- Email: [email protected]
Effect of Al content on the corrosion behavior of Mg–Al alloys in aqueous solutions of different pH
Electrochimica Acta 55 (2010) 6651–6658
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Effect of Al content on the corrosion behavior of Mg–Al alloys in aqueous solutions of different pH Ghada M. Abady b , Nadia H. Hilal b , Mohmmed El-Rabiee b , Waheed A. Badawy a,∗ a b
Chemistry Department, Faculty of Science, University of Cairo, 12613 Giza, Egypt Chemistry Department, Faculty of Science, Fayoum University, Fayoum, Egypt
a r t i c l e
i n f o
Article history: Received 28 April 2010 Accepted 4 June 2010 Available online 15 June 2010 Keywords: Corrosion Electrochemical impedance spectroscopy Magnesium–aluminum alloys Open-circuit potential Passivity Polarization techniques
a b s t r a c t The effect of systematic increase of Al content on the electrochemical behavior of the Mg–Al alloys in aqueous solutions of different pH was investigated. Different electrochemical methods such as opencircuit potential measurements, polarization techniques and electrochemical impedance spectroscopy, EIS, were used to investigate the electrochemical behavior of the alloys in aqueous solutions. The results have shown that Mg–5Al is easily corroded due to the microgalvanic effect between ␣-phase and -phase, its corrosion rate is even higher than that of Mg itself. The increase of Al content increases the corrosion resistance of the alloy due to the formation of the -phase (Mg17 Al12 ) together with the Mg ␣-phase. The ranking of the corrosion rate of these alloys was Mg–5Al > Mg > Mg–10Al ∼ = Mg–15Al. The corrosion rates of the alloys in acidic solutions are pronouncedly high compared to those measured in neutral or basic solutions. The impedance measurements are in consistence with the polarization techniques and the impedance data were fitted to theoretical data obtained according to an equivalent circuit model describing the electrode/electrolyte interface. © 2010 Elsevier Ltd. All rights reserved.
1. Introduction Due to their low density, high strength/weight ratio and excellent dimensional stability, magnesium alloys are always used in different applications such as aerospace, automotive, cellular phones, recreational equipments and computer industries, where weight reduction is important [1–3]. The activity and poor corrosion resistance of magnesium and its alloys restrict their wide applications. In the last decades, many research works had concentrated on the effects of microstructure and environmental factors on the corrosion behavior of magnesium and its alloys [4–15]. Magnesium can quickly develop an oxide film on the surface in air, but this oxide, MgO, with the Pilling–Bedworth ratio of 0.81 can only provide limited protection [9]. In wet air and aqueous solutions, oxide and hydroxide layers of Mg, MgO/Mg(OH)2 , can form spontaneously on the surface. Mg(OH)2 has the Pilling–Bedworth ratio of 1.77, and cannot provide enough corrosion protection to the metallic surface [13]. The morphology and structure of the film formed on pure magnesium after 1 h immersion in water were investigated by transmission electron microscopy, TEM, and was found to consist of three layers [6]. An inner cellular layer of 0.4–0.6 nm
∗ Corresponding author. Tel.: +20 2 3567 6558; fax: +20 2 3568 5799. E-mail addresses: [email protected], [email protected] (W.A. Badawy). 0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2010.06.007
thickness followed by a middle dense layer of 20–40 nm composed mainly of MgO, then an outer platelet-like layer of Mg(OH)2 . Both pure magnesium and its alloys have poor pitting resistance in aqueous solutions containing aggressive anions, especially chloride ions. The oxide and hydroxide films tend to breakdown chemically in solutions containing chloride, sulfate or bromide. The protective nature of the oxide films depends on the formation parameters, the chemical compositions of the metal or the alloy and the corrosive medium [16,17]. The high reactivity of Mg has a detrimental effect on coating qualities including adhesion, pore density and uniformity. It is therefore important to get more information about the electrochemical behavior of the metal and its alloys. Such information is necessary to understand the corrosion mechanism of these materials in aqueous solutions. In this paper, we are reporting on the effect of Al as an important alloying element with Mg on the electrochemical behavior of the metal and the formed alloys. The corrosion parameters were calculated and the mechanism of the corrosion process was discussed. 2. Experimental The working electrodes were made from commercial grade Mg and Mg–Al rodes, mounted into glass tubes by two-component epoxy resin leaving a surface area of 0.385 cm2 to contact the solution. The mass-spectrometric analysis of the electrodes used is presented in Table 1. The electrochemical cell was a three-
6652
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
Table 1 Mass-spectrometric analysis of the investigated materials (mass%). Alloys
Mg
Al
Cu
Ni
Fe
Mg Mg–5Al Mg–10Al Mg–15Al
99.9 94.9 89.9 84.9
– 5.0 10.0 15.0
0.03 0.03 0.03 0.03
0.03 0.03 0.03 0.03
0.04 0.04 0.04 0.04
electrode all-glass cell, with a platinum counter electrode and saturated calomel, SCE, reference electrode. Before each experiment, the working electrode was polished mechanically using successive grades emery papers up to 2000 grit. The electrode was washed thoroughly with triple distilled water, and transferred quickly to the electrolytic cell. The electrochemical measurements were carried out in stagnant, naturally aerated aqueous solutions of pH 2 (62.5 mL 0.2 mol L−1 boric acid + 62.5 mL 0.05 mol L−1 borax + 62.5 mL 0.4 mol L−1 H3 PO4 and 52.5 mL H2 O), pH 7 (127.8 mL 0.1 mol L−1 COOKC6 H4 COOH + 122.2 mL 0.1 mol L−1 HCl) and pH 12 (100 mL 0.2 mol L−1 KH2 PO4 + 163.7 mL 0.2 mol L−1 NaOH). Before each measurement, the pH of the test electrolyte was controlled by a standard pH-meter. The polarization experiments and electrochemical impedance spectroscopic investigations, EIS, were performed using a Voltalab PGZ 100 “All-in-one” potentiostate/galvanostate system. The potentials were referred to the standard potential of the SCE (0.245 V vs. nhe). All the polarization experiments were carried out using a scan rate of 10 mV s−1 . The impedance, Z, and phase shift, , were measured in the frequency domain 0.1–105 Hz. The superimposed AC-signal was 10 mV peak to peak amplitude. To achieve reproducibility, each experiment was carried out at least twice. Details of experimental procedures are as described elsewhere [18,19]. 3. Results and discussion 3.1. Open-circuit potential measurements The open-circuit potential of Mg and Mg–Al alloys with different Al contents was traced over 180 min in stagnant, naturally aerated aqueous solutions of different pH covering the acidic, neutral and basic media. The electrode potential for all alloys was found to shift towards less negative values and the steady state was reached in less than 20 min from electrode immersion in neutral solutions. In acidic and basic electrolytes it needs more than 100 min from electrode immersion in the electrolyte to be achieved as presented in Fig. 1(a–c). This means that the electrode surface is subjected to more corrosion and passivation processes in acidic and basic solutions and stabilizes readily in neutral solution as will be confirmed by the potentiodynamic and impedance measurements.
Fig. 1. Variation of the open-circuit potential of Mg electrode and the Mg–Al alloys with time in stagnant naturally aerated aqueous solutions at 25 ◦ C. (a) pH = 2, (b) pH = 7 and (c) pH = 12.
