Corrosion Science 139 (2018) 215–226
Contents lists available at ScienceDirect
Corrosion Science journal homepage: www.elsevier.com/locate/corsci
Correlation of intergranular corrosion behaviour with microstructure in AlCu-Li alloy ⁎
T
⁎
Jia-lei Huanga, Jin-feng Lia, , Dan-yang Liua, Rui-feng Zhangc, , Yong-lai Chenb, Xu-hu Zhangb, Peng-cheng Mab, Rajeev Kumar Guptad, Nick. Birbilisc a
School of Materials Science and Engineering, Central South University, Changsha 410083, China Aerospace Research Institute of Materials and Processing Technology, Beijing 100076, China c Department of Materials Science and Engineering, Monash University, Clayton, VIC, 3800, Australia d Department of Chemical and Biomolecular Engineering, Corrosion Engineering Program, The University of Akron, Akron, OH, 44325, USA b
A R T I C LE I N FO
A B S T R A C T
Keywords: Aluminium Intergranular corrosion (IGC) Open circuit potential (OCP) TEM
The IGC behaviour, OCP and microstructure of AA1460 (Al-3.12Cu-2.14Li-0.12Sc-0.12Zr, in wt.%), following HT1 (heat treatment without predeformation) and HT2 (heat treatment with predeformation) tempers were investigated. Both δ′ (Al3Li) and T1 (Al2CuLi) phases were precipitated within grains during heat treatment, while T1 phase also precipitated in the vicinity of grain boundaries. The evolution of inter- and intra-granular microstructures in AA1460 influenced the local and global electrochemical characteristics, which in turn influenced the temper dependent corrosion morphologies of AA1460. A correlation between OCP and corrosion mode was proposed, which may be used to compare the IGC sensitivity of AA1460 with different tempers.
1. Introduction Lithium (Li) containing aluminium (Al) alloys alloys possess low density, high specific strength, and low fatigue-crack growth rates. The combination of these properties make them suitable for aerospace applications [1,2]. Modern Li containing Al-alloys nominally contain copper (Cu), and such Al-Cu-Li alloys are age-hardenable. Therefore, it is important to understand and manipulate the heat treatment of Al-CuLi alloys to optimise both the mechanical properties and corrosion performance. Studying the effect of ageing on the corrosion of Al-alloys has attracted significant research attention [3–9]. Several heat treatment processes such as the overaged tempers T73, T74, T76, and retrogression and re-ageing (RRA) have been developed to improve the corrosion resistance of some Al alloys [10–13]. Yang et al. [14] compared the corrosion behaviour of Al–Zn–Mg–Cu alloys after various ageing treatments and reported an improvement in corrosion resistance by ageing treatments resulting in discontinuously distributed precipitates at grain boundaries. Wang et al. [15] observed a narrow precipitate free zone (PFZ), coarsened and discontinuous grain boundary precipitates (GBPs), and decreased SCC susceptibility in AlZn-Mg alloys after two-stage ageing. Correlations between heat treatment and sensitization in 5xxx series Al-alloy were studied by Zhang et al. [16–18], who proposed an empirical model for the evolution of
⁎
grain boundary precipitates to rationalize sensitization dependent intergranular corrosion (IGC). Intergranular corrosion behaviour of Al alloys has generally been reported to be dictated by the composition, distribution, nearest neighbor distance, and size of GBPs [19–24]. Ageing treatments have a significant influence on the corrosion behaviour of Al-Cu-Li alloys [25–28]. For an Al-Cu-Li alloy (AA2096) with T8-ageing at 160 °C, Connolly et al. [29,30] reported existence of two stress corrosion cracking (SCC) susceptibility “windows” in the severely under-aged and over-aged conditions. However, the peak-aged condition did not display SCC sensitivity. Li et al. [31] and Liu et al. [32] also revealed that the corrosion mode of Al-3.7Cu-1.2Li and Al2.7Cu-1.7Li alloys in IGC solution evolved with ageing time in the following order: General IGC, localised IGC, pitting corrosion with IGC, and that this evolution process was dependent on the ageing temperature. It was reported that with T8 ageing treatment at 155℃, the AA2050 (an Al-Cu-Li alloy) became progressively susceptible to intragranular corrosion in 0.7 M NaCl solution [33] (as opposed to IGC). This was posited to be a result of the formation and growth of both intergranular and intragranular T1 precipitates, which decreased the copper content in solid solution – apparently balancing the electrochemical behaviour of the grains and grain boundaries, Nevertheless, it was reported that peak-aged AA2050 was immune to IGC, whilst the over-aged condition had no significant influence on the IGC sensitivity [34]. Ott et al. attributed the IGC of AA2050 in the overaged condition
Corresponding authors. E-mail addresses:
[email protected] (J.-f. Li),
[email protected] (R.-f. Zhang).
https://doi.org/10.1016/j.corsci.2018.05.011 Received 9 April 2017; Received in revised form 9 May 2018; Accepted 9 May 2018 0010-938X/ © 2018 Elsevier Ltd. All rights reserved.
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
Table 1 Aging treatments applied to aluminium alloy AA1460 sheet. Aging treatments
Aging temperature
Aging time (h)
HT1
145℃ 160℃ 175℃ 130℃ (1st step) followed by 160℃ (2nd step)
0.5, 5,10, 18, 24, 34, 44, 68, 105 0.5, 5, 10, 14, 22, 29, 33, 39, 51, 96 0.5, 5, 13, 17, 25, 29, 37, 42, 96, 112 4, 10, 20 (at 130℃) 4, 8, 2, 20, 45, 76 (at 160℃ after at 130℃ for 20 h)
HT2
Fig. 1. Representative cross-sectional corrosion morphologies of aluminium alloy 1460 immersed for 6 h in.1 M NaCl + 0.1 M h2O2 solution representing: (a) general IGC - Type A (HT1-aged at 145 ℃ for 10 h), (b) local IGC - Type B (HT2-aged at 160 ℃ for 8 h following at 130℃ for 20 h) and three different types of pitting corrosion corresponding to (c) Type C (HT1-aged at 175 ℃ for 0.5 h), (d) Type D (HT1-aged at 175 ℃ for 112 h) and (e) Type E (HT2-aged at 160℃ for 76 h following at 130℃ for 20 h).
