- Email: [email protected]
Flexural Mechanical Properties of Functional Gradient Hydroxyapatite Reinforced Polyetheretherketone Biocomposites
Journal of Materials Science & Technology 32 (2016) 34–40
Contents lists available at ScienceDirect
Journal of Materials Science & Technology j o u r n a l h o m e p a g e : w w w. j m s t . o r g
Flexural Mechanical Properties of Functional Gradient Hydroxyapatite Reinforced Polyetheretherketone Biocomposites Yusong Pan *, Yan Chen, Qianqian Shen Laboratory of Multiscale Materials and Molecular Catalysis, School of Material Science & Engineering, Anhui University of Science and Technology, Huainan 232001, China
A R T I C L E
I N F O
Article history: Received 21 June 2015 Received in revised form 24 August 2015 Accepted 27 August 2015 Available online 1 December 2015 Key words: Hydroxyapatite Functional gradient HA/PEEK biocomposites Flexural strength Flexural modulus
Functional gradient hydroxyapatite reinforced polyetheretherketone is one of the most promising orthopedic implant biomaterials. In this study, functional gradient hydroxyapatite reinforced polyetheretherketone biocomposites were prepared by layer-by-layer method with the incorporation of hot press molding technology. Studies on the flexural mechanical properties of the functional gradient biocomposites revealed that the flexural stress–stain behavior of the biocomposites presented linear elastic characteristics. The fracture mechanism of the functional gradient biocomposites was predominated by brittle rupture. Furthermore, both flexural strength and break strain of the functional gradient HA/PEEK biocomposites obviously decreased with the rise of the total HA content. The effect of hydroxyapatite concentration difference between adjacent layers (HCDBAL) on the flexural strength obviously relied on the level of HCDBAL and total HA content in the functional gradient HA/PEEK biocomposites. The higher the total HA content in the functional gradient biocomposites is, the less the influence degree of HCDBAL on the flexural strength is. Moreover, total HA content and HCDBAL played synergistic influence on the flexural modulus of the functional gradient HA/PEEK biocomposites. Copyright © 2015, The editorial office of Journal of Materials Science & Technology. Published by Elsevier Limited. All rights reserved.
1. Introduction Polyetheretherketone (PEEK) is one of the most high-performance thermoplastic materials, which is a semi-crystalline thermoplastic polymer with an approximate crystallinity of 30%–35%[1,2]. PEEK has been used in a wide range of applications due to its outstanding properties such as superior mechanical properties, excellent wear-resistance, environmental resistance and thermal stability[3–5]. Especially, PEEK and PEEK composites used as implant materials for replacing and repairing hard tissues of human body have attracted increasing attention because of their excellent mechanical and biocompatibility properties[6–8]. Compared with traditional orthopedic implants such as metallic and ceramic implants, one of the benefits of PEEK based composites is that they could overcome the disadvantages of metallic and ceramic implants currently adopted in orthopedic application. It is well known that the modulus of the traditional orthopedic alloys and ceramic used in hard tissue reconstruction is 10–20 times greater than that of the bone. Thus, it will inevitably be a modulus of the implant mismatch with that of natural bone, thereby producing stress shielding effect, which would lead to the aseptic loosening of implants and bone loss. Moreover,
* Corresponding author. Assoc. Prof., Ph.D.; Tel.: +86 5546668649; Fax: +86 5546668649. E-mail address: [email protected] (Y. Pan).
many studies revealed that the electrochemical reaction of the metal alloy joint implants would occur and thus release metal ions into the natural tissue. The toxicity of metal ions released from metal joint still remains controversial[9–11]. Recently, PEEK and its composites have been widely used in orthopedic applications including the femoral component of total hip replacements, bone anchors, cervical total disc arthroplasty, intervertebral cages, dental implant systems and fracture fixation plates[12–14]. However, the mechanical properties of PEEK used in most loadbearing orthopedic implant should be further improved. Moreover, the traditional methods of stabilizing the prosthesis such as bone cement fixation are mostly substituted by bioactive fixation. The bioactive fixation technology requires prosthesis with good biocompatibility as well as excellent bioactivity. Lots of efforts have been devoted to improving the mechanical and bioactive properties of polyetheretherketone[15,16]. One of the most effective methods is to incorporate bioactive ceramics such as tricaclium phosphate, hydroxyapatite and bioactive glass-ceramics with PEEK polymer[17,18]. It should be noted that hydroxyapatite is an excellent candidate reinforcement and bioactive component for biocomposites on account of its similar constituent as that of inorganic component in natural bone and outstanding bioactive characteristics. Polymeric implant combination with HA promotes new bone growth from the existing bony walls (osteoconductive) onto it, thus stabilizing the prosthesis within a short period of time. In recent years, hydroxyapatite reinforced polyetheretherketone has been demonstrated to show
http://dx.doi.org/10.1016/j.jmst.2015.11.011 1005-0302/Copyright © 2015, The editorial office of Journal of Materials Science & Technology. Published by Elsevier Limited. All rights reserved.
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
35
Fig. 1. Schematic diagram of the functional gradient HA/PEEK biocomposites.
excellent bioactive performance[19]. Most of studies verified that the mechanical strength of PEEK can be importantly improved by the addition of certain content of hydroxyapatite particles in the PEEK matrix[20,21]. On the other hand, it is necessary to incorporate a larger amount of HA particles into the PEEK matrix to improve the bioactive properties of the composites. However, the excess of HA particles generally results in the brittleness of the composites accompanied with a significant deterioration of mechanical strength. Thus, there is conflict in simultaneously optimizing the mechanical and bioactive properties of hydroxyapatite reinforced polyetheretherketone biocomposites. The functional gradient design is an effective method to solve the conflict between mechanical and bioactive performances of the HA/PEEK biocomposites. Here, we selected hydroxyapatite particles as bioactive constituents as well as reinforcement and the polyetheretherketone as matrix to prepare functional gradient HA/ PEEK biocomposites. The functional gradient HA/PEEK biocomposite is composed of symmetrical five layers, which are respectively denoted as outer layer, middle layer and central layer (Fig. 1). The HA particle content in the functional gradient biocomposites presents increasing trend from central to the outer layer. The outer layer with the highest HA content endows the functional gradient biocomposites with outstanding bioactivity. The central layer with lowest HA content offers the functional gradient biocomposites appropriate mechanical strength and suitable ductility. The middle layer with moderate HA content makes the stress effectively transfer among the layers in the functional biocomposites. According to this design, the functional gradient HA/PEEK biocomposites with an outer layer containing high HA content and inner layer containing low HA contents would realize the simultaneous optimization on mechanical and bioactive properties of the biocomposites by altering the layer HA content in the composites. Currently, many works have attempted to improve the mechanical and bioactive properties of polyetheretherketone. However, it is a pending question to simultaneously optimize the mechanical and bioactive properties of PEEK. According to the authors’ knowledge, little attention has been devoted to solve the conflict between mechanical and bioactive properties of PEEK material through functional gradient design. The purpose of this initial investigation was to estimate the mechanical properties of the functional gradient HA/ PEEK biocomposites under bending load. The influence of various factors on the flexural mechanical properties was investigated.
purchased from SCM Industrial Chemical Co., Ltd. Calcium hydroxide (Ca(OH)2) and phosphoric acid (H3PO4) were produced and purchased from Siopharm Chemical Reagent Co., Ltd. All the chemical reagents were of analytical grade.
