ShakirGatea HooputraDamageModel
ShakirGatea HooputraDamageModel
Keywords: PEEK, Ductile Fracture, SPIF, Stress Triaxiality, Hooputra Damage Model
Abstract. This research investigates the use of Hooputra-based damage model to predict fracture
of polyether-ether-ketone (PEEK) material and its application to single point incremental forming
(SPIF) process under different conditions. Flat PEEK sheets are used to examine the influence of
temperature changes on the Hooputra fracture strain-stress triaxiality damage curve and to verify
the capability of the Hooputra ductile damage model in predicting fracture in uniaxial tensile tests
and the SPIF process. The research showed that the formulas used to calculate stress triaxiality
based on geometric dimensions are insufficient with notched samples because the radius of the
notched sample is no longer circular after plastic deformation. Temperature alters the Hooputra
damage curve; therefore, the Hooputra ductile damage formula should be developed to consider
the effect of temperature. When the Hooputra damage curve is established according to the
temperature effect, the uniaxial fracture is precisely predicted at different temperatures. Hooputra
ductile damage model could be developed to capture the fracture initiation and propagation in SPIF
process.
Introduction
Polyether-ether-ketone (PEEK) is a versatile thermoplastic polymeric material commonly used in
industrial applications due to its excellent combination of physical and mechanical properties [1].
In addition, PEEK presents a viable option for medical implants since it has a comparable elastic
modulus to cortical bone and is biocompatible [2]. PEEK's industrial and medical applications
could be expanded by using novel forming processes such as single point incremental forming
(SPIF). SPIF is a potential technique that could be developed as an alternative production solution
for PEEK materials with significant lead time and cost benefits. To employ the SPIF to produce
high-quality parts of PEEK, precise prediction of mechanical response and fracture behaviour of
PEEK materials is important.
The stress triaxiality and plastic strain are considered to be the two most significant variables
in crack initiation, growth, and coalescence in ductile fracture of materials [3]. As a result, various
ductile fracture models based on stress triaxiality, and fracture strain were established. Johnson
and Cook established a model to predict ductile fractures at high strain rates and temperatures, and
the fracture strain is related to stress triaxiality [4]. A void growth model was developed by
Kanvinde et al. to evaluate the ductile fracture in the range of high stress triaxiality [5]. Brünig et
al. established a damage model as a function of stress triaxiality [6]. Peng et al. suggested an
uncoupled ductile fracture model based on stress triaxiality and lode angle [7].
Based on the macroscopic strains and stresses a ductile fracture model was presented by
Hooputra et al. [8]. The fracture strain in this model is a function of stress triaxiality. Many
numerical investigations have been conducted based on Hooputra model. To select and calibrate
the appropriate fracture criterion for a specific application, seven different fracture models were
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of
this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under license by Materials
Research Forum LLC.
1596
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
employed. Hooputra damage model was one of them. To achieve the goal, 15 different
experimental tests on 2024-T351 aluminium covering a stress triaxiality range of -0.3 to 0.9 were
performed. The Hooputra damage model was discovered to be very promising because it
accurately predicts fracture in all types of experimental tests [9]. To determine which model could
be used to predict the forming limits of 2024-T3 aluminium alloy, eight ductile fracture models
were considered and calibrated. The Hooputra ductile damage model was shown to be more robust
against over-fitting and could be used when the number of calibration data points is small [10].
The experimental tests required to determine the parameters of Hooputra ductile damage model of
St14 steel could be neglected, and the parameters of Hooputra ductile damage model could be
predicated using a forming limit diagram [11]. To analyse the impact of mechanical discontinuities
on the crashworthiness performance of aluminium, the Hooputra damage model was combined
with the Müschenborn-Sonne forming limit diagram. It was discovered that incorporating it led to
more reliable crash simulations [12]. To predict the ductile fracture of metal alloys, a model based
on stress triaxiality, and shear stress ratio was proposed. The proposed model was compared to the
Hooputra ductile damage model, and there was a good correlation [13].
