Content uploaded by Isyna Izzal Muna
Author content
All content in this area was uploaded by Isyna Izzal Muna on Dec 06, 2024
Content may be subject to copyright.
Characterization of the thermal and mechanical properties of additively
manufactured carbon ber reinforced polymer exposed to above-zero and
sub-zero temperatures
Isyna Izzal Muna
a
, Magdalena Mieloszyk
a,*
, Ruta Rimasauskiene
b
a
Institute of Fluid-Flow Machinery, Polish Academy of Sciences, Fiszera 14, Gdansk, 80-231, Poland
b
Department of Production Engineering, Faculty of Mechanical Engineering and Design, Kaunas University of Technology, Studentu 56, Kaunas, 51424, Lithuania
ARTICLE INFO
Handling editor: SN Monteiro
Keywords:
Thermal treatment
Additive manufacturing
Mechanical characterization carbon ber
reinforced polymer structural integrity
ABSTRACT
This experimental work aims to study the thermal degradation of carbon ber reinforced polymer (CFRP) printed
using additive manufacturing. The printed samples were exposed to various thermal modes (prolonged and
cyclic) and magnitudes (above- and sub-zero degrees). The inuence of temperature on CFRP was investigated
using static tensile testing (Young’s modulus and tensile strength) supported by scanning electron microscope
and differential scanning calorimetry. It is revealed that Young’s modulus and tensile strength of samples were
all degraded after all thermal treatments. Observing the morphological structure on the surface revealed changes
in the degradation processes due to thermal treatments. A list of the dominant degradation types for each
analysed thermal treatment is presented in the paper. It was observed that similar mechanical parameters values
can be linked with different material degradation processes.
1. Introduction
Additive manufacturing (AM), also known as three-dimensional (3D)
printing, has revolutionized the fabrication of complex 3D structures
across various industries by enabling the creation of intricate geometries
through a layer-by-layer (bottom-up) approach. This contrasts with
conventional manufacturing methods that typically employ a subtrac-
tive (top-down) technique. Among the numerous applications, carbon
ber reinforced polymer (CFRP) composites have gained signicant
traction due to their exceptional mechanical properties. These materials
are now indispensable in high-performance sectors such as aeronautics
[16], automotive [27], civil [26] and marine [19]. Material extrusion,
commonly known as Fused Deposition Modeling (FDM), is widely used
for creating complex geometries in CFRP composites [2,9]. The benets
of the FDM method are easy-to-use, relatively low budget, and possible
parts customization. However, the nal elements contain a relatively
high amount of voids that negatively inuence the mechanical proper-
ties of the nal product. One of the solutions was designed by Rima-
ˇ
sauskas et al. [18]. It is based on the 1 modication of the 3D printer and
special preparation of ber reinforcement before the printing process.
The proposed procedure results in enhancing the internal structural
integrity due to the better adhesion between ber bundles inside the
laminate.
Continuous carbon ber (CCF) is increasingly recognized as an
optimal reinforcement material for CFRP composites among other car-
bon ber types such as short carbon, whiskers, due to its superior
strength-to-weight ratio, rigidity, fatigue resistance, and corrosion
resistance [3,10]. These properties make CCF an ideal substitute for
metals in aerospace structures, enhancing performance while reducing
weight. The use of thermoplastics such as polyetheretherketone (PEEK),
polylactic acid (PLA), and acrylonitrile butadiene styrene (ABS) as ma-
trix materials further optimizes the manufacturing process by improving
melt processability, reducing costs, and eliminating the need for com-
plex curing cycles. This combination has led to the emergence of ber
reinforced thermoplastics as promising materials for various
high-performance applications. The advancement of producing CFRP
composites as a strong and lightweight structure using AM technology
enables printing aerospace parts and implementation in an actual
airplane application [14]. The aircraft parts are subjected to prolonged
and cyclic mechanical loads and temperature uctuations ranging from
−50 to 150 ◦C, which can withstand ight at supersonic speeds. In the
automotive industry, these settings reect the thermal loads experi-
enced under prolonged engine heat exposure and varying driving con-
ditions, ensuring the composite can endure stable high temperatures and
* Corresponding author.
E-mail addresses: imuna@imp.gda.pl (I.I. Muna), mmieloszyk@imp.gda.pl (M. Mieloszyk), ruta.rimasauskiene@ktu.lt (R. Rimasauskiene).
Contents lists available at ScienceDirect
Journal of Materials Research and Technology
journal homepage: www.elsevier.com/locate/jmrt
https://doi.org/10.1016/j.jmrt.2024.12.028
Received 9 August 2024; Received in revised form 29 November 2024; Accepted 2 December 2024
Journal of Materials Research and Technology 33 (2024) 9832–9842
Available online 4 December 2024
2238-7854/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (
http://creativecommons.org/licenses/by/4.0/ ).
frequent temperature uctuations without degradation.
Current literature predominantly focuses on the thermal effects on
polymeric composites manufactured through conventional methods,
leaving a notable gap in the understanding of how 3D-printed CFRP
composites respond to similar conditions [1,6,8,12,15,20]. Research has
shown varying results, with some studies indicating minor mechanical
strength reductions under thermal cycling, while others report signi-
cant degradation under prolonged thermal exposure. Lafarie-Frenot and
Ho [8] carried out an experimental work on a carbon/epoxy laminated
plate [0
3
∕90
3
]
s
subjected to 1000 thermal cycles (−50 to 150 ◦C) under
nitrogen or oxygen. It was found that the specimens were gradually
damaged by transverse matrix cracks which accumulated and propa-
gated during the test. The results showed that the orientation of the
edges referred to as bers and the layer locations within the lay-up have
a signicant impact on the transverse matrix cracking of composite
laminates subjected to heat cycles. Handwerker et al. [6] found that
applying heat to ber reinforced polymer (FRP) composites improved
their mechanical properties such as stiffness and interlaminar strength
in the build-up orientation even though the microscopical analysis
showed that the test samples still contained a big ratio of air and void.
