- Email: [email protected]
Impact and fracture analysis of fish scales from Arapaima gigas
Materials Science and Engineering C 51 (2015) 153–157
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Short Communication
Impact and fracture analysis of fish scales from Arapaima gigas F.G. Torres ⁎, M. Malásquez, O.P. Troncoso Department of Mechanical Engineering, Pontificia Universidad Catolica del Peru, Av. Universitaria 1801, Lima 32, Peru
a r t i c l e
i n f o
Article history: Received 16 October 2014 Received in revised form 15 January 2015 Accepted 23 February 2015 Available online 25 February 2015 Keywords: Fish scale Arapaima gigas Impact behaviour
a b s t r a c t Fish scales from the Amazonian fish Arapaima gigas have been characterised to study their impact and fracture behaviour at three different environmental conditions. Scales were cut in two different directions to analyse the influence of the orientation of collagen layers. The energy absorbed during impact tests was measured for each sample and SEM images were taken after each test in order to analyse the failure mechanisms. The results showed that scales tested at cryogenic temperatures display fragile behaviour, while scales tested at room temperature did not fracture. Different failure mechanisms have been identified, analysed and compared with the failure modes that occur in bone. The impact energy obtained for fish scales was two to three times higher than the values reported for bone in the literature. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Fish scales are an example of collagen based biological tissues. They are plywood-like structures of closely packed collagen fibre layers reinforced with a mineral phase of calcium deficient hydroxyapatite [1]. They provide fish with a flexible and protective skin that allows for changes in shape. Recently, the interest in fish scales as a model material for the development of bioinspired composite materials has increased. Some of the studies reported in the literature include the characterisation of fish scales from Pagrus major [1], Arapaima gigas [2], Cyprinus carpio [3], Lates calcarifer [4], Labeo rohita [5] and Atractosteus spatula [6]. According to Ikoma et al. [7] type I collagen fibrils are the main organic component in fish scales. Scales are composed of several layers of parallel collagen fibre bundles, with varying fibre orientations in the different layers [8]. The mineral phase in fish scales is mainly calcium deficient hydroxyapatite (HAP). HAP forms plate-like nano-crystals that could act as a nano-reinforcement in a collagen matrix. Torres et al. [2] have reported that the mineral content of scales varies along their cross-section, which accounts for a harder outer surface and a more flexible inner layer. Most studies performed on fish scales have reported tensile tests results. Torres et al. [2] measured the maximum tensile strength and Young's modulus of dry and humid A. gigas scales, and found that the Young's modulus of dry scales was 1.38 GPa while the modulus of humid scales was 0.83 GPa. Lin et al. [8] have also performed tensile tests on A. gigas scales. They reported an elastic modulus of 1.2 GPa for ⁎ Corresponding author. E-mail address: [email protected] (F.G. Torres).
http://dx.doi.org/10.1016/j.msec.2015.02.034 0928-4931/© 2015 Elsevier B.V. All rights reserved.
dry scales and 0.1 GPa for hydrated scales. Ikoma et al. [1] have reported an elastic modulus of 2.2 GPa for scales of P. major. Indentation tests at different scales (macro-, micro- and nano-) have also been used to assess the mechanical properties of fish scales. Lin et al. [8] have reported an average hardness value in the internal layers of A. gigas scales of 200 MPa, increasing to about 550 MPa in the external layer. Indentation tests carried out by Torres et al. [9] showed that the hardness of the cross section of A. gigas scale follows a cyclic sawtooth shaped pattern. The higher hardness values correspond to the laminates where fibres are perpendicular to the indentation plane. Bruet et al. [10] found that the scales from Polypterus senegalus have hardness values that increase with distance from the inner surface to the outer surface, from 0.54 GPa to 4.5 GPa. Besides being hard and strong, fish scales are also a very tough material. In this paper we carry out impact tests with samples at three different conditions to determine their response with regard to impact loads. Considering that fish scales display a laminate type structure, the experiments conducted under impact conditions have been analysed to determine whether fish scales behave in a similar way to standard fibre reinforced polymer composites. Failure analysis using scanning electron microscopy was also used to complement this analysis.
2. Experimental part 2.1. Materials Scales from the Amazonian fish A. gigas (body weight 100–150 kg) were washed and stored in standard conditions (20 °C and 80% of
154
F.G. Torres et al. / Materials Science and Engineering C 51 (2015) 153–157
2.4. Scanning electron microscopy (SEM) The fracture surfaces from impact testing specimens were studied using a FEI — QUANTA 200 SEM (Hillsboro, OR) in low vacuum with a voltage of 30 kV and a working distance in the range of 9.9–11.2 mm. The specimens were mounted onto metal stubs and were viewed in the SEM without any coating. 3. Results and discussion
Fig. 1. Sample directions: longitudinal (1) and transverse (2).
