- Email: [email protected]
Galvanic corrosion of carbon steel coupled to antimony
Corrosion Science 68 (2013) 162–167
Contents lists available at SciVerse ScienceDirect
Corrosion Science journal homepage: www.elsevier.com/locate/corsci
Galvanic corrosion of carbon steel coupled to antimony J. Soltis ⇑, K.A. Lichti Quest Integrity Group, P.O. Box 38096, Wellington, New Zealand
a r t i c l e
i n f o
Article history: Received 16 September 2012 Accepted 8 November 2012 Available online 19 November 2012 Keywords: A. Carbon steel B. Polarisation
a b s t r a c t Galvanic corrosion of carbon steel coupled to antimony was studied in aerated and N2-purged electrolytes at ambient and 60 °C temperatures. Free corrosion potential of antimony and carbon steel shifts to more active values with increasing temperature and N2 purging of the electrolyte. Under all experimental conditions, antimony remains less electronegative than carbon steels. Aeration and temperature affect potentiodynamic behaviour of both materials. As a consequence, the corrosion current for the antimony–carbon steel couple increases with increasing temperature and with aeration. There was a good agreement between the corrosion currents obtained through the Evans’ experiment and super-imposed potentiodynamic scans. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Antimony has many uses. For example, it is used in the semiconductor industry in manufacturing electronic devices through electrodeposition [1,2], or it is used in pH-indicator electrodes [3]. A number of publications describe the electrochemistry of antimony [4–9] with respect to its use as an alloying additive in lead alloys for casting lead-acid battery grid components [10]. Little work, however, has been reported on antimony in corrosion related studies. In general, antimony exhibits stability [11,12] in the atmosphere and high resistance to sulphuric, nitric and hydrochloric acids at ambient temperature, and general resistance to various environments wherein the antimony trioxide, Sb2O3, is sparingly soluble. In the middle of 20th century, these attributes prompted development of antimony plating technology [13–15]. However, the technology never reached a commercial success predominantly due to the inherent brittleness of antimony and its low tensile strength. Le et al. [16] reported that addition of 0.1% of antimony to low-alloy steel markedly improved corrosion behaviour, which was ascribed to the formation of an oxide film, i.e. Sb2O5, on the steel surface. On the other hand, Golubev et al. [17] reported localised corrosion of carbon steel in an industrial setting in association with the process related deposition of non-continuous (defective) layers of metallic antimony. The process of antimony deposition has been referred to as cementation [18,19], and in general it is the reduction of antimonyl ions, SbO+, which occurs in association with oxidation of iron. The process can be described by the following electrochemical reactions:
⇑ Corresponding author. Tel.: +64 4 978 6641; fax: +64 4 978 9930. E-mail address: [email protected] (J. Soltis). 0010-938X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.corsci.2012.11.008
Fe ! Fe2þ þ 2e
ðoxidationÞ
ð1Þ
and þ
SbO þ 2Hþ þ 3e ! SbðsÞ þ H2 O ðreductionÞ
ð2Þ
Similar process may also occur in other industrial settings, where presence of antimony in the ionic form is intrinsic to process media, for instance geothermal fluids [20,21]. In the current work, we have studied electrochemical behaviour of a conventional carbon steel and antimony in a chloride containing environment, with the focus to establish severity of the localised corrosion when such materials are coupled together. The preliminary results cover the effect of temperature and the level of solution aeration on the corrosion current density. The on-going work will expand on the effects of higher temperature, i.e. >60 °C, solution pH and its composition with regards to concentration of chloride and ferric ions, presence of low level H2S, and the effect of cathodic–anodic area ratio. It will also include potential-pH equilibrium diagrams for the relevant systems and environmental conditions. 2. Experimental 2.1. Electrode preparation High purity (99.999%) antimony (Goodfellow Materials, England) and conventional carbon steel (AISI 1018), both in the form of rods with 6 mm diameter were used as experimental materials in as received condition. Electrodes were prepared from approximately 20 mm long pieces, which were cut from the supplied materials. One end of each specimen was then soldered to a copper lead to provide electrical connection, degreased with ethanol and finished by rinsing using acetone. Such prepared specimens were
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
163
then mounted in K36 coating and laminating epoxy resin (Nuplex Industries Ltd., New Zealand). Attention was paid to avoid introducing bubbles at the metal-resin interface, as these would interfere with experiments; prepared electrodes that contained such bubbles were discarded. Electrodes were then stored in a desiccator and prior to each measurement the surfaces ground using 1000 grid emery paper and rinsed with distilled water.
2.2. Electrochemical measurements All electrochemical measurements were obtained by using a model PGZ 100 potentiostat (Radiometer Analytical) connected to a personal computer via an RS-232 interface and controlled through the VoltaMaster 4 electrochemical software (Radiometer Analytical). A calomel electrode with saturated KCl (SCE) was used as the reference (all potentials reported were referred to SCE) and a platinum electrode as the auxiliary. Measurements were conducted using a typical three-electrode electrochemical cell with a double-wall arrangement. Such a custom-made design allowed for a hydronic heating of the experimental electrolyte, with both the heating mantle (Electrothermal, UK) and the peristaltic pump (Watson-Marlow 505S, UK) located outside the Faraday cage to minimize electromagnetic interference. The experimental electrolyte was prepared using distilled water and analytical grades of sodium and potassium chlorides (Scharlau Chemie S.A., Spain) at the ratio of 3.7, with the total concentration of chloride ions of 1000 mg dm3. The pH of the electrolyte was adjusted to 5.4 using 0.05 mol dm3 solution of sulphuric acid. In the case of measurements conducted in deaerated electrolytes, high purity nitrogen gas was used for purging for 1 h before starting a measurement and for shielding atmosphere in the cell during the measurements. Potentiodynamic scans for both materials were conducted after a 900 s hold at the free corrosion potential, Ecorr, and then initiated from the adopted Ecorr in either anodic or cathodic direction at a scan rate of 0.1 mV s1. The corrosion current that was expected to take place at the mixed potential of antimony–carbon steel couple was measured using the Evans’ experiment, where the more electronegative material (carbon steel) was used as the working electrode and the less electronegative material (antimony) was the auxiliary electrode. The galvanostatic steps, 0.1–0.2 lA, were then applied to such an arrangement and individual potentials of both electrodes measured against the SCE using the following condition: d(WORK)/dt and d(AUX)/dt < 3 mV min1 or Maximal-Step-Duration equals 30 s. Experiments were repeated three times for each experimental condition and the corrosion current, Icorr, for each measurement established as the intersection of corresponding potential vs. current curves.
3. Results and discussion 3.1. Free corrosion potential Figs. 1 and 2 show respective records of Ecorr for uncoupled antimony and carbon steel electrodes exposed to the aerated and N2-purged electrolyte at ambient temperature, i.e. 20 °C, and at temperature of 60 °C. Ecorr of antimony remained constant throughout the entire period of monitoring under all experimental conditions. Similar behaviour was also observed for carbon steel exposed to the N2-purged electrolyte at both temperatures. Under aerated conditions Ecorr of carbon steel showed a decaying trend, which became more pronounced with increasing temperature. In general, such a behaviour, i.e. decaying trend, is typical for
Fig. 1. Records for the free corrosion potential, Ecorr, of antimony in the aerated and N2-purged electrolyte at 20 and 60 °C.
