Fully Reversed Axial Notch Fatigue Behaviour of Virgin and Recycled Polypropylene Compounds

Available Available online online at at www.sciencedirect.com www.sciencedirect.com

ScienceDirect

Available online www.sciencedirect.com Available online at at www.sciencedirect.com Structural Structural Integrity Integrity Procedia Procedia 00 00 (2016) (2016) 000–000 000–000

ScienceDirect ScienceDirect

Procedia Structural 2 (2016) 2255–2262 Structural IntegrityIntegrity Procedia 00 (2016) 000–000

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

www.elsevier.com/locate/procedia

21st European Conference on Fracture, ECF21, 20-24 June 2016, Catania, Italy

Fully Reversed Axial Notch Fatigue Behaviour of Virgin and Recycled Polypropylene Compounds Thermo-mechanical modeling of aaa, M. high pressure turbine blade of an aa bb aa G. Meneghetti *, M. Ricotta Sanità , D. Refosco airplane gas turbine engine Department Department of of Industrial Industrial Engineering, Engineering, University University of of Padova, Padova, via via Venezia Venezia 1, 1, 35131 35131 Padova Padova (Italy) (Italy)

XV Portuguese Conference on Fracture, PCF 2016, 10-12 February 2016, Paço de Arcos, Portugal

11

2 2

Electrolux Electrolux spa, spa, C.so C.so L. L. Zanussi Zanussi 30, 30, 33080 33080 Porcia Porcia (Pordenone), (Pordenone), Italy Italy

P. Brandãoa, V. Infanteb, A.M. Deusc*

a

Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal IDMEC, Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal In paper, the behaviour of polypropylene compounds, characterized by different fractions of c In this this paper,Department the fatigue fatigue behaviour Engineering, of different differentInstituto polypropylene compounds, characterized byAv. different fractions of recycled recycled CeFEMA, of Mechanical Superior Técnico, Universidade de Lisboa, Rovisco Pais, 1, 1049-001 Lisboa, material, were analysed. Fully reversed fatigue tests were carried out material, were analysed. Fully reversed fatigue tests were carried out on on three three different different polypropylene polypropylene (PP) (PP) compounds, compounds, namely namely aa Portugal

