- Email: [email protected]
Mechanical Behavior of Polymeric Materials for Mounting Systems of Solar-thermal Collectors
Available online at www.sciencedirect.com
ScienceDirect Materials Today: Proceedings 10 (2019) 476–484
www.materialstoday.com/proceedings
AEM2017
Mechanical Behavior of Polymeric Materials for Mounting Systems of Solar-thermal Collectors Joerg Fischera,*, Patrick R. Bradlera, Sandra Leitnera, Gernot M. Wallnera, Reinhold W. Langa a
Institute of Polymeric Materials and Testing, Johannes Kepler University Linz, Altenbergerstrasse 69, 4040 Linz, Austria
Abstract Three unreinforced and two glass fiber reinforced polymeric materials were tested with regard to their mechanical and fatigue crack growth (FCG) behavior. In the group of the unreinforced plastics, one elastomer modified copolymer (acrylic ester-styreneacrylonitrile; ASA), one polymer blend of ASA and polycarbonate (PC) as well as one polymer blend of polybutylene terephthalate (PBT) and PC were investigated. As to the reinforced plastics, two glass fiber reinforced polyamide (PA) grades, differing in glass fiber length and glass fiber content, were characterized. Monotonic tensile tests and cyclic fracture mechanics tests were conducted at service relevant temperatures (0°C, 23°C, 60°C). For the three unreinforced plastics, significantly lower values for Young’s modulus and strength were obtained. Conversely, these materials revealed much higher strain-at-break values. Furthermore, glass fiber content dependent mechanical properties were determined for the reinforced plastics. With increasing temperatures, the Young’s moduli and strengths of all polymeric materials decreased distinctly. In terms of fatigue crack growth (FCG), the unreinforced plastics revealed an inferior FCG resistance at all temperatures. The FCG behavior of the reinforced plastics was influenced by the fiber orientation. Tests with the initial crack in mold direction exhibited a superior FCG resistance for the PA grade with the higher fiber content. In contrast, with the initial crack in cross direction, both glass fiber reinforced PA grades showed a similar FCG behavior. These effects are related to differences in the layer thicknesses of equally oriented fibers, which were obtained via microscopic investigation of the fatigue fracture surfaces. © 2019 Elsevier Ltd. All rights reserved. Selection and/or Peer-review under responsibility of AEM2017. Keywords: plastics, solar-thermal collector, mounting system, mechanical properties, fatigue crack growth
* Corresponding author. Tel.: +43-732-2468-6617; fax: +43-732-2468-6613. E-mail address: [email protected] 2214-7853 © 2019 Elsevier Ltd. All rights reserved. Selection and/or Peer-review under responsibility of AEM2017.
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
477
1. Introduction Mounting systems of solar-thermal collectors are currently constructed from aluminum or stainless steel [1,2]. In order to reduce system costs and component weights, in many applications, conventional materials will be replaced by advanced polymeric material solutions. However, for an appropriate material replacement, various aspects must be carefully considered. Besides mechanical loads (e.g., weight of the mounted solar-thermal collector, snow weight, wind load, hail), materials for mounting and framing must withstand environmental loads (e.g., temperatures, environmental media) [1-3]. These loading conditions can adversely affect the structural performance of polymeric materials [4-10]. Given the importance of understanding the material behavior under service relevant loading conditions, previous material research focused on the effect of superimposed mechanical and environmental loads [10-14]. The main objective of this research was the characterization of the mechanical and fatigue behavior of polymeric materials for mounting systems of solar-thermal collectors under application relevant temperatures. Therefore, monotonic tensile tests and cyclic fracture mechanics tests were carried out at three different temperatures (0°C, 23°C, 60°C) in air environment. Furthermore, the fatigue fracture surfaces were investigated by microscopic characterization. 2. Background In order to characterize the fatigue crack growth (FCG) behavior of plastics, the well-established concept of linear elastic fracture mechanics is applied [10-16]. In this concept, the fatigue crack growth rate (da/dN) is plotted against the stress intensity factor range ∆K. While the FCG rate indicates the crack growth per cycle, the ∆KI-value, described in eq. 1, is the difference between maximum and minimum stress intensity factor (KI,max-KI,min) of a cycling loading profile in loading mode I (pure tensile loads). In this equation, “∆σ” describes the global loading, “a” the crack length and “Y” a factor which is dependent on the geometry of the specimen used.
K I a Y
(1)
Region II of the double-logarithmic FCG kinetics curve represents the stable crack growth regime of the material. Comparing various plastics or different test conditions, enhanced FCG resistance is indicated by a shift of the FCG curve to higher ∆K-values. Furthermore, also a decrease in the slope of the FCG curve reveals improved FCG behavior [17,18]. 3. Experimental In this research, three unreinforced and two reinforced commercial polymeric materials were characterized. The three unreinforced polymeric materials included an elastomer-modified copolymer (acrylic ester-styreneacrylonitrile, ASA) and two polymer blends (acrylic ester-styrene-acrylonitrile/polycarbonate, ASA/PC and polybutylene terephthalate/polycarbonate, PBT/PC). In the category of the reinforced materials one short glass fiber reinforced polyamide grade with a glass fiber content of 30 w% (PA6-GF30) and one long glass fiber reinforced polyamide grade with a glass fiber content of 40 w% (PA6-LGF40) were investigated. An overview of material designation and the material composition is provided in Table 1. For all materials, mechanical properties and fatigue crack growth resistances were tested at three different temperatures (0°C, 23°C, 60°C) in air environment. In order to gain mechanical properties, tensile tests were carried out with a screw-driven universal tensile testing machine Zwick Z020 (Zwick GmbH & Co. KG, Germany) equipped with a temperature chamber to establish stress-strain curves at various temperatures. An extensometer was used to measure the Young’s Modulus between the strain range of 0.05% and 0.25%. Up to a strain of 0.25%, the testing speed was set to 1 mm/min. Further testing speed until break of the specimens was fixed with 50 mm/min. For all monotonic tensile tests, 1B multi-purpose specimens according to ISO-527-2 were used. Specimens were tested to evaluate Young’s modulus, strength and strain-at-break.