Table 1 Impact results from transverse and longitudinal samples. Condition
Direction
Energy absorbed (kJ/m2)
Dry sample/room temperature Dry sample/room temperature Dry sample/cryogenic temperature Dry sample/cryogenic temperature Humid sample/room temperature Humid sample/room temperature
Transverse Longitudinal Transverse Longitudinal Transverse Longitudinal
42.49 ± 6.55 35.11 ± 3.07 25.71 ± 4.41 14.32 ± 2.13 17.08 ± 3.65 11.92 ± 0.58
relative humidity). They were around 70–75 mm in length and around 1 mm thick.
2.2. Sample preparation Rectangular samples were cut off from A. gigas scales in two different directions. Fig. 1 shows a schematic view of the longitudinal and transverse directions. The samples measured 40 mm in length (L), 8 mm in width (W) and 1.5 mm in thickness, which is in agreement with the L/W relation recommended in the ASTM 4812 standard. The edges of each sample were sanded to avoid possible residual stresses.
2.3. Impact tests Charpy impact tests were performed in a CEAST Resil Impactor (Norwood, MA) with un-notched specimens. The maximum impact energy capacity of the machine was 7.5 J. The anvil gap was 22 mm. Three different conditions were used in this study (Table 1). Dry samples were tested at room temperature and at cryogenic conditions whereas humid samples were tested at room temperature. Humid samples were prepared by soaking dry specimens in distilled water 24 h before the tests. For cryogenic conditions, the specimens were kept in liquid nitrogen for 10 min and immediately after mounted onto the anvil of the impact testing equipment. A minimum of three samples were tested for each condition.
Impact tests results are depicted in Table 1. Three types of fracture were observed in the tests: a) full fracture; b) major fracture with tie; c) minor fracture (most of the cross section was still intact). The samples tested at room temperature did not undergo full fracture, as indicated by the total separation of the two halves of the sample. Fig. 2a and b shows representative longitudinal and transverse samples after impact tests at room temperature. They suffered minor fractures and showed no visible external change of shape. By contrast, samples tested at cryogenic conditions underwent full and major fractures. Fig. 3a and b shows representative samples with major fractures. The two parts of the specimens are held together only by the external layers. In Fig. 3c and d we can see samples which underwent full fracture. In this case total separation of the specimens can be observed. The impact energy values measured in this study (Table 1) are two to three times higher than those of other collagen based structures reported in the literature. Kovan [11] has reported the impact energy of the mandible bone, with values in the range of 8–20 kJ/m2. Reilly and Currey [12] reported impact energy values of 6–24 kJ/m2 for bovine tibia and humerus samples. In all cases the values reported were obtained at room temperature. The results in Table 1 show that the impact energy was lower for the samples tested at cryogenic conditions in both directions. Moreover, transverse samples at cryogenic conditions absorbed more energy than longitudinal samples at the same conditions. The energy absorbed during impact was around 65% higher for the transverse samples. At cryogenic conditions the molecular mobility of the collagen helices decreases with regard to room temperature. Specimens with reduced molecular mobility find it difficult to deform plastically at high velocity loading, as it occurs during impact. In addition, fracture mechanisms typical of brittle fracture also take place and further decrease the overall impact energy. Specimens cut in what we have defined as the transverse direction of the scales (Fig. 1), have shown higher impact energy (21.02% higher at room temperature and 79.54% higher for cryogenic tests) with regard to longitudinal samples. Several reports in the literature have described fish scales as plywood like structures formed by collagen lamellae arranged in layers, the lamellae being assembled from mineralized collagen fibrils [8]. Other authors have reported a Bouligand type (twisted plywood) structure [13,14]. It is reasonable to think that a Bouligand type arrangement ensures that the scale has in-plane isotropy in mechanical response [8]. However, results reported in the literature are contradictory. Yang et al. [14] found that the tensile strength of the scales is 50% higher in the longitudinal than in the transverse direction
Fig. 2. Representative longitudinal (a) and transverse (b) samples after impact tests at room temperature.
F.G. Torres et al. / Materials Science and Engineering C 51 (2015) 153–157
155
Fig. 3. Types of fractures observed after impact tests at cryogenic conditions: partial fracture of longitudinal sample (a), partial fracture of transverse sample (b), total fracture of longitudinal sample (c) and total fracture of transverse sample (d).
