- Email: [email protected]
The mechanical properties and microstructure of the bionic alloy–ceramic laminated composite
Materials and Design 33 (2012) 300–305
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
The mechanical properties and microstructure of the bionic alloy–ceramic laminated composite Guodong Shi a, Zhanjun Wu a, Zhi Wang a,⇑, Jun Liang b a School of Aeronautics and Astronautics, Faculty of Vehicle Engineering and Mechanics, State Key Laboratory of Structural Analysis for Industrial Equipment, Dalian University of Technology, Dalian 116024, PR China b Center for Composite Materials, Harbin Institute of Technology, Harbin 150001, PR China
a r t i c l e
i n f o
Article history: Received 11 June 2011 Accepted 19 July 2011 Available online 26 July 2011 Keywords: B. Laminates E. Mechanical properties F. Microstructure
a b s t r a c t In the present work, the bionic alloy–ceramic laminated composite was fabricated by electron beam– physical vapor deposition method. The ingots of Ni–20Co–12Cr–4Al (wt.%) and ZrO2–8 mol%Y2O3 were used as the sources of the alloy layer and ceramic layer, respectively. The laminated composite was generally destroyed within the ceramic layer when the interlaminar strength was determined, which revealed that the excellent interface bonding between the ceramic layer and the alloy layer. The obvious diffusion interfaces between the ceramic and alloy layers were readily detected, which was favorable to the mechanical properties of the laminated composite. In the heat treatment process, the diffusion of the flaws within the ceramic layer and/or alloy layer to the interface between the ceramic layer and alloy layer was easier compared with the occurrence of interlaminar diffusion. It was confirmed by the X-ray diffractometer that the reaction of the ceramic layer with alloy layer was simple physical diffusion. The tensile strength of the laminated composite increased first and then decreased as the heat treatment time increased, which was attributed to the mutual reaction of the increase in the relative density with the formation of the flaws located at the interface. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction A major problem in the service of ceramics as structural materials is their low toughness [1]. Even though many attempts have been used to increase their toughness, including incorporation of fibers [2], whiskers [3] or particles reinforcements [4], and ZrO2 phase transformation reinforcing and so on [5], up to date the brittleness of ceramics has not been overcome in nature. In the research on the structure of natural biomaterials, such as turtleback, shell and nacres, it has been found that these natural biomaterials have very reasonable laminated structures which give them many excellent properties, such as good carrying capacity, good toughness and so on [6–9]. Over the past two decades, bionics has had a profound influence on the material science and engineering, because the unique structures, compositions and correspondingly excellent properties of biology gave researchers many clues to improve the properties of materials or increase the reliability of structural components [10]. One way of preventing properties of ceramic component is to design a structure with dense and strong layers separated by weak or soft layers of the different materials [11]. The weak interface in these laminated composites serves to deflect the propagating crack and reduces its stress ⇑ Corresponding author. Tel./fax: +86 411 84706791. E-mail address: [email protected] (Z. Wang). 0261-3069/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2011.07.049
intensity. Another way of increasing reliability is the use of intermittent interlayer [12]. One of the requirements for achieving high properties of the laminated composites is the good chemical and physical bonding between the interface materials. Electron beam–physical vapor deposition (EB–PVD) is one of the advanced techniques for the fabrication of the thin coatings [13]. In the late 1980s, EB–PVD techniques was being used to deposit the top coat onto rotating blades [14]. This deposition method resulted in a columnar microstructure with the columns running perpendicular to the work piece surface, which imparted a high degree of strain tolerance making them ideal for applications involving frequent thermal cycles [15]. In the present work, the bionic alloy–ceramic laminated composite was fabricated by EB–PVD. The ingots of Ni–20Co–12Cr– 4Al (wt.%) and ZrO2–8 mol%Y2O3 were used as the sources of the alloy layer and ceramic layer, respectively. The mechanical properties and microstructure of the alloy–ceramic laminated composite were investigated in detail. Furthermore, the laminated composite was heat treated in order to improve the tensile strength through reducing residual stress. 2. Experimental procedure During deposition process, the stainless steel substrate with the diameter of 1 m and surface roughness of 1.0 rotated at 6 rpm
301
G. Shi et al. / Materials and Design 33 (2012) 300–305
around the vertical axis and the substrate temperature was maintained at 650 °C approximately. The process pressure was in the range of 6–10 103 Pa. The ingots of Ni–20Co–12Cr–4Al (wt.%) and ZrO2–8 mol%Y2O3 were used as the sources of the alloy layer and ceramic layer, respectively. The ingots of ZrO2–8 mol%Y2O3 and Ni–20Co–12Cr–4Al (wt.%) were evaporated alternately to produce the alloy–ceramic laminated composite and the thicknesses of the different layers were controlled by their deposition time. The tensile strength tests were carried out by an INSTRON-5569 universal materials testing machine with a crosshead displacement speed of 0.05 mm/min at room temperature. Nanoindentation was performed using a Hysitron TriboIndenter (Hysitron Inc., Minneapolis, MN) with a 100 lm conospherical diamond fluid tip. The laminated composite was heat treated at 1050 °C for different times in Ar atmosphere, and heating rate of 10 °C/min and cooling rate 5 °C/min was lower than the heating cooling rates of the laminated composite in the fabrication process, respectively. The indenter tip is connected to the transducer, which was displacement-controlled, and data were presented as load versus displacement plots for loading, hold, and unloading. For all samples, indentation started with the nanoindenter tip approaching the sample from off contact. In this work, five specimens were tested to get an average value. The microstructures of the specimens were examined using SEM (FEI QUANTA200). Moreover, the phase composition of the laminated composite was determined by X-ray diffractometer (Philips, X’Pert-MRD). The interlaminar strength was evaluated according to GB-T3903.8-2005.