Annealing the parts for 6 h at 200 ◦C, which is the melting point of
Polyamide 6 (PA6) as the used matrix material, produced the best
results.
Pascual-Gonz´
alez et al. [15] attempted to decrease the porosity
content in the CFRP composites with the same thermoplastic material
PA6, using post-processing temperature. It was revealed that without
changing dimensional sizes, the treated parts subjected to 150 ◦C
improved interlaminar strength by 145% and reduced porosity by
approximately 87%. Abdullah et al. [1] revealed that after thermal cy-
clic exposures, the material possesses a small decrease in mechanical
strength. The tensile strength of the tested samples decreased as the
number of thermal cycles increased. Tensile properties decreased after
5600 thermal cycles by up to 8.5% when compared to the untreated
sample group.
In this experimental research, those research gaps are being studied
by examining the mechanical and thermal properties of CFRP compos-
ites fabricated with a modied FDM printer. By subjecting the 3D-
printed samples to prolonged and cyclic thermal treatments, both
above and below 0 ◦C, this research will employ techniques such as
tensile testing, differential scanning calorimetry (DSC), scanning elec-
tron microscopy (SEM), and optical imaging to evaluate the mechanical
behavior of the composites and their structural integrity. The ndings
will enhance the understanding of degradation processes and provide
insights into optimizing the manufacturing and treatment of 3D-printed
CFRP parts for high-performance applications in aerospace, automotive,
and other industries.
2. Materials and 3D-printing method
2.1. Materials
In this experimental work, unidirectional (UD) CFRP composites
were manufactured with the modied FDM printer MeCreator2 at
Kaunas University of Technology, Lithuania according to the procedure
designed by Rimaˇ
sauskas et al. [18]. The matrix agent was thermo-
plastic PLA (Polymaker), and the CCF reinforcement was T300D-3000
(Toray, France) with 3000 carbon ber tow. Table 1 displays the ma-
trix and ber mechanical properties based on information from material
suppliers. A total of 45 3D-printed CFRP samples were produced, with
the 3D-printed specimens illustrated in Fig. 1. Each specimen was
designed with dimensions of 150 ×13 ×2 mm, and the carbon ber
content in the composite was approximately 18.2%. This estimated ber
content was determined based on the tool-path length used during
printing and can be measured as the weight ratio of carbon ber to the
overall composite specimen [23]. This method provides a precise esti-
mation of the carbon ber distribution within the matrix.
2.2. Printing process and parameters
Pre-impregnation procedure was used for the reinforcement using
CCF to improve the printing performance and lament bonding. PLA
pellets were dissolved in a 90 g/10 g dichloromethane solution from
Euro-chemicals using a magnetically charged LBX H01 mini-stirrer at
600 rpm. The virgin CCF tow (non-impregnated) was run through this
polymer and concurrently dried using an air gun at 220 ◦C.
The extrusion printing head has been modied with two inputs (one
for the ber component and one for the matrix material) and one output,
enabling the material made from polymers to be fused with the ber
throughout the printing process. The impregnated CCF was fed directly
to the printing nozzle through the printing head. Liquid PLA is joined
with impregnated CF and pumped continuously through the printing
nozzle while the polymer melts inside the mixing chamber of the heating
control unit. The modied FDM printer device and conguration of the
3D-printed specimen are presented in Fig. 2.
The optimal temperature at the printing head for melting PLA and
creating a bond with CCF was 220 ◦C. The PLA material reinforced with
carbon ber is extruded through the printing nozzle on the borosilicate
glass printing bed that is mounted on the Al plate. The rectilinear pattern
of the printing process was selected, and the extrusion width was set to
be 1.4 mm forming nine continuous parallel lines in each deposited
layer. The layer height was set to have four layers for each 3D-printed
specimen. After printing the composite sample, the cutter will sepa-
rate the lament spool and the 3D-printed part. The borosilicate glass
should be removed from the Al plate bed to cool before extracting the
3D-printed sample with blades. The printing process parameters are
shown in Table 2.
3. Experimental methods and equipment
3.1. Thermal treatment
After manufacturing, nine groups of 3D-printed CFRP specimens
were thermally exposed at different magnitudes (above and subzero
degrees Celsius) and exposition times (cyclic and stable). Table 3 pre-
sents the sample groups under various thermal conditions. Within each
thermal group, there are ve individual specimens, providing a suf-
cient sample size to ensure accurate and reliable testing results. This
approach is crucial for understanding how varying thermal exposures
inuence the mechanical behavior of CFRP materials. It is important to
note that thermal treatments at temperatures above zero degrees Celsius
are referred to as hot thermal treatments, while those below zero de-
grees Celsius are referred to as cold thermal treatments.
For hot thermal group, the experimental work was conducted at
Kaunas University of Technology, Lithuania, with a universal environ-
mental oven, “Memmert” Model UFB-400. This oven provided a con-
stant renewal of oxygen from ambient air. The temperatures in the hot
thermal loading were selected to ensure they remain above the glass
transition temperature (Tg) yet below the melting temperature (Tm) of
PLA polymer as matrix material used in the specimens which are about
60 ◦C and 160 ◦C, respectively. For the hot stable group, the specimens
were placed in a preheated oven at a desired temperature and duration
(e.g., 65 ◦C 6 h) and allowed to cool naturally to room temperature
afterwards.
For the hot cyclic group, the temperature cycles were manually
Table 1
Mechanical properties of composite components.