478
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484 Table 1. Material designation and material composition. Material designation
Material
Reinforcement type
Reinforcement content
ASA
Acrylic ester-styrene-acrylonitrile copolymer
-
-
ASA-PC
Polymer blend of acrylic ester-styreneacrylonitrile copolymer and polycarbonate
-
-
PBT-PC
Polymer blend of polybutylene terephthalate and polycarbonate
-
-
PA6-GF30
Polyamide 6
Short glass fibers
30 w%
PA6-LGF40
Polyamide 6
Long glass fibers
40 w%
All fracture mechanics tests were carried out with compact type (CT) specimens (see Fig 1) manufactured out of injection molded plaques. Considering the fiber orientation injection molding of glass fiber reinforced polymeric materials leads to a multi-layer structure [19]. To investigate the influence of fiber orientation, specimens were produced with the initial crack in mold direction (see Fig. 1, left) and in cross direction (see Fig. 1, right). Tests were performed in a media containment testing device (see Fig. 2a), originally developed by Schoeffl et al. [20], which allows for in-situ tests with superimposed mechanical and environmental loading. For the cyclic crack growth tests, the electro-dynamic testing machine ElectroPuls E3000 (Instron Inc., USA) was used. The cyclic tests were performed using a sinusoidal loading profile with a frequency of 5 Hz (schematically illustrated in Fig. 2b) and an R-ratio (ratio between the minimum and maximum applied stress) of 0.1.
Fig. 1. Illustration of sample extraction for compact type (CT) specimens with the initial crack in mold direction (left) and cross direction (right).
a)
b)
Fig. 2. (a) In-situ media testing device for an electro-dynamic testing machine and (b) a compact type (CT) specimen with sinusoidal loading profile.
Depending on material and temperature, the applied maximum force, representing the initial stress intensity factor, was chosen appropriately. To detect the crack length, the setup was equipped with a specific adapted camera and software system. Out of the established data, the fatigue crack growth rates, da/dN, and the stress intensity factor range, ΔKI, describing the local crack tip stress field, were investigated.
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
479
For the microscopic investigation of the fatigue fracture surfaces, the stereo microscope SZX16 (Olympus Austria GmbH, Austria) and the confocal laser microscope LEXT OLS 4000 (Olympus Austria GmbH, Austria) were used. With the confocal laser microscope, the fracture surfaces were analyzed by layered scanning using a laser with a wavelength of 405 nm and an objective with a magnification of 100x. For the wide-field observation of the surface, the software stitched the high-resolution images. 4. Results and Discussion The temperature dependent stress-strain curves for the three unreinforced (ASA, ASA-PC, PBT-PC) and the two glass fiber reinforced (PA6-GF30, PA6-LGF40) plastics are depicted in Fig. 3. Out of these stress-strain curves, the values for Young’s modulus, strength and strain-at-break were determined (see. Fig. 4). At all three testing temperatures, the reinforced polymeric materials revealed significantly higher Young’s moduli (factor 3 to 6) and strengths (factor 3 to 4). Conversely, the obtained strain-at-break values were much lower (factor 2 to 27). An increase in testing temperature led to declining Young’s moduli. The determined values are ranging from 2620 MPa to 2000 MPa for ASA, from 2430 MPa to 1940 MPA for ASA-PC, from 2390 MPa to 1680 MPa for PBT-PC, from 9220 MPa to 5600 MPa for PA6-GF30 and from 12680 MPa to 8490 MPa for PA6-LGF40. While the lowest values were determined for the polymer blend PBT-PC, the highest values were received for the long glass fiber reinforced PA grade PA-LGF40. Hence, a higher reinforcement content was accompanied by an increase in Young’s modulus. Also for the strengths, a decrease with higher temperatures was obtained. However, increasing temperatures resulted in higher strain-at-break values ranging from 8.1% to 29.5% for ASA, from 19.9% to 91.5% for ASA-PC, from 83.4% to 126.2% for PBT-PC, from 5.4% to 7.2% for PA6-GF30 and from 4.9% to 4.7% for PA6-LGF40. For PBTPC, by far the highest strain-at-break values were achieved. In the glass fiber reinforced grades, the fibers have a dominating influence on the strain-at-break. Thus, a brittle failure and small strain-at-break values were obtained for PA6-GF30 and PA-6LGF40. 250
unreinforced plastics ASA ASA-PC PBT-PC reinforced plastics PA6-GF30 PA6-LGF40
200 150 100
0°C
50 0
stress (MPa)
23°C
200 150 100 50 0 60°C
200 150 100 50 0
0
20
100
120
140
strain (%)
Fig. 3. Temperature dependent stress-strain curves for the three unreinforced (ASA, ASA-PC, PBT-PC) and the two glass fiber reinforced (PA6GF30, PA6-LGF40) plastics.
480
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
Young's modulus (GPa)
ASA
ASA-PC
PBT-PC
PA6-GF30
PA6-LGF40
16 12 8 4 0
strength (MPa)
250 200 150 100 50
strain-at-break (%)
0 160 120 80 40 0 23°C
0°C
60°C
temperature (°C)
Fig. 4. Temperature dependent values of Young’s modulus (top), strength (middle) and strain-at-break (bottom) for the five plastics.