216
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
1.5 h in a salt bath followed by water quenching at room temperature. Some of the quenched samples were artificially aged at three different temperatures (designated here as the HT1 temper). A two-step artificial ageing (designated here as the HT2 temper) was also performed, where quenched samples were subjected to 4% plastic deformation through cold rolling before artificial aging. The ageing parameters of for HT1 and HT2 tempers are presented in Table 1.
to the presence of large S′ (Al2CuMg) phase precipitates, and attendant Cu segregation, at grain boundaries [35]. Generally, the corrosion behaviour of Al alloys is determined by the microstructure (precipitate type, size and solute segregation), which in turn can be reflected by measurable properties such as electrical conductivity and the open circuit potential (OCP) [36–39]. For a given electrolyte, the OCP depends on the unique characteristics of intermetallics (including constituent particles, dispersoids, and precipitates) [40,41] and the matrix (determined by solute content, and a lesser extent, grain size, dislocation density, etc.) [39,42,43]. It was reported that the higher SCC and IGC resistance in the Li-containing Al-alloys AA8090 and AA1441, in contrast to other Li-containing Al-alloys, may be attributed to lower dislocation densities and the enhanced precipitation of the phases including δ, T1 and S′ [44–46]. Previous studies have reported the IGC behaviour of Al-Li alloys aged for different times or at different temperatures through the OCP variation [31–33], but no extensive discussion on the OCP-IGC relationship for Al-Cu-Li alloys has been reported thus far. The present work reports the OCP difference that can originate from the various microstructural features caused by different ageing treatments and a correlation between IGC behaviour and the OCP.
2.2. Tensile testing According to ASTM-E8M, the gauge dimensions for tensile specimens were 30 mm in length and 8 mm in width. Tensile testing was carried out using an MTS 858 testing machine (MTS System Corporation, Eden Prairie, MN, USA) at a strain rate of 2 mm/min. Three tensile tests were conducted for each ageing condition. 2.3. IGC testing The IGC test of the sample with different tempers was carried out according to the 7998-2005 GB/T standard. The IGC solution was 1 M NaCl + 0.1 M H2O2. The samples were immersed in the IGC solution at 35 ± 2℃ for 6 h. Afterwards, the sectional surface of the corroded samples was observed by a Leica DMILM metallographic microscope after milling and polishing. The maximum IGC depth was measured. Three cross sectional surfaces were investigated for each sample.
2. Experimental 2.1. Materials The material studied in this work was AA1460 sheet with a thickness of 2 mm and a chemical composition of: Al- 2.14 Li - 3.12 Cu - 0.12 Sc - 0.12 Zr (wt.%). Solution treatment was performed at 530 ℃ for
2.4. Electrochemical testing The open circuit potentials (OCP) and potentiodynamic polarisation
Fig. 2. OCP and tensile strength variation as a function of ageing time, presented with corresponding cross-sectional micrographs typical of the corrosion morphologies of aluminium alloy 1460 with ageing at 145℃. The insert shows the corresponding potentiodynamic polarisation curves in 3.5% NaCl solution at near neutral pH. 217
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
where the IGC was centered in localized area. The difference between the two types of corrosion is that the general IGC has larger continuous corrosion area than local IGC. Despite the occurrence of IGC, three different types of pitting-like corrosion morphologies were observed with HT1 and HT2 ageing in AA1460. The first pitting corrosion type corresponds to Type C (presented in Fig. 1c), where several discontinuous pits with slight IGC around their rim were noticed. This type of corrosion occurs mostly in early stages of ageing. An example of second type of pitting (Type D) is shown in Fig. 1d. where pits have larger corrosion depth and corroded area beneath the surface. Significant amount of IGC fissures were observed on the pitting facets. The third pitting like corrosion (Type E, Fig. 1e) was mainly corrosion within grains, where tiny pits coalesce to form larger pits. Additionally, only a relatively small presence of IGC was observed from cross-sectional SEM analysis. The two pitting types (Type D and Type E) were observed to be more common in later stages of ageing. The corrosion mode of AA1460 was highly dependent on the ageing time, temperature and type (HT1 or HT2), which will be discussed in the following sections.