Elastic modulus
(GPa)
Elastic strength
(MPa)
Strain at
failure (%)
Density (g/
cm
3
)
Matrix 2.636 46.6 1.9 1.17
Fiber 230 3530 1.5 1.76
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9833
regulated using the controller settings of the oven, based on real-time
thermocouple readings. The specimens were subjected to multiple
thermal cycles within a specic temperature range (e.g., between 50 and
70 ◦C for six cycles) according to a thermal cyclic plan. Each cycle
included two sets of temperatures, with a 10-min dwell time, a cooling
rate of 2.5 ◦C per minute, and a heating rate of 1 ◦C per minute. The
heating rate was manually controlled and monitored through thermo-
couple feedback to maintain accuracy. The cooling process occurred
naturally, with the heat source turned off and the cooling rate carefully
observed. For the cold thermal group, the experimental condition was
performed at the Institute of Fluid-Flow Machinery, Polish Academy of
Sciences, Poland with an automated environmental chamber “MyDis-
covery” model DM600C (Angelantoni Test Technologies Srl, Italy).
For the cold stable group, the specimens were placed in a chamber at
programmed temperatures and duration before being naturally cooled
to room temperature. For the cold cyclic group, the environmental
chamber was automatically regulated for a specic temperature range
and several thermal cycles following the thermal cyclic plan. Each cycle
had two desired temperatures with 10 min of dwelling time, a cooling
rate of 5 ◦C per minute, and a heating rate of 2 ◦C per minute. The
environmental chamber and oven are presented in Fig. 3. The environ-
mental chamber parameters are shown in Table 4.
The temperatures for the cold thermal loading in this experiment
were selected to mimic the thermal conditions experienced by aerospace
components during high-altitude ights and space missions. At such
altitudes, aerospace structures, including satellite bodies and other high-
performance materials, can be exposed to temperatures that dip signif-
icantly below freezing, especially when transitioning from direct sun-
light to the shadow of Earth or other celestial bodies. The testing
protocol for cold thermal loading followed the MIL-STD-883 standard,
specically Method 1010. The −20 ◦C threshold is commonly chosen as
a balance between reecting the extreme cold conditions experienced
and considering the potential impact on mechanical integrity, without
exceeding the operational temperature limits of the material [22].
Several studies have explored the impact of temperature gradients,
such as 5 ◦C or 10 ◦C, in evaluating the effects of temperature cycling on
the properties of CFRP material [4,21,24]. A temperature gradient of
this scale, often used in experimental setups, crucial for understanding
how materials respond to thermal stress over repeated cycles. For
example, gradients of 10 ◦C allow researchers to assess the expansion
and contraction behavior of materials, simulating conditions like ther-
mal fatigue and the thermal expansion mismatches that occur in
real-world applications [4,24].
It is important to note that the time experienced by samples treated
at prolonged above and sub-zero degrees is consistent, but for cyclic
temperatures, it varies due to the differences in temperature range. It is
worth mentioning that the CFRP sample maintained its shape during the
6-h treatment at 145 ◦C, despite this temperature being close to the
melting point of PLA, primarily due to the reinforcing role of carbon
bers. These bers provide signicant structural integrity and me-
chanical strength, preventing the PLA matrix from deforming signi-
cantly. Carbon bers also aid in evenly distributing heat throughout the
composite, reducing the risk of localized overheating and deformation.
3.2. Surface analysis
SEM and optical microscopy analysis were carried out to evaluate the
microstructure and macrostructure of specimens due to thermal treat-
ment. The optical microscope, Nikon Eclipse LV100ND, was used to
examine the micro-morphological structures outtted with a high-
denition color camera (Nikon DS-Ri2). The data was prepared and
processed using imaging software (NIS Elements 4.5.1.00, Nikon Europe
B.V., Amstelveen, The Netherlands). The largest sample size seen under
the optical microscope was 150 mm square. A scanning electron mi-
croscope (FE-SEM SU5000, Hitachi Co., Tokyo, Japan) was utilized to
observe the micro-damage that resulted from tensile testing on various
specimen groups. The largest specimen that the SEM could measure had
dimensions of 80 mm ×200 mm. The study used a digital microscope
with 500×magnication (Levenhuk, DTX 500 LCD, Warsaw, Poland).
3.3. Thermal analysis
DSC measures the energy absorbed (endotherm) or released (exo-
therm) as a function of time or temperature and it is suitable for
detecting the effects of thermal degradation by observing the melting
and crystallization process. The important thermal phases are studied to
determine the polymer crystallinity: glass transition temperature (Tg)
following ASTM standard E1356, cold crystallization (Tcc), and melting
Fig. 1. AM CFRP specimens printed with a modied FDM printer prior to thermal and mechanical testing.
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9834
point (Tm) temperatures following ASTM standard E794. Tg is indicated
by a shift of the baseline from the initial DSC curves and reported Tg was
based on the observed midpoint temperature. An exothermic peak in-
dicates a cold crystallization where an exothermic reaction (heat
release) occurs, while an endothermic peak (heat adsorption) refers to
the melting temperature in which an endothermic reaction takes place.
A DSC equipment, TA Instruments Q2000, was utilized to analyze the
thermal properties of CFRP specimens under controlled and isothermal
conditions. The sample for each thermally treated group was chopped
into small pieces and measured with a precision scale. About 10 mg of
sample from each treatment group was placed in an Al hermetic pan and
inserted into the DSC cell. A nitrogen atmosphere was supplied to the
test chamber at a ow speed of 50 mL/min for the cooling process, while
an electrically heated furnace was used for heating. The DSC measure-
ment for each sample consisted of two times of the heating process and
one time of the cooling process. The rst heating scan in DSC is used for
removing the thermal history of the polymer which might have gone
through during its synthesis and post-processing steps. Tg value differ-
ence after the second heating run could be insignicant whereas Tm and
Fig. 2. 3D-printing setup of the modied FDM printer and lament spools (A); extruder (B); printing nozzle (C); 3D-printed specimen conguration (D).
Table 2
Printing parameters.
Parameter Value
Nozzle temperature 220 ◦C
Bed temperature 90 ◦C
Interior inll 100%
Inll pattern Rectilinear
Printing speed 3 mm/s
Nozzle diameter 1.5 mm
Extrusion multiplier 0.7
Primary layer height 0.5 mm
Table 3
Sample groups.