3.2. Potentiodynamic measurements Fig. 2 presents the potentiodynamic polarization behavior of Mg, and the three different Mg–Al alloys in the acidic solutions of pH 2 (Fig. 2a), neutral solutions of pH 7 (Fig. 2b) and basic solutions of pH 12 (Fig. 2c). The corrosion parameters i.e. corrosion potential, Ecorr , corrosion current density, icorr , and corrosion resistance, Rcorr , were calculated from polarization data and presented in Table 2. The results show that the presence of small amounts of Al (≤5%) in the Mg matrix did not improve the corrosion resistance of the material. In the contrary the corrosion rate of Mg–5Al is higher than that of Mg itself (cf. Fig. 3 and Table 2). The increase of the Al content leads to an increase in the corrosion resistance of the alloy. The Mg–10Al alloy has shown the lowest corrosion rate, especially in neutral and basic solutions. This observation can be explained
on the basis that, the addition of Al to Mg produces a secondary intermetallic phase, -Mg17 Al12 , together with the single ␣-phase matrix of pure Mg. As the Al content increases up to 8%, the matrix becomes anodically active and the rate of corrosion increases due to microgalvanic effects [20]. Above this ratio the corrosion resistance increases due to the increased amount of the -phase, which decreases the microgalvanic effect occurring between ␣- and phases [8,21,22]. As can be seen from Fig. 3 and the data in Table 2, the investigated materials can be ranked according to their rates of corrosion in neutral and basic solutions in the following order Mg–5Al > Mg > Mg–10Al ∼ = Mg–15Al. The presence of microgalvanic effect between the ␣-phase and the -phase that formed due to the presence of the small amount of Al is responsible for the rel-
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
6653
Fig. 2. Potentiodynamic polarization curves of Mg and the Mg–Al alloys in stagnant naturally aerated aqueous solutions at a scan rate of 10 mV s−1 at 25 ◦ C. (a) pH = 2, (b) pH = 7 and (c) pH = 12.
atively high corrosion rate recorded with the Mg–5Al alloy. The -phase acts, generally, as an effective cathode which accelerates the corrosion of the adjacent ␣-phase [21,22]. The maximum galvanic current and hence the high rate of corrosion occur at a certain ratio of anodic area to cathodic area. As it can be expected for active metals, the rate of the corrosion of these metallic materials in acidic medium is relatively high compared to that in neutral or basic solutions; it is more than 10 times its value in basic solutions. The data presented in Fig. 3 show clearly that the three investigated materials have relatively low corrosion rates in basic solutions. This can be explained by the formation of a barrier layer of Mg(OH)2 , which
is insoluble in basic solutions [23]. In acidic solutions, the barrier layer is completely soluble and hence relatively high corrosion rates were recorded. In neutral solutions, the barrier magnesium hydroxide layer is partially soluble and so a decrease in the corrosion rate was recorded. 3.3. Effect of temperature The effect of temperature on the corrosion rates of Mg, Mg–5Al, Mg–10Al and Mg–15Al was investigated in stagnant, naturally aerated solutions of pH 2, 7 and 12 using the potentiodynamic
Table 2 Corrosion parameters of Mg and the Mg–Al alloys after 90 min of the electrode immersion in stagnant, naturally aerated aqueous solutions of pH 2, 7 and 12 at 25 ◦ C. Ecorr /mV
Rct / cm2
icorr /A cm−2
Ba /mV dec−1
Bc /mV dec−1
Corrosion rate/m Y−1
pH = 2 Mg Mg–5Al Mg–10Al Mg–15Al
−1457.6 −1728.4 −1458.2 −1537.5
87.5 79.8 84.9 118.5
229.8 265.6 166.2 134.7
124.9 135.9 109.4 101.1
−128.8 −136.6 −104.3 −98.5
5245 4089 3793 3073
pH = 7 Mg Mg–5Al Mg–10Al Mg–15Al
−1550.0 −1701.4 −1492.9 −1524.0
253.9 368.3 411.1 320.8
68.2 84.2 61.4 75.8
137.7 236.7 212.7 222.7
−86.4 −148.3 −112.3 −86.2
1555 1921 1401 1730
pH = 12 Mg Mg–5Al Mg–10Al Mg–15Al
−1426.2 −1520.5 −1294.2 −1406.4
5.09 15.3 1.72 1.83
195.9 9.70 52.4 80.7
−108.6 −128.0 −78.6 −52.7
116 348 39 41
4250 751 6520 8480
6654
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
Fig. 3. Corrosion rate for Mg and the Mg–Al alloys as a function of pH at 25 ◦ C.
polarization technique. In general, the corrosion rate which is represented by the corrosion current density, icorr , increases with the increase of temperature, and the process follows the familiar Arrhenius equation [24]. d log icorr Ea =− R d(1/T ) where Ea is the molar activation energy of the process and R is the gas constant (8.314 J mol−1 K−1 ). The relations between log icorr and 1/T for the investigated alloys are presented in Fig. 4. The activation energy, Ea , values of the corrosion processes in the different solutions were calculated and presented in Table 3. The low activation energy calculated in acidic solutions indicates the high corrosion tendency in this medium, which agrees well with both the polarization and impedance measurements. Generally, the activation energy values are less than 40 kJ mol−1 . This means that the rate determining step for the dissolution process is a one electron transfer process [25,26]. 3.4. The electrochemical impedance measurements EIS is essentially a steady state technique that is capable of accessing relaxation phenomena whose relaxation times vary over orders of magnitudes and permits single averaging within a single experiment to obtain high precision levels. The open-circuit impedance of Mg and its alloys was traced over 120 min from electrode immersion in the test solutions. Typical data for the Mg and Mg–Al alloys in aqueous solutions of pH 2, 7 and 12 are presented as Bode plots in Fig. 5. Bode plots are recommended as standard impedance plots, since the whole impedance data are equally represented and the phase angle, , as a sensitive parameter for indicating the presence of any additional time constants for interfacial phenomena appears explicitly [27–30]. To have more insight on the impedance data, the impedance vs. frequency data were plotted separately from the phase shift vs. frequency data. The two figures recorded in acidic solution are assigned as Fig. 5a and a* Table 3 Activation energy of Mg and the different Mg–Al alloys after 90 min of the electrode immersion in stagnant, naturally aerated aqueous solutions of pH 2, 7 and 12. Materials
Mg Mg–5Al Mg–10Al Mg–15Al
Ea /kJ mol−1 pH 2
pH 7
pH 12
8.33 6.25 10.00 10.68
9.55 7.13 10.39 11.27
10.41 8.17 11.81 11.80
Fig. 4. Arrhenius plots of Mg and the Mg–Al alloys in stagnant naturally aerated aqueous solutions. (a) pH = 2, (b) pH = 7 and (c) pH = 12.
for the impedance and phase, respectively. It is clear that in acidic solution continuous corrosion is taking place for all materials and relatively small total impedance, Z, at low frequency was recorded. The phase shift shows two clear maxima indicating the presence of two time constants controlling the corrosion process. In neutral and basic solution passivation phenomena could be recorded and the value of absolute impedance, Z, and the diameter of the phase maximum show that the corrosion mechanism is dependent on the corroding material. The rate of corrosion
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
6655
Fig. 5. Bode plots [a–c represent log Z vs. log f; and a*–c* represent log f vs. ] of Mg and Mg–Al alloys in stagnant naturally aerated aqueous solutions at 25 ◦ C. (a, a*) pH = 2, (b, b*) pH = 7 and (c, c*) pH = 12.