2. Experimental
2.3. Flexural mechanical property measurement
2.1. Materials
A three-point bending test was employed to evaluate the flexural properties such as flexural strength, flexural strain and modulus of functional gradient HA/PEEK biocomposites on mechanical test equipment (Model: CMT-5105, Shenzhen SANS Material Detection
Polyetheretherketone powder with the average diameter of 50 μm (VICTREX® PEEK, 450PF) was manufactured by VICTREX PIC, U.K. and
2.2. Preparation of functional gradient HA/PEEK biocomposites In order to fabricate functional gradient HA/PEEK biocomposites, the HA–PEEK composite powder were first prepared by in-situ synthesis process according to our previous work[22]. The main processing schedule is described as follows. First, a certain amount of Ca(OH)2 and PEEK powder was mixed in distilled water at 80 °C for 30– 50 min by an electric agitator. Then H3PO4 solution was slowly added into Ca(OH)2 and PEEK solution at 90 °C under electric agitator stirring. The amount of H3PO4 to be added was controlled with Ca/P molar ratio of 1.67. After Ca(OH)2 and H3PO4 had thoroughly reacted, the solution was aged at 80 °C for 10–12 h. Subsequently, the solution was rinsed and filtered repeatedly using distilled water until the pH value of the filtrate was close to 7. Finally, the residue was dried in an oven at 120 °C to constant weight. According to this procedure, HA–PEEK composite powder with various HA contents were harvested. Functional gradient HA/PEEK biocomposites were prepared by layer-by-layer casting method with the incorporation of hot pressure molding technology. First, HA–PEEK composite powder with different HA concentrations was symmetrically stacked into five layers in a steel die with rectangular shape. The HA concentration in the HA–PEEK composite powder presents an increasing trend from central to the outer layer along with the thickness direction. The mass of the HA–PEEK composite powder in the central layer is about twice the mass of the HA–PEEK composite powder in the middle and outer layers, respectively. Then the samples were heated with thermoforming equipment from room temperature to 380–390 °C and held there for 30–50 min to ensure that the PEEK in the composite powder was completely melted. Subsequently, 4–8 MPa pressure was applied to the sample and the pressure was kept for 15 min at temperature of 380–390 °C. After then, the temperature was lowered to 300 °C and then the pressure was removed. Finally, the sample was cooled down in air to room temperature. According to this process schedule, functional gradient HA/PEEK biocomposites could be fabricated. Fig. 1 illustrates the schematic diagram of functional gradient HA/PEEK biocomposites.
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
36
Co., Ltd.). The crosshead speed was 1 mm/min. Strip specimens of 70 mm in length and 25 mm in width were used. Three specimens were tested for each type of biocomposites and the average value of the flexural strength and modulus was calculated. It should be noted that the pure PEEK sample was not fractured during the bending test due to its excellent ductility. Taking this into consideration and according to the test standard (ISO 178:2003), the flexural strength of the pure PEEK sample was evaluated at 1.5 times of conventional deflection. According to the test standard (ISO 178:2003), flexural stress of the functional gradient biocomposites can be calculated using the following equation:
σf =
3FL 2bh 2
(1)
where σ f is the flexural stress, and F, L, b and h are the applied force, span, width and thickness of the sample, respectively. The flexural modulus of the functional gradient biocomposites can be calculated by the following equation:
Ef =
σ f2 − σ f1 ε f2 − ε f1
(2)
where E f is the flexural modulus and σ f2 and σ f1 are the flexural stress measured at flexural strain ε f2 and ε f1 .
3. Results and Discussion 3.1. Stress–strain behavior Fig. 2 shows the flexural stress–strain behavior of HA/PEEK and functional gradient HA/PEEK biocomposites with different hydroxyapatite contents. Some results can be concluded from Fig. 2. First, the flexural stress changing behavior of the HA/PEEK and functional gradient HA/PEEK biocomposites shows a linear increase with the rise of strain (Fig. 2(a, b)). Second, for all the functional gradient biocomposites, the stress yielding behavior of the biocomposites was not observed in the flexural stress–strain curves (Fig. 2(a, b)). Such results revealed that the HA/PEEK and functional gradient HA/ PEEK biocomposites exhibit linear elastic characteristics under the bending load. The fracture mechanism of the functional biocomposites is probably dominated by the brittle fracture. Lastly, it can be concluded from the insert graphs in Fig. 2(a, b) that the flexural break strain of the functional gradient HA/PEEK biocomposites presents decreasing trend with the rise of total HA content in the PEEK matrix. For example, the flexural break strains of the HA/PEEK and functional gradient HA/PEEK biocomposites decreased from 2.67% and 1.82% to 0.69% and 0.73%, respectively, while total HA content in the biocomposites increased from 10 wt% to 40 wt% (Fig. 2(a, b)). It is well known that the mechanical properties of the polymer/ceramic composites strongly depend on the mechanical properties of the inorganic components. Generally, the fracture strain of the ceramics such as hydroxyapatite and tricalcium phosphate reinforcements is much lower than that of the PEEK matrix due to the brittleness of the inorganic ceramics[23–25]. Then addition of inorganic ceramics in the PEEK matrix would result in the flexural break strain reduction. The change behavior of flexural break strain with total HA content in functional biocomposites indicates that the ductility decreases and brittleness increases with the rise of total HA content in the functional biocomposites. The effect of hydroxyapatite concentration difference between adjacent layers (HCDBAL) on the flexural stress–strain behavior is shown in Fig. 3. It can be obviously concluded from Fig. 3 that the change behavior of the flexural stress exhibits linear increasing trend
Fig. 2. Stress–strain behavior of HA/PEEK and functional gradient HA/PEEK biocomposites: (a) HA/PEEK, (b) functional gradient HA/PEEK with 10 wt% HCDBAL. (0/10/20 represents that the functional gradient biocomposite is composed of symmetric five layers. HA content in the outer layer is 20 wt%, HA content in the middle layer is 10 wt% and that in the central layer is 0 wt%. The total HA content in the sample is 10 wt%. HCDBAL is the abbreviation of hydroxyapatite concentration difference between adjacent layers.)
with the rise of strain under all the HCDBAL in the functional HA/ PEEK biocomposites. The change behavior of flexural stress with HCDBAL revealed that the HCDBAL would not change the elastic characteristic of the functional HA/PEEK biocomposites. Moreover, it can be found in the inset of Fig. 3 that the flexural break strain decreases with the rise of HCDBAL. The flexural break strain decreases from 1.53% to 1.15% while HCDBAL rises from 0 wt% to 15 wt%. Such a result indicates that the brittleness of the functional biocomposites increases with the rise of HCDBAL in the biocomposites. 3.2. Flexural strength Fig. 4 shows the effect of total hydroxyapatite content on the flexural strength and flexural break strain of the functional HA/PEEK biocomposites under various HCDBAL. It can be obviously concluded from Fig. 4 that both flexural strength and flexural break strain of the functional HA/PEEK biocomposites decrease with the rise of total HA content in the biocomposites under all the HCDBAL. For example, while HCDBAL is 5 wt%, the flexural strength and flexural
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
37
Fig. 3. Effect of HCDBAL on the flexural stress–strain behavior (total HA content in the PEEK matrix is 20 wt%; 15/20/25 represents that the functional gradient biocomposite is composed of symmetric five layers. HA content in the outer layer is 25 wt%, HA content in the middle layer is 20 wt% and that in the central layer is 15 wt%. The total HA content in the sample is 20 wt%. HCDBAL is the abbreviation of hydroxyapatite concentration difference between adjacent layers).
break strain of the functional gradient HA/PEEK biocomposites decreased from 103.32 MPa and 2.25% to 57.27 MPa and 0.86%, respectively, while total hydroxyapatite content in the functional gradient biocomposites increased from 5 wt% to 40 wt% (Fig. 4(b)). The change behavior of the flexural strength with the rise of total HA content is mainly attributed to two factors. One is ascribed to the inhomogeneous distribution and the agglomeration of HA particles in the PEEK matrix with the rise of total HA content in the functional gradient biocomposites, which would ultimately result in deterioration of the tensile strength[26]. The other arises from the deficient interfacial bonding strength between HA particles and PEEK matrix due to their poor physical compatibility. With the rise of total HA content, the adverse influence of the deficient interfacial bonding strength on the tensile strength was more significant and ultimately deteriorates the flexural mechanical properties of the functional gradient HA/PEEK biocomposites[27,28]. Thus, the deficient bonding strength between HA particles and PEEK matrix leads to the flexural strength deterioration with the rise of total HA content. Furthermore, comparing Fig. 4(a–c), it can be concluded that the flexural strength change behavior of the biocomposites with the rise of total HA content in the biocomposites is importantly influenced by HCDBAL. For example, while total HA content rises from 10 wt% to 40 wt%, the flexural strength of the uniform HA/PEEK biocomposites (0 wt% HCDBAL) decreases from 127.53 MPa to 46.48 MPa, almost decreasing by 64% (Fig. 4(a)). On the other hand, while total HA content rises from 10 wt% to 40 wt%, the flexural strengths of the functional gradient HA/PEEK biocomposites with 5 wt% and 10 wt% HDCBAL respectively decrease from 97.69 MPa and 96.13 MPa to 57.27 MPa and 60.10 MPa, separately decreasing by 41% and 38% (Fig. 4(b, c)). Such a result indicates that the flexural strength decreasing rate of the functional gradient HA/PEEK biocomposites with the rise of total HA content is obviously slower than that of uniform HA/PEEK biocomposites (0 wt% HDCBAL). That is, the existence of a gradient layer in the functional gradient HA/ PEEK biocomposites is beneficial to slowing down the flexural strength decline of the functional biocomposites with the rise of total HA content in the functional biocomposites. It should be noted that reducing the flexural strength decline rate with the rise of total HA content in the functional biocomposites is conducive to the
Fig. 4. Effect of total HA content on the flexural strength and flexural break strain of functional biocomposites under different HCDBAL: (a) 0 wt%, (b) 5 wt%, (c) 10 wt%.
