Accepted Manuscript Effect of temperature on static and low velocity impact properties of thermoplastic composites Luigi Sorrentino, Davi Silva de Vasconcellos, Marco D'Auria, Fabrizio Sarasini, Jacopo Tirillò PII:
S1359-8368(16)32400-3
DOI:
10.1016/j.compositesb.2017.01.010
Reference:
JCOMB 4821
To appear in:
Composites Part B
Received Date: 22 October 2016 Revised Date:
9 December 2016
Accepted Date: 8 January 2017
Please cite this article as: Sorrentino L, de Vasconcellos DS, D'Auria M, Sarasini F, Tirillò J, Effect of temperature on static and low velocity impact properties of thermoplastic composites, Composites Part B (2017), doi: 10.1016/j.compositesb.2017.01.010. 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
Effect of Temperature on Static and Low Velocity Impact Properties of Thermoplastic Composites
1
RI PT
Luigi Sorrentino1,*, Davi Silva de Vasconcellos1, Marco D'Auria1, Fabrizio Sarasini2, Jacopo Tirillò2
Istituto per i Polimeri, Compositi e Biomateriali, Consiglio Nazionale delle Ricerche, P.le E. Fermi 1, 80055 Portici (NA), Italy 2 Dipartimento di Ingegneria Chimica, Materiali e Ambiente and UDR INSTM, Università di Roma La Sapienza, Via Eudossiana 18, 00184, Roma, Italy
+390817758850
M AN U
ABSTRACT
SC
*Corresponding author (Luigi Sorrentino):
[email protected], Tel. +390817758844, Fax
In this work, thermoplastic composites based on poly(ethylene 2,6-naphthalate) (PEN) have been investigated with the aim to elucidate the effect of temperature on static and impact properties. The matrix was reinforced with four different high performing woven fabrics based on carbon, Twaron, Vectran, basalt fibres. Composites were manufactured by using the film stacking technique, alternating layers of balanced plain weave fabrics (0/90) and films of amorphous PEN, keeping the
TE D
fibre volume fraction around 40%. The compression moulding process was optimized to obtain optimal fabric impregnation and to keep the void content lower than 1%. The structural response was evaluated at 20 °C, 60 °C and 100 °C by means of static flexural and low velocity impact tests. Dynamic mechanical scans (DMA) were also performed to evaluate the stiffness of the laminates at
EP
temperatures ranging from – 80 °C to 230 °C. The flexural modulus and strength of laminates resulted to be very high in proportion with the fibre stiffness. The flexural behaviour was affected by the
AC C
temperature but a limited reduction of the stiffness (lower than 20%) was exhibited by all but PEN/Vectran up to temperatures of 100 °C. PEN/carbon resulted the best performing system at each testing temperature (flexural modulus equal to 38.8 GPa, flexural strength equal to 714 MPa at 20 °C). Results from DMA tests demonstrated a satisfactory agreement with static tests (differences within ± 13%, except that for PEN/Vectran). The impact resistance of the composites was found to depend on the fibre type, temperature and interface strength. PEN/basalt resulted the best performing composite followed by PEN/Twaron (perforation energy at 20 °C: 44.3 J and 38.2 J, respectively). The PEN/carbon laminate exhibited the lowest impact performance but, unlike PEN/basalt and PEN/Vectran, the impact resistance improved with the increasing temperature.
1
ACCEPTED MANUSCRIPT Keywords: A. Thermoplastic resin; A. Polymer-matrix composites (PMCs); B. High-temperature properties; B. Impact behaviour
1. INTRODUCTION The increasing usage of composite materials to replace metal alloys in structural components
RI PT
in the aerospace, automotive and other demanding industrial applications has been dominated by thermosetting matrices. Although the thermosetting systems usually exhibit superior tensile, shear and compressive strengths, several problems have been pointed out related to thermoset-based composites, namely, inferior performance in damage tolerance and hot/wet
SC
stability. This limited stability has been highlighted for the three widely used thermosetting matrices, namely epoxy, polyester and vinyl ester. For instance, usually the matrix in almost
M AN U
all composites for seawater applications is polyester or vinyl ester resin even if polyester and the related fibre/matrix interface typically suffers degradation by hydrolysis reaction of unsaturated groups within the resin [1–4] whilst chemical degradation in vinyl ester-based composites is present but to a much lower extent [2,5]. Significant progress has been made and is still in progress [6–11] for improving the fracture toughness and damage tolerance properties of thermosetting systems but other problems associated with hot/wet stability and
TE D
manufacture cost still remain largely unresolved. These specific needs have focused attention on the potential use of thermoplastic matrix systems also triggered by the continuous development of high performance thermoplastic polymers. Thermoplastic matrix composites exhibit distinct advantages over thermosetting ones in terms of recyclability, high specific
EP
strength and stiffness, corrosion resistance, enhanced impact toughness, cost effectiveness and flexibility of design [12]. For high demanding applications, in earlier work in fiber reinforced
AC C
thermoplastic composites, amorphous polyethersulphone (PES) and polyetherimide (PEI) were commonly used via solvent impregnation but it was easily recognized that solvent resistance was a key issue for aerospace applications and the development shifted to semicrystalline polymers such as polyether ether ketone (PEEK), and polyether ketone ketone (PEKK), poly phenylene sulphide (PPS) which allowed additional manufacturing routes such as film/fabric stacking, melt or powder impregnating processes. These high performance thermoplastic matrix systems show high glass transition temperatures due to the stiff cyclic chains in the polymer backbone and are able to meet a wide range of end-use temperature applications. Due to both high Tg and high melt viscosity, a relatively high processing temperature is needed to consolidate the related composites and therefore are more difficult to 2
ACCEPTED MANUSCRIPT process than most other thermoplastics. Among high performance thermoplastic polymers, poly(ethylene 2,6-naphthalate) (PEN) exhibits higher affordability with a sufficiently high Tg (~120°C) and a melting temperature of about 265 °C. This lower processing temperature, which enables reduced energy requirements during composite manufacturing coupled with its semicrystalline nature, good thermal stability [13], low moisture absorption [14] and solvent
RI PT