There are some research publications in the literature that illustrate the SPIF process's capacity
to successfully deform PEEK material [14]. A novel incremental sheet forming (ISF) process was
proposed to analyze the thermal formability of PEEK and investigate the geometric accuracy,
twisting, and fracture mechanisms. The results showed that as the temperature increased, the
accuracy of ISF part decreased and twisting angle increased. Striations marking is a common
fracture mode in hot PEEK ISF [15]. An alternate spiral toolpath based on linear interpolation was
developed to enhance the accuracy of SPIF of PEEK parts by reducing twisting in the deformed
parts [16]. In comparison to traditional processes, the ISF process could be used to manufacture a
customised product from PEEK, such as a cranial plate, with technological and economic
advantages [2]. Cold SPIF of PEEK sheets was performed to evaluate the effect of ISF parameters
on formability and to determine the best parameters for manufacturing the cranial plate and
cheekbone. The findings of this study demonstrated the capability of the SPIF process in the
biomedical field [17].
In this scientific context, precise selection of fracture strain and stress triaxiality values can
improve the prediction of ductile fracture models that were established based on the plastic strain
and stress triaxiality. The main motivation for this work is to evaluate the effect of temperature on
the fracture strain and stress triaxiality of PEEK and to implement these values in the Hooputra
ductile damage model to verify its accuracy to predict the ductile fracture of PEEK under different
strain states e.g., uniaxial tensile test and the plane strain of SPIF process.
Material and experimental procedures
PEEK 450G extruded sheets with a thickness of 3 mm were used in this study's experimental
testing. The glass transition and melting temperatures were determined using differential scanning
calorimetry. PEEK has glass transition and melting temperatures of 149 ⁰C and 340 ⁰C,
respectively. Tensile testing of standard samples (smooth) was performed by INSTRON testing
machine in accordance with ASTM/D638/sample type V to determine the mechanical properties
of PEEK sheets. The dimensions of the standard PEEK tensile sample were set to 63.5 mm total
length, 9.53 mm gauge length, and 3.18 mm the narrow section width. All tests were performed at
a constant strain rate of 1.75 ×10-3 S-1. Four flat notched samples with varying notch radiuses (0.5,
1, 3, and 5 mm) were machined and tested to obtain different stress triaxialities. The gauge length
width in smooth tensile sample and at notch root was designed to be 3.18 mm for each sample.
Tensile tests on smooth and notched samples were performed at room temperature, 80 ⁰C, 120 ⁰C,
and 150 ⁰C to examine the effect of temperature on fracture strain and stress triaxiality. Tensile
samples were examined in a temperature chamber. Tensile samples were painted with a white and
1597
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
black stochastic pattern, and strains were captured using video measurement during the tensile
tests. Table 1 reports the mechanical properties of PEEK sheet at room temperature.
To validate the accuracy of the Hooputra model in predicting ductile fracture in SPIF under
plane strain conditions, an SPIF test of PEEK sheets was performed using a HURCO CNC milling
machine equipped with a heating system to produce a hyperbolic truncated cone with varying angle
(from 40º to 90º) (see Fig. 1). The PEEK blank has a size of 150 × 150 mm and a thickness of 3
mm. To improve the surface quality of the produced part, the ROCOL RDT grease compound was
used as a lubricant between the high-speed steel forming tool and the PEEK sheet. The SPIF test
was carried out with a tool diameter of 10 mm, a step size of 0.5 mm, a feed rate of 1000 mm/min,
and a spindle speed of 100 rpm. An alternate spiral toolpath was used to reduce the twisting
phenomenon in PEEK SPIF.
Fig. 1. (a) Geometric shape of the hyperbolic cone, and (b) SPIF experimental fixture.
1598
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
The Hooputra ductile damage model is utilized in this study to predict fractures in both tensile
and SPIF tests. The onset of fracture in the Hooputra ductile damage model is based on the
nucleation, growth, and coalescence of voids. In this concept, fracture strain is only a function of
stress triaxiality. The following is the Hooputra ductile damage model.
where D is the damage variable that ranges from 0 (virgin material) to 1 (fractured material). Crack
initiation and propagation are modelled using element deletion. The element is eliminated from
the FE model when the damage variable equals 1.