Abstract Abstract b

42 42 wt% wt% calcium calcium carbonate carbonate filled filled PP PP (EA209), (EA209), aa 42 42 wt% wt% calcium calcium carbonate carbonate filled filled polypropylene polypropylene containing containing 25% 25% recycled recycled PP PP (R2025) (R2025) and and aa 42 42 wt% wt% calcium calcium carbonate carbonate filled filled 100% 100% recycled recycled polypropylene polypropylene (R2100). (R2100). Both Both plain plain and and notched notched samples samples were were tested. In Abstract tested. In particular, particular, the the notch notch sensitivity sensitivity was was investigated investigated on on double-edge double-edge notched notched specimens specimens machined machined from from 5-mm-thick 5-mm-thick injected =1.65), aa 22 injected moulded moulded plates. plates. Three Three different different notch notch geometries geometries were were analysed, analysed, namely namely aa 10 10 mm mm circular circular notch notch radius radius (K (Ktt=1.65), =3.17) and a 0.5 mm V-notch radius (K =5.97). During the experimental tests, the fatigue damage mm U-notch radius (K During their operation, modern aircraft engine components are subjected to increasingly demanding operating t t mm U-notch radius (Kt=3.17) and a 0.5 mm V-notch radius (Kt=5.97). During the experimental tests, the fatigue conditions, damage especially high pressure turbine blades. Such conditions cause these partsfracture to undergo different of time-dependent evolution was by on board microscope and, after failure, surfaces were analysed as evolution wasthemonitored monitored by using using on (HPT) board travelling travelling microscope and, after failure, fracture surfaces weretypes analysed as well. well. In In degradation, one of which creep. A model the finite element method (FEM) was are developed, in order toThe be able to predict view of body it concluded that analysed PP notch presence of view of this this extensive extensive body of ofisevidence, evidence, it was was using concluded that the the analysed PP compounds compounds are notch insensitive. insensitive. The presence of therecycled creep behaviour HPT blades. Flight behaviour data records a specific aircraft, by aConsequently, commercial aviation 25% PP the with respect to compound made of in 25% recycled PP slightly slightlyofinfluenced influenced the fatigue fatigue behaviour with(FDR) respectfor to the the compound madeprovided of virgin virgin PP. PP. Consequently, in the the company, useddesign-stress-life to obtain thermalcurve and was mechanical threeand different cycles. In order to createcharacterised the 3D model present paper, aa single proposed for EA209 R2025 plain notched compounds, present paper,were single design-stress-life curve was proposeddata for for EA209 and R2025 flight plain and and notched compounds, characterised needed for the FEM analysis,13a and HPT blade scrap was scanned, and its chemical composition and A,50% material properties were by equal to to 11 11 MPa. MPa. by an an inverse inverse slope, slope, k, k, equal equal to to 13 and aa reference reference net net stress stress amplitude amplitude evaluated evaluated at at 22 million million cycles, cycles,  A,50%,, equal obtained. The data that was gathered was fed into the FEM model and different simulations were run, first with a simplified 3D Conversely, Conversely, aa down-graded down-graded stress-life stress-life design design curve curve was was determined determined for for R2100 R2100 compound, compound, having having k=16 k=16 and and  A,50% =8 MPa. MPa. A,50%=8 rectangular block damage shape, inanalysis order tohighlighted better establish the model, and then with the realevolution 3D meshwere obtained from the on blade scrap. The Finally, the fatigue that damage mechanisms and their independent the type of Finally, fatigue behaviour damage analysis that damage mechanisms and their evolution the typesuch of a overallthe expected in termshighlighted of displacement was observed, in particular at the trailing were edge independent of the blade. on Therefore material and notch radius and consisted of void formation and coalescence. material and notch radius and consisted of void formation and coalescence. model can be useful in the goal of predicting turbine blade life, given a set of FDR data. © © 2016 2016 The The Authors. Authors. Published Published by by Elsevier Elsevier B.V. B.V. © 2016, PROSTR (Procedia StructuralofIntegrity) Hosting Committee by Elsevier Ltd. All rights reserved. Peer-review under responsibility the Scientific of ECF21. Peer-review under responsibility of the Scientific Committee © 2016 The Published by Elsevier B.V. Peer-review underAuthors. responsibility of the Scientific Committee of ECF21.of ECF21. Peer-review under responsibility of the Scientific Committee of PCF 2016.

Keywords: Polypropylene, Recycled Recycled polypropylene, polypropylene, fatigue, fatigue, notch notch sensitivity, sensitivity, damage damage evolution evolution Keywords: Polypropylene,

Keywords: High Pressure Turbine Blade; Creep; Finite Element Method; 3D Model; Simulation.

* * Corresponding Corresponding author. author. Tel.: Tel.: +0039-049-827-6751; +0039-049-827-6751; fax: fax: +0039-827-6785. +0039-827-6785. E-mail E-mail address: address: [email protected] [email protected] 2452-3216 2452-3216 © © 2016 2016 The The Authors. Authors. Published Published by by Elsevier Elsevier B.V. B.V.

Peer-review under responsibility the * Corresponding Tel.: +351of Peer-review underauthor. responsibility of218419991. the Scientific Scientific Committee Committee of of ECF21. ECF21. E-mail address: [email protected]

2452-3216 © 2016 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the Scientific Committee of PCF 2016. 2452-3216 © 2016, PROSTR (Procedia Structural Integrity) Hosting by Elsevier Ltd. All rights reserved. Peer-review under responsibility of the Scientific Committee of ECF21. 10.1016/j.prostr.2016.06.282

2256 2

G. Meneghetti et al. / Procedia Structural Integrity 2 (2016) 2255–2262 Author name / Structural Integrity Procedia 00 (2016) 000–000

