Materials and Design 92 (2016) 632–642

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matde s

On the feasibility of a friction-based staking joining method for

polymer–metal hybrid structures
A.B. Abibe a, M. Sônego b, J.F. dos Santos a, L.B. Canto b, S.T. Amancio-Filho a,c,⁎
a
Helmholtz-Zentrum Geesthacht, Centre for Materials and Coastal Research, Institute of Materials Research, Materials Mechanics, Solid State Joining Processes, Geesthacht, Germany
b
Graduate Program in Materials Science and Engineering (PPG-CEM), Federal University of São Carlos (UFSCar), São Carlos, SP, Brazil
c
Hamburg University of Technology, Institute of Polymer Composites, Hamburg, Germany

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

Article history: The increased use of hybrid structures to reduce weight currently faces the limitations of traditional joining
Received 9 December 2015 methods. Consequently there is a niche for development of new joining techniques, which can reduce or over-
Accepted 14 December 2015 come some of the existing limitations. This paper presents for the first time the new Friction-based Injection
Available online 17 December 2015
Clinching Joining technique (F-ICJ), describing the microstructure and changes in local properties of joints be-
tween polyetherimide (PEI) and aluminum alloy 6082-T6. A shear layer around the rotating tool composes a
Keywords:
polymer thermomechanically affected zone (PTMAZ), which presents pores as a result of evolution of gaseous
Processing technologies
Staking
products. The PTMAZ shows decreases of 8% to 12% in local strength compared to the base material, as measured
Polymer–metal structures by microhardness. Ultimate forces of 1419 ± 43 N in lap shear and 430 ± 44 N in cross tensile were achieved for
Hybrid joining F-ICJ joints. These levels are similar to ultrasonic staking joints of the same material combination, but the hollow
Friction joining design of F-ICJ stakes accounts for improved strength-to-weight ratio (18% in lap shear, 21% in cross tensile).
Polyetherimide Although the F-ICJ process currently requires longer cycles (7.5 s) than state-of-the-art ultrasonic staking
(2.8–2.9 s), generated results indicate that the F-ICJ process is a competitive staking joining method with poten-
tial for improvement.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction
There is an increasing trend to reduce vehicle weight in the trans-
portation industry because of environmental issues such as the require-
ment to reduce CO2 emissions [1], and economic concerns such as the
trend of rising fuel prices [2]. The most common strategy for weight re-
duction is the use of advanced lightweight materials such as polymer
composites, engineering thermoplastics and lightweight alloys [3].
This broader selection of materials leads to the use of hybrid structures,
where the dissimilarity among materials creates a design and assembly
challenge for engineers.
The traditional techniques to join polymeric and metallic parts are
mechanical fastening and adhesive bonding. The use of adhesives pro-
motes a continuous contact between the surface of the parts, which re-
sults in structures with uniform stress distribution and good mechanical
performance under tension, compression and shear [4]. However, the
need for extensive surface treatments adds costs and is time consuming.
Moreover, solvent evaporation, which is usually found in thermosetting
adhesive systems, may result in health and safety hazards [5].
⁎ Corresponding author at: Max-Planck Strasse 1, D-21502 Geesthacht, Germany.
E-mail addresses: andre.abibe@hzg.de (A.B. Abibe), mrl.sonego@gmail.com
(M. Sônego), jorge.dos.santos@hzg.de (J.F. dos Santos), leonardo@ufscar.br (L.B. Canto),
sergio.amancio@hzg.de (S.T. Amancio-Filho).
Mechanical fastening is largely used because of its simplicity and ease
to disassemble. The main limitation is the hole, which acts as a stress
concentrator [3] and leads to earlier crack nucleation, which may result
in premature failure. The difficulty in joining dissimilar materials and
the limitations of the traditional techniques to join complex geometries
create a niche for new joining techniques [6].
Staking joining methods are a traditional way to join dissimilar ma-
terials. They are based on the deformation of a polymeric stud, whose
diameter fits into a through-hole of a joining partner. This stud is subse-
quently formed into an anchoring element, which is designated as a
stake and is responsible for the mechanical anchoring of the joint. This
forming step can be performed at room temperature using sheer pres-
sure, which is known as cold staking [7]. Cold staking requires high join-
ing forces and is more efficient with ductile polymers, which do not
break during forming by compression. Staking can also be performed
using various energy sources to soften or melt the stud prior to forming.
Common energy sources are hot air [8], ultrasonic [9], and infrared radi-
ation [10]. The “hot staking” methods generate less residual stresses [7]
and are applicable to almost any thermoplastic. The main applications of
staked joints are in secondary and tertiary structures of automotive in-
dustry, electronic devices, and household items. State-of-the-art staking
is a simple and low-cost technique, whose main limitations are the poor
surface finishing, and the residual stresses and recovery effects in the
case of cold staking [7]. Staking techniques have been studied by several