Group name Description
Intact Untreated samples
HS-A Hot stable at 65 ◦C for 6 h
HS-B Hot stable at 145 ◦C for 6 h
HC-A Hot cyclic between 50 and 70 ◦C with 6 cycles
HC-B Hot cyclic between 140 and 150 ◦C with 6 cycles
CS-A Cold stable at 0 ◦C, for 6 h
CS-B Cold stable at −20 ◦C, for 6 h
CC-A Cold cyclic between −5 and 0 ◦C for with 12 cycles
CC-B Cold cyclic between −20 and −15 ◦C with 12 cycles
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9835
Tc value difference could be very distinct difference in their respective
values between the two runs. The DSC program is shown in Table 5.
A DSC measuring cell consists of a furnace and an integrated sensor
with designated positions for the sample and reference pans. The sample
is placed in an Al pan, and the sample pan and an empty reference pan
are placed on small platforms within the DSC chamber. Thermocouple
sensors lie below the pans and they are connected to thermocouples.
This allows for recording both the temperature difference between the
sample and reference side (DSC signal) and the absolute temperature of
the sample or reference side. The functional principle of an interior
chamber of DSC is shown in Fig. 4.
DSC was used to determine the glass transition temperature (Tg) of
the polymeric composite samples in accordance with ASTM E1356. The
melting temperature and crystallization temperature will also be ana-
lysed according to ASTM E794. The purpose of the test was to determine
whether there was any change in the crosslinking of the PLA polymer,
which could signify degradation as a result of the conditioning process.
A DSC equipment (TA Instruments Q2000) was used to perform thermal
analysis of the composite samples under controlled and isothermal
conditions. The sample for each thermally treated group was chopped
into small pieces to t inside the pan due to the stiffness of the ber-
reinforced raw lament.
One representative sample was prepared for each treatment group in
the DSC analysis. The prepared samples were cut into small workpieces
with a weight of about 10 mg for each sample. Each cut sample then was
placed in an Al hermetic pan and inserted into a DSC cell. A nitrogen
atmosphere was supplied to the test chamber at a ow rate of 50 mL/min
for the cooling process while an electrically heated furnace is used for
heating. The temperature was ramped from 20 to 200 ◦C and then cooled
back to 20 ◦C at a heating and cooling rate of 10 ◦C/min. The mea-
surement for each sample consisted of two times of the heating process
and one time of the cooling process. The following program was used:
hold equilibrium at 24 ◦C, ramp at 10 ◦C/min to 200 ◦C, hold isotherm
for 2 min, ramp at 10 ◦C/min to 40 ◦C, hold isotherm for 2 min, and
ramp back at 10 ◦C/min to 200 ◦C. The DSC equipment and DSC cell can
be seen in Fig. 5.
3.4. Mechanical testing
After thermal treatment, tensile testing was carried out on a Tilnius
Olsen H25KT universal testing machine with hydraulic grips of 25 kN.
An extensometer was calibrated with a displacement rate of 2 mm/min
against strain gauges to determine tensile strength and strain for all
group samples. Four locations on the samples were labelled with a
marker 15 mm from the center to measure the elastic strain. Gripping
tabs made of PLA were printed separately, having dimensions 50 mm in
length, 12.5 mm in width, and 2 mm in thickness, and each specimen
was then glued with two tabs at the front side and two at the backside.
Five samples for each group of thermal treatment were required to
determine the mean value for the mechanical properties of the speci-
mens according to the ASTM D3039 standard. The specimen prepared
Fig. 3. An automated environmental chamber for cold thermal treatment (left) and an air-circulated oven for hot thermal treatment (right).
Table 4
Environmental chamber parameters.
Parameter Value
Temperature [◦C] −75 ÷180
Temperature uctuations [◦C] ±0.1 ◦C ÷ ± 0.3 ◦C
Temperature change rate [◦C/min]
Heating 4.5
Cooling 4
Relative humidity [%] 10 ÷98
Relative humidity uctuations [%] 1 ÷3
Table 5
DSC program steps.
Step Temperature [◦C] Heating rate [◦C/min]
Hold equilibrium 24
Ramp-up 200 10
Hold isotherm for 2 min
Ramp-down 40 10
Hold isotherm for 2 min
Ramp-up 200 10
Fig. 4. Schematic diagram of DSC interior chamber.
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9836
for tensile testing and tensile test setup are shown in Fig. 6.
The ASTM D3039 standard was used for calculating the modulus of
elasticity (Young’s modulus) in materials testing. The tensile modulus
and strength values of the CFRP specimens were calculated by Equation
(1) and Equation (2), respectively.
σ
=Fmax
A(1)
E=ΔF L0
AfΔL(2)
where F
max
denotes the maximum tensile force given during the test (N),
A
f
is the cross-sectional area of the composite specimen (mm
2
), ΔF and
ΔL are the differences in the force and extension between two strain
points, respectively, and L
0
is the gauge length of the specimen.
4. Results and discussion
The DSC results, tensile testing, and SEM analysis will be presented
in this section.
4.1. Thermal analysis
The mechanism of the heating process between the reference pan and
the sample pan is presented in Fig. 4. When the DSC measuring cell is
being heated, the reference side—typically an empty pan—heats up
more quickly than the sample side because of the sample’s heat capacity
(C
p
); for example, the reference temperature (Tr, green) rises slightly
more quickly than the sample temperature (Ts, green). The two curves
behave similarly while heating at a steady rate until a sample reaction
occurs. In this example, the sample melts at t
1
. The temperature of the
sample does not vary during melting; nevertheless, the temperature of
the reference side remains constant and continues to rise linearly. When
melting is complete, the sample temperature begins to rise again, with a
linear increase commencing at time t
2
.
The DSC curves for the initial heating, cooling, and second heating
cycles of intact (untreated) and thermally treated CFRP specimens at
prolonged and cyclic temperatures are shown in Figs. 7 and 8, and the
values obtained from the tests are shown in Table 6.
