Journal of Materials Science & Technology  2020 , 40 (0): 107-112 https://doi.org/10.1016/j.jmst.2019.08.045

Enhancing the bake-hardening responses of a pre-aged Al-Mg-Si alloy by trace Sn additions

Gang Lua, Shuai Niea, Jianjun Wangab*, Ying Zhangd, Tianhai Wua, Yujie Liua, Chunming Liuac*

a Key Laboratory for Anisotropy and Texture of Materials (MoE), School of Materials Science and Engineering, Northeastern University, Shenyang 110819, China
b The State Key Laboratory of Rolling and Automation, Northeastern University, Shenyang 110819, China
c Research Center for Metallic Wires, Northeastern University, Shenyang 110819, China
d Shenyang Industrial Transformation Upgrading Promotion Center, Shenyang 110083, China

Corresponding authors:   *Corresponding authors at: Key Laboratory for Anisotropy and Texture of Mate-rials (MoE), School of Materials Science and Engineering, Northeastern University,Shenyang, 110819, China.E-mail addresses: wangjj@smm.neu.edu.cn (J. Wang), cmliu@mail.neu.edu.cn(C. Liu).*Corresponding authors at: Key Laboratory for Anisotropy and Texture of Mate-rials (MoE), School of Materials Science and Engineering, Northeastern University,Shenyang, 110819, China.E-mail addresses: wangjj@smm.neu.edu.cn (J. Wang), cmliu@mail.neu.edu.cn(C. Liu).*Corresponding authors at: Key Laboratory for Anisotropy and Texture of Mate-rials (MoE), School of Materials Science and Engineering, Northeastern University,Shenyang, 110819, China.E-mail addresses: wangjj@smm.neu.edu.cn (J. Wang), cmliu@mail.neu.edu.cn(C. Liu).

Received: 2019-07-27

Revised:  2019-08-27

Accepted:  2019-08-29

Online:  2020-03-01

Copyright:  2020 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology

More

Abstract

Effects of additions of trace Sn on the bake-hardening responses of a pre-aged Al-0.85Mg-0.85Si (in wt%) alloy were investigated through mechanical tests, differential scanning calorimetry, electrical resistivity and transmission electron microscopy. Results indicate that trace Sn additions reduced the number density of pre-aging clusters by inhibiting the formation of unstable counterpart during pre-aging treatment, leading to low strength and high supersaturation of solute atoms. In a subsequent paint-bake treatment, the presence of highly supersaturated solute atoms and high concentrated free vacancies moderated the activation energy barrier of βʺ phase and thus kinetically accelerated the formation of βʺ. Consequently, the trace Sn additions enhanced the bake-hardening responses of the pre-aged alloys significantly. The Sn-containing pre-aged Al-Mg-Si alloys with low strength and great bake-hardening responses hold promising potential for automotive body skin applications.

Keywords: Al-Mg-Si alloy ; Pre-aging ; Trace Sn ; Bake-hardening responses ; Precipitation kinetics

0

PDF (2259KB) Metadata Metrics Related articles

Cite this article Export EndNote Ris Bibtex

Gang Lu, Shuai Nie, Jianjun Wang, Ying Zhang, Tianhai Wu, Yujie Liu, Chunming Liu. Enhancing the bake-hardening responses of a pre-aged Al-Mg-Si alloy by trace Sn additions[J]. Journal of Materials Science & Technology, 2020, 40(0): 107-112 https://doi.org/10.1016/j.jmst.2019.08.045

1. Introduction

6xxx series Al-Mg-Si alloys are the most promising candidates of materials for automotive body owing to the combination of light weight and favorable engineering properties [[1], [2], [3], [4], [5], [6]]. Particularly the bake-hardening responses (BHR) caused by a large number of metastable (semi-)coherent, nanosized precipitates formed during paint-bake treatment are critical to the in-service performance. However, the inevitable natural aging (NA) stemmed from the logistics procedures has cast negative impacts on BHR of 6xxx series Al alloys [7,8]. During NA, a large number of natural aging clusters, namely Cluster(1) [9], are produced, leading to increment in initial strength and deterioration of formability [10]. However, Cluster(1) cannot transform directly into βʺ phase or serve as nuclei of βʺ due to the great differences from βʺ in composition (Mg/Si ratio) or smaller size than the critical value, and will dissolve in the matrix during subsequent paint-bake treatments [[11], [12], [13], [14], [15]]. Meanwhile, a larger number of solute atoms are consumed during NA, impeding precipitation in subsequent paint-bake treatment. Thus, NA significantly reduces BHR and even softens Al-Mg-Si alloys during paint-bake treatments [16].

