the cache of ${baseHref}. It is a snapshot of the page. The current page could have changed in the meantime.
Tip: To quickly find your search term on this page, press Ctrl+F or ⌘-F (Mac) and use the find bar.

Brazilian Journal of Chemical Engineering - THE EFFECT OF SMALL AMOUNTS OF ELEMENTS ON SHAPES OF POTENTIODYNAMIC AND POTENTIOSTATIC CURVES OF AISI 304L AND AISI 316L STAINLESS STEELS IN CHLORIDE MEDIA

SciELO - Scientific Electronic Library Online

 
vol.14 issue2Mechanism and Kinetics for the Dissolution of Apatitic Materials in Acid SolutionsThe Gas Balance Technique and the Respiratory Coefficient Variability in Cultures of Aspergillus awamori NRRL 3112 author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Brazilian Journal of Chemical Engineering

Print version ISSN 0104-6632

Braz. J. Chem. Eng. vol. 14 no. 2 São Paulo June 1997

http://dx.doi.org/10.1590/S0104-66321997000200002 

THE EFFECT OF SMALL AMOUNTS OF ELEMENTS ON SHAPES OF POTENTIODYNAMIC AND POTENTIOSTATIC CURVES OF AISI 304L AND AISI 316L STAINLESS STEELS IN CHLORIDE MEDIA

 

D. Pulino-Sagradi1 and N. Alonso-Falleiros2

1Departamento de Materiais, Faculdade de Engenharia Mecânica, UNICAMP, 13083-970,
e-mail: debora@fem.unicamp.br, São Paulo, Brazil
2Departamento de Engenharia Metalúrgica e de Materiais, Escola Politécnica, USP,
e-mail: nealonso@usp.br, São Paulo, Brazil

 

(Received: September 16, 1996; Accepted: February 19, 1997)

 

ABSTRACT - Samples of high purity grade and commercial purity grade type AISI 304L and AISI 316L steels were studied by the potentiodynamic and potentiostatic techniques in a naturally aerated 3.5% NaCl aqueous solution at a controlled temperature of (23±2)°C. The anodic polarization curves of the potentiodynamic technique showed that not always is it possible to determine pitting potential: most of the curves of commercial purity grade steels displayed a smooth curvature in the region where the current density should increase sharply. The density current versus time potentiostatic curves also showed different shapes according to the purity grade steels: for the commercial purity grade steels, the current density showed large oscillations with time (related to unstable pits), whereas for the high purity grade steels, a regular behavior of current density as a function of time was found (related to stable pits).
KEYWORDS: Stainless steels, pitting corrosion, potentiodynamic, potentiostatic.

 

INTRODUCTION

One purpose for evaluating a material with respect to pitting corrosion is to obtain information for comparing and selecting materials. The resistance of a given material to pitting is generally determined by the critical potentials measured by different electrochemical methods.

In the potentiodynamic method the applied potential is scanned to noble values and the respective current is measured. This technique allows the measurement of the pitting potential (Ep) concerned with pit nucleation (Szklarska-Smialowska and Janik-Czachor, 1971). However, the potentiodynamic curves depend on experimental variables such as potential scanning rate (Shibata, 1983, Janik-Czachor, 1980, Leckie, 1970, Bond and Lizlovs, 1968, Manning, 1980) and surface condition of the sample (Barbosa et al., 1991, Streicher, 1956, Manning et al., 1971). Therefore, it is necessary to be cautious in seeking absolute values of the Ep by the potentiodynamic method. To illustrate this, it is interesting to note that the definition of Ep in the literature is often confusing. This is due to the shapes that the curves can display. For example, Bird et al. (1988) and Cavalcanti et al. (1986) obtained the Ep by an extrapolation back to the passive current density for the rising curve observed in the early stages of pitting. On the other hand, Qvarfort (1988) defined two Ep values from the same curve: the first relating to a certain current density and the other to a current density ten times higher.

The potentiostatic method can be used as an alternative method in analyzing the susceptibility of a material to pitting. With this technique a constant potential is applied and the current is measured as a function of time. According to Dayal et al. (1980), this method provides not only a single Ep value, but also critical potentials correlated with stable and unstable pits.

The aim of this work is to study the effect of small differences in chemical composition on the shapes of potentiodynamic and potentiostatic curves. Four austenitic stainless steels were tested with different combinations of Mo, C, P, S, Si, Mn, N, Cu, Ni and Cr, but without changing AISI 304L and AISI 316L specifications.