Fig. 2. Records for the free corrosion potential, Ecorr, of carbon steel in the aerated and N2-purged electrolyte at 20 and 60 °C.
formation of porous hydroxide layers, which slow the rate of metal dissolution, but do not protect it from further dissolution [22]. The values of Ecorr, recorded after 900 s of monitoring, were plotted as a function of temperature in Fig. 3, which shows that antimony remains less electronegative material (cathodic) than carbon steel under all experimental conditions. Ecorr of antimony and carbon steel shift to more active values with both increasing temperature and N2 purging. This observation is not surprising since the Ecorr in the N2-purged environment would be controlled by the H–H+ half-cell reaction, which is on the active side of the standard half-cell potentials scale, when compared to the O2– OH half-cell reaction that would dictate Ecorr in aerated environment. Experimentally, such a shift in Ecorr due to level of solution aeration was reported for instance for AISI Type 304 stainless steel [23] and high purity aluminium [24]. The effect of temperature is also related to the concentration of oxygen in the electrolyte, since its concentration in aqueous solutions generally decreases with increasing temperature; for instance it falls from 14.2 mg dm3 at 1 °C to 6.4 mg dm3 at 40 °C in water [25]. Consequently, diffusion limited current associated with the oxygen reduction reaction decreases with decreasing oxygen concentration, which results in less-efficient consumption of electrons and a negative shift in Ecorr.
164
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
Fig. 3. Free corrosion potential, Ecorr, of carbon steel and antimony measured after 900 s plotted as a function of temperature for the aerated and N2-purged electrolytes.
Furthermore, the effect of temperature on Ecorr is also evident from Eq. (3) that shows that temperature, T, is of fundamental importance in electrochemical processes [26], and as such contributed to the observed dependence of Ecorr:
dE DS ¼ dT zF
ð3Þ
where DS is the change in entropy, z is the valence and F is the Faraday’s constant. Similarly, the concentration dependence of the electrode potential given by the Nernst equation includes temperature dependence in the term RT/zF, where R is the universal gas constant.
have minimal effect on the behaviour within the diffusioncontrolled region. However, an increase in anodic current density is apparent under N2-purged conditions within the IR-controlled region. Considering that dissolved oxygen would not affect resistivity or conductivity of the experimental electrolytes, it may be argued that it is the interfacial process [27], i.e. enhanced formation of the surface oxide film under presence of dissolved oxygen, that slows down charge transfer and leads to the observed differences in the anodic currents within the IR-controlled regions. Also, the anodic current density increases with increasing temperature, which is a generally accepted trend [28] ascribed to the reduction of diffusion layer thickness with increasing temperature. Hydrogen evolution dominates the cathodic behaviour under N2-purged conditions; although, at 20 °C there is some evidence of limiting behaviour, possibly due to presence of a limited amount of oxygen. Hydrogen evolution also dominates the cathodic process under aerated conditions at potentials lower than about 1 VSCE, but the mass-transport-limited oxygen reduction, ilim, is the main cathodic process at more noble potentials. Under both aerated and N2-purged conditions, the cathodic current density shifts to higher values with increasing temperature, which can be ascribed to the effect of temperature on the corrosion kinetics through its effect on the diffusion coefficient, thickness of the diffusion layer, and under aerated conditions also on the solubility of oxygen [29–31]. In general, dependence of diffusion coefficient, D, on temperature can be described through the Arrhenius equation [32] using the following form:
ED D ¼ AD exp RT
ð4Þ
where AD is the pre-exponential factor, ED is the activation energy. 3.3. Potentiodynamic polarisation – antimony
Fig. 4 compares anodic and cathodic potentiodynamic polarisation scans for the carbon steel in aerated and N2-purged electrolytes at 20 and 60 °C. Apart of the shift in Ecorr, discussed in the previous paragraph, as expected, anodic behaviour is dominated by active dissolution (Eq. (1)), at both temperatures. For the scans conducted at the same temperature, aeration and/or N2-purging
Comparison of anodic and cathodic potentiodynamic polarisation scans for antimony under similar experimental conditions as for carbon steel is made in Fig. 5. At ambient temperature and anodic potentials, antimony shows passive behaviour, with the passive current density, ipass, of the order 30 lA cm2 for aerated environment and increasing to about 40 lA cm2 under N2-purged conditions; ipass of a similar order of magnitude has been reported for instance for AISI Type 321 stainless steel in deaerated 0.01 mol dm3 S2 O32 -containing chloride-based neutral solution
Fig. 4. Potentiodynamic polarisation curves for carbon steel exposed to aerated and N2-purged electrolytes at 20 and 60 °C.
Fig. 5. Potentiodynamic polarisation curves for antimony exposed to aerated and N2-purged electrolytes at 20 and 60 °C.
3.2. Potentiodynamic polarisation – carbon steel
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
165
In regards to the corrosion process of antimony in acidic solutions, a three-stage reaction has been proposed for the oxidation [7]:
condition for carbon steel; although, the temperature related-shift to higher ilim appears to be of the same order of magnitude between the two materials. As discussed in great detail elsewhere [39–42], reduction of oxygen is a rather complex phenomenon, which can be associated with an exchange involving two or four electrons and the formation of numerous intermediates, for instance superoxide, O2, hydroxyl radical, OH, and peroxide, H2O2 or HO2. Also, oxygen reduction is strongly affected by adsorption of oxygen molecule, which when adsorbed to a metal surface weakens the O–O bond and the reduction products accelerate the kinetics of the oxygen reduction reaction [41]. Assuming that the N2-purging produced the same drop in oxygen concentration in electrolytes during experiments involving carbon steel and antimony, it may be argued that the difference in cathodic ilim measured between antimony and carbon steel may possibly relate to a different behaviour of molecular oxygen when interacting with different metal surfaces. For instance, Hellsing and Gao [43] reported on such differences for the molecular chemisorption of oxygen on copper, nickel and platinum, which were ascribed to the electronic structure of the metals, i.e. the number of unpaired d electrons.