3. Results and discussions 3.1. Microstructure The laminated composite was generally destroyed along the ceramic layer, as shown in Fig. 1A, when the interlaminar strength was determined, which revealed that the ceramic layer appeared to be well bonded with the alloy layer. The outstanding interface bonding was favorable to the mechanical properties of the laminated composite. It could be seen from Fig. 1A that 8 mol%Y2O3 stabilized ZrO2 particles appeared to be uniformly deposited on the alloy layer. Furthermore, the flaws, such as pore and pit, were not observed in the ceramic layer. Fig. 1B shows the SEM image of the fractured surface perpendicular to the alloy–ceramic interface of the laminated composite. The ceramic layer of uniform thickness was separated by the alloy layer of uniform thickness. Fig. 2 shows the SEM images of the polished cross section of the laminated composite before and after the heat treatment. No flaws, such as pore and crack, were observed in the ceramic layer of the polished cross section of the laminated composite before and after the heat treatment, which also revealed that ceramic particles
appeared to be uniformly deposited on the alloy layer. The obvious diffusion interfaces between the ceramic layer and alloy layer were readily detected, which also indicated the excellent interface bonding between the ceramic layer and the alloy layer. As heat treatment time increased, the obvious flaws such as pit and pore occurred in the interface between the ceramic and alloy layers and the amount of the flaws increased gradually. The formation of the flaws was attributed to the diffusion of the particles during high temperature heat treatment [16]. In the heat treatment process, the diffusion of the flaws within the ceramic layer and/or alloy layer to the interface between the ceramic layer and alloy layer was easier compared with the occurrence of the interlaminar diffusion [17]. In the heat treatment process, the diffusion of the flaws within the ceramic layer and/or alloy layer to the interface between the ceramic layer and alloy layer was favorable to improve the relative density of the ceramic layer and/or alloy layer, whereas was unfavorable to the interface bonding between the ceramic layer and alloy layer. Fig. 3 shows the XRD spectra obtained from the polished cross section of the laminated composite before and after the heat treatment. Apparently, the phase analysis indicated the predominant phases for the laminated composite were t-ZrO2 for the ceramic layer and c-Ni phase for the alloy layer. Compared with the XRD spectra for the polished cross section of the laminated composite before the heat treatment, the obvious change in the XRD spectra was not detected as the heat treatment time increased. Such result indicated that the reaction of the ceramic layer with alloy layer was simple physical diffusion. The SEM images of the fractured surfaces of the tensile specimen for the laminated composite are shown in Fig. 4. The staged fracture mode resulted from the crack deflection prior to final failure of the tensile specimen were readily observed. Such fracture mode would dissipate a lot of energy and improve the mechanical properties of the laminated composite [3]. Many flat fracture regions resulted from the brittle fracture mode in the fractured surfaces of the tensile specimen for the laminated composite were also detected in the fractured surfaces, as shown in Fig. 4B. The obvious plastic deformation was not observed in the flat fracture region, which was ascribed to the weak grain boundary bonding of the alloy particles [18]. Furthermore, the fractured surface of the alloy layer was blade-like shrank, which indicated the ductile fracture of the laminated composite. The interface debonding phenomenon was easily observed in the ductile fracture region of the alloy layer, whereas the interface debonding did not occur in the brittle fractured region, which revealed that the ceramic layer was well bonded with the alloy layer in the brittle fractured region. The columnar crystals with sharp edges and corners were observed in ceramic layer, which were the typical characteristics of brittle intergranular fracture [19], as shown in Fig. 4C. The ceramic layer was fractured to many fragments due to the crack initiation and
B
A
2µm
50µm
Fig. 1. The SEM images of the fractured surface parallel (A) and perpendicular (B) to the alloy–ceramic interface of the laminated composite.
302
G. Shi et al. / Materials and Design 33 (2012) 300–305
A
B Alloy layer
Ceramic layer 5µm
5µm
Fig. 2. The SEM images of the polished cross section of the laminated composite before (A) and after (B) the heat treatment for 4 h.
Fig. 3. The XRD spectra obtained from the polished cross section of the laminated composite before and after the heat treatment.
propagation. The separation of the ceramic layers in the fractured surface was due to the plastic deformation of the alloy layer and the interface debonding, which indicated that the interlaminar strength was less than the strength of the ceramic layer or alloy layer. The ceramic layer of columnar structure has a good strain tolerance and the ceramic layer with the alloy layer could be bent and deformed, so that it could greatly reduce the residual stress caused by the differences of the physical properties in the alloy and ceramic layers. The SEM images of the fractured surfaces for the specimen after the heat treatment are shown in Fig. 5B–D. In order to investigate the change in the microstructure, SEM micrograph of the fractured surfaces for the specimen before the heat treatment (has been described in Fig. 4A) is shown in Fig. 5A for comparison. It could be seen from Fig. 5 that the brittle fracture region gradually reduced as the heat treatment time increased because the bonding strength of the grain boundary in the alloy layer was improved by the atom diffusion and grain boundary migration process. These diffusions were favorable to improve the relative density of the ceramic layer and alloy layer, whereas it would reduce the interface bonding strength due to the diffusion of the flaws to the interface between the ceramic layer and alloy layer [16,17]. Compared with the
A
50µm
C
B
15µm
15µm
Fig. 4. SEM images of the fractured surfaces of the tensile specimen for the laminated composite.