 

EXPERIMENTAL METHODS

 

Preparation of Test Specimens

To undertake this work, samples of two heats of AISI 304L and two of AISI 316L stainless steels were tested. The compositions and respective identifications adopted for each steel are assembled in Table 1. Based on C, N, P, S, Si, Mn and Cu as impurities contained in steels, it was possible to divide the heats in Table 1 into two groups: one characterized by commercial purity grade, identified as 304L and 316L, and the other characterized by high purity grade, identified as 304LL and 316LL. The commercial purity grade steels were provided by Aços Villares SA (Brazil) and the high purity grade steels, by Faculty of Engineering, University of Hokkaido (Japan).

The samples were solution treated and submitted to Standard Practice A of ASTM A 262(American Society for Testing and Materials, 1990) to verify the absence of a sensitized microstructure. Then, the samples were cut and their faces were abraded with silicon carbide papers to a 600-grit finish. The specimens were finally embedded in bakelite by hot compression. However, before mounting, the samples were passivated (Shibata and Takeyama, 1977) to avoid crevice corrosion at the bakelite-metal interface during electrochemical tests. Before each electrochemical test, the surfaces of the samples (around 1 cm2) were again grinded with 600-grit paper to renew the surface.

 

Electrochemical Tests

Two kinds of electrochemical methods were performed: potentiodynamic and potentiostatic. In both cases the tests were conducted with a PAR 273 potentiostat with a saturated calomel electrode (SCE) as the reference, graphite rods as counter electrodes and a naturally aerated 3.5% NaCl aqueous solution as the electrolyte at a controlled temperature of (23±2)°C.

In the potentiodynamic tests, the sample was polarized anodically from the corrosion potential and the current density was recorded continuously up to 1000 mA/cm2. The scanning rates were 1 mV/s for 304LL and 316LL steels and 0.15 mV/s to 1 mV/s for 304L and 316L steels. After the experiments, the surfaces of the samples were examined in an optical microscope magnified 50 to 500 times.

In the potentiostatic tests, the sample was first polarized anodically from the corrosion potential with a scanning rate of 1mV/s up to the chosen potential. This potential was held for 8 h, or less if the current density reached 1000 mA/cm2. The current density was recorded as a function of time every 15s (but only the main representative points of each phenomenon were plotted). After the experiments, the surfaces of the samples were also examined in an optical microscope.

 

EXPERIMENTAL RESULTS

 

Potentiodynamic Tests

The pitting potential (Ep) is the potential whereby the current increases sharply (Szklarska-Smialowska and Janik-Czachor, 1971). This value should be theoretically obtained by potentiodynamic tests. However, it was observed that the curve shape depends on the type of material.

For the high purity grade steels (304LL and 316LL) polarized with a scanning rate of 1 mV/s, the anodic polarization curves showed an abrupt increase of current, clearly defining the Ep value (Figure 1).

The commercial purity grade steels (304L and 316L) did not show the same behavior. A smooth curvature in the region of the Ep was observed (Figure 2). This occurred for 80% of the experiments on 304L steels and for 100% of the experiments on 316L (of a total of 20 experiments for each material). In these cases, any effort to determine the pitting potential can become arbitrary.

 

Table 1: Chemical composition (weight %) of the steels studied

 

C

Cr

Ni

Mo

Mn

Si

P

S

N

Cu

304L

0.014

18.5

10.3

0.30

1.76

0.45

0.035

0.021

0.032

0.21

304LL

0.010

18.5

15.0

-

< 0.01

< 0.01

0.003

0.004

0.020

-

316L

0.021

16.7

13.0

2.16

1.56

0.53

0.033

0.030

0.026

0.19

316LL

0.015

18.5

15.1

2.47

< 0.05

< 0.01

0.004

0.003

0.017

< 0.01

 

Figure 1: Potentiodynamic curves of the (a) 304LL and (b) 316LL, both with a scanning rate of 1 mV/s.

 

Figure 2: Potentiodynamic curves of the (a) 304L and (b) 316L, both with a scanning rate of 1 mV/s.

 

To confirm whether the chemical composition is the explanation for this behavior, other scanning rates were used for these steels. The scanning rate was 0.2 mV/s for the 304L steels. The number of round curves decreased slightly (from 80% to 65% of 20 experiments). For 316L steel, the scanning rates used were 0.5 mV/s, 0.2 mV/s and 0.15 mV/s (with 5 curves at each rate). In one curve, polarized with 0.5 mV/s, the Ep value was defined.