In Fig. 5, the effect of temperature on the fatigue crack growth (FCG) curves of the three unreinforced and the two glass fiber reinforced plastics is plotted. For all materials, temperature increase revealed a deterioration of the FCG resistance. 10-1 ASA
10-2 10-3 10-4
0°C 23°C 60°C
-5
10
10-6 10-1 PA6-GF30
10-2 -3
10
10-3
10-4
0°C 23°C 60°C
10-5
da/dN (mm/cycle)
da/dN (mm/cycle)
ASA-PC
10-2
10-6 PBT-PC
10-2 10-3 0°C 23°C 60°C
-5
10
0°C 23°C 60°C
10-6 PA6-LGF40
10-2
10-4 10-5
R = 0.1, f = 5 Hz, crack in mold direction
a)
10
-5
R = 0.1, f = 5 Hz, crack in mold direction
10-3
10-4
10-6
10-4
1
10-6
10 0.5
K (MPa m )
b)
0°C 23°C 60°C
1
10
K (MPa m0.5)
Fig. 5. Fatigue crack growth kinetics for (a) the three unreinforced (ASA, ASA-PC, PBT-PC) and (b) the two glass fiber reinforced (PA6-GF30, PA6-LGF40) plastics tested at various temperatures.
This effect is reflected by a shift of the crack growth curve to lower ∆KI-values at similar FCG rates. A comparison between the FCG curves for the temperatures 60°C and 0°C at a comparable crack growth rate of 2E-
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
481
4 mm/cycle revealed differences between the ∆KI-values with a factor of 2.1 for ASA, 1.9 for ASA-PC, 1.4 for PBT-PC, 1.5 for PA6-GF30 and 1.4 for PA6-LGF40. While the reinforced plastics exhibited similar FCG curve slopes at all temperatures, for the unreinforced plastics smaller slopes were found with higher temperatures. Due to the different slopes, the curves tend to converge at higher crack growth rates leading to diminishing factors for the ∆KI-value differences at similar FCG rates. A comparison between the fatigue crack growth resistances of the five polymeric materials for each testing temperature is illustrated in Fig. 6. Additionally, the effect of different fiber orientation on the FCG behavior is shown. As to the material comparison, at all temperatures, the glass fiber reinforced plastics revealed a superior FCG resistance. In contrast, for ASA an inferior FCG behavior was detected. Comparing the results of PA6-LGF40 and ASA, which were obtained with tests with the initial crack in mold direction, at a crack growth rate of 2E4 mm/cycle, differences between the ∆KI-values with a factor of 9.4 at 60°C, 7.2 at 23°C and 6.4 at 0°C were determined. In terms of the effect of fiber length and fiber content, the PA grade with the long glass fiber reinforcement and the higher fiber content showed an enhanced FCG behavior in the tests with the initial crack in mold direction. At all temperatures, for PA6-LGF40 higher ∆KI-values with a factor of 1.3 were determined. However, tests with the initial crack in cross direction revealed a similar FCG resistance for both glass fiber reinforced plastics. While for PA6-GF30 at a comparable crack growth rate an increase of the stress intensity factor range with the factor 1.2 was achieved, the FCG behavior of PA6-LGF40 decreased with a factor of 0.9. 10-1 10-2 10-3
ASA ASA-PC PBT-PC PA6-GF30 PA6-LGF40
10-3 10-4
10-5
10-5
10-2 10-3 10-4 10-5 10-6
R = 0.1, f = 5 Hz, crack in mold direction
10-4 10-5
10-2
10-3
10-3
10-4
10-4
10-5
10-5 1
10-6
10
K [MPa m0.5]
b)
23°C
10-3
10-6
60°C
0°C
10-2
10-2
10-6
ASA ASA-PC PBT-PC PA6-GF30 PA6-LGF40
10-6
23°C
da/dN (mm/cycle)
da/dN (mm/cycle)
10-2
10-4
10-6
a)
10-1
0°C
R = 0.1, f = 5 Hz, crack in cross direction 60°C
1
10
K (MPa m0.5)
Fig. 6. Temperature dependent fatigue crack growth kinetics for the three unreinforced (ASA, ASA-PC, PBT-PC) and the two glass fiber reinforced (PA6-GF30, PA6-LGF40) plastics tested with the crack in (a) mold direction and (b) cross direction.
Fatigue fracture surfaces of the two unreinforced plastics ASA and PBT-PC are depicted in Fig. 7. A line marks the transition between the stable and the unstable crack growth. At 0°C and 23°C the position of the transition is similar. Nevertheless, at the highest temperature of 60°C the transition line appears after higher crack lengths reflected by a higher distance between the initial crack length (on the left side of the images) and the marking line. The boxes are indication the positions on the fracture surface where the confocal microscopic investigations were conducted. The gained images are illustrated in Fig 8. In these images slip bands are revealing the crack front
482
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
advance of the slow crack growth. For both materials (ASA and PBT-PC), with higher temperatures increasing slip band thicknesses were obtained.
Fig. 7. Fatigue fracture surfaces (initial crack on the left) of ASA (left) and PBT-PC (right) after tests at various temperatures together with marks for the transition between stable and unstable crack growth (line) and for the areas used for microscopic analyzes (boxes).
a)
b)
Fig. 8. Microscopic fatigue fracture surfaces (marked areas in Fig. 7) of (a) ASA and (b) PBT-PC after tests at various temperatures.
In Fig. 9, the fiber orientation distributions across the specimen thickness are shown by fatigue fracture surface images of specimens of PA6-GF30 (left) and PA6-LGF40 (right). In all images, again, the transition between stable and unstable crack growth is marked with a line. With higher testing temperatures, an increase of the crack length due to stable crack growth was achieved. However, a direct comparison between the layer thicknesses of equally oriented fibers exhibits a thicker central layer for PA6-LGF40 (about 35% of the specimen thickness) compared to PA6-GF30 (about 25% of the specimen thickness). This is related to a difference in mold filling behavior of these two materials, which is mainly caused by the different glass fiber lengths.