mechanical and bioactive properties of the functional biocomposites simultaneous optimization. This result is consistent with our initial design idea to simultaneously optimize the mechanical and bioactive performances of the biocomposites through the design of gradient layers in the biocomposites. Moreover, it is interestingly found from Fig. 4 that the decline rate of flexural break strain of the functional gradient HA/PEEK
38
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
biocomposites with the rise of total HA content in the biocomposites is obviously less than that of uniform HA/PEEK biocomposites. For example, while total HA content in the uniform HA/PEEK biocomposites (0 wt% HCDBAL) rises from 10 wt% to 40 wt%, the flexural break strain decreases from 2.67% to 0.69%, almost decreasing by 74%. Conversely, while total HA content rises from 10 wt% to 40 wt%, the flexural break strains of the functional gradient HA/ PEEK biocomposites with 5 wt% and 10 wt% HDCBAL separately decrease from 1.89% and 1.82% to 0.87% and 0.73%, respectively, decreasing by 53% and 59%. Such a result reveals that the existence of a gradient layer in the functional gradient biocomposites could importantly delay the plasticity to brittleness transmission of the functional biocomposites with the rise of total HA content. Thus, it could also explain that the flexural strength decline rate of the functional gradient HA/PEEK biocomposites with the rise of total HA content is less than that of uniform HA/PEEK biocomposites. Fortunately, the delay of the plasticity to brittleness transmission of the functional biocomposites is benefical to resisting the catastrophic fracture of the orthpaedic implants. The effect of HCDBAL on the flexural strength of the functional gradient HA/PEEK biocomposites with different total HA content is shown in Fig. 5. It can be concluded that the effect of HCDBAL on the flexural strength of the functional biocomposites can be obviously divided into two regions. While the total HA content in the functional gradient biocomposites remains constant, the effect of HCDBAL at low value region (0–15 wt%) on the flexural strength of the functional biocomposites is obviously less than that at the high value region (15 wt%–20 wt%). For example, while HCDBAL rises from 0 to 15 wt%, the flexural strength of the functional HA/PEEK biocomposites with 20 wt% and 30 wt% total HA content shows little variation. On the other hand, while HCDBAL rises from 15 wt% to 20 wt%, the flexural strength of the functional HA/PEEK biocomposites with 20 wt% and 30 wt% total HA content respectively decreases from 76.98 MPa and 62.69 MPa to 67.74 MPa and 53.92 MPa, almost decreasing by 12% and 14%. The flexural mechanical properties of the functional gradient HA/ PEEK biocomposites importantly depended on the two factors. One is the flexural strength of the monolayer in the functional gradient biocomposites and the other is the efficiency of the stress
transmission among the interface between inner-layers in the functional gradient HA/PEEK biocomposites. According to the composite principle of the mechanical strength of the composites materials, the strength of the functional gradient HA/PEEK biocomposites can be expressed in the following equation. 5
σ fc = ∑ σ icv ic
(3)
i =1
where σ fc represents the flexural strength of the functional gradient HA/PEEK biocomposites and σ ic and v ic denote the flexural strength of the monolayer, which is composed of uniform HA/PEEK biocomposites, and the volume percent of the monolayer, respectively. It can be deduced from Eq. (3) that the higher the flexural strength of the monolayer is, the higher the flexural strength of the functional gradient HA/PEEK biocomposites is. While total HA content in the functional gradient biocomposites holds constant, in comparison with uniform HA/PEEK biocomposites (0 wt% HCDBAL), the functional gradient HA/PEEK biocomposites is composed of various monolayers with different HA contents, in which some monolayers contain lower HA content than uniform HA/PEEK and other monolayers contain higher HA content than uniform HA/PEEK composites. Just as aforementioned (Fig. 4(a)), the flexural strength of the uniform HA/PEEK biocomposites decreases with the rise of HA content. Then, in comparison with uniform HA/PEEK biocomposites, the flexural strength of the monolayer with lower HA content in the functional gradient biocomposites is higher than that of uniform HA/PEEK biocomposites. Conversely, the flexural strength of the monolayer with higher HA content in the functional gradient biocomposites is lower than that of uniform HA/PEEK. Based on Eq. (3), in comparison with uniform HA/PEEK biocomposites, the higher flexural strength of the monolayer would offset the lower flexural strength of the monolayer in the functional gradient HA/PEEK biocomposites. Thus, the flexural strength of the functional gradient HA/PEEK biocomposites has little change with the rise of HCDBAL while HCDBAL is in the low concentration region (0–15 wt%). On the other hand, the difference in the microstructure and components on the interface, which is composed of adjacent monolayer in the functional gradient biocomposites, becomes more and more obvious with the rise of HCDBAL. Such change behavior of the interface in the functional gradient HA/PEEK biocomposites would result in the ineffective stress transmission among the layers[29–31]. Thus, the flexural strength of the functional gradient biocomposites decreases while HCDBAL in the functional gradient biocomposites exceeds a certain concentration (larger than 15 wt%). According to the above discussion, it can be concluded that the mechanical properties of the monolayer in the functional gradient biocomposites predominated the flexural strength of the functional gradient HA/ PEEK biocomposites and the stress would effectively transform among monolayer while HCDBAL is in the low concentration region. While HCDBAL is in the high concentration region, the ineffective stress transmission among the monolayer controls the flexural strength of the functional gradient HA/PEEK biocomposites. Thus, the flexural strength of the functional gradient biocomposites decreases while HCDBAL exceeds a certain concentration. It should be noted that the change behavior of the flexural strength of the functional gradient HA/PEEK biocomposites with HCDBAL provides us an effective method to simultaneously optimize the mechanical and biological properties of the biocomposites through controlling the total HA content and HCDBAL in the functional gradient HA/PEEK biocomposites. 3.3. Flexural modulus
Fig. 5. Effect of HCDBAL on the flexural strength of functional gradient biocomposites with different total HA contents.
The effect of total HA content on the flexural modulus of the functional gradient biocomposites with various HCDBAL is shown in
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
39
flexural modulus of the functional gradient biocomposites first increases and then decreases with the rise of total HA content. On the other hand, while HCDBAL in the functional gradient biocomposites rises further, the flexural modulus of the biocomposites presents increasing trend with the rise of total HA content in the functional gradient biocomposites. For example, while total HA content in the functional gradient biocomposites rises from 5 wt% to 40 wt%, the flexural modulus of the functional gradient biocomposites with 5 wt% HCDBAL first increases from 4.18 GPa to 7.97 GPa and then decreases to 6.94 GPa. The flexural modulus of the functional gradient biocomposites reaches a maximum while total HA content is 35 wt% (Fig. 6(b)). As for the functional gradient HA/PEEK biocomposites with 10 wt% HCDBAL, the flexural modulus monotonically increases from 5.69 GPa to 8.88 GPa while total HA content increases from 10 wt% to 40 wt% (Fig. 6(c)). The flexural modulus reflects the capability of the biocomposites to resist the deformation under external flexural force. It is generally accepted that the higher the rigidity of the material, the higher the modulus of the material. The functional gradient HA/PEEK biocomposite is composed of two components such as inorganic hydroxyapatite reinforcement and polymeric polyetheretherketone matrix. The rigidity of the inorganic hydroxyapatite is much higher than that of the polymer PEEK matrix. Thus, the flexural modulus of the functional gradient biocomposites presents increasing trend with the rise of total HA content in the functional biocomposites. Fig. 7 shows the effect of HCDBAL on the flexural modulus of the functional gradient HA/PEEK biocomposites with two different total HA contents. It can be concluded that the change behavior of the flexural modulus with HCDBAL obviously relies on the total HA content in the functional gradient HA/PEEK biocomposites. While total HA content in the functional gradient biocomposites is 20 wt%, the flexural modulus of the functional gradient biocomposites increases with the rise of HCDBAL. The flexural modulus increases from 5.53 GPa to 6.60 GPa while HCDBAL increases from 0 to 20 wt%. Conversely, while total HA content in the functional gradient biocomposites is 30 wt%, the flexural modulus of the functional gradient biocomposites decreases with the rise of HCDBAL. The flexural modulus decreases from 9.09 GPa to 5.42 GPa while HCDBAL increases from 0 to 20 wt%.
