Mechanical, thermal and bio-compatibility studies of PAEK-hydroxyapatite nanocomposites

journal of the mechanical behavior of biomedical materials 49 (2015) 1 –11

Available online at www.sciencedirect.com

www.elsevier.com/locate/jmbbm

Research Paper

Mechanical, thermal and bio-compatibility studies of PAEK-hydroxyapatite nanocomposites Pratik Roy, R.R.N. Sailajan The Energy and Resources Institute (TERI), SRC, Bangalore 560071, India

art i cle i nfo

ab st rac t

Article history:

In this study high performance bone analogue has been developed using poly(aryl ether)

Received 14 January 2015

ketone, poly(dimethyl siloxane) and reinforced with nanohydroxyapatite as biocompatible

Received in revised form

filler. Compressive, tensile and flexural properties have shown sustained improvement up

17 April 2015

to 7% of nanohydroxyapatite loading. The mechanical properties were further analyzed

Accepted 20 April 2015

using micromechanical theories for good interfacial adhesion between matrix and filler.

Available online 28 April 2015

The composites are cytocompatible and revealed multiple layers of apatite formation in

Keywords:

simulated body fluid. The thickness of apatite layer increased with increase in nanohy-

Poly(aryl ether) ketone

droxyapatite loading in the composite. Poly(dimethyl siloxane) has been grafted with

Poly(dimethyl siloxane)

phosphate group to enhance compatibility with nanohydroxyapatite. Nanohydroxyapatite

Nanohydroxyapatite

has been treated with silane to enhance compatibility and facilitate dispersion in the

Mechanical properties

matrix as observed through transmission electron microscopy, scanning electron microscopy and X-Ray diffraction studies. & 2015 Elsevier Ltd. All rights reserved.

1.

Introduction

fillers in biocompatible or bio-inert matrices, which is necessary for promoting bone regeneration and forming strong interfacial

Hydroxyapatite is well known for its high biocompatibility and

fixation between host tissue and implant (Utzschneider et al.,

osteoconductivity. It is similar in chemical structure to the

2010). Composites with hydroxyapatite and a polymer such as

inorganic composition of human bone and thus it is often used

high-density polyethylene (HDPE), poly(methyl methacrylate)

in bone reconstruction, maxillofacial and periodontal surgery

(PMMA), Polyactive™, polylactic acid (PLLA), polyhydroxybuty-

(Hench, 1998). However, hydroxyapatite alone has certain short-

rate (PHB) and polyether ether ketone (PEEK) have been used as

comings such as poor load bearing properties, difficulty in

potential materials for bone tissue replacement (Abu Bakar

casting into the desired shape and its tendency to migrate from

et al., 2003). Among them, special attention has been focused on

the implanted sites (Nikpour et al., 2012; Yamaguchi et al.,

the biocomposites with the PEEK matrix. It has superior

2001). Polymers have been used to improve the mechanical

mechanical and chemical properties, such as high strength,

properties of hydroxyapatite (compressive strength, Young’s

good wear resistance, fracture toughness, fatigue properties

modulus and fracture toughness). Bioactive polymer compo-

along with excellent chemical resistance (Ha et al., 1997). PEEK

sites can be prepared by incorporating bioactive hydroxyapatite

can replace metals in load-bearing orthopaedic and spinal

n

Corresponding author. Tel: þ91 80 25356590. E-mail address: [email protected] (R.R.N. Sailaja).

http://dx.doi.org/10.1016/j.jmbbm.2015.04.022 1751-6161/& 2015 Elsevier Ltd. All rights reserved.

2

journal of the mechanical behavior of biomedical materials 49 (2015) 1 –11

implants due to biocompatibility, safety and biomechanical properties (Sagomonyants et al., 2008). They are also considered as the ideal material to apply in spinal cages and artificial joints (Williams, 2008). Moreover, PEEK has an elastic modulus which is in the same range as cortical bone which enables improved transfer of osteogenic strains to the bone tissue (Skinner, 1988). Radiolucency of PEEK enables improved post-operative assessment of fusion by eliminating radiographic artifacts caused by using metallic scaffolds and implants (Eck et al., 2000; Diedrich et al., 2001). Abu Bakar et al. (2003) developed PEEKhydoxyapatite composites as an alternative material for loadbearing orthopaedic applications and demonstrated that they possess sound bioactive properties. Yu et al. (2005) suggested that bioactivity of the composite increases with the increase in hydroxyapatite content. However, the tensile strength and the strain to failure of the composites decreased substantially with the addition of hydroxyapatite particles (Abu Bakar et al., 2003). Gabriel et al. (2006) prepared compression-molded PEEK composite reinforced with hydroxyapatite whiskers. Increased hydroxyapatite whisker reinforcement resulted in increased elastic modulus, but decreased ultimate tensile strength, strain-to-failure and work-to-failure. There are a couple of implantable devices based on PEEK and hydroxyapatite available in the market such as Zenivas by Solvay Plastics and hydroxyapatite coating on PEEK (Accentus Medical) in order to improve its bioactivity. In the present study, we have developed a high performance polyaryl ether ketone (PAEK) based nanocomposites with nanohydroxyapatite (nHA), polydimethyl siloxane (PDMS) and reinforced with nano carbon fibre (NCF). Dispersion of filler particles can improve interfacial adhesion and this leads to better stress sharing which in turn enhances the load bearing properties of the nanocomposites. Thus, in order to facilitate better dispersion and interfacial adhesion between the components of the composites, phosphate functionalization of PDMS and silane treatment of nHA and NCF has been carried out. Therefore, mechanical and thermal properties of these composites loaded with varied amounts of nHA have been studied. The cytotoxicity of the composite material and apatite formation in simulated body fluid (SBF) has also been examined.

