Accepted Manuscript

Mechanical comparison of new composite materials for aerospace applications

C. Barile, C. Casavola, F. De Cillis

PII: S1359-8368(18)30548-1
DOI: [Link]
Reference: JCOMB 6180

To appear in: Composites Part B

Received Date: 15 February 2018

Revised Date: 27 August 2018
Accepted Date: 28 October 2018

Please cite this article as: Barile C, Casavola C, De Cillis F, Mechanical comparison of new composite
materials for aerospace applications, Composites Part B (2018), doi: [Link]
[Link].2018.10.101.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Mechanical comparison of new composite materials

for aerospace applications
C. Barile 1, C. Casavola 1,*, F. De Cillis 1

1Politecnico di Bari – Dipartimento di Meccanica, Matematica e Management – Viale Japigia 182, 70126 Bari,
Italy

PT
*Correspondence: [Link]@[Link]; Tel.: +39-080-5962787

RI
Abstract: Composite materials are becoming the most useful material for aircraft structures. Their
main advantage is connected to the possibility of deeply reducing weight and costs by maintaining
high performances in terms of strength and security. The second major advantage in using this

SC
them depends on the possibility they could be proper designed to guarantee services they are made
to. Many ways to combine them lead to the necessity of planning experimental tests in order to
evaluate the real both elastic and plastic mechanical properties and to compare their variability as

U
function of the fiber type, matrix type and manufacturing technology involved for realizing them.
In this paper, a comparison between two innovative Carbon Fiber Reinforced Plastic materials was
AN
done. They differ, one from each other, for the matrix type (PEEK and BENZOXAZINE) and for the
manufacturing process used to assemble the matrix with the reinforcement (Compression Molding
and Resin Transfer Molding). On the other hand, the resin percentage weight content of both
materials is maintained constant for all the tests: it is 42 % for PEEK matrix and 64 % for
M

BENZOXAZINE matrix. The aim of the work is to critically analyze the results in order to get
useful information for choosing the best one intended for designing and making the back section of
fuselage of a regional aircraft. The component will consist of a front portion with structural aims
D

(zoom phase) and a back part able to withstand to elevated temperatures.

TE

Keywords: Carbon Fibers Reinforced Polymer (CFRP); experimental tests; mechanical properties

1. Introduction
Composite materials, intended as high-performance materials for engineering applications, are
EP

increasing in use in the last years [1]. They are non-homogeneous and anisotropic materials which
require special consideration for determining physical and mechanical properties respect to metal
ones [2]. The most important benefit respect to the metal alloys is connected to the possibility of
C

designing them ad hoc for the specific application they are intended to. This feature is strictly
connected to the nature of composite materials. In fact, they consist at least of two separate phases
AC

which contribute together to the final properties: the reinforcement and the matrix. Differently from
metallic materials, the “parts” of composites remain distinct from each other at the macroscopic level
and are strongly affected by the damage mechanism of plies.
Several technological processes and lay-up distributions are considered for each particular
application. The aim is to get the extreme structural exploitation of fibers matching the industrial
requirements in terms of performance. The use of composite materials in aerospace field is
representative of the elevated degree of complexity in mechanical design, because almost unlimited
combinations of matrix and fiber patterns exist, but also because they could fail at not predictable
loads neither by perfectly elastic nor perfectly plastic theories [3]. Consequently, experimental tests
continue to play a significant role in the qualification process of new composite materials by using
traditional [4] or hybrid techniques [5]. Aircraft structural design consists of various levels of
structural testing, starting from specimen tests and finishing to full-scale structure tests [6]. For this
reason, it becomes important to plan a full experimental campaign considering different kind of tests
 ACCEPTED MANUSCRIPT
but also different kind of composite materials, in order to choose the proper combination
material/lay-up ensuring the best outcomes.
Tensile and compression tests allow to evaluate the most representative properties of
composite materials to be used in aircraft structures. Many of these tests must be arranged in a good
experimental campaign. In general, the laws of deformation and failure of Carbon Fiber Reinforced
Polymers (CFRPs) are described adequately in many studies, generalized in handbooks and reviews
[7-9]. Anyway, not many papers focus the attention on the analysis of mechanical response by
comparing composite materials made by the same kind of fibers combined with different matrices
and/or different manufacturing technologies. The latter also depends on the nature of matrix

