Available online at www.sciencedirect.com
Materials & Design
Materials and Design 29 (2008) 1419–1426 www.elsevier.com/locate/matdes
Technical Report
Residual mechanical properties of carbon/polyphenylenesulphide composites after solid particle erosion Tamer Sinmazc¸elik a
a,c,*
, Sinan Fidan b, Volkan Gu¨nay
c
_ Kocaeli University, Mechanical Engineering Department, Veziroglu Campus, 41040 Izmit, Turkey b _ Kocaeli University, School of Civil Aviation, Aslanbey Campus, 41285 Izmit, Turkey c TUBITAK-MRC, Materials Institute, 41470 Gebze, Turkey Received 17 May 2007; accepted 24 September 2007 Available online 1 October 2007
Abstract The objective of the study is to investigate the residual mechanical properties of cross-ply carbon fibre reinforced polyphenylenesulphide (C-PPS) composites after particle erosion. Angular silica sand particles with the size of 150–200 lm are driven by a static pressure of 1.5, 3 and 4.5 bar and are accelerated along a 50 mm long ceramic nozzle of 5 mm diameter at room temperature. The average velocity of the silica sand at these pressures at the nozzle tip was measured as 20, 40 and 60 m/s with respect to the air pressure. Composite samples clamped on to the specimen holder. The samples on specimen holder were subjected to particle flow at impingement angles between 15 and 90. Erodent mass flow was measured as 4.25, 6.25 and 9 g/s for average velocities of 20, 40 and 60 m/s, respectively. Wear rates were measured by means of weight loss with an electronic balance with an accuracy of 0.1 mg after 2–30 s of particle erosion. The impingement angle was found to have a significant influence on erosion rate. Composite material showed semi-ductile erosion behaviour, with a maximum erosion rate at impingement angle of 45. The morphology of eroded surfaces was examined by using scanning electron microscope (SEM). Possible erosion mechanisms were discussed. The erosion behaviour was not only considered as a material loss but also as exposuring of repeated impacts by means of particles. Great differences were observed between the initial and post-erosion flexural strength of the material. It was concluded that the minimum residual strength values were determined for the samples eroded at impingement angle of 45. Also the samples are found having lower residual flexural properties at higher impingement angles. 2007 Elsevier Ltd. All rights reserved.
1. Introduction Polymer composites have been used for a long time with an increased demand in various engineering fields, particularly in aerospace applications, due to their high specific mechanical properties as compared to the other conventional materials. Also these composites are finding further applications that subjected to solid particle erosion. Examples of such applications can be summarized as; pipe line carrying sand slurries in petroleum refining, helicopter * Corresponding author. Address: Kocaeli University, Mechanical _ Engineering Department, Veziroglu Campus, 41040 Izmit, Turkey. Tel.: +90 262 3351148; fax: +90 262 3352812. E-mail address:
[email protected] (T. Sinmazc¸elik).
