Wear behavior and corrosion properties of Age-hardened AA2010 aluminum alloy

2.1. Sample Preparation and heat treatment process

Before the aging process, samples were cut and prepared for microstructural investigations, hardness, wear, and corrosion tests. The spectral analysis of AA2010 alloy is given in Table 1. AA2010 alloy was solution treated at 500 °C for 1 hour and aged at 160 °C for 16, 18, and 22 hours. Polished samples were etched with aqueous NaOH solution (10 g NaOH+50 ml distilled water) and examined under an optical microscope.

2.2. Hardness and Wear Tests

Macro hardness of the heat treated samples was measured by Vickers indenter under 2 kg-force loads. In addition, the micro Vickers hardness test was applied on the worn surface under 300 g-force loads.

Wear test of heat treated samples was carried out with CSM instruments (Fig. 1), ball-on-disc wear-test apparatus under 2 N loads at room temperature. Alumina ball with a diameter of 6 mm was used as a counterpart. Each test was performed with a constant sliding speed of 5 cm/s and the track radius was 2.5 mm. Wear loss was measured every 100 m distance with the aid of a Mitutoyo profilometer. Total sliding distance of the wear test was 500 m.

2.3. Corrosion test

Corrosion test was conducted by linear polarization resistance method using Metrohm DropSens Potentiostat (Fig. 2). The corrosion rate of heat treated AA2010 alloy was determined by Tafel curves. 30 mm² area of each sample was immersed in 100 ml 3.5 wt.% sodium chloride (NaCl) solution for corrosion test. The three-electrode corrosion cell system is composed of the AA2010 sample as a working electrode, Ag/AgCl electrode as a reference electrode and a graphite electrode.

Before starting the test, open circuit potential (OCP) was measured for 60 seconds. Then Tafel curves from a cathodic potential of -0.8 V to an anodic potential of 1 V were plotted. All measurements were carried out with the scanning rate of 0.001 V/s.

3.1. Corrosion and microstructure

The microstructure image of AA2010 alloy is given in Fig. 3a. Micro-scaled Al-Fe intermetallic particles were observed by SEM-EDX (Fig. 3b). According to Al-Fe phase diagram (Baker, 1992Baker, H. (1992). ASM Handbook Volume 3: Alloy Phase Diagrams. ASM International. ), the given composition indicates the formation of FeAl and/or Fe₃Al intermetallic particles. This prediction was proved by the XRD analysis. A peak of Fe-Al intermetallic phase was detected by XRD. In addition, the main precipitation hardening phases of 2XXX series aluminum alloy such as Al₂Cu and Al₂CuMg were observed (Fig. 4). The intensity of the peaks of Mg₂Si and MgZn₂ precipitation hardening phases was found to be less than Al₂Cu and Al₂CuMg. Hence, the determinant factor of corrosion in terms of microstructure is thought to be the change of size and distribution of Al₂Cu and Al₂CuMg as discussed in the following paragraphs.

Open-circuit potential (OCP) graphic indicates the tendency of dissolution of 22 h aged sample with a more negative OCP (approximately -0.76 V). 16 h and 18 h aged samples have approximately -0.71 V OCP, hence less tendency of corrosion is expected for these samples (Fig. 5).

Tafel curves of AA2010 alloy, aged for 16, 18 and 22 hours are given in Fig. 6. Jcorr (current density) and Ecorr (corrosion potential) are given in Table 2. Ecorr is almost same for all samples (Fig. 6 and Table 2). The corrosion rate directly depends on the current density (Jcorr) corresponding to the intersection point of the anodic and cathodic slopes of the Tafel curve (Bilgiç, 2018Bilgiç, S. (2018). Galvanic Corrosion. The Eurasia Proceedings of Science, Technology, Engineering and Mathematics. SRES Publishing, pp. 259-262.). Both the current density and corrosion rate of 16 h and 18 h aged samples are approximately equal to each other. However, increasing the aging time further increased the corrosion current and accordingly corrosion rate.