303
G. Shi et al. / Materials and Design 33 (2012) 300–305
B
A
50µm
50µm
D
C
50µm
50µm
Fig. 5. SEM images of the fractured surfaces for the specimen before (A) and after the heat treatment for 0.5 h (B), 2 h (C) and 4 h (D).
Fig. 6. The tensile strength and interminar strength of the laminated composite before and after the heat treatment.
fractured surface of the specimen before the heat treatment, the obvious separation of the interface was observed in the fractured surface of the specimen for the heat treatment for 0.5 h and the amount of the interface separation further increased, whereas the width of the interface separation decreased with increasing heat treatment time to 2 h. As the heat treatment time increased to 4 h, the amount of the interface separation decreased, whereas the width of the interface separation increased, which was attributed to the weakness in the interface bonding due to the diffusion of the flaws to the interface between the ceramic layer and alloy layer.
the tensile strength decreased slightly with increasing the heat treatment time to 4 h. However, the tensile strength for the heat treatment of 4 h was still greater than that for the laminated composite before the heat treatment, which was attributed to the reduction of the residual stress in the laminated composite. The tensile strength of the laminated composite was mostly dependent on the strength of the alloy layer because the incompact bonding of the ceramic particles in the ceramic layer, as shown in Fig. 1A. The interlaminar strength with increasing heat treatment time is also shown in Fig. 6. As the heat treatment time increased, the interlaminar strength decreased gradually because the increase in the heat treatment time was favorable to the diffusion of the flaws the interface between the ceramic layer and alloy layer. In the last decade the nanoindentation proposed by Pharr et al. has evolved as a promising tool for the structure–property characterization of tissues and biomaterials [20]. In order to confirm the tensile strength of the laminated composite dependent on the strength of the alloy layer and clarify the effect of the heat treatment time on the strength of the alloy layer, the yield strength of the alloy layer was calculated by the nanoindentation techniques. In the nanoindentation measurement process, the total work, Wt, released by the indenter could be divided into the elastic work and plastic work, and Giannakopoulos and Suresh [21] has proposed as:
We hr ¼1 Wt hmax
ð1Þ
where hmax and hr are the maximum penetration depth of the indenter into the specimen and the residual depth (after complete unloading), respectively, We is the elastic work. The hmax and hr could be determined from the nanoindentation load–displacement curves, and the load–displacement relationship was calculated as:
3.2. Mechanical properties 2
Fig. 6 shows the tensile strength of the laminated composite before and after the heat treatments. As the heat treatment time increased to 2 h, the tensile strength increased gradually, whereas
Pmax ¼ C hmax
ð2Þ
where C and Pmax are the curvature of the indentation and maximum load, respectively. The literature has confirmed that the
304
G. Shi et al. / Materials and Design 33 (2012) 300–305 Table 1 The mechanical properties of the alloy layer in the laminated composite treated for different time.
Fig. 7. The nanoindentation load–displacement curves of the alloy layer in the laminated composite treated for different times.
curvature of the indentation for the elastic–plastic materials is dependent on the strength of the materials as follows [22]:
C¼
Pmax 2 hmax
¼ M1 r0:29 1 þ
ry r0:29
E ln þ M2
ry
ð3Þ
where ry and r0.29 the yield strength determined at uniaxial compression and the stress corresponding to the plastic strain of 0.29, respectively. The constant M1 and M2 are 6.618 and 0.875 for Pharr’s indenter [22]. In the indentation process, the ratio of the residual depth to the maximum penetration depth was to characterize the degree of material deformation as follows:
r0:29 ry 0:29E
¼ 1 0:142
2 hr hr 0:957 hmax hmax
ð4Þ
The nanoindentation load–displacement curves are shown in Fig. 7 and the yield strength of the alloy layer was calculated and is listed in Table 1. Young’s modulus did not show change as the heat treatment time increased. In contrast to the modulus, the hardness and yield strength increased first and then decreased as the heat treatment time increased. In the heat treatment process, the flaws of the smaller size could diffuse from the alloy layer to the interface, whereas the flaws of the larger size could not be eliminated, which resulted in the reduction of the tensile strength relative to the yield strength measured by the nanoindentation techniques. However, the yield strength measured by the nanoindentation techniques has change trend similar to that of tensile strength of the laminated composite as the heat treatment time increased, which confirmed that the tensile strength of the laminated composite was mostly dependent on the strength of the alloy layer. The increase in the density of the alloy layer with the increasing heat treatment time was favorable to the increase in the strength of the alloy layer; but on the other hand, the decrease in the interface bonding strength would lead to the reduction in the effect of Orowan strengthening on the alloy layer. The flaws such as pit and pore generally occurred in the materials fabricated by EB– PVD [13–15], which led to the reduction of the relative density. The hardness and strength of the materials was reduced by the low relative density and the flaws located at the interface. The brittle fracture of the alloy layer was activated by the flaws. The amount of the flaws in the alloy layer was reduced in the heat treatment process with the increasing heat treatment time, which was favorable to improve the strength of the alloy layer because the proportion of the brittle fracture of the alloy layer was reduced. Furthermore, the mechanical properties of the laminated composite were also affected by the interface bonding strength because