resistance, have recently drawn attention to this polymer as a matrix for high performance composites [15–22]. The possibility of using thermoplastic based composite materials in loadbearing and high demanding applications where the service temperatures are likely to be higher than 80 °C, has stimulated the need for data on such materials at these elevated
SC
temperatures. This aspect is particularly important because many mechanical properties of the resulting composites are matrix-dominated, such as flexural and compression strengths, interlaminar fracture toughness and/or compression-after-impact performance. Especially the
M AN U
response to low velocity impact is one of the most detrimental loadings for laminates. Most of the studies about the impact performance and damage tolerance of thermoplastic-based composites deal with PEEK-based composites [23–28] and only few authors reported the impact behavior of PPS-based laminates [23,27–29]. The response to impact loading is also significantly influenced by the temperature because polymers can behave in a more ductile or
TE D
brittle way. In literature, there are few studies focused on the impact response of composites at low and/or high temperatures and most studies deal with thermosetting-based composites [28,30–36]. Ibekwe et al. [37] investigated the velocity impact behavior and the residual load carrying ability of E-glass reinforced epoxy laminated composite with unidirectional and
EP
cross-ply stacking sequences. Temperature showed a significant effect on the low velocity impact responses of laminated composites with more impact damage induced in specimens impacted at lower temperatures. Gómez-del Río et al. [31] examined the response of carbon
AC C
fibre-reinforced epoxy laminates at low impact velocity and in low temperature conditions (from -20 to 150 °C). The results showed that the effect of decreasing temperature was similar to an increase in impact energy causing larger matrix cracking and delamination extension, deeper indentation and more severe fibre–matrix debonding and fibre fracture on the opposite side. In addition the authors highlighted also the important role played by the stacking sequence with more severe damage in cross-ply and quasi-isotropic laminates compared to plain woven ones. Bibo et al. [30] assessed the impact performance of carbon fibre reinforced epoxy composites when subjected to low velocity impacts from room temperature up to 150 °C. The authors found that the effect of increasing temperature, although lower than the glass transition temperature, did not influence the penetration impact resistance of laminates whilst 3
ACCEPTED MANUSCRIPT it had an effect for non-penetrating impacts, increasing the size of delaminations damage. Hirai et al. [38] investigated the impact resistance and compression-after-impact of vinyl ester-matrix composites reinforced with woven E-glass fabric in the temperature range from 65°C to 100°C. The authors found that the variation in matrix properties caused by different temperatures resulted in significant changes not only in the load/displacement characteristics
RI PT
but also in the extent of damage. The overall impact response was dominated by the poor mechanical properties of the matrix at high temperatures which reduced the impact-damage resistance and damage tolerance of the laminates. The effect of temperature on the low velocity impact response of thermoplastic based composites has received even less attention
SC
in literature [30,39,40]. In this framework, the aim of this work is to investigate the effect of temperature on the static and low velocity impact properties of high performance thermoplastic composites based on a PEN matrix meant to be used up to 100 °C. Due to the
M AN U
lower processing temperature, PEN allows also the possibility to be reinforced by polymer fibres, thus limiting the detrimental effects on fibres due to high temperature exposure. In this regard PEN matrix has been reinforced by four different high performing woven fabrics, namely carbon, basalt, Vectran and Twaron. Since the impact resistance of composites depends significantly also on the interfacial strength of the composites, the different PEN-
TE D
based composites investigated in the present study possess the versatility to be tailored for different applications in terms of strength, stiffness and impact response. 2. MATERIALS AND METHODS
EP
2.1. Raw materials
The thermoplastic matrix used for the laminate preparation is the Poly(ethylene 2,6-
AC C
naphthalate) homopolymer (PEN) (Teonex TN8065S from Teijin Kasei, Japan) with density of 1.35 g/cm3 (at 25°C), a glass transition temperature of 120°C and melting temperature of 265°C. Amorphous PEN presents maximum flexural strain of 4.94±0.01%, flexural modulus of 2.20±0.09 GPa and maximum flexural strength 93.4±0.8 MPa, all evaluated with threepoint bending tests [21].
The carbon fibre woven fabric (type CC160P) was supplied by SAATI S.p.a., Italy. This is a plain weave fabric with surface weight of 160 g/m2 and yarn count of 4.0 ends/cm (warp and weft). According to the supplier data-sheet, the employed carbon fibres are high strength carbon fibre 3K 2000 dtex. These fibres are characterized by a density of 1.76 g/cm3, an
4
ACCEPTED MANUSCRIPT average diameter of 7 µm, Young's modulus of 235±5 GPa, and tensile strength of 4.1±0.3 GPa. Basalt fibres (BAS 220.1270.P) were supplied by Basaltex-Flocart NV, Belgium in the form of a plain weave fabric with surface weight of 220 g/m2, yarn count of 7.2 ends/cm (warp and weft) and nominal thickness of 0.13 mm. According to the supplier data-sheet, the basalt
RI PT
fibres employed (KVT150tex13-I) have a density of 2.67 g/cm3, an average diameter of 13±1.5 µm, Young's modulus of 85±2 GPa and melting point of 1350±100 °C.
Woven fabrics based on aramid and Vectran fibres were both supplied by G. Angeloni srl, Italy. The aramid reinforcement (type KK220P) is a plain weave fabric with surface weight of
SC
224 g/m2 and yarn count of 6.5 ends/cm (warp and weft). According to the supplier datasheet, the aramid fibres employed are standard Twaron 2200, manufactured by Teijin (Japan) with density of 1.45 g/cm3, Young's modulus of 70±10 GPa and melting point of 500 °C.
M AN U
Vectran based fabric (type VcT200T) is a 2x2 twill weave fabric with surface weight of 201 g/m2 and yarn count of 6.0 ends/cm (warp and weft). The Vectran HT fibres, manufactured by Kuraray – Japan, have a density of 1.4 g/cm3, an average diameter of 23 µm, Young's modulus of 75±7 GPa, tensile strength of 3.2 GPa and maximum service temperature of 250°C.