http://dx.doi.org/10.1016/j.matdes.2015.12.087
0264-1275/© 2015 Elsevier Ltd. All rights reserved.
633 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
authors as an alternative to join hybrid structures. Hahn and Finkeldey
[9] studied ultrasonic staking and hot-air staking for glass fiber rein-
forced polyamide. Yeh, Schott [8] used hot-air staking to join various
combinations of engineering thermoplastics. Beute [10] introduced a
more efficient and fast concept for infrared staking. DeSouza [11] used
design of experiments to create an optimized design guideline for heat
staking for industrial production.
Other advanced technologies for hybrid joining use structures simi-
lar to stakes to create form-locked joints or additional mechanical an-
choring in a structure. Brückner et al. [12,13] presented ultrasonic
upsetting as an alternative to traditional staking, with the ability to pro-
duce hollow stake-like joints with few defects. Gude et al. [14] describe
the thermoclinching process, which is able to create a stake-like struc-
ture to join laminate composites to other materials by punching
through a notch in the pre-heated composite plate. Grujicic et al. [15,
16] introduced the direct-adhesion polymer–metal hybrid concept
using different approaches to improve overmolded hybrid components,
such as millimeter-sized impressions on the metal substrate to improve
mechanical interlocking [17]. Lambiase and Di Ilio [18] and Lambiase
[19] explored traditional clinching to join metals to amorphous poly-
mers. Pre-heating was shown to improve polymer formability and geo-
metrical aspects of the joints.
Hot staking methods are the state of the art, and are still not applied
to load-bearing lightweight structures. A new approach to obtain hot
staked joints is presented in this paper, exploring the use of highly
energy-efficient frictional heating. This paper describes for the
first time the principles of the Friction-based Injection Clinching Joining
(F-ICJ) process. Its feasibility with engineering materials relevant to the
transport industry (AA6082-T6 aluminum alloy and polyetherimide —
PEI) is demonstrated. An analysis of F-ICJ joints is performed in terms
of microstructure, local changes in the processed material, and global
mechanical properties.
2. Experimental approach
2.1. Materials
The polyetherimide (PEI) is a commercial-grade Duratron PEI
U1000, which was supplied by Quadrant Plastics (Switzerland). It is an
amorphous thermoplastic with high mechanical strength and stiffness.
It exhibits excellent thermal stability with glass transition temperature
ranging from 215 to 220 °C. The main applications of PEI are in the air-
craft, medical, packaging, and automotive industry [20].
The metallic partner is the 6082-T6 aluminum alloy (AA6082),
supplied by Aalco Metals (United Kingdom). This material is a
precipitation-hardenable alloy, whose main alloying elements are mag-
nesium (0.6–1.2%) and silicon (0.7–1.3%) [21]. AA6082 is a medium-
strength aluminum alloy with excellent corrosion resistance, good
formability, weldability and machinability. It is typically used in highly
stressed structures such as trusses, bridges, cranes, and transportation
applications [21].
The specimens were machined from 6.35-mm-thick extruded PEI
plaques and 2-mm-thick rolled AA6082 plates. Specimen geometries
are described in Fig. 1. The polymeric part had a 4 mm height and
6 mm diameter stud that protruded from its surface, which was fitted
into the machined 6 mm diameter through-hole on the metallic part.
A surface radius of 0.3 mm was used at the base of the stud. The metallic
part had a chamfered cavity design. Square specimens as shown in
Fig. 1a were used for microstructural analysis and for the investigation
of local properties. Fig. 1b and c show the specimen geometries used
for mechanical testing.
2.2. Joining process
The F-ICJ joints were produced using a cylindrical, non-consumable
tool which was coupled with a high-rotational-speed friction welding
Design
Design
92 (2016)
92 (2016)
632–642
632–642 633
machine (RSM 400, Harms & Wende, Germany). The process parame-
ters were monitored using the RQ-Fuzzy software, which was integrat-
ed with the RSM 400 system. An infrared thermal camera
(ImageIR8800, InfraTec, Germany) was used to monitor the tempera-
ture evolution during joining.
The tool was machined from AISI 316 L stainless steel bars. The tool
design is schematically shown in Fig. 2. A conical-pin tool design was
chosen to heat a sufficient volume of the stud and efficiently deform
it. A conical-pin tool creates hollow stakes, which are ideal for load-
bearing, larger stakes (diameter N 5 mm) [22]. Compared to a flat-tool
design, the conical-pin tool can better distribute the generated heat in
the stud volume while maintaining short cycles. In addition, it helps
pushing the material outwards to improve the cavity filling of the
F-ICJ joints [23,24].
Energy input at each joining condition was calculated as the applied
mechanical work (Emech), using Eq. (1) [25–27]. The equation is com-
posed of a frictional (Ef) and a deformational component (Ed). The fric-
tional contribution to mechanical work generation is the product of the
torque M and rotational speed ω, integrated over the time t. The defor-
mational contribution is the product of the axial force F and the defor-
mation rate υ, integrated over time. The integrals result for each
component in the total torque Mtotal and the displacement Δx.
Z Z
t t
E mech ¼ E f þ Ed ¼ Mω dt þ Fυ dt ¼ Mtotal ω þ FΔx ½ J ð1Þ
t0 t0
A torque sensor (model 9049, Kistler, Switzerland) was used to mea-
sure torque response during the joining process, so that Mtotal could be
determined. Rotational speed ω and axial force F were set up and mon-
itored at the controlling interface of the RSM400 system. The displace-
ment Δx can be determined by the designs of the specimens (Fig. 1)
and conical-pin tool (Fig. 2b).
2.3. Characterization of the joints
The joint microstructure was investigated with the specimens
shown in Fig. 1a. Samples were cut 1 mm from its center, mounted in
low-cure temperature epoxy resin, ground and polished according to
standard materiallographic sample preparation procedures. Microstruc-
tural analysis was performed using reflected-light optical microscopy
(RLOM) transmitted-light optical microscopy (TLOM, thin sections of
about 1 mm), and scanning electron microscopy with secondary elec-
trons (SEM, model Quanta FEG 650, FEI, USA). The metallic partner
was electrolytically etched with Barker reagent (5 mL HBF4 in 200 mL
distilled water, U = 30 V for 120 s) to reveal the grain structure.
Local mechanical properties of the joint were investigated through
microhardness indentations on embedded joint cross sections from
the specimens shown in Fig. 1a. The indentation testing procedure
was performed using a Zwick ZHV (Zwick Roell, Germany) microhard-
ness tester according to ASTM: E384 using an indentation load of
0.495 N, indentation distance of 200 μm, and a holding time of 15 s.
Changes in physical–chemical properties were investigated by com-
paring the behavior of PEI base material and joined PEI when submitted
to differential scanning calorimetry (DSC) and thermogravimetric anal-
ysis (TGA) tests. DSC tests were carried out with a DSC 200 F3 Maia
(Netzsch, Germany) under nitrogen atmosphere (20 mL/min). Two
heating cycles from 30 °C to 250 °C at a heating rate of 20 °C/min
were used. TGA tests were carried out with a TG 209 F3 Tarsus (Netzsch,
Germany) under nitrogen atmosphere (20 mL/min). A heating rate of
20 °C/min was used in the range from 25 °C to 800 °C. For both tests
samples were extracted with a scalpel from the process-affected re-
gions, using a mass of 5.0 ± 0.1 mg. Three runs were performed for
each joining condition.
The global mechanical properties of the joints were evaluated
through single lap shear testing on overlap joint specimens (Fig. 1b)
using a Zwick 1478 (Zwick Roell, Germany) universal testing machine
634 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
Design
Design
92 (2016)
92 (2016)
632–642
632–642 634

Fig. 1. Configuration of F-ICJ specimens: (a) joint for microstructural analyses and local properties testing; (b) overlap specimen for the lap shear strength test and (c) cross tensile
specimen.

at room temperature and a crosshead speed of 2 mm/min; the test
procedure was based on ASTM: D5961. The clamps were aligned to
the longitudinal axis of the specimen. A digital image correlation system
(DIC) – Aramis (GOM, Germany) – was used to monitor the sample
strain distribution during the tests.
The mechanical strength of the stake head was evaluated through
cross tensile testing (specimens from Fig. 1c) using a Zwick 1478
(Zwick Roell, Germany) universal testing machine at room temperature
and a crosshead speed of 2 mm/min; the test procedure was based on
DIN ISO 14272 [28]. In this configuration, the top metallic plate was
fixed, whereas the lower polymeric plate was pulled away.
3. Principles of Friction-based Injection Clinching Joining (F-ICJ)
F-ICJ is a new approach to the ICJ technique to join hybrid structures
of thermoplastics and metals [24]. ICJ processes produce joints by
heating and deforming a polymeric stud that is inserted into a
through-hole to create a stake that mechanically anchors the joining