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
483
Fig. 9. Fatigue fracture surface (initial crack on the left) of PA6-GF30 (left) and PA6-LGF40 (right) after tests at various temperatures together with marks for the transition between stable and unstable crack growth (line).
5. Summary and Conclusions In this research, the mechanical properties and the fatigue crack growth (FCG) resistance of three unreinforced and two glass fiber reinforced plastics were characterized at service relevant temperatures (0°C, 23°C, 60°C). The unreinforced polymeric materials included an acrylic ester-styrene-acrylonitrile copolymer (ASA), a polymer blend of ASA and polycarbonate (ASA-PC) as well as a polymer blend of polybutylene terephthalate and PC (PBT-PC). The reinforced plastics comprised a short glass fiber reinforced polyamide grade with a fiber content of 30 w% (PAGF30) and a long glass fiber reinforced polyamide grade with a fiber content of 40 w% (PA-LGF40). In order to achieve mechanical properties, monotonic tensile tests were conducted. For the reinforced plastics significantly higher values for Young’s modulus and strength as well as lower strain-at-break values were obtained. With increasing temperatures, Young’s moduli and strengths decreased distinctly and the strain-at-break values increased. Fracture mechanics tests were performed to investigate the FCG behavior of the five polymeric materials. For the reinforced plastics, a superior FCG resistance was obtained. Additionally, the influence of the fiber orientation on the fatigue crack growth was investigated. While tests with the initial crack in cross direction revealed a similar FCG resistance, tests with the initial crack in mold direction showed a significantly better performance for PALGF40. This behavior can be explained by the microscopic detected differences in the layer thicknesses of equally oriented fibers. The microscopic investigation of the fatigue fracture surfaces of ASA and PBT-PC indicated higher slip band thicknesses with increasing testing temperatures. Acknowledgements This research work was performed in the cooperative research project MidTempColl. The project was funded by the Austrian Climate and Energy Fund (KLI.EN) within the program "Neue Energien 2020" and the funding was administrated by the Austrian Research Promotion Agency (FFG). References [1] M. Koehl, M.G. Meir, P. Papillon, G.M. Wallner, S. Saile (Eds.), Polymeric Materials for Solar Thermal Applications, first ed., Wiley-VCH, 2012. [2] R. Erfurth, J. Göring, T. Delzer, Tragkonstruktionen für Solaranlagen – Planungshandbuch zur Aufständerung von Solarkollektoren, first ed., Solarpraxis AG, Berlin, D. 2001. [3] Tragkonstruktionen für Solaranlagen: Planungshandbuch zur Aufständerung von Solarkollektoren. Berlin: Solarpraxis Supernova AG, (2001). [4] J.B. DeCoste, F.S. Malm, V.T. Wallder, Ind. Eng. Chem. 43 (1), 1951, pp. 117–121. [5] R.W. Lang, A. Stern, G. Doerner, Angew. Makromol. Chemie 247 (1), 1997, pp. 131–145. [6] A. Stern, M. Novotny, R.W. Lang, Polymer Testing 17 (6), 1998, pp. 403–422. [7] A. Stern, F. Asanger, R.W. Lang, Polymer Testing 17 (6), 1998, pp. 423–441. [8] R.W. Lang, G. Pinter, W. Balika, 3R international 44, 2005, pp. 33–41. [9] V. Altstädt; The Influence of Molecular Variables on Fatigue Resistance in Stress Cracking Environments. In: Kausch HH, editor. Intrinsic molecular mobility and toughness of polymers, vol. 188. Berlin, London: Springer, pp. 105–152, 2005. [10] J. Fischer, P.R. Bradler, M. Schlaeger, G.M. Wallner, R.W. Lang, Energy Procedia 91, 2016, 27–34. [11] P.F. Schoeffl, R.W. Lang, International Journal of Fatigue 72, 2015, 90-101.
484
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
[12] J. Fischer, P.R. Bradler, R.W. Lang, G.M. Wallner, Fatigue crack growth resistance of polypropylene in chlorinated water at different temperatures, Conference Proceedings of the 18th Plastic Pipes Conference, 2016. [13] P.R. Bradler, J. Fischer, M. Schlaeger, G.M. Wallner, R.W. Lang; Conference Proceedings of EuroSun 2016, ISES. [14] P.R. Bradler, J. Fischer, G.M. Wallner R.W. Lang, Energy Procedia 119, 2017, 828-834. [15] R.W. Lang; Applicability of linear elastic fracture mechanics to fatigue in poymers and short-fiber composites. Dissertation. Berhlehem (US), (1980). [16] T.L. Anderson; Fracture mechanics: Fundamentals and applications. Boca Raton, FL: CRC Press, (2017). [17] R.W. Lang, G. Pinter, W. Balika; Konzept zur Nachweisführung für Nutzungsdauer und Sicherheit von PE-Druckrohren bei beliebiger Einbausituation; 3R International (1-2/2005):32–41, 2005. [18] M. Haager, G. Pinter, Lang RW; Estimation of Slow Crack Growth Behavior in Polyethylene after Stepwise Isothermal Crystallization; Macromol. Symp.;217(1):383–90 2004. [19] R.W. Lang, J.A. Manson, R.W. Hertzberg; Effect of short glass fibers and particulate fillers on fatigue crack propagation in polyamides; Polym. Eng. Sci.;22(15):982–7, 1982. [20] P.F. Schoeffl, P.R. Bradler, R.W. Lang. Polymer Testing, 40 (2014) 225–33.
