Procedia Engineering 81 ( 2014 ) 586 – 591
1877-7058 © 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and peer-review under responsibility of the Department of Materials Science and Engineering, Nagoya University
doi: 10.1016/j.proeng.2014.10.044
ScienceDirect
Available online at www.sciencedirect.com
11th International Conference on Technology of Plasticity, ICTP 2014, 19-24 October 2014,
Nagoya Congress Center, Nagoya, Japan
Microstructure evolution of friction boundary layer during
extrusion of AA 6060
Vidal Sanabriaa,*, Soeren Muellera, Walter Reimersb
a Extrusion Research and Development Center, Technical University of Berlin,Gustav-Meyer-Alle 25, 13355 Berlin, Germany
b Department of Materials Science and Technology, Technical University of Berlin, Ernst-Reuter-Platz 1, D-10587 Berlin, Germany
Abstract
The tribological behavior between the aluminium alloy AA 6060 and the hot working steel 1.2344 has been recently studied
using a new axial friction test for extrusion processes. With this new test, the friction forces as well as the specimens’ plastic
deformations generated during sliding tests can be investigated. Selected specimens tested at high temperature (300-500 °C),
normalized normal stress (ın/kfo=1.5) and high relative speed (0.1, 50 mm/s) have been sectioned and the microstructure on
their mid-planes has been investigated. Light optical microscope analysis of the friction boundary layer revealed highly
stretched grains and a thickness variation of the shear layer from 750 to ȝP depending on testing conditions. Moreover,
EBSD analysis showed a grain refinement in the high shear zone especially at 300 °C with a grain size about ȝP'ynamic
recrystallization was observed at 400 °C and an abnormal grain growth at 500 °C. Hardness measurements revealed a light
hardening effect of 8 % at 300 °C as well as a softening effect of 6 % at 400 °C and 500 °C.
© 2014 The Authors. Published by Elsevier Ltd.
Selection and peer-review under responsibility of Nagoya University and Toyohashi University of Technology.
Keywords: Friction boundary layer; Microstructure evolution; Aluminium alloy AA6060; Axial friction test; Extrusion
1. Introduction
The accurate calculation of the friction force plays an important role in extrusion processes. In the direct
extrusion, the friction represents a great portion of the total extrusion force and therefore it becomes relevant for the
total extrusion power estimation. Moreover, most of the friction force is generated due to the relative movement
and the strong junction between the billet and the container (Mueller, 2004). On the other hand, the friction also
takes place in the die, where multiple variables interact to control the material flow especially in portholes (Valberg,
2014) and multi-hole dies (Mueller et al., 2012).
* Corresponding author. Tel.: +49-30-31472516 ; fax: +49-30-31472503
E-mail address: vidal.sanabria@fzs-berlin.de
© 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and peer-review under responsibility of the Department of Materials Science and Engineering, Nagoya University
587
Vidal Sanabria et al. / Procedia Engineering 81 ( 2014 ) 586 – 591
The combination of variables such as temperature, sliding speed, normal force and surface conditions, defines
the mechanical and chemical interactions generating the sticking and sliding friction mechanisms (Wuttke, 1987).
When sticking friction occurs the contact surfaces remain together because of the strong junction force, whereas the
sub layers of the softer material experience a severe shear deformation. This phenomenon produces a peripheral
shear zone, which has been displayed in extrusion processes using grid pattern techniques. For instance, Kammler
et al. (2012) revealed a thick peripheral shear zone around 10 mm in the border of the billet. Moreover Valberg
(2009) found a thin shear zone in the inlet side of the bearing channel related with the sticking zone. According to
the Tresca friction model, in presence of sticking behavior the friction stress should be equivalent to the shear
strength, and therefore proportional to the flow stress. Moreover, the flow stress in hot working processes is not
only affected by the temperature and the strain rate in accordance with the Zener-Hollomon model, but also by
microstructural changes. Different mechanisms of recovery and recrystallization are present during hot deformation,
especially in metals with high stacking fault energy such as aluminium and it alloys (Humphreys and Hatherly,
2004). Dynamic recrystallization was reported by Kayser et al. (2010), who observed a continually decrease of the
main grain size along the center path of a cylindrical billet during extrusion of AA6060 aluminium alloy. Moreover,
Donati et al. (2013) found strongly elongated grains on the profile surface due to the double effect of the high
deformation firstly in the shear zone and posteriorly in the bearing channel caused by the high friction.
Recently, friction experiments with the aluminium alloy AA6060 have been carried out using a new axial
friction test (Sanabria et al., 2014). The normalized normal stress (ın/kfo=1.5) and temperature (300 °C – 500 °C)
were enough to generate sticking conditions as was also found by Widerøe et al. (2012). The results revealed a thin
aluminium layer on the die, as well as a thin peripheral shear zone in the specimens. The focus of the present work
is the microstructural investigation of the AA6060 aluminium alloy during friction tests. Special emphasis is put on
the grain size evolution and the hardness in the thin boundary layer under high temperature and severe shear
deformation.