Fig. 2. Design of the conical-pin tool: (a) tool features; (b) tool geometry.
635 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
Design
Design
92 (2016)
92 (2016)
632–642
632–642 635

partners. Cavities are used inside of the through-hole to improve the

mechanical interlocking between the parts [29]. The originally devel-
oped technology of ICJ uses electrical heating (E-ICJ) as an energy source
[29,30], while the current method is based on frictional heating (F-ICJ)
[24], which is faster and more energy-efficient.
A schematic drawing of the F-ICJ process in its basic configuration is
shown in Fig. 3. In Fig. 3a, the parts are aligned with the tool, and the
thermoplastic stud is inserted into the cavity of the metallic partner.
The rotating tool approaches the stud (Fig. 3b) and presses against it
to generate frictional heat and soften or melt the stud (“stud meltdown”
stage of the friction phase — Fig. 3c). The stud meltdown stage creates
the first softened/melted polymer volumes and is finished as the tool
reaches its final axial position. A “dwell time” stage of the friction
phase follows (Fig. 3d), in which the tool rotates in the same axial posi-
tion, generating viscous heating in the softened/melted polymer. The
dwell time stage reduces the polymer molten viscosity, which improves
the cavity filling by facilitating material flow. At the end of the friction
phase, the rotational speed is decelerated until stopping, whereas fur-
ther axial pressure is applied (consolidation phase, Fig. 3e). The consol-
idation under pressure is used to avoid relaxation effects during cooling,
which in turn would lead to dimensional changes. Following this, the
Fig. 4. Schematic monitoring diagram of tool displacement, rotational speed, and pressure
tool retracts and the staked joint is formed (Fig. 3f).
during a typical joining cycle, describing the process phases.
In the F-ICJ process, it is possible to use various tool designs to create
different stake geometries. For example, a conical-pin tool produces a
hollow stake, whereas a flat tool creates a solid stake with a flat surface.
The tool design should be chosen according to the thermal and rheolog- phases based on a schematic monitoring diagram of a typical joining
ical properties of the polymer, the geometries of the joining partners, cycle.
and their hole clearances. The metallic through-hole is designed with Fig. 4 shows the typical curves of rotational speed (blue triangles),
cavities such as threads and/or a chamfer to improve the mechanical pressure (green disks) and axial tool displacement (black squares) of
interlocking of the polymer and increase the contact surface of the a typical F-ICJ joining cycle. When the spindle reaches the set value of
stake with the metallic part. In an optimized process, the polymeric rotational speed it moves the tool towards the stud applying pressure.
stud volume efficiently fills these cavities increasing the mechanical an- During a short solid–solid friction period no tool displacement towards
choring, which results in an improved joint mechanical performance the stud is measured. With the increase of temperature the stud softens,
[31]. and the tool displacement rate increases until it is constant during the
stud meltdown stage of the friction phase. The stud meltdown stage
3.1. Process phases ends when the tool displacement is constant, which corresponds to
complete meltdown of the stud. The friction phase continues during
In a simple configuration, the F-ICJ process can be divided into two the dwell time stage. During the dwell time the tool does not move fur-
main phases. The friction phase (with “stud meltdown” and “dwell ther towards the joining partners, but it still rotates and presses against
time” stages), which is characterized by the tool rotation and the appli- the softened/melted polymer, generating further heating. Rotational
cation of axial pressure, and the consolidation phase with equal or speed is kept constant at the set value during the friction phase, but is
higher axial pressure without tool rotation. Fig. 4 depicts the F-ICJ decreased to zero at its end, when the consolidation phase starts. Axial

Fig. 3. Steps of the F-ICJ process: (a) positioning of the polymeric stud into the metallic cavity; (b) approach of the rotating tool; (c) stud meltdown stage of the friction phase; (d) dwell
time stage of the friction phase; (e) consolidation of the polymer under pressure in the metallic cavity; (f) staked joint.
636 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
Design
Design
92 (2016)
92 (2016)
632–642
632–642 636

pressure is maintained (or increased) during cooling in the consolida-
tion phase to minimize dimensional recovery. After the consolidation
phase the process ends with the retraction of the tool.
3.2. Controlling parameters and variables of the process
The main controlling parameters of F-ICJ are rotational speed (RS),
frictional time (FT), frictional pressure (FP), consolidation pressure
(CoP), and consolidation time (CoT). These parameters are indicated
in Fig. 4. In the friction phase, RS, FT, and FP are active in heat generation.
During the consolidation phase, CoP and CoT define the final shape of
the stake and are important to reduce the recovery effects, polymer
shrinkage, and avoid adherence of non-consolidated polymer volumes
on the tool surface [7]. In the current configuration, PEI/AA6082-T6
joints were successfully obtained within the following processing win-
dow: rotational speed of 7500–20,000 rpm, frictional pressure of 0.2–
0.5 MPa, frictional time of 2500–5000 ms, consolidation pressure of
0.6–1.0 MPa, and consolidation time of 2500–5000 ms. Joint formation
was studied in this process window according to the working parameter
range of the joining equipment.
These controlling parameters influence the energy input and thus
the polymer flow during stake formation. Several process variables –
such as stud meltdown, stud meltdown rate and heating time (HT in
Fig. 4) – and microstructural variables – volume of process-affected re-
gions, volume of porosities – are therefore dependent on the choice of
controlling parameters. Detailed descriptions of these variables and
how they can be influenced by the process parameters can be found
elsewhere [32].
4. Results and discussion
4.1. Joint microstructure
Fig. 5a shows the surface view of a typical PEI and AA6082-T6 F-ICJ
joint. Fig. 5b is a schematic drawing of the cross-sectional cut from (a),
displaying the microstructural zones usually present in hollow-stake
F-ICJ joints produced with a conical-pin tool. A transmitted-light micro-
graph of a thin section of such a joint is shown in Fig. 5c. Details of the
polymer–polymer and metal–polymer interfaces are displayed in
Fig. 5d and e, respectively.
The rotating tool when brought into contact with the polymer
creates a shear layer around the tool, similar to the ones observed in
Friction Stir Welding (FSW) of thermoplastics [33–35]. In the shear
layer, frictional heat and shearing soften/melt the polymer and

