Achieving high temperature deformation for Al-Mg alloys processed by severe plastic deformation.

Comaneci, Radu ; Zaharia, Luchian ; Chelariu, Romeu 等

1. INTRODUCTION

There is now a great interest in developing highly formable Al-Mg based alloys for application leading to lightweight vehicles. Significant efforts have been made to improve the formability and strength of Al-Mg alloys. Wrought Al-Mg alloys (5xxx series) are non-heat-treatable. For a small amount of Mg, strength properties of these alloys are diminishing; but they have good corrosion resistance and weldability. Because of that, their commercial applications are restricted to very specific area. Scandium is the only alloying equilibrated, thermally stable, coherent L12 phase, i.e. [Al.sub.3]Sc, in aluminum. Thus, [Al.sub.3]Sc precipitate can be used as grain boundary resistant barrier to coarsening (Kyung-Tae et al., 2005, Musin et al., 2004).

The principle of developing (super)plasticity through SPD relies on the properties varying inversely with the submicronic grain size. On review of earlier works (Noda et al., 2004, Kyung-Tae et al., 2006) it is shown that to develop superplasticity, the material has to attain ultrafine grain sizes and retain the ultrafine sizes at elevated testing temperatures. If these ultrafine grains show reasonable stability at elevated temperatures, the alloy may exhibit a capability for achieving good ductility.

The present work thus initiated to explore the potential of achieving high temperature deformation for Al-1.5Mg alloy without the third alloying element, trough ECAP technique. By improvement of mechanical properties and formability of Al-Mg alloys it is possible to introduce these non-heat-treatable but very good corrosion resistant alloys in near net shape forming processes.

2. EXPERIMENTAL PROCEDURE

Ingots of Al-1.35Mg with dimensions of 20x24x150mm were given a homogenization heat-treatment at 500[degrees]C for 24 hours (this resulted in coarsening of about 100 [micro]m) and then machined at final dimensions of 10x10x60mm.

These billets were pressed in an equal-channel angular die of 10 mm width channel, with a [phi] = 90[degrees] channel intersection angle and a [psi] = 28[degrees] curvature on the outer side of channel intersection, see Figure 1. The metal was subjected to a simple share strain in the bisect plan of the channels, under relative low pressure compared to the traditional extrusion process. The workpieces were pressed through the die 6 passes via route BC (rotation of sample by 90[degrees] in same direction after every pass).

[FIGURE 1 OMITTED]

The extrusions were carried out at 293 K at a speed of 8.75 mm/s. In order to reduce the friction, a solid lubricant was used. The total equivalent strain accumulated for N = 6 passes was [approximately equal to] 6 according to well-known Equation (1).

[s.sub.N] - N / [square root of 3][2 cot ([phi] / 2 + [psi] / 2) + [psi]cso ([phi] / 2 + [psi] / 2)] (1)

where the significance of terms are revealed in Figure 1.

Samples of 24mm total length with 1.25mm gauge height, 10mm gauge length and 2.5mm gauge width were machined parallel to the longitudinal axis of the pressed rods by electro-discharge machine. The dimensions were established according to the rule of proportional rectangular tensile specimens:

[L.sub.g] = k - [square root of ([A.sub.g])]

where k = 5.65 (from ASTM Standard EN 10002-1) and [L.sub.g] and [A.sub.g] are the gauge length and gauge area respectively. The dimensions are imposed by the final dimensions of the work pieces after ECAP process which step down, Figure 2.

[FIGURE 2 OMITTED]

Tensile test is performed at 463, 493 and 523K at an initial strain rate of 1.68 x [10.sup.-3][s.sup.-1] and 8.4 x [10.sup.-5][s.sup.-1]. The samples were subjected to tensile tests through a mounting device, Figure 3, under controlled temperature conditions using a universal INTRON 3382 testing machine and an environmental chamber INSTRON 3119-506. The samples were held at nominal temperatures for 120min.

[FIGURE 3 OMITTED]

3. RESULTS AND DISCUSSION

The Figure 4 shows the typical true stress--true strain curve for the ECAP processed Al-Mg alloy (six passes) at an initial strain rate of 8.4 x [10.sup.-5][s.sup.-1] for 220[degrees]C.