2. Experimental procedures
Small cylinders with initial diameter 7.8 mm and length 10.5 mm were extracted from homogenized cast billets
of AA6060 aluminium alloy. In each experiment, the sample was heated up inside a steel liner in 20 minutes and
set up by two counteracting stems with a constant strain rate of 0.0006 s-1 (Fig. 1a). The stems were fixed and the
liner was pulled down 5 mm with a constant slide speed of 0.1 and 50 mm/s, whereas the friction force was
measured (Fig. 1b). Around 90 seconds after each friction test the specimens were quenched in water. The
experiments were carried out at 300 °C, 400 °C and 500 °C and the samples were axially compressed until 3900 N,
1600 N and 760 N respectively, to achieve a normalized normal (radial) pressure around ın/kfo=1.5. Each
specimen was longitudinally sectioned, and the mid-plane (Fig. 1c) polished and etched before analyzing using
light optical microscopy. The mid-planes were posterior electrolytic polished and analyzed with electron
backscatter diffraction (EBSD). Additionally, the Vickers microhardness was measured DWȝPȝP
ȝPȝPȝPDQGȝPfrom the border to the center (normal force 1 N). A detailed description of the
axial friction test was reported by Sanabria et al. (2014).
Fig. 1. Schematic procedure of the axial friction test. a) Set up, b) friction test, c) tested sample.
a) b)
ın
ıa
Sample
Liner
Stem
b)
IJ
10 mm
Ø 8 mm
boundary layer
boundary layer
c)
588 Vidal Sanabria et al. / Procedia Engineering 81 ( 2014 ) 586 – 591
3. Results and discussion
After the friction experiments, the specimens were sectioned and the microstructure was revealed. Fig. 2a shows
the typical friction boundary layer of the studied cylindrical specimens, which were subjected to an axial friction
VWUHVVIJ under sticking conditions. The friction boundary layer (~2.7 mm) is formed by a nondeformed area a and
the shearing layer h. In the area a no shearing is visible, however, microstructural changes may be possible. On the
other hand, in the shearing layer a low and high shear zones are generated due to the sticking friction. Moreover,
the low shear zone is difficult to delimit because it is the transition section between the non-deformed and highly
deformed grains. Nevertheless, the high shear zone is well defined and characterized by a grain refinement as a
consequence of the severe shearing. The plastic deformations on the top and bottom of the specimen are generated
by the tension and compression forces respectively, produced during the friction test. Therefore, the plastic
deformation at the top and bottom areas was neglected for the shearing measurements.
Fig. 2a, 2b and 2c show the friction boundary layers of specimens tested with a high sliding speed of 50 mm/s
at 300 °C, 400 °C and 500 °C. Measurements revealed that the shear zone h increases with the temperature,
probably due to the strength reduction at higher temperatures. Thus, h was 1300 ȝPDW&EXW5ȝPDQG
16ȝPDW&DQG&UHVSHFWLYHO\$QRWKHUH[SHULPHQWVFDUULHGRXWat 0.1 mm/s revealed no significant
variation of h at 400 °C and 500 °C, however a reduction of 40 % was found at 300 °C.
Fig. 2. Microstructure of the entire friction boundary layer after friction test at 50 mm/s and ın/kfo=1.5 (LOM). a) Deformed zones in the
friction boundary layer, b) at 300 °C, c) at 400 °C and d) at 500 °C.
IJ
h
a
a) 4
3
2
1
Low
shear
zone
b)
h
c)
h
d)
h
High
shear
zone
1 mm 1 mm 1 mm
589
Vidal Sanabria et al. / Procedia Engineering 81 ( 2014 ) 586 – 591
Fig. 3. Microstructure of the friction shear layer at the middle of the sample at ın/kfo=1.5 (LOM). a) at 300 °C, 0.1 mm/s, b) at 400 °C,
0.1 mm/s, c) at 500 °C, 0.1 mm/s, d) at 300 °C, 50 mm/s, e) at 400 °C, 50 mm/s and f) at 500 °C, 50 mm/s.