curves of the specimens with different tempers were analysed in neutral 3.5% NaCl solution using a CHI660B electrochemical workstation. OCP testing was carried out using a saturated calomel electrode (SCE) and recorded for 10 min. Potentiodynamic polarisation tests were performed in a three-electrode configuration electrochemical cell with a potential scan rate of 1 mV/s. A platinum plate served as the counter electrode. The test area of the sample was 0.785 cm2. For each sample, OCP and potentiodynamic polarization tests were carried out three times. 2.5. Microstructural observation The aged specimens were mechanically ground to a thickness of 70–80 μm firstly and then thinned by electropolishing using an MTP-1 twin-jet electropolisher with 30% HNO3 + 70% CH3OH at −40 ℃ to −30 ℃. All the microstructure observations were carried out by a Tecnai G220 Super Twin transmission electron microscope (TEM) operating at 200 kV. 3. Results and discussion 3.1. Representative corrosion modes of 1460 Al-Cu-Li alloy
3.2. Dependence of corrosion, OCP and microstructure on HT1 ageing temperature and time
Ageing treatment has a significant influence on the microstructure and corrosion performance of Al-Cu-Li alloys. Of the aging conditions studied in this work, the corrosion modes (corrosion morphologies) were specified into five types (Fig. 1). As shown in Fig. 1a, the IGC appeared on the nearly entire surface, and this is defined as general IGC (Type A). In contrast, the local IGC (Type B) is presented in Fig. 1b
3.2.1. Corrosion and OCP The corrosion modes (corrosion morphologies) and OCP of AA1460 were dependent on the HT1 ageing temperature and time. The influence of HT1 ageing time (ranging from 0.5 to 112 h) and temperature (145 ℃, 160 ℃ and 175 ℃) on OCP and tensile strength is shown in Figs. 2, 3, and 4. The inset shows the potentiodynamic polarisation
Fig. 3. OCP and tensile strength variation as a function of ageing time, presented with corresponding cross-sectional micrographs typical of the corrosion morphologies of aluminium alloy AA1460 with ageing at 160℃. The insert shows the corresponding potentiodynamic polarisation curves in 3.5% NaCl solution at near neutral pH. 218
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
Fig. 4. OCP and tensile strength variation as a function of ageing time, presented with corresponding cross-sectional micrographs typical of the corrosion morphologies of aluminium alloy AA1460 with ageing at 175℃. The insert shows the corresponding potentiodynamic polarisation curves in 3.5% NaCl solution at near neutral pH.
the less noble direction with ageing. After 112 h of ageing at 160 ℃, the OCP was lowered −0.726 VSCE. The evolution of OCP, potentiodynamic polarisation curves, tensile strength, and corrosion morphologies with ageing time after ageing at 175℃ are displayed in Fig. 4. The peak strength was achieved after ageing for ∼ 36 h (at 175℃). Accompanying the changes in strength, the corrosion mode also changed from Type B + (C/ D) and then Type D. The OCP decreased with ageing time, reaching −0.736 VSCE following 112 h of ageing. The corrosion mode, maximum IGC depth and OCP of the alloy aged for various times at 145 ℃, 160 ℃ and 175 ℃ are shown in Table 2. The key observations from the Table 2 can be summarised as following: (1) The decrease in OCP was faster at the initial ageing stage followed by sluggish decreases with further ageing. (2) The corrosion mode changes from Type C → Type A → Type B → Type D with increasing the ageing time. (3) The change in corrosion modes was accelerated with increasing HT1 ageing temperature.
curves in 3.5% NaCl solution with near neutral pH. The ageing stage defined in this paper can be determined by the tensile strength curves. Peak-ageing stage was not achieved at 145 ℃ of ageing temperature, which was due to low precipitation kinetics. Under-, peak- and overaged stages were clearly distinguished at 175 ℃ ageing temperature. Tensile strength and OCP evolution of AA1460 as a function of ageing time and selected representative corrosion morphologies of the alloy aged at 145 ℃ are shown in Fig. 2. The tensile strength increased with increasing the ageing time. The corrosion mode also changed with the aging time. The alloy showed pitting corrosion and slight local IGC (corresponding to Type C) after ageing for 0.5 h. It can be seen from the Fig. 2 that the corrosion mode turned to Type A after 5 h of ageing treatment. After ageing for 5–68 h, the main corrosion mode is general IGC (Type A). The corrosion mode is denoted as Type A + (C). In this paper, non-bracketed letters represent the dominating corrosion types and bracketed letters represent the minor corrosion modes. When the ageing time was further extended to 116 h, the dominating corrosion mode was changed to local IGC (corresponding to Type B). The OCP dropped from −0.592 VSCE (solutionized state) to −0.645 VSCE within 5 h of ageing. Further ageing caused a slow decrease in OCP and it decreased to −0.699 VSCE after 105 h of ageing. In the case of ageing at 160℃, the tensile strength and OCP variation of AA1460 as a function of ageing time, and selected representative corrosion morphologies, are shown in Fig. 3. Owing to a higher ageing temperature, the development of strength was accelerated, however the over-aged condition was not reached following 105 h of ageing. It was observed that the corresponding corrosion modes evolved sequentially from Type C to Type A + (C), Type A, and then Type B + (D) as ageing time increased. Similar to specimens aged at 145 ℃, the OCP evolved to
3.2.2. Microstructural evolution with aging time and temperature The selected area electron diffraction (SAED) patterns and intragranular TEM images of AA1460 with HT1 tempers from different aging temperatures (145 ℃ to 175 ℃) are shown in Fig. 5. The SAED pattern along [001]Al for AA1460 aged at 145 ℃ for 0.5 h indicated the precipitation of δ′ phase, with a large number of fine δ′ precipitates observed in the TEM dark field (DF) image along < 100 > Al direction (Fig. 5a). For ageing at 160℃ for 29 h, a large number of δ′ phase (Fig. 5b) and a low population of precipitated T1 phase was found within the grains (Fig. 5c). It is noted the δ′ phase formed at 160℃ had a significantly larger size than that formed during 219
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
Table 2 Corrosion mode, maximum IGC depth and OCP of aluminium alloy AA1460 with different tempers. Temper
Aging time
Corrosion mode
Maximum IGC depth/μm OCP/VSCE
HT1at 145℃
0.5 h 5h 10 h 18 h 24 h 34 h 44 h 68 h 105 h
C A + (C) A + (C) A + (C) A + (C) A + (C) A + (C) A B + (C)
None −0.618 98.09 −0.645 107.81 −0.650 111.56 −0.655 96.23 −0.657 112.38 −0.660 123.98 −0.676 127.73 −0.688 77.81 −0.699
HT1 at 160℃
0.5 h 5h 10 h 14 h 22 h 29 h 33 h 39 h 64 h 112 h
C A + (C) A A A A B + (D) B + (D) B + (D) B + (D)
None −0.643 146.25 −0.655 151.76 −0.657 152.58 −0.661 139.81 −0.670 138.63 −0.677 96.78 −0.688 83.44 −0.696 66.74 −0.705 None −0.726
HT1at 175℃
0.5 h 5h 13 h 17 h 25 h 29 h 37 h 42 h 96 h 112 h
C B+ B+ B+ B+ B+ B+ B+ D D
None −0.658 88.01 −0.665 110.75 −0.677 140.05 −0.681 111.04 −0.686 72.19 −0.690 62.70 −0.708 None −0.711 None −0.718 None −0.736
130℃/ 4 h 130℃/10 h 130℃/20 h 130℃/20 h + 160℃/4 h 130℃/20 h + 160℃/8 h 130℃/20 h + 160℃/12 h 130℃/20 h + 160℃/20 h 130℃/20 h + 160℃/45 h 130℃/20 h + 160℃/76 h
A A A+C D + (B) D + (B) E + (D) E + (D) E + (D) E + (D)
HT2
(C) (C) (C) (D) (D) (D) (D)
94.04 −0.641 126.86 −0.652 111.02 −0.661 51.56 −0.703 54.14 −0.711 None −0.727 None −0.739 None −0.746 None −0.762
A, B, C, D represent different corrosion types (as shown in Fig. 1) respectively. Non-bracketed letters represent the dominating corrosion types and bracketed letters represent the minor corrosion modes.