Fig. 6. Effect of total HA content on the flexural modulus of the functional gradient HA/PEEK biocomposites: (a) 0 wt% HCDBAL, (b) 5 wt% HCDBAL, (c) 10 wt% HCDBAL.
Fig. 6. It can be concluded that the effect of total HA content on the flexural modulus of the functional gradient HA/PEEK biocomposites obviously depended on the level of HCDBAL. While HCDBAL in the functional gradient biocomposites varies from 0 wt% to 5 wt%, the
Fig. 7. Effect of HCDBAL on the flexural modulus of functional gradient HA/PEEK biocomposites.
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
40
functional gradient biocomposites was lower than 5 wt%, the flexural modulus of the functional gradient biocomposites first increased and then decreased with the rise of total HA content. On the other hand, while HCDBAL in the functional gradient biocomposites was more than 5 wt%, the flexural modulus of the biocomposites presented increasing trend with the rise of total HA content in the functional gradient biocomposites. (4) The effect of HCDBAL on the flexural modulus also relied on the total HA content in the functional gradient HA/PEEK biocomposites. While total HA content in the functional gradient biocomposites was 20 wt%, the flexural modulus of the functional gradient biocomposites increased with the rise of HCDBAL. Conversely, while total HA content in the functional gradient biocomposites was 30 wt%, the flexural modulus of the functional gradient biocomposites decreased with the rise of HCDBAL.
Acknowledgment The authors gratefully acknowledge the financial support by the National Natural Science Foundation of China (No. 51175004). Fig. 8. Effect of HCDBAL on the flexural break strain of functional gradient HA/PEEK biocomposites.
The effect of HCDABL on the flexural break strain of the functional gradient biocomposites is shown in Fig. 8. It is interestingly found from Fig. 8 that the change behaviors of the flexural break strain and flexural modulus with HCDBAL present opposite trend. For example, while total HA content in the functional gradient biocomposites is 20 wt%, the flexural break strain of the functional gradient biocomposites decreases with the rise of HCDBAL. On the other hand, while total HA content in the functional gradient biocomposites is 30 wt%, the flexural break strain of the functional gradient biocomposites increases from 0.82% to 1.05% while HCDBAL increases from 0 to 20 wt%. It is well known that the flexural modulus reflects the ability of the material to resist flexural deformation under the flexural force loading. The lower the flexural modulus of the materials, the higher the flexibility of the materials. Thus, the change behavior of the flexural modulus and flexural strain with HCDBAL is opposite. 4. Conclusions (1) The flexural stress–strain behavior of functional gradient HA/ PEEK biocomposites presented linear elastic characteristics. Both the flexural strength and break strain of the functional gradient biocomposites decreased with the rise of total HA content. (2) The effect of HCDBAL on the flexural strength of the functional gradient HA/PEEK biocomposites obviously relied on the value of HCDBAL and total HA content in the functional gradient biocomposites. The higher the total HA content in the functional gradient biocomposites is, the less the influence degree of HCDBAL on the flexural strength is. (3) The effect of total HA content on the flexural modulus obviously depended on the HCDBAL. While HCDBAL in the
References [1] A. Jonas, R. Legras, J.P. Issi, Polymer 32 (1991) 3364–3370. [2] Y.C. Ke, Y.B. Zheng, Z.W. Wu, Chin. J. Mater. Res. 10 (1996) 205–209. [3] C.S. Michacl, E.M. James, G.B. Jennifer, M.K. Steve, M.R. Clare, Biomaterials 31 (2010) 9156–9162. [4] X.Z.K. Loy, S.K. Sinha, Wear 296 (2012) 681–692. [5] X.Q. Pei, B. Roland, B. Michael, Wear 304 (2013) 109–117. [6] A. Godara, D. Raabe, S. Green, Acta Biomater. 3 (2007) 199–220. [7] D.J. Jaekel, D.W. Macdonald, S.M. Kurtz, J. Mech. Behav. Biomed. Mater. 4 (2011) 1275–1282. [8] F.E.I. Halabi, J.F. Rodriguez, L. Rebolledo, E. Hurtos, M. Doblare, J. Mech. Behav. Biomed. Mater. 4 (2011) 1819–1832. [9] F. Heidenau, W. Mittelmeier, R. Detsch, M. Haenle, F. Stenzel, G. Ziegler, H. Gollwitzer, J. Mater. Sci. Mater. Med. 16 (2005) 883–888. [10] V.K. Raghunathan, J.N.A. Tettey, E.M. Ellis, M.H. Grant, J. Biomed. Mater. Res. A 88 (2009) 543–550. [11] X.Z. Mao, A.A. Wong, R.W. Crawford, Med. J. Aust. 194 (2011) 649–651. [12] S.M. Kurtz, J.N. Devine, Biomaterials 28 (2007) 4845–4869. [13] H. Xin, D. Shepherd, K.D. Dearn, Open Biomed. Eng. J. 6 (2012) 73–79. [14] M.G. Thomas, J.M. Hans, A.C. Janet, S. Jens, F. Bernhard, B. Wilhelm, J. Volkmar, Biomaterials 31 (2010) 523–531. [15] C.M. Han, E.J. Lee, H.E. Kim, Y.H. Koh, K.N. Kim, Y. Ha, S.U.S.U. Kuh, Biomaterials 31 (2010) 3465–3470. [16] C. Yao, D. Storey, T.J. Webster, Int. J. Nanomed. 2 (2007) 487–492. [17] L. Petrovic, D. Pohle, H. Munstedt, T. Rechtenwald, K. Schlegel, A.S. Rupprecht, J. Biomed. Sci. 13 (2006) 41–46. [18] G.M. Wu, W.D. Hsiao, S.F. Kung, Surf. Coat. Technol. 203 (2009) 2755–2758. [19] B.D. Hahn, D.S. Park, J.J. Choi, J. Ryu, W.H. Yoon, J.H. Choi, J.W. Kim, Appl. Surf. Sci. 283 (2013) 6–11. [20] L.C. Gabriel, W.M. Yue, R.K. Roeder, Biomaterials 28 (2007) 927–935. [21] A. Kalambettu, S. Dharmalingam, Mater. Chem. Phys. 147 (2014) 168–177. [22] Y.S. Pan, Q.Q. Shen, Y. Chen, Micro Nano Lett. 8 (2013) 357–361. [23] S. Kobayashi, T. Murakoshi, Adv. Compos. Mater. 23 (2014) 163–177. [24] M. Brillhart, B.L. Gregory, J. Botsis, Polymer 32 (1991) 1605–1611. [25] P.J. Rae, E.N. Brown, E.B. Orler, Polymer 48 (2007) 598–615. [26] Y.S. Pan, J. Wang, C.L. Pan, Micro Nano Lett. 7 (2012) 880–884. [27] S.M. Tang, P. Cheang, M.S. Abubakar, K.A. Khor, K. Liao, Int. J. Fatigue 26 (2004) 49–57. [28] T.L. Conrad, D.J. Jaekel, S.M. Kurtz, J. Biomed. Mater. Res. Part B Appl. Biomater. 101 (2013) 576–583. [29] B.S. Huang, G.H. Zhao, Z.H. Wang, Appl. Mech. Mater. 556–562 (2014) 725–728. [30] H. Qing, Comput. Mater. Sci. 89 (2014) 102–113. [31] Z.K. Lei, Z.L. Wang, W. Qiu, J. Adhes. Sci. Technol. 28 (2014) 1467–1479.