0261-3069/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2007.09.003
rotor blades, pump impeller blades, high speed vehicles and aircraft operating in desert environments, water turbines, aircraft engine blades, etc. [1–3]. The most important factors influencing the erosion rate of the composite materials can be summarized under four categories. (1) The properties of the target materials (matrix material properties and morphology, reinforcements type, amount and orientation, interface properties between the matrices and reinforcements, etc.). (2) Environment and testing conditions (temperature, chemical interaction of erodent with the target). (3) Operating parameters (angle of impingement, impinging velocity, particle flux – mass per unit time – etc.). (4) The properties of the erodent (size, shape, type, hardness, etc.) [1,3,4]. Evaluation of the particle erosion behaviour of any polymer composite is not easy,
1420
T. Sinmazc¸elik et al. / Materials and Design 29 (2008) 1419–1426
because the erosion resistance of the material can be evaluated after investigating the combination of many parameters. In general, thermoplastic matrix composites exhibit a ductile erosive wear (plastic deformation, ploughing, ductile tearing). The behaviour of ductile materials is characterized by maximum erosion at low impingement angles (15–30). On the other hand, thermosetting matrix composites erode in a brittle manner (generation and propagation of surface lateral cracks, matrix micro-cracking, fibre–matrix debonding, fibre breakage and material removal) characterized by maximum erosion under normal impingement angle (90). The particle erosion rates of the thermoplastic matrix polymer composites is strongly influenced by the reinforcement type, amount and orientation. By means of increase in volume ratio of reinforcements in thermoplastic composites, semi ductile erosion behaviour occurs and maximum erosion rates shift from 15 to 60. However, as stated above, this failure classification is not definitive because the erosion behaviour of composites depends strongly on the experimental conditions and the composition of the target material. It is well known that impingement angle is one of the most important parameters in erosion behaviour. It is reported in the literature that, when the erosive particles hit the target at low angles, the impact force can be divided in two constituents: one parallel (Fp) to surface of the material and other vertical (Fv). Fp controls the abrasive and Fv is responsible for the impact phenomenon. As the impact angle shifts towards to 90, the effects of Fv become marginal. It is obvious that, in the case of normal erosion, all available energy is dissipated by impact and kinetic energy loss [3,5]. Many researchers accept that beside the material loss, particle erosion is in analogy with repeated impact. In the case of erosion, the impact process is caused by many fast moving small particles whereas low-energy repeated impact (also called impact fatigue) is usually generated by a large mass of low velocity [6]. Therefore, it is very valuable to study the erosion resistance of advanced composites, to find methods to improve their resistance, to describe their property degradation and damage growth characteristics and finally to model their residual mechanical properties [6]. Unlike traditional materials such as metals and ceramics, polymer composites exhibit unique damage characteristics. When they are subjected to low-energy impacts by particles, there might be no damage indication on surfaces by visual evaluation but internal damage may have already occurred. This damage can have an adverse effect on material performances and structural integrity [7]. There are many studies in the literature investigates the post-impact properties of the polymer composites. Tai et al. reported that low-energy impact loading affects the fatigue behaviour of the carbon/epoxy composites [8]. In other study, impact damages appears at near the bottom
of the laminate consist of fibre breaks, intralaminar cracks, as well as interlaminar delaminations [9]. In recent years experimental studies mainly focussed on the effects of impact loads on damage morphology and residual mechanical properties of polymer composites [10–17]. If solid particles impinge against a target surface and cause local damage combined with material removal should be evaluated as a design parameter if the worn structure also subjected to loading. Therefore prediction of residual mechanical properties after particle erosion is very important because the structure should remain in satisfactory structural integrity during and after particle erosion [6]. A literature survey showed that detailed investigation on solid particle erosion behaviour of polyphenylenesulphide composites has not been reported. Also it is evident that, PPS has immense potential for structural applications and it has not been exploited. Hence, comprehensive and systematic study of erosion behaviour of PPS and its composites is required. The objectives of this study are to investigate the solid particle erosion characteristics of C-PPS composites and to investigate the low-energy impact damages of the particles and residual mechanical properties of the composite materials after particle erosion. 2. Experimental 2.1. Materials Cross-ply [0/90]3s carbon fibre reinforced polyphenylenesulphide (PPS) composites used in this study were kindly supplied by Cetex NijverdalHOLLAND as 400 · 400 mm laminated plaques with the total thickness of 2 mm. The volumetric carbon fibre proportion was 51%. Per ply thickness of the laminates was 0.31 mm and areal weight of the composites was 479 g/m2. The commercial name of the material is CF0286. Test samples of 40 mm · 40 mm · 2 mm in dimensions were cut using a diamond cutter from these moulded plaques. Before the erosive wear tests all specimens were cleaned with acetone, balanced at electronic balance with the accuracy of 0.1 mg at room temperature. Great care was taken to ensure clean surface before and after wear tests. Sand and dust particles were cleaned after erosion test with air blasting, and weighed carefully.