ScienceDirect Materials Today: Proceedings 10 (2019) 476–484
www.materialstoday.com/proceedings
AEM2017
Mechanical Behavior of Polymeric Materials for Mounting Systems of Solar-thermal Collectors Joerg Fischera,*, Patrick R. Bradlera, Sandra Leitnera, Gernot M. Wallnera, Reinhold W. Langa a
Institute of Polymeric Materials and Testing, Johannes Kepler University Linz, Altenbergerstrasse 69, 4040 Linz, Austria
Abstract Three unreinforced and two glass fiber reinforced polymeric materials were tested with regard to their mechanical and fatigue crack growth (FCG) behavior. In the group of the unreinforced plastics, one elastomer modified copolymer (acrylic ester-styreneacrylonitrile; ASA), one polymer blend of ASA and polycarbonate (PC) as well as one polymer blend of polybutylene terephthalate (PBT) and PC were investigated. As to the reinforced plastics, two glass fiber reinforced polyamide (PA) grades, differing in glass fiber length and glass fiber content, were characterized. Monotonic tensile tests and cyclic fracture mechanics tests were conducted at service relevant temperatures (0°C, 23°C, 60°C). For the three unreinforced plastics, significantly lower values for Young’s modulus and strength were obtained. Conversely, these materials revealed much higher strain-at-break values. Furthermore, glass fiber content dependent mechanical properties were determined for the reinforced plastics. With increasing temperatures, the Young’s moduli and strengths of all polymeric materials decreased distinctly. In terms of fatigue crack growth (FCG), the unreinforced plastics revealed an inferior FCG resistance at all temperatures. The FCG behavior of the reinforced plastics was influenced by the fiber orientation. Tests with the initial crack in mold direction exhibited a superior FCG resistance for the PA grade with the higher fiber content. In contrast, with the initial crack in cross direction, both glass fiber reinforced PA grades showed a similar FCG behavior. These effects are related to differences in the layer thicknesses of equally oriented fibers, which were obtained via microscopic investigation of the fatigue fracture surfaces. © 2019 Elsevier Ltd. All rights reserved. Selection and/or Peer-review under responsibility of AEM2017. Keywords: plastics, solar-thermal collector, mounting system, mechanical properties, fatigue crack growth
* Corresponding author. Tel.: +43-732-2468-6617; fax: +43-732-2468-6613. E-mail address: [email protected] 2214-7853 © 2019 Elsevier Ltd. All rights reserved. Selection and/or Peer-review under responsibility of AEM2017.
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
477
1. Introduction Mounting systems of solar-thermal collectors are currently constructed from aluminum or stainless steel [1,2]. In order to reduce system costs and component weights, in many applications, conventional materials will be replaced by advanced polymeric material solutions. However, for an appropriate material replacement, various aspects must be carefully considered. Besides mechanical loads (e.g., weight of the mounted solar-thermal collector, snow weight, wind load, hail), materials for mounting and framing must withstand environmental loads (e.g., temperatures, environmental media) [1-3]. These loading conditions can adversely affect the structural performance of polymeric materials [4-10]. Given the importance of understanding the material behavior under service relevant loading conditions, previous material research focused on the effect of superimposed mechanical and environmental loads [10-14]. The main objective of this research was the characterization of the mechanical and fatigue behavior of polymeric materials for mounting systems of solar-thermal collectors under application relevant temperatures. Therefore, monotonic tensile tests and cyclic fracture mechanics tests were carried out at three different temperatures (0°C, 23°C, 60°C) in air environment. Furthermore, the fatigue fracture surfaces were investigated by microscopic characterization. 2. Background In order to characterize the fatigue crack growth (FCG) behavior of plastics, the well-established concept of linear elastic fracture mechanics is applied [10-16]. In this concept, the fatigue crack growth rate (da/dN) is plotted against the stress intensity factor range ∆K. While the FCG rate indicates the crack growth per cycle, the ∆KI-value, described in eq. 1, is the difference between maximum and minimum stress intensity factor (KI,max-KI,min) of a cycling loading profile in loading mode I (pure tensile loads). In this equation, “∆σ” describes the global loading, “a” the crack length and “Y” a factor which is dependent on the geometry of the specimen used.
K I a Y
(1)
Region II of the double-logarithmic FCG kinetics curve represents the stable crack growth regime of the material. Comparing various plastics or different test conditions, enhanced FCG resistance is indicated by a shift of the FCG curve to higher ∆K-values. Furthermore, also a decrease in the slope of the FCG curve reveals improved FCG behavior [17,18]. 3. Experimental In this research, three unreinforced and two reinforced commercial polymeric materials were characterized. The three unreinforced polymeric materials included an elastomer-modified copolymer (acrylic ester-styreneacrylonitrile, ASA) and two polymer blends (acrylic ester-styrene-acrylonitrile/polycarbonate, ASA/PC and polybutylene terephthalate/polycarbonate, PBT/PC). In the category of the reinforced materials one short glass fiber reinforced polyamide grade with a glass fiber content of 30 w% (PA6-GF30) and one long glass fiber reinforced polyamide grade with a glass fiber content of 40 w% (PA6-LGF40) were investigated. An overview of material designation and the material composition is provided in Table 1. For all materials, mechanical properties and fatigue crack growth resistances were tested at three different temperatures (0°C, 23°C, 60°C) in air environment. In order to gain mechanical properties, tensile tests were carried out with a screw-driven universal tensile testing machine Zwick Z020 (Zwick GmbH & Co. KG, Germany) equipped with a temperature chamber to establish stress-strain curves at various temperatures. An extensometer was used to measure the Young’s Modulus between the strain range of 0.05% and 0.25%. Up to a strain of 0.25%, the testing speed was set to 1 mm/min. Further testing speed until break of the specimens was fixed with 50 mm/min. For all monotonic tensile tests, 1B multi-purpose specimens according to ISO-527-2 were used. Specimens were tested to evaluate Young’s modulus, strength and strain-at-break.