In high purity grade steels, it is often possible to obtain curves where the Ep value is well defined with the scanning rate of 1 mV/s. For commercial purity grade steels, smaller scanning rates must be used to obtain some curves where it is possible to define the Ep.

The examination of the surface of the samples after the tests showed that in the 304L and 304LL steels, pits have larger diameters and greater depths and they had lower densities in comparison with the 316L and 316LL steels. Moreover, comparing pits of 304LL and 316LL with pits of 304L and 316L, respectively, it was observed that the pits of high purity grade steels have larger diameters and greater depths in spite of having lower densities in comparison with respective commercial purity grade steels. In short, it was shown that the presence of Mo and impurity atoms simultaneously decreases the size (diameter and depth) of pits and increases their density on the surface.

 

Potentiostatic Tests

The results of the potentiodynamic tests proved that this method does not always allow the determination of the Ep. Thus, it is necessary to use another technique such as the potentiostatic method. Using this method, a relationship between chemical composition and shape of the current density versus time curves was also observed. These curves were described on a log-log plot. In this representation the points were brought close together during longer periods of time. Therefore, this analysis can occasionally hide some phenomena which can occur. On the other hand, a log-log plot simplifies the determination of short incubation times (about few minutes or less) on a large scale.

For the high purity grade steels (304LL and 316LL), a regular behavior of current density as a function of time was found; that is, for all constant applied potentials, the current density does not oscillate. This is illustrated in Figures 3a and 3b. For 304LL steel, it was found that up to the applied potential of 320 mV(SCE) the current density is about 10 mA/cm2, decreasing slightly with time. The fluctuations of current density that occurred in short periods of time are negligible. They always remain at passive values. At 330 mV(SCE), the same behavior occurs until a determined time is reached. At that value the current density begins to increase sharply. Finally, for applied potentials of 350 mV(SCE) or higher, the current density increases continuously throughout the test. The higher the potential applied, the higher is the current density increasing rate. The 316LL steel showed the same behavior, but at different potentials.

Such regular behavior of current density as a function of time was not found in commercial purity grade steels. These materials showed large oscillations of current density with time (fluctuations of current density about 1000 times in a few seconds). This is illustrated in Figure 4. For 304L steel, at applied potentials of 0 to 180 mV(SCE), the current density showed a tendency to drop during the first minutes. After that, the density current began to oscillate from passive values (about 10 mA/cm2 or lower) to active values (higher than 100 mA/cm2). These fluctuations increased as the applied potential increased. Such behavior was also found for 200 mV(SCE) to 240 mV(SCE). However, in these cases, the amplitude of oscillations was greater and after a certain period the current density increased continuously (Figure 4). Finally, at 250 mV(SCE), the current density increased sharply and instantaneously without oscillating. Analogous behavior was observed in the 316L steel, but at different potentials.

Figure 3: Potentiostatic curves of the (a) 304LL and (b) 316LL for some applied potentials.

 

Figure 4: Potentiostatic curves of the (a) 304L for an applied potential of 200 mV(SCE) and (b) 316L for an applied potential of 230 mV(SCE).

 

Figure 5 schematically represents current density as a function of time found for each steel at each applied potential, and shows if pitting occurred.

For high purity grade steels, pits were found at those potentials where the current density increases continuously. In this case, this rise in current density is linked to nucleation and the resultant growth of pits. As this phenomenon occurs in a stable way, it can be associated with a stable pit nucleation potential.

For commercial purity grade steels, pits occurred at some potentials where the current density does not increase continuously, such as 120 mV(SCE), 160 mV(SCE) and 180 mV(SCE) for 304L and 210 mV(SCE) for 316L. At these potentials, high oscillations of current density occurred. In these cases, pits probably nucleate during fluctuations of current, because at the former potentials the current density reached about 100 mA/cm2 or higher. Since the current increased only during fluctuations, it can be deduced that these pits sustained consecutive processes of growth and repassivation during rises and falls of current density, respectively. This phenomenon can be related to unstable pit nucleation potential because the growth of pits did not happen in a continuous and stable way. However, the oscillations of current at lower potentials without pitting (such as 50 mV(SCE) and 150 mV(SCE) for 316L steel) probably are related to the nucleation and repassivation of pits so small that they cannot be observed with an optical microscope. These oscillations without pitting may also denote an instability of passive film.