As in the case of the shear layer h at 300 °C, the high shear zone also increases with the sliding speed. Fig. 3a
and 3d show how the high shear zone LQFUHDVHVIURPȝPWRȝPZKHQWKHVOLGLQJVSHHGis varied from
0.1 mm/s to 50 mm/s respectively. Furthermore, the high shear zone remains almost constant with the sliding
VSHHGZLWKYDOXHVRIȝPDQGȝPDW&DQG&UHVSHFWLYHO\7KHUHVXOWVVKRZWKDWh and the high
shear zone have a similar tendency of increasing with higher sliding speeds at 300 °C. On the other hand, h and the
high shear zone are not sensible to the sliding speed at 400 °C and 500 °C. The grain size evolution in the high
shear zone revealed bigger grains at higher temperatures and fine grains at higher sliding speed (Fig. 4). The
smallest grains were found in the specimens tested at 300 °C and at a sliding speed of 50 mm/s (Fig. 4d). The high
strain at 300 °C generates interaction and multiplication of the dislocations, increasing the grain boundary area and
producing a grain refinement (Humphreys and Hatherly, 2004). Moreover, when the same experiment but at
0.1 mm/s is carried out, bigger grains (4 ȝP ZHUH appeared, possible due to dynamic recovery (Fig. 4a). At
400 °C a grain fragmentation occurs possibly due to dynamic recrystallization, generating grain size of 5 ȝPDW
50 mm/s (Fig. 4e). However at 0.1 mm/s ȝPwas measured as a result of sub grain growth during the friction test.
Nevertheless, at 500 °C dynamic recrystallization with discontinuous and abnormal sub grain growth was observed
in the specimens tested at 0.1 mm/s (Fig. 4c) and at 50 mm/s (Fig. 4f). Measurements carried out in a
homogeneous area reveal grain size RIȝPDWPPVDQGȝPDWPPVBecause the grain growth occurs
very fast at high temperature (500 °C), it is possible that the abnormal grain growth (Fig. 4f) was caused by static
recrystallization during the 90 s before quenching (Humphreys and Hatherly, 2004).
Fig. 4. Microstructure of the high shear zone at the middle of the sample at ın/kfo=1.5 (EBSD-Band contrast). a) at 300 °C, 0.1 mm/s, b) at
400 °C, 0.1 mm/s, c) at 500 °C, 0.1 mm/s, d) at 300 °C, 50 mm/s, e) at 400 °C, 50 mm/s and f) at 500 °C, 50 mm/s.
ȝP
ȝP
ȝP
ȝP
ȝP
ȝP
d)
a) b)
e)
c)
f)
300 °C 400 °C 500 °C
a) b)
d) f)
e)
c)
0.1 mm/s 50 mm/s
300 °C 400 °C 500 °C
High
shear
zone
Low
shear
zone
590 Vidal Sanabria et al. / Procedia Engineering 81 ( 2014 ) 586 – 591
The grain size in the zone a was also measured. Though, this zone was not directly subjected to shearing its
grain size was also affected during the friction tests. The zone a is characterized by big grains with dendritic
formed by the cast process. Hence, the grain size of the specimens tested at 0.1 mm/s are: 139, 129 and 172 ȝPDW
300, 400 and 500 °C respectively (Figs. 5, 6 and 7). Additionally, experiments carried out at a sliding speed of 50
mm/s showed a grain size reduction of almost 15 % at 300 and 500 °C, and even 30 % at 400 °C. The grain size
reduction could be due to the increase of dislocation density as a result of higher induced stress at 50 mm/s.
Additionally, this effect could also be caused by reducing the time for grain growth, considering that at 0.1 mm/s
the test is carried out in 50 s.
In order to investigate the possible changes in the mechanical properties especially in the shear layer at different
temperatures and sliding speeds, radial hardness mappings of the tested specimens were carried out. Fig. 5 depicts
how at 300 °C the hardness increases progressively in the shear zone h until about 8 % comparing with the
hardness in the specimen’s center. Moreover, the hardness increases at higher sliding speed at the same
temperature. These results at 300 °C agree with the Hall–Petch relation, observing higher hardness values only
where the fine grains (shear zone) were found.
In the experiments carried out at 400 °C and 500 °C a drastic reduction of grain size was also observed from the
zone a to the high shear zone about 94 % and 90 % respectively (Fig. 6,7). However, despite the grain refinement
in the high shear zone, a hardness reduction about 6 % at 400 and 500 °C was found. Nevertheless, the hardness
reduction in this case is due to the drop of the dislocation density as a result of the dynamic recrystallization at high
temperatures and strain. On the other hand, the hardness values measured at 500 °C were about 25 % higher than at
400 °C possibly due to solid solution strengthening. When the temperature is higher than the solvus temperature
(Tsolvus=400 °C for AA6060) the precipitations and impurity atoms are dissolved into the solid solution.
Therefore, the dislocation movement is restricted by the lattice strain field interactions between dislocations and
the impurity atoms (Garrett, et al., 2004).
Grain size ȝP
Hardness (HV)
Length of the boundary layer (mm)
*UDLQVL]HȝP
Hardness (HV)
Length of the boundary layer (mm)
Fig. 6. Hardness and grain size of the friction boundary
layer at 400 °C.