PT
(thermoset or thermoplastic), whose choice could deeply affect the final properties.
In this paper, a comparison between two CFRPs has been carried out: CARBON/PEEK and
CARBON/BENZOXAZINE. In particular, they are made by applying Compression Molding (CM)

RI
technology on the thermoplastic material, and Resin Transfer Molding (RTM) technology for the
thermoset one. Tensile, compressive and open hole tensile tests have been carried out on specimens
realized by using the two materials. The aim is to investigate the laws of deformation and failure of

SC
layered CFRPs in order to define the better solution, in terms of performance required for the
engineering industry [10], and especially for the aerospace one. Results getting from the
experimental campaign will be used by the companies involved for validating the design phase of
the back section of fuselage of a regional aircraft, consisting of a front portion with structural aims

U
(zoom phase) and a back part able to withstand to elevated temperatures. These considerations limit
the possibility of widely share the data, according to a Non-Disclosure Agreement (NDA) signed by
AN
all the companies and research centers involved in the project.

2. Materials and Methods

M

Two innovative composite materials are experimentally tested. They are made as follows:
1. Carbon / PEEK (HTA40 - 5HS) hereinafter named Material 1 – it is a laminate composite with
thermoplastic matrix PEEK (density 1,30 g/cm3) and High Tenacity carbon fibers - Density
1.76 g/cm3 - with 3000 fibers per tow and "Five Harness Fabric" weaving; manufacturing
D

technology - Compression Molding (Isothermal compression molding at 390 °C and 20 bar).

2. Carbon / BENZOXAZINE (5HS HTA40 / BZ9110) hereinafter named Material 2 - it is a
TE

laminate composite with thermosetting matrix BENZOXAZINE BZ9110 (density 1,22 g/cm3)
and High Tenacity carbon fibers - Density 1.76 g/cm3 - with 12000 fibers per tow and "Five
Harness Fabric" weaving; manufacturing technology - Resin Transfer Molding (resin
EP

injection at 110 °C and 8 bar, and curing treatment for 90 minutes at 180 °C).
The table below shows the properties of both materials (Tab. 1).

Table 1. Material properties of 2 CFRP

C

Tensile Tensile Filament Number

AC

MATERIAL Density Manufacturing

Fiber Strength Modulus diameter Matrix of fiber Weaving
3
PROPERTIES [g/cm ] Technology
[MPa] [GPa] [µm] per tow

TENAX Thermoplastic Compression

Material 1 3950 238 7.0 1.76 3000 5HS
HTA-40 PEEK Molding (CM)

Thermosetting
TENAX Resin Transfer
Material 2 3950 238 7.0 1.76 BENZOXAZINE 12000 5HS
HTA-40 Molding (RTM)
BZ9110

CM is a reliable manufacturing process that uses metal molds with the application of external
pressure. In the compression molding process, an engineered composite lay-up is placed in the open
mold cavity, and when it is closed a consolidating force is applied. The pressure was continuously
 ACCEPTED MANUSCRIPT
kept constant during the whole cure cycle, occurring in an oven. The combination of heat and
pressure produces a composite part with low voids content and high fibers volume fraction, whose
finishing surface is similar to a net shape. Compression molding can yield composite parts having
the optimal mechanical properties possible depending on the combination of constituent phases.
This type of molding process allows to manufacture complex geometries components (e. g. holes)
without requiring machined post-mold processes. This technique produces good integrated
composite structure (carbon fiber, aramid fiber or fiberglass) needing of fewer knit lines and lower
fiber-length degradation respect to the injection molding.
RTM is an increasingly used type of molding, mainly applied to mold components having large

PT
surfaces, complex shapes and smooth finishes. It is a low pressure closed molding process for low
volume production. It represents a compromise between the slower contact molding processes and
the faster compression molding process, which requires high tooling costs. Continuous strand mats

RI
and woven reinforcement are laid dry in the mold bottom half. The mold is clamped, and a low
viscosity resin is pumped with the aim of pushing put the air. Common matrix resins include
polyester, vinyl ester, epoxy, and phenolics. Advantages respect to contact molding methods are

SC
connected to the possibility of obtaining uniform thicknesses, two finished sides and low emissions.
For optimum surface finishing, a gel coat could be applied to the mold surface prior to molding.
High quality parts produced by this method include automotive body parts, bathtubs, and
containers.