Fig. 4. SEM image: panoramic view of longitudinal samples complete fracture under cryogenic condition. (a) Mark zone indicates where brittle fracture occurred. (b) Marked zones indicate where fibre breakage occurred (circle) and where brittle fracture occurred (rectangle).
whereas Zhu et al. [15] reported that specimens with an orientation of 45° or 90° to the longitudinal axis have higher strengths (60 MPa) than the 0° orientation (40 MPa). The impact tests performed in this study have confirmed the anisotropic behaviour of fish scales as the transverse direction is tougher than the longitudinal direction. So far it is not clear if this anisotropy could be attributed to the fibre orientation or to any failure mechanism such as delamination. The main failure mechanisms observed in the SEM micrographs are described below. The fracture surface depicted in Fig. 4a corresponds to a specimen that suffered complete fracture. Here we can observe the
rough external layer of the scale and the internal layers formed from collagen fibres. The fracture surface shows the different orientations of collagen fibres bundles parallel to each other. In the same figure, the rectangle shows the region where brittle fracture occurred across the entire fracture surface. Fig. 4b is a detail view of the same sample shown in Fig. 4a. It shows the fracture of individual collagen fibres. Tearing between individual fibres can also be observed on the marked rectangle. Fibre breakage takes place away from the crack plane after fibre fracture occurred. On the right side of the figure, the ellipse indicates the region where brittle fracture occurred in the sample.
Fig. 5. SEM image: panoramic view of longitudinal samples partial fracture tested at cryogenic conditions. (a) Mark zones where fibre delamination and crack propagation occurred. (b) Mark zone indicates where tearing occurred.
156
F.G. Torres et al. / Materials Science and Engineering C 51 (2015) 153–157
In Fig. 5a and b we can observe the partial fracture of a longitudinal sample. The film which binds the two pieces is formed by collagen fibrils. Fig. 5a shows how delamination between collagen fibres occurred (rectangles). This failure mechanism starts before the scale fractures, and decreases the maximum fracture energy. After delamination occurs, the laminate structure suffers crack propagation on the zones where delamination is visible. In Fig. 5b, the marked rectangle shows that tearing between collagen fibres was also present in longitudinal samples. Fish scales display much higher impact energy compared to bone, at least 2 to 3 times higher. Both, bone and fish scales are made out of collagen and hydroxyapatite. The higher impact energy in fish scales compared to bone, both dense and porous, is mainly due to the fact that fish scales are built in a similar way to a laminate fibre reinforced composite material. The failure mechanisms observed in the fish scales samples that have undergone impact show some correspondence with those mechanisms that occur in fibre reinforced composites [16]. These include fibre breakage, delamination and matrix crack propagation. The higher content of collagen observed in the internal layers of fish scales (83% [10]) compared with 50–55% [17] of lamellar bone accounts for a more ductile behaviour in fish scales when subjected to impact. It is important to notice that such an increase in toughness does not compromise strength [11,12]. In the wet state this advantage is even more noticeable, since water acts as a plasticizing agent in collagen increasing its free volume [18], thus promoting a more rubbery behaviour under mechanical deformation. This effect has been verified for instance for mussel byssus, which are tough fibres formed mainly by collagen [19]. Figs. 6 and 7 show SEM micrographs obtained from a wet specimen that had undergone an impact test. In both figures, it can be easily observed that none of the previously described failure mechanisms, such as delamination, fibre breakage and crack propagation, have occurred. 4. Conclusions We have studied the impact behaviour of the scales from A. gigas. The results showed that dry and humid scales tested at room temperature did not undergo full fracture (total separation of two halves of the samples). By contrast, samples tested at cryogenic conditions did undergo major or full fracture, displaying different failure mechanisms such as delamination, fibre breakage, crack propagation and brittle fracture. All samples tested at low temperatures displayed fragile behaviour. With regard to the orientation of the specimens within the scales, the tests
Fig. 7. SEM image: panoramic view of transverse samples surface tested at humid condition.
performed in this study showed that for all the environmental conditions assessed, transverse samples absorbed higher levels of energy during impact than longitudinal samples. We have compared the impact energy from tests in this study with the impact energy reported for bone in the literature, which is another structure made out of collagen and hydroxyapatite. The impact energy obtained from the tests reported here for fish scales was two to three times higher compared to the impact energy reported for bone in the literature. This could be due to the fact that fish scales have a structure similar to a laminate fibre reinforced composite material, which improves the dissipation of energy during impact. Acknowledgements The authors would like to thank the Peruvian Science and Technology Program (159-FINCyT-IB-2013), the TWAS (RG/PHYS/LA 12-011) and the Vice-Rectorate for Research of the Pontificia Universidad Catolica del Peru (VRI-0096) for financial support. References
Fig. 6. SEM image: panoramic view of longitudinal samples surface tested at humid condition.