Great efforts have been contributed to develop feasible methods for tackling the issues of NA in terms of BHR, such as pre-aging (PA) [17,18] and microalloying Sn into Al matrix [19,20]. PA that is conducted at 70-120 °C immediately after quenching promotes the formation of larger-sized pre-aging clusters with a composition closed to βʺ, i.e. Cluster(2) and suppresses the formation of Cluster(1) during subsequent NA [[21], [22], [23], [24]]. Cluster(2) transforms directly into βʺ or serves as nucleation sites for βʺ formation during subsequent paint-bake treatment and enhances the precipitation kinetics and BHR of alloys [22]. Sn microalloying was proposed to inhibit NA in Al-Mg-Si alloys recently [20]. It was reported that NA of Al-Mg-Si alloys was inhibited within 2 weeks by addition of parts-per-million Sn and the aging kinetics and peak hardness of artificial aging at 180 °C were enhanced. However, Cheng et al. [25] revealed that although trace Sn additions could suppress the early-stage clustering kinetics during NA, it also retarded the artificial aging kinetics and reduced T6 hardness. Werinos et al. [19,26,27] reported that the maximum quenched-in solubility of Sn, which depended on the content of Mg and Si, controlled the maximum retardation of NA, and a Sn-containing dilute Al-Mg-Si alloy with NA stability over six months was demonstrated. However, strength of the Al alloy after paint-bake treatment was insufficient to serve as automotive body skins.

Although previous studies have unveiled some individual effects of PA and Sn microalloying on aging properties, effects of Sn on BHR of pre-aged Al-Mg-Si alloys have scarcely been studied. In the present research, BHR of two pre-aged Al-Mg-Si alloys without and with Sn was studied. Effects of trace Sn additions on mechanical strength after PA were analyzed, and the formation of pre-aging clusters and the precipitation kinetics of β" were discussed.

2. Experimental

Al-Mg-Si alloys without and with Sn, namely alloys 1# and 2#, respectively, were cast and homogenized at 540 °C for 16 h, hot (450 °C) and then cold rolled down to 1 mm thick sheets. Chemical compositions were measured by inductively coupled plasma emission spectrometry and listed in Table 1. All the samples were solution treated at 555 °C for 20 min in an air circulation furnace, followed by water quenching to room temperature. Then two types of aging treatments were applied, i.e. NA at room temperature for 7 d (T4), and pre-aged at 90 °C for 5 h followed by a 7 d NA (T4P). Finally, a simulated paint-bake treatment, referring to an artificial aging at 190 °C for 30 min, was conducted to all samples.

Table 1   Chemical composition of two alloys investigated (wt%).

AlloysMgSiFeZnCrTiSnAl
1#Sn-free0.870.850.090.160.070.06<0.01Bal.
2#Sn-containing0.860.850.090.160.060.060.037Bal.

New window

Vickers micro hardness was measured by using a Wilson Wdpert 401MVD Vickers Hardness (HV) instrument with a load of 100 g and dwell time of 10 s. Five measurements of each group were carried out for reproducibility (n = 5). Tensile tests were performed in accordance with GB/T 228.1-2010standards using an AG-Xplus100kN Material Testing Machine at room temperature with a crosshead speed of 2 mm/min. Tensile specimens were machined along the longitudinal direction of the alloy sheets. Differential scanning calorimetry (DSC) tests were conducted on the disk-shaped samples of 5 mm in diameter, using a SETARAM DSC 131 from 40 to 400 °C under an argon atmosphere. In order to calculate the precipitation activation energy of βʺ, DSC tests were conducted at three different heating rates of 10, 15 and 20 °C/min, respectively. Electrical resistivity was measured at room temperature using a PZ-60A portable digital eddy current conductometer. The measurements of electrical resistivity were repeated three times for reproducibility (n = 3). Transmission electron microscopy (TEM) micrographs were obtained on a JEM-2100 F TEM operated at 200 kV. All TEM images were acquired along <001>Al zone axes. TEM samples were mechanically polished to approximate 100 μm in thickness, and twin-jet polished in an electrolyte of 30% nitric acid in methanol between -30 and -20 °C under 16 V.