TE D
2.2. Composite manufacturing
Thin films of PEN were prepared by compression moulding from pellets pressed at 300 °C and 50 bars for 5 minutes, by using a hydraulic hot plate press (model P 300P, Collin GmbH,
EP
Germany). Films with an average thickness of 150 µm were produced and quenched to keep the polymer in the amorphous state. All materials (PEN pellets, films and reinforcing fabrics) were dried in oven by applying vacuum at 120°C for at least 6h. Neat PEN plates have been
AC C
produced as thermal reference for DMA tests by using the same temperature and pressure profiles used for the laminates (see below). Composite plates where prepared by using the film stacking process. Layers of reinforcing fabric are alternatively stacked (0/90) between layers of PEN films in a square mould with fixed dimensions (160 mm x 160 mm as length and width, respectively) and then pressed by using the hydraulic hot plate press. Temperature and pressure profiles were optimized to obtain a very good impregnation of fibres. The processing sequence consists of a heating step at 300 °C under a pressure of 0 MPa for two minutes in order to melt the PEN film, followed by a compression step at 300 °C and 2.54 MPa for five minutes to force the fabric’s impregnation. At the end of the compression step, the mould with the composite plate is 5
ACCEPTED MANUSCRIPT quenched to hinder the formation of crystals in the polymer and obtain an almost amorphous matrix for the preparation of the laminates. The need for keeping PEN in the amorphous state is due to the fact that the presence of crystals increases the local density of the polymer and induces residual stresses that can generate microcracks in the laminate during the cooling process. Laminates exhibited a very low void content and good fabric impregnation, as
RI PT
confirmed by the morphological analysis performed with the SEM and reported elsewhere [41].
The main characteristics of each laminate configuration (mean values and variance of fibre volume content, density and thickness) are reported in Table 1. Due to the different surface
SC
weights of fabrics, some laminate configurations showed a lower thickness. In particular, carbon and basalt based composites showed a significantly lower thickness with respect to aramid and Vectran based laminates. Since the composite thickness has a strong influence on
M AN U
impact properties[42–44] further composite plates based on carbon and basalt fabrics were produced with a higher number of fabric layers (but keeping constant the fibre volume content) in order to have a homogeneous thickness among all tested systems (Table 1). Samples for all characterizations were cut from composite plates by using a milling machine. Cut surfaces were finished with a polishing paste to smooth the edges.
TE D
2.3. Characterization procedures
A morphological analysis with a scanning electron microscope (SEM) has been performed on samples to investigate the impregnation quality of the production process and the damage
EP
modes of samples after the mechanical characterization. SEM observations with a Quanta 200 FEG from FEI (Eindhoven, The Nederlands) were performed on finely polished cross sections, sputter coated with gold/palladium. The composite density was evaluated by means
AC C
a hydrostatic density measuring apparatus (balance AB104-S with density determination kit supplied by Mettler-Toledo International Inc. – USA) on specimens 43 mm long and 12 mm wide. The specimen thickness is reported in Table 1, along with the mean value and the standard deviation of composite density. The void volume content has been calculated from the densities of components according to the ASTM D2734 standard. Static three-point bending tests were performed according to the ASTM D790 standard, by means of a universal testing machine (model 4304 from SANS – China, now MTS – USA) equipped with a 30 kN load cell. Flexural tests were performed at 20°C, 60°C and 100°C, after conditioning the samples for 60 minutes. The loading nose and supports of the threepoint bending fixture had cylindrical surfaces with radius of curvature equal to 5 mm. Five 6
ACCEPTED MANUSCRIPT samples for each laminate configuration, having dimensions of 100 mm x 12.7 mm (length x width) were tested. The span-to-depth ratio used was 32 to 1. The strain rate of the outer fibres during the laminates testing was equal to 0.01 min-1. Load-deflection curves were plotted to determine average values and standard deviations of flexural strength (σB), flexural modulus of elasticity (EB) and strain at break (εB).
RI PT
Thermo-dynamic mechanical properties of both matrix and composites were evaluated with a dynamic mechanical analyzer (TRITEC 2000 DMA, from Triton Technology, UK) in
accordance with ASTM D 5023. DMA tests were performed with a three-point bending configuration (dynamic displacement amplitude of 50 µm, frequency of 1 Hz) in a
SC
temperature range from -80°C to 230 °C (or until the occurrence of signal drop) and using a heating rate of 2 °C/min. Neat PEN specimens were cut with dimensions of 32 mm x 10 mm x 1.0 mm and a support span of 20 mm was used (span-to-depth ratio of 20:1). Composite
M AN U
specimens for DMA characterization were 43 mm long and 12 mm wide. The thickness of each laminate is reported in Table 1. The support span was set equal to 25 mm, thus assuring a minimum span-to-depth ratio of 16:1.
Impact tests were performed by using a falling dart impact testing machine, model Fractovis Plus from CEAST (Pianezza – TO, Italy) with a hemispherical impact head (diameter equal to
TE D
12.7 mm). Specimens had dimensions of 60 mm x 60 mm and were clamped on a circular support with inner diameter of 40 mm and outer diameter of 60 mm. Low velocity impact tests were performed at 20°C, 60°C and 100°C, after conditioning the samples for 60 minutes. An impact energy of 95 J was used in order to reach the penetration threshold for each
EP
laminate. The impactor mass was 11.9 kg and the velocity was set to 4.00 m/s.
AC C
3. RESULTS AND DISCUSSION 3.1. Composite preparation and fibre impregnation The composite production process was optimized to limit the void content below 1% (Table 1). Several SEM qualitative observations were performed on different samples for each type of laminate and voids were never detected. Although the sizing of all fabrics were specifically developed for epoxy matrices, hence the surface treatment was not specifically suited to PEN, fibre wetting was pretty good in carbon and basalt based composites. The compatibility between matrix and fibres has been qualitatively evaluated with SEM, since it has a direct relationship with both flexural and impact performances of laminates. In Figure 1 SEM micrographs taken from fractured surfaces after flexural tests are shown. Carbon (Figure 1A) 7
ACCEPTED MANUSCRIPT and basalt fibres (Figure 1B) showed a good fibre wetting and PEN remained stuck on fibres also after the fibre pull out. Aramid fibres (Figure 1C) showed poor wetting and very clean fibre surfaces were detected while Vectran fibres exhibited moderate wetting with the peculiar multi-fibrillar structure clearly evident in fibres protruding from the fractured surfaces (Figure 1D).
RI PT
3.2. Three-point bending tests
The average values and standard deviations of the measured mechanical parameters (flexural modulus, flexural strength and strain at break) from static flexural tests performed at different temperatures on composites with 8 fabric layers are summarized in Table 2. Representative
SC
stress/strain curves from flexural tests at 20 °C, 60 °C and 100 °C are shown in Figure 2 for
among the 5 tested specimens.
