- Email: [email protected]
epoxy composites via thermoplastic polyurethane nonwoven fabric
Accepted Manuscript The remarkably enhanced particle erosion resistance and toughness properties of glass fiber/epoxy composites via thermoplastic polyurethane nonwoven fabric Guangchao Lv, Na Zhang, Ming Huang, Changyu Shen, Jose Castro, Kunlun Tan, Xianhu Liu, Chuntai Liu PII:
S0142-9418(18)30708-6
DOI:
10.1016/j.polymertesting.2018.06.005
Reference:
POTE 5502
To appear in:
Polymer Testing
Received Date: 30 April 2018 Revised Date:
4 June 2018
Accepted Date: 4 June 2018
Please cite this article as: G. Lv, N. Zhang, M. Huang, C. Shen, J. Castro, K. Tan, X. Liu, C. Liu, The remarkably enhanced particle erosion resistance and toughness properties of glass fiber/epoxy composites via thermoplastic polyurethane nonwoven fabric, Polymer Testing (2018), doi: 10.1016/ j.polymertesting.2018.06.005. 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
The remarkably enhanced particle erosion resistance and toughness properties of glass fiber/epoxy composites via
RI PT
thermoplastic polyurethane nonwoven fabric Guangchao Lva, Na Zhanga*, Ming Huanga, Changyu Shena, Jose Castroc, Kunlun Tand, Xianhu Liua, Chuntai Liua,b*
National Engineering Research Center for Advanced Polymer Processing Technology,
b
M AN U
Zhengzhou University, Zhengzhou 450002, China
SC
a
Advanced Research Center for Polymer Processing Engineering of Guangdong
Province Guangdong Industry Polytechnic, Guangzhou 510300, China c
Department of integrated systems Engineering, The Ohio state university, 1971 Neil
d
TE D
Avenue, Columbus, Ohio 43210, USA
PGTEX CHINA CO., LTD, Changzhou, Jiangshu, Xinbei District, 213135, China
ABSTRACT: In this work, a new method of preparing glass fiber/epoxy (GF/EP)
EP
composites with excellent solid particle erosion resistance and fracture toughness
AC C
properties via thermoplastic polyurethane nonwoven fabric (TNF) was reported. TNF was prepared by virtue of the melt-blown (MB) process, and the composites were fabricated by vacuum-assisted resin transfer molding (VARTM) technique. Solid particle erosion characteristics of the composites were investigated in a confined space by impinging angular silica particles with a size about 300 µm. Compared with conventional GF/EP composites, the erosive wear resistance of TNF/glass fiber/epoxy
*
Corresponding authors. E-mail address: [email protected]; [email protected]
ACCEPTED MANUSCRIPT (TNF/GF/EP) composites were improved by 99%, 108%, 205% and 493% at the impingement angles of 20°, 30°, 45° and 90°, respectively. Meanwhile, increase of 78% in GIC and 115% in GⅡC were achieved. Microstructure characterizations,
RI PT
including scanning electron microscopy (SEM) and optical microscopy were performed to investigate the variations of microstructure and further to establish the relationship between epoxy matrix and glass fibers. This work suggests a promising
SC
routine to improve particle erosion resistance and fracture toughness by interleaving
M AN U
the laminates with lightweight thermoplastic polyurethane nonwoven fabrics. Key words: Polymer matrix composites; Erosive wear; Interlaminar toughened; Glass fiber; Erosion rate; Thermoplastic polyurethanes. 1. Introduction
TE D
Recently, fiber-reinforced polymers (FRPs) with high modulus rate and high strength rate are continuously replacing more and more traditional materials, for instance, plastics, wood and metallic materials [1-5]. In general, the FRPs are always
EP
applied in severe working environments, the erosive wear behavior of the FRPs are of
AC C
vital importance. Therefore, great efforts have been made to improve the erosive wear resistance of FRPs via particulate fillers [6-10]. Bagci et al. [6] investigated the erosive wear behavior of glass fiber/epoxy composites (GFECs) in the presence of Al2O3+SiO2 particulates, and Panda et al. [7] studied the influence of AlN on the erosive wear response of GFECs. Above hybrid composites, however, suffers from the intrinsic limitation that the improvement of erosion resistance is limited and/or the manufacturing processes become difficult [11,12]. In view of this, a periodically
ACCEPTED MANUSCRIPT interleaved fiber laminate system, wherein thin protective layers are interleaved into each ply, has been used [13, 14]. The protective layer can effectively protect the fiber from solid particle erosion, which is benefit by the high strength of carbon nanofibers
RI PT
(CNFs) and their nanoscale structure. At the same time, considering the laminate feature of FRP composites, an attempt has also been made to improve the inter-laminar fracture toughness over the past
SC
years [15-17]. The existence of insufficient fracture toughness and delamination may
M AN U
lead to complete failure of the composite structure. Many solutions have been proposed to improve toughness by virtue of changing the materials structure and components used in the composites. Specifically, a ductile polymer microfiber fabric inserted into the interlayer has been employed in this study to investigate the effect.
TE D
Microfiber fabrics are highly porous and good for avoiding impeding the flow of resin during the filling process. In addition, good bonding with the matrix is available due to the particularly high specific surface area. Yasaee et al. [18] investigated the
EP
Mode-Ⅱ inter-laminar fracture toughness by various interply strips such as polyimide
AC C
thermoplastic films, thermoset epoxy films, glass/epoxy prepreg and chopped Aramid fibers, reporting a 200% Mode-Ⅱ inter-laminar fracture toughness improvement. The investigation by Yi et al. [19] showed that composites interleaved with Aramid Nonwoven Fabric (ANF) had better fracture energy under Mode-I and Mode-Ⅱ loading (GIC and GⅡC) as compared with the reference one composite. Hamer et al. [20] found that the significant enhanced GIC and GⅡC could be realized by adding electro-spun nylon reinforced with MWCNTs into carbon/epoxy composites.
ACCEPTED MANUSCRIPT Thermoplastic polyurethanes (TPUs) have attracted a significant attention due to their good processability and wear resistant property deriving from their peculiar molecular structure [21-23]. This makes TPUs suitable for protective coating or films
RI PT
with minimal erosive wear. To the best of our knowledge and experience, researches on erosive wear resistance behaviors of TNF reinforced GF/EP composites have not been found in the open literatures. In view of this, we studied the use of a thin
SC
protective layer of TNF for enhancing the solid particle erosion resistance of GF/EP
M AN U
composites in this study. TNF were prepared by the melt-blown process [24-26], and the composites were fabricated by the VARTM technique. Unexpectedly, the TNF can not only enhance composite wear resistance, but also improve composite Mode-I and Mode-Ⅱ inter-laminar fracture toughness. Underlying possible wear mechanism
TE D
was also investigated by the scanning electron microscopy and optical microscopy views. These features may make TNF applicable for FRP composites in the fields where both excellent wear resistance and high toughness properties are required
EP
simultaneously.
AC C
2. Experimental 2.1. Materials
Polyether-based thermoplastic polyurethane (TPU) (Elastollan 1185A), the density
is 1.12 g/cm3 and a melt flow index is 17.5 g/10 min (215 °C, 10 kg), which was bought from BASF Co. Ltd., China. E-Glass multi-axial fabrics, ECW800 (0/+45/90/-45), used as the long fiber reinforcement was provided by Taishan fiber glass INC., China. The epoxy resin used was LT-5078A, and a diamine curing agent,
ACCEPTED MANUSCRIPT LT-5078B with an amine value of about 420-520 mg [KOH]/g, provided by Wells Advanced Materials Co. Ltd., China. Blocky, sharp edged silica with the average diameter is 300 µm (Fig. S1), which were selected as the erodent particles, offered by
M AN U
SC
RI PT
Baige Co. Ltd., China.
TE D
Fig. 1. (a) TPU nonwoven fabric, (b) SEM image of TPU nonwoven fabric.
Fig. 2. Lay-up sequences of flat laminates for erosion test (a, b) and fracture
EP
toughness test (c, d), fiber orientations of E-Glass multi-axial fabrics (e).
AC C
2.2. Preparation of TNF, EP and composites Melt-blown (MB) pilot line shown schematically in Fig. S2. MB TPU webs were
gathered at die-to-collector distances (DCD) of 50 cm with a collector speed of 10.3 m/min, the air and die temperatures were set at 245 and 220 °C, respectively. The thickness of TNF is approximately 0.2 mm. The digital image of TPU nonwoven fabric and its SEM image are displayed in Fig. 1 (a, b), respectively. The hybrid resin was degassed in vacuum oven for about 10 min until there are no
ACCEPTED MANUSCRIPT bubbles, transferred to a Polytetrafluoroethylene (PTFE) mold and cured at 25 °C (room temperature) for 24 h, post cured at 120 °C for an additional 3 h and then pure EP specimen can be obtained after stripping.