selected corrosion morphologies of AA1460 with HT2 ageing are displayed in Fig. 7. The alloy showed general IGC (corresponding to Type A) after the 1st step ageing at 130℃. The OCP decreased to −0.66 VSCE after the 1st step ageing time up to 20 h. After the 2nd step ageing at 160℃ for 4–8 h (total ageing time of 24–28 h), the corrosion mode changed to Type D + (B). The yield strength increased with the progress of ageing and attained a plateau after ageing treatment for 28 h and the corrosion morphology transformed to pitting and intragranular corrosion (Type E + (D)) where no IGC was found. Meanwhile, an obvious drop in OCP to about −0.73 VSCE was observed during the initial 2nd step ageing time of 12 h, then it decreased to about −0.76 VSCE with a much slower rate with further 2nd step aging treatment for 64 h (total ageing time of 96 h). The corrosion mode, OCP as well as the IGC depth of the alloy with HT2 ageing are listed in Table 2. It was found that the HT2 ageing lead to a more negative OCP compared to HT1 ageing, and the corrosion morphologies swiftly changed into pitting corrosion (Type E + (D)) from general IGC (Type A).
ageing at 145℃. A large number of intragranular δ′ phase precipitates were formed within grains after ageing at a higher temperature of 175 ℃ (Fig. 5d). The size, distribution, and number of T1 precipitates was altered at 175℃. A considerable population of T1 phase was observed within grain interiors, and also at sub-grain boundaries (Fig. 5e). The population density of intragranular δ′ precipitates decreased with little increase in precipitate size (Fig. 5f), when the ageing time was extended to 96 h. Furthermore, T1 precipitates were found both at sub-grain boundaries (Fig. 5g) and within sub-grain interiors (Fig. 5h). The dark field (DF) TEM images of regions around grain boundaries in the alloys with the above mentioned ageing conditions are shown in Fig. 6. It was difficult to find precipitates at the GBs when the alloy was aged at 145℃ for 0.5 h. In the alloy aged at 160℃ for 29 h, T1 precipitates were mainly formed at the GBs (Fig. 6a). It was shown that T1 precipitates were continuously distributed along most of the grain boundaries. Ageing at higher temperature and for longer time (96 h at 175℃) caused grain boundaries with either no T1 precipitates or smaller amount of it, while massive precipitation at sub-grain boundaries and grain interiors (Fig. 6b,c).
3.3.2. Microstructural evolution in HT2 ageing The SAED patterns and TEM images of AA1460 with HT2 ageing for different times are shown in Fig. 8. After HT2 ageing at 130 ℃ for 4 h, a large number of fine δ′ phase were precipitated within grain (Fig. 8a) and a few precipitates were found at grain boundaries. After the 2nd step ageing at 160℃ for 12 h (32 h aging treatment in
3.3. Effect of HT2 ageing on corrosion, OCP evolution and microstructure 3.3.1. Corrosion and OCP evolution The OCP and strength evolution as a function of ageing time and 220
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
Fig. 5. SAED patterns and intragranular DF TEM images of aluminium alloy AA1460 aged at (a) 145℃ for 0.5 h, (b,c) 160℃ for 29 h, (d,e) 175℃ for 29 h and (f,g,h) 175℃ for 96 h. (a,b,d,f) show δ′ precipitates, with the beam direction parallel to < 001 > Al; (c,e,g,h) show T1 precipitates, with the beam direction is parallel to < 112 > Al.