In the rst heating process, the Tg of the untreated (intact) CFRP
group is very close to composites treated in a stable temperature at 0 ◦C
(CS-A) and −20 ◦C (CS–B), as well as cyclic temperatures between 0 and
5 ◦C (CC-A). Similar values of Tg were also observed for specimens
subjected to cyclic heating (HC-A and HC-B). This means that these
groups will take a longer time to change their material state from solid to
a soft-rubbery material after the thermal treatment which makes them
more brittle.
However, with thermal cycling at extremely low temperatures be-
tween −20 and −15 ◦C (CC–B) and stable high temperatures (HS-A and
HS-B) the Tg value noticeably dropped.
The Tcc point of the intact group and CS-A are signicantly higher
than for the other treated groups. Cold crystallization is the process of
rapidly cooling a crystalline plastic from its liquid state, resulting in the
freezing of polymer chains in their amorphous state. Crystallization is
characterized by crystal nucleation and nucleus growth [25]. However,
cold crystallization did not occur for the higher magnitude of hot ther-
mal treatment where the specimens were exposed to cyclic temperature
(HC–B) and stable temperature (HS–B). This happens because the crys-
tals do not have enough time to form. The formation of crystals creates
an endothermic peak between the Tg and the melting point when
reheating a material in this state. The cold crystallization process is
distinguished by two characteristics the promotion of nucleation as the
supercooled glassy state gradually gains mobility with increasing tem-
perature and the presence of a maximum temperature for nucleation
above which the cold crystallization process is diffusion limited [13].
It was observed that all sample groups had a double melting tem-
perature peak during the rst heating run. This is owing to the super-
positioning of melting and recrystallization processes, which causes
this phenomenon. When most crystallization occurs during the cool-
down process, the remaining amorphous regions lack place and chain
mobility, resulting in defective crystals. These defective crystals begin to
melt, but almost simultaneously, recrystallization occurs, generating a
new crystal structure that melts almost quickly, creating the second peak
[6].
During phase transitions, CFRP samples exhibited two distinct peaks
(and onset points), which may indicate more than one form or crystal
structure. PLA has slow crystallization kinetics. In the case of PLA, three
different crystallization processes can be identied with adequate
experimental conditions: the classical cold crystallization and two pro-
cesses associated with the non-reversing exotherms [5]. The DSC graphs
obtained in this experiment exhibit unexpected and extraneous results,
such as non-reversing heat-ow curves with two exothermic processes
and a larger endotherm in the middle. The multi-exothermic processes
result from several crystalline states of PLA (
α
and β).
The DSC curve produced lower values in the second heating process
than in the rst run. This phenomenon is because the relaxation or
molecular rearrangement already occurred in the rst heating run. The
Fig. 5. DSC equipment and DSC cell.
Fig. 6. Tensile testing setup (A); and CFRP specimen with tabs (B).
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9837
cooling process imparts/equilibrates the previously known history at a
known rate from the rst heating before heating again. As a result, any
changes detected in the reheating curve between identical materials are
not due to previous thermal history effects but rather the differences in
the actual internal materials (e.g., molecular weight).
In the cooling cycle, the melt crystallization peak for each sample
group can be observed and determined by the peak, which denotes the
crystalline nature of the polymer. However, there are some cases when
the reinforcement materials constrain the PLA chains so much that the
heat capacity jump becomes undetectable in DSC. Nucleation on the
sample borders, regional connement, temperature gradient, and melt
ow are all important aspects of polymer crystallization. Similarly,
foreign chemicals in the pure PLA matrix affect PLA crystallization by
facilitating or impeding chain movement. Furthermore, due to the
presence of ordered crystallites and thicker lamellae, higher crystallinity
generates higher stiffness in both materials, preventing the material
from slipping into the crystal blocks [7]. In contrast, reduced crystal-
linity produces a higher ductility with reduced strength and stiffness
[17,29].
4.2. Mechanical characterization analysis
Fig. 9 depict the typical stress-strain curves obtained from the tensile
tests of CFRP composites of the above-zero and sub-zero degrees group,
respectively. While some graphs depict a trend of linear elasticity, others
began to break at elongation. This type of behavior is common in semi-
crystalline polymers. When the yield strength is exceeded, the chains in
the amorphous sections straighten out and stop interconnecting with
one another [6].
The tensile stress-strain graphs show that the untreated composite
specimen reached the highest stress level, followed by the prolonged
temperature group exposed to 65 ◦C. The lowest stress level was attained
from a specimen group subjected to a cyclic temperature between 50
and 70 ◦C with a considerable strain level indicating more elasticity. The
premature failure of 3D-printed samples at a strain of 0.75%, despite the
higher elongation at break for each constituent of ber and PLA (>1%),
is likely due to a combination of processing defects like voids and
porosity, stress concentrations from geometric irregularities, residual
stresses from the thermal history of printing, and the inherent anisot-
ropy of the 3D-printed composite structure. These factors collectively
contribute to a lower effective strain capacity of the 3D-printed samples
compared to the individual materials.
The corresponding mechanical parameters (Young’s modulus and
tensile strength) are shown in Fig. 10. The untreated samples showed the
highest tensile strength mean and Young’s modulus. The tensile prop-
erties of specimens then decreased after various modes of thermal
treatment at different temperature ranges. For the stable thermal
treatment, the sample group treated to the above-zero temperature
Fig. 7. DSC graphs of thermally treated and non-treated 3D-printed specimens for above-zero degrees treatment groups: (a) First heating; (b) Cooling; (c) Sec-
ond heating.
Fig. 8. DSC graphs of thermally treated and non-treated 3D-printed specimens for sub-zero degrees treatment groups: (d) First heating; (e) Cooling; (f) Sec-
ond heating.
Table 6
DSC results of CFRP composite samples.
