Materials Science and Engineering C 39 (2014) 315–324
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Friction stir processing of magnesium–nanohydroxyapatite composites with controlled in vitro degradation behavior B. Ratna Sunil a, T.S. Sampath Kumar a,⁎, Uday Chakkingal a, V. Nandakumar b, Mukesh Doble b a b
Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600036, India Department of Biotechnology, Indian Institute of Technology Madras, Chennai 600036, India
a r t i c l e
i n f o
Article history: Received 24 August 2013 Received in revised form 1 February 2014 Accepted 1 March 2014 Available online 12 March 2014 Keywords: Friction stir processing Magnesium Nano-hydroxyapatite Biodegradation Metal matrix composites
a b s t r a c t Nano-hydroxyapatite (nHA) reinforced magnesium composite (Mg–nHA) was fabricated by friction stir processing (FSP). The effect of smaller grain size and the presence of nHA particles on controlling the degradation of magnesium were investigated. Grain refinement from 1500 μm to ≈3.5 μm was observed after FSP. In vitro bioactivity studies by immersing the samples in supersaturated simulated body fluid (SBF 5×) indicate that the increased hydrophilicity and pronounced biomineralization are due to grain refinement and the presence of nHA in the composite respectively. Electrochemical test to assess the corrosion behavior also clearly showed the improved corrosion resistance due to grain refinement and enhanced biomineralization. Using MTT colorimetric assay, cytotoxicity study of the samples with rat skeletal muscle (L6) cells indicate marginal increase in cell viability of the FSP-Mg–nHA sample. The composite also showed good cell adhesion. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Metallic biomaterials are mostly preferred for load bearing applications compared to ceramics and polymers. Especially for temporary tissue applications, magnesium based materials are the promising choice due to its degradability in physiological environment and load bearing capacity. This avoids the subsequent operation to remove the implant after healing of the fractured bone [1–4]. The mechanical properties of magnesium are also closer to natural bone which minimizes stress shielding effect that is generally associated with other metallic systems. But magnesium corrodes quickly in physiological environment and produces magnesium hydroxide by evolving hydrogen during the process of degradation [5,6]. If the rate of degradation is low, the produced hydrogen gas is absorbed by the tissue as it forms. If the degradation is too rapid, the gas may form a sub-facial pocket at the implant and tissue interface. However, it has been reported that the hydrogen gas evolved during the degradation of magnesium completely disappears after 6 weeks and generally does not cause complications [2,7]. But the material–tissue interactions that depend on the cell activities are affected due to the rapid degradation [8]. So, controlling the rapid degradation is the critical issue in developing magnesium based biodegradable implants. The degradation rate of magnesium can be altered by developing new alloys and composites, surface coatings and modifying the ⁎ Corresponding author at: Medical Materials Laboratory, Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600036, India. Tel.: +91 44 22574772. E-mail address: [email protected] (T.S. Sampath Kumar).
http://dx.doi.org/10.1016/j.msec.2014.03.004 0928-4931/© 2014 Elsevier B.V. All rights reserved.
microstructure [9–12]. Hydroxyapatite (HA), a calcium phosphate mineral phase, has attracted a great interest for the past two decades due to its excellent biocompatibility, bioactivity and osseointegration [13]. There are few reports on the fabrication of magnesium–HA composites from powder metallurgy route [14,15]. Also, considerable work has been done on coating the surface of different magnesium alloys with HA to improve the corrosion resistance [10]. However, as the degradation initiates from the surface of the magnesium, introducing HA into the surface can give better protection since it makes the substrate itself more bioactive. In the present paper, we demonstrate the use of friction stir processing (FSP), a solid-state processing technique developed from friction stir welding (FSW) [16] to disperse nano-hydroxyapatite (nHA) particles into pure magnesium to fabricate fine grained Mg–nHA composite. In FSP, a cylindrical rotating tool consisting of a small pin is inserted into the material surface and moved to cause dynamic recrystallization due to intense plastic deformation resulting in significant grain refinement [17]. The stirring action of the FSP tool can be used to fabricate metal matrix composites by incorporating secondary phase particles during the process. Mishra et al. [18] explained the composite fabrication using FSP in preparing 5083Al-based SiC reinforced surface composite and subsequently, significant work has been reported on the formation of different metal matrix composites (MMCs). Table 1 lists the brief development in magnesium based composites using FSP. As evident from the table, the fabrication of magnesium based composites by FSP are limited and in particular, studies targeted for biomedical applications are lacking. In the present work, Mg–nHA composites were prepared using the FSP process. The effects of the fine grain structure and
316
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Table 1 List of work carried out to develop magnesium based composites by friction stir processing in chronological order. S. no.
