SIMTech technical reports (STR_V11_N1_01_FTG) Volume 11 Number 1 Jan-Mar 2010

1

High temperature formability in AA5083 and AA6061

A. E. W. Jarfors, Y. Aue-U-Lan, S. Castagne, J. Liu1, and M. J. Tan1

1 School of Mechanical & Aerospace Engineering, Nanyang Technological University, Singapore

Abstract – In this paper, considering the limited applications in non-superplastic materials, two typical alloys of AA5083 and AA6061 were investigated and compared during high temperature tensile tests to study their formability. The results of tensile tests and microscopy are shown which indicate the deformation properties under different test conditions. It is found that the flow stress coupled with the dynamic grain growth is related with the temperatures and strain rates. Furthermore, the highest strain rate sensitivity (m value) is obtained. Hence, the results would be analysed so as to apply those non-superplastic mate-rials for a new concept superplastic forming.

Keywords: Aluminum alloy, Superplastic & non- superplastic forming, Flow stress, Strain rate sensi-tivity, Elongation

1 BACKGROUND Generically, the aluminum alloys of

heat-treatable and high strength are common used materials, such as 7075, 7475, 2024, 5083 and 6061 [1]. Superplastic materials are polycrystalline solids which have the ability to undergo large uniform strains prior to failure [2].

As a commercial alloy with corrosion resistance and moderate to high strength, the 5083 aluminum alloy is widely used in structural sheet-metal applica-tions [3].

In order to obtain great formability and me-chanical properties, the materials with superplasticity should be specially prepared. A lot of processing approaches including rolling, forging, extrusion, ECAE, FSP and various techniques have been used to produce fine recrystallised grain size and the obser-vation of superplasticity [4-7]. However, those thermo-mechanical processing methods would limit the Superplastic Form (SPF) to great development due to relatively higher time and cost consuming [8]. Based on the limitations, a more formable material containing fine and stable structure but low cost is of great need, or move forward with the process devel-opment by expanding the conventional materials to be used as the SPF application.

2 OBJECTIVE In this paper, considering the limited applications

in non-superplastic materials, two typical alloys of

AA5083 and AA6061 were investigated during high temperature tensile tests to study their formability. The results of the tensile test coupled with optical microscopy have been investigated, which indicate the deformation properties under different test condi-tions. The flow stress as a function of temperature and strain rate has been shown to establish the formability of these two alloys. Furthermore, the highest strain rate sensitivity (m-value) is obtained, which would verify the feasibility of being used in a new concept SPF.

3 EXPERIMENTAL

The rolled AA5083 and AA6061 sheets with thickness of 3 mm were used for high temperature tensile test and microstructure observation. Tensile specimens with 20 mm gauge lengths and 4 mm gauge widths were machined from the sheets. In order to obtain better formability during test, all the AA5083 and AA6061 specimens have been annealed for 2 hours at the temperatures of 345ºC and 415ºC, re-spectively. For AA5083, an aging treatment at 150ºC for 24 hours was proved to be helpful to successfully reveal the microstructure by decorating the grain boundary with precipitates. The samples were then mechanically polished and etched with Graf Sergeant etchant (15.5 mL HNO3, 0.5 mL HF, 3 g Cr2O3, and 84 mL H2O) [3].

Figure 1 shows the typical optical microstructure of AA5083 in the annealed condition. The grains were somewhat brick shaped rather than equiaxed. The grain size was measure by the linear intercept method [6]. It is about 13.6 µm in average.

Fig. 1. Micrographs of AA5083 in the annealed condition.

A. E. W. Jarfors et al

2

Uniaxial tensile test of the specimens were per-formed using Instron 8500 testing machine equipped with an electrical resistance furnace chamber. The specimens were tested in the temperature range of 300ºC, 400ºC and 500ºC at constant crosshead dis-placement velocity, with the corresponding initial strain rate from 0.0003s-1 to 0.3s-1.

4 RESULTS & DISCUSSION

4.1 High Temperature Tensile Tests

Figure 2 shows the true stress-strain plots for a series of initial strain rates at temperature of 300ºC. It can be observed from Fig. 1, for both materials, the flow stress appears to increase with decreasing strain rate. It has been reported that the enhanced strain hardening at slower strain rates may because of dy-namic grain growth. However, the curves exhibit relatively weaker strain hardening effects especially at strain rates lower than 0.1s-1. Furthermore, it shows lower strain hardening during constant speed tests compared with the corresponding constant strain rate tests [3]. The curves show the similar profiles at dif-ferent initial stain rates under higher temperature of 400ºC or 500ºC.

0.0 0.1 0.2 0.3 0.4 0.5 0.60

40

80

120

160

200

3E-1 1E-1 2E-2 1E-2 1E-3 1E-4

True

stre

ss, M

Pa

True strain

AA5083(a)

(a) AA5083

0.0 0.1 0.2 0.3 0.4 0.50

20

40

60

80

100

True

stre

ss, M

Pa

True strain

3E-1 1E-1 2E-2 1E-2 1E-3

AA6061(b)

(b) AA6061

Fig. 2. True stress-true strain tensile tests curves at 300ºC and various initial strain rates.