Results and discussion
Effect of plastic deformation and temperature on stress triaxiality
There are two approaches for determining the stress triaxiality of notched tensile samples. The first
approach is based on the geometric dimensions of the specimen, and certain formulas were
developed to compute the stress triaxiality, such as the Bridgman formula for rounded tensile
samples and the Yuanli Bai formula [18] for flat samples. However, these formulas are no longer
strictly relevant because the dimensions of tensile samples do not remain constant during the
tensile test. Another approach is to use numerical simulation, where stress triaxiality is defined as
the ratio of hydrostatic stress to Von Mises equivalent stress. The stress triaxiality in uniaxial
tensile samples and in the SPIF process of a hyperbolic truncated cone is calculated using FE
simulation in this work.
The position of stress triaxiality was measured using the minimum cross section of a flat tensile
sample. As a result, the FE approach was used to examine the distribution of stress triaxiality on
the minimum cross section through the thickness. To demonstrate that the empirical formulas that
was used to capture stress triaxiality due to geometric dimensions are no longer strictly applicable,
the FE simulation of smooth and notched samples (notch radius (R)=0.5, 1, 3, 5 mm) was run
under the assumption that the material is elastic, then the simulation with elastic plastic properties
was run. The stress triaxiality values are constant under the elastic condition because the
dimensions of the tensile sample are not changed during the simulation and the sample returns to
its original shape when the load is removed. As a result, the stress triaxiality under elastic condition
is 0.33, 0.4, 0.43, 0.53, and 0.58 for smooth, R5 mm, R3 mm, R1 mm, and R0.5 mm samples,
respectively. Fig. 2 (a) depicts the influence of plastic deformation at room temperature on the
stress triaxiality distribution at the centre of an R5 mm notched sample. The stress triaxiality is
uniform under elastic conditions because the dimensions of the notched sample are not changed,
whereas under elastic-plastic deformation the radius of the notched sample is no longer circular
after plastic deformation, so the stress triaxiality changes through the thickness of the sample and
reaches its maximum value at the centre. The behaviour of all notched samples is the same, and
the values of stress triaxiality at the centre of the other samples are presented in Table 2. The
plastic deformation has no effect on the stress triaxiality of the smooth sample because in the
tensile test of the PEEK smooth sample, the necking begins at the beginning of the test and extends
throughout the gauge length of the tensile sample as shown in Fig. 2 (b), which means that the
1599
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
gauge length lines return to be parallel throughout the test and the fracture occurs when the
molecular chains reach a certain length.
Gro
wth
of
Ne
Nec
cki
kin
ng
g
Fig. 2. (a) Stress triaxiality distribution over the thickness of a notched sample (R5 mm) under
elastic and elastic plastic conditions and (b) mechanical behaviour of smooth sample.
PEEK is temperature sensitive, and raising the temperature below the glass transition
temperature enhances its formability. To investigate the effect of temperature on stress triaxiality,
tensile tests on smooth and notched samples were performed at various temperatures, including
room temperature (22 ⁰C), 80 ⁰C, 120 ⁰C, and 150 ⁰C. The influence of temperature on the stress
triaxiality of a notched sample (R5mm) is shown in Fig. 3 (a). It can be noted that the stress
triaxiality decreases as temperature rises, which indicates that as PEEK formability increases with
temperature, the stress triaxiality decreases. Temperature has the same effect on stress triaxiality
in all notched samples, and the effect of temperature on smooth samples is identical to the effect
of plastic deformation; the stress triaxiality of smooth samples does not change with temperature.
Table 2 shows the experimental fracture strain and stress triaxiality results at various temperatures.