decreases with the increase of the amount of Al in the alloy. The Bode plots in neutral solution are presented as Fig. 5b and b* for the impedance and phase, respectively and those in basic solution are presented as Fig. 5c and c* in the same sequence. In all solutions relatively higher impedance values were recorded for Mg–10Al and Mg–15Al (cf. Fig. 5). Lower impedance values were recorded for Mg and Mg–5Al. The phase maximum is broader in the case of Mg–15Al and Mg–10Al compared to Mg and Mg–5Al, which means a higher corrosion resistance in case of both alloys. It means that the corrosion resistance of the alloy increases with the increase of the Al content. The impedance data were analyzed using the dispersion formula for a simple equivalent circuit model consisting of a parallel combination of a capacitor, Cdl , and a resistor, Rct , in series with a resistor, Rs , representing the solution resistance. The electrode impedance, Z, in this case is represented by the mathematical formulation:
Z = Rs +
Rct 1 + (2fRct Cdl )
˛
where ˛ denotes an empirical parameter (0 ≤ ˛ ≤ 1) and f is the frequency in Hz. The above relation takes into account the deviation from the ideal capacitor, RC, behavior in terms of a distribution of time constants due to surface inhomogeneties, roughness effect, and variations in properties or compositions of surface layers [31–33]. The impedance spectra obtained experimentally were analyzed using software provided with the electrochemical workstation. According to AC circuit theory, an impedance plot obtained for a given electrochemical system can be correlated to one or more equivalent circuits [34]. The impedance data for the Mg alloys in aqueous solutions of different pH were analyzed using the equivalent circuit shown in Fig. 6. In this model a combination Rf Cf was introduced to the simple Randle’s cell to account for the presence of surface film and a Warburg impedance, Zw , to account for the diffusion process recorded in the basic solutions. The calculated equivalent circuit parameters for Mg, Mg–5Al, Mg–10Al and Mg–15Al alloys are presented in Table 4. The corrosion rate is inversely proportional to Rct , where higher Rct values were obtained for Mg–15Al and Mg–10Al and lower values for Mg and Mg–5Al. In general, Cdl , values for the magnesium alloys is low and can be attributed to the formation of relatively thick and compact pro-
tective film on the metal surface [35,36]. In the present study, an increase in Rct is not accomplished by the reduction of Cdl , however, the Cdl values are increased with an increase in Rct values. Similar observation was recorded on La additions to AZ91 alloy [35]. It has been shown that these alloys cannot form effective protective film in aqueous solutions. The Bode phase plots of the investigated materials in acidic solutions (Fig. 5a*) show two phase maxima at low and high frequencies, which assign the presence of two time constants controlling the corrosion process. The low frequency loop was attributed to a charge transfer resistance and a double-layer capacitance at the electrode surface, while the high frequency loop was attributed to the presence of a partially formed surface film of Mg(OH)2 [37–39]. In neutral solutions, a suppressed phase maximum in the low frequency region was present and the high frequency region phase maximum dominates (cf. Fig. 5b*). This means that the corrosion rate of Mg and its alloys decreases by the increase of the solution pH, which can be attributed to the accumulation of corrosion products on the metallic surface. The
Fig. 6. Equivalent circuit models for fitting of the experimental impedance data of Mg and its alloys, Rs = solution resistance, Rct = charge transfer (corrosion) resistance, Cdl = electrode capacitance, Rf = film resistance, Cf = film capacitance and Zw is the Warburg impedance. (a) Model for fitting the data in neutral solutions. (b) Model for fitting the data in basic solutions, where Zw was introduced to account for the diffusion controlled process.
6656
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
Table 4 Equivalent circuit parameters of Mg and the different Mg–Al alloys after 90 min of the electrode immersion in stagnant, naturally aerated aqueous solutions of pH 2, 7 and 12 at 25 ◦ C. pH
Sample
Rs /
Rct /k cm2
Cdl /F cm−2
2
Mg Mg–5Al Mg–10Al Mg–15Al
73.2 62.2 81.3 84.1
2.219 2.670 7.94 11.83
35.9 10.2 6.3 7.61
7
Mg Mg–5Al Mg–10Al Mg–15Al
53.7 57.2 51.1 53.8
0.66 0.42 0.38 0.88
12
Mg Mg–5Al Mg–10Al Mg–15Al
13.7 24.9 10.1 5.1
2.58 0.53 2.37 5.59
9.98 5.94 6.76 8.67 12.3 9.5 16.8 35.9
Rf / cm2
9.90
Cf /F cm−2
40.3
333.0 356.6 389.5 408.6
23.9 11.2 81.7 61.5
385
75
corrosion products thus formed are more stable in neutral solutions but soluble in acidic solutions [23]. In basic solutions only one phase maximum is recorded (cf. Fig. 5c*). The disappearance of the second phase maximum and the relatively high impedance values are good indications for the presence of a stable protective layer on the metallic surface in this medium. The presence of a diffusion process in basic solutions was also recorded as linear part at low frequencies in the Nyquist diagram presented in Fig. 7. The effect of Al on the corrosion behavior of Mg can be clearly understood by comparing the potential–pH diagram of Mg and Mg–Al presented in Fig. 8a and b [40]. According to the presented diagrams, Mg corrodes in aqueous solutions through the reduction of water with the formation of Mg(OH)2 that is soluble in acidic solutions leading to the recorded high corrosion rates in these media. In basic solutions a stable Mg(OH)2 protective film is formed on the metallic surface which inhibits the corrosion process and lower corrosion rates are recorded. 3.5. Surface morphology At pH ≥ 4 Al2 O3 ·3H2 O is formed on the alloy surface leading to an inhibition of the corrosion process and stabilizing the Mg(OH)2 layer. In the pH range 8–12, Mg aluminate is formed and a blocking of the active centers by Mg(OH)2 takes place. The formation of microstructures due to the presence of the alloying elements i.e. Al, leads to the stabilization of the alloy and increases its corrosion resistance [10,41–43]. The role of Al in the formation of the
Fig. 7. Nyquist plot for Mg and Mg alloys in stagnant naturally aerated basic solution of pH 12 at 25 ◦ C.
Fig. 8. Potential–pH diagrams of Mg (a) and Mg–Al alloy (b).
stabilizing -phase is presented on the scanning electron micrographs of Figs. 9 and 10. In Fig. 9 the polished surfaces of Mg (Fig. 9a) and Mg–10Al (Fig. 9b) alloys are presented. It is clear from this figure that the magnesium surface consists of a smooth easily corroding magnesium matrix i.e. ␣-phase. The presence of Al as alloying element leads to the formation of large areas of the phase i.e. Mg17 Al12 intermetallic. The large area of the intermetallic in Mg–10Al alloy compared to the Mg–5Al leads to a decrease in the corrosion rate due to modification of the anodic to cathodic areas that are inhibited in low ratio Al alloys. The comparison between the corrosion behavior of the Mg itself and the Mg–10Al alloy is presented clearly in Fig. 10. The micrograph of Fig. 10a shows clearly that the Mg surface is subjected to severe corrosion even in neutral solutions. The Mg–10Al alloy under the same conditions shows only localized attack which leads to the recorded lower corrosion rate (cf. Fig. 10b). It is clear from the presented data that Mg is active and oxidized easily to the divalent cation. The oxidation process is taking place in two steps in accordance with a one electron transfer kinetics [44,45] which is consistent with the low activation energy of the corrosion process [18,26]. The first charge transfer step is the formation of the Mg+ ion which is readily oxidized to Mg2+ in the presence of hydrogen ions in acidic solutions or water in neutral solutions according the: Mg → Mg+ + e
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
6657
Fig. 9. Scanning electron micrographs of mechanically polished Mg (a) and Mg–10Al alloy (b).