M AN U
each type of composite. These curves have been selected to represent the average behaviour
The static mechanical behaviour of the different PEN composites was dependent on the fabrics and the interface strength. In particular, the fibre type mainly affected the flexural modulus, while the flexural strength was significantly influenced by the compatibility between the matrix and the fibres. PEN/carbon composites exhibited the highest flexural modulus among all samples with an almost perfectly brittle elastic behaviour, and the lowest
TE D
strain at break. PEN/basalt composites follow, presenting a brittle behaviour with limited plasticity. PEN/aramid composites showed an initial stiffness similar to that of PEN/basalt composites but a subsequent low plastic plateau and high strain at break. PEN/Vectran
EP
composites were found to show the lowest flexural modulus and strength among all composites, reaching the limit of strain at 5% specified by the ASTM D790 standard with almost no visible sign of damage. PEN/Vectran composites were also tested at a higher strain
AC C
rate (calculated for the outer fibres and equal to 0.10 min-1) 10 times faster than the previous one, as indicated by the standard in case no rupture occurs within the 5 % limit strain. In this testing conditions the mechanical behaviour was almost the same of that at 0.01 min-1, and no evidences of fibre breakage was detected. The value of strain at break higher than 5% in PEN/Vectran composites was in accordance with results found in literature [22]. The low flexural response of vectran based composites is likely to be due to the production process of the composite. In fact, the proximity of the impregnation temperature for producing the laminate with the melting temperature of the liquid crystalline polymer could have modified the microstructure of fibres, thus inducing alterations in the fibre properties and hence a reduction of their mechanical performance. 8
ACCEPTED MANUSCRIPT The relative trend of the flexural strength is similar to that of the flexural modulus (Table 2). Carbon based laminates exhibited a very high flexural strength (714±53 MPa), with the failure occurring on the lower side of the sample due to the tensile failure of fibres (Figure 3A). Basalt based system exhibited lower flexural strength values with respect to PEN/carbon ascribed to the occurrence of a compressive failure of upper layers in the laminate (Figure 3B)
RI PT
which hindered the exploitation of the high strength of the fibres. PEN/aramid laminates showed significantly lower flexural strength when compared to carbon and basalt composites and an early compressive failure of upper layers in the laminate was detected (Figure 3C). PEN/vectran laminates did not show any visible damage, as confirmed from the shape of the
SC
flexural stress versus flexural strain curve (Figure 3D) even though they showed the lowest flexural strength among all systems.
The rise of the temperature caused a general reduction in all flexural properties but the
M AN U
dependence of flexural modulus and strength on the temperature was different for all systems. In Table 2 the values of the flexural parameters evaluated at the different temperatures are reported while the normalized flexural modulus and strength, obtained by dividing the absolute values at each temperature by the values at 20 °C, are reported in Figure 4A and Figure 4B, respectively. At 60 °C and 100 °C the PEN/carbon laminate showed a limited
TE D
reduction of the flexural modulus (-6.4% and -17.0%, respectively) while a sharp drop of the flexural strength was detected (-37% and -46.8%, respectively). PEN/basalt showed a moderate reduction of the flexural modulus (up to -20.2% at 100 °C) but interestingly the flexural strength was only slightly reduced at 60°C and 100°C (-6.8% and -25.2%).
EP
PEN/aramid showed a moderate reduction of its flexural modulus (up to -20.3% at 100°C) while its flexural strength underwent a severe decrease at 100°C (-42.4%). PEN/Vectran was the most sensitive laminate to the temperature increase, showing an abrupt reduction of both
AC C
flexural modulus and strength. As regards the static mechanical behaviour, the most interesting system is PEN/basalt since it showed a good retention of its ambient temperature performance up to 100°C. A major role in the decrease of the flexural strength could be ascribed to the reduction of the interface strength with the temperature, as also confirmed by the low velocity impacts results. 3.3. Dynamic mechanical tests DMA tests were performed to estimate the influence of temperature on the elastic behaviour of the composites, even if the absolute value of the storage modulus in bending measured by this technique usually differs from that measured by static mechanical tests. Nevertheless, the 9
ACCEPTED MANUSCRIPT use of the three-point bending configuration for the DMA testing can give results very close to those from static testing [45]. Since samples for the DMA characterization are small, discrepancies in the absolute value of storage modulus in bending could also be related to specimen characteristics. To compensate these issues DMA tests were performed on at least four specimens for neat PEN and each PEN composite configuration.
RI PT
In Figure 5 representative curves of storage modulus versus temperature are presented for composites and neat amorphous PEN. In general, the stiffness of samples, intended as the absolute value of the storage modulus, is mostly dependent on the reinforcement properties while its dependence on the temperature is due to the polymer as confirmed by the similarity
SC
of slopes of neat PEN and PEN reinforced with basalt or carbon fibres. The storage modulus dropped above 100°C for all composites, with carbon and aramid being the best performing systems. This behaviour confirms a good retention of the stiffness for these materials up to the
M AN U
glass transition temperature of the polymer (around 120 °C). Moreover, the DMA plot shows a clear crystallization process occurring during heating (storage modulus increases at temperatures higher than 165 °C). The signal magnitude after the crystallization completion allows speculating that composites prepared by allowing a complete crystallization of the PEN matrix could be capable to sustain high loads up to 230 °C.
TE D
The dependence of the storage modulus on the temperature is linear and monotonically decreasing for carbon and basalt based systems, but for the other two composites, being aramid and Vectran fibres polymeric in nature, a complex behaviour was detected. The PEN/Aramid system shows a slightly increasing elastic response (storage modulus) with the
EP
temperature up to the signal drop at 120 °C, while the PEN/Vectran system exhibits a raising storage modulus up to -20 °C and then a decreasing trend until 120 °C is reached. The slope of the storage modulus curve after crystallization above 165 °C was found to decrease for all
AC C
systems. It is worth noting that the slope of the PEN/Vectran curve above -20 °C was steeper with respect to both basalt and carbon composites. These peculiar behaviours could probably be related to the coefficients of thermal expansion, that are negative for both aramid and Vectran fibres, but further investigations should be performed to clarify this phenomenon. A satisfactory agreement between static flexural modulus and storage modulus was detected for all composites (differences within ± 13%) except for PEN/Vectran, for which a significant difference of the storage modulus value and the flexural modulus at 60 °C was measured.