In addition to unstable pits, the 304L and 316L steels also showed stable pits. At potentials of 200 mV(SCE) or higher for the 304L and from 230 mV(SCE) for the 316L steel, there was a continuous and stable rise in the current density after a certain period.

 

DISCUSSION

The results of the potentiodynamic tests showed two kinds of anodic curves, depending on material and scanning rate: curves with a well-defined Ep and round curves with no clear definition of the Ep.

Round curves were obtained for commercial purity grade steels, mainly with high scanning rates. According to Manning (1980), when the scanning rate is high, there is less time for the thickening of passive film. Potentials prone to result in the adsorption of chloride ions are attained quickly (Skorchelletti, 1976).

 

High purity steels

Commercial purity steels

   
   
   
log t

304LL: up 350

316LL: up 570

304LL: up 250

316LL: up 340

Figure 5: Schematic current density (j) versus time (t) curves obtained from potentiostatic tests. The numbers refer to applied potential, mV(SCE).

 

Both facts therefore contribute to the dissolution of metal. This causes a gradual rise of current density with the increasing potential during scanning. This fact could explain the round form of curves as a function of high scanning rates, but not as a function of material. Considering that impurities are harmful to corrosion resistance (Fontana and Greene, 1978, Stewart and Williams, 1992), it can be assumed that, due to the lower purity grade of 304L and 316L steels, these materials probably have a less resistant passive film or have more affinity with chloride ions (this affinity can be translated into a higher number of active sites for adsorption of these ions). Then, in the low purity steels that have more weak sites, pits can form at lower potentials and more metastable pitting transients are seen. This is probably related not only to a dependence on pit nucleation rate with potential (Broli and Holtan, 1977), but also to a dependence on pit growth rate with potential. This explanation could thus justify the shape of round curves as a function of kind of material. However, other explanations could be possible and more detailed investigations need to be done.

In any event, curves with well-defined Ep were obtained in commercial purity grade steels polarized with low scanning rates and in high purity grade steels. In spite of commercial purity grade steels having a lower purity grade, the reduction of the scanning rate allows the formation of a passive film of better quality than that obtained with higher rates (Manning, 1980). Thus, there is an enhancement of resistance to pitting even at higher potentials where the adsorption (Skorchelletti, 1976) of chloride ions is better. Thus, the pit nucleation and growth will only occur with a certain critical potential where the adsorption of chloride ions is also critical.

The results of the potentiostatic tests also showed two types of curves: curves without significant oscillations of current density, typical of high purity grade steels, and curves with large oscillations, typical of commercial purity grade steels. Both types of curves can be discussed as a function of the passive film quality of each material and its relationship to the stability of pits.

First, pits can be classified in two groups: stable and unstable, with respective critical potentials (Dayal et al., 1980). According to Frankel et al. (1989), in both cases, the mechanism of nucleation is the same but the mechanism of growth is different. At nucleation, the passive film breaks down and metal contacts electrolyte. Initially, there is a some pit growth under the passive film (Frankel et al., 1989). This means that in the metal surface there is a small hole under which there is a pit. However, the dissolution of metal inside the pit creates high stress on the remainder of the film due to the gradient of concentration between the outside and inside of the electrolyte. This stress can break down the passive film. Thus, the initial size of the hole increases causing the mixing of the outside and inside solution of the pit. This may provoke pit repassivation. This process is expressed by oscillations in current density (Frankel et al., 1989) and can be related to the "birth and death" of an unstable pit. Figure 6, based on Frankel et al. (1989), schematically shows this mechanism.

In any event, if the remainder of the film above the pit is strong enough to resist the stress during a given period, a salt film can be deposited inside the pit. This phenomenon can stabilize the growth of the pit (Frankel et al., 1989). Therefore, nucleation and growth of a stable pit occur (Figure 6).

According to this theory, strong passive films promote the event of stable pits, while films with low mechanical resistance tend to create conditions for unstable pits to occur. This can explain why the commercial purity grade steels showed curves with large oscillations. Their passive films are probably less resistant than those of high purity grade steels, promoting break and repassivation of passive film. This process appears as fluctuations of the current density with time. On the other hand, high purity grade steels showed curves without significant oscillations because their passive films are more resistant and, thus, less unstable events occur.