The true stresses as a function of strain rate at different temperatures are plotted in Fig. 3. For both materials, it can be found that the curves show typical sigmoidal profiles. The flow stress appears to be somewhat related with the strain rate, while it increase with increasing strain rate. Besides, the two materials vary very much from each other especially during higher strain rates.

The strain rate sensitivity (m value), calculated from slopes of the log stress versus log strain rate, ( εσ &logd/logdm = ), are plotted in Fig. 4 respec-tively. The main trends for data on m value are as follows: (1) Generally, the m values peaks increase as the

temperature increasing except AA5083. As result of weak strain hardening and grain growth effects during testing at 300ºC, the alloy behaves high stress at higher strain rate but low stress at lower strain rate.

(2) For superplastic behaviour, m value would be greater than or equal to 0.3 [1,2]. However, this would not happen because of the non-superplastic materials. Unlike the superplastic materials, both AA5083 and AA6061 show the maximum vales at higher strain rate. A maximum value of 0.41 is obtained at 300ºC for AA5083, while 0.24 for AA6061 at 500ºC.

10-4 10-3 10-2 10-1 100

10

100

1000300oC400oC400oC

Flow

stre

ss, M

Pa

Strain rate, s-1

(a) AA5083

(a) AA5083

10-4 10-3 10-2 10-1 100

10

100

1000300oC400oC400oC

Flow

stre

ss, M

Pa

Strain rate, s-1

(a) AA5083

(b) AA6061

Fig. 3. Strain rate dependence of flow stress at different temperatures.

High temperature formability in AA5083 and AA6061

3

-4 -3 -2 -1 00.0

0.1

0.2

0.3

0.4

0.5 300oC 400oC 500oC

m v

alue

Logarithm of strain rate

(a) AA5083

(a) AA5083

-3 -2 -1 00.0

0.1

0.2

0.3

0.4

0.5 300oC 400oC 500oC

m v

alue

Logarithm of strain rate

(b) AA6061

(b) AA6061

Fig. 4. The m values plotted as a function of log strain rate at different temperatures.

The total elongation of test samples is used as a

measure of the formability of the material, and is typically a function of strain rate and temperature. The elongation for both alloys is plotted as a function of log initial strain rate and is shown in Fig. 5. Generally, the tensile elongation increases with increasing strain rate and test temperature. This trend is different from the superplastic materials with slower strain rate showing larger elongations [7,9-11]. However, the elongations do not vary a great from different rates during test at 300ºC or 400ºC. That is probably due to the relatively less cavitation and weaker strain hard-ening effects especially at high rate.

10-4 10-3 10-2 10-1 100

20

40

60

80

100

120 300oC 400oC 500oC

Elo

ngat

ion,

%

Initial strain rate, s-1

(a) AA5083

(a) AA5083

10-3 10-2 10-1 100

20

40

60

80

100

120 300oC 400oC 500oC

Elo

ngat

ion,

%

Initial strain rate, s-1

(b) AA6061

(b) AA6061 Fig. 5. Tensile elongation as a function of initial strain rate at different temperatures.

It is true that there would be much more cavities

and grain growth at 500ºC. The largest elongations are 93.25% and 69.00% respectively for AA5083 and AA6061 at 500ºC. Besides, the m value confers a high resistance to neck development and results in the high tensile elongations [2], so that the maximum elonga-tion is coincide with the ranges of highest m value distributions.

4.2 Microstructure

Typical micrographs of AA5083 tested at 300ºC and initial strain rate of 10-4s-1 in the grip and gauge regions are shown in Fig. 6. Not surprisingly, the grains appear to be somewhat coarser in the samples deformed due to the grain growth at high temperature. The increase in-grain growth caused by different temperature is evident by comparing the microstruc-ture with Fig. 7. Similarly to the superplastic materials [10], the alloy shows much more grain growth at 500ºC. The average grain size of grip region is about 22.5 µm at 300ºC, while some of the grains grow as high as 200 µm at 500ºC. The microstructure also contains a wide range of grain sizes, which are com-posed of the new, smaller grains formed due to re-crystallisation. It is interesting to note that the grains have higher aspect ratio along the tensile direction at 300ºC.

It has been reported that the failure is mostly in-ternal cavitation at the lower strain rates; but it changes to external necking at higher rates [10]. As a result, there are more cavities with the decreasing of strain rate. Furthermore, there are much more cavities and grain growth at higher temperature of 500ºC. The fracture tips have jagged appearance especially at lower rates. That is most likely because the sample fractures along the intergranular cavities.

A. E. W. Jarfors et al

4

It is found that less grain growth and weaker strain hardening could happen in specified testing condition. Besides, the dislocation creep seems play a role during deformation [11]. Although neither of the two non-superplastic materials can behave as high as the elongation of superplasticty, the relative low flow stress, ductility, microstructure, temperature and strain rate would be considered for the formability behaviour. Hence, a new approach could be applied by using those non-superplastic materials for a new concept superplastic forming.