Optical microscope images of corroded surfaces clearly indicate the formation of intergranular corrosion and pitting corrosion (Fig. 7). The pitting corrosion occurs due to the galvanic coupling of the cathodic intermetallic particles and anodic aluminum matrix. Al-Fe intermetallic particles arise from the interaction between alloying elements and impurities during the casting process. They are not affected by solution heat treatment or aging (Andreatta et al., 2004Andreatta, F., Terryn, H., De Wit, J.H.W. (2004). Corrosion behaviour of different tempers of AA7075 aluminium alloy. Electrochim. Acta. 49 (17-18), 2851-2862. https://doi.org/10.1016/j.electacta.2004.01.046.). However, intermetallic particles have a significant effect on corrosion. They cause discontinuities in the surface film formed in aqueous solution and promote the local breakdown of the passive oxide film. Once the oxide layer is damaged, dissolution of the aluminum matrix accelerates (Chen et al., 1996Chen, G.S., Gao, M., Wei, R.P. (1996). Microconstituent-induced pitting corrosion in aluminum alloy 2024-T3. Corrosion 52 (1), 8-15. https://doi.org/10.5006/1.3292099. ; Mishra and Balasubramaniam, 2007Mishra, A.K., Balasubramaniam, R. (2007). Corrosion inhibition of aluminum alloy AA 2014 by rare earth chlorides. Corros. Sci. 49 (3), 1027-1044. https://doi.org/10.1016/j.corsci.2006.06.026. ; Krishna et al., 2017Krishna, K.G., Das, G., Venkateswarlu, K., Kumar, K.H. (2017). Studies on Aging and Corrosion Properties of Cryorolled Al-Zn-Mg-Cu (AA7075) Alloy. Trans. Indian Inst. Met. 70 (3), 817-825. https://doi.org/10.1007/s12666-017-1064-3. ). Micro scaled Al-Fe intermetallic particles (Fig. 3b) worked as a cathode resulting in anodic dissolution of aluminum matrix (Yasakau et al., 2007Yasakau, K.A., Zheludkevich, M.L., Lamaka, S.V., Ferreira, M.G.S. (2007). Role of intermetallic phases in localized corrosion of AA5083. Electrochim. Acta 52 (27), 7651-7659. https://doi.org/10.1016/j.electacta.2006.12.072.), hence they caused pitting corrosion of AA2010 samples independently of aging time (Fig. 7 and Fig. 8). Apart from the Al-Fe intermetallic particles, 2XXX series aluminum alloys include strengthening Al₂Cu, and Al₂CuMg precipitations (Wang and Starink, 2005Wang, S.C., Starink, M.J. (2005). Precipitates and intermetallic phases in precipitation hardening Al-Cu-Mg-(Li) based alloys. Int. Mater. Rev. 50 (4), 193-215. https://doi.org/10.1179/174328005X14357. ; Burt, 2015Burt, V. (2015). Corrosion in the Petrochemical Industry. ASM International.) as detected by XRD analysis in the present study. Here, Al2Cu phase takes place as cathode side in galvanic coupling of aluminum matrix and precipitate, hence accelerates the oxidation of aluminum and formation of corrosion pits (Burt, 2015Burt, V. (2015). Corrosion in the Petrochemical Industry. ASM International.; Vieira et al., 2011Vieira, A.C., Pinto, A.M., Rocha, L.A., Mischler, S. (2011). Effect of Al₂Cu precipitates size and mass transport on the polarisation behaviour of age-hardened Al-Si-Cu-Mg alloys in 0.05 M NaCl. Electrochim. Acta 56 (11), 3821-3828. https://doi.org/10.1016/j.electacta.2011.02.044. ). However, Al₂CuMg tends to precipitate along the grain boundary and generates a copper depleted zone. Both Al₂CuMg phase and copper depleted zone along the boundaries exhibit anodic characteristic in comparison with the surrounding matrix (Burt, 2015Burt, V. (2015). Corrosion in the Petrochemical Industry. ASM International.). Consequently, it was thought that intergranular corrosion was promoted in the overaged sample by the coarsened Al₂CuMg precipitations and generated copper depleted zone. In addition, anodic precipitates within the grain boundary such as Mg₂Si and MgZn₂ (Gharavi et al., 2015Gharavi, F., Matori, K.A., Yunus, R., Othman, N.K., Fadaeifard, F. (2015). Corrosion behavior of Al6061 alloy weldment produced by friction stir welding process. J. Mater. Res. Technol. 4 (3), 314-322. https://doi.org/10.1016/j.jmrt.2015.01.007. ; Liu et al., 2016Liu, Y., Mol, J.M.C., Janssen, G.C.A.M. (2016). Combined Corrosion and Wear of Aluminium Alloy 7075-T6. J. Bio. Tribo. Corros. 2 (2), 2-9. https://doi.org/10.1007/s40735-016-0042-3. ) that were detected by XRD promote the intergranular corrosion. The overaged sample which was aged for 22 h showed maximum corrosion rate owing to the total effect of pitting and intergranular corrosion. Both the pitting and intergranular corrosions are the dominant corrosion mechanisms for overaged sample, whereas 18 h aged sample mainly suffered from pitting corrosion (Fig. 7). Even so, pitting corrosion damaged the overaged sample more than 18 h aged one as it is seen in SEM images (Fig. 8).