To clarify why stress triaxiality decreases with rising temperature in notched samples, Fig. 3 (b)
compares the FE results of stresses distribution at necking area of R5 mm notched sample at room
temperature with one at 150 ⁰C. It is assumed that σ1 is the tensile stress, which obtained from the
experimental tensile test and decreases as the testing temperature rises. σ2 and σ3 denote the stress
in the width and thickness directions, respectively. σ2 and σ3 are close to zero before necking and
increase when necking begins. Because the yield stress of PEEK is greater at room temperature
than at 150 ⁰C, high values of σ2 and σ3 are required to compress the sample and induce necking
in the sample at room temperature, whereas the required stresses (σ2 and σ3) are decreased at 150
⁰C as shown in the figure. When σ2 and σ3 are reduced, the hydrostatic stress decreases, resulting
in a decrease in stress triaxiality.
1600
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
Fig. 3. Sensitivity of stress triaxiality (a) and stresses distribution (b) to temperature changes of
R5 mm notched sample.
Table 2. Fracture strain-stress triaxiality (η) of PEEK at different notches temperatures.
Temp. Smooth R5 R3 R1 R0.5
[⁰C] Strain η Strain η Strain η Strain η Strain η
22 1.29 0.33 0.52 0.93 0.38 1.04 0.15 1.20 0.07 1.37
80 1.55 0.33 0.62 0.79 0.59 0.96 0.16 1.04 0.09 1.28
120 1.86 0.33 0.78 0.73 0.67 0.78 0.22 0.93 0.12 1.19
150 1.94 0.33 1.33 0.55 0.96 0.60 0.35 0.86 0.24 1.03
Determination of Hooputra ductile damage parameters
The Hooputra ductile damage model is utilised in this work to predict the fracture initiation and
propagation in uniaxial tensile testing and the SPIF process. To calculate the Hooputra parameters
at different temperatures the values of fracture strain and corresponding stress triaxiality from
Table 2 are plotted as five isolated points at each temperature. To obtain the best match between
Eq.1 and the five isolated points, a curve fitting procedure is carried out using MATLAB tools.
The best-fitting values of a, b, and c represent the parameters of the Hooputra ductile damage
model. Fig. 4 depicts the Hooputra damage curves at various temperatures based on data from
Table 2, and Table 3 presents the relevant Hooputra parameters. According to Fig. 4 and Table 3,
the temperature affects the Hooputra damage curve and Hooputra parameters; hence, the Hooputra
ductile damage model should be modified to account for temperature impacts.
1601
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
1602
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
in Hooputra damage model, which means the Hooputra ductile damage curve is constant in FE
simulation, and according to the findings of this study, this curve changes with temperature and
the safe region expands with increasing temperature. Particularly, in the SPIF FE model, the stress
triaxiality-fracture strain values are compared with the Hooputra ductile damage curve, and if the
value is below the curve, no damage occurs; however, if the value is above the curve, damage
occurs in the FE model; however, the values of stress triaxiality-fracture strain change with
increasing temperature (fracture strain increases and stress triaxiality decreases) due to the plastic
deformation and friction. As a result, the Hooputra ductile damage model should be developed to
account for the effect of temperature in order to be utilised to predict fracture in polymeric
materials.
Fig. 5. (a) Uniaxial damage evolution at various temperatures, and (b) comparison of numerical
and experimental stress-strain curves at various temperatures.
Conclusions
The influence of temperature on stress triaxiality and Hooputra ductile damage curve of PEEK
was addressed in this work, and the Hooputra ductile damage results of fracture prediction of
uniaxial tensile test and SPIF process were validated by the experimental work. The findings of
this research can be summarised as follows:
• The stress triaxiality formulas based on geometric dimensions are not sufficient for notched
samples because the radius of the notched sample is no longer circular after plastic
deformation.
• The Hooputra ductile damage curve of PEEK material is temperature sensitive and changes
with deformation temperature.
• When the influence of temperature is addressed in Hooputra parameter calculations, the
uniaxial fracture is accurately predicted using the Hooputra ductile damage model. The
fracture in the SPIF process is predicted with a percentage of error; this could be because
the influence of friction-generated temperature was not taken into account in the FE model
and the Hooputra ductile damage curve is constant throughout the FE simulation.
• According to this investigation, the Hooputra ductile damage model should be modified to
take temperature into account in order to apply this model to predict fracture initiation and
propagation in polymeric materials.