total), the fraction of δ′ precipitates was increased (Fig. 8b) and T1 phase were precipitated within grain (Fig. 8c). It is important to notice that T1 precipitates at grain boundaries had a significantly larger interspacing (Fig. 8d) than that in the alloy treated with HT1 ageing at 160℃ for 29 h (Fig. 6a). With further aging (2nd step aging for 72 h), the intragranular δ′ precipitates coarsened and many θ ′′ (Al2Cu) precipitates formed which were flanked by a pair of lenticular δ′ precipitates (Fig. 8e). Moreover, number and size of T1 precipitates within the grains increased (Fig. 8f) and the fraction of GB T1 precipitates was further decreased (Fig. 8g). Compared to HT1 ageing treatment at 160 ℃, the precipitation of T1 phase within the grains of HT2-aged alloy was accelerated, which could be confirmed through the comparison between Figs. 5c and 8d. In addition, their fraction at the grain boundaries was decreased and the spacing was increased (Figs. 6a, 8e). The observed distribution of T1 precipitates within grain and at grain boundaries after HT1 and HT2
ageing was attributed to the pre-deformation prior to ageing treatments. To overcome the nucleation energy barrier during ageing process, semi-coherent T1 precipitates prefer to nucleate at the sites of dislocation, sub-grain boundaries, grain boundaries and stacking faults [47]. Consequently, T1 precipitates were principally found at sub-grain boundaries and grain boundaries after HT1 ageing in AA1460 (Figs. 5e,g and 6) which did not undergo any prior cold rolling. It should be noted that cold rolling could introduce defects such as dislocations into the matrix which will increase the nucleation sites for T1 precipitates [48,49]. Consequently, the precipitation of T1 phase was accelerated and their population density within grains was increased by HT2 ageing as reported in this paper. Additionally, increased number density of nucleation sites created overlapping diffusion fields as T1 precipitates grew, which enhanced the ageing kinetics [49]. Furthermore, the increase in the volume fraction of T1 precipitates within grains combined
221
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
Fig. 6. DF TEM images of grain boundaries in aluminium alloy AA1460 aged at (a) 160℃ for 29 h, (b) 175℃ for 29 h and (c) 175℃ for 96 h. The beam direction is parallel to < 112 > Al.
IGC, local IGC and pitting, respectively. Therefore, the corresponding corrosion mode could be determined when the OCP evolved into a certain zone with ageing time extension. Throughout the ageing process, the higher ageing temperature resulted in more negative OCP and the corrosion mode changed rapidly from IGC to pitting (grain corrosion). Overall, the ageing time interval during general IGC and local IGC occurrence was narrower. Moreover, the above phenomenon was more obvious for HT2 ageing treatment. It is of interest that the OCP values are related to IGC sensitivity (IGC area). The more negative OCP values resulted in lower sensitivity. It should be emphasised that this relation is mainly valid within the ageing range from under-aged stage to over-aged stage and may only be suitable for comparing the IGC sensitivity of a certain Al-Li alloy with different tempers. By comparing the difference between OCP and microstructures of the alloy with various tempers (Table 2), it was found that more δ′ and T1 precipitates lead to a greater decrease in the open circuit potential. δ′ precipitation leads to the OCP shift to a negative direction at initial ageing stage. Further ageing (at higher temperature or for a longer time) leads to the decrease in T1 precipitation and the OCP with a greater ageing time extension. This OCP decrease with increase in ageing time was mainly attributed to the more negative potentials of δ′ and T1 precipitates comparing with the matrix. This result echoes with the previous work from Li et al [41,51] who reported that the corrosion
with the enhanced ageing kinetics had been shown to lower the volume fraction of GB T1 precipitates due to the introduction of heterogeneous matrix nucleation sites and the larger change in volume-free energy (ΔGv) and the accommodation of the {1 1 1} shear strain for T1 help for its preferential nucleation on dislocations [50]. 3.4. Correlation of corrosion mode with OCP and ageing process For all different tempers (HT1 at different temperature and HT2), the corrosion mode of AA1460 followed a similar trend with prolonged ageing time: pitting corrosion with slight local IGC → general IGC → local IGC with pitting → pitting corrosion characterized with grain corrosion. As the ageing progressed, the IGC depth declined after an initial ascent. The OCP, closely related to the corrosion mode, decreased with increasing ageing. By consideration of the corrosion mode and OCP evolution shown in Figs. 4, 5 and Table 2, a corrosion diagram for describing the correlation of corrosion mode with OCP and ageing processes was plotted in Fig. 9. The OCP decreased with increasing the ageing time at the three temperatures tested. According to the OCP values, the plot was divided into four regions, which were denoted as Zone I (−0.59 to −0.64 VSCE), Zone II (−0.64 to −0.68 VSCE), Zone III (−0.68 to −0.71VSCE) and Zone IV (−0.72 to −0.77VSCE), respectively. These four regions were related to four different corrosion types, pitting with IGC, general
222
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
Fig. 7. OCP and tensile strength evolution as a function of total HT2 ageing time, presented with selected corresponding corrosion morphologies of aluminium alloy 1460. The insert shows the potentio-dynamic polarisation curves in 3.5% NaCl solution at near neutral pH. The time at horizontal ordinate is the total ageing time at 1st step of 130℃ and 2nd step of 160℃.