RI PT
VARTM was selected to manufacture the GF/EP and TNF/GF/EP composites. The glass fibers and TNF were laid-up in laminate sequences as shown in Fig. 2. Vacuum was adopted to compel the bag to compress tightly against the fiber stack before mold
SC
filling. Make sure there is no leaking in the whole device. The epoxy hybrid was
M AN U
degassed in vacuum for about 10 min, the resin was infused into the no leakage mold afterwards. The specimen was subsequently cured at 25°C for 24 h and then post cured at 120 °C for an additional 3 h. The fiber volume fraction of the composites is 56 %.
TE D
2.3. Erosive Wear Testing
The erodents were accelerated to high velocities by a static pressure and directed onto a test specimen as illustrated in Fig. S3. It mainly includes an air compressor, an
EP
erodent feeder, a sand-blasting gun, and a specimen holder. The prescribed angle (α)
AC C
can be obtained by adjusting the spray gun and the sample holder. Square sample with size of 30 mm×30 mm was cut for solid particle erosion tests. The factors for the erosion trials are listed in Table 1. Before and after the erosion tests, the specimen was cleaned by acetone, dried, and weighed to an accuracy of 0.1 mg by virtue of an electronic balance (FA1004, China), subtract the two experimental data to determine mass loss. The weight erosion rate ( E ) was then obtained through the following formula [27]:
ACCEPTED MANUSCRIPT E=
∆w we
(1)
where ∆w is the weight loss of the test specimen and we is the mass of erodent particles (i.e., testing time × erodent feed rate). The smaller the erosion rate, the
represents an average gained from seven measurements.
SC
Table 1. Erosion test conditions.
Erodent size (µm)
Silica sand
M AN U
Test parameters Erodent
RI PT
better the erosive wear resistance of the composite was. The reported result
300
Erodent feed rate (g/s)
4.98
Test temperature
RT (25 °C)
60
Nozzle diameter (mm)
6.5
Air pressure (MPa)
0.345
EP
TE D
Distance from nozzle to sample (mm)
AC C
2.4. Testing for fracture toughness According to ASTM D5528, GIC was measured by Double Cantilever Beam (DCB)
test. Fig. 3a shows the schematic image for the DCB test. The DCB specimens were 150 mm long and 25 mm wide (containing 50 mm length pre-cracks, which were prepared by PTFE film). The piano hinges (used to apply the load) were attached to each specimen under pressure through a cyanoacrylate adhesive. The DCB test were executed at cross head speed of 1 mm/min on a common testing machine (Instron 5585), while the load and crack extension were recorded. At least seven specimens
ACCEPTED MANUSCRIPT were tested to drive the mean data. Mode-I critical energy release rate, GIC, was calculated using a modified beam theory (MBT) way:
GIC =
3Pδ 2 b( a + ∆ )
(2)
RI PT
Where δ is the displacement of the load point, b is the sample width, a is the crack length, P is the load, and ∆ is the horizontal axis intercept from a - C 1/3
SC
curve. The compliance, C , is the ratio of the load point displacement to the applied load, δ P .
M AN U
GⅡC was measured by end-notch flexure (ENF) test as shown in Fig. 3b. The dimension of the samples was 140 mm (in length) × 25 mm (in width), containing 50 mm length pre-cracks. The specimen was loaded twice and was loaded in displacement control at cross head speed of 1 mm/min on a common testing machine
TE D
(Instron 5585). The first loading with the span length 70 mm was stopped until crack progression occurred. Then the sample was reloaded with the span length 100 mm,
EP
until a sudden load drop was observed. At least seven samples were tested. Mode-Ⅱ
AC C
critical energy release rate, GⅡC, was figured out by equation (3):
where P
GIIC =
9 Pδa 2 2b(2 L3 + 3a 3 )
(3)
is the maximum load got from the straight part of the second
load–displacement curve, b is the width of the sample, a is the beginning crack length (25 mm), L is a half of the span length (50 mm), and δ is the load point displacement.
ACCEPTED MANUSCRIPT
Mode-Ⅱ test.
SC
2.5. Microscopy
RI PT
Fig. 3. Geometry and dimensions of fracture toughness sample: (a) Mode-Ⅱ test; (b)
The eroded surfaces and fracture surfaces of fracture toughness samples were
M AN U
coated with a thin layer of gold, and examined using a scanning electron microscope (SEM, ZEISS EVO-18). The eroded surfaces of the samples were also characterized by a Leica DVM6 optical microscopy, which provide both 3D display and plan display of the surface structures.
TE D
3. Results and discussion 3.1. Erosion performance
EP
Fig. 4 is the erosion rates of EP and its composites have been drawn into a curve with an angle as a function. Ductile and brittle erosion behavior can be distinctly
AC C
varying from this graph. EP and TNF/GF/EP composite demonstrate maximum erosion rate at 30° and minimum at 90° impingement angle, showing ductile erosion behavior. Whereas the maximum erosion rate of GF/EP composite appears at 90°, which reflects typical brittle erosion behavior. According to the literature [28], the maximum erosion rate was at about 30° and 90° impingement angles for ductile and brittle erosions, respectively. The erosion rate of GF/EP composite is higher than that of EP regardless of impingement angles. The combine of glass fibers decreases the
ACCEPTED MANUSCRIPT solid particle erosion resistance of EP and alters the erosive wear behavior of the pure EP owing to the brittle nature of glass fibers. TNF/GF/EP composite exhibits better erosive wear resistance than that of pure EP: the erosion rate of TNF/GF/EP
RI PT
composites was decreased by 10%, 22%, 6% and 21% at impingement angles of 20°, 30°, 45° and 90°, respectively. On the other hand, compared to GF/EP composites, the erosion resistance of TNF/GF/EP composites was enhanced by 99%, 108%, 205% and
TE D
M AN U
SC
493% at impingement angles of 20°, 30°, 45° and 90°, respectively.
Fig. 4. Erosion rates of EP and its composites as a function of impingement angle.
EP
Fig. 5 shows the mass loss of different samples as a function of erosion time at the impingement angles of 30° and 90°, respectively. The curve shows a steady state, and
AC C
the rate of mass loss is proportional to the time of erosion. Although EP and TNF/GF/EP displayed ductile erosion behavior, no incubation period was discovered in the experiment process [29].
RI PT
ACCEPTED MANUSCRIPT
Fig. 5. The mass loss of different samples as a function of erosion time at different
SC
impingement angles: (a) 30°, (b) 90°. 3.2. Morphology of eroded surface
M AN U
Fig. 6 (a-d) presents the eroded surfaces of GF/EP composite at 30°, 60° and 90° impingement angles. The erosion behavior of GF/EP composite samples is brittle in nature. The failure mode in GF/EP composite is a complicated process including surface micro-cracking, surface matrix removing, fiber/matrix de-bonding, fiber
TE D
fracture and material removal [30]. In addition, many micro-cracks produced by the erodent particles can be observed in glass fibers and many tiny fragments of fibers
EP
also can be discovered on the surface of erosion. At oblique impingement angles, it has been observed that the impact force can be divided into two-part: one is parallel
AC C
(Fp) to the target surface and the other is vertical (Fv). Fp is in charge of the abrasive and Fv mainly controls the impact force.
ACCEPTED MANUSCRIPT Impinging direction
TE D
M AN U
SC
RI PT
Impinging direction
Fig. 6. SEM images of eroded GF/EP (left column), EP (center column) and TNF/GF/EP (right column) at impingement angles of: (a), (e), (i) 30°, (b), (f), (j) 60°,
EP
(c), (d), (g), (h), (k), (l) 90°.
AC C
At 30° where erodent particles had small Fv to the target surface, the surface matrix was chipped away mainly via the abrasive action of Fp of impinging particles (Fig. 6a). In addition, many cracks can be clearly observed, which were resulted from the small Fv of particles. Fewer fibers were broken and eroded away by Fv, most of glass fibers were kept firmly in place. Good adhesion between matrix and glass fibers can be observed. With the impingement angle increases, the abrasive action of Fp steps down, while the vertical component turns more crucial. This leads to widespread damage
ACCEPTED MANUSCRIPT and detachment of glass fibers at 60° impact (Fig. 6b). At 90°, direct particles impact action produced the highest kinetic energy which was consumed by a lot of micro-cracks and glass fibers plastic deformation (Fig. 6c). The interfacial de-bonding
RI PT
between glass fibers and epoxy resin was also discovered. Owing to repeated impacts, the brittle glass fibers formed tiny wear debris but still adhered to the eroded surface. Especially in the enlarged area, as shown in Fig. 6d, some deep craters (see the white
SC
arrows) were formed, which indicated that due to the lack of adequate support from
M AN U
the matrix, the direct impact causes a large amount of de-bonding and breakage of glass fibers，while the rough surface causing larger removal of materials. Hence, for GF/EP composite, it is concluded that the vertical component of the velocity vector plays extremely vital role in damage [31].