U
Experimental campaign consists in testing 40 specimens carried out by using the two different
CRFP materials. Specimens obtained by the two materials were subjected to three types of
AN
mechanical tests, in order to determine tensile, compression and open-hole tensile properties. All
tests were performed by using a 250 kN servo-hydraulic testing machine, with hydraulic clamping.
M
D
TE
C EP
AC

Figure 1. Compression Modulus: Test Setup

Tensile tests were performed according to standard ASTM D3039 [11]. Five specimens of
Material 1 and five of Material 2 were tested at a constant crosshead displacement equal to 2 mm/min.
Material 1 laminate consists of 8 plies assembled according to [(0,90)4]s lay-up. Material 2 laminate
consists of 6 plies assembled according to [(0,90)3]s lay-up. All the tested specimens were cut at 0 °
respect to the load direction. Specimens were long 250 mm and had a rectangular cross-section for
both materials. In particular, the section had dimensions 25 mm x 2,48 mm for Material 1 and 25 mm
 ACCEPTED MANUSCRIPT
x 2,35 for Material 2. The different thicknesses are connected to the standard requirements. Their
difference doesn’t affect the test results because the resin percentage weight content for both
materials is maintained constant for all the tests: it is 42 % for PEEK matrix and 64 % for
BENZOXAZINE matrix. The specimens were clamped in testing machine by using resins tabs
according to the standard suggestions [11]. Two strain gauges for each side of specimens were
applied to monitor strain behavior. On each side, one was positioned half-way axially and the
second was applied transversely, according to the standard [11].
Compression tests were performed according to SACMA SRM-1 [12]. Ten specimens for each
material were tested at a constant crosshead displacement equal to 1 mm/min. Five were cut at 0 °

PT
and five at 90 ° respect to the load direction. For both the materials, the number of plies was 8 and
the laminate lay-up was assembled according to [(0,90)4]s scheme. Specimens were long 80 mm and
had a rectangular cross-section for both materials. In particular, the section had sizes 15 mm x 2,48

RI
mm for Material 1 and 15 mm x 3,04 mm for Material 2. A standard loading system was used. It was
an anti-buckling compression system positioned between two flat surfaces firmly connected to the
testing machine grips (Fig. 1). In order to define the compression modulus one strain gauge (1.5 mm

SC
base, 350 Ohm) was positioned at the middle section of test samples which didn’t have any tabs [12].
Open hole tensile tests were carried out in accordance with ASTM D5766/D5766M [13]. Five
specimens of each material were tested at a constant crosshead displacement equal to 2 mm/min.
(Fig. 2). They were cut at 0° respect to the load direction and had the same lay-up [(0,90) ] .

U
Specimens were 200 mm long and had a rectangular cross-section holed at the middle-length for
both materials. In particular, the section had sizes 36 mm x 2,48 mm for Material 1 and 36 mm x 3,04
AN
mm for Material 2. All the holes had a 6 mm diameter in order to respect the suggested standard
ratios D/h (diameter/thickness) between 1,5 and 3, and w/D (width/diameter) equal to 6.
All test specimens have a size and geometry according to indications provided by the
corresponding standard.
M
D
TE
C EP
AC

Figure 2. Open Hole Tensile: Test Setup

3. Results
Results obtained are critically compared according to the specific test they refer to in order to
define the best material for the specific purpose.
 ACCEPTED MANUSCRIPT
Figure 3 reports the complete test evolution of the five samples of each material in order to
underline the tests repeatability.
Figure 4 and 5 show a graphical comparison of the two CFRPs subjected to tensile load. They
report the stress-strain behavior of one representative sample of each material and the mean value of
elastic properties of both them. It is possible to observe that: the ultimate tensile strength as well as
the Poisson ratio are higher for Material 1 than Material 2, conversely the Young modulus is slightly
greater for Material 2 than Material 1.