(a) Grip region

(b) Gauge region

Fig. 6. Micrographs of AA5083 tested at 300ºC and initial strain rate of 10-4s-1.

(a) Grip region

(b) Gauge region

Fig. 7. Micrographs of AA5083 tested at 500ºC and initial strain rate of 10-4s-1.

5 CONCLUSIONS

AA5083 and AA6061 specimens have been an-nealed for testing the formability at high temperature and different rates. The mechanical properties cou-pled with microstructure have indicated the formabil-ity behaviour. Experimental results also prove the capability of applying those non-superplastic materi-als for a new concept superplastic forming. The con-clusions are summarised as follows: 1. Both of the two alloys exhibit relatively weaker

strain hardening effects especially at relatively lower strain rates. Generally, the flow stress in-creases with increasing strain rate.

2. The m peaks increase as the temperature in-creasing, except AA5083 at 300ºC. Both AA5083 and AA6061 show the maximum values at higher strain rate. A maximum m value of 0.41 is ob-tained at 300ºC for AA5083, while 0.24 for AA6061 at 500ºC.

3. The largest elongations are 93.25% and 69.00% respectively for AA5083 and AA6061 at 500ºC. The elongation is also coinciding with the ranges of highest m value distributions.

4. The grains appear to be somewhat coarser in the samples deformed due to the grain growth at high temperature. The grain size is about 13.6 µm in annealed condition, while it grows up to 22.5 µm at 300ºC and as high as 200 µm at 500ºC.

5. Recrystallisation happens during deformation. There are more cavities at higher temperature and lower strain rate. The sample fractures along the intergranular cavities.

6 INDUSTRIAL SIGNIFICANCE

The assesment of the material properties at extreme tyemperature has significance tot hew tool design when forming these traditionally non- superplastic materials. Superplastic forming is now entering Singapore with the new investment in jet engine manufacturing and as such the process is gaining in importance.

High temperature formability in AA5083 and AA6061

5

The MRO business in Singapore could also venture into high mix low volume manufacturing using superplastic forming using the cost advantage iof the non-superplastic grades of materials for legacy parts for the aircraft nmanufacturer to strengthen the position as a total MRO hub.

The current work assesses these low cost materials to provide data for die design and is as such an integral part in the knowlegde base necessary for the transition to happen.

REFERENCES [1] T.G. Nieh, J. Wadsworth, O.D. Sherby, Superplasticity

in Metals and Ceramics, Cambridge: Cambridge University Press, 1997.

[2] J. Pilling, N. Ridley, Superplasticity in Crystalline Solids, London: The Institute of Metals, 1989.

[3] R. Verma, P.A. Friedman, A.K. Ghosh, S. Kim, and C. Kim, “Characterization of Superplastic Deformation Behavior of a Fine Grain 5083 Al Alloy Sheet”, Met-allurgical Mater. Transac. A, vol. 27A, pp. 1889-1898, 1996.

[4] K. Kannan, J.S. Vetrano, and C.H. Hamilton, “Effects of Alloy Modification and Thermomechanical Proc-essing on Recrystallization of Al-Mg-Mn Alloys”, Metallurgical Mater. Transac. A, vol. 27A, pp. 2947- 2957, 1996.

[5] Y. Wu, L.D. Castillo, and E.J. Lavernia, “Superplastic Behavior of Spray-Deposited 5083 Al”, Metallurgical Mater. Transac. A, vol. 28A, pp. 1059-1068, 1997.

[6] D.H. Bae, and A.K. Ghosh, “Grain Size and Tem-perature Dependence of Superplastic Deformation in an Al-Mg Alloy Under Isostructural Condition”, Acta Materialia, vol. 48, pp. 1207-1224, 2000.

[7] R.M. Cleveland, A.K. Ghosh, and J.R. Bradley, “Comparison of Superplastic Behavior in Two 5083 Aluminum Alloys”, Mater. Sci. Eng. A, vol. 351, pp. 228-236, 2003.

[8] P.A. Friedman, S.G. Luckey, W.B. Copple, R. Allor, C.E. Miller, and C. Young, “Overview of Superplastic Forming Research at Ford Motor Company”, J. Mater. Eng. Perform., vol. 13, pp. 670-677, 2004.

[9] W.J. Kim, Y.K. Sa, H.K. Kim, and U.S. Yoon, “Plastic Forming of the Equal-Channel Angular Pressing Procesed 6061 Aluminum Alloy”, Mater. Sci. Eng. A, vol. 487, pp. 360-368, 2008.

[10] P.A. Friedman, and W.B. Copple, “Superplastic Re-sponse in Al-Mg Sheet Alloys”, J. Mater. Eng. Per-form., vol. 13, pp. 335-347, 2004.

[11] K.T. Park, D.Y. Hwang, S.Y. Chang, and D.H. Shin, “Low-Temperature Superplastic Behavior of a Sub-micrometer-Grained 5083 Al Alloy Fabricated by Se-vere Plastic Deformation”, Metallurgical Mater. Transac. A, vol. 33A, pp. 2859-2867, 2002.