2.

Experimental

2.1.

Materials

followed by drop wise addition of 5 ml of APTS. Then the reaction was allowed to occur in a locally fabricated microwave reactor (Enerzi Microwave Systems, India) for 1 h under reflux. After reaction, the modified nHA was separated by centrifugation (8000 rpm, 5 min) and washed with chloroform and absolute alcohol. The resulting modified nHA powder was dried in a vacuum oven at 60 1C for at least 24 h before use. Similar silane surface treatment was given to NCF to enhance dispersion and compatibility with the other components of the nanocomposites.

2.3.

Phosphate grafted PDMS (f-PDMS) was prepared by mixing MAEP and PDMS (10 ml of MAEP for 5 g of PDMS) in Brabender (Plasticorder, CMEI, 16CMESPL, East Germany) at 140 1C with rotor speed of 10 rpm for 5 min. The blend of PAEK/nHA/NCF and f-PDMS was prepared by melt mixing in Brabender (Plasticorder, CMEI, 16CMESPL, East Germany) at 372 1C with rotor speed of 20 rpm for 10 min by varying percentage of silane treated nHA (0%, 1%, 3%, 5%, 7%, 8% and 10%).

2.4.

2.2.

Silane treatment of nHA and NCF

10 g of nHA powder was mixed with 300 ml of dimethylformamide (DMF) and the mixture was stirred at reflux temperature (153 1C) under nitrogen atmosphere for 45 min. This was

Compression molding

The blends obtained from Brabender plasticorder were pressed into sheets in a compression molding machine (Compression molding press, Santec, India) at 18 MPa pressure and 372 1C. The heating time was kept at 10 min and the curing time at 20 min.

2.5.

Characterization

2.5.1.

Fourier transform infrared spectroscopy (FTIR)

The Fourier transform infrared spectroscopy (FTIR) spectra of nanocomposites were recorded between 600 and 4000 cm  1 using a Bruker ALPHA FT-IR Spectrometer. The samples were coated on a potassium bromide (KBr) plate and dried in a vacuum oven at 120 1C before it was tested.

2.5.2.

X-ray diffraction (XRD)

X-ray diffraction measurements (XRD) for the composites were performed using advanced diffractometer [PAN alytical, XPERT-PRO] equipped with Cu-Kα radiation source (X¼0.154 nm). The diffraction data were collected in the range of 2θ¼ 2– 601 using a fixed time mode with a step interval of 0.051.

2.5.3. PAEK (MW ¼133,100) powder used in this study was kindly gifted by Gharda Chemicals, Mumbai (India). PDMS (MW ¼1198), 2methacryloxy Ethyl Phosphate (MAEP), 3-aminopropyl triethoxy silane (APTS) and NCF used in this study was procured from Global Nanotech, Mumbai (India). nHA (particle size up to 20 nm) was purchased from J.K. Impex, Mumbai (India). All other chemicals were procured from S.d.Fine Chem, Bangalore (India) and were used as received.

Blend preparation

Scanning electron microscopy (SEM)

Scanning Electron Microscopy [SEM] (FEI QUANTA 200 microscope) was used to study the morphology of the fractured specimens. The specimens were gold sputtered prior to microcopy (JEOL, SM-1100E).

2.5.4.

Transmission electron microscopy (TEM)

Transmission electron microscopy (TEM) for nanocomposites was performed using a JEOL, Model 782, operating at 200 kV. TEM specimens were prepared by dispersing the composite powders in methanol by ultrasonication. A drop of the suspension was put on a TEM support grid (300 mesh copper grid coated with carbon). After drying in air, the composite powder remained attached to the grid and was viewed under the transmission electron microscope.

journal of the mechanical behavior of biomedical materials 49 (2015) 1–11

3

Fig. 1 – FTIR spectrum of PAEK, PDMS, nHA, NCF and composite with 1%, 3% and 10% nHA.

2.6.

Mechanical properties

2.6.1.

Tensile properties

The tensile properties of the nanocomposites were measured using Zwick UTM (Zwick Roell, ZHU, 2.5) with Instron tensile flat surface grips at a cross head speed of 5 mm/min. The tensile tests were performed as per ASTM: D 638 method. The specimens tested were of rectangular shape having length, width and thickness of 8 cm, 1.5 cm and 0.4 cm, respectively. A minimum of five specimens were tested for each variation in composition of the blend and results were averaged.

2.6.2.

Flexural properties

The flexural properties of the nanocomposites were measured using Zwick UTM (Zwick Roell, ZHU, 2.5) with a test

speed of 5 mm/min. The tests were performed as per ASTM: D 790-10 method. The samples were having a length of 5 cm, width of 1.5 cm and a thickness of 0.4 cm. At least five specimens were tested for each variation in composition of the blend and results were averaged.

2.6.3.