[FIGURE 4 OMITTED]

The complete results of performed tensile tests for a strain rate of 1.68 x [10.sup.-3][s.sup.-1] and 8.4 x [10.sup.-5][s.sup.-1] are presented in Fig. 5 and 6. passes ECAP (test performed at strain rate 8.4 x [10.sub.-5][s.sup.-1])

Extensive strain hardening takes place initially. After reaching a maximum, the flow stress continously decreases until failure. The increase of temperature leads not to a shift of the peak stress to a higher strain as we expect. That because the lower strain rate sensitivity of alloy in the testing temperatures range. A steady-state flow was not found at all examined temperatures despite the fact that the value of elongation-to-failure is relatively high. No superplasticity was reported.

We found maximum tensile ductility (122%) at a middle of the testing temperatures range for the lower strain rate.

As is known from the Considere criterion (Wang & Ma, 2003) inhomogeneous deformation (necking) sets in when:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

where a and e are true stress and true strain, respectively.

For rate-sensitive materials, the presence of the strain rate sensitivity of the flow stress (m) helps to sustain the homogeneous deformation. The Hart's instability criterion is (Wang & Ma, 2003):

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)

These equations indicate that for a high strength material (large [sigma]), which is the case for UFG metals, sufficiently large strain hardening and/or strain rate hardening need to be present to sustain the uniform straining before the onset of localized deformation. A simple derivation based on Eq. 3 shows that the strain hardening exponent (n) corresponds to the true strain at necking under uniaxial tension. The strain hardening and strain rate hardening properties are therefore important for stable plastic deformation. It seems that without coherent [Al.sub.3]Sc dispersoids which are highly effective in pinning of boundaries no exceptional elongations takes place because of the extensive static growth of structure takes place.

4. CONCLUSIONS

Due to ultra-fine grain structure achieved by ECAP, the Al-1.35Mg alloy exhibit low flow stress and large degree of ductility at high temperature and it becomes suitable for near net shape forming processes. The high formability expands the possibility of product design together with superior transferability of the die surface to the metal sheet.

5. REFERENCES

Kyung-Tae, P.; Chong, S. L. and Dong, H. S. (2005). Superplastic Behavior of as--Equal Channel Angular Pressed 5083 Al and 5083 Al--0.2 Sc Alloys, Materials Science Forum, Vols. 475-479, 2005, pp. 2937-2940.

Kyung-Tae, P.; Chong, S. L.; Yong S. K. and Dong H. S. (2006). Superplastic Deformation of Ultrafine Grained Al Alloy Processed by ECAP and Post-Rolling, Materials Science Forum, Vols. 503-504,2006, pp. 119-124.

Musin, F.; Kaibyshev, R.; Motohashi, Y. and Itoh, G. (2004). Superplastic behaviour and microstructure evolution in a commercial ultra-fine grained Al-Mg-Sc alloy, Materials Science Forum, Vols. 447-448, 2004, pp. 417-422.

Noda, M.; Funami, K.; Hirohashi, M. and Kobayashi, M. (2004) Effect of Grain Size and Microstructure on Appearance of Low Temperature Superplasticity in Al-Mg Alloy, Materials Science Forum, Vols. 447-448, 2004, pp. 435-440.

Wang, Y.M. and Ma. E. (2003). Strain Hardening and Strain Rate Sensitivity of Ultrafine-Grained Metals, Journal of Metastable and Nanocrystalline Materials, Vol. 17, 2003, pp. 55-64.

Fig. 5. Evolution of mechanical properties for Al-1.35Mg after 6
passes ECAP (test performed at strain rate 1.68 x [10.sup.-3]
[s.sup.-1])

 190[degrees]C 220[degrees]C 250[degrees]C

Rm [MPa] 90.89 56.07 39.02

Rp0.2 [MPa] 73.01 41.19 32.15

A [%] 27.89 42.40 47.63

Note: Table made from bar graph.

Fig. 6. Evolution of mechanical properties for Al-1.35Mg after 6

 190[degrees]C 220[degrees]C 250[degrees]C

Rm [MPa] 71.12 68.57 58.79

Rp0.2 [MPa] 38.36 34.48 31.41

A [%] 66.29 122.57 107.55

Note: Table made from bar graph.