Fig. 10. Scanning electron micrographs of Mg (a) and Mg–10Al alloy (b) after 90 min immersion in naturally aerated neutral solutions of pH 7 at 25 ◦ C.
References 2Mg+ + 2H+ → 2Mg2+ + H2 2Mg+ + 2H2 O → 2Mg2+ + 2OH− + H2 This leads to the formation of the corrosion product Mg(OH)2 , which is stable in neutral and basic solutions and leads to the pronounced increase of the corrosion resistance and impedance of the metallic surface in these solutions. The presence of Al leads to the formation of -Mg17 Al12 intermetallic, which is more stable than ␣-Mg matrix. This intermetallic acts as an affective barrier and as an active cathode for the ␣-matrix. As the ratio of Al increases the ratio of -matrix increases and a continuous oxide film is formed on its surface which inhibits the corrosion process [10,43]. 4. Conclusions Mg–10Al has reasonably high corrosion resistance compared to Mg or Mg–5Al. The increased corrosion resistance is due to the formation of large areas of the Mg17 Al12 intermetallic, which leads to the formation of microstructures that stabilizes the alloy and increases its corrosion resistance. Any increase in the Al content e.g. Mg–15Al had little effect on the corrosion characteristics. In neutral and basic solutions the surface Mg(OH)2 layer is stable and the formation of Al2 O3 ·3H2 O and also Mg aluminates stabilizes the alloy surface and hence low corrosion rates are recorded. The calculated activation energy of the corrosion process is 10 kJ mol−1 , which assigns a one electron charge transfer as a rate determining step. The dissolved Mg+ ions are readily transformed into the stable Mg2+ and the formation of Mg(OH)2 . This corrosion product is stable in neutral and basic solutions and completely soluble in acidic solutions.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]
Y. Zhang, C. Yan, F. Wang, et al., Surf. Coat. Technol. 161 (2002) 36. S.J. Kim, Y. Mizutani, R. Ichino, et al., Surf. Coat. Technol. 169/170 (2003) 143. Z. Zhang, L.J. Zhang, J.J. Fan, et al., Electrochim. Acta 52 (2007) 5325. N. Pebere, C. Riera, F. Dabosi, Electrochim. Acta 35 (1990) 555. G.L. Makar, J. Kruger, Int. Mater. Rev. 38 (1993) 138. J.H. Nordien, S. Ono, N. Masuko, K. Nisancioglu, J. Electrochem. Soc. 142 (1995) 3320. G. Song, A. Atrens, D. John, X. Wu, J. Nairn, Corros. Sci. 39 (1997) 1981. G. Song, A. Atrens, X. Wu, B. Zhang, Corros. Sci. 40 (1998) 1769. G. Song, A. Atrens, M. Dargusch, Corros. Sci. 41 (1999) 249. R. Ambat, N.N. Aung, W. Zhou, Corros. Sci. 42 (2000) 1433. G. Baril, N. Pebere, Corros. Sci. 43 (2001) 471. G. Baril, C. Blanc, N. Pebere, J. Electrochem. Soc. 148 (2001) 489. P. Schmutz, V. Guillaumin, R.S. Lillard, J.A. Lillard, G.S. Frankel, J. Electrochem. Soc. 150 (2003) 99. S. Mathieu, C. Rapin, J. Steinmetz, P. Steinmetz, Corros. Sci. 45 (2003) 2741. N. LeBozec, M. Jonsson, D. Thierry, Corrosion 60 (2004) 356. B. Luan, J.E. Gray, J. Alloys Compd. 336 (2002) 88. Z. Shi, G. Song, A. Atrens, Corros. Sci. 47 (2005) 2760. W.A. Badawy, F.M. Al-Kharafi, A.S. El-Azab, Corros. Sci. 41 (1999) 709. R.M. El-Sherif, K.M. Ismail, W.A. Badawy, Electrochim. Acta 49 (2004) 5139. O. Lunder, J.E. Lein, T.K. Aune, K. Nisancioglu, Corrosion 45 (1989) 741. G. Song, A. Atrens, Adv. Eng. Mater. 1 (1999) 33. M. Schutze, W. Cahn, P. Haasen, E.J. Kramer (Eds.), Material Science and Technology Series, Wiley-VCH, Weinheim, 2000, p. 131. W.S. Loose, in: L.M. Pidgeon, J.C. Mathes, N.E. Woldmen (Eds.), Corrosion and Protection of Magnesium, ASM Int., Materials Park, OH, 1946, p. 173. R.G. Martimer, Physical Chemistry, The Benjamin/Cummings Publishing Company, Inc, Redwood City, CA, 1983, p. 822. A. Wieckowski, E. Ghali, Electrochim. Acta 30 (1985) 1423. W.A. Badawy, F.M. Al-Kharafi, J.R. Al-Ajmi, J. Appl. Electrochem. 30 (2000) 693. W.A. Badawy, S.S. El-Egamy, K.M. Ismail, Br. Corros. J. 28 (1993) 133. A.J. Brock, G.C. Wood, Electrochim. Acta 12 (1967) 395. D.D. Macdonald, S. Real, S.I. Smedly, M. Uraquidi-Macdonald, J. Electrochem. Soc. 135 (1988) 2410. C.M.A. Brett, Corros. Sci. 33 (1992) 203. A.E. Bohe, J.R. Vilche, K. Juettner, W.J. Lorenz, W. Paatsch, Electrochim. Acta 34 (1989) 1443. K. Hladky, L.M. Calow, J.L. Dawson, Br. Corros. 15 (1980) 20. J. Hitzig, J. Titz, K. Juettner, W.J. Lorenz, E. Schmidt, Electrochim. Acta 29 (1984) 287.
6658
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
[34] J.R. Macdonald, Impedance Spectroscopy, John Wiley & Sons, New York, 1987. [35] Y. Fan, W. Guohua, G. Hongtao, L. Guanqun, Z. Chunquan, J. Mater. Sci. 41 (2006) 5409. [36] R. Udhayan, D.P. Bhatt, J. Power Sources 63 (1996) 103. [37] F. Rosalbino, E. Angelini, S. De Negri, A. Saccone, S. Delfino, Intermetallics 13 (2005) 55. [38] F. Rosalbino, E. Angelini, S. De Negri, A. Saccone, S. Delfino, Intermetallics 14 (2006) 1487. [39] Z. Xuehua, H. Yuanwei, W. Zhongling, C. Qiurong, G. Fuxing, Corros. Sci. 48 (2006) 4223.
[40] M. Pourbaix, Atlas of Electrochemical Equilibrium Diagram in Aqueous Solutions, NACE, Houston, TX, 1966. [41] S. Mathieu, C. Rapin, J. Steinmetz, P.A. Steinmetz, Corros. Sci. 45 (2003) 2741. [42] G. Song, A.L. Bowles, D.H. St. John, Mater. Sci. Eng. A 366 (2004) 74. [43] Y. Song, D. Shan, R. Chen, E.-H. Han, Corros. Sci. 51 (2009) 1087. [44] A.J. Bard, Encyclopedia of Electrochemistry of the Elements, vol. VIII, Marcel Dekker, New York, 1978. [45] W.A. Badawy, N.H. Helal, M.M. El-Rabiee, H. Nady, Electrochem. Acta 55 (2010) 1880.