Compression properties

The compressive properties of the nanocomposites were performed as per ASTM: D 695 by Zwick UTM (Zwick Roell, ZHU, 2.5) with a pre load of 4.5 kN and a test speed of 5 mm/min. The samples were having a length of 5 cm, width of 1.5 cm and a thickness of 0.4 cm. A minimum of five specimens were tested for each variation in composition of the blend and the results were averaged.

4

journal of the mechanical behavior of biomedical materials 49 (2015) 1 –11

2.7.

Thermal analysis

2.7.1.

Thermogravimetric analysis (TGA)

Thermogravimetric analyses (TGA) were carried out for the nanocomposites using Perkin-Elmer Pyris Diamond 6000 analyzer [Perkin Elmer Inc., USA] in a Nitrogen atmosphere. The samples were subjected to a heating rate of 10 1C/min in a heating range of 30–600 1C with Al2O3 as reference material.

2.8.

Bioactivity

2.8.1.

In vitro bioactivity evaluation in simulated body fluid

In vitro bioactivity of the composites was determined by soaking them in a simulated body fluid (SBF). This solution was prepared by dissolving NaCl, NaHCO3, KCl, K2HPO4  3H2O, MgCl2  6H2O and CaCl2 in deionized water and buffered with (CH2OH)3CNH2 and

HCl (1 N) to adjust the pH value at 7.4, following the method described by Kokubo et al. (1990). PAEK composites with different loadings of nHA were immersed in SBF solution for 0 (control), 7, 14 and 28 days at human body temperature (37 1C). At the end of soaking time in SBF solution, the composite samples were removed and rinsed with deionized water and dried in a hot air oven to investigate for the formation of hydroxyapatite layer on the surface of the samples. The apatite formation was observed by using scanning electron microscopy.

2.9.

In vitro cytotoxicity test

2.9.1.

Cell line and culture medium

RAW 264.7 (Mouse, Macrophages) and L-929 (Mouse, connective tissue) cell line, were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% inactivated

Fig. 2 – XRD diffractograms of PAEK, PDMS, NCF, nHA and composite containing 1% and 8% nHA.

5

journal of the mechanical behavior of biomedical materials 49 (2015) 1–11

Fetal Bovine Serum (FBS), penicillin (100 IU/ml), streptomycin (100 mg/ml) and amphotericin B (5 mg/ml) in a humidified atmosphere of 5% CO2 at 37 1C until confluent. The cells were dissociated with TPVG solution (0.2% trypsin, 0.02% EDTA, 0.05% glucose in PBS) for L 929 and scrapped for RAW 264.7 with cell scraper. The stock cultures were grown in 25 cm2 culture flasks and all experiments were carried out in 96 microtitre plates.

3.

Results and discussion

3.1.

FTIR analysis

Determination of cell viability by MTT assay

The monolayer cell culture was trypsinized/scrapped with cell scrapper and the cell count was adjusted to 1.0  105 cells/ ml using DMEM containing 10% FBS. To each well of the 96 well microtitre plate, 0.1 ml of the diluted cell suspension (approximately 10,000 cells) was added. After 24 h, when a partial monolayer was formed, the supernatant was discarded and the monolayer was washed once with the culture medium. Different nanocomposites concentrations of 100 ml were added on to the partial monolayer in microtitre plates. The plates were then incubated at 37 1C for 24 h in 5% CO2 atmosphere. The microscopic examination for these samples was carried out and the observations were noted every 24 h. After 24 h, the nanocomposite solutions in the wells were discarded and 50 μl of MTT in phosphate buffered saline (PBS) was added to each well. The plates were gently shaken and incubated for 3 h at 37 1C in 5% CO2 atmosphere. The supernatant was removed and 100 μl of propanol was added and the plates were gently shaken to solubilize the formed formazan. The absorbance was measured using a micro plate reader at a wavelength of 540 nm and the percentage of cell death was calculated. The cell death is directly proportional to the levels of tumor necrosis factor alpha (TNF-α) in the cell supernatants. The percentage growth inhibition was calculated using the following formula and concentration of test

Relative Compressive Strenfth

% Growth inhibition ¼ 100
 mean OD of individual test group  100  mean OD of control group

4.0

Turcsanyi Experimental RCS Nicolais-Narkis

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0%

2%

4%

6%

8%

10%

12%

Percentage of nHA

Fig. 1 shows the FTIR analysis of the composites. The FTIR spectra of neat components are also given in the figure for comparison. The FTIR spectrum of neat PAEK depicts the characteristic C¼ O peak at 1650 cm  1 and its symmetric stretching at 928 cm  1 (Rath et al., 2007). The C–O stretching is seen at 1012 cm  1 and the peaks at 685, 764 and 848 cm  1 is due to –C–H stretching. For neat PDMS, the main stretching vibrations for Si–O, Si–OH and C–H bonds can be seen respectively at 866, 1013 and 2963 cm  1 (Hanoosh and Abdelrazaq, 2009). For silane treated nHA, the asymmetric stretching vibration absorption peak is seen at 2935 cm  1 while the bending vibration at 1676 cm  1 is attributed to –N–H bond. The stretching of amine group is seen at 3429 cm  1 (Zhang et al., 2005). The FTIR spectra for composites loaded with 1%, 3% and 10% nHA are also shown in the figure. In all these composites, the peak at 2963 cm  1 seen in neat PDMS was not observed, and the peak for amine Table 1 – Values of constants for micromechanical models for relative compressive strength (RCS). Model

a

b

Nicolais–Narkis Turcsanyi

5 7.2

 50  45

5.0

Relative Compressive Modulus

2.9.2.

drug needed to inhibit cell growth by 50% (CTC50) values is generated from the dose–response curves for each cell line.