[1] T. Ikoma, H. Kobayashi, J. Tanaka, D. Walsh, S. Mann, Microstructure, mechanical, and biomimetic properties of fish scales from Pagrus major, J. Struct. Biol. 142 (2003) 327–333. [2] F.G. Torres, O.P. Troncoso, J. Nakamatsu, C.J. Grande, C.M. Gómez, Characterization of the nanocomposite laminate structure occurring in fish scales from Arapaima gigas, Mater. Sci. Eng. C 28 (2008) 1276–1283. [3] R. Duan, J. Zhang, X. Du, X. Yao, K. Konno, Properties of collagen from skin, scale and bone of carp (Cyprinus carpio), Food Chem. 112 (2009) 702–706. [4] S. Sankar, S. Sekar, R. Mohan, Sunita Rani, J. Sundaraseelan, T.P. Sastry, Preparation and partial characterization of collagen sheet from fish (Lates calcarifer) scales, Int. J. Biol. Macromol. 42 (2008) 6–9. [5] R. Nadeem, T.M. Ansari, A.M. Khalid, Fourier transform infrared spectroscopic characterization and optimization of Pb(II) biosorption by fish (Labeo rohita) scales, J. Hazard. Mater. 156 (2008) 64–73. [6] P.G. Allison, M.Q. Chandler, R.I. Rodriguez, B.A. Williams, R.D. Moser, C.A. Weiss Jr., A.R. Poda, B.J. Lafferty, A.J. Kennedy, J.M. Seiter, W.D. Hodo, R.F. Cook, Mechanical properties and structure of the biological multilayered material system, Atractosteus spatula scales, Acta Biomater. 9 (2013) 5289–5296. [7] T. Ikoma, H. Kobayashi, J. Tanaka, D. Walsh, S. Mann, Physical properties of type I collagen extracted from fish scales of Pagrus major and Oreochromis niloticus, Int. J. Biol. Macromol. 32 (2003) 199–204. [8] Y.S. Lin, C.T. Wei, E.A. Olevsky, M.A. Meyers, Mechanical properties and the laminate structure of Arapaima gigas scales, J. Mech. Behav. Biomed. Mater. 4 (2011) 1145–1156. [9] F.G. Torres, E. Le Bourhis, O.P. Troncoso, J. Llamoza, Structure–property relationships in Arapaima gigas scales revealed by nanoindentation tests, Polym. Polym. Compos. 22 (2014) 369–374.
F.G. Torres et al. / Materials Science and Engineering C 51 (2015) 153–157 [10] B.J.F. Bruet, J. Song, M.C. Boyce, C. Ortiz, Materials design principles of ancient fish armour, Nat. Mater. 7 (2008) 748–756. [11] V. Kovan, An assessment of impact strength of the mandible, J. Biomech. 41 (2008) 3488–3491. [12] G.C. Reilly, J.D. Currey, The effects of damage and microcracking on the impact strength of bone, J. Biomech. 33 (2000) 337–343. [13] E.A. Zimmermann, B. Gludovatz, E. Schaible, N.K. Dave, W. Yang, M.A. Meyers, R.O. Ritchie, Mechanical adaptability of the Bouligand-type structure in natural dermal armour, Nat. Commun. 4 (2013) 2634. [14] W. Yang, V.R. Sherman, B. Gludovatz, M. Mackey, E.A. Zimmermann, E.H. Chang, E. Schaible, Z. Qin, M.J. Buehler, R.O. Ritchie, M.A. Meyers, Protective role of Arapaima gigas fish scales: structure and mechanical behavior, Acta Biomater. 10 (2014) 3599–3614.