2.2. Test procedures The room temperature erosion test facility used in the present investigation. Angular silica sand particles with the size of 150–200 lm are driven by a static pressure of 1.5, 3 and 4.5 bar accelerated along a 50 mm long ceramic nozzle of 5 mm diameter. The average velocity of the silica sand at these pressures at the nozzle tip was measured as 20, 40 and 60 m/s respect to the air pressure. Composite samples clamped on to the specimen holder 40 mm away from the nozzle (Fig. 1). The samples on specimen holder were subjected to a particle flow at a given impingement angles between 15 and 90. Erodent mass flow was measured 4.25, 6.25 and 9 g/s for 20, 40 and 60 m/s, respectively. Wear rates were measured by means of weight loss at electronic balance with the accuracy of 0.1 mg after 2–30 s of erosion. The residual flexural strength (RFS) of the samples were characterized by means of three point bending tests according to ISO 178. The samples with the dimensions of 10 · 40 · 2 (in mm) were tested in a Universal testing apparatus (Instron 4411) using a cross head velocity of 10 mm/min and span length of 32 mm.
T. Sinmazc¸elik et al. / Materials and Design 29 (2008) 1419–1426
1421
Fig. 1. A schematic illustration of particle erosion test rig.
2.3. Characterization of the eroded samples The eroded surface was inspected in a Jeol JSM-6335F field emission scanning electron microscope (SEM). The samples were gold-sputtered in order to reduce charging of the surface.
3. Results and discussion Steady state wear rates were reached during the erosion periods for each impingement angles. No incubation period was observed at any impingement angles and particle velocities. Also it was observed that the mass loss of the eroded samples were proportional with the erosion time at each impingement angle (15, 30, 45, 60 and 90) and particle speed (20, 40 and 60 m/s). One of the example of these investigations is illustrated in Fig. 2. It is shown that the mass loss is linearly increased as a function of erosion time (between 2 and 30 s) for each different particle speed at impingement angle of 90.
As it is well known, the impingement angle has a great influence on the particle erosion. In order to monitoring the effect of impingement angles, experimental results are illustrated in Fig. 3. The mass losses of C-PPS composites are illustrated as a function of impingement angles at different particle speeds after erosion time of 30 s. The maximum wear rates were observed around the impingement angle of 45 for each particle velocity. This result was the evidence of the semi-ductile nature of C-PPS composite materials similar to other reinforced polymers [5,18,19]. Un-reinforced thermoplastic polymeric PPS matrix with a ductile nature expected to show a maximum erosion around 15– 30 impingement angle. However, due to the carbon fibre content (having a brittle nature) the maximum erosion rate shifted to larger angles and realized around 45. Fig. 4 illustrates the fibre orientations of composite material with respect to the particle flow direction. The angle of impingement, a, varied between the 15 and 90 during the investigations. Especially for the acute impingement angles such as 15 and 30 the orientation of the first
0.18
v= 60 m/s v= 40 m/s v= 20 m/s
0.16 0.14
0.30
v= 60 m/s v= 40 m/s v= 20 m/s
0.25
0.10
Mass loss (gr)
Mass loss (gr)
0.12
0.08 0.06
0.20
0.15
0.10
0.04 0.02
0.05
0.00 0.00 0
5
10
15
20
25
30
Time of erosion (s) Fig. 2. The mass loss of PPS composite as a function of erosion time at different particle speeds at impingement angle of 90.
15
30
45
60
75
90
Impingement angle (degree)
Fig. 3. The mass loss of PPS composite as a function of different impingement angles at different particle speeds after erosion time of 30 s.