Experimental RCM Kerner Sato-Furukawa Halpin-Tsai Modified Mori Tanaka

4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5

B- Value

0.0 0

9 8 7 6 5 4 3 2 1 0

2

4

6

8

10

12

Percentage of nHA in the composite

Fig. 4 – Plot of relative compressive modulus versus percentage of nHA in the composite.

Table 2 – Values of constants for micromechanical models for relative compressive modulus (RCM). 0%

2%

4%

6%

8%

10%

Percentage of nanohydroxyapatite

Fig. 3 – (a) Plot of relative compressive strength versus percentage of nHA in the composite. (b) Plot of B value versus percentage nHA in the composite.

12%

Model

a

b

Parameter

Value

Kerner Modified Halpin–Tsai Sato–Furukawa

6.5 5.7 35.5

 31  31  187

– – ξ

– – 0.1

6

journal of the mechanical behavior of biomedical materials 49 (2015) 1 –11

stretching at 3429 cm  1 was also not present indicating that the aminosilane functionalized nHA has undergone interactions with other components of the blend.

particles and thus a second order nonlinear equation is used as given below (Eq. (2)) (Kumar et al., 2011):   ∅0 ¼ a ∅ Af þ b ð∅ Af Þ2 ð2Þ

3.2.

In Eq. (2), Ø is the volume fraction of filler i.e. nHA and is determined as follows:

XRD analysis

The X-Ray diffractograms for the nanocomposites are given in Fig. 2. Neat PAEK has mainly four diffraction peaks at 2θ value of 1101, 1111, 2001 and 2111 (Lai et al., 2007). Neat nHA has a single crystalline peak at a 2θ value of 25.61 while the flexible PDMS is predominantly amorphous with a small broad peak at a 2θ value of 12.31 indicating that there are no hard segments. Similar observation has been made by Bai et al. (2008). NCF has a small broad peak and sharp crystalline peak 2θ values of 9.421 and 19.361 respectively. The diffractograms for nanocomposites loaded with 1% and 8% nHA are also shown in the figure. In both the nanocomposites, the crystalline peak of nHA at 2θ value of 25.61 and peak of NCF at lower 2θ value of 9.421 disappeared indicating that both nHA and NCF have exfoliated in the nanocomposite.

3.3.

Mechanical properties

3.3.1.

Relative compressive strength (RCS)

The relative compressive strength (compressive strength of the nanocomposite/compressive strength of the composite without nHA) of the nanocomposites are shown in Fig. 3(a). The compressive strength of the composite without nHA is 55.7 MPa. The RCS value increases as nHA content increased and reach an optimal value of 3.2 with 7% nHA loading. Higher loadings of nHA are detrimental to compressive strength values. Two commonly used micromechanical models have been used to further assess the interfacial interactions between the blend components. The first is the Nicolais and Narkis (1971) model which assumes that there exists no filler-matrix adhesion, and the equation is given below:   ð1Þ RCS ¼ 1 1:21 ∅02=3 In Eq. (1), Ø0 is the corrected volume fraction which takes into account the effect of size and aspect ratio of nHA

∅i ¼

Wi =ρi ΣWi =ρi

ð3Þ

where wi and ρi are the weight fraction and density of component i, respectively. The density values of PAEK, HA, PDMS, NCF are measured to be 1.3 g/cm3, 3.12 g/cm3, 1.13 g/cm3 and 2.1 g/cm3, respectively. The aspect ratio of nHA i.e. Af is taken to be 3.1 (Wang et al., 2002). The theoretical values obtained from the Nicolais–Narkis model do not match with the experimental values indicating that there exist interfacial interactions between the blend components. The other model is the Turcsanyi model which introduces a parameter, B, to assess interfacial adhesion. The expression for the Turcsanyi model (Turcsányi et al., 1988) is RCS ¼

1 ϕ0 expðBϕ0 Þ 1 þ 2:5 ϕ0

ð4Þ

The value of B has been determined by trial and error to match with the experimental results. The value of B has been calculated for each composition containing varied loading of nHA. A plot of B versus percentage nHA has been given in Fig. 3(b). The theoretical values from the Turcsanyi model also match closely with the experimental value as shown in Fig. 3(a). From Fig. 3(b), it can be seen, the B value is highest at 7% nHA loading. The optimal RCS value has also been obtained at 7% nHA loading. The higher the value of B, the better is the interfacial adhesion as reported similarly by Bliznakov et al. (2000). It has been envisaged that surface treatment of filler enhances interfacial adhesion as compared to that of the composites loaded with untreated fillers. The values of a and b for Eq. (2) for both the models have been given in Table 1.

3.3.2.

Relative compressive modulus (RCM)

The relative compressive modulus (modulus of the nanocomposite/modulus of the composite without nHA) of the

Fig. 5 – SEM micrograph showing compressive fracture of composite containing 0% nHA (a) and 7% nHA (b).