Fig. 5. (a) Surface view of a PEI/AA6082 F-ICJ joint; (b) schematic view of the cross-section from (a), showing typical microstructural zones of an F-ICJ joint with a hollow stake produced
using a conical-pin tool; (c) transmitted-light micrograph of a joint (TLOM); (d) detail of the polymer–polymer interfaces in the PEI stake–the PTMAZ–PHAZ border is seen as a dark line
across the stake diameter; the PHAZ–BM theoretical boundary is defined by a dotted line (TLOM); (e) polymer–metal interface at the chamfer cavity. Overlay of RLOM (PEI) and polarized
RLOM (AA6082, Barker etchant). Joint processed with RS: 7500 rpm; FT: 2500 ms; FP: 0.5 MPa; CoT: 5000 ms; CoP: 1.0 MPa.
637 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
promote material flow. The shear layer is therefore a polymer
thermomechanically-affected zone (PTMAZ), as also described in
other joining processes [36–38]. Its extension can be defined by the
dark-line boundary visible in the polymeric stake (Fig. 5c, detail in
Fig. 5d). The thermomechanical treatment at this volume during the
F-ICJ process affects the microstructure and properties of the polymer
[39]. Depending on the polymeric material, temperatures, and the strain
rates, effects such as recrystallization, chain reorientation, and thermal
degradation can occur in the polymer.
Adjacent to the PTMAZ there is a volume of polymer, which does not
flow but is affected by heat conducted from the PTMAZ. In welding/join-
ing processes this zone is traditionally described as the polymer heat-
affected zone (PHAZ, theoretical boundary marked by a dotted line in
Fig. 5d) [36,37]. The temperature at the PHAZ achieved during the join-
ing process is below the polymer glass transition (Tg) or crystalline
melting (Tm) temperature. The mild heating in this volume may cause
effects such as microcracking due to thermal stresses, and/or local
strengthening due to physical aging [26,39]. The interface between the
PTMAZ and the PHAZ can be noticed in Fig. 5c and d, respectively, by
the shape of the shear layer, seen as a dark line across the stake. In
other processes, the PTMAZ–PHAZ interface is visible through sharp dif-
ferences in the microstructure due to material flow, such as partial fiber
breakage in FSW of polypropylene composites [40] and residual stresses
(by photoelasticity) in FSW of poly(ethylene-terephthalate-glycol) [35]
and poly(methyl methacrylate) [33]. The extension of the PHAZ and its
boundary with the unaffected volume of the base material (BM) is nota-
bly difficult to define because most effects only occur in the macromo-
lecular scale [32]. Moreover, in low-thermal-conductivity materials
such as polymers, the heat flow from the PTMAZ can only affect small
volumes.
In cases where the tool-feeding rate is too fast (high frictional pres-
sures), a solid volume directly below the PTMAZ can experience cold
forming. The hydrostatic pressure applied by the tool over the soft-
ened/melted volume compresses these solid volumes above the yield-
ing point, creating topological defects which increase free volume, and
Fig. 6. (a) RLOM micrograph of the cross section of a PEI/AA6082 F-ICJ joint showing volumetric flaws in the PTMAZ (RS: 12,000 rpm; FT: 5000 ms; FP: 0.3 MPa; CoT: 5000 ms; CoP:
0.7 MPa); (b) SEM images showing smooth internal walls of the pores, which can be an indication of evolution of gaseous products from partial thermal degradation [42,45];
(c) thermographs of a specimen prior to joining (left-hand side) and at the end of the friction phase (right-hand side), showing the peak temperature at the PTMAZ (shear layer) close
to 380 °C; (d) F-ICJ processing temperature
andand
Design
Design
92 (2016)
92 (2016)
632–642
632–642 637
therefore change the material properties [39,41]. This volume is defined
as plastically-deformed zone (PDZ) [32], but since it only occurs in spe-
cific conditions it is not further investigated in this paper.
Thermomechanical treatment at the shear layer (PTMAZ) of PEI/
AA6082 F-ICJ joints tends to generate volumetric flaws in this volume.
These are observed as porosities along the PTMAZ volume (Fig. 6a).
Pores in polymer joints are commonly associated with the evolution of
structural water, entrapment of air, products of thermal degradation,
or differential thermal expansion along the affected volume [42,43].
PEI parts used in this work were not dried prior to joining and contained
0.48 ± 0.00% structural water in weight, therefore structural water may
have vaporized and constituted part of the pores. Air entrapment is an
unlikely cause for pores in F-ICJ, because of the centrifugal forces in-
volved in the shear layer. The hypotheses of thermal degradation and
differential thermal expansion were thoroughly studied by Tan et al.
for laser-joined PA6-CF/steel joints [42]. It was shown that porosities
originating from differential thermal expansion present irregular
rough inner walls, whereas evolution of gaseous products from thermal
degradation creates pores with smooth inner walls. For PEI/AA6082 F-
ICJ joints the porosities show rather smooth inner walls (SEM images,
Fig. 6b), supporting the idea that they are a product of gaseous products.
The temperatures achieved in the PTMAZ during the F-ICJ process dur-
ing the friction phase are at the range of 250–380 °C [32], as shown by
the thermograph in Fig. 6c. This range of temperature is far below the
onset of decomposition for PEI (TGA curve, Fig. 6d); nevertheless, deg-
radation by chain scission was detected in PEI joined by F-ICJ by Sônego
et al. [44] through size exclusion chromatography (SEC). Possible gas-
eous products result from the evolution of low-molar-mass compounds
at the range of 250 °C to 400 °C, such as cyclical monomers and 1,3-
phenylenebisphtalammide [45]. Mechanisms of thermal degradation
for PEI may also be activated by structural water, generating cyclical
monomers and other small hydrocarbons [46]. Therefore, the pores in
PEI/AA6082 F-ICJ joints are believed to be mostly caused from the evo-
lution of structural water and gaseous products from initial stages of
thermal degradation.
range compared to PEI's decomposition behavior (TGA) [32].
638 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
Design
Design
92 (2016)
92 (2016)
632–642
632–642 638