Composite
Study/observations
Reference
1 2 3 4 5 6 7
AZ91 Mg alloy/SiC powder AZ61 Mg alloy/nano-SiO2 Multi walled carbon nanotubes (MWCNTs)/AZ31 Mg alloy Al-rich thixoformed AZ91D Mg alloy AZ91 Mg alloy/SiC powder AZ31 Mg alloy/nano-Al2O3 AZ91 Mg alloy/SiC and Al2O3 powders
Lee et al. [19] Lee et al. [20] Morisada et al. [21] Chen et al. [22] Asadi et al. [23] Azizieh et al. [24] Asadi et al. [25]
8
i) AZ31B Mg alloy/carbon fibers ii) AZ91D Mg alloy/carbon fibers
Achieved fine grain structure and improved wear properties Grain refinement, high hardness, high strain rate and superplasticity Grain refinement and improved hardness Grain refinement and improved corrosion resistance Effect of processing parameters on microstructure. Effect of rotational speed and probe profile on microstructure and hardness Microstructural observations Hardness measurements Effect of the matrix characteristics, experimental parameters on the microstructural changes Fragmentation and distribution of the carbon fibers, recrystallization, solutionizing and precipitation
distributed nHA particles resulting from FSP on biomineralization, degradation and cytotoxicity to the rat skeletal muscle (L6) cells were investigated. The feasibility of the FSP technique to be adopted for biomedical applications is also discussed. 2. Materials and methods 2.1. Processing Friction stir processing was performed on commercially available pure magnesium (M/s. Exclusive magnesium, Hyderabad, India) sheets of size 100 × 100 × 5 mm cut from the billet. Prior to the processing, the sheets were annealed at 340 ºC for 30 min followed by furnace cooling. Annealed pure magnesium was coded as Mg. The chemical composition of the pure magnesium as confirmed by inductively coupled plasma atomic emission spectroscopy (ICP-AES, Optima 5300DV, PerkinElmer, USA) is 0.003% Al, 0.001% Zn, 0.002% Fe, 0.008%Mn and the remaining being magnesium in atomic percentage. The tool used in the present study is made of hardened H-13 tool steel consisting a tapered pin with a diameter varying from 3 to 5 mm over 2.7 mm length. The shoulder diameter of the tool is 15 mm. Optimized parameters were obtained by processing with different combinations of load, tool speed and its traverse speed to attain defect free stir zone. The traverse speed of rotating tool along the traverse axis was 12 mm/min with a rotating speed of 1200 rpm. 5000 N load was applied during the process. The processed sample was coded as FSP Mg. In order to produce surface composite, a shallow groove of 1 mm width and 2 mm depth was machined on the surface of the Mg sheets using milling cutter and the groove was filled with nHA powder. The nHA powder used in the present study was synthesized by microwave irradiation method as reported by Rameshbabu et al. [27]. Then FSP was carried out with the pin and tool plunged into the groove. Same processing parameters used for FSP Mg sample were adopted to produce the composite and coded as FSP-Mg–nHA. 2.2. Characterization Samples of size 20 × 10 × 5 mm were cut from the annealed sample and across the stir zone of FSPed samples for microstructural observations. They were then mechanically polished using emery papers up to 2000 grade and washed using ethanol. Further, the samples were polished using diamond paste (1–3 μm grit size) with the help of disk polishing machine (Binpol-VTD, Chennai Metco, India). Then the polished samples were etched with a solution comprised of 5 ml acetic acid, 5 g picric acid, 10 ml water and 100 ml ethyl alcohol for 20 to 60 s. After polishing and before etching, the samples were cleaned ultrasonically in ethanol to remove any residues generated from polishing. The microstructural observations were carried out using optical microscope (Vertimet-CP, Chennai Metco, India), scanning electron microscope (SEM, FEI Quanta 200, Netherlands) operated at 30 kV. Energy dispersive X-ray (EDS) mapping was also carried out to confirm the distribution of calcium and phosphorous in the FSP-Mg–nHA sample. For TEM
Martens et al. [26]
observations, samples were cut from the center of the processed zone and thin slices were cut from the samples using an automated slow speed saw. Then they were mechanically thinned to a thickness of 100 μm using fine graded emery papers. Then the thin disks were subjected to electro-polishing using a twin-jet electro-polishing facility using an electrolyte (mixture of 1% perchloric acid and 99% ethanol) until a hole was formed. The adjacent areas of these holes having a thickness less than 100 nm were examined by transmission electron microscope (TEM, Philips CM12, Holland) operated at 120 kV. Samples of weight 25 mg were cut from the stir zone as well as from the unprocessed regions and dissolved in 2% HNO3 aqueous solution. The elemental composition of the solution was analyzed by ICP-AES method to assess the possible dissolution of iron from the FSP tool into the magnesium sheets during the process. For wettability studies, all the samples were mechanically polished using emery papers up to 2000 grade, cleaned with ethanol, dried and the contact angles were measured (Easy DROP, KRUSS, Germany) using distilled water as the solvent at five different locations under ambient conditions. 2.3. In vitro bioactivity Specimens of size 10 × 10 × 5 mm were cut from the central part of the stirred zone and dried at 60 °C for 2 h after ultrasonic cleaning in ethanol. Super saturated concentrations (SBF 5 ×) have been used in the present study to accelerate the mineralization and to quickly assess the role of mineral phases on the degradation behavior of the samples. The samples were immersed in SBF 5 × and kept in a constant water bath at a temperature of 37 °C for 72 h to study the bioactivity. The ion concentrations of SBF 5 × are shown in Table 2 and compared with that of SBF as reported by Kokubo et al. [28]. The reagent grade chemicals (NaCl, NaHCO3, KCl, K2HPO4, MgCl2·6H2O, CaCl2 and Na2SO4 (Merck, India.)) with an appropriate weight were dissolved in deionized water as per the recommended sequence. Then the solution was buffered at pH = 7.4 with tris-hydroxymethyl aminomethane (TRIS) and appropriate amount of 1 M HCl at 37 °C [28]. Each sample was immersed in 50 ml of SBF 5× solution (the ratio of the SBF volume to the sample apparent surface area is more than 1:10). After different Table 2 Ion concentrations of the SBF 5×. Ion
Na+ K+ Mg2+ Ca2+ Cl− HCO− 3 HPO2− 4 2− SO4 pH
Ion concentrations (mM) Blood plasma
SBF
SBF 5×
142 5 1.5 2.5 103 27 1.0 0.5 7.2–7.4
142 5 1.5 2.5 147.8 4.2 1.0 0.5 7.4
710 25 7.5 12.5 739 21 5.0 2.5 7.4
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
intervals of time, the samples were removed from the SBF and gently rinsed with de-ionized water. The deposited phases on the surfaces of the samples after immersion were characterized by X-ray powder diffractometer (D8 DISCOVER, Bruker, USA) with Cu Kα radiation (λ = 1.54 Å) at a scanning rate of 1 step/s and a step size of 0.1°/step. SEM and EDS analysis of the immersed samples were also carried out.
317
3. Results 3.1. Microstructure
Samples of size 10 × 10 × 1 mm were exposed to rat skeletal muscle (L6) cells (National Centre for Cell Sciences, Pune, India) for 72 h to estimate the cytotoxicity quantitatively using MTT [3-(4,5dimethylthiazole-2-yl)-2,5-diphenyl tetrazolium bromide] colorimetric assay [30]. The methodology for carrying cytotoxicity and cell adhesion studies was reported in our earlier paper [31]. The experiments were carried in triplicate and the cytotoxicity was expressed as percent cell viability with respect to the control (polystyrene tissue culture plates). The attachment and proliferation of the cells were observed using SEM.