7

journal of the mechanical behavior of biomedical materials 49 (2015) 1–11

1000

250

Tensile Modulus

200

700 150

600 500

100

400 300

50

200

Flexural Modulus (MPa)

Tensile Strength

800

Tensile Strength (MPa)

Tensile Modulus (MPa)

900

1600

150

1400

130

1200

110

1000

90

800

70 50

600 Flexural Modulus

400

30

Flexural Strength

10

200

100 0 0%

2%

4%

6%

8%

10%

Flexural Strength (MPa)

Fig. 6 – TEM micrograph of nanocomposite containing 3% nHA (a) and 7% nHA (b).

0 12%

-10 12%

0 0%

2%

4%

6%

8%

10%

Percentage od nHA in the composite

Percentage of nHA in the composite

Fig. 8 – Plot of flexural strength and modulus versus percentage of nHA in the composite.

Fig. 7 – Plot of tensile strength and modulus versus percentage of nHA in the composite.

120

RCM ¼ 1 þ

ϕ0 15ð1 υÞ ϕm ð8 10υÞ

100 (a) (b)

Weight loss (%)

nanocomposites is shown in Fig. 4. The compressive modulus of the composite without nHA is 943.9 MPa. The RCM value doubled to an optimum value of 2.02 at 7% nHA loading. Addition of rigid hydroxyapatite particles has a stiffening effect on the chains leading to enhanced modulus values. Predictive theories have been used to further analyse the obtained experimental results. The first is the modified Kerner’s model (Kerner, 1956) for polymers incorporated with rigid spherical particles. The modified Kerner’s equation for estimating modulus is given in the below equation:

80

(c) (d)

PDMS (a)

60

(e)

PAEK (b) PAEK+PDMS (c) Composite with 1% nHA (d)

40

(f)

Composite with 8% nHA (e) Pure nHA (f)

20

0 0

ð5Þ

In Eq. (5), υ is the matrix Poison ratio taken to be 0.4 and Øm is the volume fraction of matrix without nHA. From Fig. 4, it can be noticed that the trend matches with the experimental values. The values of Kerner model show the theoretical upper bound values of RCM and are thus higher than the experimental values. Similar observation for ceramic metal composites has been reported by Hsieh and Tuan (2007). The second model is the widely popular Halpin–Tsai model (Halpin and Kardos, 1976; Ayatollahi et al., 2011; Lai et al., 2014; Kothmann et al., 2015) modified for nanocomposites and is given in the following equation:

100

200

300

400

500

600

700

800

Temperature (°C)

Fig. 9 – TGA thermograms of pure nHA, PDMS, PAEK and composites with different loadings of nHA.

RCM ¼

3 1 þ 2ϕ0 ηe 5 1 þ η T ϕ0 þ 8 1 ηe ϕ0 8 1 ηT ϕ0

ð6Þ

In Eq. (6) ηe ¼

δ 1 δ þ 2Af

and ηT ¼

δ 1 δþ2

ð7Þ

8

journal of the mechanical behavior of biomedical materials 49 (2015) 1 –11

In Eq. (7), δ is the ratio of filler modulus to matrix modulus (without nHA). The modulus of nHA has been taken to be 54,000 MPa (Wang et al., 2004). The theoretical values from the Halpin–Tsai model shows a good fit with the experimental values. The model assumes good matrix-filler adhesion. The trend and the optimal value also match with the observed results. The third model is the composition dependent Sato– Furukawa model which includes an adhesion parameter ξ. The ξ value varies from 0 to 1 for perfect to poor adhesion, respectively. The expression for the ;Sato–Furukawa model (Sato and Furu Kawa, 1963) is given in the following equation: " RCM ¼

1þ

! # ϕ0 ð2=3Þ ϕ0ð2=3Þψξ ð1 ψξÞ   1  ϕ01=3 ϕ0 2 2 ϕ0ð1=3Þ

where,

7Days

ð8Þ

 ψ¼

ϕ0 3



0

0

1 þ ϕ ð1=3Þ  ϕ ð2=3Þ 0 0 1 ϕ ð1=3Þ þ ϕ ð2=3Þ

! ð9Þ

The theoretical values for the Sato Furukawa model are also plotted in Fig. 4. This model gives a closer fit with the observed values as compared to either modified Kerner’s model or the Halpin–Tsai model. Further, ξ value of 0.1 suggests good interfacial interactions. The values for a and b along with the parameter for all the above models are given in Table 2. The surface treatment of fillers with silane introduces an organic amine group which interacts with the other components of the blend thereby, anchoring them together. The SEM morphology of the compressive fractured surfaces (0% nHA) shows brittle fractured surface with a large number of elongated voids formed by debonding of particles as shown in Fig. 5(a). The

14Days

28Days

Fig. 10 – SEM morphologies of the composite (A) (5% nHA), (B) (8% nHA) and (C) (10% nHA) in SBF, after 7, 14 and 28 days of immersion.