4.2. Local changes in the polymeric stake

Thermomechanical effects on the polymeric material can locally

modify its properties. Fig. 7 shows a microhardness distribution map
(left-hand side) overlaid with the micrograph (right-hand side) of an
F-ICJ joint.
The microhardness analysis provided additional information on the
polymer microstructural zones created by the F-ICJ process. For the
joint in Fig. 7, two zones could be defined in the polymer partner,
which correspond to the PTMAZ and BM zones. Although a PHAZ is ex-
pected to be present beyond the boundary of the PTMAZ, the indenta-
tions could not define an intermediate region between the PTMAZ and
BM. The PTMAZ–PHAZ and PHAZ–BM boundaries and their properties
can possibly be defined by nanoindentation testing, or chemical struc-
ture mapping (by Raman or infrared spectroscopy). The lowest micro-
hardness values are observed at the thermomechanically-affected
zone (PTMAZ, 226 ± 8 MPa), which is the region that was most affected
by the process. Adjacent to the PTMAZ is an unaffected volume with
Fig. 8. Average local strength (microhardness) at the PTMAZ for different levels of energy
base material properties (BM, 244 ± 5 MPa). Other processing condi-
input Emech, and microhardness range of the base material (gray rectangle). The linear
tions with energy input (Emech, from Eq. (1)) ranging from 325 J to fitting (dashed line) shows no direct correlation between the variables.
2489 J result in local hardness at the PTMAZ of 212 MPa to 226 MPa
(Fig. 8), which represent reductions of 8% to 12% in relation to the BM
region. In Fig. 8 it is seen that the microhardness does not change signif- received PEI, indicating that no extensive degradation occurs. This result
icantly with changes in Emech; these results show that the mechanism of indicates that, although partial degradation is present [44] in F-ICJ
strength reduction in the PTMAZ seems not to be depend on the applied joints, the changes observed in local mechanical properties are probably
energy input. not a result of thermal degradation but a result of less dense chain pack-
The change in microhardness in the process-affected volumes com- ing, which creates more free volume in the PTMAZ and reduces local
pared to the base material may be associated to physical chemical ef- strength [39,50].
fects such as changes in chain packing density (free volume) because
of alterations at the molecular scale such as chain rotation/reorientation 4.3. Global mechanical properties
[47] or partial degradation by chain scission [48,49]. These topological
changes leading to increased free volume occur because of the F-ICJ joints were subjected to single lap shear testing. For overlap
thermomechanical treatment and different cooling rates that were im- joints of materials with different stiffness (or thickness), the forces
posed by the process [39,50]. Degradation in thermomechanical pro- cause a secondary bending moment, as shown in Fig. 10a. The secondary
cesses is a result of the shear and heat imposed on the polymer. Low bending creates a stress concentration on the stake head (orange
levels of degradation were shown by Amancio-Filho [26]) and Oliveira square), while the pulling forces concentrate stresses at the base of
[51]) not to deteriorate the global mechanical performance of the joints. the stake (red circle) [32]. In order to analyze the mechanical behavior
However, excessive degradation could significantly affect the micro- of the PEI/AA6082 F-ICJ joints under lap shear, DIC measurements of
structure of polypropylene laser welds [48] and the joint strength of local Von Mises strain distribution (Fig. 10b) are associated with stages
ultrasonically-welded Spectra fabric (UHMWPE) [52]. of a typical force–displacement curve for a sound F-ICJ joint (Fig. 10c).
Thermal degradation of the joined PEI can be qualitatively investi- The force–displacement curve first shows an elastic behavior (Stage I).
gated by comparing physical–chemical properties of the PTMAZ of the As the displacement increases, the stud rotation due to secondary bend-
joints with those of the as-received PEI. Fig. 9 shows the glass transition ing is prevented by the stake head, and the strain is concentrated around
temperatures (a) and onset of decomposition (b) of the base material the stake base (Stage II). During Stage II, a radial crack nucleates and prop-
compared to two F-ICJ joints of different energy input levels. Processing agates around the stake head (orange square and arrow, Fig. 10a). As this
condition C1 (sparse blue crosshatching, 1415 ± 47 J) generated 52% crack reaches a critical size, the stake head stops contributing to load bear-
more mechanical work than condition C2 (dense green crosshatching, ing and a decrease in force is observed (Stage III). At this point there is an
930 ± 23 J). Nevertheless, the PTMAZ of neither joints presented signif- intensification of the secondary bending effect, as seen by the increase in
icant changes in Tg or onset of decomposition when compared to as- local strain through DIC (green and yellow areas in Fig. 10b, Stage III). At

Fig. 7. Juxtaposition of the microhardness distribution map and micrograph (RLOM) for a typical PEI/AA6082 F-ICJ joint. The boundary of the PTMAZ is visible through the change in local
strength. Joint processed using RS: 8000 rpm; FT: 5000 ms; FP: 0.3 MPa; CoT: 2500 ms; CoP: 0.7 MPa.
639 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
Design
Design
92 (2016)
92 (2016)
632–642
632–642 639

shows typical force–displacement curves for two failure modes ob-

Fig. 9. Comparison of physical–chemical properties of as-received PEI and PTMAZ material
from F-ICJ joints: (a) glass transition temperature (DSC, second heating); (b) onset of de-
composition (TGA).
Stage IV, the load is sustained by the bearing of the stake on the internal
cavities of the joining partner; the strain is intensified by the secondary
bending moment, and is now redistributed along the width of the poly-
meric plate. A crack propagates through the thickness of the base plate
(red circle and arrow, Fig. 10a), and it fails as a result from secondary
bending (Stage V), as displayed in Fig. 10d.
Cross tensile testing was performed to evaluate the mechanical
strength of the stake head. After the specimen was fixed, the polymer
plate was pulled, and the metal plate remained stationary. Fig. 11a
served in cross-tensile testing of F-ICJ joints.
An approximately linear behavior is observed in the curves of
Fig. 11a. This is associated with the polymer plate bending during the
test as shown in Fig. 11b. The effect creates stresses at the stake base
(red circles, Fig. 11b) and at the stake head (orange squares, Fig. 11b).
These are competing failure mechanisms in cross tensile testing, with
head pull-out being facilitated by flaws in the stake head [32]. The bend-
ing causes nucleation and fast propagation of cracks in the stake base re-
gion through the plate thickness (Fig. 11b, red lines), which leads to
base plate bending failure (Fig. 11c), without apparent damage to the
stake head. However, porosities as shown in Figs. 5 and 6 may influence
crack behavior (Fig. 11b, orange lines), leading to a head pull-out failure
(Fig. 11d).
An initial benchmarking study was carried out to assess the mechan-
ical performance of joints produced by the new F-ICJ technique in com-
parison to well-established technologies. For this, the same materials
and specimen geometries were joined by F-ICJ and ultrasonic staking,
and the microstructures (one replicate) and mechanical performances
(three replicates) of the joints were compared. The microstructures of
the joints are shown in Fig. 12a. While F-ICJ creates a hollow stake by
using the conical-pin tool, the ultrasonically-staked joints were created
with a knurled tool and leading to solid stakes. In both cases joints flush
to the aluminum surface were obtained. Fig. 12b compares the mechan-
ical performance of these joints in lap shear and cross tensile configura-
tions. Both joints achieved similar mechanical performance. In terms of
strength-to-weight ratio F-ICJ joints present an improvement due to
their lighter design (Fig. 12c). However, in the current stage of develop-
ment, F-ICJ joining cycles are longer than ultrasonic staking. For the
comparison the F-ICJ joints were produced in 7.5 s, while the ultrasonic
staking joints were produced in 2.8–2.9 s, including holding time.