3. Results

3.1. Mechanical properties

Mechanical properties of the experimental alloys in T4 and T4P were shown in Fig. 1. As seen from Fig. 1(a), hardness of alloy 1# and 2# in T4 was 83 HV and 79 HV, respectively, which were higher than those in T4P (76 HV and 66 HV). Particularly, hardness of alloy 2# in T4P was only 66 HV, lower than that in T4 by 13 HV. After paint-bake treatment, hardness of the two alloys reached to the same level of 106 HV. BHR of two alloys in T4P was 30 HV and 40 HV, respectively, much higher than those in T4 (13 HV and 8 HV). It is evident that BHR of Al-Mg-Si alloys was improved by PA and trace Sn additions reinforced the positive effects of PA on BHR. In addition, hardness of Sn-containing alloy in T4P was lower than that of Sn-free alloy.

Fig. 1.   Mechanical properties of two alloys in different conditions: (a) Vickers hardness; (b) tensile properties.

Effects of lean additions of Sn on BHR of the pre-aged alloys were further confirmed through tensile tests at ambient temperature, as shown in Fig. 1(b), and the detailed mechanical properties were shown in Table 2. Before paint-bake treatments, yield strength of alloy 2# was 122 MPa, lower than that of alloy 1# by 25 MPa. However, after paint-bake treatments, yield strength of both alloys increased to 257 MPa. BHR of alloy 2# reached to 135 MPa, whilst that of alloy 1# was 110 MPa. Such mechanical strength of both alloys after paint-bake treatments is desirable for in-service dent resistance, and the low strength of alloy 2# after T4P indicates promising formability.

Table 2   Mechanical properties of two alloys under different conditions.

StateAlloyYield strength (MPa)Ultimate Tensile Strength (MPa)Elongation (%)BHR (MPa)
Before-BHAfter-BHBefore-BHAfter-BHBefore-BHAfter-BH
T4P1#14725726933434.524.7110
2#12225724132934.525135

New window

3.2. DSC analysis

As is well known, increment in strength during paint-bake treatment is ascribed to the formation of precipitates, which could be interpreted through analyzing the thermal events in DSC plots. In order to study the mechanisms accounting for the effects of trace Sn additions on BHR, DSC analyses for both alloys in T4P were conducted with the heating rate of 10 °C/min, as shown in Fig. 2. There were one faint endothermic valley and two exothermic peaks, referred to valley A, peak B and peak C, respectively. According to previous research [9,12], the endothermic valley A is attributed to the dissolution of clusters, the exothermic peaks B and C are corresponding to the main strengthening phase βʺ and the over-aged phase β', respectively.

Fig. 2.   DSC traces of two alloys in T4P with heating rate of 10 °C /min.

As seen from Fig. 2, the endothermic valley A in the DSC trace of alloy 2# was smaller than that of alloy 1#, indicating that less clusters in alloy 2# dissolved than those in alloy 1# during the heating process, namely, the pre-aging clusters in alloy 2# were more stable. Many studies [9,[11], [12], [13], [14], [15]] suggest that the clusters with a composition (Mg/Si ratio) closed to β" evolved into β" during paint-bake treatment, while those with a composition (Mg/Si ratio) differing from β" dissolved in Al matrix. And the clusters with a similar composition (Mg/Si ratio) to β" have a larger size than those with a composition (Mg/Si ratio) differing from β" [[11], [12], [13], [14], [15],28]. Accordingly, it could be speculated that there were fewer pre-aging clusters with a size smaller than the critical value in alloy 2# than in alloy 1#. Thus, most pre-aging clusters in alloy 2# transformed directly into βʺ enhancing precipitation kinetics, which is consistent with the remarkable shift of exothermic peak B with reducing temperature. Small sub-peaks B' following the main exothermic peak B could be seen in both DSC curves, and the sub-peak B' of alloy 2# was smaller than that of alloy 1#. According to Takaki et al. [12] and Sawa et al. [29], the sub-peak was associated with βʺ formed after dissolution of preexisting clusters. It also implied that the pre-aging clusters in alloy 2# were more stable than those in alloy 1#. It should be noted that the exothermic peaks C of the two alloys emerged almost at the same temperature, indicating that trace Sn additions had slight effects on the precipitation kinetics of β'.