9

journal of the mechanical behavior of biomedical materials 49 (2015) 1–11

nanocomposite loaded with 7% nHA shows brittle fracture surface with a large number of small elongated voids uniformly spread in the entire surface [Fig. 5(b)] giving the look of porous surface. The uniform dispersion helped in stress sharing from matrix to filler leading to improved mechanical properties as shown in Figs. 3 and 4. The dispersion of nHA in the composite is seen clearly in the TEM micrographs shown in Fig. 6. Fig. 6(a) and (b) shows the TEM micrographs of nanocomposites loaded with 3% and 7% nHA, respectively. The dark regions indicate nHA are seen uniformly spread throughout the surface as both small and large agglomerates. However, these long streams of agglomerated particles favour load transfer mechanism at the interface as suggested by Dorigato and Pegoretti (2012).

Tensile and flexural properties

3.3.3.

Fig. 7 shows the tensile strength and tensile modulus of the nanocomposites. Tensile modulus reaches an optimum at 7% nHA loading and the value increased by 2.4 times higher than that of the modulus of composite without nHA. The rigid nHA particles have a stiffening effect and thus inhibit free movement of chains. A similar trend is observed for tensile strength of the nanocomposite. The maximum value is 3.3 times (for 7% nHA loading) higher than that of composite without nHA. The enhanced interfacial adhesion led to better stress transfer from matrix to filler leading to improved tensile properties. The nanocomposite loaded with 7% nHA exhibited the maximum flexural strength and modulus as shown in Fig. 8. The flexural strength of the composites showed a significant increase from 31.03 MPa (without nHA) to 107.35 MPa (with 7% nHA) indicating that the flexural strength improved more 35 RAW Cell line

30

L-929 Cell line

Cytotoxicity (%)

25 20 15 10 5 0

62.5

125

250

500

1000

Nanocomposite concentration (µg/ml)

Fig. 11 – Influence of PAEK-nHA composite on RAW-264.4 and L-929 cell lines metabolic activity as measured using MTT assay (data are presented as the means 72 standard deviation).

than three times. The flexural modulus also improved from 975.21 MPa (for 0% nHA) to 1387.17 MPa i.e. the flexural modulus improved by about one and a half times.

3.4.

Thermal analysis

3.4.1.

Thermogravimetric analysis (TGA)

Fig. 9 illustrates the thermogravimetric analyses of neat components and the composites. Pure PAEK shows an onset degradation temperature at 494.5 1C (with 4.4% weight loss and reaches a 46% weight loss at 697 1C). Neat PDMS shows degradation at 400 1C (2% weight loss) and 30% weight loss is observed at 697 1C due to cleavage of Si–CH3 bonds (Madsen et al., 2013) while nHA starts degrading at 397 1C (with 1.7% weight loss) and continues to degrade up to 478 1C (with 91.5% weight loss). The weight loss in the initial stage could be due to vaporization of water of crystallization while at higher temperatures, breakage of CO23  and HPO4 in hydroxyapatite occur (Zhang et al., 2005). The blend without nHA shows two-step degradation at 438 1C (with 0.7% weight loss) and 554 1C (with 18.2% weight loss). The nanocomposite loaded with 8% nHA shows improved thermal stability with an onset degradation temperature 525 1C, while 18% weight loss occurs only at 584 1C.

3.5.

Bioactivity

3.5.1.

In vitro bioactivity evaluation in simulated body fluid

Fig. 10(A–C) shows the morphology of SEM analysis of the surface of PAEK-nHA composite sheets (5%, 8% and 10% nHA) after 7, 14 and 28 days of incubation, respectively. SEM micrograph shows the apatite formation was present after 7 days of immersion in SBF. A bone-like apatite layer plays an important role in establishing the bone-bonding interface between biomaterials and the living tissue (Ge et al., 2004). Fig. 10(A) showed tiny crystal formation on composite surface which corresponds to apatite growth (Lu and Leng 2005). The size and number of the apatite particles formed on the composite C (10% nHA) was larger than those of the particles on the composite A and B (5% & 8% nHA). With the increase in immersion period in SBF, the thickness of apatite formation increased on composite surface to macroscopic size. After 28 days of immersion in SBF the entire surface of all the composites were covered with apatite crystals, though some larger particles were noticed on the surface of composite C (10% nHA). It has been observed that higher nHA loading and longer immersion in SBF facilitates apatite formation. Further, phosphate functionalized PDMS has also shown better affinity with nHA, thereby, promoting apatite formation.

Table 3 – Effect of test composite on lipopolysaccharide (LPS) induced TNF-α production (% Cell death) in RAW-264.7 cell line. Sample

Concentration tested (mg/ml)

% Cell death

PAEK composite with 10% nHA

1000 500

2.7670.89 4.2570.86

10

journal of the mechanical behavior of biomedical materials 49 (2015) 1 –11

3.6.

Cytotoxicity

3.6.1.

MTT assay

Biocompatibility of the prepared nanocomposites was examined by MTT assay. This is a colorimetric assay that measures the reduction of yellow 3-(4,5-dimethythiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) by mitochondrial enzyme succinate dehydrogenase. The MTT enters the cells and passes into the mitochondria where it is reduced to an insoluble, coloured (dark purple) formazan product. The number of viable cells was found to be proportional to the extent of formazan production by the cells used (Francis and Rita, 1986). In this study RAW-264.7 and L-929 cell lines were examined for cytotoxicity of varied concentration of the composite loaded with 10% nHA. Fig. 11 shows the percentage of cytotoxicity for both the cell lines. Percentage of cytotoxicity increased by 50.94% by increasing the concentration of nanocomposite from 62.5 mg/ml to 125 mg/ml. However, for higher concentration of nanocomposite specimen i.e. 500 mg/ml and 1000 mg/ml, the rise in cytotoxicity percentage is only 20.53% for RAW-264.7 cell line. However, for L-929 cell line the corresponding percentages of cytotoxicity levels have been found to increase by 61.96% and 41.53% respectively. Composite specimen with 10% nHA was found to be nontoxic against both cell lines with CTC50 showing more than 1000 mg/ml value. Percentage of cell death (RAW-264.7 cell line) was also calculated by measuring the production of TNF-α which showed a very negligible value (Table 3). Based on the above results it can be deduced that the nanocomposite is nontoxic.