TE D
The eroded surface of EP and TNF/GF/EP composite was viewed by SEM analysis at 30°, 60° and 90° impingement angles (Fig. 6 (e-l)). At 30° (see Fig. 6e and i), the thin layers of surface matrix were cut off mainly due to the remarkable cutting effects.
EP
In this angle, the cutting action of impinging particles results in a large number of
AC C
‘lips’, which were shown by means of the white features. These ‘lips’ prone to be in the same direction as the erodent particles. By comparison, the surface protected via the TNF, the epoxy resin and TNF were firmly bound and the eroded surface presents less ‘lips’. At the impingement angle of 60°, the ‘lips’ were not visible (see Fig. 6f and j). It is interesting to note that larger fragment of surface materials of pure EP were cut off (see the white arrows on Fig. 6f), mainly due to the higher vertical component of the velocity vector. From Fig. 6g, a number of craters caused by erodent particles can
ACCEPTED MANUSCRIPT be seen at 90° impingement angle, which proves that a large amount of surface material is completely chipped off. However, there are few craters of TNF/GF/EP composite after erosion (Fig. 6k). It is up to the presence of TNF contribute a flexible
RI PT
way to protect the surface of the TNF/GF/EP composite. The 3D network structure of TNF interleaved into the composite can absorb and consume the most of the impact particle energy. A correlation was found between the erosive wear resistance and the
SC
rebound resilience, defined as the absorbed impact energy of the impact particles.
M AN U
Generally, elastomers usually show a better solid particle erosion resistance because of less crack propagation and more elastic/plastic deformation [32]. There are some exposed TPU fibers in TNF/GF/EP composite after erosion as revealed in Fig. 6k. Compared the higher magnification SEM micrographs of EP and TNF/GF/EP
TE D
composite (Fig. 6(h, l)), detail surface morphology and the potential erosive mechanisms are elucidated. It can be concluded that the micro-cracking and related separate ‘blocks’ of epoxy resin materials are caused by repeat impacts of erodent
EP
particles. The occurrence of micro-cracks will further generate the formation of wear
AC C
debris. The TPU fibers perform the ‘bridging’ function as a network between these ‘blocks’, which can transfer the force to the net (Fig. 6l). This feature definitely helps to protect the ‘blocks’ from erosion. Fig. 7 displays 3D views of eroded surface with various geometries on different
samples. As seen in Fig. 7a, the formation and extrusion of chip cause the removal of pure epoxy at 30° impact angle, and form a lot of ‘lips’. However, for TNF/GF/EP composite (see Fig. 7c), it is evident that the eroded surface shows less ‘lips’. Fig. 7b
ACCEPTED MANUSCRIPT and d present the eroded surface of EP and TNF/GF/EP at 90° impact angle. It is interesting to note that the obvious difference of scale bars, i.e., height from crest to trough between 30° and 90° of EP and TNF/GF/EP, this is mainly due to the smaller
RI PT
plastic deformation of EP. From Fig. 7e, most of glass fibers were kept tightly in place and good adhesion was observed between epoxy and glass fibers. However, at 90°, the epoxy in GF/EP was easily removed, leading to breakage of the fibers and
AC C
EP
TE D
M AN U
SC
fiber-epoxy de-bonding (Fig. 7f).
Fig. 7. Three-dimensional views at impingement angles of 30° (left column) and 90° (right column) of eroded (a), (b) EP, (c), (d) TNF/GF/EP and (e), (f) GF/EP.
3.3. Inter-laminar fracture toughness of Mode-I The typical load-displacement curves for DCB test is given in Fig. 8. For both composites with and without TNF interleaves, the loading force initially raised
ACCEPTED MANUSCRIPT linearly as the inter-laminar pre-crack growth, followed by a linear deviation, and ends with a load drop. More interestingly, the composite modified by TNF displays a conspicuous higher Pmax as compared to GF/EP composite. The average fracture
RI PT
toughness energy under Mode-I loading, GIC, of both composites DCB samples are demonstrated in Fig. 9. There is a remarkable increase (about 78%) in GIC value for
M AN U
SC
the TNF/GF/EP composite, compared with the composite without the TNF interleave.
Fig. 8. Typical load-displacement curves during DCB test of GF/EP and
AC C
EP
TE D
TNF/GF/EP.
Fig. 9. Comparison between GF/EP and TNF/GF/EP, for the GIC in Mode-I tests.
The interfacial fracture surfaces of the GF/EP and TNF/GF/EP specimens under
Mode-I test were examined by SEM and presented in Fig. 10. There is obvious difference between the failure mode of two kinds of the specimens. The fracture surface of the reference specimen (Fig. 10a) is relatively smooth, indicating low crack
ACCEPTED MANUSCRIPT growth capacity of the epoxy due to continuous structure of the epoxy matrix layer [33]. Weak interfacial strength and de-bonding are the main failure mechanisms in GF/EP composites. On the contrary, the matrix of the composite that comprises the
RI PT
TNF as interleaf shows ductile failure (Fig. 10b). The significantly increased toughness in TNF/GF/EP composite is due to the enhanced interlayer material interface with the epoxy resin, which is generally sufficient thereby forming strong
SC
bonding between the TPU fibers and the epoxy, leading to a transfer of the impact
M AN U
force and avoiding delaminate of TPU fibers and epoxy. Fiber breakage can be found in Fig. 10c. Due to the extra energy required for TNF/GF/EP interface failure, this
TE D
strong bonding inhibit the layered expansion.
EP
Fig. 10. SEM images of fracture surfaces in Mode-I test: (a) GF/EP (×500), (b)
AC C
TNF/GF/EP at low magnification (×500), (c) TNF/GF/EP at high magnification (×5000).
3.4. Inter-laminar fracture toughness of Mode-Ⅱ Typical load-displacements curves of the ENF specimens are given in Fig. 11. For
the reference specimen, the drop is very dramatically, which indicates once the initial crack is present, crack propagates rapidly. The composite comprises the TNF as interleaf in its in-between, the decrease of the load displacement curve is relatively flat and the critical load is very high, which indicates that the crack growth of these
ACCEPTED MANUSCRIPT composites encountered higher resistance. Fig. 12 exhibited the average values of GⅡC for both ENF samples. A significant enhance in the GⅡC is clearly observed. The average GⅡC value for TNF/GF/EP samples was 115% higher than that for the
M AN U
SC
RI PT
composite without TNF interleave.
TE D
Fig. 11. Typical load-displacement curves during ENF test of GF/EP and TNF/GF/EP.
EP
Fig. 12. GⅡC of GF/EP and TNF/GF/EP in Mode-Ⅱ.
AC C
The interfacial fracture surfaces of ENF samples with and without interleaves under Mode-Ⅱ test are demonstrated in Fig. 13a and b. The fracture surface of the reference sample (Fig. 13a) presents a smooth surface which shows a lower fracture resistance. In addition, fractured glass fibers are found for shear induced bending under Mode-Ⅱ test. However, the hybrid composite comprises the TNF as interleaf in its in-between exhibits rough fracture surface, as revealed in Fig. 13b. The pulled-off of TPU fiber and the deflection behavior of the crack will increase the unevenness of the material
ACCEPTED MANUSCRIPT fracture surface, and the higher the surface unevenness of the sample, the greater the
RI PT
fracture energy required.
TNF/GF/EP (×200).
M AN U
4. Conclusions
SC
Fig. 13. SEM images of fracture surfaces in Mode-Ⅱ: (a) GF/EP (×200), (b)
This work suggests a promising routine to enhance particle erosion resistance and fracture toughness by interleaving the laminates with lightweight thermoplastic polyurethane nonwoven fabrics (TNF). According to the results of the research, the
TE D
conclusions are as follows:
1. Thermoplastic polyurethane nonwoven fabric was prepared by a melt-blown
EP
process and used to prepare the TNF/GF/EP composite through vacuum-assisted resin transfer molding process.
AC C
2. TNF/GF/EP composite demonstrated maximum erosion rate at 30° and minimum at 90° impingement angle, showing ductile erosion behavior, whereas the maximum erosion rate of GF/EP composite appears at 90°, which reflects typical brittle erosion behavior. 3. Compared with conventional GF/EP composites, the erosion resistance of TNF/GF/EP composites were improved by 99%, 108%, 205% and 493% at the impingement angles of 20°, 30°, 45° and 90°, respectively. The 3D network structure
ACCEPTED MANUSCRIPT of TNF interleaved into the composite can help to absorb and consume the energy to a larger extent. 4. Toughening of GF/EP composites was also achieved by interleaving TNF. It offered
Mode-I and 115% for Mode-Ⅱ.
RI PT
the best overall performance, especially for improved fracture toughness, 78% for
5. The enhancement in both GIC and GⅡC is mainly due to the improved resistance to
SC
crack growth in the interlayer which requires extra energy.