Sb þ H2 O ! SbOHðadÞ þ Hþ þ e
ð7Þ
3.4. Corrosion current – evans experiments
SbOHðadÞ ! SbOðadÞ þ Hþ þ e
ð8Þ
þ
ð9Þ
Results from the measurements of the corrosion current for the carbon steel–antimony couple at 20 and 60 °C are shown in Figs. 6 and 7. Both figures indicate relatively good reproducibility of obtained measurements; although, greater scatter in the values of corrosion current were observed in the measurements associated with the aerated electrolyte and temperature of 60 °C (Fig. 7). Fig. 8 depicts the dependence of the corrosion current density, icorr, as a function of temperature for both aerated and N2-purged electrolytes, with the measured values, and calculated average and standard deviation values listed in Table 1. It is evident that both temperature and aeration increase the values of icorr. Whilst the effect of increasing temperature on increase of icorr is approximately two-folded in N2-purged environment, i.e. from 2.09 to 4.87 lA cm2, about six times higher corrosion current density is obtained for the couple under aerated conditions, i.e. 7.5 lA cm2 at 20 °C increasing to 40.2 lA cm2 at 60 °C. The magnitude of the corrosion current is determined by the kinetics of the system, which means that the rate of the overall corrosion process is controlled by the slowest step, which in this case
[33]. Wikstrom and Nobe [34] reported similar passive behaviour for antimony in acidic (pH 0–2) chloride containing solutions, with ipass decreasing from approximately 101–103 A cm2 with increasing pH. At 60 °C, anodic behaviour is initially represented by active dissolution and then followed by a distinct active-topassive transition at around 100 mVSCE under both aerated and N2-purged conditions. However, slightly higher passivation current density is observed for this transition in N2-purged electrolyte. According to El Wakkad and Hickling [35], ‘passivation’ of antimony is associated with formation of a thick Sb2O5 film between the metallic antimony and the outer film of Sb2O3, with the similar view also reported elsewhere [10,34,36] and in general making a reference to the following electrochemical reactions:
4 Sb þ 6 H2 O ! 2 Sb2 O3 þ 12 Hþ þ 12 e
ð5Þ
and
3 Sb2 O5 þ 12 Hþ þ 12 e ! 3 Sb2 O3 þ 6 H2 O
SbOðadÞ ! SbO þ e
ð6Þ
The fact that higher current density is observed under N2-purged conditions, points to contribution of oxygen to the overall process of oxidation under aerated conditions, mainly supply of oxygen atoms to form Sb–O–Sb bonds. For such a process we propose the following reaction to occur involving oxygen:
4Sb þ 2H2 O þ O2 ! 4SbOH
ð10Þ
However, it is important to realise that since both the oxides Sb2O3 and Sb2O5 are sparingly soluble to aqueous environment, for antimony there is no protection domain in potential-pH equilibrium diagram due to passivation by an oxide film [37]. Similarly as in the case of carbon steel, cathodic behaviour of antimony also shows dependence on N2-purging and temperature. Whilst the effect of temperature remains the same, as already discussed for the carbon steel case, there are some fundamental differences associated with antimony. More specifically, under N2-purging and at both experimental temperatures measured cathodic current densities show limiting behaviour, which can be ascribed to hydrogen evolution and formation of gaseous stibine [37], SbH3, which can also be accompanied by formation of solid antimony hydride [38], Sb2H2. The following electrochemical reactions might be considered for the antimony based products:
Sb þ 3Hþ þ 3e ! SbH3
ð11Þ
2Sb þ 2Hþ þ 2e ! Sb2 H2
ð12Þ
On the other hand, in aerated solutions cathodic current densities show a transition between two limiting behaviours, with a shift in this transition to more noble potentials from about 600 mVSCE at 60 °C to 500 mVSCE at 20 °C. At potentials closer to Ecorr, the cathodic behaviour is dominated by the reduction of dissolved oxygen and mass-transport-limited current density, ilim. This situation continues with further lowering of the potential until the transition point and increase in cathodic current density due to reduction of antimony to hydrides, as per Eqs. (11) and (12). Under aerated conditions and at the ambient temperature, ilim was about 4 lA cm2, which is lower than the mass-transportlimited current density of 20 lA cm2 measured under the same
Fig. 6. Potential as a function of the applied current for the carbon steel–antimony couple during the Evans’ experiment in aerated and N2-purged solution at 20 °C; icorr indicated by the intersection of the E vs. i curves of antimony (Sb) and carbon steel (CS).
166
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
low, where the slightly higher icorr at 60 °C is ascribed to the temperature-related decrease in thickness of the diffusion layer. In the case of uniform corrosion of carbon steel, such corrosion current densities would represent corrosion rates of about 24 and 56 lm y1 at 20 and 60 °C; for instance, this may be considered within allowable corrosion rates for 30 y life of a plant. However, in the case of localised galvanic corrosion, the accumulated loss on carbon steel could be a few orders of magnitude higher. Under aerated conditions, the oxygen reduction reaction in acidic solutions proceeds at an appreciably faster rate than the hydrogen evolution reaction either through the formation of peroxide, H2O2, as an intermediate or directly to water [44]:
Fig. 7. Potential as a function of the applied current for the carbon steel–antimony couple during the Evans’ experiment in aerated and N2-purged solution at 60 °C; icorr indicated by the intersection of the E vs. i curves of antimony (Sb) and carbon steel (CS).
O2 þ 2Hþ þ 2e ! H2 O2
ð13Þ
H2 O2 þ 2Hþ þ 2e ! 2H2 O
ð14Þ
The corrosion current is limited by the rate of diffusion of dissolved oxygen, and since D increases with increasing temperature the corrosion current, and thus the corrosion rate increases as well. As the rate of diffusion is proportional to the concentration of dissolved oxygen, the corrosion current is also proportional to the dissolved oxygen concentration and, as already discussed earlier, increasing temperature decreases solubility of oxygen in aqueous solutions, which tends to decrease the corrosion rate. {Note: oxygen is being removed at the boiling point.} Overall, as summarised by Porte [31], the corrosion current increases up to a maximum temperature of about 80 °C, and then decreases due to the reduced solubility of oxygen, which becomes the most important factor; however, in a closed system, the corrosion current would increase with increasing temperature until all the oxygen is consumed through the oxygen reduction reaction. 3.5. Corrosion current – prediction through polarisation curves
Fig. 8. Corrosion current, icorr, plotted as a function of temperature for the aerated and N2-purged electrolytes. The error bars indicate standard deviation.
Galvanic corrosion current can be roughly predicted by super-imposing polarisation curves of the anode and the cathode materials. Figs. 9 and 10 depict such an attempt for carbon steel and antimony under considered experimental conditions at 20 and 60 °C, respectively. The approximated corrosion current densities are summarised in Table 2. Upon comparison of the results in Tables 1 and 2, it is evident that icorr values obtained through super-position of potentiodynamic scans and through the Evans’ experiment are relatively comparable. Theoretically, such a fairly
Table 1 Results for the icorr (lA cm2) for the carbon steel-antimony couple. Temperature: 20 °C
Temperature: 60 °C
Aerated
N2-purged
Aerated
N2-purged
8.15 7.12 7.24
2.09 2.13 2.05
37.15 39.59 43.86
4.74 4.85 5.02
Average 7.50
2.09
40.20
4.87
Standard deviation 0.56
0.04
3.39
0.14
is the rate of reduction reactions occurring at the cathode. Under N2-purged conditions, it would be the rates of hydrogen evolution and possibly the reduction of antimony to hydrates. However, these reactions are very slow due to the low concentration of hydrogen ions at the experimental pH of 5.4. Therefore, the measured corrosion current densities in the absence of oxygen are
Fig. 9. Super-imposed respective cathodic and anodic scans for antimony (Sb) and carbon steel (CS) for aerated and N2-purged electrolytes at 20 °C.
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
167
References
Fig. 10. Super-imposed respective cathodic and anodic scans for antimony (Sb) and carbon steel (CS) for aerated and N2-purged electrolytes at 60 °C.