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Effect of Al content on the corrosion behavior of Mg–Al alloys in aqueous solutions of different pH Ghada M. Abady b , Nadia H. Hilal b , Mohmmed El-Rabiee b , Waheed A. Badawy a,∗ a b
Chemistry Department, Faculty of Science, University of Cairo, 12613 Giza, Egypt Chemistry Department, Faculty of Science, Fayoum University, Fayoum, Egypt
a r t i c l e
i n f o
Article history: Received 28 April 2010 Accepted 4 June 2010 Available online 15 June 2010 Keywords: Corrosion Electrochemical impedance spectroscopy Magnesium–aluminum alloys Open-circuit potential Passivity Polarization techniques
a b s t r a c t The effect of systematic increase of Al content on the electrochemical behavior of the Mg–Al alloys in aqueous solutions of different pH was investigated. Different electrochemical methods such as opencircuit potential measurements, polarization techniques and electrochemical impedance spectroscopy, EIS, were used to investigate the electrochemical behavior of the alloys in aqueous solutions. The results have shown that Mg–5Al is easily corroded due to the microgalvanic effect between ␣-phase and -phase, its corrosion rate is even higher than that of Mg itself. The increase of Al content increases the corrosion resistance of the alloy due to the formation of the -phase (Mg17 Al12 ) together with the Mg ␣-phase. The ranking of the corrosion rate of these alloys was Mg–5Al > Mg > Mg–10Al ∼ = Mg–15Al. The corrosion rates of the alloys in acidic solutions are pronouncedly high compared to those measured in neutral or basic solutions. The impedance measurements are in consistence with the polarization techniques and the impedance data were fitted to theoretical data obtained according to an equivalent circuit model describing the electrode/electrolyte interface. © 2010 Elsevier Ltd. All rights reserved.
1. Introduction Due to their low density, high strength/weight ratio and excellent dimensional stability, magnesium alloys are always used in different applications such as aerospace, automotive, cellular phones, recreational equipments and computer industries, where weight reduction is important [1–3]. The activity and poor corrosion resistance of magnesium and its alloys restrict their wide applications. In the last decades, many research works had concentrated on the effects of microstructure and environmental factors on the corrosion behavior of magnesium and its alloys [4–15]. Magnesium can quickly develop an oxide film on the surface in air, but this oxide, MgO, with the Pilling–Bedworth ratio of 0.81 can only provide limited protection [9]. In wet air and aqueous solutions, oxide and hydroxide layers of Mg, MgO/Mg(OH)2 , can form spontaneously on the surface. Mg(OH)2 has the Pilling–Bedworth ratio of 1.77, and cannot provide enough corrosion protection to the metallic surface [13]. The morphology and structure of the film formed on pure magnesium after 1 h immersion in water were investigated by transmission electron microscopy, TEM, and was found to consist of three layers [6]. An inner cellular layer of 0.4–0.6 nm
∗ Corresponding author. Tel.: +20 2 3567 6558; fax: +20 2 3568 5799. E-mail addresses: [email protected], [email protected] (W.A. Badawy). 0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2010.06.007
thickness followed by a middle dense layer of 20–40 nm composed mainly of MgO, then an outer platelet-like layer of Mg(OH)2 . Both pure magnesium and its alloys have poor pitting resistance in aqueous solutions containing aggressive anions, especially chloride ions. The oxide and hydroxide films tend to breakdown chemically in solutions containing chloride, sulfate or bromide. The protective nature of the oxide films depends on the formation parameters, the chemical compositions of the metal or the alloy and the corrosive medium [16,17]. The high reactivity of Mg has a detrimental effect on coating qualities including adhesion, pore density and uniformity. It is therefore important to get more information about the electrochemical behavior of the metal and its alloys. Such information is necessary to understand the corrosion mechanism of these materials in aqueous solutions. In this paper, we are reporting on the effect of Al as an important alloying element with Mg on the electrochemical behavior of the metal and the formed alloys. The corrosion parameters were calculated and the mechanism of the corrosion process was discussed. 2. Experimental The working electrodes were made from commercial grade Mg and Mg–Al rodes, mounted into glass tubes by two-component epoxy resin leaving a surface area of 0.385 cm2 to contact the solution. The mass-spectrometric analysis of the electrodes used is presented in Table 1. The electrochemical cell was a three-
6652
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
Table 1 Mass-spectrometric analysis of the investigated materials (mass%). Alloys
Mg
Al
Cu
Ni
Fe
Mg Mg–5Al Mg–10Al Mg–15Al
99.9 94.9 89.9 84.9
– 5.0 10.0 15.0
0.03 0.03 0.03 0.03
0.03 0.03 0.03 0.03
0.04 0.04 0.04 0.04
electrode all-glass cell, with a platinum counter electrode and saturated calomel, SCE, reference electrode. Before each experiment, the working electrode was polished mechanically using successive grades emery papers up to 2000 grit. The electrode was washed thoroughly with triple distilled water, and transferred quickly to the electrolytic cell. The electrochemical measurements were carried out in stagnant, naturally aerated aqueous solutions of pH 2 (62.5 mL 0.2 mol L−1 boric acid + 62.5 mL 0.05 mol L−1 borax + 62.5 mL 0.4 mol L−1 H3 PO4 and 52.5 mL H2 O), pH 7 (127.8 mL 0.1 mol L−1 COOKC6 H4 COOH + 122.2 mL 0.1 mol L−1 HCl) and pH 12 (100 mL 0.2 mol L−1 KH2 PO4 + 163.7 mL 0.2 mol L−1 NaOH). Before each measurement, the pH of the test electrolyte was controlled by a standard pH-meter. The polarization experiments and electrochemical impedance spectroscopic investigations, EIS, were performed using a Voltalab PGZ 100 “All-in-one” potentiostate/galvanostate system. The potentials were referred to the standard potential of the SCE (0.245 V vs. nhe). All the polarization experiments were carried out using a scan rate of 10 mV s−1 . The impedance, Z, and phase shift, , were measured in the frequency domain 0.1–105 Hz. The superimposed AC-signal was 10 mV peak to peak amplitude. To achieve reproducibility, each experiment was carried out at least twice. Details of experimental procedures are as described elsewhere [18,19]. 3. Results and discussion 3.1. Open-circuit potential measurements The open-circuit potential of Mg and Mg–Al alloys with different Al contents was traced over 180 min in stagnant, naturally aerated aqueous solutions of different pH covering the acidic, neutral and basic media. The electrode potential for all alloys was found to shift towards less negative values and the steady state was reached in less than 20 min from electrode immersion in neutral solutions. In acidic and basic electrolytes it needs more than 100 min from electrode immersion in the electrolyte to be achieved as presented in Fig. 1(a–c). This means that the electrode surface is subjected to more corrosion and passivation processes in acidic and basic solutions and stabilizes readily in neutral solution as will be confirmed by the potentiodynamic and impedance measurements.
Fig. 1. Variation of the open-circuit potential of Mg electrode and the Mg–Al alloys with time in stagnant naturally aerated aqueous solutions at 25 ◦ C. (a) pH = 2, (b) pH = 7 and (c) pH = 12.