Fig. 1 shows the optical macro- and micro-images of Mg and bright field TEM image of nHA powder used in the present study. Before FSP, the average grain size of annealed pure magnesium (Mg) was measured as ≈1500 μm. In the macroscopic image, the grains appeared with different contrast (white, gray and black) after etching. The grain size was varying from few hundreds of micrometers (Fig. 1(b)) to less than 2 mm (Fig. 1(a)). The nHA particles were of acicular morphology with 15–20 nm width and 60–80 nm length (Fig. 1(c)). The SEM images of FSPed samples observed at the cross section (Fig. 2(a) and (b)) clearly indicate that FSP results in grain refinement up to 3.5 μm in both the FSP Mg and FSP-Mg–nHA composite. Fig. 2(b) shows the SEM image of a selected area from FSP-Mg–nHA and Fig. 2(c) to (g) shows the elemental maps corresponding to the selected area obtained by EDS. These elemental maps confirm that the white particles appearing in the selected area are composed of calcium, phosphorous and oxygen which indicate nHA. Typical photograph of the stir zone of FSP-Mg–nHA sample is shown in Fig. 3(a). The optical macroscopic image (Fig. 3(b)) obtained at the cross section of FSP-Mg–nHA sample shows typical depth and width of the stir zone, thermo-mechanical affected zone (TMAZ) and heat affected zone (HAZ). The distribution of nHA, near the surface (Fig. 3(b)) and the accumulation of more nHA beneath the surface (Fig. 3(e)) can be observed from the SEM image obtained at the cross section of FSP-Mg–nHA sample. The EDS analysis (shown in Fig. 3(c) and (e)) corresponding to the white particles indicates the presence of calcium and phosphorous which confirms the incorporation of nHA into the matrix. TEM images of FSPed samples shown in Fig. 4(a) and (b) indicate the grain refinement after FSP in both the samples. The embedded nHA particles represented by white arrows and dislocations as indicated by black arrows can be seen within the grains (Fig. 4(b) and (c)). Based on the morphology and size of individual nHA crystals shown in Fig. 1, the small particles shown in Fig. 4(b) and (c) were identified to be nHA individual crystals. Also, the corresponding SAED pattern (Fig. 4(d)) was indexed and this confirms the presence of nHA along with Mg. Even though agglomerated nHA particles are observed in SEM, this could not be observed in TEM. This is because during the sample preparation for TEM observations, these large agglomerated particles may not remain within the thin disks. But some individual nHA crystals may stay within the thin foil as observed in the TEM images. The area that is observed in TEM is also very small compared to SEM.
2.6. Statistical analysis
3.2. Wettability
The experiments were carried in triplicate to evaluate the difference in cell viability and the statistical analysis was carried out using oneway ANOVA analysis. The percent cell viability of the samples was shown as the mean ± standard deviation and a value of p b 0.05 was considered to be statistically significant.
The Mg, FSP Mg and FSP-Mg–nHA samples show variation in contact angles (θ) (Fig. 5). The Mg sample surface was found to be hydrophilic with water contact angles of 76.7° (±1.8, N = 5). The contact angles for FSP Mg and FSP-Mg–nHA samples were measured as 63.1° (±4.7, N = 5) and 62.2° (± 3.4, N = 5) respectively. The surface energy Es
2.4. Corrosion behavior 2.4.1. Electrochemical test The corrosion behavior of the samples was studied by potentiodynamic polarization tests (Model K0235, EG&G Princeton Applied Research, USA) conducted in SBF 5 × solution at room temperature (30 °C). The samples were metallographically polished using emery sheets up to 2000 grade and ultrasonically cleaned with ethanol. The counter electrode was made of graphite and the saturated calomel electrode (SCE) was used as the reference electrode. About one square cm area of the sample (working electrode) was exposed to the solution for 30 min prior to the beginning of the experiments to establish open circuit potential. Then, with a scanning rate of 5 mV/s, the test was completed between the potentials −2.5 to 0.5 V. 2.4.2. Immersion test The degradation behavior of the samples was studied by measuring the weight loss after immersing them in SBF 5 × (in a volume of 500 ml) at 37 °C for 24, 48 and 72 h (weight loss = (weight before immersion − weight after removing the corrosion products) / surface area) [29]. The samples were removed from SBF 5× and cleaned using boiling solution of 180 g/l chromic acid to remove the surface corrosion products. pH change in the solution at different intervals of time was also measured (pH 700, Eutech instruments, Singapore) to monitor the change of ionic concentration during the degradation of the samples. 2.5. Cytotoxicity and cell adhesion
Fig. 1. a) Optical macro-image, b) optical micro-image of as annealed pure magnesium, and c) TEM image of nano-hydroxyapatite.
318
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Fig. 2. SEM images observed at the cross section of a) FSP Mg and b) FSP-Mg–nHA (white arrows indicate large particles of nHA accumulated) and EDS mappings of Mg–nHA sample showing the distribution of individual elements: c) magnesium, d) calcium, e) phosphorous, and f) oxygen and g) the corresponding merged images.
has been calculated from the contact angles using the following equation [32]: Es ¼ Evl cosθ
ð1Þ
where Evl is the surface energy between water and air under ambient condition (i.e., 72.8 mJ/m2 at 20 °C) for pure water and θ is the static contact angle. The FSP Mg and FSP-Mg–nHA samples were found to have higher surface energy (32.84 ± 4.7 and 33.89 ± 3.8 respectively)
Fig. 3. a) Typical photograph of stir zone of FSP-Mg–nHA sample at the surface, b) optical macroscopic image of FSP-Mg–nHA sample at the cross section, c) SEM image of FSP-Mg–nHA observed at the cross section (near the surface) in high magnification, d) cross section in low magnification and e) beneath the surface in high magnification showing agglomerated nHA.
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
319
Fig. 4. TEM images of a) FSP Mg and b) FSP-Mg–nHA. c) Magnified image of FSP-Mg–nHA composite (white arrows indicate individual nHA crystals and black arrows indicate the dislocations) and d) the corresponding SAED pattern.
than Mg sample (16.73 ± 2.2) (Fig. 5). Surface energies are nearly identical for FSP Mg and FSP-Mg–nHA. 3.3. In vitro bioactivity Fig. 6 shows the SEM images of the samples immersed in SBF 5× and the corresponding EDS analysis of the chemical composition of the phases. Actually, EDS analysis does not give exact atomic ratio of the phases quantitatively formed on the surfaces. But a relative elemental composition in atomic percentages that is obtained from EDS analysis
Fig. 5. Surface energies of the samples calculated from the water contact angles (typical photographs of the water droplets on the surface of the samples are shown in the graph).