Table 2 Results for the approximate icorr (lA cm2) for the carbon steel-antimony couple. Temperature: 20 °C
Temperature: 60 °C
Aerated
N2-purged
Aerated
N2-purged
8.5
2.1
85.0
3.5
accurate prediction has been discussed elsewhere [45], when Ecorr of the anode and cathode are at least 120 mV apart, which is the case for the system studied in the current work.
4. Conclusions In this work we provide an insight into the galvanic corrosion of carbon steel coupled to antimony in aerated and N2-purged electrolytes at ambient and 60 °C temperatures. Measurements of free corrosion potentials show that (i) antimony and carbon steel show a shift in their Ecorr to more active values with increasing temperature and N2 purging, and (ii) antimony remains less electronegative than carbon steel under all experimental conditions. Aeration and/or N2 purging, and temperature affect potentiodynamic behaviour of both antimony and carbon steel, and thus the corrosion current for the antimony-carbon steel couple. More specifically, the corrosion current increases with increasing temperature and aeration. There was a good agreement between the corrosion currents obtained for the antimony-carbon steel couple through the Evans’ experiment and super-imposition of polarisation scans.
Acknowledgement This work was funded by the New Zealand Foundation for Research, Science and Technology under Contract No. C05X0704.
[1] M.-H. Yang, I.-W. Sun, J. Appl. Electrochem. 33 (2003) 1077–1082. [2] W. Yang, H. Cang, Y. Tang, J. Wang, Y. Shi, J. App. Electrochem. 38 (2008) 537– 542. [3] G. Edwell, Electrochim. Acta 24 (1972) 595–605. [4] D. Pavlov, M. Bojinov, T. Laitinen, G. Sundholm, Electrochim. Acta 36 (1991) 2081–2086. [5] D. Pavlov, M. Bojinov, T. Laitinen, G. Sundholm, Electrochim. Acta 36 (1991) 2087–2092. [6] T. Laitinen, H. Revitzer, G. Sundholm, J. Vilhunen, D. Pavlov, M. Bojinov, Electrochim. Acta 36 (1991) 2093–2102. [7] S. Laihonen, T. Laitinen, G. Sundholm, A. Yli-Pentti, Electrochim. Acta 35 (1990) 229–238. [8] M. Metikos-Hukovic, B. Lovrecek, Electrochim. Acta 23 (1980) 717–723. [9] M. Metikos-Hukovic, B. Lovrecek, Electrochim. Acta 23 (1978) 1371–1376. [10] M. Metikos-Hukovic, R. Babic, S. Omanovic, J. Electroanal. Chem. 374 (1994) 199–206. [11] A. Bregman, Met. Prog. 59 (1951) 245–249. [12] G.R. Schaer, W.H. Safranek, C.L. Faust, Plating 45 (1958) 139–147. [13] R.M. Burns, W.W. Bradley, Protective Coatings for Metals, 2nd ed., Reinhold Publishing Co., New York, NY, 1955. [14] K.G. Soderberg, H.L. Pinkerton, Plating 37 (1950) 254–259. [15] S.S. Abd El Rehim, A. Awad, A. El Sayed, J. Appl. Electrochem. 17 (1987) 156– 164. [16] D.P. Le, W.S. Ji, J.G. Kim, K.J. Jeong, S.H. Lee, Corros. Sci. 50 (2008) 1195–1204. [17] V.G. Golubev, Z.K. Vasileva, Z.K. Murzova, L.A. Bolsun, T.E. Golosova, Khim. Volokna 3 (1988) 12–16. [18] R. Piontelli, L. Fagnani, Korr. und Metallsch 19 (1925) 113–121. [19] A.L. Pitman, M. Pourbaix, N. De Zoubov, in: M. Puorbaix (Ed.), Atlas of Electrochemical Equilibria in Aqueous Solutions: Antimony, Pergamon Press, Brussels, 1966, p. 524. [20] K.L. Brown, S.F. Simmons, Geothermics 32 (2003) 619–625. [21] G. Zhang, C. Liu, H. Liu, Z. Jin, G. Han, L. Li, Geothermics 37 (2008) 73–83. [22] W.S. Tait, An Introduction to Electrochemical Corrosion Testing for Practising Engineers and Scientist, PairODocs Publications, Racine, WI, 1994. [23] J.B. Lee, A.K. Agrawal, R.W. Staehle, EPRI Report NP 1741, Electric Power Institute, Palo Alto, California, 1981. [24] J. Soltis, N.J. Laycock, D. Krouse, Corros. Sci. 53 (2011) 7–10. [25] R.F. Weiss, Deep Sea Res. 17 (1970) 721–735. [26] P. Grundler, A. Kirbs, L. Dunsch, Chem. Phys. Chem. 10 (2009) 1722–1746. [27] Z. Ahmad, Principles of Corrosion Engineering and Corrosion Control, Elsevier Science & Technology Books, Oxford, UK, 2006. [28] E.L. Liening, in: Process Industries Corrosion: Electrochemical Corrosion Testing Techniques, NACE International Inc., Houston, TX, 1986. [29] V.V. Gerasimov, G.V. Akimov, I.L. Rozenfeld, Bul, Acad. USSR Div. Chem. Sci. 5 (1956) 9–11. [30] G.T. Skaperdas, H.H. Uhlig, Ind. Eng. Chem. 34 (1942) 748–754. [31] H.A. Porte, Technical Note N-907, Naval Civil Eng. Laboratory, Port Hueneme, CA, 1967. [32] H.R. Corti, L.N. Trevanib, A. Anderko, in: D.A. Palmer, R. Fernandez-Prini, A.H. Harvey (Eds.), Aqueous Systems at Elevated Temperatures and Pressures: Physical Chemistry in Water, Steam and Hydrothermal Solutions, Elsevier Ltd., 2004, p. 321. [33] Y.M. Liou, S.Y. Chiu, C.L. Lee, H.C. Shih, J. Appl. Electrochem. 29 (1999) 1377– 1381. [34] L.L. Wikstrom, K. Nobe, Corrosion 31 (1975) 364–369. [35] S.E.S. El Wakkad, A. Hickling, J. Phys. Chem. 57 (1953) 203–208. [36] L.L. Wikstrom, N.T. Thomas, K. Nobe, J. Electrochem. Soc. 122 (1975) 1201– 1206. [37] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, National Association of Corrosion Engineers, Houston, TX, 1974. [38] D.T. Hurd, Introduction to Chemistry of the Hydrides, Wiley, New York, NY, 1952. [39] S. Strbac, R.R. Adzic, Electrochim. Acta 41 (1996) 2903–2908. [40] J. Xu, W. Huang, R.L. McCreer, J. Electroanal. Chem. 410 (1996) 235–242. [41] H.H. Yang, R.L. McCreer, J. Electrochem. Soc. 147 (2000) 3420–3428. [42] J.P. Hoare, The Electrochemistry of Oxygen, Interscience Publisher, New York, NY, 1968. [43] B. Hellsing, S. Gao, Solid State Commun. 90 (1994) 223–228. [44] V.S. Bagotskij, L.N. Nekrasov, N.A. Shumilova, Espehy Khimii 34 (1965) 1697– 1703. [45] H.P. Hack, in: R. Baboian (Ed.), Corrosion Tests and Standards Application and Interpretation: Galvanic, American Society for Testing and Materials, Philadelphia, PA, 1995, p. 186.