Group
name
Glass
transition
temperature
Tg [
◦
C]
Cold
crystallization
temperature Tcc
[
◦
C]
Melting
temperature
Tm [
◦
C]
Crystallization
temperature Tc
[
◦
C]
Intact 64.88 118.67 158.09 60.80
HS-A 61.58 98.29 162.91 60.07
HS-B 63.23 –161.39 59.8
HC-A 65.43 114.38 157.82 58.97
HC-B 64.19 –164.83 59.66
CS-A 64.73 118.07 157.89 59.02
CS-B 64.61 94.88 158.92 59.93
CC-A 64.33 95.68 159.06 58.42
CC-B 62.55 94.72 158.64 60.76
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9838
achieved higher ultimate tensile strength than the sub-zero temperature.
The lowest Young’s modulus and ultimate tensile strength were ob-
tained from the sample group treated at −20 ◦C. Treatment at 65 ◦C
signicantly reduced the ultimate tensile strength and Young’s modulus.
However, at 145 ◦C Young’s modulus was reduced but the ultimate
tensile strength appeared to be slightly higher than the treated sample at
65 ◦C.
The mean values in the bar plots represented the patterns of each
printed group specimen and the range of their effects on tensile qualities.
It is assumed that the decrease in mechanical strength and elastic
modulus under thermally stable and cyclic loading was caused by the
difference in coefcients of thermal expansion (CTE) between the matrix
and ber [11]. The material degraded due to the thermal stress caused
by the CTE disparity, which might lead to bers pulling out due to
ber-matrix debonding. The deviation in the mechanical properties
observed in the results can be attributed to the inherent variability in the
3D-printing process, particularly when using a modied 3D printer for
CFRP samples. This type of 3D printer, while capable of 3D-printing
CFRP composites, often faces challenges related to repeatability in
terms of both the quality and dimensions of the printed samples.
The tensile tests results (reduction of Young’s modulus and tensile
strength) are consistent with the DSC ndings, where reduced thermal
stability (lower Tg) and altered crystallinity were observed. The
different levels of crystallinity between sample groups can help explain
the observed mechanical differences. Furthermore, in the research study
of thermosetting systems [28], crystallization behavior can be analysed
similarly in thermoplastic-based composite systems. In these systems,
the post-manufacturing crystallization is crucial, where residual thermal
conditions continue to affect the crystalline content and, in turn, the
material’s nal mechanical performance. The degree of crystallization
post-processing can inuence the toughness and load-distribution ca-
pabilities of the composite, just as in thermosetting polymers where
cross-linking determines the nal properties.
4.3. Morphological analysis
Before and after the thermal loading exposures, morphological
studies were carried out using an optical microscope device, and the
microstructure of the various specimen groups after tensile testing was
investigated using a scanning electron microscope.
4.3.1. Optical microscopy
Fig. 11 presents the surface structure at the microscopic scale from
each heating group before and after thermal loading. This optical
analysis was solely conducted at elevated temperatures due to its direct
correlation with the thermal treatment procedure proposed post-
manufacture. The PLA matrix material looked slightly ner and
smoother on the prolonged and cyclic treatment samples at lower tem-
peratures (HS-A and HS-B). However, these specimens showed no dis-
torted shapes (some crease structure along specimen length) because of
insufcient high-temperature exposure. The thermal treatment at higher
temperatures showed a more obvious difference on the surface after the
treatment [12].
The CFRP sample subjected to continuous heating at 145 ◦C reveals
signicant drying out of the polymer matrix. This thermal exposure
causes a reduction in the moisture content of the matrix, leading to
changes in its physical properties. This change is evident in the DSC
results, where the melting peak area is notably reduced. The smaller
area under the melting peak indicates a decrease in the thermal energy
required for phase transitions, which is consistent with the loss of vol-
atiles and moisture from the polymer matrix during prolonged heating.
4.3.2. Scanning electron microscopy
The samples after tensile testing were examined using SEM. Micro-
graphs are presented in Fig. 12. The polymer structure of the he un-
treated sample shows no signs of breakage or deterioration. Upon close
examination, it exhibits a pristine surface with no observable defects or
aws. PLA subjected to cyclic heating between 50 and 70 ◦C (HC-A)
Fig. 9. Average stress-strain curves for thermal treatment group: (a) above-zero, (b) sub-zero.
Fig. 10. Experimental results of the tensile strength and Young’s modulus of 3D-printed samples for all thermal treatment groups.
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9839
Fig. 11. Optical micrographs of the specimen before and after the prolonged temperature at 65 ◦C (HS-A) and 145 ◦C(HS-B); and cyclic temperature between 50 and
70 ◦C (HC-A) and between 140 and 150 ◦C (HC–B).
Fig. 12. SEM photos of the untreated and treated specimen group following destructive tensile testing.
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9840
underwent signicant structural degradation, evidenced by a conspic-
uous fracture bisecting the material. This fracture observed prominently
along the middle, suggests that the polymer experienced considerable
stress and strain during the cyclic heating process, resulting in macro-
scopic damage. Conversely, samples subjected to cyclic thermal treat-
ment between 140 and 150 ◦C (HC–B) displayed larger and more evenly
distributed fractures in the polymer matrix. After prolonged thermal
exposure to 145 ◦C (HS–B), the sample exhibited matrix thinning and
larger, more widespread cracks. In contrast, the polymer parts treated
with prolonged temperature at 65 ◦C (HS-A) displayed microcracks,
indicating localized weakening within the material. The occurrence of
microcracks implies that the PLA underwent thermal expansion or
contraction, leading to stress concentrations and subsequent crack for-
mation. These ndings underscore the sensitivity of PLA to temperature
uctuations and the importance of controlled processing conditions in
ensuring material integrity.
The higher temperatures during cyclic and prolonged thermal
treatments (HC–B and HS-B), exacerbate the damage to polymer com-
ponents. This indicates that elevated temperatures intensify the degra-
dation process, leading to more severe structural damage in the polymer
material. The fracture appeared to exist uniformly in almost all areas of
the matrix material. In addition to that the bers were also showing
some warping behavior and were no longer coated by the material made
from the pre-impregnation process. This can be interpreted as the brit-
tleness observed in the material treated at higher temperatures can be
primarily attributed to the increase in crystallinity. Crystallinity restricts
the polymer chains’ movement, making the material less exible and
more prone to brittle failure. This behavior is conrmed by DSC results
where the cold crystallization was not presented in groups HC-A and HC-
B. This is because there is no crystal nucleation which will increase the
chain mobility with the increasing temperature. In the cold thermal
treatment case, the prolonged temperature CS-B created some notice-
able gaps in the bers, and matrix crack was also presented with some
ber pull-out. For the treatment CS-A, bers seemed to remain in their
initial arrangement and the gaps between bers were not observed.