Fig. 10. Single lap shear testing of an PEI/AA6082 F-ICJ joint: (a) schematic drawing of the main forces acting on the joint, and crack nucleation and propagation points; (b) Von Mises strain
distribution at the specimen surface during stages of the test; (c) typical force–displacement curve; (d) failure by secondary bending. Joint processed using RS: 8000 rpm; FT: 5000 ms; FP:
0.3 MPa; CoT: 5000 ms; CoP: 1.0 MPa.
640 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
Design
Design
92 (2016)
92 (2016)
632–642
632–642 640

Fig. 11. Cross tensile testing and the observed failure mechanism for sound PEI/AA6082 F-ICJ joints: (a) typical force–displacement curves during cross tensile testing. Head pull-out joint
processed using RS: 8000 rpm; FT: 5000 ms; FP: 0.5 MPa; CoT: 5000 ms; CoP: 0.7 MPa. Base plate bending joint processed using RS: 12,000 rpm; FT: 2500 ms; FP: 0.5 MPa; CoT: 2500 ms;
CoP: 1.0 MPa; (b) schematic drawing of the crack nucleation points and propagation paths for each failure mode; (c) surface view of the base plate bending failure; (d) surface view of the
head pull-out failure.
5. Concluding remarks polymer (PTMAZ). A heat affected zone (PHAZ) is present adjacent to
the PTMAZ; however, in this study its extension and properties could
This paper presented the primary features of the new Friction-based not be identified. Additional tests must be done to characterize the
Injection Clinching Joining technique (F-ICJ) and demonstrated its feasi- PHAZ. Volumetric flaws such as pores are generally present in the
bility for hybrid joints of polyetherimide (PEI) and 6082-T6 aluminum PTMAZ, and were showed to be a product of evolution of gaseous prod-
alloy. The process phases, joining mechanisms, and general principles ucts from initial steps of polymer thermal degradation.
of the F-ICJ technology were explained for the first time in the literature. Differences in the local mechanical properties at the process-
The microstructural changes related to the process were described, and affected polymer volumes were identified using microhardness testing.
associated changes to local mechanical and physical chemical properties A decrease in local strength in the PTMAZ in relation to the base material
were explored. The mechanical behavior of typical F-ICJ joints was de- is observed, but the decrease appears to be independent of the energy
scribed, and a comparison to state-of-the-art staking was carried out. input. As shown by thermal analyses (DSC and TGA), physical chemical
F-ICJ joints are produced by means of frictional heating and axial properties of the PTMAZ material do not differ from as-received PEI.
pressure, controllable by the choice of process parameters. The generat- Therefore the local strength reduction seems not to be a product of ex-
ed thermomechanical work softens the polymeric stud, creating a shear tensive thermal degradation, but of less-packed chain configuration
layer around the rotating tool in which the polymer flows to create a with more free volume.
stake. Cooling under pressure is used in a consolidation phase. During The global mechanical performance of the joints was evaluated
a typical F-ICJ process no significant microstructural changes are present using single lap shear and cross tensile tests. F-ICJ joints tested in lap
in the metal partner; on the other hand, the shear layer around the ro- shear fail as a result of stress concentration related to secondary bend-
tating tool constitutes a thermomechanically affected zone in the ing. Cross tensile failure of sound joints occurs by base plate bending,