157
[15] D. Zhu, L. Szewciw, F. Vernerey, F. Barthelat, Puncture resistance of the scaled skin from striped bass: collective mechanisms and inspiration for new flexible armor designs, J. Mech. Behav. Biomed. Mater. 24 (2013) 30–40. [16] L.R. Xu, A.J. Rosakis, Impact failure characteristics in sandwich structures: part I: basic failure mode selection, Int. J. Solids Struct. 39 (2002) 4215–4235. [17] V. Ziv, H.D. Wagner, S. Weiner, Microstructure–microhardness relations in parallelfibered and lamellar bone, Bone 18 (1996) 417–428. [18] G.I. Tseretely, O.I. Smirnova, DSC study of melting and glass transition in gelatins, J. Therm. Anal. 38 (1992) 1189–1201. [19] O.P. Troncoso, F.G. Torres, C.J. Grande, Characterization of the mechanical properties of tough biopolymer fibres from the mussel byssus of Aulacomya ater, Acta Biomater. 4 (2008) 1114–1117.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Short Communication
Impact and fracture analysis of fish scales from Arapaima gigas F.G. Torres ⁎, M. Malásquez, O.P. Troncoso Department of Mechanical Engineering, Pontificia Universidad Catolica del Peru, Av. Universitaria 1801, Lima 32, Peru
a r t i c l e
i n f o
Article history: Received 16 October 2014 Received in revised form 15 January 2015 Accepted 23 February 2015 Available online 25 February 2015 Keywords: Fish scale Arapaima gigas Impact behaviour
a b s t r a c t Fish scales from the Amazonian fish Arapaima gigas have been characterised to study their impact and fracture behaviour at three different environmental conditions. Scales were cut in two different directions to analyse the influence of the orientation of collagen layers. The energy absorbed during impact tests was measured for each sample and SEM images were taken after each test in order to analyse the failure mechanisms. The results showed that scales tested at cryogenic temperatures display fragile behaviour, while scales tested at room temperature did not fracture. Different failure mechanisms have been identified, analysed and compared with the failure modes that occur in bone. The impact energy obtained for fish scales was two to three times higher than the values reported for bone in the literature. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Fish scales are an example of collagen based biological tissues. They are plywood-like structures of closely packed collagen fibre layers reinforced with a mineral phase of calcium deficient hydroxyapatite [1]. They provide fish with a flexible and protective skin that allows for changes in shape. Recently, the interest in fish scales as a model material for the development of bioinspired composite materials has increased. Some of the studies reported in the literature include the characterisation of fish scales from Pagrus major [1], Arapaima gigas [2], Cyprinus carpio [3], Lates calcarifer [4], Labeo rohita [5] and Atractosteus spatula [6]. According to Ikoma et al. [7] type I collagen fibrils are the main organic component in fish scales. Scales are composed of several layers of parallel collagen fibre bundles, with varying fibre orientations in the different layers [8]. The mineral phase in fish scales is mainly calcium deficient hydroxyapatite (HAP). HAP forms plate-like nano-crystals that could act as a nano-reinforcement in a collagen matrix. Torres et al. [2] have reported that the mineral content of scales varies along their cross-section, which accounts for a harder outer surface and a more flexible inner layer. Most studies performed on fish scales have reported tensile tests results. Torres et al. [2] measured the maximum tensile strength and Young's modulus of dry and humid A. gigas scales, and found that the Young's modulus of dry scales was 1.38 GPa while the modulus of humid scales was 0.83 GPa. Lin et al. [8] have also performed tensile tests on A. gigas scales. They reported an elastic modulus of 1.2 GPa for ⁎ Corresponding author. E-mail address: [email protected] (F.G. Torres).
http://dx.doi.org/10.1016/j.msec.2015.02.034 0928-4931/© 2015 Elsevier B.V. All rights reserved.
dry scales and 0.1 GPa for hydrated scales. Ikoma et al. [1] have reported an elastic modulus of 2.2 GPa for scales of P. major. Indentation tests at different scales (macro-, micro- and nano-) have also been used to assess the mechanical properties of fish scales. Lin et al. [8] have reported an average hardness value in the internal layers of A. gigas scales of 200 MPa, increasing to about 550 MPa in the external layer. Indentation tests carried out by Torres et al. [9] showed that the hardness of the cross section of A. gigas scale follows a cyclic sawtooth shaped pattern. The higher hardness values correspond to the laminates where fibres are perpendicular to the indentation plane. Bruet et al. [10] found that the scales from Polypterus senegalus have hardness values that increase with distance from the inner surface to the outer surface, from 0.54 GPa to 4.5 GPa. Besides being hard and strong, fish scales are also a very tough material. In this paper we carry out impact tests with samples at three different conditions to determine their response with regard to impact loads. Considering that fish scales display a laminate type structure, the experiments conducted under impact conditions have been analysed to determine whether fish scales behave in a similar way to standard fibre reinforced polymer composites. Failure analysis using scanning electron microscopy was also used to complement this analysis.
2. Experimental part 2.1. Materials Scales from the Amazonian fish A. gigas (body weight 100–150 kg) were washed and stored in standard conditions (20 °C and 80% of
154
F.G. Torres et al. / Materials Science and Engineering C 51 (2015) 153–157
2.4. Scanning electron microscopy (SEM) The fracture surfaces from impact testing specimens were studied using a FEI — QUANTA 200 SEM (Hillsboro, OR) in low vacuum with a voltage of 30 kV and a working distance in the range of 9.9–11.2 mm. The specimens were mounted onto metal stubs and were viewed in the SEM without any coating. 3. Results and discussion
Fig. 1. Sample directions: longitudinal (1) and transverse (2).