Furthermore, DSC analyses were performed at two different heating rates, i.e. 15 and 20 °C /min, respectively, where the temperature of peak B was defined as peak temperature of βʺ. Peak temperatures of βʺ in the DSC curves at various heating rates were concluded in Table 3. The Kissinger method [30] was used to calculate the formation activation energy of the main precipitation phase β" in such two alloys, which could be described as follows:

$\frac{α}{ T_{p}^2}=(\frac{Rk}{E})exp(\frac{-E}{ RT_p} $(1)

where α, k, Tp, E and R represented the heating rate (K/min), the reaction constant, the peak temperature of the given process, the activation energy and the universal gas constant, respectively. The formation activation energy E for βʺ could be determined from the plot of ln (α/$T_p^2$) versus 1/Tp, as shown in Fig. 3. The formation activation energy for βʺ in alloy 1# and 2# were 219.8 kJ/mol and 155.3 kJ/mol, respectively. Obviously, trace Sn additions reduced the formation activation energy of βʺ significantly, which was corresponding to the observation of βʺ peak shifting to lower temperature in Fig. 2.

Table 3   The peak temperatures of βʺ in the DSC curves performed at different heating rates.

AlloyHeating rate (°C/min)Peak temperature (°C)
1#10270.2
15274.3
20277.7
2#10255.5
15260.5
20265.3

New window

Fig. 3.   lnα/Tp2 versus 1/Tp for βʺ occurred during continuous heating at various heating rates.

3.3. Microstructure investigation

Fig. 4 showed TEM images and selected area electron diffraction (SAED) of the Al-Mg-Si alloys in T4P and T4P + BH. For both alloys in T4P (Fig. 4(a) and (b)), there were no any precipitates observed in the TEM images, while in T4P + BH (Fig. 4(c) and (d)), large numbers of dot-like precipitates and some needle-like fine precipitates were observed, and the number densities of precipitates in both alloys were almost identical, which contributed to the almost identical strength after paint-bake treatment. In order to reveal microstructure of the observed precipitates, high-resolution transmission electron microscopy (HRTEM) images of two types of precipitates were analyzed, as shown in Fig. 5. According to the HRTEM images and fast Fourier transform (FFT) patterns, these precipitates were coherent with Al matrix. The needle-like precipitates were β" phases and possessed an orientation relationship with Al matrix: [0 $\bar{1}$ 0]β"//[001]Al; (601)β"//(200)Al; ($\bar{4}$ 03)β"//(020)Al, and the dot-like precipitates were GP zones with a disordered structure.

Fig. 4.   TEM images of two alloys in different states: (a) alloy 1# in T4P; (b) alloy 2# in T4P; (c) alloy 1# in T4P + BH; (d) alloy 2# in T4P + BH.

Fig. 5.   HRTEM images and corresponding FFT patterns of precipitates: (a, b) β”; (c, d) GP zone.

3.4. Electrical resistivity determination

Although slight differences were seen from the TEM images of both alloys in T4P, their mechanical properties varied dramatically. Therefore, it could be speculated that the differences in mechanical properties of two pre-aged alloys were due to the pre-aging clusters. Electrical resistivity was closely correlated to the microstructure of Al alloys, in particular at the initial period of aging. As such it is a key indicator of the characteristics of clustering. Electrical resistivity, according to the Matthiessen’s laws [31], could be expressed as:

ρ=ρpure(T)+$\sum_{i}$ρiCippt(2)

where ρpure(T) is the temperature dependent resistivity, ρi is the specific resistivity of the ith solute and Ci is the concentration of the solute, and ρppt is the contribution of precipitates to resistivity. Thus, the change of electrical resistivity during PA could be described as:

Δρ=$\sum_{i}$ρiΔCi+Δρppt(3)

where $\sum_{i}$ρiΔCi and Δρppt represent the contribution of solute atoms depletion, which is detrimental to electrical resistivity [32], and pre-aging clusters formation, which is beneficial to electrical resistivity [31], respectively. It is noted that the total electrical resistivity of Al-Mg-Si alloys during PA always increases at low temperatures (≤ 120 °C) [31,33,34]. Thus, it is conclusive that the changes in electrical resistivity were dominated by pre-aging clusters formation. According to the results of Esmaeili et al. [35], electrical resistivity of Al alloys is a function of the number density of clusters formed during the early stages of aging. Therefore, the differences in electrical resistivity before and after PA reveal the variations of the number density of pre-aging clusters. Fig. 6 shows increasing electrical resistivity of both alloys during PA. It was obvious that the increase in electrical resistivity of alloy 2# was smaller than that of alloy 1#, suggesting that fewer clusters were formed in alloy 2# than that in alloy 1# during PA, which corresponded to the lower T4P strength of alloy 2#.

