Materials Chemistry and Physics 130 (2011) 1357–1360

Contents lists available at SciVerse ScienceDirect

Materials Chemistry and Physics

journal homepage: www.elsevier.com/locate/matchemphys

The effect of extrusion speed on the structure and corrosion properties of aged
and non-aged 6063 aluminum alloy
A. Aytaç a,b , B. Daşçılar a , M. Usta a,∗
a
Department of Materials Science and Engineering, Gebze Institute of Technology, 41400 Gebze, Kocaeli, Turkey
b
Department of Chemistry, Faculty of Science, Gazi University, 06500 Teknikokullar, Ankara, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: Extrusion is used in processing of Al alloy for consolidation, redistribution of reinforcements, and shape
Received 14 January 2011 forming. The important parameter that controls the extrusion process is extrusion rate, which is a function
Received in revised form 13 May 2011 of extrusion equipment and parameters. Homogenized AA6063 alloy billets were extruded at different
Accepted 11 September 2011
ram speeds related to extrusion speed (3, 6, 10 mm s−1 ) and then aged at 185 ◦ C for 6 h. The extruded
samples were studied for their corrosion, microstructure, and mechanical properties. The effects of extru-
Keywords:
sion speed and aging on the corrosion behavior of AA6063 were investigated using dynamic polarization
A. Alloys
and impedance techniques in 0.5 M NaCl aqueous solution. The surface structures were then character-
C. Electrochemical techniques
C. Hardness
ized by means of surface profilometer. The electrochemical measurements showed that the decreasing
C. Corrosion test ram speed for the aged sample improved the corrosion resistance. In addition, the aging improved the
corrosion resistance of aluminum in 0.5 NaCl environments.
© 2011 Elsevier B.V. All rights reserved.

1. Introduction lines, pick-up and tearing [12]. In literature, many works were done
to eliminate those defects with changing extrusion speeds [13–17].
Extrusion speeds are controlled by the press capacity and incip- However, the effect of extrusion speeds on the corrosion behavior
ient melting depending on the temperatures. The press capacity of the materials is missing although this behavior is very important
is dominant at low temperatures, whereas the incipient melting in use of extruded material. The increase in extrusion speed can be
becomes the controlling parameter at high temperatures. The oper- advantage in order to increase the productivity. However, increas-
ating window is further shifted toward lower extrusion speeds in ing the extrusion speed may be disadvantage in terms of corrosion
metal matrix composites as a result of stick–slip tearing, which is behavior of metals. In this context, the current study is achieved to
related to low ductility of metal matrix composites and scouring investigate this behavior.
action of the reinforcements on the die land [1]. Brusethaug et al.
[2] defined the extrudability as the maximum speed that could be 2. Experimental
achieved before the surface tears became apparent, and reported
2.1. Materials
that the extrusion speed limit for A357 Al/55 SiCp composite was
30–40% less than that for unreinforced alloy. Selseth and Lefstand Fig. 1 shows typical temperature/time profile for the current study used. Almost
[3] attributed the decrease in extrusion speed limit to the complex- 4 m-long cylindrical billets were cast by direct-chill casting process. Then, they were
ity of the extrudated shape. Lieblich et al. [4] observed a decrease in cut in proper size and homogenized at 585 ◦ C for 8 h. Preheating before extrusion
was applied at 500 ◦ C. Direct extrusion operation was maintained by using simple
the speed and increase in temperature during extrusion when the
solid die at different ram speeds of 3, 6 and 10 mm s−1 . Since the ram speed is related
extrusion ratio was increased. Extrusion was used as metal form- to extrusion speed for a given extrusion shape and it is more common to use the
ing technique for 6XXX, 7XXX and 2XXX series aluminum alloys extrusion speed than the ram speed, the results are given in terms of extrusion speed
[1,3,5–11]. in the present paper. After the extrusion, each profile extruded for different speed
was cut in half and half of them were artificially aged at 185 ◦ C for 6 h, while the rest
There are numerous limitations on the production of aluminum
were not aged. The aged and non-aged samples were exposed to corrosion study as
alloy profiles. Maximum extrusion speed is important for higher explained below. Table 1 shows the chemical composition of 6063 alloy used in the
productivity. The higher the maximum extrusion speed, the higher study. Magnesium and silicon are the alloying elements, whereas Fe is not alloying
productivity will be. Within the proper extrusion speed, the pro- element in 6xxx series aluminum alloys.
ductivity can be increased by eliminating surface defects such as die
2.2. Hardness and surface roughness

Hardness properties of the extruded aluminum alloy sample were carried out
∗ Corresponding author. Tel.: +90 2626051782; fax: +90 2626538490. with MTV-10 hardness tester machine using HV scale. Surface roughness char-
E-mail address: [email protected] (M. Usta). acterization was performed on the surface using the Veeco Dektak 8 Advanced

0254-0584/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.matchemphys.2011.09.029
1358 A. Aytaç et al. / Materials Chemistry and Physics 130 (2011) 1357–1360

Table 1
Chemical composition of alloy by weight percentage.