PT
RI
U SC
AN
M
D
TE
C EP
AC

Figure 3. Tensile trends of Material 1 and Material 2

These results seem to asses that the type of resin affects the elastic properties (E) and
alternatively the manufacturing process influences the plastic properties (σ). Generally, it is known
that the thermosetting resins (Material 2) is characterized by higher resistance and hardness than the
thermoplastic ones (Material 1), as confirmed by [14]. Namely Material 1 is more ductile than 2, that is
to say it has a lower Young modulus and at the same time a higher Poisson ratio by combining a
higher rate of transversal contraction with a less pronounced longitudinal elongation. On the other
hand, the unexpected attitude of ultimate tensile strength could be explained only by considering
 ACCEPTED MANUSCRIPT
the different manufacturing process of composite laminates. As previously described, in fact, the
Material 1 was obtained by using a CM process, while the Material 2 was realized by means of RTM
one. The main difference between them refers to the pressure level of process. RTM involves only
low pressure during the injection which helps in obtaining more uniform components and thus
inducing high mechanical strength [10].

PT
RI
U SC
AN
M

Figure 4. Tensile Test: Graphical comparison

In order to better show the differences between both materials, the numerical values of tensile
results are reported in Table 2. All values indicate the error and the percentage difference of Material
D

1 respect to Material 2 too.

TE

Table 2. Numerical results of tensile tests

Maximum Tensile Young Modulus

TENSILE TEST Poisson's Ratio
Stress [MPa] [MPa]
EP

Material 1 898.3 ± 51.9 59790 ± 576 0.073 ± 0.010

C

Material 2 739.8 ± 15.7 62949 ± 589 0.059 ± 0.004

Error (%)
AC

21% 5% 24%
1 over 2

The mode and the location of failure of all the specimens are reported in Figure 6 and 7. The
predominant type of failure is codified as Lateral (L) At grip/tab (A) Top (T) or Bottom (B) for both
materials, according to the standard typical modes [11]. Another observation concerns with the
percentage type of failures for both materials: it seems that Material 1 predominant failure is located
at top or at the bottom of the specimen (4 specimens over 5 show LAT or LAB failure), on the other
hand for Material 2 the predominant one is located at the gage section (5 specimens over 5 show a
breakage in correspondence of the middle section).
 ACCEPTED MANUSCRIPT

PT
RI
SC
Figure 5. Tensile Test: Mechanical Properties Comparison

U
AN
M
D
TE
EP

Figure 6. Tensile Test: Carbon/PEEK failure modes of specimens

C

Looking at the compression tests, the stress-strain trend seems to be quite similar for both
materials (Fig. 8 e 9). Anyway, by analyzing the compressive moduli, it is possible to note that a
AC

perfect isotropic behavior is accounted only for Material 1, despite Material 2 (Fig. 10). The difference
of modulus values is neglectable in the first case and is about 6000 MPa for the second one. This
unforeseen result suggests a higher compressive stiffness in 90° respect to 0°, indicating the presence
of some effects able to improve the mechanical strength only in one direction, even if the laminate is
balanced and symmetric. Considering the phases composition of two materials, the only difference
between them is connected to the type of matrix and manufacturing process.
 ACCEPTED MANUSCRIPT

PT
RI
SC
Figure 7. Tensile Test: Carbon/Benzoxazine failure modes of specimens

Another important consideration refers to the comparable values of compressive and tensile

U
moduli of both CFRP materials. According to the scientific literature, the basic assumption of CFRP
tape is based on the statistics that compressive modulus is fraction lower than the tensile one [15].
AN
This attitude is disproved for CFRP fabrics. In particular, by using a Five Harness Satin Fabric it is
possible to improve the compressive properties up to the tensile one, ensuring better performances
for specific purpose.
M
D
TE
C EP
AC