can be used to compare the level of mineralization on the samples immersed in SBF under the same conditions after different intervals of time. Mg shows wide degraded area with cracks over the surface (Fig. 6(a)) after 24 h of immersion. The magnified image (Fig. 6(d)) shows the deposited white phases. The chemical composition of the precipitates over Mg sample indicates the presence of Mg, Al, Zn, O, P, Ca, and Cl elements. The elements, Mg, Al, and Zn are originally from the sample and the Ca, P and Cl are from the SBF. FSP Mg has the deposition of white precipitates (Fig. 6(b)) and the magnified image shows Mg(OH)2 flakes along with precipitates in the shape of tiny clusters (Fig. 6(e)). The surface of FSP-Mg–nHA sample was covered by white precipitates of spherical morphology after 24 h of immersion (Fig. 6(c) and (f)). The presence of Ca and P was found to be more on FSPed samples and the effect was more on FSP-Mg–nHA sample. The XRD analysis (Fig. 7) confirms these spheres as hydroxyapatite (HA). As the immersion time increased to 48 h and 72 h, the surface of Mg was observed to be degraded and covered with discontinuous flakes of Mg(OH)2 (Fig. 6(g)). Also, from the corresponding EDS analysis, the presence of Ca and P indicate the deposition of apatite after 72 h. A thick and dense apatite formation was observed on many locations for FSP-Mg– nHA sample (Fig. 6(i)). However, the entire sample surface was not covered with apatite like coating within 72 h. But on FSP Mg sample, a combination of different morphologies appeared (Fig. 6(h)), which may be due to the presence of both Mg(OH)2 and HA. Fig. 8 shows the relative comparison of Ca, P and Cl elements from the EDS analysis on the surface of the samples after 24 h, 48 h and 72 h of immersion. The chlorine content was observed to be low as the deposition of Ca and P increased on both the FSPed samples. After 72 h of immersion, the Ca/P atomic ratio was calculated as 0.64, 0.95 and 1.35 for Mg, FSP Mg and Mg– nHA samples respectively. Peaks corresponding to (002), (121), (030), (130), (222), (123), (231) and (004) planes of HA were identified from the XRD patterns respectively (Fig. 7). These peaks are prominent for FSP-Mg–nHA.
320
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Fig. 6. SEM morphologies of the samples after in vitro bioactivity test: a) Mg after 24 h of immersion, b) FSP Mg after 24 h of immersion, c) FSP-Mg–nHA after 24 h of immersion, d), e) and f) are the magnified images of the corresponding samples, g) Mg after 72 h of immersion, h) FSP Mg after 72 h of immersion and i) FSP-Mg–nHA after 72 h of immersion.
3.4. Corrosion behavior 3.4.1. Electrochemical test Fig. 9(a) and (b) shows the potentiodynamic polarization curves of the samples before and after 72 h of immersion in SBF 5 × solution and the corresponding parameters are listed in Table 4. Compared to Mg sample, the corrosion potentials (Ecorr) of FSPed samples shifted towards more positive potential and the corrosion current density (Icorr) was lower than that of the Mg sample which indicates improved corrosion resistance for both the FSPed samples. Similar trend was observed for the samples after 72 h of immersion in SBF 5×.
Fig. 7. XRD patterns of the samples after immersing in SBF 5× for 72 h.
3.4.2. Immersion test The weight loss after different intervals of time for all the samples is shown in Fig. 9(c). The Mg sample undergoes more weight loss compared to FSP Mg and FSP-Mg–nHA composite, while the composite showed lower values than FSP Mg. From the pH measurements (Fig. 9(d)), determined for every 12 h, it can be understood that the increment was more for Mg compared to the other samples. The pH was increased and the rate of pH change was reduced due to the biomineralization on all of the samples as the immersion time increased to
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
321
Fig. 8. Elemental compositions of the samples obtained by EDS analysis after in vitro bioactivity test.
72 h. Both the FSP Mg and FSP-Mg–nHA have similar pH variation as the immersion time increased to 72 h. 3.5. Cytotoxicity and cell adhesion Fig. 10 shows the percentage viability of the L6 cells exposed to the samples for 72 h. All the materials showed negligible toxicity. Based on the statistical analysis (p b 0.05), there is no significant difference between Mg, FSP Mg and FSP-Mg–nHa samples with respect to cell viability. Fig. 11 shows the morphologies of the adhered L6 cells on the sample surfaces. Better adhesion of interconnected L6 cells on the FSPMg–nHA surface can be clearly observed when compared to other samples. 4. Discussion Melting of magnesium does not occur during FSP (i.e. temperature reached during the process is lower than 650 °C, which is the melting temperature of magnesium) [16,23]. Also, as reported by Rameshbabu et al. [27], nHA used in the present study is stable at this temperature. The crystallite size of nHA measured by using Scherrer's formula is around 32 nm. Agglomeration of nHA particles and more amount of nHA accumulation beneath the surface (Fig. 3) are due to the high surface energy associated with nanoparticles and the difference in the material flow in thickness direction during the process. Even though, the width and the depth of the groove produced on the workpiece before FSP to fill nHA powder is 1 mm and 2 mm respectively, the width of HA distribution will be nearly the same as to the width of the stir zone due to the characteristic of the material flow during FSP [16]. As the
FSP tool moves in the transverse direction, the material captured in the swirl zone beneath the tool pin undergoes a vortex flow around the pin from the front to the rear in the extrusion zone and this result in a non-uniform depth of the processed zone [33,34]. The crosssectional micrograph also shows skewed depth of the stir zone (Fig. 3(b)). The EDS analysis near the surface region as well as far below the surface at the boundary of the stir zone confirms the presence of HA particles (Fig. 3(d) and (e)). Hence, it can be concluded that the effective region of HA distribution in Mg after FSP is not limited to 1 mm width but distributed throughout the stir zone of around 12–15 mm width and 2–2.5 mm depth. TEM observations (Fig. 4) revealed that nHA was distributed within the grain as individual crystals. This is due to entrapped nHA crystals during the evolution of new grains due to dynamic recrystallization in FSP. Hence, from SEM and TEM observations, it can be concluded that the distribution of nHA in Mg matrix happened at different length scales varying from well dispersed crystals within the grains to agglomerates in the Mg matrix. It is evident from the ICP-AES results (Table 3) that the dissolution of Fe from FSP tool into the workpiece after FSP is almost negligible. The value of 0.003 wt.% Fe is well within the limit of 0.015 wt.% Fe which can lead to abnormal galvanic corrosion [35]. Wettability influences the quality of protein adsorption and cell adhesion on the implant surface [36]. In the present study, the effect of the surface roughness was negligible on the wettability as all the samples were mechanically polished using emery papers to induce the same level of roughness before performing the measurements. Therefore, the variation in contact angles is due to the difference in grain sizes only. Nearly similar surface energies were found for FSP Mg and FSPMg–nHA. These observations, indicate that the FSP induced grain
322
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Fig. 9. Corrosion behavior of the samples estimated by electrochemical test and immersion test: a) potentiodynamic polarization curves of the samples before immersion test, b) potentiodynamic polarization curves of the samples after 72 h of immersion in SBF 5×, c) weight loss comparison and d) pH change during the immersion test.