3.2. Potentiodynamic measurements Fig. 2 presents the potentiodynamic polarization behavior of Mg, and the three different Mg–Al alloys in the acidic solutions of pH 2 (Fig. 2a), neutral solutions of pH 7 (Fig. 2b) and basic solutions of pH 12 (Fig. 2c). The corrosion parameters i.e. corrosion potential, Ecorr , corrosion current density, icorr , and corrosion resistance, Rcorr , were calculated from polarization data and presented in Table 2. The results show that the presence of small amounts of Al (≤5%) in the Mg matrix did not improve the corrosion resistance of the material. In the contrary the corrosion rate of Mg–5Al is higher than that of Mg itself (cf. Fig. 3 and Table 2). The increase of the Al content leads to an increase in the corrosion resistance of the alloy. The Mg–10Al alloy has shown the lowest corrosion rate, especially in neutral and basic solutions. This observation can be explained
on the basis that, the addition of Al to Mg produces a secondary intermetallic phase, -Mg17 Al12 , together with the single ␣-phase matrix of pure Mg. As the Al content increases up to 8%, the matrix becomes anodically active and the rate of corrosion increases due to microgalvanic effects [20]. Above this ratio the corrosion resistance increases due to the increased amount of the -phase, which decreases the microgalvanic effect occurring between ␣- and phases [8,21,22]. As can be seen from Fig. 3 and the data in Table 2, the investigated materials can be ranked according to their rates of corrosion in neutral and basic solutions in the following order Mg–5Al > Mg > Mg–10Al ∼ = Mg–15Al. The presence of microgalvanic effect between the ␣-phase and the -phase that formed due to the presence of the small amount of Al is responsible for the rel-
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
6653
Fig. 2. Potentiodynamic polarization curves of Mg and the Mg–Al alloys in stagnant naturally aerated aqueous solutions at a scan rate of 10 mV s−1 at 25 ◦ C. (a) pH = 2, (b) pH = 7 and (c) pH = 12.
atively high corrosion rate recorded with the Mg–5Al alloy. The -phase acts, generally, as an effective cathode which accelerates the corrosion of the adjacent ␣-phase [21,22]. The maximum galvanic current and hence the high rate of corrosion occur at a certain ratio of anodic area to cathodic area. As it can be expected for active metals, the rate of the corrosion of these metallic materials in acidic medium is relatively high compared to that in neutral or basic solutions; it is more than 10 times its value in basic solutions. The data presented in Fig. 3 show clearly that the three investigated materials have relatively low corrosion rates in basic solutions. This can be explained by the formation of a barrier layer of Mg(OH)2 , which
is insoluble in basic solutions [23]. In acidic solutions, the barrier layer is completely soluble and hence relatively high corrosion rates were recorded. In neutral solutions, the barrier magnesium hydroxide layer is partially soluble and so a decrease in the corrosion rate was recorded. 3.3. Effect of temperature The effect of temperature on the corrosion rates of Mg, Mg–5Al, Mg–10Al and Mg–15Al was investigated in stagnant, naturally aerated solutions of pH 2, 7 and 12 using the potentiodynamic
Table 2 Corrosion parameters of Mg and the Mg–Al alloys after 90 min of the electrode immersion in stagnant, naturally aerated aqueous solutions of pH 2, 7 and 12 at 25 ◦ C. Ecorr /mV
Rct / cm2
icorr /A cm−2
Ba /mV dec−1
Bc /mV dec−1
Corrosion rate/m Y−1
pH = 2 Mg Mg–5Al Mg–10Al Mg–15Al
−1457.6 −1728.4 −1458.2 −1537.5
87.5 79.8 84.9 118.5
229.8 265.6 166.2 134.7
124.9 135.9 109.4 101.1
−128.8 −136.6 −104.3 −98.5
5245 4089 3793 3073
pH = 7 Mg Mg–5Al Mg–10Al Mg–15Al
−1550.0 −1701.4 −1492.9 −1524.0
253.9 368.3 411.1 320.8
68.2 84.2 61.4 75.8
137.7 236.7 212.7 222.7
−86.4 −148.3 −112.3 −86.2
1555 1921 1401 1730
pH = 12 Mg Mg–5Al Mg–10Al Mg–15Al
−1426.2 −1520.5 −1294.2 −1406.4
5.09 15.3 1.72 1.83
195.9 9.70 52.4 80.7
−108.6 −128.0 −78.6 −52.7
116 348 39 41
4250 751 6520 8480
6654
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
Fig. 3. Corrosion rate for Mg and the Mg–Al alloys as a function of pH at 25 ◦ C.
polarization technique. In general, the corrosion rate which is represented by the corrosion current density, icorr , increases with the increase of temperature, and the process follows the familiar Arrhenius equation [24]. d log icorr Ea =− R d(1/T ) where Ea is the molar activation energy of the process and R is the gas constant (8.314 J mol−1 K−1 ). The relations between log icorr and 1/T for the investigated alloys are presented in Fig. 4. The activation energy, Ea , values of the corrosion processes in the different solutions were calculated and presented in Table 3. The low activation energy calculated in acidic solutions indicates the high corrosion tendency in this medium, which agrees well with both the polarization and impedance measurements. Generally, the activation energy values are less than 40 kJ mol−1 . This means that the rate determining step for the dissolution process is a one electron transfer process [25,26]. 3.4. The electrochemical impedance measurements EIS is essentially a steady state technique that is capable of accessing relaxation phenomena whose relaxation times vary over orders of magnitudes and permits single averaging within a single experiment to obtain high precision levels. The open-circuit impedance of Mg and its alloys was traced over 120 min from electrode immersion in the test solutions. Typical data for the Mg and Mg–Al alloys in aqueous solutions of pH 2, 7 and 12 are presented as Bode plots in Fig. 5. Bode plots are recommended as standard impedance plots, since the whole impedance data are equally represented and the phase angle, , as a sensitive parameter for indicating the presence of any additional time constants for interfacial phenomena appears explicitly [27–30]. To have more insight on the impedance data, the impedance vs. frequency data were plotted separately from the phase shift vs. frequency data. The two figures recorded in acidic solution are assigned as Fig. 5a and a* Table 3 Activation energy of Mg and the different Mg–Al alloys after 90 min of the electrode immersion in stagnant, naturally aerated aqueous solutions of pH 2, 7 and 12. Materials
Mg Mg–5Al Mg–10Al Mg–15Al
Ea /kJ mol−1 pH 2
pH 7
pH 12
8.33 6.25 10.00 10.68
9.55 7.13 10.39 11.27
10.41 8.17 11.81 11.80
Fig. 4. Arrhenius plots of Mg and the Mg–Al alloys in stagnant naturally aerated aqueous solutions. (a) pH = 2, (b) pH = 7 and (c) pH = 12.
for the impedance and phase, respectively. It is clear that in acidic solution continuous corrosion is taking place for all materials and relatively small total impedance, Z, at low frequency was recorded. The phase shift shows two clear maxima indicating the presence of two time constants controlling the corrosion process. In neutral and basic solution passivation phenomena could be recorded and the value of absolute impedance, Z, and the diameter of the phase maximum show that the corrosion mechanism is dependent on the corroding material. The rate of corrosion
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
6655
Fig. 5. Bode plots [a–c represent log Z vs. log f; and a*–c* represent log f vs. ] of Mg and Mg–Al alloys in stagnant naturally aerated aqueous solutions at 25 ◦ C. (a, a*) pH = 2, (b, b*) pH = 7 and (c, c*) pH = 12.