boundary precipitates were continuously distributed, which can provide a continuous channel for IGC. These two factors determine that the alloy is highly susceptible to IGC. Further ageing will lead to the precipitation of T1 within grains, which is accompanied with a further lowering of OCP. The potential difference is therefore lowered to a certain level for which IGC does not occur. Meanwhile, this process also causes the coarsening and increase in spacing between the precipitates, leading to discontinuous distribution of the grain boundary T1 precipitates, which cannot provide a continuous channel for IGC. As a result of these two factors, potential difference and precipitate distribution, the IGC sensitivity is lowered. In addition, the precipitated T1 phase within grains accelerate the corrosion of the grain matrix [41,51]. Conversely, the main corrosion mode is transferred to pitting featured with grain corrosion corresponding to type D and type E. Increase in ageing temperature, owing to higher diffusion coefficients of Cu and Li atoms, accelerates ageing kinetics resulting in precipitation of T1 within grains, whilst simultaneously GB T1 precipitates become more discontinuous. Accordingly, the corrosion mode evolution becomes faster with ageing time. Pre-deformation associated with HT2 ageing accelerates the nucleation of T1 precipitates in the grain interior, also increasing their volume fraction. A greater decrease in OCP and a smaller potential difference between grain interior and grain boundary precipitates are achieved. This is accompanied with a decrease in the fraction of T1 on
potential of T1 precipitate in NaCl solution was a few millivolts (about 221 mV) lower than that of the matrix. Although the potential of δ′ precipitate was not reported, it is expected to be more negative than that of the alloy matrix due to its higher content of active Li element. Therefore, the precipitation of δ′ phase should lower the overall potential of the alloy upon initial ageing, and the precipitation of T1 phase accompanied with the reduction of Cu content in the matrix leads to a further decrease in the overall potential of the alloy during the later stages of ageing. This OCP lowering associated with δ′ and T1 precipitation has also been observed in AA8090, AA1441, AA2050, AA2099 and other Al-Cu-Li-X alloys [31–33,45,46]. Therefore, the alloy with higher temperature ageing or HT2 ageing treatment possesses a more negative OCP due to the increased volume fraction of T1 precipitates within the grains during the same ageing time. The OCP of the alloy mainly reflects the potential of grains as the area fraction of grain boundaries is not significant [52]. Additionally, the OCP of T1 is more negative than the potential of the grain interior [41,51,53]. According to results obtained herein for AA1460 specifically, the evolution of corrosion mode with OCP can be explained as follows: for HT1 temper, early ageing stage (160 ℃ for 0.5 h) had relatively more positive OCP value (about −0.59 to −0.64 VSCE) and a small of amount of GB precipitates. Therefore the alloy was not susceptible to IGC. With further ageing, the OCP was lowered to about −0.68 VSCE and the potential difference was still appreciable. Concomitantly, the grain
223
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
Fig. 8. SAED patterns and TEM images of aluminium alloy 1460 with HT2 aging (a) at 130℃ for 4 h, (b,c,d) at 160℃ for 12 h and (e,f,g) at 160℃ for 76 h following at 130℃ for 20 h. (a,b,e) DF TEM images showing intragranular δ′ precipitates, imaged with the beam direction parallel to < 001 > Al; (c,f) DF TEM images showing intragranular T1 precipitates, imaged with the beam direction is parallel to < 112 > Al; (d) BF TEM image showing grain boundary T1 precipitates; (g) DF image showing grain boundary T1 precipitates, the direction is parallel to < 112 > Al.
1 The temper dependent corrosion mode for AA1460 evolved in the following order, with ageing time: Pitting corrosion with minor localised IGC, General IGC, Local IGC and pitting (characterised by corrosion of the grain interior). In addition, both a higher HT1 ageing temperature and the HT2 ageing condition accelerated the aforementioned corrosion mode evolution. 2 Concomitant with the precipitation of δ′ (Al3Li) and T1 phase (Al2CuLi), the OCP of AA1460 shifted towards the less noble (negative) direction. This was associated with enhanced anodic kinetics from potentiodynamic polarisation testing. Higher HT1 ageing temperatures, or the HT2 ageing treatment, enhanced the precipitation of δ′ and T1. 3 A phenomenological IGC corrosion diagram relating OCP evolution to corrosion modes was presented, where the various OCP ranges were correlated to the corrosion mode revealed. A first order
GBs. These factors contribute to the characteristics of the corrosion mode evolution with HT2 ageing time (Fig. 5). According to the results herein, the prevalence of IGC was dependent on the microstructure, whilts the OCP evolution reflected the microstructural variation within grains (which was also accompanied by microstructural variation of GBs). The corrosion mode of a specific Al-Cu-Li alloy in IGC solution could therefore can be related to the OCP values. However, this assertion warrants further investigations in regards to its broader validity amongst Al-Cu-Li alloys. 4. Conclusions The corrosion behaviour, microstructure and tensile strength of AA1460 with HT1 (145 ℃ to 175 ℃) and HT2 tempers was investigated. Based on the study herein, the following observations were made.
224
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