Contents lists available at ScienceDirect
Journal of Materials Science & Technology j o u r n a l h o m e p a g e : w w w. j m s t . o r g
Flexural Mechanical Properties of Functional Gradient Hydroxyapatite Reinforced Polyetheretherketone Biocomposites Yusong Pan *, Yan Chen, Qianqian Shen Laboratory of Multiscale Materials and Molecular Catalysis, School of Material Science & Engineering, Anhui University of Science and Technology, Huainan 232001, China
A R T I C L E
I N F O
Article history: Received 21 June 2015 Received in revised form 24 August 2015 Accepted 27 August 2015 Available online 1 December 2015 Key words: Hydroxyapatite Functional gradient HA/PEEK biocomposites Flexural strength Flexural modulus
Functional gradient hydroxyapatite reinforced polyetheretherketone is one of the most promising orthopedic implant biomaterials. In this study, functional gradient hydroxyapatite reinforced polyetheretherketone biocomposites were prepared by layer-by-layer method with the incorporation of hot press molding technology. Studies on the flexural mechanical properties of the functional gradient biocomposites revealed that the flexural stress–stain behavior of the biocomposites presented linear elastic characteristics. The fracture mechanism of the functional gradient biocomposites was predominated by brittle rupture. Furthermore, both flexural strength and break strain of the functional gradient HA/PEEK biocomposites obviously decreased with the rise of the total HA content. The effect of hydroxyapatite concentration difference between adjacent layers (HCDBAL) on the flexural strength obviously relied on the level of HCDBAL and total HA content in the functional gradient HA/PEEK biocomposites. The higher the total HA content in the functional gradient biocomposites is, the less the influence degree of HCDBAL on the flexural strength is. Moreover, total HA content and HCDBAL played synergistic influence on the flexural modulus of the functional gradient HA/PEEK biocomposites. Copyright © 2015, The editorial office of Journal of Materials Science & Technology. Published by Elsevier Limited. All rights reserved.
1. Introduction Polyetheretherketone (PEEK) is one of the most high-performance thermoplastic materials, which is a semi-crystalline thermoplastic polymer with an approximate crystallinity of 30%–35%[1,2]. PEEK has been used in a wide range of applications due to its outstanding properties such as superior mechanical properties, excellent wear-resistance, environmental resistance and thermal stability[3–5]. Especially, PEEK and PEEK composites used as implant materials for replacing and repairing hard tissues of human body have attracted increasing attention because of their excellent mechanical and biocompatibility properties[6–8]. Compared with traditional orthopedic implants such as metallic and ceramic implants, one of the benefits of PEEK based composites is that they could overcome the disadvantages of metallic and ceramic implants currently adopted in orthopedic application. It is well known that the modulus of the traditional orthopedic alloys and ceramic used in hard tissue reconstruction is 10–20 times greater than that of the bone. Thus, it will inevitably be a modulus of the implant mismatch with that of natural bone, thereby producing stress shielding effect, which would lead to the aseptic loosening of implants and bone loss. Moreover,
* Corresponding author. Assoc. Prof., Ph.D.; Tel.: +86 5546668649; Fax: +86 5546668649. E-mail address: [email protected] (Y. Pan).
many studies revealed that the electrochemical reaction of the metal alloy joint implants would occur and thus release metal ions into the natural tissue. The toxicity of metal ions released from metal joint still remains controversial[9–11]. Recently, PEEK and its composites have been widely used in orthopedic applications including the femoral component of total hip replacements, bone anchors, cervical total disc arthroplasty, intervertebral cages, dental implant systems and fracture fixation plates[12–14]. However, the mechanical properties of PEEK used in most loadbearing orthopedic implant should be further improved. Moreover, the traditional methods of stabilizing the prosthesis such as bone cement fixation are mostly substituted by bioactive fixation. The bioactive fixation technology requires prosthesis with good biocompatibility as well as excellent bioactivity. Lots of efforts have been devoted to improving the mechanical and bioactive properties of polyetheretherketone[15,16]. One of the most effective methods is to incorporate bioactive ceramics such as tricaclium phosphate, hydroxyapatite and bioactive glass-ceramics with PEEK polymer[17,18]. It should be noted that hydroxyapatite is an excellent candidate reinforcement and bioactive component for biocomposites on account of its similar constituent as that of inorganic component in natural bone and outstanding bioactive characteristics. Polymeric implant combination with HA promotes new bone growth from the existing bony walls (osteoconductive) onto it, thus stabilizing the prosthesis within a short period of time. In recent years, hydroxyapatite reinforced polyetheretherketone has been demonstrated to show
http://dx.doi.org/10.1016/j.jmst.2015.11.011 1005-0302/Copyright © 2015, The editorial office of Journal of Materials Science & Technology. Published by Elsevier Limited. All rights reserved.
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
35
Fig. 1. Schematic diagram of the functional gradient HA/PEEK biocomposites.
excellent bioactive performance[19]. Most of studies verified that the mechanical strength of PEEK can be importantly improved by the addition of certain content of hydroxyapatite particles in the PEEK matrix[20,21]. On the other hand, it is necessary to incorporate a larger amount of HA particles into the PEEK matrix to improve the bioactive properties of the composites. However, the excess of HA particles generally results in the brittleness of the composites accompanied with a significant deterioration of mechanical strength. Thus, there is conflict in simultaneously optimizing the mechanical and bioactive properties of hydroxyapatite reinforced polyetheretherketone biocomposites. The functional gradient design is an effective method to solve the conflict between mechanical and bioactive performances of the HA/PEEK biocomposites. Here, we selected hydroxyapatite particles as bioactive constituents as well as reinforcement and the polyetheretherketone as matrix to prepare functional gradient HA/ PEEK biocomposites. The functional gradient HA/PEEK biocomposite is composed of symmetrical five layers, which are respectively denoted as outer layer, middle layer and central layer (Fig. 1). The HA particle content in the functional gradient biocomposites presents increasing trend from central to the outer layer. The outer layer with the highest HA content endows the functional gradient biocomposites with outstanding bioactivity. The central layer with lowest HA content offers the functional gradient biocomposites appropriate mechanical strength and suitable ductility. The middle layer with moderate HA content makes the stress effectively transfer among the layers in the functional biocomposites. According to this design, the functional gradient HA/PEEK biocomposites with an outer layer containing high HA content and inner layer containing low HA contents would realize the simultaneous optimization on mechanical and bioactive properties of the biocomposites by altering the layer HA content in the composites. Currently, many works have attempted to improve the mechanical and bioactive properties of polyetheretherketone. However, it is a pending question to simultaneously optimize the mechanical and bioactive properties of PEEK. According to the authors’ knowledge, little attention has been devoted to solve the conflict between mechanical and bioactive properties of PEEK material through functional gradient design. The purpose of this initial investigation was to estimate the mechanical properties of the functional gradient HA/ PEEK biocomposites under bending load. The influence of various factors on the flexural mechanical properties was investigated.
purchased from SCM Industrial Chemical Co., Ltd. Calcium hydroxide (Ca(OH)2) and phosphoric acid (H3PO4) were produced and purchased from Siopharm Chemical Reagent Co., Ltd. All the chemical reagents were of analytical grade.