Heat treatment time (h)
0
0.5
2
4
Modulus (GPa) Hardness (GPa) Yield strength (GPa)
243 5.56 85.7
241 5.65 87.2
240 5.86 89.8
244 5.81 86.9
the laminated composite included a lot of the alloy–ceramic interface. As the heat treatment time increased, the amount of the flaws located at the interface increased gradually, which would reduce the mechanical properties of the laminated composite. Therefore, the tensile strength of the laminated composite increased first and then decreased as the heat treatment time increased, which was attributed to the mutual reaction of the increase in the relative density with the formation of the flaws located at the interface. Further work is continuing to investigate and report the effect of the laminated structure on the thermophysical properties of the bionic alloy–ceramic laminated composite. 4. Conclusions In the present work, the bionic alloy–ceramic laminated composite was fabricated using the ingots of Ni–20Co–12Cr–4Al (wt.%) and ZrO2–8 mol%Y2O3 as the sources of the alloy layer and ceramic layer, respectively. The mechanical properties and microstructure of the laminated composite were investigated in detail. The laminated composite was generally destroyed along the ceramic layer when the interlaminar strength was determined, which revealed that the ceramic layer appeared to be well bonded with the alloy layer. The flaws, such as pore and crack, were not observed in the polished cross section of the laminated composite. Compared with the fractured surface feature of the specimen before the heat treatment, the obvious separation was observed in the fractured surface of the specimen after the heat treatment, which was attributed to the weakness in the interface bonding due to the diffusion of the flaws to the interface between the ceramic layer and alloy layer. The tensile strength of the laminated composite increased first and then decreased as the heat treatment time increased, which was attributed to the mutual reaction of the increase in the relative density with the formation of the flaws located at the interface. Acknowledgments This work was supported by China Postdoctoral Science Foundation Funded Project (20100481220) and the Fundamental Research Funds for the Central Universities (3014-852001 and DUT10ZDG05) and the National Natural Science Foundation of China (51002019 and 91016024). References [1] Wang Z, Hong CQ, Zhang XH, Sun X, Han JC. Microstructure and thermal shock behavior of ZrB2–SiC–graphite composite. Mater Chem Phys 2009;113:338–41. [2] Lee KS, Jang KS, Park JH, Kim TW, Han IS, Woo SK. Designing the fiber volume ratio in SiC fiber-reinforced SiC ceramic composites under Hertzian stress. Mater Des 2011;32:4394–401. [3] Wu P, Zheng Y, Zhao YL, Yu HZ. Effect of SiC whisker addition on the microstructures and mechanical properties of Ti(C, N)-based cermets. Mater Des 2011;32:951–6. [4] Meng SH, Chen HB, Hu JH, Wang ZW. Radiative properties characterization of ZrB2–SiC-based ultrahigh temperature ceramic at high temperature. Mater Des 2011;32:377–81. [5] Zhu T, Li WJ, Zhang XH, Hu P, Hong CQ, Weng L. Damage tolerance and R-curve behavior of ZrB2–ZrO2 composites. Mater Sci Eng A 2009;516:297–301.
G. Shi et al. / Materials and Design 33 (2012) 300–305 [6] Chen B, Peng X, Wang JG, Wu X. Laminated microstructure of Bivalva shell and research of biomimetic ceramic/polymer composite. Ceram Int 2004;30:2011–4. [7] Chen RF, Wang CA, Huang Y, Le HR. An efficient biomimetic process for fabrication of artificial nacre with ordered-nanostructure. Mater Sci Eng C 2008;28:218–22. [8] Mirzababaee M, Tahani M. Accurate determination of coupling effects on free edge interlaminar stresses in piezoelectric laminated plates. Mater Des 2009;30:2963–74. [9] Fu YM, Zhang P, Yang F. Interlaminar stress distribution of composite laminated plates with functionally graded fiber volume fraction. Mater Des 2010;31:2904–15. [10] Pinto Correia IF, Mota Soares CM, Mota Soares CA, Herskovits J. Analysis of adaptive shell structures using a refined laminated model. Compos Struct 2004;66:261–8. [11] Almeida FS, Awruch AM. Design optimization of composite laminated structures using genetic algorithms and finite element analysis. Compos Struct 2009;88:443–54. [12] Krstic Z, Krstic VD. Young’s modulus, density and phase composition of pressureless sintered self-sealed Si3N4/BN laminated structures. J Eur Ceram Soc 2008;28:1723–30. [13] Schulz U, Schmücker M. Microstructure of ZrO2 thermal barrier coatings applied by EB–PVD. Mater Sci Eng A 2000;276:1–8.
305
[14] Wellman RG, Nicholls JR. Erosion, corrosion and erosion–corrosion of EB PVD thermal barrier coatings. Tribol Int 2008;41:657–62. [15] Peng H, Guo HB, He J, Gong SK. Oxidation and diffusion barrier behaviors of double-layer NiCoCrAlY coatings produced by plasma activated EB–PVD. Surf Coat Technol 2011;205:4658–64. [16] Haynes JA, Zhang Y, Cooley KM, Walker L, Reeves KS, Pint BA. Hightemperature diffusion barriers for protective coatings. Surf Coat Technol 2004;188–189:153–7. [17] Mulligan C, Blanchet TA, Gall D. Control of lubricant transport by a CrN diffusion barrier layer during high-temperature sliding of a CrN–Ag composite coating. Surf Coat Technol 2010;205:1350–5. [18] Sciti D, Bellosi A, Esposito L. Bonding of zirconia to super alloy with the active brazing technique. J Eur Ceram Soc 2001;21:45–52. [19] Kameda J, Nishiyama Y. Combined effects of phosphorus segregation and partial intergranular fracture on the ductile–brittle transition temperature in structural alloy steels. Mater Sci Eng A 2011;528:3705–13. [20] Pharr GM, Oliver WC, Brotzen FR. On the generality of the relationship among contact stiffness, contact area, and elastic modulus during indentation. J Mater Res 1992;7:613–7. [21] Giannakopoulos AE, Suresh S. Determination of elastoplastic properties by instrumented sharp indentation. Scripta Mater 1999;40:1191–8. [22] Suresh S, Giannakopoulos AE. A new method for estimating residual stresses by instrumented sharp indentation. Acta Mater 1998;46:5755–67.