[15] Y.L. Wang, H.C. Jiang, Z.M. Li, D.S. Yan, D. Zhang, L.J. Rong, Two-stage double peaks ageing and its effect on stress corrosion cracking susceptibility of Al-Zn-Mg alloy, J. Mater. Sci. Technol. (2017). [16] R. Zhang, M.A. Steiner, S.R. Agnew, S.K. Kairy, C.H.J. Davies, N. Birbilis, Experiment-based modelling of grain boundary β-phase (Mg2Al3) evolution during sensitisation of aluminium alloy AA5083, Sci. Rep. 7 (2017). [17] R. Zhang, S.P. Knight, R.L. Holtz, R. Goswami, C.H.J. Davies, N. Birbilis, A survey of sensitization in 5xxx series aluminum alloys, Corrosion 72 (2016) 144–159. [18] R. Zhang, Y. Zhang, Y. Yan, S. Thomas, C.H.J. Davies, N. Birbilis, A materials the effect of reversion heat treatment on the degree of sensitisation for aluminium alloy AA5083, Corros. Sci. 126 (2017) 324–333. [19] N. Ott, Y.M. Yan, S. Ramamurthy, S. Kairy, N. Birbilis, Auger electron spectroscopy analysis of grain boundary microchemistry in an Al–Cu–Li alloy, Scr. Mater. 119 (2016) 17–20. [20] R.G. Buchheit, J.P. Moran, G.E. Stoner, Electrochemical behavior of the T1(Al2CuLi) intermetallic compound and its role in localized corrosion of Al–2%Li–3% Cu alloys, Corrosion 50 (1994) 120–130. [21] R.F. Zhang, R.K. Gupta, C.H.J. Davies, A.M. Hodge, M. Tort, K. Xia, N. Birbilis, The influence of grain size and grain orientation on sensitization in AA5083, Corrosion 72 (2015) 160–168. [22] K.D. Ralston, N. Birbilis, C.H.J. Davies, Revealing the relationship between grain size and corrosion rate of metals, Scri. Mater. 63 (2010) 1201–1204. [23] B. Decreus, A. Deschamps, P. Donnadieu, J.C. Ehrström, On the role of microstructure in governing fracture behaviorof an aluminum–copper–lithium alloy, Metall. Mater. Trans. A 586 (2013) 418–427. [24] B. Decreus, A. Deschamps, F.D. Geuser, P. Donnadieu, C. Sigli, M. Weyland, The influence of Cu/Li ratio on precipitation in Al–Cu–Li–x alloys, Acta. Mater. 61 (2013) 2207–2218. [25] B.J. Connolly, J.R. Scully, Corrosion cracking susceptibility in Al-Li-Cu alloys 2090 and 2096 as a function of isothermal aging time, Scr. Mater. 42 (2000) 1039–1045. [26] J.G. Brunner, N. Birbilis, K.D. Ralston, S. Virtanen, Impact of ultrafine-grained microstructure on the corrosion of aluminium alloy AA2024, Corros. Sci. 57 (2012) 209–214. [27] R.G. Buchheit, Local dissolution phenomena associated with S phase (Al [sub2] CuMg) particles in aluminum alloy 2024-T3, Electrochem. Soc. 144 (1997) 2621–2628. [28] N.L. Sukiman, X. Zhou, N. Birbilis, A.E. Hughes, J.M.C. Mol, S.J. Garcia, G.E. Thompson, Durability and corrosion of aluminium and its alloys: overview, property space techniques and developments, Alum. Alloy New Trends Fabr. Appl. 2012 (2016) 47–97. [29] B.J. Connolly, J.R. Scully, Corrosion cracking susceptibility in Al-Li-Cu alloys 2090 and 2096 as a function of isothermal ageing time, Scr. Mater. 42 (2000) 1039–1045. [30] B.J. Connolly, The transition from localized corrosion to SCC of Al-Li-Cu alloy AA2096 as a function of isothermal ageing heat treatment at 160℃ ageing, PhD Dissertation, University of Virginia, 2002. [31] J.F. Li, N. Birbilis, D.Y. Liu, Y.L. Chen, X.H. Zhang, C. Cai, Intergranular corrosion of Zn-free and Zn micro-alloyed Al-xCu-yLi alloys, Corros. Sci. 105 (2016) 44–57. [32] Q. Liu, R.H. Zhu, D.Y. Liu, Y. Xu, J.F. Li, Y.L. Chen, X.H. Zhang, Z.Q. Zheng, Correlation between artificial ageing and intergranular corrosion sensitivity of a new Al-Cu-Li alloy sheet, Mater. Corros. 68 (2017) 65–76. [33] V. Proton, J. Alexis, E. Andrieu, J. Delfosse, A. Deschamps, F. De Geuser, M.C. Lafont, C. Blanc, The influence of artificial ageing on the corrosion behaviour of a 2050 aluminium-copper-lithium alloy, Corros. Sci. 80 (2014) 494–502. [34] C. Henon, S. Rouault, Comparison of corrosion performance and mechanisms of AlCu alloys with and without Li addition, ICAA13 (2012) 431–436. [35] N. Ott, S.K. Kairy, Y. Yan, N. Birbilis, Evolution of grain boundary precipitates in an Al-Cu-Li alloy during ageing, Metall. Mater. Trans. A 48 (2017) 51–56. [36] X.L. Zhang, L. Zhang, G.H. Wu, W.C. Liu, C.C. Shi, J.S. Tao, J.W. Sun, Microstructural evolution and mechanical properties of cast Al-2Li-2Cu-0.5Mg0.2Zr alloy during heat treatment, Mater. Charact. 132 (2017) 312–319. [37] T.C. Tsai, T.H. Chuang, Atmospheric stress corrosion cracking of a superplastic 7475 aluminum alloy, Metall. Mater. Trans. A 27 (1996) 2617–2627. [38] T.C. Tsai, T.H. Chuang, Relationship between electrical conductivity and stress corrosion susceptibility of 7075 and 7475 aluminum alloys, Corrosion 52 (1996) 414–416. [39] Z.X. Wang, P. Chen, H. Li, B.J. Fang, R.G. Song, Z.Q. Zheng, The intergranular corrosion susceptibility of 2024 Al alloy during re-ageing after solution treating and cold-rolling, Corros. Sci. 114 (2016) 156–168. [40] B. Davó, J.J. de Damborenea, Use of rare earth salts as electrochemical corrosion inhibitors for an Al-Li-Cu (8090) alloy in 3.56% NaCl, Electrochem. Acta 49 (2004) 4957–4965. [41] J.F. Li, C.X. Li, Z.W. Peng, W.J. Chen, Z.Q. Zheng, Corrosion mechanism associated with T1 and T2 precipitates of Al-Cu-Li alloys in NaCl solution, J. Alloys Compd. 460 (2008) 688–693. [42] J.G. Brunner, N. Birbilis, K.D. Ralston, S. Virtanen, Impact of ultrafine-grained microstructure on the corrosion of aluminium alloy AA2024, Corros. Sci. 57 (2012) 209–214. [43] Q. Meng, G.S. Frankel, Effect of Cu content on corrosion behavior of 7xxx series aluminum alloys, Corrosion (Houston Tx) 60 (2004) 1403–1411. [44] K.S. Ghosh, S. Mukhopadhyay, B. Konar, B. Mishra, Study of ageing and electrochemical behaviour of Al-Li-Cu-Mg alloys, Mater. Corros. 64 (2013) 890–901. [45] K.S. Ghosh, K. Das, U.K. Chatterjee, Correlation of stress corrosion cracking behaviour with electrical conductivity and open circuit potential in Al-Li-Cu-Mg-Zr alloys, Mater. Corros. 58 (2007) 181–188. [46] K.S. Ghosh, K. Das, U.K. Chatterjee, Electrochemical behaviour of retrogressed and reaged (RRA) 8090 and 1441 Al-Li-Cu-Mg-Zr alloys, J. Appl. Electrochem. 36
Fig. 9. Phenomenological corrosion diagram relating OCP evolution to observed corrosion mode. The potential intervals in the diagram which correspond to different corrosion modes were estimated from Table 2.