1422
T. Sinmazc¸elik et al. / Materials and Design 29 (2008) 1419–1426
Fig. 4. Schematic illustration of particle flow direction respect to the fibre directions of cross-ply composites.
layer has a great role. As reported in literature [1], relationship between the fibre directions and particle flow direction has a great importance. It is reported that the fibres in composites, subjected to particle flow, break in bending. If the particle flow direction parallel with respect to the fibre orientation, bending requires particle indentation into the composite. This indentation induces compressive stresses. At this state, fibre shows a very high resistance to micro bending. This phenomena results in local removal of fibre and resin material from the impacted surface. On the other hand, it is reported that, for transverse particle impact, the particle exerts a lateral loading for the fibres. Because of the lower resistance to bending moment, bundles of fibres get bent and brake easily. This causes high erosive wear at 90 orientation. Also, in case of transverse erosion, high interfacial tensile stresses are generated by particle impacts. This causes intensive debonding and breakage of the fibres, which are not sufficiently supported by the matrix. The continuous impact of particles on the fibres breaks them with a formation of cracks perpendicular to their length. The bending of fibres is possible because the surrounding matrix and supporting fibres have been removed [1]. In the present case (as illustrated in Fig. 4) the erosive particles at first step hit the 1st layer, which has a parallel fibres to the particle flow direction. The materials show a higher resistance to the fibre breaking compared to transverse loading. After 1st layer is completely removed the fibres and matrix material belongs to 2nd layer will subjected to particle erosion in transverse direction. In 2nd layer brittle carbon fibres fracture easily from the out of plane impact forces and therefore they show least resistance to erosion and also they do not absorb much impact energy in the fracture process. However, efforts were previously focussed on the study of the influence of the material rather than the operating parameters, such as angle of impingement of erodent, particle velocity, etc. Angle of impingement is the most important and widely studied parameter in the case of materials in the literature [18]. The SEM image of the eroded surface at impingement angle of 15 is seen in Fig. 5. The erosive particles hit the 1st layer with the particle speed of 20 m/ s. The fibres in 1st layer, are oriented in parallel direction with respect to the particle flow (illustrated by white
Fig. 5. SEM image of the eroded sample at impingement angle of 15 after erosion time of 30 s (v = 20 m/s).
arrow). Because of the acute impingement angle, the particles exert parallel forces (Fp) for the fibres on the material surface, which result in dominantly abrasive wear. The vertical constituent (Fv) is very small compared to the other constituent that result in a very small impact phenomenon [5]. As expected, due to the parallel hits of the erodent at low angle, not much fibre cracking is occurred. Because of the small impact component of the erodent, the fibres resist to cracking. As seen in Fig. 5 there is no cracks on the fibres (vertical to their length). On the other hand there is a matrix removal between the fibres, which weakens the fibre supporting mechanism. Insufficiently supported fibres are broken and may move from their original positions. But at the 1st layer, many of the fibres keep their original direction and continue to load bearing facilities. As the impact angle of the particles shifts towards larger angles, the effects of Fv become marginal. Fig. 6 illustrates the wear morphology of the material after normal erosion (at 90) at low particle speed of 20 m/s. At this impingement angle there is no parallel constituent of particle speed and force (Fp). This why, no abrasive wear behaviour is expected. Each erodent particle has a totally vertical speed and exerts totally ‘‘low-energy repeated impact’’ loading for the material surface. Kinetic energies of the particles are dissipated by the impact and micro-cracking [3,5]. Although the erodent particle has a same small impingement speed, compared to Fig. 5, there was an extensive fibre cracking occurred after normal erosion (Fig. 6). Many fibres were cracked into small fragments and they were removed from their places partly with the surrounding matrix like spalled fragments. When comparing two surface morphologies, it is clear that the wear morphology, in Fig. 6, is quite different compared to Fig. 5. On the other hand, the effects of the particle impacts on the subsurface of the composites should be noted. At acute angle, due to the smaller vertical constitu-
T. Sinmazc¸elik et al. / Materials and Design 29 (2008) 1419–1426
Fig. 6. SEM image of the eroded sample at impingement angle of 90 after erosion time of 30 s (v = 20 m/s).