10
ACCEPTED MANUSCRIPT 3.4. Low velocity impact tests The impact energy needed to completely perforate the laminates was evaluated and is reported in Table 3. The perforation energy depends on the fibre type and on the total amount of reinforcement, which can be calculated as the product of the laminate thickness and the fibre volume. Among the 8 layers composite samples, PEN/Aramid8L and PEN/Vectran8L were the
RI PT
laminates with the highest perforation energies (38.2 J and 29.4 J, respectively), while PEN/Carbon8L exhibited the lowest performance (8.7 J). PEN/Basalt8L showed an
intermediate, albeit very good, performance. The ratio between the perforation energies of PEN/Basalt8L and PEN/Carbon8L was around 3. The specific perforation energy was
SC
additionally calculated for all samples with the aim to verify the capability of the composites to dissipate impact energy per unit mass (Table 3). As expected PEN/Aramid was the best performing laminate, followed by PEN/Vectran and PEN/Basalt ones.
M AN U
Since the dissipated energy is proportional to the total amount of reinforcement, PEN/Basalt14L and PEN/Carbon12L samples were prepared by keeping constant the fibre volume content and reaching a laminate thickness comparable to PEN/Aramid8L and PEN/Vectran8L composites. As a result, the perforation energy of PEN/Basalt14L and PEN/Carbon12L samples raised (to 44.3 J and 15.3 J, respectively) with respect to
TE D
PEN/Basalt8L and PEN/Carbon8L. Comparing the perforation energy values of thick PEN/Basalt and PEN/Carbon laminates to those of composites with similar thickness (namely PEN/Aramid8L and PEN/Vectran8L), PEN/Basalt14L composite showed the highest impact perforation energy while the PEN/Carbon12L still showed the lowest value (Figure 6A). These
EP
results indicate that, at constant fibre volume percentage and thickness (i.e. actual fibre volume content), the PEN/Basalt system has the highest impact resistance by a wide margin
AC C
among all tested laminates, and it is followed by PEN/Aramid. The ratio between the perforation energies of PEN/Basalt14L and PEN/Carbon12L was again around 3, hence the relative performance is preserved after the increase of the actual amount of fibres. The comparison of the specific perforation energy among thick samples showed a very interesting result. PEN/Aramid was the best performing laminate but the specific perforation energy of PEN/Basalt14L was higher than that of PEN/Vectran8L, and only 14% lower than PEN/Aramid8L. The performance of the PEN/Basalt laminate is particularly remarkable when compared to that of PEN/Aramid because the latter system has a lower interface strength, as clearly demonstrated by the very wide damaged area on the back surface of the impacted sample (Figure 7). A good fibre/polymer compatibility is qualitatively detectable in the 11
ACCEPTED MANUSCRIPT PEN/Basalt composite (small damaged area along with very few damages far from the impact zone) while the PEN/Aramid system shows a large damaged area around the impact point. A cross shaped pattern far from the impact point is also visible in the PEN/Vectran laminate, while PEN/Carbon did not show any damage far from the impact zone. It is important to note that the contribution to the impact energy dissipation operated by the interface, as assessed by
RI PT
Sorrentino et al. in [46] for polypropylene/glass fibre composites where the lower the interface strength the higher the impact resistance, is very high. The authors showed that
composites with a weak interface strength between polymer and fibres were able to withstand much higher loads and absorb higher impact energy before being penetrated with respect to
SC
composites with a strong interface strength. As previously stated, the interface strength
between PEN and basalt fibres was quite good, according to the qualitative SEM analysis (Figure 1), while PEN and aramid fibres did not exhibit suitable compatibility and
M AN U
consequently their interface strength was weak. This situation suggests that by decreasing the basalt/PEN interface strength it would be possible to increase the impact resistance, thus potentially leaving the PEN/Basalt system as the best performing laminate also in terms of specific impact resistance.
Temperature did not show a univocal effect on the impact response and in some cases an
TE D
increased perforation energy was measured. In fact, PEN/Carbon and PEN/Aramid showed an increase of the perforation energy with the increase of temperature (Figure 6B and C). PEN/Basalt slightly reduced its capability to withstand impact loads while PEN/Vectran showed a clear drop of the impact response with the rise of the testing temperature (Figure 8).
EP
The analysis of the force versus time curves (Figure 9) shows that, among thick samples, the peak load of PEN/Aramid exhibited the highest value among all systems at all testing temperatures. PEN/Basalt14L showed the second highest peak load, followed by
AC C
PEN/Vectran8L and PEN/Carbon12L. High temperature tests were performed on 8 layer systems. The PEN/Aramid and PEN/Vectran composites showed a reduction of the peak load at 60 °C and 100 °C while the peak load of PEN/Basalt only dropped at 100 °C. PEN/Carbon showed an interesting behaviour because its peak load increased with the temperature, a phenomenon typically related to the weakening of the fibre/matrix interface strength and in agreement with the significant reduction of the static flexural strength at 60 °C and 100 °C. Contary to [47] where the authors state that “the typically low strain to break of fibres might suggest that energy absorption is governed by fibre breakage as temperature increases and this indicates that with increasing temperature fewer and fewer fibres are actually breaking.”, the actual behaviour of the tested thermoplastic systems evidenced that the low-velocity impact 12
ACCEPTED MANUSCRIPT performance is influenced by the type of fibre surface treatment as also reported in [48]. Figure 10 summarizes the different behaviours of the systems with the temperature: reduction of the peak but enlargement of the curve shape (PEN/Aramid), reduction of both peak load and curve shape (PEN/Basalt), and increase of peak load and curve shape (PEN/Carbon).
RI PT
4. CONCLUSIONS The static and low velocity impact behaviours of thermoplastic composites based on PEN and high performance woven fabrics with a fibre volume content of 40% were assessed. The
flexural modulus was in direct proportion with the fibre type and volume content, and the PEN/Carbon composite was the best performing system, followed by PEN/Basalt and
SC
PEN/Aramid. PEN/Vectran exhibited a poor mechanical performance, likely due to some degradation occurred to the fibres during the composite production. The flexural strength was
M AN U
dependent on the interface strength, but at 20 °C its trend was consistent with the flexural modulus. At 60 °C and 100 °C the flexural properties of all laminates decreased, but the PEN/Basalt laminate showed the lowest strength reduction among all systems. The static mechanical response was in good agreement with DMA tests, showing the capability of all laminates to withstand high loads at temperature as high as 100 °C. The low velocity impact resistance was dependent on the fibre type, the fibre/matrix interface
TE D
strength and the laminate thickness. Unexpectedly, the best performance was shown by basalt based laminates. In fact, PEN/Basalt was able to withstand a higher impact energy before perforation than both polymeric and carbon fibres based systems. Interestingly PEN/Carbon,
EP
although characterized by the lowest impact resistance, exhibited an increase of the impact performance with the temperature, along with PEN/Aramid. A significant role in this phenomenon is believed to be played by the coupled contribution of the thermoplastic matrix
AC C
and the variation of the fibre/matrix interface strength with the temperature. 5. ACKNOWLEDGEMENTS This study has been carried out with financial support from MIUR Ministry (Italy) within the TECOP project (PON02_00029_3206010). Authors would like to thank Mr Fabio Docimo for its contribution to the preparation of all tested samples.