correlation between ageing time, corrosion mode, and OCP was evident. IGC sensitivity of AA1460 decreased as the OCP was shifted in less noble direction. Acknowledgements The author would like to thank the ARIMPT (Aerospace Research Institute of Materials and Processing Technology) funded this project. Particular thanks are owed to Shichen Li for his assistance in operating the TEM microscope and providing continued encouragement and support. Finally, Thanks are also given to the technical staff at the Central South University who assisted in the manufacture of test pieces and in the use of laboratory equipment. NB is supported by Woodside Energy. References [1] T. Dursun, C. Soutis, Recent developments in advanced aircraft aluminium alloys, Mater. Des. 56 (2014) 862–871. [2] R.J. Rioja, J. Liu, The evolution of Al-Li base products for aerospace and space applications, Metall. Mater. Trans. A 43 (2012) 3325–3337. [3] R.F. Zhang, Y. Qiu, Y.S. Qi, N. Birbilis, A closer inspection of a grain boundary immune to intergranular corrosion in a sensitised Al-Mg alloy, Corros. Sci. (2018). [4] O. Gharbi, N. Birbilis, K. Ogle, Li reactivity during the surface pretreatment of Al-Li alloy AA2050-T3, Electrochim. Acta 243 (2017) 207–219. [5] N.D. Alexopoulos, Z. Velonaki, C.I. Stergiou, S.K. Kourkoulis, The effect of artificial ageing heat treatments on the corrosion-induced hydrogen embrittlement of 2024 (Al–Cu) aluminium alloy, Corros. Sci. 102 (2016) 413–424. [6] G.M. Scamans, N. Birbilis, R.G. Buchheit, Corrosion of aluminum and its alloys, in: Tony Richardson (Ed.), Shreir’s Corrosion, 2010, pp. 1975–2008. [7] U. Donatus, M. Terada, C.R. Ospina, F.M. Queiroz, A.F.S. Bugarin, I. Costa, On the AA2198-T851 alloy microstructure and its correlation with localized corrosion behaviour, Corros. Sci. 131 (2018) 300–309. [8] V. Proton, J. Alexis, E. Andrieu, J. Delfosse, M.C. Lafont, C. Blanc, Characterisation and understanding of the corrosion behaviour of the nugget in a 2050 aluminium alloy friction stir welding joint, Corros. Sci. 73 (2013) 130. [9] J. Proton, E. Alexis, C. Andrieu, J. Blanc, L. Delfosse, G. Lacroix, Odemer, Influence of post-welding heat treatment on the corrosion behavior of a 2050-T3 aluminumcopper-lithium alloy friction stir welding joint, Electrochem. Soc. 158 (2011) C139–C147. [10] J.R. Davis, Corrosion of Aluminium and Aluminium Alloys, ASM International, Materials Park, OH, 1999. [11] J. Thompson, E.S. Tonkins, V.S. Agarwala, A heat treatment for reducing corrosion and stress corrosion cracking susceptibilities in 7xxx aluminium alloys, Mater. Perform. 26 (1987) 45–52. [12] B. Cina, Reducing the susceptibility of alloys particularly aluminium alloys to stress corrosion cracking, US Patent, 1974. [13] Y.L. Wang, Q.L. Pan, L.L. Wei, B. Li, Y. Wang, Effect of retrogression and reaging treatment on the microstructure and fatigue crack growth behaviour of 7050 aluminum alloy thick plate, Mater. Des. 55 (2014) 857–863. [14] W.C. Yang, S.X. Ji, Q. Zhang, M.P. Wang, Investigation of mechanical and corrosion properties of an Al-Zn-Mg-Cu alloy under various ageing conditions and interface analysis of η′ precipitate, Mater. Des. 85 (2015) 752–761.
225
Corrosion Science 139 (2018) 215–226
J.-l. Huang et al.
Al2CuLi (T1) precipitation, Metall. Trans. A 22 (1991) 299–306. [51] J.F. Li, Z.Q. Zheng, S.C. Li, W.J. Chen, W.D. Ren, X.S. Zhao, Simulation study on function mechanism of some precipitates in localized corrosion of Al alloy, Corros. Sci. 49 (2007) 2436–2449. [52] A.A. Gazder, W.Q. Cao, C.H.J. Davies, E.V. Pereloma, An EBSD investigation of interstitial-free steel subjected to equal channel angular extrusion, Mater. Sci. Eng. A 497 (2008) 341–352. [53] N. Birbilis, M.K. Cavanaugh, R.G. Buchheit, Electrochemical behavior and localized corrosion associated with Al7Cu2Fe particles in aluminum alloy 7075-T651, Corros. Sci. 48 (2006) 4202–4215.
(2006) 1057–1068. [47] K.H. Lee, Y.J. Lee, K. Hiraga, Precipitation behavior in the early stage of aging in an Al-Li-Cu-Mg-Zr-Ag (Weldalite 049) alloy, Mater. Res. 14 (1999) 384–389. [48] B.I. Rodgers, P.B. Prangnell, Quantification of the influence of increased prestretching on microstructure-strength relationships in the Al-Cu-Li alloy AA2195, Acta Mater. 108 (2016) 55–67. [49] J.F. Li, Z.H. Ye, D.Y. Liu, Y.L. Chen, X.H. Zhang, X.Z. Xu, Z.Q. Zheng, Influence of pre-deformation on aging precipitation behavior of three Al-Cu-Li alloys, Acta Metall. Sin. (Engl. Lett.) 30 (2017) 133–145. [50] W.A. Cassada, G.J. Shiflet, E.A. Starker Jr, The effect of plastic deformation on
226