ent of the particle impingement, the effect of impact phenomena is smaller. Greater part of the particle’s energy spent for removing the matrix between the fibres, and fracture them in limited extend. The fibres in 1st layer which were hit by particles at 15 strongly resist to the particles and limited subsurface deformations that described in literature [7–17] are expected. On the other hand, the lowenergy repeated impact loading of the particles is dominant in Fig. 6 compared to Fig. 5. Beside the remarkable difference in wear morphology at first layer, due to the maximum kinetic energy transfer by the particles at the impingement angle of 90, there was a possible damage on the composite such as fibre cracks, matrix micro cracks, delaminations and also residual stresses may have occurred in the material [7–17]. The effect of particle speed is seen by comparing the wear morphologies of Figs. 5 and 7. Silica particles, with a 60 m/s speed hit the target material approximately three times faster than the silica particles with 20 m/s speed. Higher speed means higher kinetic energy. In this case, silica particles with 60 m/s gives approximately nine timers higher kinetic energy than particles with 20 m/s according to the kinetic energy equation of Ek ¼ 12 m v2 . If remembered, the fibres are more sensitive to impact loadings rather than parallel loadings to their original directions; it is possible to understand the main difference between the morphology of Figs. 7 and 5. Due to higher kinetic energy of the particles, which are impinging to the fibres vertically, more fibres cracked after micro bending and particle indentation [1]. On the other hand, due to the parallel constituents of the particle flow, remarkable matrix and small fibre fragments removals are observed. After erosion, most of the fibres kept their orientations after they are fractured. The continuous impact of silica particles on the composite surface resulted in multiple fracture of matrix and
1423
Fig. 7. SEM image of the eroded sample at impingement angle of 15 after erosion time of 30 s (v = 60 m/s).
resulting in micro-cracking of the matrix. However, exposure of fibres is not apparent. Small platelets or flakes, which are detached from the surface due to continuous impact, are still adhered to the surface. The effects of high speed and high impingement angle of the particles (90) are observed in Fig. 8. It is quite clear from the micrographs that roughening of the surface during erosion around the wear scar is reflected. A typical erosion mark can be clearly seen. It is obvious that in the case of normal erosion all available energy is dissipated by impact and micro-cracking. As a result of the particle impingements fibres are remarkably fractured into the small fragments. Also remarkable matrix deformations and removals are seen in Fig. 8. The fractured small fibre fragments randomly distributed in the roughly deformed matrix and loss their original directions. Remarkable high
Fig. 8. SEM image of the eroded sample at impingement angle of 90 after erosion time of 30 s (v = 60 m/s).