decreases with the increase of the amount of Al in the alloy. The Bode plots in neutral solution are presented as Fig. 5b and b* for the impedance and phase, respectively and those in basic solution are presented as Fig. 5c and c* in the same sequence. In all solutions relatively higher impedance values were recorded for Mg–10Al and Mg–15Al (cf. Fig. 5). Lower impedance values were recorded for Mg and Mg–5Al. The phase maximum is broader in the case of Mg–15Al and Mg–10Al compared to Mg and Mg–5Al, which means a higher corrosion resistance in case of both alloys. It means that the corrosion resistance of the alloy increases with the increase of the Al content. The impedance data were analyzed using the dispersion formula for a simple equivalent circuit model consisting of a parallel combination of a capacitor, Cdl , and a resistor, Rct , in series with a resistor, Rs , representing the solution resistance. The electrode impedance, Z, in this case is represented by the mathematical formulation:
Z = Rs +
Rct 1 + (2fRct Cdl )
˛
where ˛ denotes an empirical parameter (0 ≤ ˛ ≤ 1) and f is the frequency in Hz. The above relation takes into account the deviation from the ideal capacitor, RC, behavior in terms of a distribution of time constants due to surface inhomogeneties, roughness effect, and variations in properties or compositions of surface layers [31–33]. The impedance spectra obtained experimentally were analyzed using software provided with the electrochemical workstation. According to AC circuit theory, an impedance plot obtained for a given electrochemical system can be correlated to one or more equivalent circuits [34]. The impedance data for the Mg alloys in aqueous solutions of different pH were analyzed using the equivalent circuit shown in Fig. 6. In this model a combination Rf Cf was introduced to the simple Randle’s cell to account for the presence of surface film and a Warburg impedance, Zw , to account for the diffusion process recorded in the basic solutions. The calculated equivalent circuit parameters for Mg, Mg–5Al, Mg–10Al and Mg–15Al alloys are presented in Table 4. The corrosion rate is inversely proportional to Rct , where higher Rct values were obtained for Mg–15Al and Mg–10Al and lower values for Mg and Mg–5Al. In general, Cdl , values for the magnesium alloys is low and can be attributed to the formation of relatively thick and compact pro-
tective film on the metal surface [35,36]. In the present study, an increase in Rct is not accomplished by the reduction of Cdl , however, the Cdl values are increased with an increase in Rct values. Similar observation was recorded on La additions to AZ91 alloy [35]. It has been shown that these alloys cannot form effective protective film in aqueous solutions. The Bode phase plots of the investigated materials in acidic solutions (Fig. 5a*) show two phase maxima at low and high frequencies, which assign the presence of two time constants controlling the corrosion process. The low frequency loop was attributed to a charge transfer resistance and a double-layer capacitance at the electrode surface, while the high frequency loop was attributed to the presence of a partially formed surface film of Mg(OH)2 [37–39]. In neutral solutions, a suppressed phase maximum in the low frequency region was present and the high frequency region phase maximum dominates (cf. Fig. 5b*). This means that the corrosion rate of Mg and its alloys decreases by the increase of the solution pH, which can be attributed to the accumulation of corrosion products on the metallic surface. The
Fig. 6. Equivalent circuit models for fitting of the experimental impedance data of Mg and its alloys, Rs = solution resistance, Rct = charge transfer (corrosion) resistance, Cdl = electrode capacitance, Rf = film resistance, Cf = film capacitance and Zw is the Warburg impedance. (a) Model for fitting the data in neutral solutions. (b) Model for fitting the data in basic solutions, where Zw was introduced to account for the diffusion controlled process.
6656
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
Table 4 Equivalent circuit parameters of Mg and the different Mg–Al alloys after 90 min of the electrode immersion in stagnant, naturally aerated aqueous solutions of pH 2, 7 and 12 at 25 ◦ C. pH
Sample
Rs /
Rct /k cm2
Cdl /F cm−2
2
Mg Mg–5Al Mg–10Al Mg–15Al
73.2 62.2 81.3 84.1
2.219 2.670 7.94 11.83
35.9 10.2 6.3 7.61
7
Mg Mg–5Al Mg–10Al Mg–15Al
53.7 57.2 51.1 53.8
0.66 0.42 0.38 0.88
12
Mg Mg–5Al Mg–10Al Mg–15Al
13.7 24.9 10.1 5.1
2.58 0.53 2.37 5.59
9.98 5.94 6.76 8.67 12.3 9.5 16.8 35.9
Rf / cm2
9.90
Cf /F cm−2
40.3
333.0 356.6 389.5 408.6
23.9 11.2 81.7 61.5
385
75
corrosion products thus formed are more stable in neutral solutions but soluble in acidic solutions [23]. In basic solutions only one phase maximum is recorded (cf. Fig. 5c*). The disappearance of the second phase maximum and the relatively high impedance values are good indications for the presence of a stable protective layer on the metallic surface in this medium. The presence of a diffusion process in basic solutions was also recorded as linear part at low frequencies in the Nyquist diagram presented in Fig. 7. The effect of Al on the corrosion behavior of Mg can be clearly understood by comparing the potential–pH diagram of Mg and Mg–Al presented in Fig. 8a and b [40]. According to the presented diagrams, Mg corrodes in aqueous solutions through the reduction of water with the formation of Mg(OH)2 that is soluble in acidic solutions leading to the recorded high corrosion rates in these media. In basic solutions a stable Mg(OH)2 protective film is formed on the metallic surface which inhibits the corrosion process and lower corrosion rates are recorded. 3.5. Surface morphology At pH ≥ 4 Al2 O3 ·3H2 O is formed on the alloy surface leading to an inhibition of the corrosion process and stabilizing the Mg(OH)2 layer. In the pH range 8–12, Mg aluminate is formed and a blocking of the active centers by Mg(OH)2 takes place. The formation of microstructures due to the presence of the alloying elements i.e. Al, leads to the stabilization of the alloy and increases its corrosion resistance [10,41–43]. The role of Al in the formation of the
Fig. 7. Nyquist plot for Mg and Mg alloys in stagnant naturally aerated basic solution of pH 12 at 25 ◦ C.
Fig. 8. Potential–pH diagrams of Mg (a) and Mg–Al alloy (b).
stabilizing -phase is presented on the scanning electron micrographs of Figs. 9 and 10. In Fig. 9 the polished surfaces of Mg (Fig. 9a) and Mg–10Al (Fig. 9b) alloys are presented. It is clear from this figure that the magnesium surface consists of a smooth easily corroding magnesium matrix i.e. ␣-phase. The presence of Al as alloying element leads to the formation of large areas of the phase i.e. Mg17 Al12 intermetallic. The large area of the intermetallic in Mg–10Al alloy compared to the Mg–5Al leads to a decrease in the corrosion rate due to modification of the anodic to cathodic areas that are inhibited in low ratio Al alloys. The comparison between the corrosion behavior of the Mg itself and the Mg–10Al alloy is presented clearly in Fig. 10. The micrograph of Fig. 10a shows clearly that the Mg surface is subjected to severe corrosion even in neutral solutions. The Mg–10Al alloy under the same conditions shows only localized attack which leads to the recorded lower corrosion rate (cf. Fig. 10b). It is clear from the presented data that Mg is active and oxidized easily to the divalent cation. The oxidation process is taking place in two steps in accordance with a one electron transfer kinetics [44,45] which is consistent with the low activation energy of the corrosion process [18,26]. The first charge transfer step is the formation of the Mg+ ion which is readily oxidized to Mg2+ in the presence of hydrogen ions in acidic solutions or water in neutral solutions according the: Mg → Mg+ + e
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
6657
Fig. 9. Scanning electron micrographs of mechanically polished Mg (a) and Mg–10Al alloy (b).