M AN U
Acknowledgement
We express our great thanks to the National Natural Science Foundation of China (11432003), Postdoctoral Fund of China (2016M592305), and Foundation of Henan (16A430010), Additional appreciation goes to National Key Research and
References
TE D
Development Programs (2016YFB0101602).
[1] R.J.A. Hamad, M.A.M. Johari, R.H. Haddad, Mechanical properties and bond
EP
characteristics of different fiber reinforced polymer rebars at elevated temperatures,
AC C
Constr. Build. Mater. 142 (2017) 521-535. [2] U. Javaid, Z.M. Khan, M.B. Khan, M. Bassyouni, M.S. Abdel-Hamid, M.H. Abdel-Aziz, Fabrication and thermo-mechanical characterization of glass fiber/vinyl ester wind turbine rotor blade, Compos. Part B Eng. 91 (2016) 257-266. [3] S. Biswas, A. Satapathy, Tribo-performance analysis of red mud filled glass-epoxy composites using Taguchi experimental design, Mater. Des. 30 (2009) 2841-2853. [4] X. Wang, Y. Pan, Y. Qin, M. Voigt, X. Liu, G. Zheng, Q. Chen, D.W. Schubert, C.
ACCEPTED MANUSCRIPT Liu, C. Shen, Creep and recovery behavior of injection-molded isotactic polypropylene with controllable skin-core structure, Polym. Test. 69 (2018). DOI: 10.1016/j.polymertesting.2018.05.040.
RI PT
[5] J. Jing, X. Liu, M. Lian, Y. Pan, Q. Chen, H. Liu, G. Zheng, Z. Guo, D.W. Schubert, C. Shen, C. Liu, Self-reinforcing and toughening isotactic polypropylene via melt sequential injection molding, Polym. Test. 67 (2018), 183-189.
SC
[6] M. Bagci, Determination of solid particle erosion with Taguchi optimization
M AN U
approach of hybrid composite systems, Tribol. Int. 94 (2016) 336-345.
[7] P. Panda, S. Mantry, S. Mohapatra, S.K. Singh, A. Satapathy, A study on erosive wear analysis of glass fiber-epoxy-AlN hybrid composites, J. Compos. Mater. 48 (2014) 107-118.
TE D
[8] A. Papadopoulos, G. Gkikas, A.S. Paipetis, N.M. Barkoula, Effect of CNTs addition on the erosive wear response of epoxy resin and carbon fibre composites, Compos. Part A Appl. Sci. Manuf. 84 (2016) 299-307.
EP
[9] S.S. Mahapatra, A. Patnaik, Study on mechanical and erosion wear behavior of
AC C
hybrid composites using Taguchi experimental design, Mater. Des. 30 (2009) 2791-2801.
[10] A. Patnaik, A. Satapathy, N. Chand, N.M. Barkoula, S. Biswas, Solid particle erosion wear characteristics of fiber and particulate filled polymer composites: A review, Wear 268 (2010) 249-263. [11] X. Liu, Y. Pan, G. Zheng, D.W. Schubert, Rheological and electrical behavior of poly(methyl methacrylate)/carbon black composites as investigated by creep recovery
ACCEPTED MANUSCRIPT in shear, Compos. Sci. Technol. 128 (2016) 1-7. [12] M. Qu, F. Nilsson, Y. Qin, G. Yang, Y. Pan, X. Liu, G.H. Rodriguez, J. Chen, C. Zhang, D.W. Schubert, Electrical conductivity and mechanical properties of melt-spun
RI PT
ternary composites comprising PMMA, carbon fibers and carbon black, Compos. Sci. Technol. 150 (2017) 24-31.
[13] N. Zhang, F. Yang, C. Shen, J. Castro, L.J. Lee, Particle erosion on carbon
SC
nanofiber paper coated carbon fiber/epoxy composites, Compos. Part B Eng. 54
M AN U
(2013) 209-214.
[14] N. Zhang, F. Yang, D. Guerra, C. Shen, J. Castro, J.L. Lee, Enhancing particle erosion
resistance
of
glass-reinforced
polymeric
composites
using carbon
nanofiber-based nanopaper coatings, J. Appl. Polym. Sci. 129 (2013) 1875-1881.
TE D
[15] Y. Zeng, H. Liu, Y.W. Mai, X. Du, Improving interlaminar fracture toughness of carbon fibre/epoxy laminates by incorporation of nano-particles, Compos. Part B Eng. 43 (2012) 90-94.
EP
[16] W. Wang, Y. Takao, T. Matsubara, H.S. Kim, Improvement of the interlaminar
AC C
fracture toughness of composite laminates by whisker reinforced interlamination, Compos. Sci. Technol. 62 (2002) 767-774. [17] M. Hojo, T. Ando, M. Tanaka, T. Adachi, S. Ochiai, Y. Endo, Modes I and Ⅱ interlaminar fracture toughness and fatigue delamination of CF/epoxy laminates with self-same epoxy interleaf, Int. J. Fatigue. 28 (2006) 1154-1165. [18] M. Yasaee, I.P. Bond, R.S. Trask, E.S. Greenhalgh, Mode-Ⅱ interfacial toughening through discontinuous interleaves for damage suppression and control,
ACCEPTED MANUSCRIPT Compos. Part A Appl. Sci. Manuf. 43 (2012) 121-128. [19] M. Guo, X. Yi, G. Liu, L. Liu, Simultaneously increasing the electrical conductivity and fracture toughness of carbon-fiber composites by using silver
RI PT
nanowires-loaded interleaves, Compos. Sci. Technol. 97 (2014) 27-33. [20] S. Hamer, H. Leibovich, A. Green, R. Avrahami, E. Zussman, A. Siegmann, D. Sherman, Mode-I and Mode-Ⅱ fracture energy of MWCNT reinforced nanofibrilmats
SC
interleaved carbon/epoxy laminates, Compos. Sci. Technol. 90 (2014) 48-56.
M AN U
[21] G. Arena, K. Friedrich, D. Acierno, E. Padenko, P. Russo, G. Filippone, J. Wagner, Solid particle erosion and viscoelastic properties of thermoplastic polyurethanes, Eur. Polym. J. 9 (2015) 166-176.
[22] D. Acierno, L. Sanguigno, G. Arena, K. Friedrich, E. Padenko, P. Russo, Erosion
TE D
behavior and mechanical properties of thermoplastic polyurethanes, Conference Proceedings, 1599 (2014) 110-113.
[23] S. Arjula, A.P. Harsha, Study of erosion efficiency of polymers and polymer
EP
composites, Polym. Test. 25 (2006) 188-196.
AC C
[24] Y.E. Lee, L.C. Wadsworth, Fiber and web formation of melt-blown thermoplastic polyurethane polymers, J. Appl. Polym. Sci. 105 (2007) 3724-3727. [25] A. Begenir, S. Michielsen, B. Pourdeyhimi, Melt-blowing thermoplastic polyurethane and polyether-block-amide elastomers: effect of processing conditions and crystallization on web properties, Polym. Eng. Sci. 49 (2009) 1340-1349. [26] Y.O. Kang, J.N. Im, W.H. Park, Morphological and permeable properties of antibacterial double-layered composite nonwovens consisting of microfibers and
ACCEPTED MANUSCRIPT nanofibers, Compos. Part B Eng. 75 (2015) 256-263. [27] J.R. Mohanty, Investigation on solid particle erosion behavior of date palm leaf fiber-reinforced polyvinyl pyrrolidone composites, J. Thermoplast. Compos. Mater.
RI PT
30 (2017) 1003-1016. [28] G.P. Tilly, A two stage mechanism of ductile erosion, Wear 23 (1973) 87-96.
[29] K. Friedrich, X. Pei, A. Almajid, Specific erosive wear rate of neat polymer films
SC
and various polymer composites, J. Reinf. Plast. Compos. 32 (2013) 631-643.
M AN U
[30] N.M. Barkoula, J. Karger-Kocsis, Solid particle erosion of unidirectional GF reinforced EP composites with different fiber/matrix adhesion, J. Reinf. Plast. Compos. 21 (2002) 1377-1388.
[31] J. Chen, I.M. Hutchings, T. Deng, S.A. Bradley, K.K. Koziol, The effect of
TE D
carbon nanotube orientation on erosive wear resistance of CNT-epoxy based composites, Carbon 73 (2014) 421-431.
[32] I.M. Hutchings, D.W.T. Deuchar, A.H. Muhr, Erosion of unfilled elastomers by
EP
solid particle impact, J. Mater. Sci. 22 (1987) 4071-4076.
AC C
[33] V. Kostopoulos, A. Kotrotsos, S. Tsantzalis, P. Tsokanas, T. Loutas, A.W. Bosman, Toughening and healing of continuous fibre reinforced composites by supramolecular polymers, Compos. Sci. Technol. 128 (2016) 84-93.
ACCEPTED MANUSCRIPT A new method of preparing glass fiber/epoxy (GF/EP) composites with thermoplastic polyurethane nonwoven fabric (TNF) was reported. The erosive wear resistance of TNF/GF/EP composites was improved.