1424
T. Sinmazc¸elik et al. / Materials and Design 29 (2008) 1419–1426
fibre/matrix interfacial deformations are occurred. Particles transferred all their kinetic energies to the material surface. This energy is spent for fibre cracking, fibre/matrix interfacial deformations, matrix deformation, matrix removals, etc. As stated previously, the main aim of this study is to investigate the residual mechanical properties of composites after particle erosion. The sample preparation procedure for residual flexural properties measurement is illustrated in Fig. 9. The square shaped composite samples (coupons) in 40 · 40 mm dimension were used in particle erosion investigations. (Fig. 9a). The erodent flow is focussed at the centre of the coupons. After erosion at obtuse angles, circular wear scars with the diameter of approximately 10 mm were observed in (Fig. 9b). After particle erosion, composite samples were cut into the rectangular samples as illustrated in Fig. 9c. At normal erosion (90) it is possible to achieve circular wear scar. On the other hand, at lower impingement angles, the elliptical wear scars were occurred (Fig. 9d). The eroded samples were placed between the supports (the span length was 32 mm) in order to measure the residual flexural strength (Fig. 9e). As indicated in Fig. 9e eroded surfaces forced to tensile stresses during the bending. The load bearing mechanism during the bending is illustrated regarding to fibre orientations in Fig. 10. The original sample thickness, b, became to ‘‘b–h’’ (Fig. 9e) and the natural axis change its original position and shifted to upwards after particle erosion. Compared to the original sample (Fig. 10a) thinner sample is subjected
to bending after erosion which means decreased number of fibres subjected to tensile and compressive loadings during the flexural loading. The sample thicknesses of (b–h) after particle erosion are measured with micrometers and flexure strengths of the samples were calculated regarding to this residual sample thicknesses. At original sample, the maximum compressive and tensile stresses during the flexure tests occurred at point ‘‘A’’ and ‘‘B’’, respectively. On the other hand, at eroded samples, maximum tensile stress occurred at point ‘‘C’’ during the bending (Fig. 10b). Point C has a great importance. As investigated in the literature [1–17] as a result of particle impingements, there may many broken fibres, matrix cracks, fibre/matrix interfacial deformations and subsurface deformations like intralaminar cracks, interlaminar delaminations occurred at and around point ‘‘C’’. As already known, the carbon fibres bear the high percentage of flexural load. Also it should be noted that these fibres has a brittle nature and tends to be broken easily as a result of particle impingement especially at oblique angles (maximum at 90). Therefore the effect of impact angle of the particles has a great importance. Relationship of the mass loss and residual flexural strength of the material regarding to impingement angles can be evaluated from Figs. 3 and 11. As presented in Fig. 3, maximum mass losses were observed at impingement angle of 45 for each impingement speed. Therefore, not surprisingly minimum residual flexural properties were achieved from the samples eroded at impingement angle of 45 (Fig. 11).
Fig. 9. Sample preparation procedure of residual flexural strength (RFS) testing: after particle erosion.
Fig. 10. Schematic illustration of RFS measurement procedure of the samples (a – the original sample, b – eroded sample).
T. Sinmazc¸elik et al. / Materials and Design 29 (2008) 1419–1426
800
Flexure strength (MPa)
700 600
20 m/s
500 400 300
40 m/s 200
60 m/s
100 0 10
20
30
40
50
60
70
80
90
Angle of impingement (o)
Fig. 11. Residual flexure strength values of eroded samples.
The impingement angle of 45 can be evaluated as an ‘‘optimum angle’’. It is possible to see the vertical impact effects of the particles and also as a result of parallel constituent of the particle there is remarkable material erosion on the surface. Not surprisingly the lower mass losses give higher residual flexural strength (RFS) values. Although the wear depth of ‘‘h’’ is very small, when we evaluate the sample thickness of the eroded samples as (b–h) it is possible to calculate the flexure strength of the material independent from the variation in thickness. If we used an ‘‘ideal’’ material which has an isotropic microstructure and there are no plastic and residual deformations and stresses of the particle erosion, it is expected to get the very close values between the erosion rates