Table 1 Impact results from transverse and longitudinal samples. Condition
Direction
Energy absorbed (kJ/m2)
Dry sample/room temperature Dry sample/room temperature Dry sample/cryogenic temperature Dry sample/cryogenic temperature Humid sample/room temperature Humid sample/room temperature
Transverse Longitudinal Transverse Longitudinal Transverse Longitudinal
42.49 ± 6.55 35.11 ± 3.07 25.71 ± 4.41 14.32 ± 2.13 17.08 ± 3.65 11.92 ± 0.58
relative humidity). They were around 70–75 mm in length and around 1 mm thick.
2.2. Sample preparation Rectangular samples were cut off from A. gigas scales in two different directions. Fig. 1 shows a schematic view of the longitudinal and transverse directions. The samples measured 40 mm in length (L), 8 mm in width (W) and 1.5 mm in thickness, which is in agreement with the L/W relation recommended in the ASTM 4812 standard. The edges of each sample were sanded to avoid possible residual stresses.
2.3. Impact tests Charpy impact tests were performed in a CEAST Resil Impactor (Norwood, MA) with un-notched specimens. The maximum impact energy capacity of the machine was 7.5 J. The anvil gap was 22 mm. Three different conditions were used in this study (Table 1). Dry samples were tested at room temperature and at cryogenic conditions whereas humid samples were tested at room temperature. Humid samples were prepared by soaking dry specimens in distilled water 24 h before the tests. For cryogenic conditions, the specimens were kept in liquid nitrogen for 10 min and immediately after mounted onto the anvil of the impact testing equipment. A minimum of three samples were tested for each condition.
Impact tests results are depicted in Table 1. Three types of fracture were observed in the tests: a) full fracture; b) major fracture with tie; c) minor fracture (most of the cross section was still intact). The samples tested at room temperature did not undergo full fracture, as indicated by the total separation of the two halves of the sample. Fig. 2a and b shows representative longitudinal and transverse samples after impact tests at room temperature. They suffered minor fractures and showed no visible external change of shape. By contrast, samples tested at cryogenic conditions underwent full and major fractures. Fig. 3a and b shows representative samples with major fractures. The two parts of the specimens are held together only by the external layers. In Fig. 3c and d we can see samples which underwent full fracture. In this case total separation of the specimens can be observed. The impact energy values measured in this study (Table 1) are two to three times higher than those of other collagen based structures reported in the literature. Kovan [11] has reported the impact energy of the mandible bone, with values in the range of 8–20 kJ/m2. Reilly and Currey [12] reported impact energy values of 6–24 kJ/m2 for bovine tibia and humerus samples. In all cases the values reported were obtained at room temperature. The results in Table 1 show that the impact energy was lower for the samples tested at cryogenic conditions in both directions. Moreover, transverse samples at cryogenic conditions absorbed more energy than longitudinal samples at the same conditions. The energy absorbed during impact was around 65% higher for the transverse samples. At cryogenic conditions the molecular mobility of the collagen helices decreases with regard to room temperature. Specimens with reduced molecular mobility find it difficult to deform plastically at high velocity loading, as it occurs during impact. In addition, fracture mechanisms typical of brittle fracture also take place and further decrease the overall impact energy. Specimens cut in what we have defined as the transverse direction of the scales (Fig. 1), have shown higher impact energy (21.02% higher at room temperature and 79.54% higher for cryogenic tests) with regard to longitudinal samples. Several reports in the literature have described fish scales as plywood like structures formed by collagen lamellae arranged in layers, the lamellae being assembled from mineralized collagen fibrils [8]. Other authors have reported a Bouligand type (twisted plywood) structure [13,14]. It is reasonable to think that a Bouligand type arrangement ensures that the scale has in-plane isotropy in mechanical response [8]. However, results reported in the literature are contradictory. Yang et al. [14] found that the tensile strength of the scales is 50% higher in the longitudinal than in the transverse direction
Fig. 2. Representative longitudinal (a) and transverse (b) samples after impact tests at room temperature.
F.G. Torres et al. / Materials Science and Engineering C 51 (2015) 153–157
155
Fig. 3. Types of fractures observed after impact tests at cryogenic conditions: partial fracture of longitudinal sample (a), partial fracture of transverse sample (b), total fracture of longitudinal sample (c) and total fracture of transverse sample (d).