13
ACCEPTED MANUSCRIPT REFERENCES Apicella A, Migliaresi C, Nicolais L, Iaccarino L, Roccotelli S. The water ageing of unsaturated polyester-based composites: influence of resin chemical structure. Composites 1983;14:387–92. doi:10.1016/0010-4361(83)90160-X.
[2]
Bradley WL, Grant TS. The effect of the moisture absorption on the interfacial strength of polymeric matrix composites. J Mater Sci 1995;30:5537–42. doi:10.1007/BF00351570.
[3]
Kootsookos A, Mouritz AP. Seawater durability of glass- and carbon-polymer composites. Compos Sci Technol 2004;64:1503–11. doi:10.1016/j.compscitech.2003.10.019.
[4]
Huang G, Sun H. Effect of water absorption on the mechanical properties of glass/polyester composites. Mater Des 2007;28:1647–50. doi:10.1016/j.matdes.2006.03.014.
[5]
Ramirez FA, Carlsson LA, Acha BA. Evaluation of water degradation of vinylester and epoxy matrix composites by single fiber and composite tests. J Mater Sci 2008;43:5230–42. doi:10.1007/s10853-008-2766-z.
[6]
Czél G, Wisnom MR. Demonstration of pseudo-ductility in high performance glass/epoxy composites by hybridisation with thin-ply carbon prepreg. Compos Part A Appl Sci Manuf 2013;52:23–30. doi:10.1016/j.compositesa.2013.04.006.
[7]
Czél G, Pimenta S, Wisnom MR, Robinson P. Demonstration of pseudo-ductility in unidirectional discontinuous carbon fibre/epoxy prepreg composites. Compos Sci Technol 2015;106:110–9. doi:10.1016/j.compscitech.2014.10.022.
[8]
Selver E, Potluri P, Hogg P, Soutis C. Impact damage tolerance of thermoset composites reinforced with hybrid commingled yarns. Compos Part B Eng 2016;91:522–38. doi:10.1016/j.compositesb.2015.12.035.
[9]
Shivakumar Gouda PS, Chatterjee V, Barhai PK, Jawali D, Rahatekar S, Wisnom MR. Improved fracture toughness in carbon fibre epoxy composite through novel pre-preg coating method using epoxy terminated butadiene nitrile rubber. Mater Des 2014;62:320–6. doi:10.1016/j.matdes.2014.05.018.
AC C
EP
TE D
M AN U
SC
RI PT
[1]
[10] Wong DWY, Lin L, McGrail PT, Peijs T, Hogg PJ. Improved fracture toughness of carbon fibre/epoxy composite laminates using dissolvable thermoplastic fibres. Compos Part A Appl Sci Manuf 2010;41:759–67. doi:10.1016/j.compositesa.2010.02.008. [11] Zeng Y, Liu HY, Mai YW, Du XS. Improving interlaminar fracture toughness of carbon fibre/epoxy laminates by incorporation of nano-particles. Compos Part B Eng 2012;43:90–4. doi:10.1016/j.compositesb.2011.04.036.
14
ACCEPTED MANUSCRIPT [12] Vaidya UK, Chawla KK. Processing of fibre reinforced thermoplastic composites. Int Mater Rev 2008;53:185–218. doi:10.1179/174328008x325223. [13] Turnbull L, Liggat JJ, Macdonald WA. Thermal degradation chemistry of poly(ethylene naphthalate) - A study by thermal volatilisation analysis. Polym Degrad Stab 2013;98:2244–58. doi:10.1016/j.polymdegradstab.2013.08.018.
RI PT
[14] Ito E, Kobayashi Y. Effects of absorbed water on physical properties of polyesters. J Appl Polym Sci 1980;25:2145–57. doi:10.1002/app.1980.070251001. [15] Zheng L, Qi J, Liu D, Zhou W. Studies on the phase structure, mechanical, rheological properties, and nonisothermal crystallization kinetics of the PEN/SCF composites. J Appl Polym Sci 2009;112:3462–9. doi:10.1002/app.
SC
[16] Kim JY, Kim DK, Kim SH. Thermal decomposition behavior of poly(ethylene 2,6naphthalate)/silica nanocomposites. Polym Compos 2009;30:1779–87. doi:10.1002/pc.20749.
M AN U
[17] Karsli NG, Yesil S, Aytac A. Effect of short fiber reinforcement on the properties of recycled poly(ethylene terephthalate)/poly(ethylene naphthalate) blends. Mater Des 2013;46:867–72. doi:10.1016/j.matdes.2012.11.009. [18] Xu C, Cao L, Chen Y. Glass fibers reinforced poly(ethylene 2,6-naphthalate)/ethylene propylene diene monomer composites: Structure, mechanical, and thermal properties. Polym Compos 2014;35:939–47. doi:10.1002/pc.22738.
TE D
[19] Sorrentino L, Cafiero L, D’Auria M, Iannace S. Cellular thermoplastic fibre reinforced composite (CellFRC): A new class of lightweight material with high impact properties. Compos Part A Appl Sci Manuf 2014;64:223–7. doi:10.1016/j.compositesa.2014.05.016.
AC C
EP
[20] Sorrentino L, Cafiero L, Iannace S. Foams from high performance thermoplastic PEN/PES blends with expanded graphite. In: D’Amore A, Acierno D, Grassia L, editors. AIP Conf. Proc., vol. 1593, AIP Publishing LLC; 2014, p. 358–61. doi:10.1063/1.4873800. [21] Sorrentino L, de Vasconcellos DS, D’Auria M, Sarasini F, Iannace S. Thermoplastic composites based on poly(ethylene 2,6-naphthalate) and basalt woven fabrics: Static and dynamic mechanical properties. Polym Compos 2015;37:n/a-n/a. doi:10.1002/pc.23449. [22] Wang J, Chen J, Dai P. Polyethylene naphthalate single-polymer-composites produced by the undercooling melt film stacking method. Compos Sci Technol 2014;91:50–4. doi:10.1016/j.compscitech.2013.11.026. [23] Ghaseminejhad MN, Parvizi-Majidi A. Impact behaviour and damage tolerance of woven carbon fibre-reinforced thermoplastic composites. Constr Build Mater 1990;4:194–207. doi:10.1016/0950-0618(90)90040-8. 15
ACCEPTED MANUSCRIPT [24] VU-KHAN HWAT. Damage Extension in Carbon Fiber/PEEK Crossply Laminates under Low VeIocity Impact. Jorurnal Compos Mater 1994;28:684–707. doi:10.1177/002199839402800801.