and RFS ratios. But if we used polymer composites with a highly anisotropic microstructure it is possible to get different results. Also when we remember the previous studies [7–17] due to the effects of low impact energy loadings of the particles there is possible subsurface deformations, delaminations, intralaminar cracks formations may occur. When we measure the thickness of the sample eroded at 90 it should be noted that the cross section of the sample has a partially deformed (especially nearby the eroded surface) structures as a result of particle hits. Not surprisingly the effect of particle impingements on the material is remarkably higher at high particle speeds. Great surface and subsurface deformations as investigated in literature [7–17] are expected for particle speed of 60 m/s. When we focussed back the results of erosive wear rates in Fig. 3 for the particle speed of 60 m/s the mass loss at impingement angle of 90 was approximately 0.163 g. Also the mass loss at impingement angle of 15 was approximately 0.09 g. That means the mass loss after erosion at 90 1.8 times than that of 15. On the other hand residual flexural strength (RFS) of the sample, which was eroded at impingement angle of 90 and particle speed of 60 m/s was measured as 170 MPa (Fig. 11). At the same particle speed,
RFS was measured as 650 MPa as seen in Fig. 11 for the sample, which is eroded at 15. The ratio between these RFS is approximately 3.8. The difference between these two ratios; (i) the ratio of mass losses between 90 and 15, (ii) the ratio of residual flexural strength between 15 and 90 gives an important feedback upon the composite deformations (included the subsurface deformations) after the particle erosion regarding to impingement angles. As a result, samples were eroded 1.8 times higher at 90 than that of 15. On the other hand the RFS at 90 3.8 times lower than that of 15. When we evaluated the wear morphologies of the samples (Figs. 5–8), mass loss values (Fig. 3) and RFS values (Fig. 11). It is possible to conclude the impingement angle effects on residual flexural strength. Not surprisingly it is expected lower flexural strength from the samples, which has a greater mass loss after particle erosion at 90 compared to 15 (Fig. 3). On the other hand because of brittle nature, fibres have a great tendency to fracture at the angles which are close to 90. As discussed in previous sections of this manuscript, the low-energy impact fatigue loadings of the particles and their subsurface deformations has a great importance at obtuse angles (maximum at 90) which is responsible for remarkable decrease in residual flexural properties of the eroded samples compared to acute angles. In order to observe the impact effect of the particles on RFS the results are re-illustrated in Fig. 12. Not surprisingly the impact effect of the particles is minimum at 15, and maximum at 90. Fig. 12 illustrates the mass loss ratios between the 90 and 15 with grey columns and residual flexural strength ratios between 15 and 90 with white columns for the each particle speed of 20, 40 and 60 m/s. Mass loss ratios between 90 and 15 are realized around 1.75 for each particle speed as seen in Fig. 12. For the particle speed of 20 m/s mass loss ratio and RFS ratio approximately has a same value around the 1.8. That is the
4.0
Mass loss ratio (90° and 15°)
3.5
Residual flexure strength ratio (15° and 90°)
3.0 2.5
Ratio
900
1425
2.0 1.5 1.0 0.5 0.0 10
20
30
40
50
60
70
Particle speed (m/s)
Fig. 12. Mass loss and RFS ratios of the eroded materials at each particle speed.
1426
T. Sinmazc¸elik et al. / Materials and Design 29 (2008) 1419–1426
evidence of subsurface deformations after particle erosion. Because at 20 m/s, particle has a lower kinetic energy which results in limited subsurface deformations after impacts of the erodent. On the other hand, as stated previously at particle speed of 60 m/s, there is a quite remarkable difference in mass loss ratio and RFS ratio, 1.8 and 3.8, respectively. The increment at RFS ratios of 20,40 and 60 m/s is not linear. But that is supported our theory because the kinetic energy of the particles at 40 m/s 4 times higher compared to 20 m/s and the kinetic energy of the particles at 60 m/s nine times higher than 20 m/s. This why, hyperbolical increment is observed in RFS ratios, which shows the close relationships between the changes in kinetic energy of the particles. 