Si Fe Cu Mn Mg Cr Ti V Al

0.38 0.165 <0.015 0.0188 0.487 <0.015 0.0081 <0.0041 Balance

Table 3
The effect of the extrusion speed on the surface roughness of the aged and non-aged
samples.

Extrusion speed/ram Surface roughness (␮m), Ra-average

speed (mm s−1 )

Non-aged Aged

3 299.45 229.27
6 357.87 282.08
10 439.18 301.37

greater way. In addition, the hardness increases with aging due to

the precipitation hardening.
Surface roughness (Ra ) is parameter that shows a surface qual-
Fig. 1. Temperature/time profile for the current study used. ity. Ra is denoted as arithmetical average (AA) [18]. The surface
roughness of the samples (Ra ) was measured using a Veeco Dek-
tak 8 Advanced Development Profilometer by scanning the area
Development Profiler surface profilometer. Samples were cleaned with ethanol, dis-
tilled water and butanol before the tests. Moreover, lens paper was used to wipe any of 300 ␮m × 500 ␮m with the scan rate of 20 s. The average values
remaining contaminants off the surface. The images were exported to ASCII text files of 48 measurements are shown in Table 3. It can be seen that the
and the surface roughness analysis was performed using custom written code. 6063 Al surface becomes less rough after aging because the surface
of oxide layer that covers the deformation surface increases dur-
2.3. Polarization measurement ing aging process. The surface becomes more homogenous with
the oxide layer. However, the surface roughness increases with the
Potentiodynamic polarization measurement and electrochemical impedance
measurements were carried out with conventional three electrodes electrochemical increase in extrusion speed.
cell. The counter and reference electrodes were a platinum plate (1 cm2 ) and Ag/AgCl
electrode, respectively. Working electrodes (5 cm × 10 cm × 0.2 cm) was sealed in 3.2. Tafel plots
Teflon holders prior to each experiment.
The polarization measurements were carried out in a 0.5 M NaCl solution at 25 ◦ C
in a Pyrex glass cell exposed to atmospheric air. A Volta Lab 80 (PGZ 301 Dynamic- Tafel plots recorded for all the studied specimens are depicted in
EIS Voltammetry) potentiostat was used in all experiments. The potential vs. current Figs. 2 and 3. From the plots it can be noted that diffusion-limiting
density curves were recorded with 5 mV s−1 . The exposed area of the samples was behavior starts in the cathodic current after 20–30 mV (Ag/AgCl)
1.04 cm2 .
of polarization from Ecorr for the samples. The corrosion current
(icorr ), cathodic (ˇc ) and anodic (ˇa ) Tafel constants derived from
2.4. Impedance measurement
the Tafel plots are summarized in Table 4. From the observed data,
The impedance measurements were carried out in the frequency region of it can be seen that the 3 mm s−1 extruded aged sample shows mini-
50 MHz to 20 kHz, taking five points per decade for different time. The real (Z ) and mum icorr value as compared to others. Non-aged samples show the
imaginary (Z ) components of the impedance spectra in the complex plane were higher corrosion rate than the aged samples and the icorr of the non-
analyzed using the Circular Regression programmer in the Volta Lab 80 potentio-
aged samples slightly increase order of 3, 10 and 6. Aging process
stat. The samples were prepared in the same way as microstructures evaluation and
polarization measurement. improved corrosion resistance of the samples. ˇc derived for aged
samples is found to be greater than ˇc derived for the non-aged
samples except for 3 mm s−1 extruded sample. The observation of
3. Results and discussion
limiting behavior of cathodic current in the composites indicates
the existence of cathodic controlled reaction, i.e., oxygen reduction
3.1. Hardness and surface roughness

The hardness of samples was measured in HV scale using MTV-

10 hardness tester at an applied load of 20 kg. The aged and
non-aged specimens were tested. Ten hardness measurements
were taken from each sample and the average value was reported.
The effect of extrusion speed on the hardness of aged and non-aged
samples is given in Table 2. As shown in table, the extrusion speed
does not change the hardness of aged and non-aged samples in a

Table 2
The effect of the extrusion speed on the hardness of the aged and non-aged samples.