2. Experimental
2.3. Flexural mechanical property measurement
2.1. Materials
A three-point bending test was employed to evaluate the flexural properties such as flexural strength, flexural strain and modulus of functional gradient HA/PEEK biocomposites on mechanical test equipment (Model: CMT-5105, Shenzhen SANS Material Detection
Polyetheretherketone powder with the average diameter of 50 μm (VICTREX® PEEK, 450PF) was manufactured by VICTREX PIC, U.K. and
2.2. Preparation of functional gradient HA/PEEK biocomposites In order to fabricate functional gradient HA/PEEK biocomposites, the HA–PEEK composite powder were first prepared by in-situ synthesis process according to our previous work[22]. The main processing schedule is described as follows. First, a certain amount of Ca(OH)2 and PEEK powder was mixed in distilled water at 80 °C for 30– 50 min by an electric agitator. Then H3PO4 solution was slowly added into Ca(OH)2 and PEEK solution at 90 °C under electric agitator stirring. The amount of H3PO4 to be added was controlled with Ca/P molar ratio of 1.67. After Ca(OH)2 and H3PO4 had thoroughly reacted, the solution was aged at 80 °C for 10–12 h. Subsequently, the solution was rinsed and filtered repeatedly using distilled water until the pH value of the filtrate was close to 7. Finally, the residue was dried in an oven at 120 °C to constant weight. According to this procedure, HA–PEEK composite powder with various HA contents were harvested. Functional gradient HA/PEEK biocomposites were prepared by layer-by-layer casting method with the incorporation of hot pressure molding technology. First, HA–PEEK composite powder with different HA concentrations was symmetrically stacked into five layers in a steel die with rectangular shape. The HA concentration in the HA–PEEK composite powder presents an increasing trend from central to the outer layer along with the thickness direction. The mass of the HA–PEEK composite powder in the central layer is about twice the mass of the HA–PEEK composite powder in the middle and outer layers, respectively. Then the samples were heated with thermoforming equipment from room temperature to 380–390 °C and held there for 30–50 min to ensure that the PEEK in the composite powder was completely melted. Subsequently, 4–8 MPa pressure was applied to the sample and the pressure was kept for 15 min at temperature of 380–390 °C. After then, the temperature was lowered to 300 °C and then the pressure was removed. Finally, the sample was cooled down in air to room temperature. According to this process schedule, functional gradient HA/PEEK biocomposites could be fabricated. Fig. 1 illustrates the schematic diagram of functional gradient HA/PEEK biocomposites.
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
36
Co., Ltd.). The crosshead speed was 1 mm/min. Strip specimens of 70 mm in length and 25 mm in width were used. Three specimens were tested for each type of biocomposites and the average value of the flexural strength and modulus was calculated. It should be noted that the pure PEEK sample was not fractured during the bending test due to its excellent ductility. Taking this into consideration and according to the test standard (ISO 178:2003), the flexural strength of the pure PEEK sample was evaluated at 1.5 times of conventional deflection. According to the test standard (ISO 178:2003), flexural stress of the functional gradient biocomposites can be calculated using the following equation:
σf =
3FL 2bh 2
(1)
where σ f is the flexural stress, and F, L, b and h are the applied force, span, width and thickness of the sample, respectively. The flexural modulus of the functional gradient biocomposites can be calculated by the following equation:
Ef =
σ f2 − σ f1 ε f2 − ε f1
(2)
where E f is the flexural modulus and σ f2 and σ f1 are the flexural stress measured at flexural strain ε f2 and ε f1 .
3. Results and Discussion 3.1. Stress–strain behavior Fig. 2 shows the flexural stress–strain behavior of HA/PEEK and functional gradient HA/PEEK biocomposites with different hydroxyapatite contents. Some results can be concluded from Fig. 2. First, the flexural stress changing behavior of the HA/PEEK and functional gradient HA/PEEK biocomposites shows a linear increase with the rise of strain (Fig. 2(a, b)). Second, for all the functional gradient biocomposites, the stress yielding behavior of the biocomposites was not observed in the flexural stress–strain curves (Fig. 2(a, b)). Such results revealed that the HA/PEEK and functional gradient HA/ PEEK biocomposites exhibit linear elastic characteristics under the bending load. The fracture mechanism of the functional biocomposites is probably dominated by the brittle fracture. Lastly, it can be concluded from the insert graphs in Fig. 2(a, b) that the flexural break strain of the functional gradient HA/PEEK biocomposites presents decreasing trend with the rise of total HA content in the PEEK matrix. For example, the flexural break strains of the HA/PEEK and functional gradient HA/PEEK biocomposites decreased from 2.67% and 1.82% to 0.69% and 0.73%, respectively, while total HA content in the biocomposites increased from 10 wt% to 40 wt% (Fig. 2(a, b)). It is well known that the mechanical properties of the polymer/ceramic composites strongly depend on the mechanical properties of the inorganic components. Generally, the fracture strain of the ceramics such as hydroxyapatite and tricalcium phosphate reinforcements is much lower than that of the PEEK matrix due to the brittleness of the inorganic ceramics[23–25]. Then addition of inorganic ceramics in the PEEK matrix would result in the flexural break strain reduction. The change behavior of flexural break strain with total HA content in functional biocomposites indicates that the ductility decreases and brittleness increases with the rise of total HA content in the functional biocomposites. The effect of hydroxyapatite concentration difference between adjacent layers (HCDBAL) on the flexural stress–strain behavior is shown in Fig. 3. It can be obviously concluded from Fig. 3 that the change behavior of the flexural stress exhibits linear increasing trend
Fig. 2. Stress–strain behavior of HA/PEEK and functional gradient HA/PEEK biocomposites: (a) HA/PEEK, (b) functional gradient HA/PEEK with 10 wt% HCDBAL. (0/10/20 represents that the functional gradient biocomposite is composed of symmetric five layers. HA content in the outer layer is 20 wt%, HA content in the middle layer is 10 wt% and that in the central layer is 0 wt%. The total HA content in the sample is 10 wt%. HCDBAL is the abbreviation of hydroxyapatite concentration difference between adjacent layers.)
with the rise of strain under all the HCDBAL in the functional HA/ PEEK biocomposites. The change behavior of flexural stress with HCDBAL revealed that the HCDBAL would not change the elastic characteristic of the functional HA/PEEK biocomposites. Moreover, it can be found in the inset of Fig. 3 that the flexural break strain decreases with the rise of HCDBAL. The flexural break strain decreases from 1.53% to 1.15% while HCDBAL rises from 0 wt% to 15 wt%. Such a result indicates that the brittleness of the functional biocomposites increases with the rise of HCDBAL in the biocomposites. 3.2. Flexural strength Fig. 4 shows the effect of total hydroxyapatite content on the flexural strength and flexural break strain of the functional HA/PEEK biocomposites under various HCDBAL. It can be obviously concluded from Fig. 4 that both flexural strength and flexural break strain of the functional HA/PEEK biocomposites decrease with the rise of total HA content in the biocomposites under all the HCDBAL. For example, while HCDBAL is 5 wt%, the flexural strength and flexural
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
37
Fig. 3. Effect of HCDBAL on the flexural stress–strain behavior (total HA content in the PEEK matrix is 20 wt%; 15/20/25 represents that the functional gradient biocomposite is composed of symmetric five layers. HA content in the outer layer is 25 wt%, HA content in the middle layer is 20 wt% and that in the central layer is 15 wt%. The total HA content in the sample is 20 wt%. HCDBAL is the abbreviation of hydroxyapatite concentration difference between adjacent layers).
break strain of the functional gradient HA/PEEK biocomposites decreased from 103.32 MPa and 2.25% to 57.27 MPa and 0.86%, respectively, while total hydroxyapatite content in the functional gradient biocomposites increased from 5 wt% to 40 wt% (Fig. 4(b)). The change behavior of the flexural strength with the rise of total HA content is mainly attributed to two factors. One is ascribed to the inhomogeneous distribution and the agglomeration of HA particles in the PEEK matrix with the rise of total HA content in the functional gradient biocomposites, which would ultimately result in deterioration of the tensile strength[26]. The other arises from the deficient interfacial bonding strength between HA particles and PEEK matrix due to their poor physical compatibility. With the rise of total HA content, the adverse influence of the deficient interfacial bonding strength on the tensile strength was more significant and ultimately deteriorates the flexural mechanical properties of the functional gradient HA/PEEK biocomposites[27,28]. Thus, the deficient bonding strength between HA particles and PEEK matrix leads to the flexural strength deterioration with the rise of total HA content. Furthermore, comparing Fig. 4(a–c), it can be concluded that the flexural strength change behavior of the biocomposites with the rise of total HA content in the biocomposites is importantly influenced by HCDBAL. For example, while total HA content rises from 10 wt% to 40 wt%, the flexural strength of the uniform HA/PEEK biocomposites (0 wt% HCDBAL) decreases from 127.53 MPa to 46.48 MPa, almost decreasing by 64% (Fig. 4(a)). On the other hand, while total HA content rises from 10 wt% to 40 wt%, the flexural strengths of the functional gradient HA/PEEK biocomposites with 5 wt% and 10 wt% HDCBAL respectively decrease from 97.69 MPa and 96.13 MPa to 57.27 MPa and 60.10 MPa, separately decreasing by 41% and 38% (Fig. 4(b, c)). Such a result indicates that the flexural strength decreasing rate of the functional gradient HA/PEEK biocomposites with the rise of total HA content is obviously slower than that of uniform HA/PEEK biocomposites (0 wt% HDCBAL). That is, the existence of a gradient layer in the functional gradient HA/ PEEK biocomposites is beneficial to slowing down the flexural strength decline of the functional biocomposites with the rise of total HA content in the functional biocomposites. It should be noted that reducing the flexural strength decline rate with the rise of total HA content in the functional biocomposites is conducive to the
Fig. 4. Effect of total HA content on the flexural strength and flexural break strain of functional biocomposites under different HCDBAL: (a) 0 wt%, (b) 5 wt%, (c) 10 wt%.