RI PT
[25] Ishikawa T, Sugimoto S, Matsushima M, Hayashi Y. Some experimental findings in compression-after-impact (CAI) tests of CF/PEEK (APC-2) and conventional CF/epoxy flat plates. Compos Sci Technol 1995;55:349–63. doi:10.1016/02663538(95)00079-8. [26] Gao SL, Kim JK. Cooling rate influences in carbon fibre/PEEK composites. Part III: Impact damage performance. Compos - Part A Appl Sci Manuf 2001;32:775–85. doi:10.1016/S1359-835X(00)00189-5.
SC
[27] Vieille B, Casado VM, Bouvet C. About the impact behavior of woven-ply carbon fiber-reinforced thermoplastic- and thermosetting-composites: A comparative study. Compos Struct 2013;101:9–21. doi:10.1016/j.compstruct.2013.01.025.
M AN U
[28] Vieille B, Casado VM, Bouvet C. Influence of matrix toughness and ductility on the compression-after-impact behavior of woven-ply thermoplastic- and thermosettingcomposites: A comparative study. Compos Struct 2014;110:207–18. doi:10.1016/j.compstruct.2013.12.008.
TE D
[29] Vedula M, Koczak MJ. Impact Resistance of Cross-Plied Polyphenylene Sulfide Composites. J Thermoplast Compos Mater 1989;2:154–63. doi:10.1177/089270578900200301. [30] Bibo GA, Hogg PJ, Kemp M. High-temperature damage tolerance of carbon fibrereinforced plastics:. 2. Post-impact compression characteristics. Composites 1995;26:91–102. doi:10.1016/0010-4361(95)90408-R.
EP
[31] Gómez-del Río T, Zaera R, Barbero E, Navarro C. Damage in CFRPs due to low velocity impact at low temperature. Compos Part B Eng 2005;36:41–50. doi:10.1016/j.compositesb.2004.04.003.
AC C
[32] Icten BM, Atas C, Aktas M, Karakuzu R. Low temperature effect on impact response of quasi-isotropic glass/epoxy laminated plates. Compos Struct 2009;91:318–23. doi:10.1016/j.compstruct.2009.05.010. [33] Aktaş M, Karakuzu R, Arman Y. Compression-after impact behavior of laminated composite plates subjected to low velocity impact in high temperatures. Compos Struct 2009;89:77–82. doi:10.1016/j.compstruct.2008.07.002. [34] Aktas M, Karakuzu R, Icten BM. Impact Behavior of Glass/Epoxy Laminated Composite Plates at High Temperatures. J Compos Mater 2010;44:2289–99. doi:10.1177/0021998310369576.
16
ACCEPTED MANUSCRIPT [35] Aktas M, Karakuzu R, Içten BM. Thermal Impact Behavior of Glass–EpoxyLaminated Composite Plates. J Thermoplast Compos Mater 2011;24:535–6. doi:10.1177/0892705710393113.
RI PT
[36] Suvarna R, Arumugam V, Bull DJ, Chambers AR, Santulli C. Effect of temperature on low velocity impact damage and post-impact flexural strength of CFRP assessed using ultrasonic C-scan and micro-focus computed tomography. Compos Part B Eng 2014;66:58–64. doi:10.1016/j.compositesb.2014.04.028. [37] Ibekwe SI, Mensah PF, Li G, Pang SS, Stubblefield MA. Impact and post impact response of laminated beams at low temperatures. Compos Struct 2007;79:12–7. doi:10.1016/j.compstruct.2005.11.025.
SC
[38] Hirai Y, Hamada H, Kim J-K. Impact response of woven glass-fabric composites—II. Effect of temperature. Compos Sci Technol 1998;58:119–28. doi:10.1016/S02663538(97)00112-7.
M AN U
[39] Alcock B, Cabrera NO, Barkoula NM, Wang Z, Peijs T. The effect of temperature and strain rate on the impact performance of recyclable all-polypropylene composites. Compos Part B Eng 2008;39:537–47. doi:10.1016/j.compositesb.2007.03.003. [40] Garcia-Gonzalez D, Rodriguez-Millan M, Rusinek A, Arias A. Low temperature effect on impact energy absorption capability of PEEK composites. Compos Struct 2015;134:440–9. doi:10.1016/j.compstruct.2015.08.090.
TE D
[41] Sorrentino L, Vasconcellos DS de, D’Auria M, Sarasini F, Tirillò J. Thermoplastic Composites Based on PEN and high performance woven fabrics. Polymer Composites, submitted (2016). Polym Compos 2016;Submitted.
EP
[42] Cantwell WJ, Morton J. Geometrical effects in the low velocity impact response of CFRP. Compos Struct 1989;12:39–59. doi:10.1016/0263-8223(89)90043-3. [43] Cantwell WJ, Morton J. The impact resistance of composite materials - a review. Composites 1991;22:347–62. doi:10.1016/0010-4361(91)90549-V.
AC C
[44] Caprino G, Lopresto V. On the penetration energy for fibre-reinforced plastics under low-velocity impact conditions. Compos Sci Technol 2001;61:65–73. doi:10.1016/S0266-3538(00)00152-4. [45] Deng S, Hou M, Ye L. Temperature-dependent elastic moduli of epoxies measured by DMA and their correlations to mechanical testing data. Polym Test 2007;26:803–13. doi:10.1016/j.polymertesting.2007.05.003. [46] Sorrentino L, Simeoli G, Iannace S, Russo P. Mechanical performance optimization through interface strength gradation in PP/glass fibre reinforced composites. Compos Part B Eng 2015;76:201–8. doi:10.1016/j.compositesb.2015.02.026.
17
ACCEPTED MANUSCRIPT [47] Salehi-Khojin A, Mahinfalah M, Bashirzadeh R, Freeman B. Temperature effects on Kevlar/hybrid and carbon fiber composite sandwiches under impact loading. Compos Struct 2007;78:197–206. doi:10.1016/j.compstruct.2005.09.005.