Extrusion speed (mm s−1 ) Hardness (HV 20)

Non-aged Aged

3 56.00 ± 2.28 73.57 ± 0.97

6 59.66 ± 1.96 82.57 ± 2.07
10 56.83 ± 0.98 77.85 ± 2.91 Fig. 2. The polarization curves of aged Al samples with three extrusion speed in
0.5 M NaCl at 25 ◦ C with scan rate of 5 mV s−1 .
 A. Aytaç et al. / Materials Chemistry and Physics 130 (2011) 1357–1360 1359

Table 4
The electrochemical parameters obtained from the Tafel slopes at 25 ◦ C (5 mV s−1 ).

Samples Extrusion speed (mm s−1 ) −Ecor (V) icor (␮A cm−2 ) ˇa (mV dec−1 ) ˇc (mV dec−1 )

Aged 3 1.434 6.71 116.3 59.6

6 1.512 27.23 323.9 91.0
10 1.412 14.34 566.0 100.4

Non-aged 3 1.539 101.6 290.1 84.5

6 1.523 102.2 200.5 81.0
10 1.517 101.0 155.2 78.1

Fig. 5. The Nyquist diagram of the aged Al in 0.5 M NaCl at 25 ◦ C.

Polarization resistance obtained from uncompleted arc in

Fig. 3. The polarization curves of non-aged Al samples with three extrusion speed Figs. 4 and 5 are quite different from each other (Table 5). However,
in 0.5 M NaCl at 25 ◦ C with scan rate of 5 mV s−1 .
in Fig. 5 the impedance curves obtained from the aged Al samples
show completed arcs. This can be explained as follows: There is
reaction. Similar to ˇc , the increase of ˇa was also observed for a barrier film formed in air upon the surface after aging at 185 ◦ C
the aged samples (Table 4). Observation of an increased value of for 6 h. This gives a completed arc due to the oxide layer. We can
ˇa and ˇc suggests that both aging and extrusion are beneficial in therefore conveniently conclude that the aluminum oxide layer on
controlling of both anodic and cathodic reactions. the surface thickens by the aging process. Moreover, the polariza-
tion resistance of aged Al samples with extruded 6 and 10 mm s−1
3.3. Impedance measurements are very close to each other like electrochemical parameter. The
change of the polarization resistance for aged Al, which was taken
Impedance spectra in 0.1 M NaCl for aged and non-aged as the resistance of the surface film, was found by the Circular
aluminum with three different extrusion speeds are shown in Regression analysis of the Volta Lab40 software. The results are
Figs. 4 and 5. As shown in Fig. 4, the impedance curves obtained shown in Table 5. Fig. 6 presented the results of EIS experiments
from non-aged Al samples show uncompleted arcs at higher fre- for (a) non-aged and (b) aged samples. There is a distinct differ-
quencies and Warburg impedance at lower frequencies. At the ence between the electrochemical responses of the samples. The
initial state there is a barrier film formed in air upon the sur- equivalent circuit representing the electrochemical behavior of the
face. This gives an uncompleted arc. The hydrogen evolution and extruded surface shows only one time constant (Fig. 6a). In this
diffusion controlled removal of OH− ions from the surface or diffu- circuit, Rct is the charge transfer resistance, Cdl is the double layer
sion of chloride ions to the surface result a Warburg impedance. capacitance and Rs is the solution resistance. In the case of the elec-
trode passivated by means of aging, the equivalent circuit should
consider two time constants in the corresponding impedance spec-
trum (Fig. 6b). In this circuit, W describes the transmission element.
From results obtained by potentiodynamic polarization and elec-
trochemical impedance spectroscopy, it can be concluded that
increasing anodic aluminum oxide with aging process plays impor-
tant role as a barrier layer toward penetration of Cl− ions.
Aging process improved corrosion resistance of the samples.
Therefore, non-aged samples show the higher corrosion rate than
the aged samples and the icorr of the non-aged samples slightly

Table 5
The change of the polarization resistance of Al.

Extrusion speed (mm s−1 ) Rp ( cm2 )

Aged Non-aged

3 50.18 100.4
6 58.20 47.88
10 56.71 16.19
Fig. 4. The Nyquist diagram of the non-aged Al in 0.5 M NaCl at 25 ◦ C.
1360 A. Aytaç et al. / Materials Chemistry and Physics 130 (2011) 1357–1360

Fig. 6. Equivalent circuits of the studied system (a) non-aged and (b) aged specimens.