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
The mechanical properties and microstructure of the bionic alloy–ceramic laminated composite Guodong Shi a, Zhanjun Wu a, Zhi Wang a,⇑, Jun Liang b a School of Aeronautics and Astronautics, Faculty of Vehicle Engineering and Mechanics, State Key Laboratory of Structural Analysis for Industrial Equipment, Dalian University of Technology, Dalian 116024, PR China b Center for Composite Materials, Harbin Institute of Technology, Harbin 150001, PR China
a r t i c l e
i n f o
Article history: Received 11 June 2011 Accepted 19 July 2011 Available online 26 July 2011 Keywords: B. Laminates E. Mechanical properties F. Microstructure
a b s t r a c t In the present work, the bionic alloy–ceramic laminated composite was fabricated by electron beam– physical vapor deposition method. The ingots of Ni–20Co–12Cr–4Al (wt.%) and ZrO2–8 mol%Y2O3 were used as the sources of the alloy layer and ceramic layer, respectively. The laminated composite was generally destroyed within the ceramic layer when the interlaminar strength was determined, which revealed that the excellent interface bonding between the ceramic layer and the alloy layer. The obvious diffusion interfaces between the ceramic and alloy layers were readily detected, which was favorable to the mechanical properties of the laminated composite. In the heat treatment process, the diffusion of the flaws within the ceramic layer and/or alloy layer to the interface between the ceramic layer and alloy layer was easier compared with the occurrence of interlaminar diffusion. It was confirmed by the X-ray diffractometer that the reaction of the ceramic layer with alloy layer was simple physical diffusion. The tensile strength of the laminated composite increased first and then decreased as the heat treatment time increased, which was attributed to the mutual reaction of the increase in the relative density with the formation of the flaws located at the interface. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction A major problem in the service of ceramics as structural materials is their low toughness [1]. Even though many attempts have been used to increase their toughness, including incorporation of fibers [2], whiskers [3] or particles reinforcements [4], and ZrO2 phase transformation reinforcing and so on [5], up to date the brittleness of ceramics has not been overcome in nature. In the research on the structure of natural biomaterials, such as turtleback, shell and nacres, it has been found that these natural biomaterials have very reasonable laminated structures which give them many excellent properties, such as good carrying capacity, good toughness and so on [6–9]. Over the past two decades, bionics has had a profound influence on the material science and engineering, because the unique structures, compositions and correspondingly excellent properties of biology gave researchers many clues to improve the properties of materials or increase the reliability of structural components [10]. One way of preventing properties of ceramic component is to design a structure with dense and strong layers separated by weak or soft layers of the different materials [11]. The weak interface in these laminated composites serves to deflect the propagating crack and reduces its stress ⇑ Corresponding author. Tel./fax: +86 411 84706791. E-mail address: [email protected] (Z. Wang). 0261-3069/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2011.07.049
intensity. Another way of increasing reliability is the use of intermittent interlayer [12]. One of the requirements for achieving high properties of the laminated composites is the good chemical and physical bonding between the interface materials. Electron beam–physical vapor deposition (EB–PVD) is one of the advanced techniques for the fabrication of the thin coatings [13]. In the late 1980s, EB–PVD techniques was being used to deposit the top coat onto rotating blades [14]. This deposition method resulted in a columnar microstructure with the columns running perpendicular to the work piece surface, which imparted a high degree of strain tolerance making them ideal for applications involving frequent thermal cycles [15]. In the present work, the bionic alloy–ceramic laminated composite was fabricated by EB–PVD. The ingots of Ni–20Co–12Cr– 4Al (wt.%) and ZrO2–8 mol%Y2O3 were used as the sources of the alloy layer and ceramic layer, respectively. The mechanical properties and microstructure of the alloy–ceramic laminated composite were investigated in detail. Furthermore, the laminated composite was heat treated in order to improve the tensile strength through reducing residual stress. 2. Experimental procedure During deposition process, the stainless steel substrate with the diameter of 1 m and surface roughness of 1.0 rotated at 6 rpm
301
G. Shi et al. / Materials and Design 33 (2012) 300–305
around the vertical axis and the substrate temperature was maintained at 650 °C approximately. The process pressure was in the range of 6–10 103 Pa. The ingots of Ni–20Co–12Cr–4Al (wt.%) and ZrO2–8 mol%Y2O3 were used as the sources of the alloy layer and ceramic layer, respectively. The ingots of ZrO2–8 mol%Y2O3 and Ni–20Co–12Cr–4Al (wt.%) were evaporated alternately to produce the alloy–ceramic laminated composite and the thicknesses of the different layers were controlled by their deposition time. The tensile strength tests were carried out by an INSTRON-5569 universal materials testing machine with a crosshead displacement speed of 0.05 mm/min at room temperature. Nanoindentation was performed using a Hysitron TriboIndenter (Hysitron Inc., Minneapolis, MN) with a 100 lm conospherical diamond fluid tip. The laminated composite was heat treated at 1050 °C for different times in Ar atmosphere, and heating rate of 10 °C/min and cooling rate 5 °C/min was lower than the heating cooling rates of the laminated composite in the fabrication process, respectively. The indenter tip is connected to the transducer, which was displacement-controlled, and data were presented as load versus displacement plots for loading, hold, and unloading. For all samples, indentation started with the nanoindenter tip approaching the sample from off contact. In this work, five specimens were tested to get an average value. The microstructures of the specimens were examined using SEM (FEI QUANTA200). Moreover, the phase composition of the laminated composite was determined by X-ray diffractometer (Philips, X’Pert-MRD). The interlaminar strength was evaluated according to GB-T3903.8-2005.