Fig. 4. SEM image: panoramic view of longitudinal samples complete fracture under cryogenic condition. (a) Mark zone indicates where brittle fracture occurred. (b) Marked zones indicate where fibre breakage occurred (circle) and where brittle fracture occurred (rectangle).
whereas Zhu et al. [15] reported that specimens with an orientation of 45° or 90° to the longitudinal axis have higher strengths (60 MPa) than the 0° orientation (40 MPa). The impact tests performed in this study have confirmed the anisotropic behaviour of fish scales as the transverse direction is tougher than the longitudinal direction. So far it is not clear if this anisotropy could be attributed to the fibre orientation or to any failure mechanism such as delamination. The main failure mechanisms observed in the SEM micrographs are described below. The fracture surface depicted in Fig. 4a corresponds to a specimen that suffered complete fracture. Here we can observe the
rough external layer of the scale and the internal layers formed from collagen fibres. The fracture surface shows the different orientations of collagen fibres bundles parallel to each other. In the same figure, the rectangle shows the region where brittle fracture occurred across the entire fracture surface. Fig. 4b is a detail view of the same sample shown in Fig. 4a. It shows the fracture of individual collagen fibres. Tearing between individual fibres can also be observed on the marked rectangle. Fibre breakage takes place away from the crack plane after fibre fracture occurred. On the right side of the figure, the ellipse indicates the region where brittle fracture occurred in the sample.
Fig. 5. SEM image: panoramic view of longitudinal samples partial fracture tested at cryogenic conditions. (a) Mark zones where fibre delamination and crack propagation occurred. (b) Mark zone indicates where tearing occurred.
156
F.G. Torres et al. / Materials Science and Engineering C 51 (2015) 153–157
In Fig. 5a and b we can observe the partial fracture of a longitudinal sample. The film which binds the two pieces is formed by collagen fibrils. Fig. 5a shows how delamination between collagen fibres occurred (rectangles). This failure mechanism starts before the scale fractures, and decreases the maximum fracture energy. After delamination occurs, the laminate structure suffers crack propagation on the zones where delamination is visible. In Fig. 5b, the marked rectangle shows that tearing between collagen fibres was also present in longitudinal samples. Fish scales display much higher impact energy compared to bone, at least 2 to 3 times higher. Both, bone and fish scales are made out of collagen and hydroxyapatite. The higher impact energy in fish scales compared to bone, both dense and porous, is mainly due to the fact that fish scales are built in a similar way to a laminate fibre reinforced composite material. The failure mechanisms observed in the fish scales samples that have undergone impact show some correspondence with those mechanisms that occur in fibre reinforced composites [16]. These include fibre breakage, delamination and matrix crack propagation. The higher content of collagen observed in the internal layers of fish scales (83% [10]) compared with 50–55% [17] of lamellar bone accounts for a more ductile behaviour in fish scales when subjected to impact. It is important to notice that such an increase in toughness does not compromise strength [11,12]. In the wet state this advantage is even more noticeable, since water acts as a plasticizing agent in collagen increasing its free volume [18], thus promoting a more rubbery behaviour under mechanical deformation. This effect has been verified for instance for mussel byssus, which are tough fibres formed mainly by collagen [19]. Figs. 6 and 7 show SEM micrographs obtained from a wet specimen that had undergone an impact test. In both figures, it can be easily observed that none of the previously described failure mechanisms, such as delamination, fibre breakage and crack propagation, have occurred. 4. Conclusions We have studied the impact behaviour of the scales from A. gigas. The results showed that dry and humid scales tested at room temperature did not undergo full fracture (total separation of two halves of the samples). By contrast, samples tested at cryogenic conditions did undergo major or full fracture, displaying different failure mechanisms such as delamination, fibre breakage, crack propagation and brittle fracture. All samples tested at low temperatures displayed fragile behaviour. With regard to the orientation of the specimens within the scales, the tests
Fig. 7. SEM image: panoramic view of transverse samples surface tested at humid condition.
performed in this study showed that for all the environmental conditions assessed, transverse samples absorbed higher levels of energy during impact than longitudinal samples. We have compared the impact energy from tests in this study with the impact energy reported for bone in the literature, which is another structure made out of collagen and hydroxyapatite. The impact energy obtained from the tests reported here for fish scales was two to three times higher compared to the impact energy reported for bone in the literature. This could be due to the fact that fish scales have a structure similar to a laminate fibre reinforced composite material, which improves the dissipation of energy during impact. Acknowledgements The authors would like to thank the Peruvian Science and Technology Program (159-FINCyT-IB-2013), the TWAS (RG/PHYS/LA 12-011) and the Vice-Rectorate for Research of the Pontificia Universidad Catolica del Peru (VRI-0096) for financial support. References
Fig. 6. SEM image: panoramic view of longitudinal samples surface tested at humid condition.