Fig. 12. Comparison of F-ICJ to ultrasonic staking of PEI/AA6082 joints: (a) overview of the microstructure; (b) ultimate lap shear and cross tensile forces; (c) strength-to-weight ratios. F-
ICJ processing conditions RS: 7500 rpm; FT: 2500 ms; FP: 0.5 MPa; CoT: 5000 ms; CoP: 1.0 MPa.
641 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
which shows the high strength of the stake that was formed in F-ICJ.
However, porosities might weaken the stake head, effectively changing
the failure behavior to head pull-out.
Friction-based Injection Clinching Joining technique (F-ICJ) has been
shown to be a promising joining technology for polymer–metal hybrid
structures. A comparison to state-of-the-art ultrasonic staking indicates
that alternative designs, such as hollow stakes combined with cavities in
the metallic part as used for F-ICJ, have potential to be applied industri-
ally. Results of this comparison showed that F-ICJ joints had ultimate lap
shear (1419 ± 43 N) and cross tensile (430 ± 44 N) forces similar to ul-
trasonic staking joints (1370 ± 303 N and 405 ± 50 N, respectively), but
improved strength-to-weight performance (18% in lap shear, 21% in
cross tensile) due to the hollow profile of the formed stake. Neverthe-
less F-ICJ has still longer joining cycles (7.5 s) than ultrasonic staking
(2.8–2.9 s) in this comparison. Therefore, although F-ICJ provides im-
proving key joint characteristics such as esthetics and good mechanical
performance, further process development is needed to achieve com-
petitive cycle times.
Acknowledgments
The authors acknowledge the financial support that was provided
for the Young Investigator Group “Advanced Polymer–Metal Hybrid
Structures” (grant number VH-NG-626) by the Helmholtz Association
Germany, FAPESP (Brazil) for the MSc scholarship to M. Sônego (grant
number 2013/26293-4), and CNPq (Brazil) for the productivity research
grant to Leonardo B. Canto (grant number 304169/2014-5).
References
[1] F. Cuenot, CO2 emissions from new cars and vehicle weight in Europe; how the EU
regulation could have been avoided and how to reach it? Energ Policy 37 (2009)
3832–3842, http://dx.doi.org/10.1016/j.enpol.2009.07.036.
[2] J.R. Duflou, J. De Moor, I. Verpoest, W. Dewulf, Environmental impact analysis of
composite use in car manufacturing, CIRP Ann. Manuf. Technol. 58 (2009) 9–12,
http://dx.doi.org/10.1016/j.cirp.2009.03.077.
[3] S.T. Amancio-Filho, J.F. dos Santos, Joining of polymers and polymer–metal hybrid
structures: recent developments and trends, Polym. Eng. Sci. 49 (2009)
1461–1476, http://dx.doi.org/10.1002/pen.21424.
[4] C.S. Adderley, Adhesive bonding, Mater. Des. 9 (1988) 287–293, http://dx.doi.org/
10.1016/0261-3069(88)90006-4.
[5] T.A. Barnes, I.R. Pashby, Joining techniques for aluminium spaceframes used in auto-
mobiles: part II — adhesive bonding and mechanical fasteners, J. Mater. Process.
Technol. 99 (2000) 72–79, http://dx.doi.org/10.1016/S0924-0136(99)00361–1.
[6] R.W. Messler Jr., The challenges for joining to keep pace with advancing materials
and designs, Mater. Des. 16 (1995) 261–269, http://dx.doi.org/10.1016/0261-
3069(96)00004–0.
[7] J. Rotheiser, Joining of Plastics: Handbook for Designers and Engineers, Carl Hanser
Verlag, Munich, 1999.
[8] H.J. Yeh, C.L. Schott, J.B. Park, Experimental study on hot-air cold staking of PC, PC/
ABS and acetal samples, SPE ANTEC 1998, Society of Plastics Engineers, Atlanta,
USA 1998, pp. 1078–1083.
[9] O. Hahn, C. Finkeldey, Ultrasonic riveting and hot-air-sticking of fiber-reinforced
thermoplastics, J. Thermoplast. Compos. Mater. 16 (2003) 521–528, http://dx.doi.
org/10.1177/089270503032852.
[10] S. Beute, Infrastake — staking at the speed of light, SPE ANTEC 2009, Society of Plas-
tics Engineers, Chicago, USA 2009, pp. 1723–1727.
[11] R. DeSouza, Characterization and Guideline Development of the Heat Staking Pro-
cess (Bachelor Thesis), University of Toronto, Canada, 2009.
[12] E. Brückner, S. Friedrich, M. Gehde, U.M. Endemann, S. Motshev, Material and pro-
cess influences during the riveting of technical plastics, Fügen von Kunststoffen 2
(2013) 98–103.
[13] E. Brückner, S. Friedrich, M. Gehde, T. Fischer, G. Johannes, A. Schneider, Ultrasonic
upsetting — innovative process strategy to join hybrid material combinations, Join.
Plast. - Fügen Kunststoffen 9 (2015) 40–47.
[14] M. Gude, W. Hufenbach, R. Kupfer, A. Freund, C. Vogel, Development of novel form-
locked joints for textile reinforced thermoplastices and metallic components, J.
Mater. Process. Technol. 216 (2015) 140–145, http://dx.doi.org/10.1016/j.
jmatprotec.2014.09.007.
[15] M. Grujicic, V. Sellappan, M.A. Omar, N. Seyr, A. Obieglo, M. Erdmann, et al., An over-
view of the polymer-to-metal direct-adhesion hybrid technologies for load-bearing
automotive components, J. Mater. Process. Technol. 197 (2008) 363–373, http://dx.
doi.org/10.1016/j.jmatprotec.2007.06.058.
[16] M. Grujicic, V. Sellappan, G. Arakere, N. Seyr, M. Erdmann, Computational feasibility
analysis of direct-adhesion polymer-to-metal hybrid technology for load-bearing
body-in-white structural components, J. Mater. Process. Technol. 195 (2008)
282–298, http://dx.doi.org/10.1016/j.jmatprotec.2007.05.016.
Design
Design
92 (2016)
92 (2016)
632–642
632–642 641
[17] M. Grujicic, V. Sellappan, G. Arakere, N. Seyr, A. Obieglo, M. Erdmann, et al., The po-
tential of a clinch-lock polymer metal hybrid technology for use in load-bearing au-
tomotive components, J. Mater. Eng. Perform. 18 (2009) 893–902, http://dx.doi.org/
10.1007/s11665-008-9325-2.
[18] F. Lambiase, A. Di Ilio, Mechanical clinching of metal–polymer joints, J. Mater. Pro-
cess. Technol. 215 (2015) 12–19, http://dx.doi.org/10.1016/j.jmatprotec.2014.08.
006.
[19] F. Lambiase, Joinability of different thermoplastic polymers with aluminium AA6082
sheets by mechanical clinching, Int. J. Adv. Manuf. Technol. (2015) 1–12, http://dx.
doi.org/10.1007/s00170-015-7192-1.
[20] S. Carroccio, C. Puglisi, G. Montaudo, Polyetherimide, in: S. Thomas, P.M. Visakh
(Eds.), Handbook of Engineering and Specialty Thermoplastics, Scrivener Publishing
LLC, Salem, MT 2012, pp. 79–110.
[21] Aluminium Alloy 6082 — T6 ~ T651, Aalco Metals Ltd., 2014
[22] D.W. Eagles, Plastic hot air stake assembly, Assem. Autom. 20 (2000) 205–214,
http://dx.doi.org/10.1108/01445150010372151.
[23] M. Sônego, Análise da viabilidade da união de estruturas híbridas polímero-metal
através de rebitagem por injeção baseada em energia friccional(Bachelor Thesis)
Universidade Federal de São Carlos (UFSCar), Brazil, 2013.
[24] A.B. Abibe, S.T. Amancio-Filho, M. Sônego, J.F. Dos Santos, A Method for Joining a Plastic
Workpiece to a Further Workpiece(European Patent Application EP14182938.2) 2014.
[25] J. Altmeyer, J.F. dos Santos, S.T. Amancio-Filho, Effect of the friction riveting process
parameters on the joint formation and performance of Ti alloy/short-fibre rein-
forced polyether ether ketone joints, Mater. Des. 60 (2014) 164–176, http://dx.
doi.org/10.1016/j.matdes.2014.03.042.
[26] S.T. Amancio-Filho, Friction Riveting: Development and Analysis of a New Joining
Technique for Polymer–Metal Multi-materials Structures(PhD Thesis) Technische
Universität Hamburg-Harburg, Germany, 2007.
[27] P. Su, A. Gerlich, T.H. North, G.J. Bendzsak, Energy utilisation and generation during
friction stir spot welding, Sci. Technol. Weld. Join. 11 (2006) 163–169, http://dx.doi.
org/10.1179/174329306X84373.
[28] DIN, EN ISO 14272: Probenmaße und Verfahren für die Kopfzugprüfung an
Widerstandspunkt- und Buckelschweißungen mit geprägten Buckeln, Deutsches
Institut für Normung e.V., Brussels, 2002.
[29] Amancio-Filho ST, Dos Santos JF, Beyer M. Method and device for connecting a plas-
tic workpiece to a further workpiece. US 7,780,432 B2. USA. 2010.
[30] A.B. Abibe, S.T. Amancio-Filho, J.F. Dos Santos, E. Hage, Development and analysis of
a new joining method for polymer–metal hybrid structures, J. Thermoplast. Compos.
Mater. 24 (2011) 233–249, http://dx.doi.org/10.1177/0892705710381469.
[31] A.B. Abibe, S.T. Amancio-Filho, J.F. dos Santos, E. Hage, Mechanical and failure behav-
iour of hybrid polymer–metal staked joints, Mater. Des. 46 (2013) 338–347, http://
dx.doi.org/10.1016/j.matdes.2012.10.043.
[32] A.B. Abibe, Friction-based Injection Clinching Joining (F-ICJ): A New Joining Method
for Hybrid Lightweight Structures(PhD Thesis) Technische Universität Hamburg-
Harburg, Germany, 2015.
[33] F. Simões, D.M. Rodrigues, Material flow and thermo-mechanical conditions during
Friction Stir Welding of polymers: literature review, experimental results and em-
pirical analysis, Mater. Des. 59 (2014) 344–351, http://dx.doi.org/10.1016/j.
matdes.2013.12.038.
[34] Z. Kiss, T. Czigany, Microscopic analysis of the morphology of seams in friction stir
welded polypropylene, Express Polym. Lett. 6 (2012) 54–62, http://dx.doi.org/10.
3144/expresspolymlett.2012.6.
[35] Z. Kiss, T. Czigány, Effect of welding parameters on the heat affected zone and the
mechanical properties of friction stir welded poly(ethylene-terephthalate-glycol),
J. Appl. Polym. Sci. 125 (2012) 2231–2238, http://dx.doi.org/10.1002/app.36440.
[36] Y.H. Yin, N. Sun, T.H. North, S.S. Hu, Hook formation and mechanical properties in
AZ31 friction stir spot welds, J. Mater. Process. Technol. 210 (2010) 2062–2070,
http://dx.doi.org/10.1016/j.jmatprotec.2010.07.029.
[37] C.F. Rodrigues, L.A. Blaga, J.F. dos Santos, L.B. Canto, E. Hage Jr., S.T. Amancio-Filho,
FricRiveting of aluminum 2024-T351 and polycarbonate: temperature evolution,
microstructure and mechanical performance, J. Mater. Process. Technol. 214
(2014) 2029–2039, http://dx.doi.org/10.1016/j.jmatprotec.2013.12.018.
[38] Y.C. Lin, J.J. Liu, B.Y. Lin, Effect of tool geometry on strength of friction stir spot
welded aluminum alloys, Adv. Mater. Res. 579 (2012) 109–117, http://dx.doi.org/
10.4028/www.scientific.net/AMR.579.109.
[39] J.-C. Bauwens, Physical aging: relation between free volume and plastic deforma-
tion, in: W. Brostow, R.D. Corneliussen (Eds.), Failure of Plastics, Carl Hanser Verlag,
Munich 1986, pp. 235–258.
[40] T. Czigany, Z. Kiss, Friction stir welding of fiber reinforced polymer composites, 18th
International Conference on Composite Materials, International Committee on Com-
posite Materials, Jeju, South Korea, 2011.
[41] N. Brown, Yield behaviour of polymers, in: W. Brostow, R.D. Corneliussen (Eds.),
Failure of Plastics, Carl Hanser Verlag, Munich 1986, pp. 98–118.
[42] X. Tan, J. Zhang, J. Shan, S. Yang, J. Ren, Characteristics and formation mechanism of
porosities in CFRP during laser joining of CFRP and steel, Compos. Part B: Eng. 70
(2015) 35–43, http://dx.doi.org/10.1016/j.compositesb.2014.10.023.
[43] S.T. Amancio-Filho, J. Roeder, S.P. Nunes, J.F. dos Santos, F. Beckmann, Thermal deg-
radation of polyetherimide joined by friction riveting (FricRiveting). Part I: influence
of rotation speed, Polym. Degrad. Stab. 93 (2008) 1529–1538, http://dx.doi.org/10.
1016/j.polymdegradstab.2008.05.019.
[44] M. Sônego, A.B. Abibe, J.F. Dos Santos, L.B. Canto, S.T. Amancio-Filho, Chemical chang-
es in polyetherimide (PEI) joined by friction-based injection clinching joining (F-ICJ)
technique, Polymer Processing Society Conference (PPS 2015). Graz, Austria, 2015.
[45] S. Carroccio, C. Puglisi, G. Montaudo, Thermal degradation mechanisms of
polyetherimide investigated by direct pyrolysis mass spectrometry, Macromol.
Chem. Phys. 200 (1999) 2345–2355.
642 A.B.A.B.
Abibe
Abibe
et al.
et /al.
Materials
/ Materials
andand
Design
Design
92 (2016)
92 (2016)
632–642
632–642 642