However, the PLA matrix is slightly distorted compared to the untreated
(intact) group. Fibers warping and separation were the damage resulting
from thermal cycling CC-A. The matrix was observed to be deformed
into some unstructured arrangement. A similar morphological structure
was observed for the thermal cycling CC-B where the polymer matrix
was not only deformed but seemed to be more nucleated and formed
some thin-small membranes.
Samples subjected to high-temperature treatments exhibited signs of
ber-matrix debonding and matrix degradation, which are consistent
with the observed reductions in mechanical strength and Young’s
modulus. The morphological features provide a visual conrmation of
the damage mechanisms inferred from the DSC and tensile test results.
The fracture surfaces of the tensile-tested samples revealed different
failure modes depending on the thermal treatment. Samples treated at
higher temperatures showed brittle fracture characteristics with less
ber pull-out, indicating reduced ductility and toughness. This
morphological evidence supports the ndings from both the DSC and
tensile analyses, illustrating how thermal history impacts the fracture
behavior of CFRP materials. A list of the dominant degradation types for
each analysed thermal treatment together with the mechanical param-
eters values is presented in Table 7.
5. Conclusion
This study comprehensively examined the effects of various thermal
treatments on the mechanical properties, thermal stability, and
morphological characteristics of continuous carbon ber reinforced
polymer (CFRP) composites 3D-printed using the modied FDM printer.
Untreated CFRP samples exhibited the highest tensile strength and
Young’s modulus, indicating superior mechanical performance. Ther-
mal treatments, especially prolonged heating at 145 ◦C (HS–B) and
cyclic thermal exposure between 140 and 150 ◦C (HC–B), resulted in
signicant degradation of mechanical properties, as evidenced by
reduced tensile strength and modulus. Differential Scanning Calorimetry
(DSC) analysis revealed that thermal treatments inuenced the glass
transition temperature (Tg), crystallinity, and other thermal transitions.
Higher Tg values were observed in untreated samples, correlating with
better thermal stability and mechanical integrity.
Optical and scanning electron microscopy (SEM) revealed distinct
morphological changes in the treated samples. Scanning electron mi-
croscopy (SEM) highlighted considerable structural changes in the
matrix-ber interface, such as matrix thinning, cracks, and ber pull-
out. High-temperature treatments (HS–B and HC-B) led to severe and
widespread matrix and ber damage, while cold thermal treatment at
−20 ◦C (CS–B) resulted in noticeable gaps and cracks. SEM analysis
indicated that untreated samples maintained an intact and undamaged
structure, whereas treated samples, especially those exposed to cyclic
heating, showed signicant structural degradation, such as macro-
fractures and large, widespread cracks.
Thermal treatments signicantly impact the crystallinity and ther-
mal stability of the material, which in turn affects its mechanical per-
formance and structural integrity. The increased brittleness and reduced
mechanical strength of thermally treated samples are directly linked to
the morphological damage observed through SEM and the thermal
behavior characterized by DSC. DSC analysis indicated variations in Tg
among samples subjected to different thermal treatments. Lower Tg
values are associated with decreased polymer chain mobility, suggesting
reduced thermal stability. This stability is crucial as it correlates with
degraded mechanical properties, where materials with lower Tg
demonstrate weaker resistance to deformation under thermal stress.
These ndings emphasize the need for optimized thermal management
in the processing and application of CFRP composites to maintain their
mechanical properties and ensure long-term performance. Understand-
ing the effects of thermal treatments provides valuable insights for the
design and manufacturing of high-performance composite materials.
CRediT authorship contribution statement
Isyna Izzal Muna: Conceptualization, Methodology, Software,
Formal analysis, Investigation, Data curation, Writing – original draft.
Magdalena Mieloszyk: Conceptualization, Methodology, Validation,
Software, Investigation, Resources, Supervision, Funding acquisition,
Project administration, Writing – review & editing. Ruta Rima-
sauskiene: Conceptualization, Methodology, Investigation, Resources,
Table 7
Mechanical parameters, temperature values and the observable degradation
types.
Thermal
group
Tensile
strength
Young’s
Modulus
Tg Degradation types
[MPa] [GPa] [◦C]
Intact 226.14 ±
11.43
28.65 ±
1.14
64.88 Pristine, no polymer
damage/fracture
HS-A 217.99 ±
8.44
25.39 ±
1.45
61.58 Localized micro-crack of
matrix
HS-B 221.21 ±
6.69
23.97 ±
3.54
63.23 Matrix thinning and more
distributed crack of matrix
HC-A 204.41 ±
8.07
20.75 ±
2.55
65.43 Macro-fracture at the middle
area of matrix
HC-B 215.49 ±
9.61
25.34 ±
2.31
64.19 Widespread area of matrix
fracture
CC-A 216.80 ±
10.05
23.99 ±
1.97
64.73 Fibers warping and
separation
CC-B 188.32 ±
8.05
21.60 ±
3.05
64.61 Fibers warping and
nucleated matrix
CS-A 209.36 ±
6.97
23.05 ±
1.01
64.33 No ber pull-out and gaps,
matrix brittleness
CS-B 188.93 ±
8.54
22.67 ±
2.39
62.55 Fibers pull-out and gaps,
matrix crack
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9841
Supervision, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgement
The research was nanced by the National Science Centre, Poland in
the project entitled Thermal Degradation Processes of additively man-
ufactured structures (2019/35/O/ST8/00757), and the Polish National
Agency for Academic Exchange in the Foreign Doctoral fellowship
(PPN/STA/2021/1/00006).