478
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484 Table 1. Material designation and material composition. Material designation
Material
Reinforcement type
Reinforcement content
ASA
Acrylic ester-styrene-acrylonitrile copolymer
-
-
ASA-PC
Polymer blend of acrylic ester-styreneacrylonitrile copolymer and polycarbonate
-
-
PBT-PC
Polymer blend of polybutylene terephthalate and polycarbonate
-
-
PA6-GF30
Polyamide 6
Short glass fibers
30 w%
PA6-LGF40
Polyamide 6
Long glass fibers
40 w%
All fracture mechanics tests were carried out with compact type (CT) specimens (see Fig 1) manufactured out of injection molded plaques. Considering the fiber orientation injection molding of glass fiber reinforced polymeric materials leads to a multi-layer structure [19]. To investigate the influence of fiber orientation, specimens were produced with the initial crack in mold direction (see Fig. 1, left) and in cross direction (see Fig. 1, right). Tests were performed in a media containment testing device (see Fig. 2a), originally developed by Schoeffl et al. [20], which allows for in-situ tests with superimposed mechanical and environmental loading. For the cyclic crack growth tests, the electro-dynamic testing machine ElectroPuls E3000 (Instron Inc., USA) was used. The cyclic tests were performed using a sinusoidal loading profile with a frequency of 5 Hz (schematically illustrated in Fig. 2b) and an R-ratio (ratio between the minimum and maximum applied stress) of 0.1.
Fig. 1. Illustration of sample extraction for compact type (CT) specimens with the initial crack in mold direction (left) and cross direction (right).
a)
b)
Fig. 2. (a) In-situ media testing device for an electro-dynamic testing machine and (b) a compact type (CT) specimen with sinusoidal loading profile.
Depending on material and temperature, the applied maximum force, representing the initial stress intensity factor, was chosen appropriately. To detect the crack length, the setup was equipped with a specific adapted camera and software system. Out of the established data, the fatigue crack growth rates, da/dN, and the stress intensity factor range, ΔKI, describing the local crack tip stress field, were investigated.
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
479
For the microscopic investigation of the fatigue fracture surfaces, the stereo microscope SZX16 (Olympus Austria GmbH, Austria) and the confocal laser microscope LEXT OLS 4000 (Olympus Austria GmbH, Austria) were used. With the confocal laser microscope, the fracture surfaces were analyzed by layered scanning using a laser with a wavelength of 405 nm and an objective with a magnification of 100x. For the wide-field observation of the surface, the software stitched the high-resolution images. 4. Results and Discussion The temperature dependent stress-strain curves for the three unreinforced (ASA, ASA-PC, PBT-PC) and the two glass fiber reinforced (PA6-GF30, PA6-LGF40) plastics are depicted in Fig. 3. Out of these stress-strain curves, the values for Young’s modulus, strength and strain-at-break were determined (see. Fig. 4). At all three testing temperatures, the reinforced polymeric materials revealed significantly higher Young’s moduli (factor 3 to 6) and strengths (factor 3 to 4). Conversely, the obtained strain-at-break values were much lower (factor 2 to 27). An increase in testing temperature led to declining Young’s moduli. The determined values are ranging from 2620 MPa to 2000 MPa for ASA, from 2430 MPa to 1940 MPA for ASA-PC, from 2390 MPa to 1680 MPa for PBT-PC, from 9220 MPa to 5600 MPa for PA6-GF30 and from 12680 MPa to 8490 MPa for PA6-LGF40. While the lowest values were determined for the polymer blend PBT-PC, the highest values were received for the long glass fiber reinforced PA grade PA-LGF40. Hence, a higher reinforcement content was accompanied by an increase in Young’s modulus. Also for the strengths, a decrease with higher temperatures was obtained. However, increasing temperatures resulted in higher strain-at-break values ranging from 8.1% to 29.5% for ASA, from 19.9% to 91.5% for ASA-PC, from 83.4% to 126.2% for PBT-PC, from 5.4% to 7.2% for PA6-GF30 and from 4.9% to 4.7% for PA6-LGF40. For PBTPC, by far the highest strain-at-break values were achieved. In the glass fiber reinforced grades, the fibers have a dominating influence on the strain-at-break. Thus, a brittle failure and small strain-at-break values were obtained for PA6-GF30 and PA-6LGF40. 250
unreinforced plastics ASA ASA-PC PBT-PC reinforced plastics PA6-GF30 PA6-LGF40
200 150 100
0°C
50 0
stress (MPa)
23°C
200 150 100 50 0 60°C
200 150 100 50 0
0
20
100
120
140
strain (%)
Fig. 3. Temperature dependent stress-strain curves for the three unreinforced (ASA, ASA-PC, PBT-PC) and the two glass fiber reinforced (PA6GF30, PA6-LGF40) plastics.
480
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
Young's modulus (GPa)
ASA
ASA-PC
PBT-PC
PA6-GF30
PA6-LGF40
16 12 8 4 0
strength (MPa)
250 200 150 100 50
strain-at-break (%)
0 160 120 80 40 0 23°C
0°C
60°C
temperature (°C)
Fig. 4. Temperature dependent values of Young’s modulus (top), strength (middle) and strain-at-break (bottom) for the five plastics.