[46] H. Farong, W. Xueqiu, L. Shijin, The thermal stability of polyetherimide,
Polym. Degrad. Stab. 18 (1987) 247–259, http://dx.doi.org/10.1016/0141-
3910(87)90005-X.
[47] A. Flores, F. Ania, F.J. Baltá-Calleja, From the glassy state to ordered polymer struc-
tures: a microhardness study, Polymer 50 (2009) 729–746, http://dx.doi.org/10.
1016/j.polymer.2008.11.037.
[48] E. Ghorbel, G. Casalino, S. Abed, Laser diode transmission welding of polypropylene:
geometrical and microstructure characterisation of weld, Mater. Des. 30 (2009)
2745–2751, http://dx.doi.org/10.1016/j.matdes.2008.10.027.
[49] D.R. Rueda, F.J. Baltá Calleja, R.K. Bayer, Influence of processing conditions on the
structure and surface microhardness of injection-moulded polyethylene, J. Mater.
Sci. 16 (1981) 3371–3380, http://dx.doi.org/10.1007/BF00586299.
[50] D.R. Rueda, M.C.G. Gutiérrez, F.J.B. Calleja, S. Piccarolo, Order in the amorphous state
of poly(ethylene terephthalate) as revealed by microhardness: creep behavior and
physical aging, Int. J. Polym. Mater. Polym. Biomater. 51 (2002) 897–908, http://
dx.doi.org/10.1080/714975674.
[51] O. PHF, Estudo das propriedades e desempenho mecânico de juntas soldadas por
fricção pontual de poli (metacrilato de metila) (PMMA)(Master Thesis)
Universidade Federal de São Carlos (UFSCar), Brazil, 2012.
[52] S. Ghosh, R. Reddy, Ultrasonic sealing of polyester and spectra fabrics using thermo
plastic properties, J. Appl. Polym. Sci. 113 (2009) 1082–1089, http://dx.doi.org/10.
1002/app.30050.