AC C
EP
TE D
M AN U
SC
RI PT
Toughening of GF/EP composites was achieved by interleaving TNF.
S0142-9418(18)30708-6
DOI:
10.1016/j.polymertesting.2018.06.005
Reference:
POTE 5502
To appear in:
Polymer Testing
Received Date: 30 April 2018 Revised Date:
4 June 2018
Accepted Date: 4 June 2018
Please cite this article as: G. Lv, N. Zhang, M. Huang, C. Shen, J. Castro, K. Tan, X. Liu, C. Liu, The remarkably enhanced particle erosion resistance and toughness properties of glass fiber/epoxy composites via thermoplastic polyurethane nonwoven fabric, Polymer Testing (2018), doi: 10.1016/ j.polymertesting.2018.06.005. 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
The remarkably enhanced particle erosion resistance and toughness properties of glass fiber/epoxy composites via
RI PT
thermoplastic polyurethane nonwoven fabric Guangchao Lva, Na Zhanga*, Ming Huanga, Changyu Shena, Jose Castroc, Kunlun Tand, Xianhu Liua, Chuntai Liua,b*
National Engineering Research Center for Advanced Polymer Processing Technology,
b
M AN U
Zhengzhou University, Zhengzhou 450002, China
SC
a
Advanced Research Center for Polymer Processing Engineering of Guangdong
Province Guangdong Industry Polytechnic, Guangzhou 510300, China c
Department of integrated systems Engineering, The Ohio state university, 1971 Neil
d
TE D
Avenue, Columbus, Ohio 43210, USA
PGTEX CHINA CO., LTD, Changzhou, Jiangshu, Xinbei District, 213135, China
ABSTRACT: In this work, a new method of preparing glass fiber/epoxy (GF/EP)
EP
composites with excellent solid particle erosion resistance and fracture toughness
AC C
properties via thermoplastic polyurethane nonwoven fabric (TNF) was reported. TNF was prepared by virtue of the melt-blown (MB) process, and the composites were fabricated by vacuum-assisted resin transfer molding (VARTM) technique. Solid particle erosion characteristics of the composites were investigated in a confined space by impinging angular silica particles with a size about 300 µm. Compared with conventional GF/EP composites, the erosive wear resistance of TNF/glass fiber/epoxy
*
Corresponding authors. E-mail address: [email protected]; [email protected]
ACCEPTED MANUSCRIPT (TNF/GF/EP) composites were improved by 99%, 108%, 205% and 493% at the impingement angles of 20°, 30°, 45° and 90°, respectively. Meanwhile, increase of 78% in GIC and 115% in GⅡC were achieved. Microstructure characterizations,
RI PT
including scanning electron microscopy (SEM) and optical microscopy were performed to investigate the variations of microstructure and further to establish the relationship between epoxy matrix and glass fibers. This work suggests a promising
SC
routine to improve particle erosion resistance and fracture toughness by interleaving
M AN U
the laminates with lightweight thermoplastic polyurethane nonwoven fabrics. Key words: Polymer matrix composites; Erosive wear; Interlaminar toughened; Glass fiber; Erosion rate; Thermoplastic polyurethanes. 1. Introduction
TE D
Recently, fiber-reinforced polymers (FRPs) with high modulus rate and high strength rate are continuously replacing more and more traditional materials, for instance, plastics, wood and metallic materials [1-5]. In general, the FRPs are always
EP
applied in severe working environments, the erosive wear behavior of the FRPs are of
AC C
vital importance. Therefore, great efforts have been made to improve the erosive wear resistance of FRPs via particulate fillers [6-10]. Bagci et al. [6] investigated the erosive wear behavior of glass fiber/epoxy composites (GFECs) in the presence of Al2O3+SiO2 particulates, and Panda et al. [7] studied the influence of AlN on the erosive wear response of GFECs. Above hybrid composites, however, suffers from the intrinsic limitation that the improvement of erosion resistance is limited and/or the manufacturing processes become difficult [11,12]. In view of this, a periodically
ACCEPTED MANUSCRIPT interleaved fiber laminate system, wherein thin protective layers are interleaved into each ply, has been used [13, 14]. The protective layer can effectively protect the fiber from solid particle erosion, which is benefit by the high strength of carbon nanofibers
RI PT
(CNFs) and their nanoscale structure. At the same time, considering the laminate feature of FRP composites, an attempt has also been made to improve the inter-laminar fracture toughness over the past
SC
years [15-17]. The existence of insufficient fracture toughness and delamination may
M AN U
lead to complete failure of the composite structure. Many solutions have been proposed to improve toughness by virtue of changing the materials structure and components used in the composites. Specifically, a ductile polymer microfiber fabric inserted into the interlayer has been employed in this study to investigate the effect.
TE D
Microfiber fabrics are highly porous and good for avoiding impeding the flow of resin during the filling process. In addition, good bonding with the matrix is available due to the particularly high specific surface area. Yasaee et al. [18] investigated the
EP
Mode-Ⅱ inter-laminar fracture toughness by various interply strips such as polyimide
AC C
thermoplastic films, thermoset epoxy films, glass/epoxy prepreg and chopped Aramid fibers, reporting a 200% Mode-Ⅱ inter-laminar fracture toughness improvement. The investigation by Yi et al. [19] showed that composites interleaved with Aramid Nonwoven Fabric (ANF) had better fracture energy under Mode-I and Mode-Ⅱ loading (GIC and GⅡC) as compared with the reference one composite. Hamer et al. [20] found that the significant enhanced GIC and GⅡC could be realized by adding electro-spun nylon reinforced with MWCNTs into carbon/epoxy composites.
ACCEPTED MANUSCRIPT Thermoplastic polyurethanes (TPUs) have attracted a significant attention due to their good processability and wear resistant property deriving from their peculiar molecular structure [21-23]. This makes TPUs suitable for protective coating or films
RI PT
with minimal erosive wear. To the best of our knowledge and experience, researches on erosive wear resistance behaviors of TNF reinforced GF/EP composites have not been found in the open literatures. In view of this, we studied the use of a thin
SC
protective layer of TNF for enhancing the solid particle erosion resistance of GF/EP
M AN U
composites in this study. TNF were prepared by the melt-blown process [24-26], and the composites were fabricated by the VARTM technique. Unexpectedly, the TNF can not only enhance composite wear resistance, but also improve composite Mode-I and Mode-Ⅱ inter-laminar fracture toughness. Underlying possible wear mechanism
TE D
was also investigated by the scanning electron microscopy and optical microscopy views. These features may make TNF applicable for FRP composites in the fields where both excellent wear resistance and high toughness properties are required
EP
simultaneously.
AC C
2. Experimental 2.1. Materials
Polyether-based thermoplastic polyurethane (TPU) (Elastollan 1185A), the density
is 1.12 g/cm3 and a melt flow index is 17.5 g/10 min (215 °C, 10 kg), which was bought from BASF Co. Ltd., China. E-Glass multi-axial fabrics, ECW800 (0/+45/90/-45), used as the long fiber reinforcement was provided by Taishan fiber glass INC., China. The epoxy resin used was LT-5078A, and a diamine curing agent,
ACCEPTED MANUSCRIPT LT-5078B with an amine value of about 420-520 mg [KOH]/g, provided by Wells Advanced Materials Co. Ltd., China. Blocky, sharp edged silica with the average diameter is 300 µm (Fig. S1), which were selected as the erodent particles, offered by
M AN U
SC
RI PT
Baige Co. Ltd., China.
TE D
Fig. 1. (a) TPU nonwoven fabric, (b) SEM image of TPU nonwoven fabric.
Fig. 2. Lay-up sequences of flat laminates for erosion test (a, b) and fracture
EP
toughness test (c, d), fiber orientations of E-Glass multi-axial fabrics (e).
AC C
2.2. Preparation of TNF, EP and composites Melt-blown (MB) pilot line shown schematically in Fig. S2. MB TPU webs were
gathered at die-to-collector distances (DCD) of 50 cm with a collector speed of 10.3 m/min, the air and die temperatures were set at 245 and 220 °C, respectively. The thickness of TNF is approximately 0.2 mm. The digital image of TPU nonwoven fabric and its SEM image are displayed in Fig. 1 (a, b), respectively. The hybrid resin was degassed in vacuum oven for about 10 min until there are no
ACCEPTED MANUSCRIPT bubbles, transferred to a Polytetrafluoroethylene (PTFE) mold and cured at 25 °C (room temperature) for 24 h, post cured at 120 °C for an additional 3 h and then pure EP specimen can be obtained after stripping.