References
[1] Abdullah F, Okuyama Ki, Morimitsu A, Yamagata N. Effects of thermal cycle and
ultraviolet radiation on 3d printed carbon ber/polyether ether ketone ablator.
Aerospace 2020;7:95.
[2] Al Noman A, Kumar BK, Dickens T. Field assisted additive manufacturing for
polymers and metals: materials and methods. Virtual Phys Prototyp 2023;18:
e2256707.
[3] Bettini P, Alitta G, Sala G, Di Landro L. Fused deposition technique for continuous
ber reinforced thermoplastic. J Mater Eng Perform 2017;26:843–8.
[4] De Leon A, Sweat R. Interfacial engineering of cfrp composites and temperature
effects: a review. Mech Compos Mater 2023;59:419–40.
[5] Gracia-Fern´
andez CA, G´
omez-Barreiro S, L´
opez-Beceiro J, Naya S, Artiaga R. New
approach to the double melting peak of poly (l-lactic acid) observed by dsc. J Mater
Res 2012;27:1379–82.
[6] Handwerker M, Wellnitz J, Marzbani H, Tetzlaff U. Annealing of chopped and
continuous bre reinforced polyamide 6 produced by fused lament fabrication.
Compos B Eng 2021;223:109119.
[7] Harris L. A Study of the crystallisation kinetics in PEEK and PEEK composites.
University of Birmingham; 2011. Ph.D. thesis.
[8] Lafarie-Frenot M, Ho N. Inuence of free edge intralaminar stresses on damage
process in CFRP laminates under thermal cycling conditions. Compos Sci Technol
2006;66:1354–65.
[9] Liu Y, Sing SL. A review of advances in additive manufacturing and the integration
of high-performance polymers, alloys, and their composites. Mater Sci Addit Manuf
2023;2:1587.
[10] Maqsood N, Rimaˇ
sauskas M. Characterization of carbon ber reinforced pla
composites manufactured by fused deposition modeling. Composites Part C: Open
Access 2021;4:100112.
[11] Muna II, Mieloszyk M. Temperature inuence on additive manufactured carbon
ber reinforced polymer composites. Materials 2021;14:6413.
[12] Muna II, Mieloszyk M, Rimasauskiene R, Maqsood N, Rimasauskas M. Thermal
effects on mechanical strength of additive manufactured CFRP composites at stable
and cyclic temperature. Polymers 2022;14:4680.
[13] Muravyev NV, Vyazovkin S. The status of pyrolysis kinetics studies by thermal
analysis: quality is not as good as it should and can readily be. Thermo 2022;2:
435–52.
[14] Parmar H, Khan T, Tucci F, Umer R, Carlone P. Advanced robotics and additive
manufacturing of composites: towards a new era in industry 4.0. Mater Manuf
Process 2022;37:483–517.
[15] Pascual-Gonz´
alez C, San Martín P, Lizarralde I, Fern´
andez A, Le´
on A, Lopes C,
Fern´
andez-Bl´
azquez J. Post-processing effects on microstructure, interlaminar and
thermal properties of 3d printed continuous carbon bre composites. Compos B
Eng 2021;210:108652.
[16] Paul R, Dai L. Interfacial aspects of carbon composites. Compos Interfac 2018;25:
539–605.
[17] Perez-Martin H, Mackenzie P, Baidak A, Br´
adaigh CM ´
O, Ray D. Crystallinity
studies of PEKK and carbon bre/pekk composites: a review. Compos B Eng 2021;
223:109127.
[18] Rimaˇ
sauskas M, Kuncius T, Rimaˇ
sauskiene R. Processing of carbon ber for 3d
printed continuous composite structures. Mater Manuf Process 2019;34:1528–36.
[19] Rubino F, Nistic`
o A, Tucci F, Carlone P. Marine application of ber reinforced
composites: a review. J Mar Sci Eng 2020;8:26.
[20] Snyder AD, Phillips ZJ, Turicek JS, Diesendruck CE, Nakshatrala KB, Patrick JF.
Prolonged in situ self-healing in structural composites via thermo-reversible
entanglement. Nat Commun 2022;13:6511.
[21] Su T, Zhou H, Wang R, Chen Y, Luo W, Li M. Effects of working temperature on
mechanical performance and failure characteristics of carbon ber reinforced
plastic and adhesive–repaired titanium structures with central inclined cracks.
J Mater Eng Perform 2023;32:1824–39.
[22] Swanson TD, Birur GC. Nasa thermal control technologies for robotic spacecraft.
Appl Therm Eng 2003;23:1055–65.
[23] Tian X, Liu T, Yang C, Wang Q, Li D. Interface and performance of 3d printed
continuous carbon ber reinforced PLA composites. Compos Appl Sci Manuf 2016;
88:198–205.
[24] Tikhomirova E, Sidokhin Å. Thermal cyclic material tests planning. Strength Mater
2018;50:295–301.
[25] Turnbull D. Formation of crystal nuclei in liquid metals. J Appl Phys 1950;21:
1022–8.
[26] Van Den Einde L, Zhao L, Seible F. Use of frp composites in civil structural
applications. Construct Build Mater 2003;17:389–403.
[27] Vorderbrüggen J, Meschut G. Investigations on a material-specic joining
technology for cfrp hybrid joints along the automotive process chain. Compos
Struct 2019;230:111533.
[28] Zarrelli M, Skordos AA, Partridge IK. Toward a constitutive model for cure-
dependent modulus of a high temperature epoxy during the cure. Eur Polym J
2010;46:1705–12.
[29] Zotti A, Zuppolini S, T´
abi T, Grasso M, Ren G, Borriello A, Zarrelli M. Effects of 1d
and 2d nanollers in basalt/poly (lactic acid) composites for additive
manufacturing. Compos B Eng 2018;153:364–75.
I.I. Muna et al.
Journal of Materials Research and Technology 33 (2024) 9832–9842
9842