Fig. 10. Scanning electron micrographs of Mg (a) and Mg–10Al alloy (b) after 90 min immersion in naturally aerated neutral solutions of pH 7 at 25 ◦ C.
References 2Mg+ + 2H+ → 2Mg2+ + H2 2Mg+ + 2H2 O → 2Mg2+ + 2OH− + H2 This leads to the formation of the corrosion product Mg(OH)2 , which is stable in neutral and basic solutions and leads to the pronounced increase of the corrosion resistance and impedance of the metallic surface in these solutions. The presence of Al leads to the formation of -Mg17 Al12 intermetallic, which is more stable than ␣-Mg matrix. This intermetallic acts as an affective barrier and as an active cathode for the ␣-matrix. As the ratio of Al increases the ratio of -matrix increases and a continuous oxide film is formed on its surface which inhibits the corrosion process [10,43]. 4. Conclusions Mg–10Al has reasonably high corrosion resistance compared to Mg or Mg–5Al. The increased corrosion resistance is due to the formation of large areas of the Mg17 Al12 intermetallic, which leads to the formation of microstructures that stabilizes the alloy and increases its corrosion resistance. Any increase in the Al content e.g. Mg–15Al had little effect on the corrosion characteristics. In neutral and basic solutions the surface Mg(OH)2 layer is stable and the formation of Al2 O3 ·3H2 O and also Mg aluminates stabilizes the alloy surface and hence low corrosion rates are recorded. The calculated activation energy of the corrosion process is 10 kJ mol−1 , which assigns a one electron charge transfer as a rate determining step. The dissolved Mg+ ions are readily transformed into the stable Mg2+ and the formation of Mg(OH)2 . This corrosion product is stable in neutral and basic solutions and completely soluble in acidic solutions.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]
Y. Zhang, C. Yan, F. Wang, et al., Surf. Coat. Technol. 161 (2002) 36. S.J. Kim, Y. Mizutani, R. Ichino, et al., Surf. Coat. Technol. 169/170 (2003) 143. Z. Zhang, L.J. Zhang, J.J. Fan, et al., Electrochim. Acta 52 (2007) 5325. N. Pebere, C. Riera, F. Dabosi, Electrochim. Acta 35 (1990) 555. G.L. Makar, J. Kruger, Int. Mater. Rev. 38 (1993) 138. J.H. Nordien, S. Ono, N. Masuko, K. Nisancioglu, J. Electrochem. Soc. 142 (1995) 3320. G. Song, A. Atrens, D. John, X. Wu, J. Nairn, Corros. Sci. 39 (1997) 1981. G. Song, A. Atrens, X. Wu, B. Zhang, Corros. Sci. 40 (1998) 1769. G. Song, A. Atrens, M. Dargusch, Corros. Sci. 41 (1999) 249. R. Ambat, N.N. Aung, W. Zhou, Corros. Sci. 42 (2000) 1433. G. Baril, N. Pebere, Corros. Sci. 43 (2001) 471. G. Baril, C. Blanc, N. Pebere, J. Electrochem. Soc. 148 (2001) 489. P. Schmutz, V. Guillaumin, R.S. Lillard, J.A. Lillard, G.S. Frankel, J. Electrochem. Soc. 150 (2003) 99. S. Mathieu, C. Rapin, J. Steinmetz, P. Steinmetz, Corros. Sci. 45 (2003) 2741. N. LeBozec, M. Jonsson, D. Thierry, Corrosion 60 (2004) 356. B. Luan, J.E. Gray, J. Alloys Compd. 336 (2002) 88. Z. Shi, G. Song, A. Atrens, Corros. Sci. 47 (2005) 2760. W.A. Badawy, F.M. Al-Kharafi, A.S. El-Azab, Corros. Sci. 41 (1999) 709. R.M. El-Sherif, K.M. Ismail, W.A. Badawy, Electrochim. Acta 49 (2004) 5139. O. Lunder, J.E. Lein, T.K. Aune, K. Nisancioglu, Corrosion 45 (1989) 741. G. Song, A. Atrens, Adv. Eng. Mater. 1 (1999) 33. M. Schutze, W. Cahn, P. Haasen, E.J. Kramer (Eds.), Material Science and Technology Series, Wiley-VCH, Weinheim, 2000, p. 131. W.S. Loose, in: L.M. Pidgeon, J.C. Mathes, N.E. Woldmen (Eds.), Corrosion and Protection of Magnesium, ASM Int., Materials Park, OH, 1946, p. 173. R.G. Martimer, Physical Chemistry, The Benjamin/Cummings Publishing Company, Inc, Redwood City, CA, 1983, p. 822. A. Wieckowski, E. Ghali, Electrochim. Acta 30 (1985) 1423. W.A. Badawy, F.M. Al-Kharafi, J.R. Al-Ajmi, J. Appl. Electrochem. 30 (2000) 693. W.A. Badawy, S.S. El-Egamy, K.M. Ismail, Br. Corros. J. 28 (1993) 133. A.J. Brock, G.C. Wood, Electrochim. Acta 12 (1967) 395. D.D. Macdonald, S. Real, S.I. Smedly, M. Uraquidi-Macdonald, J. Electrochem. Soc. 135 (1988) 2410. C.M.A. Brett, Corros. Sci. 33 (1992) 203. A.E. Bohe, J.R. Vilche, K. Juettner, W.J. Lorenz, W. Paatsch, Electrochim. Acta 34 (1989) 1443. K. Hladky, L.M. Calow, J.L. Dawson, Br. Corros. 15 (1980) 20. J. Hitzig, J. Titz, K. Juettner, W.J. Lorenz, E. Schmidt, Electrochim. Acta 29 (1984) 287.
6658
G.M. Abady et al. / Electrochimica Acta 55 (2010) 6651–6658
[34] J.R. Macdonald, Impedance Spectroscopy, John Wiley & Sons, New York, 1987. [35] Y. Fan, W. Guohua, G. Hongtao, L. Guanqun, Z. Chunquan, J. Mater. Sci. 41 (2006) 5409. [36] R. Udhayan, D.P. Bhatt, J. Power Sources 63 (1996) 103. [37] F. Rosalbino, E. Angelini, S. De Negri, A. Saccone, S. Delfino, Intermetallics 13 (2005) 55. [38] F. Rosalbino, E. Angelini, S. De Negri, A. Saccone, S. Delfino, Intermetallics 14 (2006) 1487. [39] Z. Xuehua, H. Yuanwei, W. Zhongling, C. Qiurong, G. Fuxing, Corros. Sci. 48 (2006) 4223.
[40] M. Pourbaix, Atlas of Electrochemical Equilibrium Diagram in Aqueous Solutions, NACE, Houston, TX, 1966. [41] S. Mathieu, C. Rapin, J. Steinmetz, P.A. Steinmetz, Corros. Sci. 45 (2003) 2741. [42] G. Song, A.L. Bowles, D.H. St. John, Mater. Sci. Eng. A 366 (2004) 74. [43] Y. Song, D. Shan, R. Chen, E.-H. Han, Corros. Sci. 51 (2009) 1087. [44] A.J. Bard, Encyclopedia of Electrochemistry of the Elements, vol. VIII, Marcel Dekker, New York, 1978. [45] W.A. Badawy, N.H. Helal, M.M. El-Rabiee, H. Nady, Electrochem. Acta 55 (2010) 1880.