In Fig. 5, the effect of temperature on the fatigue crack growth (FCG) curves of the three unreinforced and the two glass fiber reinforced plastics is plotted. For all materials, temperature increase revealed a deterioration of the FCG resistance. 10-1 ASA
10-2 10-3 10-4
0°C 23°C 60°C
-5
10
10-6 10-1 PA6-GF30
10-2 -3
10
10-3
10-4
0°C 23°C 60°C
10-5
da/dN (mm/cycle)
da/dN (mm/cycle)
ASA-PC
10-2
10-6 PBT-PC
10-2 10-3 0°C 23°C 60°C
-5
10
0°C 23°C 60°C
10-6 PA6-LGF40
10-2
10-4 10-5
R = 0.1, f = 5 Hz, crack in mold direction
a)
10
-5
R = 0.1, f = 5 Hz, crack in mold direction
10-3
10-4
10-6
10-4
1
10-6
10 0.5
K (MPa m )
b)
0°C 23°C 60°C
1
10
K (MPa m0.5)
Fig. 5. Fatigue crack growth kinetics for (a) the three unreinforced (ASA, ASA-PC, PBT-PC) and (b) the two glass fiber reinforced (PA6-GF30, PA6-LGF40) plastics tested at various temperatures.
This effect is reflected by a shift of the crack growth curve to lower ∆KI-values at similar FCG rates. A comparison between the FCG curves for the temperatures 60°C and 0°C at a comparable crack growth rate of 2E-
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
481
4 mm/cycle revealed differences between the ∆KI-values with a factor of 2.1 for ASA, 1.9 for ASA-PC, 1.4 for PBT-PC, 1.5 for PA6-GF30 and 1.4 for PA6-LGF40. While the reinforced plastics exhibited similar FCG curve slopes at all temperatures, for the unreinforced plastics smaller slopes were found with higher temperatures. Due to the different slopes, the curves tend to converge at higher crack growth rates leading to diminishing factors for the ∆KI-value differences at similar FCG rates. A comparison between the fatigue crack growth resistances of the five polymeric materials for each testing temperature is illustrated in Fig. 6. Additionally, the effect of different fiber orientation on the FCG behavior is shown. As to the material comparison, at all temperatures, the glass fiber reinforced plastics revealed a superior FCG resistance. In contrast, for ASA an inferior FCG behavior was detected. Comparing the results of PA6-LGF40 and ASA, which were obtained with tests with the initial crack in mold direction, at a crack growth rate of 2E4 mm/cycle, differences between the ∆KI-values with a factor of 9.4 at 60°C, 7.2 at 23°C and 6.4 at 0°C were determined. In terms of the effect of fiber length and fiber content, the PA grade with the long glass fiber reinforcement and the higher fiber content showed an enhanced FCG behavior in the tests with the initial crack in mold direction. At all temperatures, for PA6-LGF40 higher ∆KI-values with a factor of 1.3 were determined. However, tests with the initial crack in cross direction revealed a similar FCG resistance for both glass fiber reinforced plastics. While for PA6-GF30 at a comparable crack growth rate an increase of the stress intensity factor range with the factor 1.2 was achieved, the FCG behavior of PA6-LGF40 decreased with a factor of 0.9. 10-1 10-2 10-3
ASA ASA-PC PBT-PC PA6-GF30 PA6-LGF40
10-3 10-4
10-5
10-5
10-2 10-3 10-4 10-5 10-6
R = 0.1, f = 5 Hz, crack in mold direction
10-4 10-5
10-2
10-3
10-3
10-4
10-4
10-5
10-5 1
10-6
10
K [MPa m0.5]
b)
23°C
10-3
10-6
60°C
0°C
10-2
10-2
10-6
ASA ASA-PC PBT-PC PA6-GF30 PA6-LGF40
10-6
23°C
da/dN (mm/cycle)
da/dN (mm/cycle)
10-2
10-4
10-6
a)
10-1
0°C
R = 0.1, f = 5 Hz, crack in cross direction 60°C
1
10
K (MPa m0.5)
Fig. 6. Temperature dependent fatigue crack growth kinetics for the three unreinforced (ASA, ASA-PC, PBT-PC) and the two glass fiber reinforced (PA6-GF30, PA6-LGF40) plastics tested with the crack in (a) mold direction and (b) cross direction.
Fatigue fracture surfaces of the two unreinforced plastics ASA and PBT-PC are depicted in Fig. 7. A line marks the transition between the stable and the unstable crack growth. At 0°C and 23°C the position of the transition is similar. Nevertheless, at the highest temperature of 60°C the transition line appears after higher crack lengths reflected by a higher distance between the initial crack length (on the left side of the images) and the marking line. The boxes are indication the positions on the fracture surface where the confocal microscopic investigations were conducted. The gained images are illustrated in Fig 8. In these images slip bands are revealing the crack front
482
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
advance of the slow crack growth. For both materials (ASA and PBT-PC), with higher temperatures increasing slip band thicknesses were obtained.
Fig. 7. Fatigue fracture surfaces (initial crack on the left) of ASA (left) and PBT-PC (right) after tests at various temperatures together with marks for the transition between stable and unstable crack growth (line) and for the areas used for microscopic analyzes (boxes).
a)
b)
Fig. 8. Microscopic fatigue fracture surfaces (marked areas in Fig. 7) of (a) ASA and (b) PBT-PC after tests at various temperatures.
In Fig. 9, the fiber orientation distributions across the specimen thickness are shown by fatigue fracture surface images of specimens of PA6-GF30 (left) and PA6-LGF40 (right). In all images, again, the transition between stable and unstable crack growth is marked with a line. With higher testing temperatures, an increase of the crack length due to stable crack growth was achieved. However, a direct comparison between the layer thicknesses of equally oriented fibers exhibits a thicker central layer for PA6-LGF40 (about 35% of the specimen thickness) compared to PA6-GF30 (about 25% of the specimen thickness). This is related to a difference in mold filling behavior of these two materials, which is mainly caused by the different glass fiber lengths.