1. Introduction The use of calcium carbonate (CaCO3) filled Polypropylene (PP) is very common in structural applications due to its low manufacturing cost, capability to be molded in complex geometries, high production rate and significantly low weight to strength ratio. Recently cost reduction was achieved either by introducing in the material a fraction of recycled PP or using totally recycled PP. While for conventional calcium carbonate filled PP some papers are available in literature dealing with its fatigue behavior and damage mechanisms (Allard et al. 1989, Zhou and Mallick 2005, Kultural and Eryurek 2007, Meneghetti et al. 2014), in authors’ knowledge, lack of information concerning the fatigue behavior of filled recycled PP exists and particularly dealing with fatigue notch sensitivity. Recently, Meneghetti et al (2015) have investigated the static and fatigue behaviour and the damage evolution of different polypropylene compounds, characterised by different fractions of recycled material. Fatigue notch sensitivity was analysed as well. It was found that all tested materials are notch insensitive under static loading and in the high cycles fatigue regime. Concerning the static damage mechanisms and their evolution, they were found to be independent on the type of material and notch radius and consisted of void formation and coalescence. As far as the sharpest notch made of the fully recycled material is considered, the same mechanisms were observed in fatigue behaviour. In the present paper, the fatigue behaviour of 42 wt% calcium carbonate filled Polypropylene (PP), a 42 wt% calcium carbonate filled PP containing 25% recycled PP and a 42 wt% calcium carbonate filled 100% recycled PP was investigated. Both plain and notched samples were tested. In particular, the notch sensitivity was investigated on double-edge notched specimens machined from 5-mm-thick injected moulded plates. Three different notch geometries were analysed, namely a 10 mm circular notch radius (Kt=1.65, referred to the net section), a 2 mm U-notch radius (Kt=3.17, referred to the net section) and a 0.5 mm V-notch radius (Kt=5.97, referred to the net section). During the fatigue tests, fatigue damage was analysed by stopping the fatigue test at a fixed number of cycles and monitoring the damage evolution by using a travelling microscope. It was found that all tested materials are notch insensitive from extremely-low to high –cycle fatigue regime and that the presence of 25% of recycled PP do not influence the fatigue material response with respect to the virgin PP. Therefore, a single fatigue design stresslife curve was proposed for EA209 and R2025 materials. On the contrary, a down-graded stress-life design curve was determined for R2100 compound. 2. Materials, specimens’ geometry and test methods The static and the fully reversed fatigue behaviour of three different material systems were analysed, namely a 42 wt% calcium carbonate filled Polypropylene (PP), here defined as EA209, a 42 wt% calcium carbonate filled PP containing 25% recycled PP (R2025) and a 42 wt% calcium carbonate filled 100% recycled PP (R2100). To evaluate the static and fatigue notch sensitivities, double-edge notched specimens were machined from 5.2-mmthick injected moulded plates, according to specimens’ geometry shown in Fig. 1, where the geometry of plain material is also shown. In this paper, notched specimens will be referred as R10, R2 and V05 for the geometries shown in Fig. 1c, Fig. 1d and Fig. 1e, respectively. To evaluate the stress concentration factor referred to the net section, 3D linear elastic finite element analyses were carried out by using 8-node solid elements (SOLID185 of commercial code ANSYS® 15) and it was found Kt=1.65, 3.17 and 5.97 for 10 mm circular notch radius, 2 mm Unotch radius and 0.5 mm V-notch radius, respectively. For each material configuration, three static tests were carried-out on plain specimens at room temperature (RT) by imposing a displacement rate equal to 1 mm/min, according to ISO 527 standard (ISO 527, 1996). Concerning tensile tests on notched samples, the applied displacement rate was reduced to maintain the linear elastic stress- rate at the notch tip in a ± 10% range with respect to the relevant plain material. The fatigue tests were conducted by imposing a sinusoidal wave form characterised by a nominal stress ratio R (defined as the ratio between the minimum and the maximum stress) equal to -1. To maintain the specimen’s temperature in the range from 20 to 32° C, test frequencies between 1 and 25 Hz were adopted, depending on the applied stress level. Surface temperature of materials was monitored by fixing 0.127 mm diameter copper - constantan thermocouples at the notch tip, using a silver-loaded conductive epoxy glue. Temperature signals generated by the thermocouples were acquired by means



G. Meneghetti et al. / Procedia Structural Integrity 2 (2016) 2255–2262 Author name / Structural Integrity Procedia 00 (2016) 000–000

2257 3

of a data logger National Instruments “Hi-speed NI USB-9162” USB carrier, operating at a maximum sample frequency of 5 Hz. All experimental tests were carried out using a MTS 858 Mini Bionix II servo-hydraulic test machine equipped with a 15 kN load cell and a 25-mm-gauge-length 634.12F-24 MTS extensometer. Finally, fatigue damage evolution was monitored by means of AM4113ZT Dino-Lite digital travelling microscope. 100 30

15

150

R60

13

= R10

100 30

25

20

(a)

=

32

20

4

20

(c)

15 150

150

R0.5

32

32

20

R2

=

=

=

=

22

(b)

(d) 47°

(e)

Fig. 1. Specimens’ geometry adopted for a) static and b) fatigue tests on plain specimens and (c-e) static and fatigue tests on notched specimens (thickness 5.2 mm).