mechanical and bioactive properties of the functional biocomposites simultaneous optimization. This result is consistent with our initial design idea to simultaneously optimize the mechanical and bioactive performances of the biocomposites through the design of gradient layers in the biocomposites. Moreover, it is interestingly found from Fig. 4 that the decline rate of flexural break strain of the functional gradient HA/PEEK
38
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
biocomposites with the rise of total HA content in the biocomposites is obviously less than that of uniform HA/PEEK biocomposites. For example, while total HA content in the uniform HA/PEEK biocomposites (0 wt% HCDBAL) rises from 10 wt% to 40 wt%, the flexural break strain decreases from 2.67% to 0.69%, almost decreasing by 74%. Conversely, while total HA content rises from 10 wt% to 40 wt%, the flexural break strains of the functional gradient HA/ PEEK biocomposites with 5 wt% and 10 wt% HDCBAL separately decrease from 1.89% and 1.82% to 0.87% and 0.73%, respectively, decreasing by 53% and 59%. Such a result reveals that the existence of a gradient layer in the functional gradient biocomposites could importantly delay the plasticity to brittleness transmission of the functional biocomposites with the rise of total HA content. Thus, it could also explain that the flexural strength decline rate of the functional gradient HA/PEEK biocomposites with the rise of total HA content is less than that of uniform HA/PEEK biocomposites. Fortunately, the delay of the plasticity to brittleness transmission of the functional biocomposites is benefical to resisting the catastrophic fracture of the orthpaedic implants. The effect of HCDBAL on the flexural strength of the functional gradient HA/PEEK biocomposites with different total HA content is shown in Fig. 5. It can be concluded that the effect of HCDBAL on the flexural strength of the functional biocomposites can be obviously divided into two regions. While the total HA content in the functional gradient biocomposites remains constant, the effect of HCDBAL at low value region (0–15 wt%) on the flexural strength of the functional biocomposites is obviously less than that at the high value region (15 wt%–20 wt%). For example, while HCDBAL rises from 0 to 15 wt%, the flexural strength of the functional HA/PEEK biocomposites with 20 wt% and 30 wt% total HA content shows little variation. On the other hand, while HCDBAL rises from 15 wt% to 20 wt%, the flexural strength of the functional HA/PEEK biocomposites with 20 wt% and 30 wt% total HA content respectively decreases from 76.98 MPa and 62.69 MPa to 67.74 MPa and 53.92 MPa, almost decreasing by 12% and 14%. The flexural mechanical properties of the functional gradient HA/ PEEK biocomposites importantly depended on the two factors. One is the flexural strength of the monolayer in the functional gradient biocomposites and the other is the efficiency of the stress
transmission among the interface between inner-layers in the functional gradient HA/PEEK biocomposites. According to the composite principle of the mechanical strength of the composites materials, the strength of the functional gradient HA/PEEK biocomposites can be expressed in the following equation. 5
σ fc = ∑ σ icv ic
(3)
i =1
where σ fc represents the flexural strength of the functional gradient HA/PEEK biocomposites and σ ic and v ic denote the flexural strength of the monolayer, which is composed of uniform HA/PEEK biocomposites, and the volume percent of the monolayer, respectively. It can be deduced from Eq. (3) that the higher the flexural strength of the monolayer is, the higher the flexural strength of the functional gradient HA/PEEK biocomposites is. While total HA content in the functional gradient biocomposites holds constant, in comparison with uniform HA/PEEK biocomposites (0 wt% HCDBAL), the functional gradient HA/PEEK biocomposites is composed of various monolayers with different HA contents, in which some monolayers contain lower HA content than uniform HA/PEEK and other monolayers contain higher HA content than uniform HA/PEEK composites. Just as aforementioned (Fig. 4(a)), the flexural strength of the uniform HA/PEEK biocomposites decreases with the rise of HA content. Then, in comparison with uniform HA/PEEK biocomposites, the flexural strength of the monolayer with lower HA content in the functional gradient biocomposites is higher than that of uniform HA/PEEK biocomposites. Conversely, the flexural strength of the monolayer with higher HA content in the functional gradient biocomposites is lower than that of uniform HA/PEEK. Based on Eq. (3), in comparison with uniform HA/PEEK biocomposites, the higher flexural strength of the monolayer would offset the lower flexural strength of the monolayer in the functional gradient HA/PEEK biocomposites. Thus, the flexural strength of the functional gradient HA/PEEK biocomposites has little change with the rise of HCDBAL while HCDBAL is in the low concentration region (0–15 wt%). On the other hand, the difference in the microstructure and components on the interface, which is composed of adjacent monolayer in the functional gradient biocomposites, becomes more and more obvious with the rise of HCDBAL. Such change behavior of the interface in the functional gradient HA/PEEK biocomposites would result in the ineffective stress transmission among the layers[29–31]. Thus, the flexural strength of the functional gradient biocomposites decreases while HCDBAL in the functional gradient biocomposites exceeds a certain concentration (larger than 15 wt%). According to the above discussion, it can be concluded that the mechanical properties of the monolayer in the functional gradient biocomposites predominated the flexural strength of the functional gradient HA/ PEEK biocomposites and the stress would effectively transform among monolayer while HCDBAL is in the low concentration region. While HCDBAL is in the high concentration region, the ineffective stress transmission among the monolayer controls the flexural strength of the functional gradient HA/PEEK biocomposites. Thus, the flexural strength of the functional gradient biocomposites decreases while HCDBAL exceeds a certain concentration. It should be noted that the change behavior of the flexural strength of the functional gradient HA/PEEK biocomposites with HCDBAL provides us an effective method to simultaneously optimize the mechanical and biological properties of the biocomposites through controlling the total HA content and HCDBAL in the functional gradient HA/PEEK biocomposites. 3.3. Flexural modulus
Fig. 5. Effect of HCDBAL on the flexural strength of functional gradient biocomposites with different total HA contents.
The effect of total HA content on the flexural modulus of the functional gradient biocomposites with various HCDBAL is shown in
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
39
flexural modulus of the functional gradient biocomposites first increases and then decreases with the rise of total HA content. On the other hand, while HCDBAL in the functional gradient biocomposites rises further, the flexural modulus of the biocomposites presents increasing trend with the rise of total HA content in the functional gradient biocomposites. For example, while total HA content in the functional gradient biocomposites rises from 5 wt% to 40 wt%, the flexural modulus of the functional gradient biocomposites with 5 wt% HCDBAL first increases from 4.18 GPa to 7.97 GPa and then decreases to 6.94 GPa. The flexural modulus of the functional gradient biocomposites reaches a maximum while total HA content is 35 wt% (Fig. 6(b)). As for the functional gradient HA/PEEK biocomposites with 10 wt% HCDBAL, the flexural modulus monotonically increases from 5.69 GPa to 8.88 GPa while total HA content increases from 10 wt% to 40 wt% (Fig. 6(c)). The flexural modulus reflects the capability of the biocomposites to resist the deformation under external flexural force. It is generally accepted that the higher the rigidity of the material, the higher the modulus of the material. The functional gradient HA/PEEK biocomposite is composed of two components such as inorganic hydroxyapatite reinforcement and polymeric polyetheretherketone matrix. The rigidity of the inorganic hydroxyapatite is much higher than that of the polymer PEEK matrix. Thus, the flexural modulus of the functional gradient biocomposites presents increasing trend with the rise of total HA content in the functional biocomposites. Fig. 7 shows the effect of HCDBAL on the flexural modulus of the functional gradient HA/PEEK biocomposites with two different total HA contents. It can be concluded that the change behavior of the flexural modulus with HCDBAL obviously relies on the total HA content in the functional gradient HA/PEEK biocomposites. While total HA content in the functional gradient biocomposites is 20 wt%, the flexural modulus of the functional gradient biocomposites increases with the rise of HCDBAL. The flexural modulus increases from 5.53 GPa to 6.60 GPa while HCDBAL increases from 0 to 20 wt%. Conversely, while total HA content in the functional gradient biocomposites is 30 wt%, the flexural modulus of the functional gradient biocomposites decreases with the rise of HCDBAL. The flexural modulus decreases from 9.09 GPa to 5.42 GPa while HCDBAL increases from 0 to 20 wt%.