3. Results and discussions 3.1. Microstructure The laminated composite was generally destroyed along the ceramic layer, as shown in Fig. 1A, when the interlaminar strength was determined, which revealed that the ceramic layer appeared to be well bonded with the alloy layer. The outstanding interface bonding was favorable to the mechanical properties of the laminated composite. It could be seen from Fig. 1A that 8 mol%Y2O3 stabilized ZrO2 particles appeared to be uniformly deposited on the alloy layer. Furthermore, the flaws, such as pore and pit, were not observed in the ceramic layer. Fig. 1B shows the SEM image of the fractured surface perpendicular to the alloy–ceramic interface of the laminated composite. The ceramic layer of uniform thickness was separated by the alloy layer of uniform thickness. Fig. 2 shows the SEM images of the polished cross section of the laminated composite before and after the heat treatment. No flaws, such as pore and crack, were observed in the ceramic layer of the polished cross section of the laminated composite before and after the heat treatment, which also revealed that ceramic particles
appeared to be uniformly deposited on the alloy layer. The obvious diffusion interfaces between the ceramic layer and alloy layer were readily detected, which also indicated the excellent interface bonding between the ceramic layer and the alloy layer. As heat treatment time increased, the obvious flaws such as pit and pore occurred in the interface between the ceramic and alloy layers and the amount of the flaws increased gradually. The formation of the flaws was attributed to the diffusion of the particles during high temperature heat treatment [16]. In the heat treatment process, the diffusion of the flaws within the ceramic layer and/or alloy layer to the interface between the ceramic layer and alloy layer was easier compared with the occurrence of the interlaminar diffusion [17]. In the heat treatment process, the diffusion of the flaws within the ceramic layer and/or alloy layer to the interface between the ceramic layer and alloy layer was favorable to improve the relative density of the ceramic layer and/or alloy layer, whereas was unfavorable to the interface bonding between the ceramic layer and alloy layer. Fig. 3 shows the XRD spectra obtained from the polished cross section of the laminated composite before and after the heat treatment. Apparently, the phase analysis indicated the predominant phases for the laminated composite were t-ZrO2 for the ceramic layer and c-Ni phase for the alloy layer. Compared with the XRD spectra for the polished cross section of the laminated composite before the heat treatment, the obvious change in the XRD spectra was not detected as the heat treatment time increased. Such result indicated that the reaction of the ceramic layer with alloy layer was simple physical diffusion. The SEM images of the fractured surfaces of the tensile specimen for the laminated composite are shown in Fig. 4. The staged fracture mode resulted from the crack deflection prior to final failure of the tensile specimen were readily observed. Such fracture mode would dissipate a lot of energy and improve the mechanical properties of the laminated composite [3]. Many flat fracture regions resulted from the brittle fracture mode in the fractured surfaces of the tensile specimen for the laminated composite were also detected in the fractured surfaces, as shown in Fig. 4B. The obvious plastic deformation was not observed in the flat fracture region, which was ascribed to the weak grain boundary bonding of the alloy particles [18]. Furthermore, the fractured surface of the alloy layer was blade-like shrank, which indicated the ductile fracture of the laminated composite. The interface debonding phenomenon was easily observed in the ductile fracture region of the alloy layer, whereas the interface debonding did not occur in the brittle fractured region, which revealed that the ceramic layer was well bonded with the alloy layer in the brittle fractured region. The columnar crystals with sharp edges and corners were observed in ceramic layer, which were the typical characteristics of brittle intergranular fracture [19], as shown in Fig. 4C. The ceramic layer was fractured to many fragments due to the crack initiation and
B
A
2µm
50µm
Fig. 1. The SEM images of the fractured surface parallel (A) and perpendicular (B) to the alloy–ceramic interface of the laminated composite.
302
G. Shi et al. / Materials and Design 33 (2012) 300–305
A
B Alloy layer
Ceramic layer 5µm
5µm
Fig. 2. The SEM images of the polished cross section of the laminated composite before (A) and after (B) the heat treatment for 4 h.
Fig. 3. The XRD spectra obtained from the polished cross section of the laminated composite before and after the heat treatment.
propagation. The separation of the ceramic layers in the fractured surface was due to the plastic deformation of the alloy layer and the interface debonding, which indicated that the interlaminar strength was less than the strength of the ceramic layer or alloy layer. The ceramic layer of columnar structure has a good strain tolerance and the ceramic layer with the alloy layer could be bent and deformed, so that it could greatly reduce the residual stress caused by the differences of the physical properties in the alloy and ceramic layers. The SEM images of the fractured surfaces for the specimen after the heat treatment are shown in Fig. 5B–D. In order to investigate the change in the microstructure, SEM micrograph of the fractured surfaces for the specimen before the heat treatment (has been described in Fig. 4A) is shown in Fig. 5A for comparison. It could be seen from Fig. 5 that the brittle fracture region gradually reduced as the heat treatment time increased because the bonding strength of the grain boundary in the alloy layer was improved by the atom diffusion and grain boundary migration process. These diffusions were favorable to improve the relative density of the ceramic layer and alloy layer, whereas it would reduce the interface bonding strength due to the diffusion of the flaws to the interface between the ceramic layer and alloy layer [16,17]. Compared with the
A
50µm
C
B
15µm
15µm
Fig. 4. SEM images of the fractured surfaces of the tensile specimen for the laminated composite.
303
G. Shi et al. / Materials and Design 33 (2012) 300–305
B
A
50µm
50µm
D
C
50µm
50µm
Fig. 5. SEM images of the fractured surfaces for the specimen before (A) and after the heat treatment for 0.5 h (B), 2 h (C) and 4 h (D).