Table 1. Static mechanical properties of tested plain materials. Material

E (MPa)

y (MPa)

b (MPa)

b

EA209

2967

19.0

/

/

R2025

2881

18.1

4.4

0.52

R2100

2382

15.4

5.5

>0.50

3. Static test results For each material and specimen’s geometry, three static tests were carried out and the mean value of the elastic modulus E, the yield stress y, the tensile stress at break b and the tensile strain at break b are listed in Table 1 and Table 2 for plain and notched material, respectively. Details of static curves can be found in Meneghetti et al (2015). Here we noticed that the case of EA209 material, the specimens’ separation has never been reached; in fact, due to their high ductility, the available stroke of the test machine actuator (100 mm) was not sufficient to separate the specimens. Moreover in the case of R2100 and EA209_R10 specimens, b was higher than the maximum deformation measurable by the adopted extensometer (>0.50). Concerning the notched specimens, stresses are

G. Meneghetti et al. / Procedia Structural Integrity 2 (2016) 2255–2262 Author name / Structural Integrity Procedia 00 (2016) 000–000

2258 4

always referred to the net section, n. By considering Table 1, one can see that the yield stress decreases of about 20% by using fully recycled PP (15.4 MPa) as compared to the original PP compound (19.0 MPa). Table 2 shows that all tested materials resulted practically insensitive to notches in static tests: even by considering the most severe notches having a theoretical stress concentration factor equal to 5.94, the observed strength reduction factor was close to 1. Table 2. Static mechanical properties of tested notched materials (stresses are referred to the net section). Material

Geometry

y (MPa)

b (MPa)

b

EA209

V05

20.0

0

0.3

R2

20.2

0

0.3

R10

20.7

2.4

>0.50

V05

18.6

0

0.3

R2

19.3

0

0.3

R10

20.4

6.2

0.2

V05

14.6

0

0.2

R2

15.5

0

0.2

R10

16.4

0

0.3

R2025

R2100

4. Fatigue test results and fatigue design curves The results of push-pull axial fatigue tests are shown in Fig. 2a, Fig 2b and Fig. 2c for EA209, R2025 and R2100 materials, respectively, in terms of applied net stress amplitude, an. It can be seen that for a given material, plain and notched specimens can be considered as a single population and then they were statistically analyzed under the hypothesis of log-normal distribution of the number of cycles to failure, Nf, and constant scatter with respect to an. The mean and the 10%-90% survival probability curves, fitting the experimental results with a confidence level of 95%, have the following equation. k an  N f  const

(1)

Therefore, as it was pointed out above concerning the static tests, also in the case of fatigue loadings the analysed materials are practically notch insensitive, at least in the fatigue regime investigated in this paper. Moreover, by considering the fatigue curves relevant to EA209 and R2025 materials, one can note that the slope k as well as the reference stress amplitude An,50% evaluated at NA=2 million cycles are very similar and then one can assume to statistically analyse EA209 and R2025 fatigue data as a single population. Fig. 2d shows the results of such statistical analysis, which confirms that the presence of 25% of recycled PP do not influence the fatigue behaviour with respect to virgin PP. Conversely, the relevant fatigue curve of R2100 presents a An,50% value 20% lower than that related to EA209 and R2025 materials and k slope 27% higher. In view of this body of evidence, two distinct design stress-life curves are proposed, one for EA209 and R2025 materials and another for R2100 material. For both curves, in the extremely low cycle fatigue behaviour, the fatigue strength was assumed equal to the material yield stress, obtained averaging the relevant yield values listed in Table 1 and 2. It was calculated y=19.5 MPa and y=15.5 MPa, for EA209 & R2025 and R2100, respectively. The upper limit of the low cycle fatigue region in terms of number of cycle, NS, was defined as the intercept between the yield stress and the stress life curve shown in Fig 2c and d for R2100 and EA209&R2025 material, respectively.