Contents lists available at SciVerse ScienceDirect
Corrosion Science journal homepage: www.elsevier.com/locate/corsci
Galvanic corrosion of carbon steel coupled to antimony J. Soltis ⇑, K.A. Lichti Quest Integrity Group, P.O. Box 38096, Wellington, New Zealand
a r t i c l e
i n f o
Article history: Received 16 September 2012 Accepted 8 November 2012 Available online 19 November 2012 Keywords: A. Carbon steel B. Polarisation
a b s t r a c t Galvanic corrosion of carbon steel coupled to antimony was studied in aerated and N2-purged electrolytes at ambient and 60 °C temperatures. Free corrosion potential of antimony and carbon steel shifts to more active values with increasing temperature and N2 purging of the electrolyte. Under all experimental conditions, antimony remains less electronegative than carbon steels. Aeration and temperature affect potentiodynamic behaviour of both materials. As a consequence, the corrosion current for the antimony–carbon steel couple increases with increasing temperature and with aeration. There was a good agreement between the corrosion currents obtained through the Evans’ experiment and super-imposed potentiodynamic scans. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Antimony has many uses. For example, it is used in the semiconductor industry in manufacturing electronic devices through electrodeposition [1,2], or it is used in pH-indicator electrodes [3]. A number of publications describe the electrochemistry of antimony [4–9] with respect to its use as an alloying additive in lead alloys for casting lead-acid battery grid components [10]. Little work, however, has been reported on antimony in corrosion related studies. In general, antimony exhibits stability [11,12] in the atmosphere and high resistance to sulphuric, nitric and hydrochloric acids at ambient temperature, and general resistance to various environments wherein the antimony trioxide, Sb2O3, is sparingly soluble. In the middle of 20th century, these attributes prompted development of antimony plating technology [13–15]. However, the technology never reached a commercial success predominantly due to the inherent brittleness of antimony and its low tensile strength. Le et al. [16] reported that addition of 0.1% of antimony to low-alloy steel markedly improved corrosion behaviour, which was ascribed to the formation of an oxide film, i.e. Sb2O5, on the steel surface. On the other hand, Golubev et al. [17] reported localised corrosion of carbon steel in an industrial setting in association with the process related deposition of non-continuous (defective) layers of metallic antimony. The process of antimony deposition has been referred to as cementation [18,19], and in general it is the reduction of antimonyl ions, SbO+, which occurs in association with oxidation of iron. The process can be described by the following electrochemical reactions:
⇑ Corresponding author. Tel.: +64 4 978 6641; fax: +64 4 978 9930. E-mail address: [email protected] (J. Soltis). 0010-938X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.corsci.2012.11.008
Fe ! Fe2þ þ 2e
ðoxidationÞ
ð1Þ
and þ
SbO þ 2Hþ þ 3e ! SbðsÞ þ H2 O ðreductionÞ
ð2Þ
Similar process may also occur in other industrial settings, where presence of antimony in the ionic form is intrinsic to process media, for instance geothermal fluids [20,21]. In the current work, we have studied electrochemical behaviour of a conventional carbon steel and antimony in a chloride containing environment, with the focus to establish severity of the localised corrosion when such materials are coupled together. The preliminary results cover the effect of temperature and the level of solution aeration on the corrosion current density. The on-going work will expand on the effects of higher temperature, i.e. >60 °C, solution pH and its composition with regards to concentration of chloride and ferric ions, presence of low level H2S, and the effect of cathodic–anodic area ratio. It will also include potential-pH equilibrium diagrams for the relevant systems and environmental conditions. 2. Experimental 2.1. Electrode preparation High purity (99.999%) antimony (Goodfellow Materials, England) and conventional carbon steel (AISI 1018), both in the form of rods with 6 mm diameter were used as experimental materials in as received condition. Electrodes were prepared from approximately 20 mm long pieces, which were cut from the supplied materials. One end of each specimen was then soldered to a copper lead to provide electrical connection, degreased with ethanol and finished by rinsing using acetone. Such prepared specimens were
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
163
then mounted in K36 coating and laminating epoxy resin (Nuplex Industries Ltd., New Zealand). Attention was paid to avoid introducing bubbles at the metal-resin interface, as these would interfere with experiments; prepared electrodes that contained such bubbles were discarded. Electrodes were then stored in a desiccator and prior to each measurement the surfaces ground using 1000 grid emery paper and rinsed with distilled water.
2.2. Electrochemical measurements All electrochemical measurements were obtained by using a model PGZ 100 potentiostat (Radiometer Analytical) connected to a personal computer via an RS-232 interface and controlled through the VoltaMaster 4 electrochemical software (Radiometer Analytical). A calomel electrode with saturated KCl (SCE) was used as the reference (all potentials reported were referred to SCE) and a platinum electrode as the auxiliary. Measurements were conducted using a typical three-electrode electrochemical cell with a double-wall arrangement. Such a custom-made design allowed for a hydronic heating of the experimental electrolyte, with both the heating mantle (Electrothermal, UK) and the peristaltic pump (Watson-Marlow 505S, UK) located outside the Faraday cage to minimize electromagnetic interference. The experimental electrolyte was prepared using distilled water and analytical grades of sodium and potassium chlorides (Scharlau Chemie S.A., Spain) at the ratio of 3.7, with the total concentration of chloride ions of 1000 mg dm3. The pH of the electrolyte was adjusted to 5.4 using 0.05 mol dm3 solution of sulphuric acid. In the case of measurements conducted in deaerated electrolytes, high purity nitrogen gas was used for purging for 1 h before starting a measurement and for shielding atmosphere in the cell during the measurements. Potentiodynamic scans for both materials were conducted after a 900 s hold at the free corrosion potential, Ecorr, and then initiated from the adopted Ecorr in either anodic or cathodic direction at a scan rate of 0.1 mV s1. The corrosion current that was expected to take place at the mixed potential of antimony–carbon steel couple was measured using the Evans’ experiment, where the more electronegative material (carbon steel) was used as the working electrode and the less electronegative material (antimony) was the auxiliary electrode. The galvanostatic steps, 0.1–0.2 lA, were then applied to such an arrangement and individual potentials of both electrodes measured against the SCE using the following condition: d(WORK)/dt and d(AUX)/dt < 3 mV min1 or Maximal-Step-Duration equals 30 s. Experiments were repeated three times for each experimental condition and the corrosion current, Icorr, for each measurement established as the intersection of corresponding potential vs. current curves.
3. Results and discussion 3.1. Free corrosion potential Figs. 1 and 2 show respective records of Ecorr for uncoupled antimony and carbon steel electrodes exposed to the aerated and N2-purged electrolyte at ambient temperature, i.e. 20 °C, and at temperature of 60 °C. Ecorr of antimony remained constant throughout the entire period of monitoring under all experimental conditions. Similar behaviour was also observed for carbon steel exposed to the N2-purged electrolyte at both temperatures. Under aerated conditions Ecorr of carbon steel showed a decaying trend, which became more pronounced with increasing temperature. In general, such a behaviour, i.e. decaying trend, is typical for
Fig. 1. Records for the free corrosion potential, Ecorr, of antimony in the aerated and N2-purged electrolyte at 20 and 60 °C.