By examining the results of both techniques, it was noted that the unstable pitting potentials obtained by the potentiostatic method coincide with the potentials where the potentiodynamic curves begin to become round. It seems that nucleation of unstable pits probably occurs at the beginning of this round region. This can therefore be the reason for the lack of definition of pitting potential in potentiodynamic curves.

 

CONCLUSIONS

(1) Because of the anodic curve shape, the potentiodynamic method does not always allow the determination of pitting potentials of AISI 304L and AISI 316L steels in 3.5% NaCl aqueous solution. For the commercial purity grade steels, the anodic curves display a smooth curvature in the region where the current should increase sharply, not defining the pitting potential.

(2) The potentiostatic method can exhibit two kinds of curves, depending on the material: (i) curves with large current density oscillations in time are related to unstable pits, typical of commercial purity grade AISI 304L and AISI 316L steels and (ii) curves without significant current density oscillations in time are related to stable pits, characteristic of the high purity grade AISI 304L and AISI 316L steels.

(a) (b)
(c) (d)
Figure 6: Proposed mechanism for stable and unstable pit occurrence. (a) nucleation of a pit under passive film; (b) stresses on passive film by means of gradient of concentration between outside and inside electrolyte; (c) if passive film has a low mechanical strength: this film breaks down due to stresses above, electrolyte inside and outside pit are mixed and this pit finally repassivates; (d) if passive film has a high mechanical strength: this film does not break down, there is a deposition of a saline layer and the pit can grow steadily.

(3) The round portion of the potentiodynamic curves, which inhibits the determination of pitting potential, begins at anodic potentials next to the unstable pitting potentials. The data obtained for the materials used in this work suggest that the lack of definition of pitting potential in the potentiodynamic method may be related to the occurrence of unstable pits. However, other phenomena cannot be ruled out.

 

ACKNOWLEDGEMENTS

The authors thank FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) for the financial support received for this work.

 

REFERENCES

American Society for Testing and Materials, Section 3, v. 02, p. 1. Philadelphia (1990).

Barbosa, M.A.; Garrido, A.; Campilho, A. and Sutherland, I., Corr. Sci. 32:179 (1991).         [ Links ]

Bird, H.E.H.; Pearson, B.R. and Brook, P.A., Corr. Sci. 28:81 (1988).         [ Links ]

Bond, A.P. and Lizlovs, E.A., J. Electrochem. Soc. 115:1130 (1968).         [ Links ]

Broli, A. and Holtan, H., Corr. Sci. 17:59 (1977).         [ Links ]

Cavalcanti, E.H.S.; Thompson, G.E.; Wood, G.C. and Dawson, J.L., 13th Seminário Nacional de Corrosão, p. 209, Rio de Janeiro, Brazil (1986). (In Portuguese.)         [ Links ]

Dayal, R.K; Parvathavarthini, N. and Gnanamoorty, J.B., Corrosion 36:433 (1980).         [ Links ]

Fontana, M.G. and Greene, N.D., Corrosion Engineering, 2nd ed., p.55. McGraw-Hill, Auckland (1978).         [ Links ]

Frankel, G.S.; Stockert, L., Hunkeler, F. and Boehni, H., Corrosion 43:429 (1987).         [ Links ]

Janik-Czachor, M., Werkst. Korros. 31:606 (1980).         [ Links ]

Leckie, H.P., J. Electrochem. Soc. 117:1152 (1970).         [ Links ]

Manning, P.E., Corrosion 36:468 (1980).         [ Links ]

Manning, P.E.; Duquette, D.J. and Savage, F., Corrosion 35:151 (1979).         [ Links ]

Metals Handbook, 8th ed., v. 1, p. 417. ASM, Metals Park (1961).         [ Links ]

Qvarfort, R., Corr. Sci. 28:135 (1988).         [ Links ]

Shibata, T., Trans. Iron Steel Inst. Jpn. 23:785 (1983).         [ Links ]

Shibata, T. and Takeyama, T., Corrosion 33:243 (1977).         [ Links ]

Skorchelletti, V.V., Theory of Metal Corrosion, p.220. Keter, Jerusalem (1976).         [ Links ]

Stewart, J. and Williams, D.E., Corr. Sci. 33:457 (1992).         [ Links ]

Srreicher, M. A., J. Electrochem. Soc. 103:375 (1956).

Szklarska-Smialowska, Z. and Janik-Czachor, M., Corr. Sci. 11:901 (1971).