G. Meneghetti et al. / Procedia Structural Integrity 2 (2016) 2255–2262 Author name / Structural Integrity Procedia 00 (2016) 000–000

(b) 30

10

an [MPa]

an [MPa]

(a) 30

5

EA209 k=12.3 V05 An,50%=10.5 MPa R2 T=1.34 R10

1

10

103

102

10

104

105

106

10

 an [MPa]

k=16.4 An,50%=8.38 MPa T=1.27

102

102

103

104

105

106

107

103

104

105

106

EA209 V05 R2 k=12.9 R10 An,50%=10.5 MPa R2025 T=1.29 V05 R2 R10

10

5

107

Nf, number of cycles to failure

1

10

102 103 104 105 106 Nf, number of cycles to failure

Fig. 2. Fatigue data and stress-life curve for a) EA209, b) R2025, c) R2100 and d) EA2090 & R2025 plain and notched specimens

30 EA209 & R2025 An,50%=10.5 MPa k=12.9 T=1.29

y=19.5 MPa y=15.5 MPa

an [MPa]

an [MPa]

10

(d) 30

R2100 V05 R2 R10

1

k=14.5 An,50%=10.6 MPa T=1.27

Nf, number of cycles to failure

(c) 30

10

R2025 V05 R2 R10

5 1

107

Nf, number of cycles to failure

5

2259 5

10

5

1

10

NS=83

NS=670

102

103

R2100 An,50%=8.38 MPa k=16.4 T=1.27

104

105

106

Nf, number of cycles to failure Fig. 3 Fatigue design curves for EA209 & R2025 compounds and for R2100 material

107

107

6

Author name / Structural Integrity Procedia 00 (2016) 000–000

G. Meneghetti et al. / Procedia Structural Integrity 2 (2016) 2255–2262

2260

5. Damage evolution under push-pull fatigue loading During the fatigue tests, the damage evolution was monitored by stopping the fatigue test at a fixed number of cycles and monitoring the specimen from both side of the sample (hereafter referred as “lateral view”). Fig. 4 shows the typical damage mechanics observed in the case of EA209_V05 samples, fatigued in low cycle fatigue regime. It can be observed that damage starts in the early stage of fatigue life (Fig4a, N/Nf= 0.81%) and consists in material whitening, which evolves as visible crazing, at N/Nf =1.47% (see Fig.4c). Then crazing propagated through the thickness (Fig. 7c, N/Nf=3.49%) and at 38.8% of the total fatigue life, some big voids are present (Fig. 4d), which increase in size (Fig.4e) up to the final failure. The same damage mechanisms were observed also in the case of extremely low cycle fatigue, as shown in Fig. 5, which refers to an EA209_V05 specimens fatigued at an=20 MPa, which failed at Nf=11 cycles. For completeness, it is worth noting that the high cycle fatigue regime of EA209 material was analysed by Meneghetti et. (2015) and appeared as material whitening, which expanded through-thethickness direction, followed by void nucleation and coalescence.

whitening

crazing

1 mm

1 mm

a)N=21 cycles (l.v)

b) N=38 cycles (l.v)

1 mm c) N=90 cycles (l.v)

1 mm d) N=1000 cycles (l.v)

1 mm e) N=2500 cycles (l.v)

Fig. 4. Fatigue damage evolution observed at the notch tip from “lateral view” (l.v.) (EA209_V05, an=18 MPa; Nf=2579).

whitening

1 mm a)N=1 cycles (l.v)

1 mm b) N=6 cycles (l.v)

1 mm c) N=8 cycles (l.v)

1 mm d) N=10 cycles (l.v)

1 mm e) N=11 cycles (l.v)

Fig. 5. Fatigue damage evolution observed at the notch tip from “lateral view” (l.v.) (EA209_V05, an=20 MPa; Nf=11).