AC C
EP
TE D
M AN U
SC
RI PT
[48] Hirai Y, Hamada H, Kim J-KJK. Impact response of woven glass-fabric composites - I. Effect of fibre surface treatment. Compos Sci Technol 1998;58:91–104. doi:10.1016/S0266-3538(97)00111-5.
18
ACCEPTED MANUSCRIPT Figure Captions
Figure 1. Wetting quality of fibres: A) Carbon, B) Basalt, C) Aramid, D) Vectran Figure 2. Effect of the temperature on the stress/strain curves of PEN composites: A)
PEN/Carbon8L, B) PEN/Basalt8L, C) PEN/Aramid8L, D) PEN/Vectran8L Figure 3. Pictures reporting laminates damage after the flexural test: A) Carbon, B) Basalt, C)
RI PT
Aramid, D) Vectran.
Figure 4. Normalized flexural modulus (A) and normalized flexural strength (B) as function of
the temperature
Figure 5. Representative curves of storage moduli (solid lines) and flexural moduli (dashed lines) versus temperature for the different composites and the neat amorphous PEN.
SC
Figure 6. Energy versus Time curves for the different composites: A) 20 °C, B) 60 °C, C) 100
°C
M AN U
Figure 7. Back surfaces of samples impacted at 100 °C: A) Carbon, B) Basalt, C) Aramid, D) Vectran Figure 8. Comparison of the Energy versus Time curves of all systems measured at 20 °C and
100 °C
Figure 9. Force versus Time curves for the different composites: A) 20 °C, B) 60 °C, C) 100
°C
Figure 10. Comparison of the Force versus Time curves of all systems measured at 20 °C and
AC C
EP
TE D
100 °C
19
ACCEPTED MANUSCRIPT Table Captions
Table 1. Summary of the physical parameters measured on laminates. Table 2. Mean values and standard deviations of flexural modulus (EB), flexural strength (σB), and flexural strain at break (εB) at the different testing temperatures.
AC C
EP
TE D
M AN U
SC
RI PT
Table 3. Values of perforation energy measured for each type of PEN composite studied.
20
ACCEPTED MANUSCRIPT
Table 1. Summary of the physical parameters measured on laminates Number of fabric plies 8
Density (g/cm3) 1.51±0.01
Thickness (mm) 1.73±0.04
Void content (%) 0.7±0.2
PEN/Basalt8L
39±2%
8
1.87±0.03
1.70±0.05
0.7±0.2
PEN/Aramid8L
41±1%
8
1.39±0.02
2.82±0.09
0.8±0.2
PEN/Vectran8L
40±1%
8
1.37±0.02
2.73±0.06
0.7±0.1
PEN/Carbon12L
41±1%
12
1.52±0.01
2.60±0.08
0.7±0.1
PEN/Basalt14L
40±1%
14
1.88±0.01
RI PT
PEN/Carbon8L
Fibre volume ratio 40±1%
2.72±0.05
AC C
EP
TE D
M AN U
SC
Composite
0.8±0.1
ACCEPTED MANUSCRIPT
Table 2. Mean values and standard deviations of flexural modulus (EB), flexural strength (σB), and flexural strain at break (εB) at the different testing temperatures.
EB (GPa) 32.2±1.9 16.2±0.6 13.7±1.2 6.5±0.9
AC C
EP
TE D
M AN U
SC
PEN/Carbon8L PEN/Basalt8L PEN/Aramid8L PEN/Vectran8L
EB (GPa) 38.8±0.9 19.8±1.5 17.2±0.7 11.4±0.4
20 °C 60 °C σB εB EB σB εB (MPa) (%) (GPa) (MPa) (%) 714±53 2.0±0.1 36.3±1.2 456±55 1.4±0.1 365±41 2.7±0.6 17.1±1.2 340±25 2.9±0.5 224±24 2.9±0.4 14.8±0.3 161±1.4 2.4±0.2 > 190 > 5.0 7.1±0.9 > 76 3.6±0.5
100 °C σB (MPa) 380±23 273±18 129±14 > 59
εB (%) 1.3±0.1 2.6±0.4 1.8±0.1 3.1±0.5
RI PT
Composite
ACCEPTED MANUSCRIPT
Table 3. Values of perforation energy measured for each type of PEN composite studied. Thickness (mm)
8 8 8 8 12 14
1.73±0.04 1.70±0.05 2.82±0.09 2.73±0.06 2.60±0.08 2.72±0.05
Perforation energy (J) 20 °C 60 °C 100 °C 8.7±0.3 11.7±0.1 16.2±0.1 24.0±0.5 22.2±1.8 21.9±0.9 38.2±0.9 43.7±1.1 47.3±1.3 29.4±0.2 24.2±0.3 19.5±0.6 15.3±0.2 44.3±0.2 -
AC C
EP
TE D
M AN U
SC
PEN/Carbon8L PEN/Basalt8L PEN/Aramid8L PEN/Vectran8L PEN/Carbon12L PEN/Basalt14L
Number of layers
Specific Perforation Energy (J/g/cm3) 20 °C 5.8 12.8 27.3 21.5 10.1 23.6
RI PT
Composite
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
Figure 1 . Wetting quality of fibres: A) Carbon, B) Basalt, C) Aramid, D) Vectran
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure 2 . Effect of the temperature on the stress/strain curves of PEN composites: A) PEN/Carbon8L, B) PEN/Basalt8L, C) PEN/Aramid8L, D) PEN/Vectran8L
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
Figure 3 . Pictures reporting laminates damage after the flexural test: A) Carbon, B) Basalt, C) Aramid, D) Vectran.
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
Figure 4 . Normalized flexural modulus (A) and normalized flexural strength (B) as function of the temperature
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
Figure 5 . Representative curves of storage moduli (solid lines) and flexural moduli (dashed lines) versus temperature for the different composites and the neat amorphous PEN.
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
Figure 6 . Energy versus Time curves for the different composites: A) 20 °C, B) 60 °C, C) 100 °C
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure 7 . Back surfaces of samples impacted at 100 °C: A) Carbon, B) Basalt, C) Aramid, D) Vectran
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
Figure 8 . Comparison of the Energy versus Time curves of all systems measured at 20 °C and 100 °C
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 9 . Force versus Time curves for the different composites: A) 20 °C, B) 60 °C, C) 100 °C
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
Figure 10 . Comparison of the Force versus Time curves of all systems measured at 20 °C and 100 °C