4.

Conclusions

In the present study, high performance nanocomposites of PAEK/PDMS/nHA with various loadings of nHA (0% to 10%) have been developed. The tensile, compressive and flexural strength of the composites showed a sustained increase with increasing nHA loading up to 10%. SEM micrographs revealed that the nHA particles were uniformly dispersed in the PAEK/ PDMS matrix and no phase-separation was observed. XRD and FTIR analyses clearly confirm the presence of silane groups and exfoliation of the filler particles. The optimal nHA loading with enhanced mechanical strength is obtained with 7% nHA loading. Cytotoxicity test confirms that the developed composite is cytocompatible. The bioactive nanocomposite demonstrated ability to induce apatite formation in simulated body fluid. Hence, these composites show promise to be used for load bearing bone substitute applications.

Acknowledgements The authors are grateful to the Indian Council of Medical Research for kindly sanctioning funds to carry out this research work.

r e f e r e n c e s

Abu Bakar, M.S., Cheang, P., Khor, K.A., 2003. Tensile properties and microstructural analysis of spheroidized hydroxyapatitepoly(etheretherketone) biocomposites. Mater. Sci. Eng. A 345, 55–63. Abu Bakar, M.S., Cheng, M.H., Tang, S.M., Yu, S.C., Liao, K., Tan, C.T., Khor, K.A., Cheang, P., 2003. Tensile properties, tension–tension fatigue and biological response of polyetheretherketone–hydroxyapatite composites for loadbearing orthopedic implants. Biomaterials 24, 2245–2250. Abu Bakar, M.S., Cheang, P., Khor, K.A., 2003. Mechanical properties of injection molded hydroxyapatite– polyetheretherketone biocomposites. Compos. Sci. Technol. 63, 421–425. Ayatollahi, M.R., Shadlou, S., Shokrieh, M.M., Chitsazzadeh, M., 2011. Effect of multiwalled carbon-nanotube aspect ratio on the mechanical and electrical properties of epoxy-based nanocomposites. Polym. Test. 30, 548–556. Bai, C., Zhang, X., Dai, J., Wang, J., 2008. Synthesis of UV crosslinkable waterborne siloxane–polyurethane dispersion PDMS-PEDA-PU and the properties of the films. J. Coat. Technol. Res. 2008 (5), 251–257. Bliznakov, E.D., White, C.C., Shaw, M.T., 2000. Mechanical properties of blends of HDPE and recycled urea-formaldehyde resin. J. Appl. Polym. Sci. 77, 3220–3227. Diedrich, O., Perlick, L., Schmitt, O., Kraft, C.N., 2001. Radiographic characteristics on conventional radiographs after posterior lumbar interbody fusion: comparative study between radiotranslucent and radiopaque cages. J. Spinal Disord. 14 (6), 522–532. Dorigato, A., Pegoretti, A., 2012. Fracture behaviour of linear low density polyethylene–fumed silica nanocomposites. Eng. Fract. Mech. 29, 213–224. Eck, K.R., Lenke, L.G., Bridwell, K.H., Gilula, L.A., Lashgari, C.J., Riew, K.D., 2000. Radiographic assessment of anterior titanium mesh cages. J. Spinal Disord. 13 (6), 501–509. Francis, D., Rita, L., 1986. Rapid colorometric assay for cell growth and survival modifications to the tetrazolium dye procedure giving improved sensitivity and reliability. J. Immunol. Methods 89, 271–277. Gabriel, L.C., Weimin, Y., Ryan, K.R., 2006. Processing and tensile properties of hydroxyapatite-whisker-reinforced polyetheretherketone. Biomaterials 28, 927–935. Ge, Z., Baguenard, S., Lim, L.Y., Wee, A., Khor, E., 2004. Hydroxyapatite–chitin materials as potential tissue engineered bone substitutes. Biomaterials 25, 1049–1058. Ha, S.W., Kirch, M., Birchler, F., Eckert, K.L., Mayer, J., Wintermantel, E., Sittig, C., Klingenfuss, P.I., Textor, M., Spencer, N.D., Guecheva, M., Vonmont, H., 1997. Surface activation of polyetheretherketone (PEEK) and formation of calcium phosphate coatings by precipitation. J. Mater. Sci. Mater. Med. 11, 683–690. Halpin, J.C., Kardos, J.L., 1976. The Halpin–Tsai equations: a review. Polym. Eng. Sci. 16, 344–352. Hanoosh, W.S., Abdelrazaq, E.M., 2009. Polydimethyl siloxane toughened epoxy resins: tensile strength and dynamic mechanical analysis. Malays. Polym. J. 4, 52–61. Hench, L.L., 1998. Bioceramics. J. Am. Ceram. Soc. 81, 705–728. Hsieh, C.L., Tuan, W.H., 2007. Thermal expansion behavior of a model ceramic–metal composite. Mater. Sci. Eng. A 460-461, 453–458. Kerner, E.H., 1956. The elastic and thermo elastic properties of composite media. Proc. Phys. Soc. B 69, 808–813. Kokubo, T., Kushitani, H., Sakka, S., Kitsuqi, T., Yamamuro, T., 1990. Solutions able to reproduce in vivo surface-structure