refinement causes the increased surface energy in FSPed samples as fine grain structure exhibits higher surface energy (due to availability of higher surface area) than coarse grained one [37,38]. From the in vitro bioactivity test, it can be understood that the localized corrosion was the reason behind the irregular pits along with large cracks of wide degraded area on the surface of Mg sample (Fig. 6(a)). A higher level of Cl observed along with the other elements is due to the formation of MgCl2 phase [6]. The results indicate the excellent bioactivity for FSPed samples, particularly for FSP-Mg–nHA sample. The nHA particles introduced into magnesium act as nucleation sites to initiate the crystallization of HA from the SBF [39]. The accelerated heterogeneous apatite crystallization in SBF achieved due to: (i) the presence of apatite nuclei by introducing nHA into the surface that avoided the necessity of apatite crystal nucleation and (ii) a surface with an optimum interfacial energy to initiate the nucleation [40,41]. The higher value of relative Ca/P ratio close to that of HA indicates the highly Table 3 Elemental composition of the samples in atomic percentage after and before FSP obtained from ICP-AES. Element
Mg Al Fe Mn Zn Ca P
As received (%)
99.99 0.0028 0.002 0.0008 0.0012 – –
After FSP (%) FSP Mg
Mg–nHA
99.98 0.002 0.001 0.0016 0.0016 – –
96.47 0.0018 0.003 0.00065 0.0007 2.184 1.336
bioactive nature of FSP-Mg–nHA composite compared to other. However, the low Ca/P ratios than the stoichiometric HA (1.67) for both the FSPed samples immersed for 72 h suggest the formation of magnesium phosphate as reported by Wang et al. [6] and in our earlier study [42]. Interestingly, FSP Mg shows peaks corresponding to magnesium phosphate, but not FSP-Mg–nHA. This may be due to the early formation of the apatite on FSP-Mg–nHA that reduced the formation of other phosphorous containing compounds. Also, the presence of lower phosphorous from the EDS analysis on FSP-Mg–nHA compared with FSP Mg (Fig. 8) after 72 h, suggest the insufficient quantity of precipitated magnesium phosphate on FSP-Mg–nHA sample, thus the corresponding peaks do not appear in XRD analysis. Reduced corrosion rate in the FSP-Mg–nHA compared to the other samples is due to the development of a quick passive layer [43] and reduced intensity of galvanic couple between grain interior and grain boundary as the grain size becomes smaller [44]. The improved corrosion behavior after bioactivity test and lower weight loss in immersion test for FSP-Mg–nHA composite among all of the samples are due to
Table 4 Electrochemical parameters of the samples obtained from the electrochemical test. Sample
Before/after biomineralization study (72 h)
Ecorr (V)
Icorr (A/cm2)
Mg FSP Mg Mg–nHA Mg FSP Mg Mg–nHA
Before
−1.88 −1.61 −1.57 −1.46 −1.38 −1.23
8.705 2.79 0.674 1.028 1.03 0.18
After
× × × × × ×
10−3 10−3 10−3 10−4 10−4 10−4
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Fig. 10. Cytotoxicity of the samples towards rat skeletal muscle (L6) cells expressed as cell viability using MTT assay (the percentage cell viability in each case was normalized to control. The control cells are the cells grown on standard polystyrene tissue culture plates at the same condition as that of other samples).
the enhanced biomineralization that reduced further degradation in the aggressive environment. Magnesium hydroxide (Mg(OH)2) is formed during the degradation of magnesium, and pH of the SBF solution increases due to the consumption of hydrogen ions and release of OH− ions as explained by Wang et al. [6,45]. Release of lower amounts of OH− ions due to lower degradation has decreased the rate of pH change in FSPed samples as
323
observed in the present study. For non-degradable metals such as commercially pure titanium, the pH change in SBF solution is negligible. But for the materials like magnesium, which degrade in SBF by releasing OH− ions; pH is increased and precipitation is initiated. As the pH increment was rapid in Mg compared with the FSPed samples, mineralization due to pH change should be more for the Mg sample. On the contrary, it is observed that the FSPed samples show more mineralization compared with Mg sample in spite of showing much lower increase in pH. Therefore, pH change alone does not increase the mineralization. Wettability plays a more important role in the nucleation and crystal growth of apatite during mineralization as explained by Das et al. [46]. Also, nHA in the Mg matrix facilitates heterogeneous nucleation and increases the rate of apatite crystal formation and growth. Hence the results indicate that while the pH variation has an influence on precipitation, the rapid mineralization achieved in FSPed samples at lower pH compared with Mg sample is due to high wettability and the presence of nHA in the Mg matrix. The MTT assay studies showed the negligible effect of all of the samples on cell viability. The FSP Mg has marginally reduced viability after 72 h. But statistically, no significant difference was observed for all the samples (p b 0.05). FSP Mg and FSP-Mg–nHA samples have shown better cell–cell connectivity and cell–material interactions compared to Mg. Among the three, FSP-Mg–nHA shows improved attachment and proliferation of the cells to cover wider surface area. The reasons behind the better adhesion of the cells on the FSP-Mg–nHA sample compared to other samples are high surface energy, embedded nHA particles and lower degradation rate. Surfaces with high wettability promote adsorption of proteins and cell adhesion to develop strong bond between the material and the tissue [47]. The HA embedded into the FSP-Mg– nHA surface enhances the osseointegration [12]. Lower degradation favors better cell adhesion and growth [8]. It is difficult for cells to
Fig. 11. SEM morphologies of the L6 cells on the surface of the samples incubated for 24 h: a) control (polystyrene tissue culture plate), b) Mg, c) FSP-Mg and d) FSP-Mg–nHA.