1603
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
1604
Material Forming - ESAFORM 2024 Materials Research Forum LLC
Materials Research Proceedings 41 (2024) 1596-1605 https://doi.org/10.21741/9781644903131-177
[6] Brünig, Michael, Oliver Chyra, Daniel Albrecht, Larissa Driemeier, and Marcílio Alves, A
ductile damage criterion at various stress triaxialities, International journal of plasticity. 24 (10)
(2008) 1731-1755. https://doi.org/10.1016/j.ijplas.2007.12.001
[7] Peng, Zengli, Haisheng Zhao, and Xin Li, New ductile fracture model for fracture prediction
ranging from negative to high stress triaxiality, International Journal of Plasticity. 145 (2021)
103057. https://doi.org/10.1016/j.ijplas.2021.103057
[8] Hooputra, H, H Gese, H Dell, and H Werner, A comprehensive failure model for
crashworthiness simulation of aluminium extrusions, International Journal of Crashworthiness. 9
(5) ( 2004) 449-464. https://doi.org/10.1533/ijcr.2004.0289
[9] Wierzbicki, Tomasz, Yingbin Bao, Young-Woong Lee, and Yuanli Bai, Calibration and
evaluation of seven fracture models, International Journal of Mechanical Sciences. 47 (4-5) (2005)
719-743. https://doi.org/10.1016/j.ijmecsci.2005.03.003
[10] Zadpoor, Amir Abbas, Jos Sinke, and Rinze Benedictus, Formability prediction of high
strength aluminum sheets, International Journal of Plasticity, 2009. 25(12): p. 2269-2297.
https://doi.org/10.1016/j.ijplas.2009.02.005
[11] Haji Aboutalebi, F and A Banihashemi, Numerical estimation and practical validation of
Hooputra’s ductile damage parameters, The International Journal of Advanced Manufacturing
Technology. 75 (2014) 1701-1710. https://doi.org/10.1007/s00170-014-6275-8
[12] Estrada, Quirino, Dariusz Szwedowicz, Jesús Silva-Aceves, Tadeusz Majewski, Julio
Vergara-Vazquez, and Alejandro Rodriguez-Mendez, Crashworthiness behavior of aluminum
profiles with holes considering damage criteria and damage evolution, International Journal of
Mechanical Sciences. 131 (2017) 776-791. https://doi.org/10.1016/j.ijmecsci.2017.07.042
[13] Zhu, Yazhi and Michael D Engelhardt, Prediction of ductile fracture for metal alloys using a
shear modified void growth model, Engineering Fracture Mechanics. 190 (2018) 491-513.
https://doi.org/10.1016/j.engfracmech.2017.12.042
[14] Zhu, Hui, Hengan Ou, and Atanas Popov, Incremental sheet forming of thermoplastics: a
review, The International Journal of Advanced Manufacturing Technology. 111 (2020) 565-587.
https://doi.org/10.1007/s00170-020-06056-5
[15] Yang, Zhiyuan, Fei Chen, Shakir Gatea, and Hengan Ou, Design of the novel hot incremental
sheet forming experimental setup, characterization of formability behavior of polyether-ether-
ketone (PEEK), The International Journal of Advanced Manufacturing Technology. 106 (2020)
5365-5381. https://doi.org/10.1007/s00170-020-05035-0
[16] Yang, Zhiyun and Fei Chen, Mechanism of twist in incremental sheet forming of
thermoplastic polymer, Materials & Design. 195 (2020) 108997.
https://doi.org/10.1016/j.matdes.2020.108997
[17] Rosa-Sainz, Ana, María Luisa García-Romeu, I Ferrer, María Beatriz Silva, and G Centeno,
On the effective peek application for customized cranio-maxillofacial prostheses: An experimental
formability analysis, Journal of Manufacturing Processes. 86 (2023) 66-84.
https://doi.org/10.1016/j.jmapro.2022.12.044
[18] Bai, Yuanli, Xiaoqing Teng, and Tomasz Wierzbicki, On the application of stress triaxiality
formula for plane strain fracture testing. J Eng MaterTechnol. 131 (2) (2009) 021002.
https://doi.org/10.1115/1.3078390
1605