Fig. 6. The tensile strength and interminar strength of the laminated composite before and after the heat treatment.
fractured surface of the specimen before the heat treatment, the obvious separation of the interface was observed in the fractured surface of the specimen for the heat treatment for 0.5 h and the amount of the interface separation further increased, whereas the width of the interface separation decreased with increasing heat treatment time to 2 h. As the heat treatment time increased to 4 h, the amount of the interface separation decreased, whereas the width of the interface separation increased, which was attributed to the weakness in the interface bonding due to the diffusion of the flaws to the interface between the ceramic layer and alloy layer.
the tensile strength decreased slightly with increasing the heat treatment time to 4 h. However, the tensile strength for the heat treatment of 4 h was still greater than that for the laminated composite before the heat treatment, which was attributed to the reduction of the residual stress in the laminated composite. The tensile strength of the laminated composite was mostly dependent on the strength of the alloy layer because the incompact bonding of the ceramic particles in the ceramic layer, as shown in Fig. 1A. The interlaminar strength with increasing heat treatment time is also shown in Fig. 6. As the heat treatment time increased, the interlaminar strength decreased gradually because the increase in the heat treatment time was favorable to the diffusion of the flaws the interface between the ceramic layer and alloy layer. In the last decade the nanoindentation proposed by Pharr et al. has evolved as a promising tool for the structure–property characterization of tissues and biomaterials [20]. In order to confirm the tensile strength of the laminated composite dependent on the strength of the alloy layer and clarify the effect of the heat treatment time on the strength of the alloy layer, the yield strength of the alloy layer was calculated by the nanoindentation techniques. In the nanoindentation measurement process, the total work, Wt, released by the indenter could be divided into the elastic work and plastic work, and Giannakopoulos and Suresh [21] has proposed as:
We hr ¼1 Wt hmax
ð1Þ
where hmax and hr are the maximum penetration depth of the indenter into the specimen and the residual depth (after complete unloading), respectively, We is the elastic work. The hmax and hr could be determined from the nanoindentation load–displacement curves, and the load–displacement relationship was calculated as:
3.2. Mechanical properties 2
Fig. 6 shows the tensile strength of the laminated composite before and after the heat treatments. As the heat treatment time increased to 2 h, the tensile strength increased gradually, whereas
Pmax ¼ C hmax
ð2Þ
where C and Pmax are the curvature of the indentation and maximum load, respectively. The literature has confirmed that the
304
G. Shi et al. / Materials and Design 33 (2012) 300–305 Table 1 The mechanical properties of the alloy layer in the laminated composite treated for different time.
Fig. 7. The nanoindentation load–displacement curves of the alloy layer in the laminated composite treated for different times.
curvature of the indentation for the elastic–plastic materials is dependent on the strength of the materials as follows [22]:
C¼
Pmax 2 hmax
¼ M1 r0:29 1 þ
ry r0:29
E ln þ M2
ry
ð3Þ
where ry and r0.29 the yield strength determined at uniaxial compression and the stress corresponding to the plastic strain of 0.29, respectively. The constant M1 and M2 are 6.618 and 0.875 for Pharr’s indenter [22]. In the indentation process, the ratio of the residual depth to the maximum penetration depth was to characterize the degree of material deformation as follows:
r0:29 ry 0:29E
¼ 1 0:142
2 hr hr 0:957 hmax hmax
ð4Þ
The nanoindentation load–displacement curves are shown in Fig. 7 and the yield strength of the alloy layer was calculated and is listed in Table 1. Young’s modulus did not show change as the heat treatment time increased. In contrast to the modulus, the hardness and yield strength increased first and then decreased as the heat treatment time increased. In the heat treatment process, the flaws of the smaller size could diffuse from the alloy layer to the interface, whereas the flaws of the larger size could not be eliminated, which resulted in the reduction of the tensile strength relative to the yield strength measured by the nanoindentation techniques. However, the yield strength measured by the nanoindentation techniques has change trend similar to that of tensile strength of the laminated composite as the heat treatment time increased, which confirmed that the tensile strength of the laminated composite was mostly dependent on the strength of the alloy layer. The increase in the density of the alloy layer with the increasing heat treatment time was favorable to the increase in the strength of the alloy layer; but on the other hand, the decrease in the interface bonding strength would lead to the reduction in the effect of Orowan strengthening on the alloy layer. The flaws such as pit and pore generally occurred in the materials fabricated by EB– PVD [13–15], which led to the reduction of the relative density. The hardness and strength of the materials was reduced by the low relative density and the flaws located at the interface. The brittle fracture of the alloy layer was activated by the flaws. The amount of the flaws in the alloy layer was reduced in the heat treatment process with the increasing heat treatment time, which was favorable to improve the strength of the alloy layer because the proportion of the brittle fracture of the alloy layer was reduced. Furthermore, the mechanical properties of the laminated composite were also affected by the interface bonding strength because
Heat treatment time (h)
0
0.5
2
4
Modulus (GPa) Hardness (GPa) Yield strength (GPa)