journal of the mechanical behavior of biomedical materials 49 (2015) 1–11

changes in bioactive glass–ceramic A–W. J. Biomed. Mater. Res. 1990 (24), 721–734. Kothmann, M.H., Zeiler, R., Rios de Anda, A., Bru¨ckner, A, Altstsdt, V., 2015. Fatigue crack propagation behaviour of epoxy resins modified with silica-nanoparticles. Polymer 60, 157–163. Kumar, S., Singh, R.B., Sharma, P.K., Beniwal, R.S., Singh, R., 2011. Non-linear effect of volume fraction of inclusions on the effective thermal conductivity of composite materials: a modified Maxwell model. Sci. Res. 1, 10–18. Lai, S.M., Wu, S.H., Lin, G.G., Don, T.M., 2014. Unusual mechanical properties of melt-blended poly(lactic acid) (PLA)/clay nanocomposites. Eur. Polym. J. 52, 193–206. Lai, Y.H., Kuo, M.C., Huang, J.C., Chen, M., 2007. On the PEEK composites reinforced by surface-modified nano-silica. Mater. Sci. Eng. A 458, 158–169. Lu, X., Leng, Y., 2005. Theoretical analysis of calcium phosphate precipitation in simulated body fluid. Biomaterials 26, 1097–1108. Madsen, F.B., Dimitrov, I., Daugaard, A.E., Hvilsted, S., Skov, A.L., 2013. Novel cross-linkers for PDMS networks for controlled and well distributed grafting of functionalities by click chemistry. Polym. Chem. 4, 1700–1707. Nicolais, L., Narkis, M., 1971. Stress–strain behavior of styrene– aerylo nitrile/glass bead composites in the glassy region. Polym. Eng. Sci. 11, 194–199. Nikpour, M.R., Rabiee, S.M., Jahanshahi, M., 2012. Synthesis and characterization of hydroxyapatite/chitosan nanocomposite materials for medical engineering applications. Composites Part B 43, 1881–1886. Rath, T., Kumar, S., Mahaling, R.N., Mukherjee, M., Das, C.K., Pandey, K.N., Saxena, A.K., 2007. Flexible composite of PEEK and liquid crystalline polymer in presence of polyphosphazene. J. Appl. Polym. Sci. 104, 3758–3765. Sagomonyants, K.B., Smith, J.M.L., Devine, J.N., Aronow, M.S., Gronowicz, G.A., 2008. The in vitro response of human

11

osteoblasts to polyetheretherketone (PEEK) substrates compared to commercially pure titanium. Biomaterials 29, 1563–1572. Sato, Y., Furu Kawa, J., 1963. A molecular theory of filler reinforcement based on concept of internal deformation. Rubber Chem. Technol. 36, 1081–1091. Skinner, H.B., 1988. Composite technology for total hip arthroplasty. Clin. Orthop. Relat. Res. 235, 224–236. Turcsa´nyi, B., Puka´nszky, B., Tu¨do¨s, F., 1988. Composition dependence of tensile yield stress in filled polymers. J. Mater. Sci. Lett. 7, 160–162. Utzschneider, S., Becker, F., Grupp, T.M., Sievers, B., Paulus, A., Gottschalk, O., Jansson, V., 2010. Inflammatory response against different carbon fiber-reinforced PEEK wear particles compared with UHMWPE in vivo. Acta Biomater. 6, 4296–4304. Wang, C.X., Zhou, X., Wang, M., 2004. Influence of sintering temperatures on hardness and Young’s modulus of tricalcium phosphate bioceramic by nanoindentation technique. Mater. Charact. 52, 301–307. Wang, X., Li, Y., Wei, J., Groot, K.D., 2002. Development of biomimetic nano-hydroxyapatite/poly(hexamethylene adipamide) composites. Biomaterials 23, 4787–4791. Williams, D.F., 2008. On the mechanisms of biocompatibility. Biomaterials 29, 2941–2953. Yamaguchi, I., Tokuchi, K., Fukuzaki, H., Koyama, Y., Takakuda, K., Monma, H., 2001. Preparation and microstructure analysis of chitosan/hydroxyapatite nanocomposites. J. Biomed. Mater. Res. A 55, 20–27. Yu, S.C., Hariram, K.P., Kumar, R., Cheang, P., Aik, K.K., 2005. In vitro apatite formation and its growth kinetics on hydroxyapatite/polyetheretherketone biocomposites. Biomaterials 2005 (26), 2343–2352. Zhang, S.M., Liu, J., Zhou, W., Cheng, L., Guo, X.D., 2005. Interfacial fabrication and property of hydroxyapatite/ polylactide resorbable bone fixation composites. Curr. Appl. Phys. 5, 516–518.