RI PT
VARTM was selected to manufacture the GF/EP and TNF/GF/EP composites. The glass fibers and TNF were laid-up in laminate sequences as shown in Fig. 2. Vacuum was adopted to compel the bag to compress tightly against the fiber stack before mold
SC
filling. Make sure there is no leaking in the whole device. The epoxy hybrid was
M AN U
degassed in vacuum for about 10 min, the resin was infused into the no leakage mold afterwards. The specimen was subsequently cured at 25°C for 24 h and then post cured at 120 °C for an additional 3 h. The fiber volume fraction of the composites is 56 %.
TE D
2.3. Erosive Wear Testing
The erodents were accelerated to high velocities by a static pressure and directed onto a test specimen as illustrated in Fig. S3. It mainly includes an air compressor, an
EP
erodent feeder, a sand-blasting gun, and a specimen holder. The prescribed angle (α)
AC C
can be obtained by adjusting the spray gun and the sample holder. Square sample with size of 30 mm×30 mm was cut for solid particle erosion tests. The factors for the erosion trials are listed in Table 1. Before and after the erosion tests, the specimen was cleaned by acetone, dried, and weighed to an accuracy of 0.1 mg by virtue of an electronic balance (FA1004, China), subtract the two experimental data to determine mass loss. The weight erosion rate ( E ) was then obtained through the following formula [27]:
ACCEPTED MANUSCRIPT E=
∆w we
(1)
where ∆w is the weight loss of the test specimen and we is the mass of erodent particles (i.e., testing time × erodent feed rate). The smaller the erosion rate, the
represents an average gained from seven measurements.
SC
Table 1. Erosion test conditions.
Erodent size (µm)
Silica sand
M AN U
Test parameters Erodent
RI PT
better the erosive wear resistance of the composite was. The reported result
300
Erodent feed rate (g/s)
4.98
Test temperature
RT (25 °C)
60
Nozzle diameter (mm)
6.5
Air pressure (MPa)
0.345
EP
TE D
Distance from nozzle to sample (mm)
AC C
2.4. Testing for fracture toughness According to ASTM D5528, GIC was measured by Double Cantilever Beam (DCB)
test. Fig. 3a shows the schematic image for the DCB test. The DCB specimens were 150 mm long and 25 mm wide (containing 50 mm length pre-cracks, which were prepared by PTFE film). The piano hinges (used to apply the load) were attached to each specimen under pressure through a cyanoacrylate adhesive. The DCB test were executed at cross head speed of 1 mm/min on a common testing machine (Instron 5585), while the load and crack extension were recorded. At least seven specimens
ACCEPTED MANUSCRIPT were tested to drive the mean data. Mode-I critical energy release rate, GIC, was calculated using a modified beam theory (MBT) way:
GIC =
3Pδ 2 b( a + ∆ )
(2)
RI PT
Where δ is the displacement of the load point, b is the sample width, a is the crack length, P is the load, and ∆ is the horizontal axis intercept from a - C 1/3
SC
curve. The compliance, C , is the ratio of the load point displacement to the applied load, δ P .
M AN U
GⅡC was measured by end-notch flexure (ENF) test as shown in Fig. 3b. The dimension of the samples was 140 mm (in length) × 25 mm (in width), containing 50 mm length pre-cracks. The specimen was loaded twice and was loaded in displacement control at cross head speed of 1 mm/min on a common testing machine
TE D
(Instron 5585). The first loading with the span length 70 mm was stopped until crack progression occurred. Then the sample was reloaded with the span length 100 mm,
EP
until a sudden load drop was observed. At least seven samples were tested. Mode-Ⅱ
AC C
critical energy release rate, GⅡC, was figured out by equation (3):
where P
GIIC =
9 Pδa 2 2b(2 L3 + 3a 3 )
(3)
is the maximum load got from the straight part of the second
load–displacement curve, b is the width of the sample, a is the beginning crack length (25 mm), L is a half of the span length (50 mm), and δ is the load point displacement.
ACCEPTED MANUSCRIPT
Mode-Ⅱ test.
SC
2.5. Microscopy
RI PT
Fig. 3. Geometry and dimensions of fracture toughness sample: (a) Mode-Ⅱ test; (b)
The eroded surfaces and fracture surfaces of fracture toughness samples were
M AN U
coated with a thin layer of gold, and examined using a scanning electron microscope (SEM, ZEISS EVO-18). The eroded surfaces of the samples were also characterized by a Leica DVM6 optical microscopy, which provide both 3D display and plan display of the surface structures.
TE D
3. Results and discussion 3.1. Erosion performance
EP
Fig. 4 is the erosion rates of EP and its composites have been drawn into a curve with an angle as a function. Ductile and brittle erosion behavior can be distinctly
AC C
varying from this graph. EP and TNF/GF/EP composite demonstrate maximum erosion rate at 30° and minimum at 90° impingement angle, showing ductile erosion behavior. Whereas the maximum erosion rate of GF/EP composite appears at 90°, which reflects typical brittle erosion behavior. According to the literature [28], the maximum erosion rate was at about 30° and 90° impingement angles for ductile and brittle erosions, respectively. The erosion rate of GF/EP composite is higher than that of EP regardless of impingement angles. The combine of glass fibers decreases the
ACCEPTED MANUSCRIPT solid particle erosion resistance of EP and alters the erosive wear behavior of the pure EP owing to the brittle nature of glass fibers. TNF/GF/EP composite exhibits better erosive wear resistance than that of pure EP: the erosion rate of TNF/GF/EP
RI PT
composites was decreased by 10%, 22%, 6% and 21% at impingement angles of 20°, 30°, 45° and 90°, respectively. On the other hand, compared to GF/EP composites, the erosion resistance of TNF/GF/EP composites was enhanced by 99%, 108%, 205% and
TE D
M AN U
SC
493% at impingement angles of 20°, 30°, 45° and 90°, respectively.
Fig. 4. Erosion rates of EP and its composites as a function of impingement angle.
EP
Fig. 5 shows the mass loss of different samples as a function of erosion time at the impingement angles of 30° and 90°, respectively. The curve shows a steady state, and
AC C
the rate of mass loss is proportional to the time of erosion. Although EP and TNF/GF/EP displayed ductile erosion behavior, no incubation period was discovered in the experiment process [29].
RI PT
ACCEPTED MANUSCRIPT
Fig. 5. The mass loss of different samples as a function of erosion time at different
SC
impingement angles: (a) 30°, (b) 90°. 3.2. Morphology of eroded surface
M AN U
Fig. 6 (a-d) presents the eroded surfaces of GF/EP composite at 30°, 60° and 90° impingement angles. The erosion behavior of GF/EP composite samples is brittle in nature. The failure mode in GF/EP composite is a complicated process including surface micro-cracking, surface matrix removing, fiber/matrix de-bonding, fiber
TE D
fracture and material removal [30]. In addition, many micro-cracks produced by the erodent particles can be observed in glass fibers and many tiny fragments of fibers
EP
also can be discovered on the surface of erosion. At oblique impingement angles, it has been observed that the impact force can be divided into two-part: one is parallel
AC C
(Fp) to the target surface and the other is vertical (Fv). Fp is in charge of the abrasive and Fv mainly controls the impact force.
ACCEPTED MANUSCRIPT Impinging direction
TE D
M AN U
SC
RI PT
Impinging direction
Fig. 6. SEM images of eroded GF/EP (left column), EP (center column) and TNF/GF/EP (right column) at impingement angles of: (a), (e), (i) 30°, (b), (f), (j) 60°,
EP
(c), (d), (g), (h), (k), (l) 90°.
AC C
At 30° where erodent particles had small Fv to the target surface, the surface matrix was chipped away mainly via the abrasive action of Fp of impinging particles (Fig. 6a). In addition, many cracks can be clearly observed, which were resulted from the small Fv of particles. Fewer fibers were broken and eroded away by Fv, most of glass fibers were kept firmly in place. Good adhesion between matrix and glass fibers can be observed. With the impingement angle increases, the abrasive action of Fp steps down, while the vertical component turns more crucial. This leads to widespread damage
ACCEPTED MANUSCRIPT and detachment of glass fibers at 60° impact (Fig. 6b). At 90°, direct particles impact action produced the highest kinetic energy which was consumed by a lot of micro-cracks and glass fibers plastic deformation (Fig. 6c). The interfacial de-bonding
RI PT
between glass fibers and epoxy resin was also discovered. Owing to repeated impacts, the brittle glass fibers formed tiny wear debris but still adhered to the eroded surface. Especially in the enlarged area, as shown in Fig. 6d, some deep craters (see the white
SC
arrows) were formed, which indicated that due to the lack of adequate support from
M AN U
the matrix, the direct impact causes a large amount of de-bonding and breakage of glass fibers，while the rough surface causing larger removal of materials. Hence, for GF/EP composite, it is concluded that the vertical component of the velocity vector plays extremely vital role in damage [31].
TE D
The eroded surface of EP and TNF/GF/EP composite was viewed by SEM analysis at 30°, 60° and 90° impingement angles (Fig. 6 (e-l)). At 30° (see Fig. 6e and i), the thin layers of surface matrix were cut off mainly due to the remarkable cutting effects.