324
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
proliferate on rapidly degrading surface as observed on Mg (Fig. 11(b)). Better corrosion resistance observed for the FSP-Mg–nHA favors the cell adhesion compared to that of Mg. The above studies demonstrate that FSP can be used to fabricate biodegradable Mg–nHA composite for temporary orthopedic applications like locking plates, bone plates and fixtures. Moreover, as the process is a solid state technique; the problems of stability of the dispersing phase that is generally encountered in melting and sintering methods can be avoided in developing new composites. However, the technique involves few machining operations like machining of groove before FSP and cutting and machining to prepare the components after processing. The economical feasibility of the technique can be addressed by designing the experiments with multiple passes with a single setup to increase the productivity using optimized tool geometry, especially the shoulder diameter, pin profile and the length. 5. Conclusions Friction stir processing was adapted successfully to achieve grain refinement and to fabricate Mg–nHA metal matrix composite. The nHA particles were found to be dispersed in pure Mg after friction stir processing. Wettability was observed to be increased substantially for the composite due to the induced fine grain structure. It has been found that fine grain structure and incorporated nHA particles have a pronounced effect on biomineralization in SBF. FSP-Mg–nHA composite showed excellent bioactivity compared to Mg and owing to the early formation of mineral phases over the surface, the localized degradation due to pitting was also controlled. Furthermore, the FSP-Mg–nHA composite was nontoxic to L6 cells and showed superior cell adhesion which is promising for hard tissue applications. Acknowledgments Authors would like to thank Dr K Prasad Rao, IIT Madras for providing the NRB supported FSP facility. References [1] M.P. Staiger, A.M. Pietak, J. Huadmai, G. Dias, Magnesium and its alloys as orthopedic biomaterials: a review, Biomaterials 27 (2006) 1728–1734. [2] F. Witte, The history of biodegradable magnesium implants: a review, Acta Biomater. 6 (2010) 1680–1692. [3] G.L. Song, Control of biodegradation of biocompatible magnesium alloys, Corros. Sci. 49 (2007) 1696–1701. [4] F. Witte, J. Fischer, J. Nellesen, H.-A. Crostack, V. Kaese, A. Pisch, B. Felix, W. Henning, In vitro and in vivo corrosion measurements of magnesium alloys, Biomaterials 27 (2006) 1013–1018. [5] F. Witte, N. Hort, C. Vogt, S. Cohen, K.U. Kainer, R. Willumeit, F. Feyerabend, Degradable biomaterials based on magnesium corrosion, Curr. Opin. Solid State Mater. Sci. 12 (2008) 63–72. [6] Y. Wang, M. Wei, J. Gao, J. Hu, Y. Zhang, Corrosion process of pure magnesium in simulated body fluid, Mater. Lett. 62 (2008) 2181–2184. [7] E.D. McBride, Magnesium screw and nail transfixion in fractures, South. Med. J. 31 (50) (1938) 508–515. [8] K. Sigrid, J.G. Brunner, B. Fabry, S. Virtanen, Control of magnesium corrosion and biocompatibility with biomimetic coatings, J. Biomed. Mater. Res. B 96B (2011) 84–90. [9] H. Hornberger, S. Virtanen, A.R. Boccaccini, Biomedical coatings on magnesium alloys — a review, Acta Biomater. 8 (2012) 2442–2455. [10] H. Wang, Y. Estrin, Z. Zúberová, Bio-corrosion of a magnesium alloy with different processing histories, Mater. Lett. 62 (2008) 2476–2479. [11] S. Shaylin, G.J. Dias, Calcium phosphate coatings on magnesium alloys for biomedical applications: a review, Acta Biomater. 8 (2012) 20–30. [12] H. Wang, Y. Estrin, H.M. Fu, G.L. Song, Z. Zúberová, The effect of pre-processing and grain structure on the bio-corrosion and fatigue resistance of magnesium alloy AZ31, Adv. Eng. Mater. 9 (2007) 967–972. [13] W. Suchank, M. Yoshimura, Processing and properties of hydroxyapatite-based biomaterials for use as hard tissue replacement implants, J. Mater. Res. 13 (1998) 94–117. [14] X. Gu, W. Zhou, Y. Zheng, L. Dong, Y. Xi, D. Chai, Microstructure, mechanical property, bio-corrosion and cytotoxicity evaluations of Mg/HA composites, Mater. Sci. Eng. C 30 (2010) 827–832. [15] F. Witte, F. Feyerabend, P. Maier, J. Fischer, M. Stormer, C. Blawert, W. Dietzel, N. Hort, Biodegradable magnesium–hydroxyapatite metal matrix composites, Biomaterials 28 (2007) 2163–2174.
[16] R.S. Mishra, Z.Y. Ma, Friction stir welding and processing, Mater. Sci. Eng. R 50 (2005) 1–78. [17] R.S. Mishra, M.W. Mahoney, S.X. McFadden, N.A. Mara, A.K. Mukherjee, High strain rate super plasticity in a friction stir processed 7075 Al alloy, Scripta Mater. 42 (2000) 163–168. [18] R.S. Mishra, Z.Y. Ma, I. Charit, Friction stir processing: a novel technique for fabrication of surface composite, Mater. Sci. Eng. A 341 (2003) 307–310. [19] W.-B. Lee, C.-Y. Lee, M.-K. Kim, J.-I. Yoon, Y.-J. Kim, Y.-M. Yoen, S.-B. Jung, Microstructures and wear property of friction stir welded AZ91Mg/SiC particle reinforced composite, Compos. Sci. Technol. 66 (2006) 1513–1520. [20] C.J. Lee, J.C. Huang, P.J. Hsieh, Mg based nano-composites fabricated by friction stir processing, Scripta Mater. 54 (2006) 1415–1420. [21] Y. Morisada, H. Fujii, T. Nagaoka, K. Nogi, M. Fukusumi, Fullerene/A5083 composites fabricated by material flow during friction stir processing, Compos. Part A 38 (2007) 2097–2101. [22] T.-J. Chen, Z.-M. Zhu, Y.-D. Li, Y. Ma, Y. Hao, Friction stir processing of thixoformed AZ91D magnesium alloy and fabrication of Al-rich surface, Trans. Nonferr. Met. Soc. 20 (2010) 34–42. [23] P. Asadi, G. Faraji, M.K. Besharati, Producing of AZ91/SiC composite by friction stir processing, Int. J. Adv. Manuf. Technol. 51 (2010) 247–260. [24] M. Azizieh, A.H. Kokabi, P. Abachi, Effect of rotational speed and probe profile on microstructure and hardnessof AZ31/Al2O3 nanocomposites fabricated by friction stir processing, Mater. Des. 32 (2011) 2034–2041. [25] P. Asadi, G. Faraji, A. Masoumi, M.K. Besharatigivi, Experimental investigation of magnesium-base nanocomposite produced by friction stir processing: effects of particle types and number of friction stir processing passes, Metall. Mater. Trans. A 42A (2011) 2820–2832. [26] A. Mertens, A. Simar, H.