[1] T. Ikoma, H. Kobayashi, J. Tanaka, D. Walsh, S. Mann, Microstructure, mechanical, and biomimetic properties of fish scales from Pagrus major, J. Struct. Biol. 142 (2003) 327–333. [2] F.G. Torres, O.P. Troncoso, J. Nakamatsu, C.J. Grande, C.M. Gómez, Characterization of the nanocomposite laminate structure occurring in fish scales from Arapaima gigas, Mater. Sci. Eng. C 28 (2008) 1276–1283. [3] R. Duan, J. Zhang, X. Du, X. Yao, K. Konno, Properties of collagen from skin, scale and bone of carp (Cyprinus carpio), Food Chem. 112 (2009) 702–706. [4] S. Sankar, S. Sekar, R. Mohan, Sunita Rani, J. Sundaraseelan, T.P. Sastry, Preparation and partial characterization of collagen sheet from fish (Lates calcarifer) scales, Int. J. Biol. Macromol. 42 (2008) 6–9. [5] R. Nadeem, T.M. Ansari, A.M. Khalid, Fourier transform infrared spectroscopic characterization and optimization of Pb(II) biosorption by fish (Labeo rohita) scales, J. Hazard. Mater. 156 (2008) 64–73. [6] P.G. Allison, M.Q. Chandler, R.I. Rodriguez, B.A. Williams, R.D. Moser, C.A. Weiss Jr., A.R. Poda, B.J. Lafferty, A.J. Kennedy, J.M. Seiter, W.D. Hodo, R.F. Cook, Mechanical properties and structure of the biological multilayered material system, Atractosteus spatula scales, Acta Biomater. 9 (2013) 5289–5296. [7] T. Ikoma, H. Kobayashi, J. Tanaka, D. Walsh, S. Mann, Physical properties of type I collagen extracted from fish scales of Pagrus major and Oreochromis niloticus, Int. J. Biol. Macromol. 32 (2003) 199–204. [8] Y.S. Lin, C.T. Wei, E.A. Olevsky, M.A. Meyers, Mechanical properties and the laminate structure of Arapaima gigas scales, J. Mech. Behav. Biomed. Mater. 4 (2011) 1145–1156. [9] F.G. Torres, E. Le Bourhis, O.P. Troncoso, J. Llamoza, Structure–property relationships in Arapaima gigas scales revealed by nanoindentation tests, Polym. Polym. Compos. 22 (2014) 369–374.
F.G. Torres et al. / Materials Science and Engineering C 51 (2015) 153–157 [10] B.J.F. Bruet, J. Song, M.C. Boyce, C. Ortiz, Materials design principles of ancient fish armour, Nat. Mater. 7 (2008) 748–756. [11] V. Kovan, An assessment of impact strength of the mandible, J. Biomech. 41 (2008) 3488–3491. [12] G.C. Reilly, J.D. Currey, The effects of damage and microcracking on the impact strength of bone, J. Biomech. 33 (2000) 337–343. [13] E.A. Zimmermann, B. Gludovatz, E. Schaible, N.K. Dave, W. Yang, M.A. Meyers, R.O. Ritchie, Mechanical adaptability of the Bouligand-type structure in natural dermal armour, Nat. Commun. 4 (2013) 2634. [14] W. Yang, V.R. Sherman, B. Gludovatz, M. Mackey, E.A. Zimmermann, E.H. Chang, E. Schaible, Z. Qin, M.J. Buehler, R.O. Ritchie, M.A. Meyers, Protective role of Arapaima gigas fish scales: structure and mechanical behavior, Acta Biomater. 10 (2014) 3599–3614.
157
[15] D. Zhu, L. Szewciw, F. Vernerey, F. Barthelat, Puncture resistance of the scaled skin from striped bass: collective mechanisms and inspiration for new flexible armor designs, J. Mech. Behav. Biomed. Mater. 24 (2013) 30–40. [16] L.R. Xu, A.J. Rosakis, Impact failure characteristics in sandwich structures: part I: basic failure mode selection, Int. J. Solids Struct. 39 (2002) 4215–4235. [17] V. Ziv, H.D. Wagner, S. Weiner, Microstructure–microhardness relations in parallelfibered and lamellar bone, Bone 18 (1996) 417–428. [18] G.I. Tseretely, O.I. Smirnova, DSC study of melting and glass transition in gelatins, J. Therm. Anal. 38 (1992) 1189–1201. [19] O.P. Troncoso, F.G. Torres, C.J. Grande, Characterization of the mechanical properties of tough biopolymer fibres from the mussel byssus of Aulacomya ater, Acta Biomater. 4 (2008) 1114–1117.