243 5.56 85.7
241 5.65 87.2
240 5.86 89.8
244 5.81 86.9
the laminated composite included a lot of the alloy–ceramic interface. As the heat treatment time increased, the amount of the flaws located at the interface increased gradually, which would reduce the mechanical properties of the laminated composite. Therefore, the tensile strength of the laminated composite increased first and then decreased as the heat treatment time increased, which was attributed to the mutual reaction of the increase in the relative density with the formation of the flaws located at the interface. Further work is continuing to investigate and report the effect of the laminated structure on the thermophysical properties of the bionic alloy–ceramic laminated composite. 4. Conclusions In the present work, the bionic alloy–ceramic laminated composite was fabricated using the ingots of Ni–20Co–12Cr–4Al (wt.%) and ZrO2–8 mol%Y2O3 as the sources of the alloy layer and ceramic layer, respectively. The mechanical properties and microstructure of the laminated composite were investigated in detail. The laminated composite was generally destroyed along the ceramic layer when the interlaminar strength was determined, which revealed that the ceramic layer appeared to be well bonded with the alloy layer. The flaws, such as pore and crack, were not observed in the polished cross section of the laminated composite. Compared with the fractured surface feature of the specimen before the heat treatment, the obvious separation was observed in the fractured surface of the specimen after the heat treatment, which was attributed to the weakness in the interface bonding due to the diffusion of the flaws to the interface between the ceramic layer and alloy layer. The tensile strength of the laminated composite increased first and then decreased as the heat treatment time increased, which was attributed to the mutual reaction of the increase in the relative density with the formation of the flaws located at the interface. Acknowledgments This work was supported by China Postdoctoral Science Foundation Funded Project (20100481220) and the Fundamental Research Funds for the Central Universities (3014-852001 and DUT10ZDG05) and the National Natural Science Foundation of China (51002019 and 91016024). References [1] Wang Z, Hong CQ, Zhang XH, Sun X, Han JC. Microstructure and thermal shock behavior of ZrB2–SiC–graphite composite. Mater Chem Phys 2009;113:338–41. [2] Lee KS, Jang KS, Park JH, Kim TW, Han IS, Woo SK. Designing the fiber volume ratio in SiC fiber-reinforced SiC ceramic composites under Hertzian stress. Mater Des 2011;32:4394–401. [3] Wu P, Zheng Y, Zhao YL, Yu HZ. Effect of SiC whisker addition on the microstructures and mechanical properties of Ti(C, N)-based cermets. Mater Des 2011;32:951–6. [4] Meng SH, Chen HB, Hu JH, Wang ZW. Radiative properties characterization of ZrB2–SiC-based ultrahigh temperature ceramic at high temperature. Mater Des 2011;32:377–81. [5] Zhu T, Li WJ, Zhang XH, Hu P, Hong CQ, Weng L. Damage tolerance and R-curve behavior of ZrB2–ZrO2 composites. Mater Sci Eng A 2009;516:297–301.
G. Shi et al. / Materials and Design 33 (2012) 300–305 [6] Chen B, Peng X, Wang JG, Wu X. Laminated microstructure of Bivalva shell and research of biomimetic ceramic/polymer composite. Ceram Int 2004;30:2011–4. [7] Chen RF, Wang CA, Huang Y, Le HR. An efficient biomimetic process for fabrication of artificial nacre with ordered-nanostructure. Mater Sci Eng C 2008;28:218–22. [8] Mirzababaee M, Tahani M. Accurate determination of coupling effects on free edge interlaminar stresses in piezoelectric laminated plates. Mater Des 2009;30:2963–74. [9] Fu YM, Zhang P, Yang F. Interlaminar stress distribution of composite laminated plates with functionally graded fiber volume fraction. Mater Des 2010;31:2904–15. [10] Pinto Correia IF, Mota Soares CM, Mota Soares CA, Herskovits J. Analysis of adaptive shell structures using a refined laminated model. Compos Struct 2004;66:261–8. [11] Almeida FS, Awruch AM. Design optimization of composite laminated structures using genetic algorithms and finite element analysis. Compos Struct 2009;88:443–54. [12] Krstic Z, Krstic VD. Young’s modulus, density and phase composition of pressureless sintered self-sealed Si3N4/BN laminated structures. J Eur Ceram Soc 2008;28:1723–30. [13] Schulz U, Schmücker M. Microstructure of ZrO2 thermal barrier coatings applied by EB–PVD. Mater Sci Eng A 2000;276:1–8.
305
[14] Wellman RG, Nicholls JR. Erosion, corrosion and erosion–corrosion of EB PVD thermal barrier coatings. Tribol Int 2008;41:657–62. [15] Peng H, Guo HB, He J, Gong SK. Oxidation and diffusion barrier behaviors of double-layer NiCoCrAlY coatings produced by plasma activated EB–PVD. Surf Coat Technol 2011;205:4658–64. [16] Haynes JA, Zhang Y, Cooley KM, Walker L, Reeves KS, Pint BA. Hightemperature diffusion barriers for protective coatings. Surf Coat Technol 2004;188–189:153–7. [17] Mulligan C, Blanchet TA, Gall D. Control of lubricant transport by a CrN diffusion barrier layer during high-temperature sliding of a CrN–Ag composite coating. Surf Coat Technol 2010;205:1350–5. [18] Sciti D, Bellosi A, Esposito L. Bonding of zirconia to super alloy with the active brazing technique. J Eur Ceram Soc 2001;21:45–52. [19] Kameda J, Nishiyama Y. Combined effects of phosphorus segregation and partial intergranular fracture on the ductile–brittle transition temperature in structural alloy steels. Mater Sci Eng A 2011;528:3705–13. [20] Pharr GM, Oliver WC, Brotzen FR. On the generality of the relationship among contact stiffness, contact area, and elastic modulus during indentation. J Mater Res 1992;7:613–7. [21] Giannakopoulos AE, Suresh S. Determination of elastoplastic properties by instrumented sharp indentation. Scripta Mater 1999;40:1191–8. [22] Suresh S, Giannakopoulos AE. A new method for estimating residual stresses by instrumented sharp indentation. Acta Mater 1998;46:5755–67.