-M. Montrieux, J. Halleux, F. Delannay, J. Lecomte-Beckers, Friction stir processing of magnesium matrix composites reinforced with carbon fibres: influence of the matrix characteristics and of the processing parameters on microstructural developments, in: W.J. Poole, K.U. Kainer (Eds.), Mg 2012. Proceedings of the 9th International Conference on Magnesium Alloys and Their Applications, July 8–12 2012, pp. 845–850, (Vancouver (Canada), http://hdl.handle.net/2268/ 120134 ). [27] N. Rameshbabu, K. Prasad Rao, T.S. Sampath Kumar, Accelerated microwave processing of nanocrystalline hydroxyapatite, J. Mater. Sci. 40 (2005) 6319–6323. [28] T. Kokubo, H. Takadama, How useful is SBF in predicting in vivo bone bioactivity? Biomaterials 27 (2006) 2907–2915. [29] L. Xu, E. Zhang, D. Yin, S. Zeng, K. Yang, In vitro corrosion behaviour of Mg alloys in a phosphate buffered solution for bone implant application, J. Mater. Sci. Mater. Med. 19 (2008) 1017–1025. [30] T. Mosmann, Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays, J. Immunol. Methods 65 (1983) 55–63. [31] B. Ratna Sunil, Arun Anil Kumar, T.S. Sampath Kumar, Uday Chakkingal, Role of biomineralization on the degradation of fine grained AZ31 magnesium alloy processed by groove pressing, Mater. Sci. Eng. C 33 (2013) 1607–1615. [32] R.N. Wenzel, Resistance of solid surfaces to wetting by water, Ind. Eng. Chem. 28 (1936) 988–994. [33] W.J. Arbegast, Modeling friction stir joining as a metalworking process, in: Z. Jin, A. Beaudoin, T.A. Bieler, B. Radhakrishnan (Eds.), Hot deformation of Aluminum Alloys III, TMS, Warrendale, PA, USA, 2003, pp. 313–327. [34] W.J. Arbegast, A flow-partitioned deformation zone model for defect formation during friction stir welding, Scripta Mater. 58 (2008) 372–376. [35] G.L. Song, A. Atrens, Corrosion mechanisms of magnesium alloys, Adv. Eng. Mater. 1 (1999) 11–33. [36] C.J. Wilson, R.E. Clegg, D.I. Leavensley, M.J. Pearcy, Mediation of biomaterial–cell interactions by adsorbed proteins: a review, Tissue Eng. 11 (2005) 1–18. [37] M. Hosokawa, K. Nogi, M. Naito, T. Yokoyama, Nanoparticle Technology Handbook, Elsevier, 2007. 19–21. [38] R. Chatopadhyay, Surface Wear: Analysis, Treatment, and Prevention, ASM International, Ohio, USA, 2001. 1–51. [39] B. Marc, L. Jacques, Can bioactivity be tested in vitro with SBF solution? Biomaterials 302 (2009) 175–2179. [40] J.A. Juhasz, S.M. Best, A.D. Auffret, W. Bonfield, Biological control of apatite growth in simulated body fluid and human blood serum, J. Mater. Sci. Mater. Med. 19 (2008) 1823–1829. [41] H.M. Kim, T. Himeno, M. Kawashita, T. Kokubo, T. Nakamura, The mechanism of biomineralization of bone-like apatite on synthetic hydroxyapatite: an in vitro assessment, J. R. Soc. Interface 1 (2004) 17–22. [42] B. Ratna Sunil, T.S. Sampath Kumar, Uday Chakkingal, Bioactive grain refined magnesium by friction stir processing, Mater. Sci. Forum 710 (2012) 264–269. [43] G.B. Hamu, D. Eliezer, L. Wagner, The relation between severe plastic deformation microstructure and corrosion behavior of AZ31 magnesium alloy, J. Alloys Compd. 468 (2009) 222–229. [44] G.R. Argade, S.K. Panigrahi, R.S. Mishra, Effects of grain size on the corrosion resistance of wrought magnesium alloys containing neodymium, Corros. Sci. 58 (2012) 145–151. [45] Y. Wang, M. Wei, J. Gao, Improve corrosion resistance of magnesium in simulated body fluid by dicalcium phosphate dihydrate coating, Mater. Sci. Eng. C 29 (2009) 1311–1316. [46] K. Das, S. Bose, A. Bandyopadhyay, Surface modifications and cell–materials interactions with anodized Ti, Acta Biomater. 3 (2007) 573–585. [47] L. Xiaomei, Y.L. Jung, J.D. Henry, D. Ravi, M.M. Andrea, A.V. Erwin, Influence of substratum surface chemistry/energy and topography on the human fetal osteoblastic cell line hFOB 1.19: phenotypic and genotypic responses observed in vitro, Biomaterials 28 (2007) 4535–4550
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Friction stir processing of magnesium–nanohydroxyapatite composites with controlled in vitro degradation behavior B. Ratna Sunil a, T.S. Sampath Kumar a,⁎, Uday Chakkingal a, V. Nandakumar b, Mukesh Doble b a b
Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600036, India Department of Biotechnology, Indian Institute of Technology Madras, Chennai 600036, India
a r t i c l e
i n f o
Article history: Received 24 August 2013 Received in revised form 1 February 2014 Accepted 1 March 2014 Available online 12 March 2014 Keywords: Friction stir processing Magnesium Nano-hydroxyapatite Biodegradation Metal matrix composites
a b s t r a c t Nano-hydroxyapatite (nHA) reinforced magnesium composite (Mg–nHA) was fabricated by friction stir processing (FSP). The effect of smaller grain size and the presence of nHA particles on controlling the degradation of magnesium were investigated. Grain refinement from 1500 μm to ≈3.5 μm was observed after FSP. In vitro bioactivity studies by immersing the samples in supersaturated simulated body fluid (SBF 5×) indicate that the increased hydrophilicity and pronounced biomineralization are due to grain refinement and the presence of nHA in the composite respectively. Electrochemical test to assess the corrosion behavior also clearly showed the improved corrosion resistance due to grain refinement and enhanced biomineralization. Using MTT colorimetric assay, cytotoxicity study of the samples with rat skeletal muscle (L6) cells indicate marginal increase in cell viability of the FSP-Mg–nHA sample. The composite also showed good cell adhesion. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Metallic biomaterials are mostly preferred for load bearing applications compared to ceramics and polymers. Especially for temporary tissue applications, magnesium based materials are the promising choice due to its degradability in physiological environment and load bearing capacity. This avoids the subsequent operation to remove the implant after healing of the fractured bone [1–4]. The mechanical properties of magnesium are also closer to natural bone which minimizes stress shielding effect that is generally associated with other metallic systems. But magnesium corrodes quickly in physiological environment and produces magnesium hydroxide by evolving hydrogen during the process of degradation [5,6]. If the rate of degradation is low, the produced hydrogen gas is absorbed by the tissue as it forms. If the degradation is too rapid, the gas may form a sub-facial pocket at the implant and tissue interface. However, it has been reported that the hydrogen gas evolved during the degradation of magnesium completely disappears after 6 weeks and generally does not cause complications [2,7]. But the material–tissue interactions that depend on the cell activities are affected due to the rapid degradation [8]. So, controlling the rapid degradation is the critical issue in developing magnesium based biodegradable implants. The degradation rate of magnesium can be altered by developing new alloys and composites, surface coatings and modifying the ⁎ Corresponding author at: Medical Materials Laboratory, Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600036, India. Tel.: +91 44 22574772. E-mail address: [email protected] (T.S. Sampath Kumar).