increase order of 3, 10 and 6. The 6 mm s−1 extruded samples have References

shown maximum icorr value for both aged and non-aged samples.
However, the behavior might be related to the phases present [1] W. Dixion, 5th International Al Extrusion Technology Seminar, The Aluminum
Association, Washington, 1992, pp. 429–436.
which are similar and more uniformly distributed after extrusion. [2] S. Brusethaug, O. Reiso, E. Ruch, in: J. Masounave, F.G. Hamel (Eds.), Fabrication
The phases can form micro-galvanic cells on the specimen surface, of Particulate Reinforced Metal Composites, ASM International, OH, 1990, pp.
thus enhancing the corrosion attack. Looking at the measured cor- 173–179.
[3] J.E. Selseth, M. Lefstand, SINTEF Report STF20 A 89029, SINTEF, Trondheim,
rosion rates for 6 mm s−1 extrusion speed can be critical and one can Norway, 1989.
assume that 3 mm s−1 is less detrimental than 6 mm s−1 . In addi- [4] M. Lieblich, D.G. Gonzalez, P. Adeva, J. Ibanez, M. Torralba, G.J. Caruna, J. Mater.
tion, from Table 2, the same behavior can be seen for hardness. Sci. Lett. 16 (1997) 726.
[5] Y. Brechet, J.D. Embury, S. Tao, L. Luo, Acta Metall. Mater. 39 (1993)
Like corrosion rate, the 6 mm s−1 extruded samples have shown 1781.
maximum hardness because of the same reason. [6] C.W. Lawrence, P.M. Mummery, J.H. Tweed, J. Mater. Sci. Lett. 12 (1993)
647.
[7] H. Lagace, D.J. Lloyd, Can. Metall. Quart. 28 (1989) 145.
4. Conclusions
[8] M.K. Surappa, J. Mater. Sci. Lett. 12 (1993) 1272.
[9] M. Gupta, R. Sikand, A.K. Gupta, Scr. Metall. Mater. 30 (1994) 1343.
1. Substantially higher surface roughness was observed for non- [10] B. Dutta, I. Samajdar, M.K. Surappa, Mater. Sci. Technol. 14 (1998) 36.
aged samples and higher hardness was found for aged samples. [11] J. Kryze, V. Djafari, D. Francois, Proceedings of the World Congress on Powder
Metallurgy SF2M, vol. III, Paris, France, 1994, pp. 1963–1966.
The extrusion speed does not change the hardness of aged and [12] M. Usta, The influence of thermal treatment on the cast structure of 6xxx alu-
non-aged samples in a greater way. In addition, the hardness minum alloys, Ph.D thesis, Rensselaer Polytechnic Institute, New York, 2001,
increases with aging due to the precipitation hardening. It can pp. 37–46.
[13] R.M. Kelly, P.F. Chaveriat, K.P. Wardlow, S.R. Claves, W.Z. Misiolek, Predictive
be seen that the 6063 Al surface becomes less rough after aging metallographic assessment of extrudability and comparative testing of 6xxx
because the surface of oxide layer that covers the deformation aluminum alloy billets, in: 8th International Al Extrusion Technology Seminar,
surface increases during aging process. vol. 2, Florida, 2004, pp. 39–51.
[14] D. Igor, J. Prgin, The effect of extrusion conditions on the surface properties of
2. The aging process improved the corrosion resistance of alu- AlMgSi (6063) hollow sections, in: 7th International Al Extrusion Technology
minum alloy. Seminar, vol. 1, Illinois, 2000, pp. 261–273.
3. In conclusion, the analysis of potentiodynamic polarization, EIS [15] T. Sheppard, On the relationship between extrusion conditions, mechani-
cal properties and surface acceptability in some hard aluminum alloys, in:
and appropriate equivalent circuit models reveal that the alu- 7th International Al Extrusion Technology Seminar, vol. 1, Illinois, 2000, pp.
minum corrosion is effectively inhibited by the formation of 307–323.
oxide layer. [16] N. Parson, J. Hankin, K. Hicklin, C. Lowett, Comparison of the extrusion perfor-
mance and product characteristics of three structural extrusion alloys: AA6061,
AA6082 and AA6005A, in: 7th International Al Extrusion Technology Seminar,
Acknowledgments vol. 2, Illinois, 2000, pp. 1–13.
[17] H. Matsuoka, Y. Katoh, M. Sakaguchi, T. Oka, M. Hirano, Effect of inter-
The authors thank to O. Faruk Deniz for helping surface rough- metallic compound distribution on extrusion surface in 6061 alloy, in:
7th International Al Extrusion Technology Seminar, vol. 2, Illinois, 2000,
ness measurements, Ahmet Nazım for helping with SEM study, pp. 99–105.
Mr. Ümit Büyükfırat and the FENİŞ Aluminum for their support [18] J.A. Schey, Introduction to Manufacturing Processes, 3rd ed., McGraw-Hill, New
during this study. York, 2000.