Fig. 6.   Electrical resistivity of two alloys in different conditions.

4. Discussion

As mentioned above, PA effectively suppressed strengthening during NA and increased BHR. In contrast, trace Sn additions not only enhanced BHR of pre-aged alloys, but also reduced mechanical strength in T4P, which was desirable for automobile body skin applications. It is well recognized that mechanical properties of Al-Mg-Si alloys were closely related to the precipitates formed in aging processes. To elucidate the dramatic improvements in mechanical properties and aging properties by trace Sn additions, the precipitation behavior of the two alloys during PA and paint-bake treatments were characterized in a combination with the free energy change of pre-aging clustering and precipitation, as shown in Fig. 7. After solution treatment, the concentration of solute atoms for two experimental alloys should be same X0, ascribing to the slight variations in composition except for Sn content. Thus, the precipitation driving force during PA for two experimental alloys was same (ΔGPA). However, free vacancies also played an important role in nucleation [36,37], and the nucleation rates increased with the concentration of free vacancies. Wolverton [38] calculated the solute-vacancy binding energy in pure Al by means of first-principles density functional theory, and found the binding energy of Sn reached up to 0.25 eV, which was greatly higher than that of Mg (-0.02 eV) and Si (0.08 eV). After quenching, numerous free vacancies were captured by solute atoms Sn, and the number of free vacancies decreased remarkably, leading to a lower nucleation rate of pre-aging clusters in alloy 2#. Consequently, fewer pre-aging clusters formed during pre-aging in alloy 2#, which is consistent with the results of resistivity measurement. It contributed to the lower strength of alloy 2# in T4P. Previous studies based on atom probe tomography [20,39,40] and high-angle annular dark-field scanning transmission electron microscopy [41] postulated that some Sn atoms appeared in the MgSi precipitates, but no individual Sn precipitates were evident, differing from the role of Sn additions in Al-Cu [42]. Due to the higher binding energy between Sn and Mg atoms [43], it was reasonable to speculate that more Mg atoms diffused into the pre-aging clusters in alloy 2# than in alloy 1#. Therefore, Mg/Si ratio of the pre-aging clusters in alloy 2# increased, resulting in high stability [7], which led to that few pre-aging clusters of alloy 2# dissolved during the heating process in DSC experiments.

Fig. 7.   Schematic free energy diagram of precipitation in two alloys.

As is well known, decomposition processes of experimental alloys cannot fully complete during such a short time period of PA. After PA, concentration of solute atoms for alloy 1# and 2# (X1 and X2) was higher than the equilibrium concentration X0' at pre-aging temperatures. Due to the higher nucleation rate of pre-aging clusters in alloy 1#, the concentration of solute atoms of alloy 1# after PA was lower than that of alloy 2# (X1<X2). As a result, the supersaturation of solute atoms in alloy 2# was higher than that in alloy 1#, therefore, the precipitation driving force during paint-bake treatment in alloy 2# was larger than that in alloy 1# (ΔGppt2>ΔGppt1). Moreover, the vacancies trapped by Sn atoms were released at the artificial aging temperature owing to a decrease in binding force between Sn and vacancies [20], accelerating the diffusion kinetics of solute atoms. Consequently, alloy 2# exhibited higher precipitation kinetics during paint-bake treatment and higher BHR than alloy 1#.

5. Conclusion

This study investigated the effects of lean additions of Sn on BHR of a pre-aged Al-0.85Mg-0.85Si (in wt%) alloy. Results indicate that such trace Sn additions significantly improved BHR of pre-aged Al-Mg-Si alloys from 110 MPa to 135 MPa. Trace Sn additions also reduced the number density of pre-aging clusters by inhibiting the formation of unstable counterparts, resulting in the lower strength in T4P. However, the formation activation energy of βʺ was reduced by the introduction of trace Sn, whilst it kinetically accelerated the formation of βʺ owing to the high supersaturation of solute atoms and the high concentration of free vacancy during paint-bake treatment. After paint-bake treatments, Sn-free and -containing alloys differed only slightly in microstructure and the finial strength. This study demonstrates that Sn-containing pre-aged Al-Mg-Si alloys with low strength and great BHR hold promising potential for automotive body skin applications.

Acknowledgement

This work was financially supported by the National Key R&D Program of China (No. 2018YFB2001800).


/