Fig. 2. Records for the free corrosion potential, Ecorr, of carbon steel in the aerated and N2-purged electrolyte at 20 and 60 °C.
formation of porous hydroxide layers, which slow the rate of metal dissolution, but do not protect it from further dissolution [22]. The values of Ecorr, recorded after 900 s of monitoring, were plotted as a function of temperature in Fig. 3, which shows that antimony remains less electronegative material (cathodic) than carbon steel under all experimental conditions. Ecorr of antimony and carbon steel shift to more active values with both increasing temperature and N2 purging. This observation is not surprising since the Ecorr in the N2-purged environment would be controlled by the H–H+ half-cell reaction, which is on the active side of the standard half-cell potentials scale, when compared to the O2– OH half-cell reaction that would dictate Ecorr in aerated environment. Experimentally, such a shift in Ecorr due to level of solution aeration was reported for instance for AISI Type 304 stainless steel [23] and high purity aluminium [24]. The effect of temperature is also related to the concentration of oxygen in the electrolyte, since its concentration in aqueous solutions generally decreases with increasing temperature; for instance it falls from 14.2 mg dm3 at 1 °C to 6.4 mg dm3 at 40 °C in water [25]. Consequently, diffusion limited current associated with the oxygen reduction reaction decreases with decreasing oxygen concentration, which results in less-efficient consumption of electrons and a negative shift in Ecorr.
164
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
Fig. 3. Free corrosion potential, Ecorr, of carbon steel and antimony measured after 900 s plotted as a function of temperature for the aerated and N2-purged electrolytes.
Furthermore, the effect of temperature on Ecorr is also evident from Eq. (3) that shows that temperature, T, is of fundamental importance in electrochemical processes [26], and as such contributed to the observed dependence of Ecorr:
dE DS ¼ dT zF
ð3Þ
where DS is the change in entropy, z is the valence and F is the Faraday’s constant. Similarly, the concentration dependence of the electrode potential given by the Nernst equation includes temperature dependence in the term RT/zF, where R is the universal gas constant.
have minimal effect on the behaviour within the diffusioncontrolled region. However, an increase in anodic current density is apparent under N2-purged conditions within the IR-controlled region. Considering that dissolved oxygen would not affect resistivity or conductivity of the experimental electrolytes, it may be argued that it is the interfacial process [27], i.e. enhanced formation of the surface oxide film under presence of dissolved oxygen, that slows down charge transfer and leads to the observed differences in the anodic currents within the IR-controlled regions. Also, the anodic current density increases with increasing temperature, which is a generally accepted trend [28] ascribed to the reduction of diffusion layer thickness with increasing temperature. Hydrogen evolution dominates the cathodic behaviour under N2-purged conditions; although, at 20 °C there is some evidence of limiting behaviour, possibly due to presence of a limited amount of oxygen. Hydrogen evolution also dominates the cathodic process under aerated conditions at potentials lower than about 1 VSCE, but the mass-transport-limited oxygen reduction, ilim, is the main cathodic process at more noble potentials. Under both aerated and N2-purged conditions, the cathodic current density shifts to higher values with increasing temperature, which can be ascribed to the effect of temperature on the corrosion kinetics through its effect on the diffusion coefficient, thickness of the diffusion layer, and under aerated conditions also on the solubility of oxygen [29–31]. In general, dependence of diffusion coefficient, D, on temperature can be described through the Arrhenius equation [32] using the following form:
ED D ¼ AD exp RT
ð4Þ
where AD is the pre-exponential factor, ED is the activation energy. 3.3. Potentiodynamic polarisation – antimony
Fig. 4 compares anodic and cathodic potentiodynamic polarisation scans for the carbon steel in aerated and N2-purged electrolytes at 20 and 60 °C. Apart of the shift in Ecorr, discussed in the previous paragraph, as expected, anodic behaviour is dominated by active dissolution (Eq. (1)), at both temperatures. For the scans conducted at the same temperature, aeration and/or N2-purging
Comparison of anodic and cathodic potentiodynamic polarisation scans for antimony under similar experimental conditions as for carbon steel is made in Fig. 5. At ambient temperature and anodic potentials, antimony shows passive behaviour, with the passive current density, ipass, of the order 30 lA cm2 for aerated environment and increasing to about 40 lA cm2 under N2-purged conditions; ipass of a similar order of magnitude has been reported for instance for AISI Type 321 stainless steel in deaerated 0.01 mol dm3 S2 O32 -containing chloride-based neutral solution
Fig. 4. Potentiodynamic polarisation curves for carbon steel exposed to aerated and N2-purged electrolytes at 20 and 60 °C.
Fig. 5. Potentiodynamic polarisation curves for antimony exposed to aerated and N2-purged electrolytes at 20 and 60 °C.
3.2. Potentiodynamic polarisation – carbon steel
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
165
In regards to the corrosion process of antimony in acidic solutions, a three-stage reaction has been proposed for the oxidation [7]:
condition for carbon steel; although, the temperature related-shift to higher ilim appears to be of the same order of magnitude between the two materials. As discussed in great detail elsewhere [39–42], reduction of oxygen is a rather complex phenomenon, which can be associated with an exchange involving two or four electrons and the formation of numerous intermediates, for instance superoxide, O2, hydroxyl radical, OH, and peroxide, H2O2 or HO2. Also, oxygen reduction is strongly affected by adsorption of oxygen molecule, which when adsorbed to a metal surface weakens the O–O bond and the reduction products accelerate the kinetics of the oxygen reduction reaction [41]. Assuming that the N2-purging produced the same drop in oxygen concentration in electrolytes during experiments involving carbon steel and antimony, it may be argued that the difference in cathodic ilim measured between antimony and carbon steel may possibly relate to a different behaviour of molecular oxygen when interacting with different metal surfaces. For instance, Hellsing and Gao [43] reported on such differences for the molecular chemisorption of oxygen on copper, nickel and platinum, which were ascribed to the electronic structure of the metals, i.e. the number of unpaired d electrons.