In summary, 22 h aged sample exhibits the maximum corrosion rate of 0.056 mm/year as predicted from OCP value, whereas 18 h aged sample with the peak strength exhibits a minimum corrosion rate of 0.039 mm/year (Table 2). Because, reaching the peak strength by artificial aging significantly reduces the intergranular corrosion susceptibility. Artificial aging up to peak strength avoids the development of the copper depleted zone by eliminating the precipitation of coarse Al₂CuMg phase along the grain boundary. As a consequence, intergranular corrosion can be got under control (Burt, 2015Burt, V. (2015). Corrosion in the Petrochemical Industry. ASM International.; Krishna et al., 2017Krishna, K.G., Das, G., Venkateswarlu, K., Kumar, K.H. (2017). Studies on Aging and Corrosion Properties of Cryorolled Al-Zn-Mg-Cu (AA7075) Alloy. Trans. Indian Inst. Met. 70 (3), 817-825. https://doi.org/10.1007/s12666-017-1064-3. ).

3.2. Hardness and wear tests

The average macro hardness of the as-received AA2010 alloy is 85.27 HV. Macro hardness values of heat treated AA2010 alloy are given in Fig. 9. Peak hardness was obtained after 18 h aging process and was found to be 120 HV on the average. 22 h aging process caused over aging and the hardness decreased to 111 HV. As it was noticed in the study of Meyveci et al. (2012)Meyveci, A., Karacan, İ., Durmuş, H., Çaligülü, U. (2012). Artificial Neural Network (ANN) Approach to Hardness Prediction of Aged Aluminium 2024 and 6063 Alloys. Mater. Test. 54 (1), 36-40. https://doi.org/10.3139/120.110290., prolonged aging time resulted in a decrease in hardness of heat treatable aluminum alloys due to the over aging phenomenon. The artificial aging heat treatment increased the hardness of AA2010 alloy by 30% - 40%.

Volumetric material loss-sliding distance graphic of aged samples are given in Fig. 10. Meyveci et al. (2011)Meyveci, A., Karacan, İ., Fırat, E.H., Çaligülü, U., Durmuş, H. (2011). Study of wear of aged aluminum alloys 2XXX and 6XXX within the manova statistical analysis. Met. Sci. Heat. Treat. 53 (3), 173-175. https://doi.org/10.1007/s11041-011-9363-5. reported that the wear loss depends on aging temperature and aging time. Up to 200 meters, minimum wear loss was observed in 18 h aged sample that exhibits peak hardness. In further stages of the wear test, overaged sample showed minimum wear loss. This finding is associated with the strain hardening ability of the material. After the wear test, the effect of cold deformation was revealed by micro hardness test. The overaged sample with the minimum hardness value exhibited the maximum hardness on the worn surface after the wear test. This, can be explained by the cold deformation ability of overaged sample. Due to its relatively low hardness, overaged sample can be deformed easier under the cyclic load of wear test. Consequently, cold deformation of the overaged sample provides better wear resistance, unexpectedly. Lindroos et al. (2015)Lindroos, M., Valtonen, K., Kemppainen, A., Laukkanen, A., Holmberg, K., Kuokkala, V.T. (2015). Wear behavior and work hardening of high strength steels in high stress abrasion. Wear 322-323, 32-40. https://doi.org/10.1016/j.wear.2014.10.018. stated that, the wear rate can decrease due to the significant surface work hardening during the repeated loading of the wear test. Similarly, De Mello et al. (2017)De Mello, J.D.B., Labiapari, W.S., Ardila, M.A.N., Oliveira, S.A.G., Costa, H.L. (2017). Strain hardening: can it affect abrasion resistance?. Tribol. Lett. 65 (2), 67. https://doi.org/10.1007/s11249-017-0850-8. asserted that higher level of strain increases the internal energy of the material that promotes surface oxidation, and therefore, it decreases the wear rate. Because, the surface oxidation reduces the friction between the metal surface and the counterpart, if the wear test is carried out in air media without a lubricant (Buckley, 1981Buckley, D.H. (1981). Tribology Series. Elsevier.). Friction coefficient values measured during the wear test support the formation of surface oxidation in overaged sample more than the others. The average values of friction coefficient (µ) were obtained 0.68, 0.64, and 0.49 for 16 h., 18 h., and 22 h. aged samples, respectively. Owing to the reduced hardness of overaged sample, counterpart easily deformed surface and increased the concentration of dislocations. Strained metal with a high level of dislocation concentration is chemically more active on the surface (Buckley, 1981Buckley, D.H. (1981). Tribology Series. Elsevier.), hence it is prone to oxidation during the wear test and reduce the friction coefficient as observed in overaged sample. SEM-EDX analysis of wear debris of 22 h aged specimen consists of aluminum and oxygen. It proves the existence of oxidation of aluminum (Fig. 11). Al₂O₃ on worn surface not only arises from the oxidation of metal surface, but also the detached particles of alumina ball as a counterpart.

Continuous scratches were observed on the worn surface of 16 and 18 h aged specimens. Wear track shows a sharp boundary between the aluminum matrix and worn surface of specimens aged for 16 and 18 h. On the contrary, the overaged sample exhibits the remains of cold deformed material along the wear track boundary (Fig. 12).