EP
In this angle, the cutting action of impinging particles results in a large number of
AC C
‘lips’, which were shown by means of the white features. These ‘lips’ prone to be in the same direction as the erodent particles. By comparison, the surface protected via the TNF, the epoxy resin and TNF were firmly bound and the eroded surface presents less ‘lips’. At the impingement angle of 60°, the ‘lips’ were not visible (see Fig. 6f and j). It is interesting to note that larger fragment of surface materials of pure EP were cut off (see the white arrows on Fig. 6f), mainly due to the higher vertical component of the velocity vector. From Fig. 6g, a number of craters caused by erodent particles can
ACCEPTED MANUSCRIPT be seen at 90° impingement angle, which proves that a large amount of surface material is completely chipped off. However, there are few craters of TNF/GF/EP composite after erosion (Fig. 6k). It is up to the presence of TNF contribute a flexible
RI PT
way to protect the surface of the TNF/GF/EP composite. The 3D network structure of TNF interleaved into the composite can absorb and consume the most of the impact particle energy. A correlation was found between the erosive wear resistance and the
SC
rebound resilience, defined as the absorbed impact energy of the impact particles.
M AN U
Generally, elastomers usually show a better solid particle erosion resistance because of less crack propagation and more elastic/plastic deformation [32]. There are some exposed TPU fibers in TNF/GF/EP composite after erosion as revealed in Fig. 6k. Compared the higher magnification SEM micrographs of EP and TNF/GF/EP
TE D
composite (Fig. 6(h, l)), detail surface morphology and the potential erosive mechanisms are elucidated. It can be concluded that the micro-cracking and related separate ‘blocks’ of epoxy resin materials are caused by repeat impacts of erodent
EP
particles. The occurrence of micro-cracks will further generate the formation of wear
AC C
debris. The TPU fibers perform the ‘bridging’ function as a network between these ‘blocks’, which can transfer the force to the net (Fig. 6l). This feature definitely helps to protect the ‘blocks’ from erosion. Fig. 7 displays 3D views of eroded surface with various geometries on different
samples. As seen in Fig. 7a, the formation and extrusion of chip cause the removal of pure epoxy at 30° impact angle, and form a lot of ‘lips’. However, for TNF/GF/EP composite (see Fig. 7c), it is evident that the eroded surface shows less ‘lips’. Fig. 7b
ACCEPTED MANUSCRIPT and d present the eroded surface of EP and TNF/GF/EP at 90° impact angle. It is interesting to note that the obvious difference of scale bars, i.e., height from crest to trough between 30° and 90° of EP and TNF/GF/EP, this is mainly due to the smaller
RI PT
plastic deformation of EP. From Fig. 7e, most of glass fibers were kept tightly in place and good adhesion was observed between epoxy and glass fibers. However, at 90°, the epoxy in GF/EP was easily removed, leading to breakage of the fibers and
AC C
EP
TE D
M AN U
SC
fiber-epoxy de-bonding (Fig. 7f).
Fig. 7. Three-dimensional views at impingement angles of 30° (left column) and 90° (right column) of eroded (a), (b) EP, (c), (d) TNF/GF/EP and (e), (f) GF/EP.
3.3. Inter-laminar fracture toughness of Mode-I The typical load-displacement curves for DCB test is given in Fig. 8. For both composites with and without TNF interleaves, the loading force initially raised
ACCEPTED MANUSCRIPT linearly as the inter-laminar pre-crack growth, followed by a linear deviation, and ends with a load drop. More interestingly, the composite modified by TNF displays a conspicuous higher Pmax as compared to GF/EP composite. The average fracture
RI PT
toughness energy under Mode-I loading, GIC, of both composites DCB samples are demonstrated in Fig. 9. There is a remarkable increase (about 78%) in GIC value for
M AN U
SC
the TNF/GF/EP composite, compared with the composite without the TNF interleave.
Fig. 8. Typical load-displacement curves during DCB test of GF/EP and
AC C
EP
TE D
TNF/GF/EP.
Fig. 9. Comparison between GF/EP and TNF/GF/EP, for the GIC in Mode-I tests.
The interfacial fracture surfaces of the GF/EP and TNF/GF/EP specimens under
Mode-I test were examined by SEM and presented in Fig. 10. There is obvious difference between the failure mode of two kinds of the specimens. The fracture surface of the reference specimen (Fig. 10a) is relatively smooth, indicating low crack
ACCEPTED MANUSCRIPT growth capacity of the epoxy due to continuous structure of the epoxy matrix layer [33]. Weak interfacial strength and de-bonding are the main failure mechanisms in GF/EP composites. On the contrary, the matrix of the composite that comprises the
RI PT
TNF as interleaf shows ductile failure (Fig. 10b). The significantly increased toughness in TNF/GF/EP composite is due to the enhanced interlayer material interface with the epoxy resin, which is generally sufficient thereby forming strong
SC
bonding between the TPU fibers and the epoxy, leading to a transfer of the impact
M AN U
force and avoiding delaminate of TPU fibers and epoxy. Fiber breakage can be found in Fig. 10c. Due to the extra energy required for TNF/GF/EP interface failure, this
TE D
strong bonding inhibit the layered expansion.
EP
Fig. 10. SEM images of fracture surfaces in Mode-I test: (a) GF/EP (×500), (b)
AC C
TNF/GF/EP at low magnification (×500), (c) TNF/GF/EP at high magnification (×5000).
3.4. Inter-laminar fracture toughness of Mode-Ⅱ Typical load-displacements curves of the ENF specimens are given in Fig. 11. For
the reference specimen, the drop is very dramatically, which indicates once the initial crack is present, crack propagates rapidly. The composite comprises the TNF as interleaf in its in-between, the decrease of the load displacement curve is relatively flat and the critical load is very high, which indicates that the crack growth of these
ACCEPTED MANUSCRIPT composites encountered higher resistance. Fig. 12 exhibited the average values of GⅡC for both ENF samples. A significant enhance in the GⅡC is clearly observed. The average GⅡC value for TNF/GF/EP samples was 115% higher than that for the
M AN U
SC
RI PT
composite without TNF interleave.
TE D
Fig. 11. Typical load-displacement curves during ENF test of GF/EP and TNF/GF/EP.
EP
Fig. 12. GⅡC of GF/EP and TNF/GF/EP in Mode-Ⅱ.
AC C
The interfacial fracture surfaces of ENF samples with and without interleaves under Mode-Ⅱ test are demonstrated in Fig. 13a and b. The fracture surface of the reference sample (Fig. 13a) presents a smooth surface which shows a lower fracture resistance. In addition, fractured glass fibers are found for shear induced bending under Mode-Ⅱ test. However, the hybrid composite comprises the TNF as interleaf in its in-between exhibits rough fracture surface, as revealed in Fig. 13b. The pulled-off of TPU fiber and the deflection behavior of the crack will increase the unevenness of the material
ACCEPTED MANUSCRIPT fracture surface, and the higher the surface unevenness of the sample, the greater the
RI PT
fracture energy required.
TNF/GF/EP (×200).
M AN U
4. Conclusions
SC
Fig. 13. SEM images of fracture surfaces in Mode-Ⅱ: (a) GF/EP (×200), (b)
This work suggests a promising routine to enhance particle erosion resistance and fracture toughness by interleaving the laminates with lightweight thermoplastic polyurethane nonwoven fabrics (TNF). According to the results of the research, the
TE D
conclusions are as follows:
1. Thermoplastic polyurethane nonwoven fabric was prepared by a melt-blown
EP
process and used to prepare the TNF/GF/EP composite through vacuum-assisted resin transfer molding process.
AC C
2. TNF/GF/EP composite demonstrated maximum erosion rate at 30° and minimum at 90° impingement angle, showing ductile erosion behavior, whereas the maximum erosion rate of GF/EP composite appears at 90°, which reflects typical brittle erosion behavior. 3. Compared with conventional GF/EP composites, the erosion resistance of TNF/GF/EP composites were improved by 99%, 108%, 205% and 493% at the impingement angles of 20°, 30°, 45° and 90°, respectively. The 3D network structure
ACCEPTED MANUSCRIPT of TNF interleaved into the composite can help to absorb and consume the energy to a larger extent. 4. Toughening of GF/EP composites was also achieved by interleaving TNF. It offered
Mode-I and 115% for Mode-Ⅱ.
RI PT
the best overall performance, especially for improved fracture toughness, 78% for
5. The enhancement in both GIC and GⅡC is mainly due to the improved resistance to
SC
crack growth in the interlayer which requires extra energy.
M AN U
Acknowledgement
We express our great thanks to the National Natural Science Foundation of China (11432003), Postdoctoral Fund of China (2016M592305), and Foundation of Henan (16A430010), Additional appreciation goes to National Key Research and
References
TE D
Development Programs (2016YFB0101602).