Figure 8. Compression Test 0°: Graphical Comparison

 ACCEPTED MANUSCRIPT

PT
RI
U SC
Figure 9. Compression Test 90°: Graphical Comparison
AN
M
D
TE
C EP
AC

Figure 10. Compression Test: Modulus Comparison

In order to better shows the differences between both materials, the numerical values of
compressive results are reported in Table 3. All values indicate the error and the percentage
difference of Material 1 respect to Material 2 too.
The type of failure of all the specimens are reported in Figure 11 and 12. The characteristic type
of failure is a crush localized indifferently at up or bottom terminal edge of the specimens for both
materials.
 ACCEPTED MANUSCRIPT
Table 3. Numerical results of compressive tests
COMPRESSIVE Compressive Compressive
MODULUS TEST Modulus 0° [MPa] Modulus 90° [MPa]

Material 1 60415 ± 7377 59099 ± 3654

Material 2 61797 ± 2009 68325 ± 4686

Error (%)
2% 14%

PT
1 over 2

RI
U SC
AN
M

Figure 11. Compression Test: Carbon/PEEK failure modes of specimens 0° and 90°
D
TE
C EP
AC

Figure 12. Compression Test: Carbon/Benzoxazine failure modes of specimens 0° and 90°

Finally, open hole tensile results are compared (Fig. 13). Due to the presence of the hole, tensile
strength decreases: comparison of samples, characterized by the same lay-up and design parameters
but without holes, indicates that a decreasing of 53% for Material 1 and of 31% for Material 2 in
ultimate tensile strength happen. The percentage decreasing, between unnotched and notched
samples have been commonly observed by many researchers [16-17].
 ACCEPTED MANUSCRIPT

PT
RI
U SC
Figure 13. Open Hole Tensile Test: Graphical comparison
AN
M
D
TE
C EP

Figure 14. Open Hole Tensile Test: Carbon/PEEK failure modes of specimens
AC

By comparing the two CFRP materials the main aspect that need to be underlined refers to the
different effects the hole induces in the mechanical strength. It results, in fact, that the ultimate open
hole tensile strength of Material 1 is lower than Material 2, and this is in contrast with the results
obtained in ultimate tensile strength. It seems that the hole manufacturing largely affects the
mechanical properties, by reducing the composite strength according with the resin nature.
In order to better shows the differences between the materials, the numerical values of
open-hole tensile results are reported in Table 4. All values indicate the error and the percentage
difference of Material 1 respect to Material 2 too.
 ACCEPTED MANUSCRIPT

Table 4. Numerical results of open-hole tensile tests

Ultimate Open-Hole Young Modulus

OPEN-HOLE TEST
Strength (MPa) (MPa)

Material 1 431.9 ± 14 56926 ± 1481

PT
Material 2 508.8 ± 8.7 57540 ± 1133

Error (%)
18% 1%
1 over 2

RI
The type of failure of all the specimens are reported in Figure 14 and 15. The predominant type
of failure is codified as Lateral (L) Gage (G) Middle (M), according to the standard typical modes

SC
[13].

U
AN
M
D
TE
EP

Figure 15. Open Hole Tensile Test: Carbon/Benzoxazine failure modes of specimens
C

4. Conclusions
AC

In this paper, a comparison of two CFRP materials was carried out. They were built by
combining the same fibers with two types of resins and two manufacturing processes. Three kinds of
experimental tests were planned for defining the better combination materials/manufacturing
process: tensile, compressive and open-hole tensile tests. Results seems to show some beneficial
effects on mechanical tensile properties due to the technological process. Despite the expected
behavior due to the thermosetting matrix, the ultimate tensile strength as well as the Poisson ratio
are higher for Material 1 than Material 2. The type of resin affects only the elastic properties (E).
On the other hand, the manufacturing methods appear to differently affect the compressive
behavior of symmetric and balanced composites along two different load directions. The results
suggest a higher compressive stiffness in 90° respect to 0° for Material 2, indicating the presence of
some beneficial effects of RTM process able to improve the mechanical strength only in one
direction, even if the laminate is balanced and symmetric.
 ACCEPTED MANUSCRIPT
Then open tensile tests didn’t show any influences of technological process respect to the real
properties of composite resins. It seems that the manufacturing hole largely affects the mechanical
properties, by reducing the composite strength according with the resin nature.
This paper opens the way to more deep investigations for these two materials, as well as for the
unlimited combinations of fibers, matrices, lay-ups and manufacturing processes. Now researchers
involved in this project are focusing the attention on the fatigue tests for different configuration
loads. It represents an interesting research fields not only for the scientific community but especially
for the industrial companies, whose aim is to produce aircraft by ensuring high performance
combined with elevated safety requirements. It is known that the great advantages coming from the