Fig. 6. Effect of total HA content on the flexural modulus of the functional gradient HA/PEEK biocomposites: (a) 0 wt% HCDBAL, (b) 5 wt% HCDBAL, (c) 10 wt% HCDBAL.
Fig. 6. It can be concluded that the effect of total HA content on the flexural modulus of the functional gradient HA/PEEK biocomposites obviously depended on the level of HCDBAL. While HCDBAL in the functional gradient biocomposites varies from 0 wt% to 5 wt%, the
Fig. 7. Effect of HCDBAL on the flexural modulus of functional gradient HA/PEEK biocomposites.
Y. Pan et al. / Journal of Materials Science & Technology 32 (2016) 34–40
40
functional gradient biocomposites was lower than 5 wt%, the flexural modulus of the functional gradient biocomposites first increased and then decreased with the rise of total HA content. On the other hand, while HCDBAL in the functional gradient biocomposites was more than 5 wt%, the flexural modulus of the biocomposites presented increasing trend with the rise of total HA content in the functional gradient biocomposites. (4) The effect of HCDBAL on the flexural modulus also relied on the total HA content in the functional gradient HA/PEEK biocomposites. While total HA content in the functional gradient biocomposites was 20 wt%, the flexural modulus of the functional gradient biocomposites increased with the rise of HCDBAL. Conversely, while total HA content in the functional gradient biocomposites was 30 wt%, the flexural modulus of the functional gradient biocomposites decreased with the rise of HCDBAL.
Acknowledgment The authors gratefully acknowledge the financial support by the National Natural Science Foundation of China (No. 51175004). Fig. 8. Effect of HCDBAL on the flexural break strain of functional gradient HA/PEEK biocomposites.
The effect of HCDABL on the flexural break strain of the functional gradient biocomposites is shown in Fig. 8. It is interestingly found from Fig. 8 that the change behaviors of the flexural break strain and flexural modulus with HCDBAL present opposite trend. For example, while total HA content in the functional gradient biocomposites is 20 wt%, the flexural break strain of the functional gradient biocomposites decreases with the rise of HCDBAL. On the other hand, while total HA content in the functional gradient biocomposites is 30 wt%, the flexural break strain of the functional gradient biocomposites increases from 0.82% to 1.05% while HCDBAL increases from 0 to 20 wt%. It is well known that the flexural modulus reflects the ability of the material to resist flexural deformation under the flexural force loading. The lower the flexural modulus of the materials, the higher the flexibility of the materials. Thus, the change behavior of the flexural modulus and flexural strain with HCDBAL is opposite. 4. Conclusions (1) The flexural stress–strain behavior of functional gradient HA/ PEEK biocomposites presented linear elastic characteristics. Both the flexural strength and break strain of the functional gradient biocomposites decreased with the rise of total HA content. (2) The effect of HCDBAL on the flexural strength of the functional gradient HA/PEEK biocomposites obviously relied on the value of HCDBAL and total HA content in the functional gradient biocomposites. The higher the total HA content in the functional gradient biocomposites is, the less the influence degree of HCDBAL on the flexural strength is. (3) The effect of total HA content on the flexural modulus obviously depended on the HCDBAL. While HCDBAL in the
References [1] A. Jonas, R. Legras, J.P. Issi, Polymer 32 (1991) 3364–3370. [2] Y.C. Ke, Y.B. Zheng, Z.W. Wu, Chin. J. Mater. Res. 10 (1996) 205–209. [3] C.S. Michacl, E.M. James, G.B. Jennifer, M.K. Steve, M.R. Clare, Biomaterials 31 (2010) 9156–9162. [4] X.Z.K. Loy, S.K. Sinha, Wear 296 (2012) 681–692. [5] X.Q. Pei, B. Roland, B. Michael, Wear 304 (2013) 109–117. [6] A. Godara, D. Raabe, S. Green, Acta Biomater. 3 (2007) 199–220. [7] D.J. Jaekel, D.W. Macdonald, S.M. Kurtz, J. Mech. Behav. Biomed. Mater. 4 (2011) 1275–1282. [8] F.E.I. Halabi, J.F. Rodriguez, L. Rebolledo, E. Hurtos, M. Doblare, J. Mech. Behav. Biomed. Mater. 4 (2011) 1819–1832. [9] F. Heidenau, W. Mittelmeier, R. Detsch, M. Haenle, F. Stenzel, G. Ziegler, H. Gollwitzer, J. Mater. Sci. Mater. Med. 16 (2005) 883–888. [10] V.K. Raghunathan, J.N.A. Tettey, E.M. Ellis, M.H. Grant, J. Biomed. Mater. Res. A 88 (2009) 543–550. [11] X.Z. Mao, A.A. Wong, R.W. Crawford, Med. J. Aust. 194 (2011) 649–651. [12] S.M. Kurtz, J.N. Devine, Biomaterials 28 (2007) 4845–4869. [13] H. Xin, D. Shepherd, K.D. Dearn, Open Biomed. Eng. J. 6 (2012) 73–79. [14] M.G. Thomas, J.M. Hans, A.C. Janet, S. Jens, F. Bernhard, B. Wilhelm, J. Volkmar, Biomaterials 31 (2010) 523–531. [15] C.M. Han, E.J. Lee, H.E. Kim, Y.H. Koh, K.N. Kim, Y. Ha, S.U.S.U. Kuh, Biomaterials 31 (2010) 3465–3470. [16] C. Yao, D. Storey, T.J. Webster, Int. J. Nanomed. 2 (2007) 487–492. [17] L. Petrovic, D. Pohle, H. Munstedt, T. Rechtenwald, K. Schlegel, A.S. Rupprecht, J. Biomed. Sci. 13 (2006) 41–46. [18] G.M. Wu, W.D. Hsiao, S.F. Kung, Surf. Coat. Technol. 203 (2009) 2755–2758. [19] B.D. Hahn, D.S. Park, J.J. Choi, J. Ryu, W.H. Yoon, J.H. Choi, J.W. Kim, Appl. Surf. Sci. 283 (2013) 6–11. [20] L.C. Gabriel, W.M. Yue, R.K. Roeder, Biomaterials 28 (2007) 927–935. [21] A. Kalambettu, S. Dharmalingam, Mater. Chem. Phys. 147 (2014) 168–177. [22] Y.S. Pan, Q.Q. Shen, Y. Chen, Micro Nano Lett. 8 (2013) 357–361. [23] S. Kobayashi, T. Murakoshi, Adv. Compos. Mater. 23 (2014) 163–177. [24] M. Brillhart, B.L. Gregory, J. Botsis, Polymer 32 (1991) 1605–1611. [25] P.J. Rae, E.N. Brown, E.B. Orler, Polymer 48 (2007) 598–615. [26] Y.S. Pan, J. Wang, C.L. Pan, Micro Nano Lett. 7 (2012) 880–884. [27] S.M. Tang, P. Cheang, M.S. Abubakar, K.A. Khor, K. Liao, Int. J. Fatigue 26 (2004) 49–57. [28] T.L. Conrad, D.J. Jaekel, S.M. Kurtz, J. Biomed. Mater. Res. Part B Appl. Biomater. 101 (2013) 576–583. [29] B.S. Huang, G.H. Zhao, Z.H. Wang, Appl. Mech. Mater. 556–562 (2014) 725–728. [30] H. Qing, Comput. Mater. Sci. 89 (2014) 102–113. [31] Z.K. Lei, Z.L. Wang, W. Qiu, J. Adhes. Sci. Technol. 28 (2014) 1467–1479.