4. Conclusions The C-PPS composites has a semi-brittle behaviour during particle erosion. Cross-ply C-PPS composite has a maximum wear rate at 45 impingement angle. Particle erosion rate of the material has a close relationship between particle speeds, and angle of impingements. The impingement angle of the particles should be taken into account as a design criteria for the polymer composites, which will subjected to particle erosion. The maximum wear rate was observed at 45, which also has a minimum residual flexural strength. On the other hand the impingement angles of 15 and 90 has a lower wear rates, compared to 45 and has a potential application capabilities. But the impingement angle at 90 has great material deformations which result in lower residual flexural properties. As a conclusion, the solid particle erosion should be thinking as a repeated low-energy impact procedure. The damage growth during particle erosion is therefore similar to impact fatigue loading. In this study also flexural testing was carried out to evaluate post-erosion properties, namely flexural strength. It was anticipated that a correlation between the particle erosion parameters, the damage magnitudes and residual flexural properties would be established. Residual flexural property of the composites after particle erosion shows that flexural strength is more sensitive to the presence of localised impact damages at higher impingement angles and higher particle velocities. The residual flexural properties after erosion at oblique angles and higher impingement speeds are smaller compared to acute angles and lower velocities. That means not only the wear rates affect the residual mechanical properties, but also the effect of fibre fracture tendency, and also (may be the most important factor) the subsurface defor-
mations; matrix cracks and deformations due to the impingement of the particles affect the results. References [1] Tewari US, Harsha AP, Hager AM, Friedrich K. Solid particle erosion of carbon fibre– and glass fibre–epoxy composites. Compos Sci Technol 2003;63:549–57. [2] Tewari US, Harsha AP, Ha¨ger AM, Friedrich K. Solid particle erosion of unidirectional carbon fibre reinforced polyetheretherketone composites. Wear 2002;252:992–1000. [3] Bhushan B. Principles and applications of tribology. New York: Wiley; 1999. [4] Rajesh JJ, Bijwe J, Tewari US, Venkataraman B. Erosive wear behavior of various polyamides. Wear 2001;249:702–14. [5] Harsha AP, Tewari US, Venkatraman B. Solid particle erosion behaviour of various polyaryletherketone composites. Wear 2003;254:693–712. [6] Barkoula NM, Papanicolaou GC, Karger-Kocsis J. Prediction of the residual tensile strengths of carbon-fibre/epoxy laminates with and without interleaves after solid particle erosion. Compos Sci Technol 2002;62:121–30. [7] Zhang ZY, Richardson MOW. Low velocity impact induced damage evaluation and its effect on the residual flexural properties of pultruded GRP composites. Compos Struct 2007;81:195–201. [8] Tai NH, Yip MC, Lin JL. Effects of low energy impact on the fatigue behaviour of carbon/epoxy composites. Compos Sci Technol 1998;58:1–8. [9] Angelidis N, Irving PE. Detection of impact damage in CFRP laminates by means of electrical potential techniques. Compos Sci Technol 2007;67:594–604. [10] Dear JP, Brown SA. Impact damage processes in reinforced polymeric materials. Compos Part A 2003;34:411–20. [11] Saez SS, Barbero E, Zaera R, Navarro C. Compression after impact of thin composite laminates. Compos Sci Technol 2005;65:1911–9. [12] Dhakal HN, Zhang ZY, Richardson MOW, Errajhi OAZ. The low velocity impact response of non-woven hemp fibre reinforced unsaturated polyester composites. Compos Struct 2007;81:559–67. [13] Johnson AF, Holzapfel M. Influence of delamination on impact damage in composite structures. Compos Sci Technol 2006;66:807–15. [14] De Morais WA, Monteiro SN, D’Almeida JRM. Effect of the laminate thickness on the composite strength to repeated low energy impacts. Compos Struct 2005;70:223–8. [15] Yuanjian T, Isaac DH. Combined impact and fatigue of glass fiber reinforced composites’’, Compos Part B, in press. [16] De Morais WA, Monteiro SN, D’Almeida JRM. Evaluation of repeated low energy impact damage in carbon–epoxy composite materials. Compos Struct 2005;67:307–15. [17] Wang S, Chung DDL, Chung JH. Impact damage of carbon fiber polymer–matrix composites, studied by electrical resistance measurement. Compos Part A 2005;36:1707–15. [18] Harsha AP, Thakre AA. Investigation on solid particle erosion behaviour of polyetherimide and its composites. Wear 2007;262:807–18. [19] Rattan R, Bijwe J. Influence of impingement angle on solid particle erosion of carbon fabric reinforced polyetherimide composite. Wear 2007;262:568–74.