Fig. 5. Hardness and grain size of the friction
boundary layer at 300 °C.
*UDLQVL]HȝP
Hardness (HV)
Length of the boundary layer (mm)
Fig. 7. Hardness and grain size of the friction boundary layer at 500 °C.
@ 50 mm/s
@ 0.1 mm/s
a Low
shear
High
shear
591
Vidal Sanabria et al. / Procedia Engineering 81 ( 2014 ) 586 – 591
4. Conclusions
The investigation of the friction boundary layer of the AA6060 aluminium alloy under extrusion conditions was
successfully carried out. Small cylindrical specimens were subjected to axial friction tests at different temperatures
(300 – 500 °C) and sliding speed (0.1 – 50 mm/s). After testing the samples were quenched in water to avoid static
recovery and static recrystallization. Microstructural analysis revealed a shear layer, which was divided into the
low and high shear zones. At 300 °C, the shear zone increased from 750 to 1300 ȝP when the sliding speed was
increased from 0.1 mm/s to 50 mm/s. The increase of the shear layer was accompanied by the reduction of the
grain size of more than 95 % in the high shear zone. Small grains of about 4 ȝP and even smaller were found in
the high shear zone at 300 °C. The high plastic deformation could cause the multiplication of the dislocations and
the increase of the grain boundary area, generating the grain refinement. Furthermore, a hardness mapping revealed
an increase of 8 % in the mechanical properties in the highly deformed zone, which is in agreement with the Hall–
Petch relation.
Contrary to the experiments at 300 °C, at higher temperature (400 and 500 °C) the shear layer h was not sensible
to the sliding speed. Due to the strength reduction at higher temperatures, the shear layer was about 1500 ȝP and
ȝP DW and 500 °C respectively. The recrystallization process decreased the dislocation density and
therefore also the hardness, for that reason a reduction about 6 % was found in the high shear zone at 400 and
500 °C. Additionally, the hardness values measured at 500 °C were about 25 % higher than at 400 °C possibly due
to solid solution strengthening.
Acknowledgments
This project (IGF – No: 17202 N/1) was funded through the AiF in the program for the enhancement of
industrial joint research of the Federal Ministry of Economics and Technology of Germany. The authors are
thankful to ZELMI Technical University of Berlin, where the EBSD measurements were carried out.
References
Donati, L., Segatori, A., El Mehtedi, M., Tomesani, L., 2013. Grain evolution analysis and experimental validation in the extrusion of 6XXX
alloys by use of a lagrangian FE code. International Journal of Plasticity 46, 70-81.
Garrett, R.P., Lin, J., Dean, T.A., 2004. An investigation of the effects of solution heat treatment on mechanical properties for AA6xxx alloys:
experimentation and modelling. International Journal of Plasticity 21, 1640-1657.
Humphreys, F.J., Hatherly, M., 2004. Recrystallization and Related Annealing Phenomena. Elsevier, Second edition, Oxford, U.K.
Kayser, T., Klusemann, B., Lambers, H.-G., Maier H.J., Svendsen, B., 2010. Characterization of grain microstructure development in the
aluminium alloy EN AW-6060 during extrusion. Materials Science and Engineering A 527, 6568-6573.
Kammler, M., Hadifi, T., Nowak, M., Bouguecha, A., 2012. Experimental and numerical investigation on metal flow during direct extrusion of
EN AW-6082. Key Engineering Materials 491, 137-144.
Mueller, K., 2004. Fundamentals of extrusion. Giesel Verlag GmbH., Iserhagen, Germany.
Mueller, S., Muehlhause, J., Maier, J., Hora, P., 2012. Experimental and numerical analysis of the friction condition in the die bearing during
aluminium extrusion. Key Engineering Materials 491, 113-119.
Valberg, H., 2009. Analysis of metal flow of aluminium through long chocked die channels. Key Engineering Materials 424, 145-152.
Valberg, H., Nolte, D., Khan, Y., 2014. On the balance of the metal flow in porthole dies with differently sized porthole channels. Key
Engineering Materials. 585, 77-84.
Sanabria, V., Mueller, S., Reimers, W., 2014. A new high speed friction test for extrusion processes. Key Engineering Materials 585, 33-39.
Sanabria, V., Mueller, S., Gall, S., Reimers, W., 2014. Investigation of friction boundary conditions during extrusion of Aluminium and
Magnesium alloys. Key Engineering Materials 611-612, 997-1004.
Widerøe, F., Welo, T., 2012. Conditions for sticking friction between aluminium allow AA6060 and tool steel in hot forming. Key Engineering
Materials. 491, 121-128.
Wuttke, W., 1987. Tribophysik, Reibung und Verschleiss von Metallen. Hanser., Muenchen, Germany.