Sb þ H2 O ! SbOHðadÞ þ Hþ þ e
ð7Þ
3.4. Corrosion current – evans experiments
SbOHðadÞ ! SbOðadÞ þ Hþ þ e
ð8Þ
þ
ð9Þ
Results from the measurements of the corrosion current for the carbon steel–antimony couple at 20 and 60 °C are shown in Figs. 6 and 7. Both figures indicate relatively good reproducibility of obtained measurements; although, greater scatter in the values of corrosion current were observed in the measurements associated with the aerated electrolyte and temperature of 60 °C (Fig. 7). Fig. 8 depicts the dependence of the corrosion current density, icorr, as a function of temperature for both aerated and N2-purged electrolytes, with the measured values, and calculated average and standard deviation values listed in Table 1. It is evident that both temperature and aeration increase the values of icorr. Whilst the effect of increasing temperature on increase of icorr is approximately two-folded in N2-purged environment, i.e. from 2.09 to 4.87 lA cm2, about six times higher corrosion current density is obtained for the couple under aerated conditions, i.e. 7.5 lA cm2 at 20 °C increasing to 40.2 lA cm2 at 60 °C. The magnitude of the corrosion current is determined by the kinetics of the system, which means that the rate of the overall corrosion process is controlled by the slowest step, which in this case
[33]. Wikstrom and Nobe [34] reported similar passive behaviour for antimony in acidic (pH 0–2) chloride containing solutions, with ipass decreasing from approximately 101–103 A cm2 with increasing pH. At 60 °C, anodic behaviour is initially represented by active dissolution and then followed by a distinct active-topassive transition at around 100 mVSCE under both aerated and N2-purged conditions. However, slightly higher passivation current density is observed for this transition in N2-purged electrolyte. According to El Wakkad and Hickling [35], ‘passivation’ of antimony is associated with formation of a thick Sb2O5 film between the metallic antimony and the outer film of Sb2O3, with the similar view also reported elsewhere [10,34,36] and in general making a reference to the following electrochemical reactions:
4 Sb þ 6 H2 O ! 2 Sb2 O3 þ 12 Hþ þ 12 e
ð5Þ
and
3 Sb2 O5 þ 12 Hþ þ 12 e ! 3 Sb2 O3 þ 6 H2 O
SbOðadÞ ! SbO þ e
ð6Þ
The fact that higher current density is observed under N2-purged conditions, points to contribution of oxygen to the overall process of oxidation under aerated conditions, mainly supply of oxygen atoms to form Sb–O–Sb bonds. For such a process we propose the following reaction to occur involving oxygen:
4Sb þ 2H2 O þ O2 ! 4SbOH
ð10Þ
However, it is important to realise that since both the oxides Sb2O3 and Sb2O5 are sparingly soluble to aqueous environment, for antimony there is no protection domain in potential-pH equilibrium diagram due to passivation by an oxide film [37]. Similarly as in the case of carbon steel, cathodic behaviour of antimony also shows dependence on N2-purging and temperature. Whilst the effect of temperature remains the same, as already discussed for the carbon steel case, there are some fundamental differences associated with antimony. More specifically, under N2-purging and at both experimental temperatures measured cathodic current densities show limiting behaviour, which can be ascribed to hydrogen evolution and formation of gaseous stibine [37], SbH3, which can also be accompanied by formation of solid antimony hydride [38], Sb2H2. The following electrochemical reactions might be considered for the antimony based products:
Sb þ 3Hþ þ 3e ! SbH3
ð11Þ
2Sb þ 2Hþ þ 2e ! Sb2 H2
ð12Þ
On the other hand, in aerated solutions cathodic current densities show a transition between two limiting behaviours, with a shift in this transition to more noble potentials from about 600 mVSCE at 60 °C to 500 mVSCE at 20 °C. At potentials closer to Ecorr, the cathodic behaviour is dominated by the reduction of dissolved oxygen and mass-transport-limited current density, ilim. This situation continues with further lowering of the potential until the transition point and increase in cathodic current density due to reduction of antimony to hydrides, as per Eqs. (11) and (12). Under aerated conditions and at the ambient temperature, ilim was about 4 lA cm2, which is lower than the mass-transportlimited current density of 20 lA cm2 measured under the same
Fig. 6. Potential as a function of the applied current for the carbon steel–antimony couple during the Evans’ experiment in aerated and N2-purged solution at 20 °C; icorr indicated by the intersection of the E vs. i curves of antimony (Sb) and carbon steel (CS).
166
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
low, where the slightly higher icorr at 60 °C is ascribed to the temperature-related decrease in thickness of the diffusion layer. In the case of uniform corrosion of carbon steel, such corrosion current densities would represent corrosion rates of about 24 and 56 lm y1 at 20 and 60 °C; for instance, this may be considered within allowable corrosion rates for 30 y life of a plant. However, in the case of localised galvanic corrosion, the accumulated loss on carbon steel could be a few orders of magnitude higher. Under aerated conditions, the oxygen reduction reaction in acidic solutions proceeds at an appreciably faster rate than the hydrogen evolution reaction either through the formation of peroxide, H2O2, as an intermediate or directly to water [44]:
Fig. 7. Potential as a function of the applied current for the carbon steel–antimony couple during the Evans’ experiment in aerated and N2-purged solution at 60 °C; icorr indicated by the intersection of the E vs. i curves of antimony (Sb) and carbon steel (CS).
O2 þ 2Hþ þ 2e ! H2 O2
ð13Þ
H2 O2 þ 2Hþ þ 2e ! 2H2 O
ð14Þ
The corrosion current is limited by the rate of diffusion of dissolved oxygen, and since D increases with increasing temperature the corrosion current, and thus the corrosion rate increases as well. As the rate of diffusion is proportional to the concentration of dissolved oxygen, the corrosion current is also proportional to the dissolved oxygen concentration and, as already discussed earlier, increasing temperature decreases solubility of oxygen in aqueous solutions, which tends to decrease the corrosion rate. {Note: oxygen is being removed at the boiling point.} Overall, as summarised by Porte [31], the corrosion current increases up to a maximum temperature of about 80 °C, and then decreases due to the reduced solubility of oxygen, which becomes the most important factor; however, in a closed system, the corrosion current would increase with increasing temperature until all the oxygen is consumed through the oxygen reduction reaction. 3.5. Corrosion current – prediction through polarisation curves
Fig. 8. Corrosion current, icorr, plotted as a function of temperature for the aerated and N2-purged electrolytes. The error bars indicate standard deviation.
Galvanic corrosion current can be roughly predicted by super-imposing polarisation curves of the anode and the cathode materials. Figs. 9 and 10 depict such an attempt for carbon steel and antimony under considered experimental conditions at 20 and 60 °C, respectively. The approximated corrosion current densities are summarised in Table 2. Upon comparison of the results in Tables 1 and 2, it is evident that icorr values obtained through super-position of potentiodynamic scans and through the Evans’ experiment are relatively comparable. Theoretically, such a fairly
Table 1 Results for the icorr (lA cm2) for the carbon steel-antimony couple. Temperature: 20 °C
Temperature: 60 °C
Aerated
N2-purged
Aerated
N2-purged
8.15 7.12 7.24
2.09 2.13 2.05
37.15 39.59 43.86
4.74 4.85 5.02
Average 7.50
2.09
40.20
4.87
Standard deviation 0.56
0.04
3.39
0.14
is the rate of reduction reactions occurring at the cathode. Under N2-purged conditions, it would be the rates of hydrogen evolution and possibly the reduction of antimony to hydrates. However, these reactions are very slow due to the low concentration of hydrogen ions at the experimental pH of 5.4. Therefore, the measured corrosion current densities in the absence of oxygen are
Fig. 9. Super-imposed respective cathodic and anodic scans for antimony (Sb) and carbon steel (CS) for aerated and N2-purged electrolytes at 20 °C.
J. Soltis, K.A. Lichti / Corrosion Science 68 (2013) 162–167
167
References
Fig. 10. Super-imposed respective cathodic and anodic scans for antimony (Sb) and carbon steel (CS) for aerated and N2-purged electrolytes at 60 °C.
Table 2 Results for the approximate icorr (lA cm2) for the carbon steel-antimony couple. Temperature: 20 °C
Temperature: 60 °C
Aerated
N2-purged
Aerated
N2-purged
8.5
2.1
85.0
3.5
accurate prediction has been discussed elsewhere [45], when Ecorr of the anode and cathode are at least 120 mV apart, which is the case for the system studied in the current work.