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
483
Fig. 9. Fatigue fracture surface (initial crack on the left) of PA6-GF30 (left) and PA6-LGF40 (right) after tests at various temperatures together with marks for the transition between stable and unstable crack growth (line).
5. Summary and Conclusions In this research, the mechanical properties and the fatigue crack growth (FCG) resistance of three unreinforced and two glass fiber reinforced plastics were characterized at service relevant temperatures (0°C, 23°C, 60°C). The unreinforced polymeric materials included an acrylic ester-styrene-acrylonitrile copolymer (ASA), a polymer blend of ASA and polycarbonate (ASA-PC) as well as a polymer blend of polybutylene terephthalate and PC (PBT-PC). The reinforced plastics comprised a short glass fiber reinforced polyamide grade with a fiber content of 30 w% (PAGF30) and a long glass fiber reinforced polyamide grade with a fiber content of 40 w% (PA-LGF40). In order to achieve mechanical properties, monotonic tensile tests were conducted. For the reinforced plastics significantly higher values for Young’s modulus and strength as well as lower strain-at-break values were obtained. With increasing temperatures, Young’s moduli and strengths decreased distinctly and the strain-at-break values increased. Fracture mechanics tests were performed to investigate the FCG behavior of the five polymeric materials. For the reinforced plastics, a superior FCG resistance was obtained. Additionally, the influence of the fiber orientation on the fatigue crack growth was investigated. While tests with the initial crack in cross direction revealed a similar FCG resistance, tests with the initial crack in mold direction showed a significantly better performance for PALGF40. This behavior can be explained by the microscopic detected differences in the layer thicknesses of equally oriented fibers. The microscopic investigation of the fatigue fracture surfaces of ASA and PBT-PC indicated higher slip band thicknesses with increasing testing temperatures. Acknowledgements This research work was performed in the cooperative research project MidTempColl. The project was funded by the Austrian Climate and Energy Fund (KLI.EN) within the program "Neue Energien 2020" and the funding was administrated by the Austrian Research Promotion Agency (FFG). References [1] M. Koehl, M.G. Meir, P. Papillon, G.M. Wallner, S. Saile (Eds.), Polymeric Materials for Solar Thermal Applications, first ed., Wiley-VCH, 2012. [2] R. Erfurth, J. Göring, T. Delzer, Tragkonstruktionen für Solaranlagen – Planungshandbuch zur Aufständerung von Solarkollektoren, first ed., Solarpraxis AG, Berlin, D. 2001. [3] Tragkonstruktionen für Solaranlagen: Planungshandbuch zur Aufständerung von Solarkollektoren. Berlin: Solarpraxis Supernova AG, (2001). [4] J.B. DeCoste, F.S. Malm, V.T. Wallder, Ind. Eng. Chem. 43 (1), 1951, pp. 117–121. [5] R.W. Lang, A. Stern, G. Doerner, Angew. Makromol. Chemie 247 (1), 1997, pp. 131–145. [6] A. Stern, M. Novotny, R.W. Lang, Polymer Testing 17 (6), 1998, pp. 403–422. [7] A. Stern, F. Asanger, R.W. Lang, Polymer Testing 17 (6), 1998, pp. 423–441. [8] R.W. Lang, G. Pinter, W. Balika, 3R international 44, 2005, pp. 33–41. [9] V. Altstädt; The Influence of Molecular Variables on Fatigue Resistance in Stress Cracking Environments. In: Kausch HH, editor. Intrinsic molecular mobility and toughness of polymers, vol. 188. Berlin, London: Springer, pp. 105–152, 2005. [10] J. Fischer, P.R. Bradler, M. Schlaeger, G.M. Wallner, R.W. Lang, Energy Procedia 91, 2016, 27–34. [11] P.F. Schoeffl, R.W. Lang, International Journal of Fatigue 72, 2015, 90-101.
484
Joerg Fischer et al / Materials Today: Proceedings 10 (2019) 476–484
[12] J. Fischer, P.R. Bradler, R.W. Lang, G.M. Wallner, Fatigue crack growth resistance of polypropylene in chlorinated water at different temperatures, Conference Proceedings of the 18th Plastic Pipes Conference, 2016. [13] P.R. Bradler, J. Fischer, M. Schlaeger, G.M. Wallner, R.W. Lang; Conference Proceedings of EuroSun 2016, ISES. [14] P.R. Bradler, J. Fischer, G.M. Wallner R.W. Lang, Energy Procedia 119, 2017, 828-834. [15] R.W. Lang; Applicability of linear elastic fracture mechanics to fatigue in poymers and short-fiber composites. Dissertation. Berhlehem (US), (1980). [16] T.L. Anderson; Fracture mechanics: Fundamentals and applications. Boca Raton, FL: CRC Press, (2017). [17] R.W. Lang, G. Pinter, W. Balika; Konzept zur Nachweisführung für Nutzungsdauer und Sicherheit von PE-Druckrohren bei beliebiger Einbausituation; 3R International (1-2/2005):32–41, 2005. [18] M. Haager, G. Pinter, Lang RW; Estimation of Slow Crack Growth Behavior in Polyethylene after Stepwise Isothermal Crystallization; Macromol. Symp.;217(1):383–90 2004. [19] R.W. Lang, J.A. Manson, R.W. Hertzberg; Effect of short glass fibers and particulate fillers on fatigue crack propagation in polyamides; Polym. Eng. Sci.;22(15):982–7, 1982. [20] P.F. Schoeffl, P.R. Bradler, R.W. Lang. Polymer Testing, 40 (2014) 225–33.