Fig.6 shows the typical damage evolution observed in the case of R2100_R10 specimens, fatigued in medium and high cycled fatigue life. Damage evolution, started in the early stage of fatigue life as material whitening, evolved, characterized by the presence of light rows and some small voids, as shown in Fig.6a (N/Nf=59.7%). Then these voids grew and coalesced (Fig 6c, N/Nf=99.0%) to form a single void (Fig 6d) that expanded through-the-thickness direction (Fig.6e, N/Nf=99.9%). It is worth noting that, the through-the-thickness crack (Fig. 7c-e) was not visible at the specimen surface analysed with front views, according to Meneghetti et al. (2015). This result was supported by the fracture surface analysis shown in Fig. 6f, where it is seen that the fatigue crack propagated more in the midthickness than near specimen’s surfaces, thus generating a curved crack front, as indicated by the dashed lines. The final external ligament failed when Nf was reached. Finally, Fig. 7 shows the typical damage evolution observed in extremely low cycle fatigue in the case of R2100_V05 specimens. One can see that fatigue damage started in the early stage of fatigue life (Fig. 7a) as material whitening, followed by formation of small voids (see Fig 7b, in the circle). Then the number of voids increased (Fig. 7c) and some of them coalesced (Fig. 7d) and propagated through the thickness up to final failure (Fig7e). Again, it can be observed (see Fig 7f) that crack propagation was not visible from the specimen surface, as mentioned above regarding R2100_R10 specimen.



G. Meneghetti et al. / Procedia Structural Integrity 2 (2016) 2255–2262 Author name / Structural Integrity Procedia 00 (2016) 000–000

1 mm a)N=168446 cycles (l.v)

b) =254038 cycles (l.v)

1 mm c) N=278983 cycles (l.v)

1 mm d) N=280636 cycles (l.v)

2261 7

1 mm e) N=281741 cycles (l.v)

f) fracture surface Fig. 6. Fatigue damage evolution observed at the notch tip from “lateral view” (l.v.) and fracture surface (R2100_R10, an=10 MPa; Nf=281980).

1 mm a)N=1 cycle (l.v)

1 mm b) N=3 cycles (l.v)

1 mm c) N=5 cycles (l.v)

1 mm d) N=6 cycles (l.v)

1 mm e) N=7 cycles (l.v)

e) fracture surface Fig. 7. Fatigue damage evolution observed at the notch tip from “lateral view” (l.v.) and fracture surface (R2100_R10, an=15 MPa; Nf=7 cycles).

6. Conclusions In this paper, the fatigue behaviour and the damage evolution of different polypropylene compounds, characterized by different fractions of recycled material, were analysed. Fully reversed fatigue tests were carried out on three different polypropylene (PP) compounds, namely a 42 wt% calcium carbonate filled PP (EA209), a 42 wt% calcium carbonate filled polypropylene containing 25% recycled PP (R2025) and a 42 wt% calcium carbonate filled 100% recycled polypropylene (R2100). The fatigue notch sensitivity was investigated as well, considering double-edge notched specimens, having 10 mm circular notch radius, 2 mm U-notch radius and 0.5 mm V-notch radius. It was found that all tested materials are notch insensitive from extremely low to high cycle fatigue regime. Moreover, at least from an engineering point of view, it was noted that the presence of 25% recycled PP did not influence the material fatigue strength, compared to that of specimens made of virgin PP. Therefore, a single design stress-life curve was proposed for both materials, based on the net-stress-amplitude. Conversely, a down-graded design stresslife was proposed for R2100 compound. Finally, it was found that damage mechanisms and their evolution were independent on the type of material and notch radius and consisted of void formation and coalescence.

2262 8

G. Meneghetti et al. / Procedia Structural Integrity 2 (2016) 2255–2262 Author name / Structural Integrity Procedia 00 (2016) 000–000

References R.C. Allard, T. Vu-Khanh, J.P. Chalifoux, 1989. Fatigue crack propagation in mica-filled polyolefins, Polym Comp 10, 62-68. Y Zhou, P.K. Mallick, 2005. Fatigue performance of injection molded talc-filled polypropylene, Polym Eng Science 45, 510-516. E. Kultural, I.B. Eryurek, 2007. Fatigue behavior of calcium carbonate filled polypropylene under high frequency loading, Mater Design 28, 816823. Meneghetti, G., Ricotta, M., Lucchetta, G., Carmignato, S., 2014. An hysteresis energy-based synthesis of fully reversed axial fatigue behaviour of different polypropylene composites. Composites: Part B 65, 17-25. Meneghetti, G., Ricotta, M., Sanità, M., Refosco, D., 2015. Notch sensitivity on fully reversed axial fatigue behaviour of different polypropylene compounds. Procedia Engineering 109, 441-449. XXIII Italian Group of Fracture Meeting, IGFXXIII.