PT
use of composite materials move in the same direction of the problems derived from not visible
damages induced by manufacturing processes in combining the two phases. The latter are
responsible of the non-predictable structure collapse.

RI
References

SC
1. Di Leif Carlsson A, Donald Adams F, Byron Pipes R. Experimental Characterization of Advanced
Composite Materials, Fourth Edition.
2. Tong L, Mouritz AP, Bannister MK. 3D fiber reinforced polymer composites. Ed. Elsevier, 2002.
3. Kucher NK, Zemtsov MP, Zarazovskii MN. Deformation behavior and strength of unidirectional carbon

U
fiber laminates. Mech Compos Mater 2006;42(5):407-418.
4. Barile C, Casavola C, Pappalettere C. The influence of stitching and unconventional fibres orientation on
AN
the tensile properties of CFRP laminates. Compos Part B: Eng 2017;110:248-254.
5. Barile C, Casavola C, Pappalettera G, Pappalettere C. Hybrid characterization of laminated wood with
ESPI and optimization methods. Conference Proceedings of the Society for Experimental Mechanics
Series, 2013;3:75-83.
M

6. Fan H, Vassilopoulos AP, Keller T. Experimental and numerical investigation of tensile behavior of
non-laminated CFRP straps. Comp Part B: Eng 2016;91:327–336.
7. Cox B, Flanagan G. Hanbook of analytical methods for textile composites. NASA Contractor Report 4750.
D

Langley Research Center, 1997.

8. Cantwell WJ, Morton J. The impact resistance of composite materials-a review. Compos 1991;22:347–362.
TE

9. Mouritz AP, Leong KH, Herszberg I. A review of the effect of stitching on the in-plane mechanical
properties of fibre-reinforced polymer composites. Compos Part A-Appl S 28A 1997:979–991.
10. Idicula M, Sreekumar PA, Kuruvilla J, Sabu T. Natural Fiber Hybrid Composites—A Comparison Between
Compression Molding and Resin Transfer Molding. Polym Composites 2009;30(10) DOI 10.1002/pc.20706
EP

11. ASTM D3039/D3039M “Standard Test Method for Tensile Properties of Polymer Composite Materials”,
ASTM International, West Conshohocken, PA, 2017, [Link].
12. ASTM D5766/D5766M “Standard Test Method for Open-Hole Tensile Strength of Polymer Matrix
Composite Laminates”, ASTM International, West Conshohocken, PA, 2011, [Link].
C

13. SACMA SRM-1R-94 “Compressive Properties of Oriented Fiber-Resin Composites”.

14. Ma Y, Yang Y, Sugahara T, Hamada H. A study on the failure behavior and mechanical properties of
AC

unidirectional fiber reinforced thermosetting and thermoplastic composites. Compos Part B: Eng
2016;99:162-172.
15. Meng M, Le HR, Rizvi MJ, Grove SM. The effects of unequal compressive/tensile moduli of composites,
Compos Struct 2015. doi: [Link]
16. Barile C, Casavola C, Pappalettere C, Tursi F. RFI composite materials behaviour. Ninth Meet “New
Trends Fatigue Fracture - NT2F9” 2010;10(3):209e13.
17. Zheng Y, Cheng X, Yasir B. Effect of stitching on plain and open-hole strength of CFRP laminates. Chin J
Aeronaut 2012;25:473-484.