4. Conclusions In this work we provide an insight into the galvanic corrosion of carbon steel coupled to antimony in aerated and N2-purged electrolytes at ambient and 60 °C temperatures. Measurements of free corrosion potentials show that (i) antimony and carbon steel show a shift in their Ecorr to more active values with increasing temperature and N2 purging, and (ii) antimony remains less electronegative than carbon steel under all experimental conditions. Aeration and/or N2 purging, and temperature affect potentiodynamic behaviour of both antimony and carbon steel, and thus the corrosion current for the antimony-carbon steel couple. More specifically, the corrosion current increases with increasing temperature and aeration. There was a good agreement between the corrosion currents obtained for the antimony-carbon steel couple through the Evans’ experiment and super-imposition of polarisation scans.
Acknowledgement This work was funded by the New Zealand Foundation for Research, Science and Technology under Contract No. C05X0704.
[1] M.-H. Yang, I.-W. Sun, J. Appl. Electrochem. 33 (2003) 1077–1082. [2] W. Yang, H. Cang, Y. Tang, J. Wang, Y. Shi, J. App. Electrochem. 38 (2008) 537– 542. [3] G. Edwell, Electrochim. Acta 24 (1972) 595–605. [4] D. Pavlov, M. Bojinov, T. Laitinen, G. Sundholm, Electrochim. Acta 36 (1991) 2081–2086. [5] D. Pavlov, M. Bojinov, T. Laitinen, G. Sundholm, Electrochim. Acta 36 (1991) 2087–2092. [6] T. Laitinen, H. Revitzer, G. Sundholm, J. Vilhunen, D. Pavlov, M. Bojinov, Electrochim. Acta 36 (1991) 2093–2102. [7] S. Laihonen, T. Laitinen, G. Sundholm, A. Yli-Pentti, Electrochim. Acta 35 (1990) 229–238. [8] M. Metikos-Hukovic, B. Lovrecek, Electrochim. Acta 23 (1980) 717–723. [9] M. Metikos-Hukovic, B. Lovrecek, Electrochim. Acta 23 (1978) 1371–1376. [10] M. Metikos-Hukovic, R. Babic, S. Omanovic, J. Electroanal. Chem. 374 (1994) 199–206. [11] A. Bregman, Met. Prog. 59 (1951) 245–249. [12] G.R. Schaer, W.H. Safranek, C.L. Faust, Plating 45 (1958) 139–147. [13] R.M. Burns, W.W. Bradley, Protective Coatings for Metals, 2nd ed., Reinhold Publishing Co., New York, NY, 1955. [14] K.G. Soderberg, H.L. Pinkerton, Plating 37 (1950) 254–259. [15] S.S. Abd El Rehim, A. Awad, A. El Sayed, J. Appl. Electrochem. 17 (1987) 156– 164. [16] D.P. Le, W.S. Ji, J.G. Kim, K.J. Jeong, S.H. Lee, Corros. Sci. 50 (2008) 1195–1204. [17] V.G. Golubev, Z.K. Vasileva, Z.K. Murzova, L.A. Bolsun, T.E. Golosova, Khim. Volokna 3 (1988) 12–16. [18] R. Piontelli, L. Fagnani, Korr. und Metallsch 19 (1925) 113–121. [19] A.L. Pitman, M. Pourbaix, N. De Zoubov, in: M. Puorbaix (Ed.), Atlas of Electrochemical Equilibria in Aqueous Solutions: Antimony, Pergamon Press, Brussels, 1966, p. 524. [20] K.L. Brown, S.F. Simmons, Geothermics 32 (2003) 619–625. [21] G. Zhang, C. Liu, H. Liu, Z. Jin, G. Han, L. Li, Geothermics 37 (2008) 73–83. [22] W.S. Tait, An Introduction to Electrochemical Corrosion Testing for Practising Engineers and Scientist, PairODocs Publications, Racine, WI, 1994. [23] J.B. Lee, A.K. Agrawal, R.W. Staehle, EPRI Report NP 1741, Electric Power Institute, Palo Alto, California, 1981. [24] J. Soltis, N.J. Laycock, D. Krouse, Corros. Sci. 53 (2011) 7–10. [25] R.F. Weiss, Deep Sea Res. 17 (1970) 721–735. [26] P. Grundler, A. Kirbs, L. Dunsch, Chem. Phys. Chem. 10 (2009) 1722–1746. [27] Z. Ahmad, Principles of Corrosion Engineering and Corrosion Control, Elsevier Science & Technology Books, Oxford, UK, 2006. [28] E.L. Liening, in: Process Industries Corrosion: Electrochemical Corrosion Testing Techniques, NACE International Inc., Houston, TX, 1986. [29] V.V. Gerasimov, G.V. Akimov, I.L. Rozenfeld, Bul, Acad. USSR Div. Chem. Sci. 5 (1956) 9–11. [30] G.T. Skaperdas, H.H. Uhlig, Ind. Eng. Chem. 34 (1942) 748–754. [31] H.A. Porte, Technical Note N-907, Naval Civil Eng. Laboratory, Port Hueneme, CA, 1967. [32] H.R. Corti, L.N. Trevanib, A. Anderko, in: D.A. Palmer, R. Fernandez-Prini, A.H. Harvey (Eds.), Aqueous Systems at Elevated Temperatures and Pressures: Physical Chemistry in Water, Steam and Hydrothermal Solutions, Elsevier Ltd., 2004, p. 321. [33] Y.M. Liou, S.Y. Chiu, C.L. Lee, H.C. Shih, J. Appl. Electrochem. 29 (1999) 1377– 1381. [34] L.L. Wikstrom, K. Nobe, Corrosion 31 (1975) 364–369. [35] S.E.S. El Wakkad, A. Hickling, J. Phys. Chem. 57 (1953) 203–208. [36] L.L. Wikstrom, N.T. Thomas, K. Nobe, J. Electrochem. Soc. 122 (1975) 1201– 1206. [37] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, National Association of Corrosion Engineers, Houston, TX, 1974. [38] D.T. Hurd, Introduction to Chemistry of the Hydrides, Wiley, New York, NY, 1952. [39] S. Strbac, R.R. Adzic, Electrochim. Acta 41 (1996) 2903–2908. [40] J. Xu, W. Huang, R.L. McCreer, J. Electroanal. Chem. 410 (1996) 235–242. [41] H.H. Yang, R.L. McCreer, J. Electrochem. Soc. 147 (2000) 3420–3428. [42] J.P. Hoare, The Electrochemistry of Oxygen, Interscience Publisher, New York, NY, 1968. [43] B. Hellsing, S. Gao, Solid State Commun. 90 (1994) 223–228. [44] V.S. Bagotskij, L.N. Nekrasov, N.A. Shumilova, Espehy Khimii 34 (1965) 1697– 1703. [45] H.P. Hack, in: R. Baboian (Ed.), Corrosion Tests and Standards Application and Interpretation: Galvanic, American Society for Testing and Materials, Philadelphia, PA, 1995, p. 186.