[1] R.J.A. Hamad, M.A.M. Johari, R.H. Haddad, Mechanical properties and bond
EP
characteristics of different fiber reinforced polymer rebars at elevated temperatures,
AC C
Constr. Build. Mater. 142 (2017) 521-535. [2] U. Javaid, Z.M. Khan, M.B. Khan, M. Bassyouni, M.S. Abdel-Hamid, M.H. Abdel-Aziz, Fabrication and thermo-mechanical characterization of glass fiber/vinyl ester wind turbine rotor blade, Compos. Part B Eng. 91 (2016) 257-266. [3] S. Biswas, A. Satapathy, Tribo-performance analysis of red mud filled glass-epoxy composites using Taguchi experimental design, Mater. Des. 30 (2009) 2841-2853. [4] X. Wang, Y. Pan, Y. Qin, M. Voigt, X. Liu, G. Zheng, Q. Chen, D.W. Schubert, C.
ACCEPTED MANUSCRIPT Liu, C. Shen, Creep and recovery behavior of injection-molded isotactic polypropylene with controllable skin-core structure, Polym. Test. 69 (2018). DOI: 10.1016/j.polymertesting.2018.05.040.
RI PT
[5] J. Jing, X. Liu, M. Lian, Y. Pan, Q. Chen, H. Liu, G. Zheng, Z. Guo, D.W. Schubert, C. Shen, C. Liu, Self-reinforcing and toughening isotactic polypropylene via melt sequential injection molding, Polym. Test. 67 (2018), 183-189.
SC
[6] M. Bagci, Determination of solid particle erosion with Taguchi optimization
M AN U
approach of hybrid composite systems, Tribol. Int. 94 (2016) 336-345.
[7] P. Panda, S. Mantry, S. Mohapatra, S.K. Singh, A. Satapathy, A study on erosive wear analysis of glass fiber-epoxy-AlN hybrid composites, J. Compos. Mater. 48 (2014) 107-118.
TE D
[8] A. Papadopoulos, G. Gkikas, A.S. Paipetis, N.M. Barkoula, Effect of CNTs addition on the erosive wear response of epoxy resin and carbon fibre composites, Compos. Part A Appl. Sci. Manuf. 84 (2016) 299-307.
EP
[9] S.S. Mahapatra, A. Patnaik, Study on mechanical and erosion wear behavior of
AC C
hybrid composites using Taguchi experimental design, Mater. Des. 30 (2009) 2791-2801.
[10] A. Patnaik, A. Satapathy, N. Chand, N.M. Barkoula, S. Biswas, Solid particle erosion wear characteristics of fiber and particulate filled polymer composites: A review, Wear 268 (2010) 249-263. [11] X. Liu, Y. Pan, G. Zheng, D.W. Schubert, Rheological and electrical behavior of poly(methyl methacrylate)/carbon black composites as investigated by creep recovery
ACCEPTED MANUSCRIPT in shear, Compos. Sci. Technol. 128 (2016) 1-7. [12] M. Qu, F. Nilsson, Y. Qin, G. Yang, Y. Pan, X. Liu, G.H. Rodriguez, J. Chen, C. Zhang, D.W. Schubert, Electrical conductivity and mechanical properties of melt-spun
RI PT
ternary composites comprising PMMA, carbon fibers and carbon black, Compos. Sci. Technol. 150 (2017) 24-31.
[13] N. Zhang, F. Yang, C. Shen, J. Castro, L.J. Lee, Particle erosion on carbon
SC
nanofiber paper coated carbon fiber/epoxy composites, Compos. Part B Eng. 54
M AN U
(2013) 209-214.
[14] N. Zhang, F. Yang, D. Guerra, C. Shen, J. Castro, J.L. Lee, Enhancing particle erosion
resistance
of
glass-reinforced
polymeric
composites
using carbon
nanofiber-based nanopaper coatings, J. Appl. Polym. Sci. 129 (2013) 1875-1881.
TE D
[15] Y. Zeng, H. Liu, Y.W. Mai, X. Du, Improving interlaminar fracture toughness of carbon fibre/epoxy laminates by incorporation of nano-particles, Compos. Part B Eng. 43 (2012) 90-94.
EP
[16] W. Wang, Y. Takao, T. Matsubara, H.S. Kim, Improvement of the interlaminar
AC C
fracture toughness of composite laminates by whisker reinforced interlamination, Compos. Sci. Technol. 62 (2002) 767-774. [17] M. Hojo, T. Ando, M. Tanaka, T. Adachi, S. Ochiai, Y. Endo, Modes I and Ⅱ interlaminar fracture toughness and fatigue delamination of CF/epoxy laminates with self-same epoxy interleaf, Int. J. Fatigue. 28 (2006) 1154-1165. [18] M. Yasaee, I.P. Bond, R.S. Trask, E.S. Greenhalgh, Mode-Ⅱ interfacial toughening through discontinuous interleaves for damage suppression and control,
ACCEPTED MANUSCRIPT Compos. Part A Appl. Sci. Manuf. 43 (2012) 121-128. [19] M. Guo, X. Yi, G. Liu, L. Liu, Simultaneously increasing the electrical conductivity and fracture toughness of carbon-fiber composites by using silver
RI PT
nanowires-loaded interleaves, Compos. Sci. Technol. 97 (2014) 27-33. [20] S. Hamer, H. Leibovich, A. Green, R. Avrahami, E. Zussman, A. Siegmann, D. Sherman, Mode-I and Mode-Ⅱ fracture energy of MWCNT reinforced nanofibrilmats
SC
interleaved carbon/epoxy laminates, Compos. Sci. Technol. 90 (2014) 48-56.
M AN U
[21] G. Arena, K. Friedrich, D. Acierno, E. Padenko, P. Russo, G. Filippone, J. Wagner, Solid particle erosion and viscoelastic properties of thermoplastic polyurethanes, Eur. Polym. J. 9 (2015) 166-176.
[22] D. Acierno, L. Sanguigno, G. Arena, K. Friedrich, E. Padenko, P. Russo, Erosion
TE D
behavior and mechanical properties of thermoplastic polyurethanes, Conference Proceedings, 1599 (2014) 110-113.
[23] S. Arjula, A.P. Harsha, Study of erosion efficiency of polymers and polymer
EP
composites, Polym. Test. 25 (2006) 188-196.
AC C
[24] Y.E. Lee, L.C. Wadsworth, Fiber and web formation of melt-blown thermoplastic polyurethane polymers, J. Appl. Polym. Sci. 105 (2007) 3724-3727. [25] A. Begenir, S. Michielsen, B. Pourdeyhimi, Melt-blowing thermoplastic polyurethane and polyether-block-amide elastomers: effect of processing conditions and crystallization on web properties, Polym. Eng. Sci. 49 (2009) 1340-1349. [26] Y.O. Kang, J.N. Im, W.H. Park, Morphological and permeable properties of antibacterial double-layered composite nonwovens consisting of microfibers and
ACCEPTED MANUSCRIPT nanofibers, Compos. Part B Eng. 75 (2015) 256-263. [27] J.R. Mohanty, Investigation on solid particle erosion behavior of date palm leaf fiber-reinforced polyvinyl pyrrolidone composites, J. Thermoplast. Compos. Mater.
RI PT
30 (2017) 1003-1016. [28] G.P. Tilly, A two stage mechanism of ductile erosion, Wear 23 (1973) 87-96.
[29] K. Friedrich, X. Pei, A. Almajid, Specific erosive wear rate of neat polymer films
SC
and various polymer composites, J. Reinf. Plast. Compos. 32 (2013) 631-643.
M AN U
[30] N.M. Barkoula, J. Karger-Kocsis, Solid particle erosion of unidirectional GF reinforced EP composites with different fiber/matrix adhesion, J. Reinf. Plast. Compos. 21 (2002) 1377-1388.
[31] J. Chen, I.M. Hutchings, T. Deng, S.A. Bradley, K.K. Koziol, The effect of
TE D
carbon nanotube orientation on erosive wear resistance of CNT-epoxy based composites, Carbon 73 (2014) 421-431.
[32] I.M. Hutchings, D.W.T. Deuchar, A.H. Muhr, Erosion of unfilled elastomers by
EP
solid particle impact, J. Mater. Sci. 22 (1987) 4071-4076.
AC C
[33] V. Kostopoulos, A. Kotrotsos, S. Tsantzalis, P. Tsokanas, T. Loutas, A.W. Bosman, Toughening and healing of continuous fibre reinforced composites by supramolecular polymers, Compos. Sci. Technol. 128 (2016) 84-93.
ACCEPTED MANUSCRIPT A new method of preparing glass fiber/epoxy (GF/EP) composites with thermoplastic polyurethane nonwoven fabric (TNF) was reported. The erosive wear resistance of TNF/GF/EP composites was improved.
AC C
EP
TE D
M AN U
SC
RI PT
Toughening of GF/EP composites was achieved by interleaving TNF.