http://dx.doi.org/10.1016/j.msec.2014.03.004 0928-4931/© 2014 Elsevier B.V. All rights reserved.
microstructure [9–12]. Hydroxyapatite (HA), a calcium phosphate mineral phase, has attracted a great interest for the past two decades due to its excellent biocompatibility, bioactivity and osseointegration [13]. There are few reports on the fabrication of magnesium–HA composites from powder metallurgy route [14,15]. Also, considerable work has been done on coating the surface of different magnesium alloys with HA to improve the corrosion resistance [10]. However, as the degradation initiates from the surface of the magnesium, introducing HA into the surface can give better protection since it makes the substrate itself more bioactive. In the present paper, we demonstrate the use of friction stir processing (FSP), a solid-state processing technique developed from friction stir welding (FSW) [16] to disperse nano-hydroxyapatite (nHA) particles into pure magnesium to fabricate fine grained Mg–nHA composite. In FSP, a cylindrical rotating tool consisting of a small pin is inserted into the material surface and moved to cause dynamic recrystallization due to intense plastic deformation resulting in significant grain refinement [17]. The stirring action of the FSP tool can be used to fabricate metal matrix composites by incorporating secondary phase particles during the process. Mishra et al. [18] explained the composite fabrication using FSP in preparing 5083Al-based SiC reinforced surface composite and subsequently, significant work has been reported on the formation of different metal matrix composites (MMCs). Table 1 lists the brief development in magnesium based composites using FSP. As evident from the table, the fabrication of magnesium based composites by FSP are limited and in particular, studies targeted for biomedical applications are lacking. In the present work, Mg–nHA composites were prepared using the FSP process. The effects of the fine grain structure and
316
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Table 1 List of work carried out to develop magnesium based composites by friction stir processing in chronological order. S. no.
Composite
Study/observations
Reference
1 2 3 4 5 6 7
AZ91 Mg alloy/SiC powder AZ61 Mg alloy/nano-SiO2 Multi walled carbon nanotubes (MWCNTs)/AZ31 Mg alloy Al-rich thixoformed AZ91D Mg alloy AZ91 Mg alloy/SiC powder AZ31 Mg alloy/nano-Al2O3 AZ91 Mg alloy/SiC and Al2O3 powders
Lee et al. [19] Lee et al. [20] Morisada et al. [21] Chen et al. [22] Asadi et al. [23] Azizieh et al. [24] Asadi et al. [25]
8
i) AZ31B Mg alloy/carbon fibers ii) AZ91D Mg alloy/carbon fibers
Achieved fine grain structure and improved wear properties Grain refinement, high hardness, high strain rate and superplasticity Grain refinement and improved hardness Grain refinement and improved corrosion resistance Effect of processing parameters on microstructure. Effect of rotational speed and probe profile on microstructure and hardness Microstructural observations Hardness measurements Effect of the matrix characteristics, experimental parameters on the microstructural changes Fragmentation and distribution of the carbon fibers, recrystallization, solutionizing and precipitation
distributed nHA particles resulting from FSP on biomineralization, degradation and cytotoxicity to the rat skeletal muscle (L6) cells were investigated. The feasibility of the FSP technique to be adopted for biomedical applications is also discussed. 2. Materials and methods 2.1. Processing Friction stir processing was performed on commercially available pure magnesium (M/s. Exclusive magnesium, Hyderabad, India) sheets of size 100 × 100 × 5 mm cut from the billet. Prior to the processing, the sheets were annealed at 340 ºC for 30 min followed by furnace cooling. Annealed pure magnesium was coded as Mg. The chemical composition of the pure magnesium as confirmed by inductively coupled plasma atomic emission spectroscopy (ICP-AES, Optima 5300DV, PerkinElmer, USA) is 0.003% Al, 0.001% Zn, 0.002% Fe, 0.008%Mn and the remaining being magnesium in atomic percentage. The tool used in the present study is made of hardened H-13 tool steel consisting a tapered pin with a diameter varying from 3 to 5 mm over 2.7 mm length. The shoulder diameter of the tool is 15 mm. Optimized parameters were obtained by processing with different combinations of load, tool speed and its traverse speed to attain defect free stir zone. The traverse speed of rotating tool along the traverse axis was 12 mm/min with a rotating speed of 1200 rpm. 5000 N load was applied during the process. The processed sample was coded as FSP Mg. In order to produce surface composite, a shallow groove of 1 mm width and 2 mm depth was machined on the surface of the Mg sheets using milling cutter and the groove was filled with nHA powder. The nHA powder used in the present study was synthesized by microwave irradiation method as reported by Rameshbabu et al. [27]. Then FSP was carried out with the pin and tool plunged into the groove. Same processing parameters used for FSP Mg sample were adopted to produce the composite and coded as FSP-Mg–nHA. 2.2. Characterization Samples of size 20 × 10 × 5 mm were cut from the annealed sample and across the stir zone of FSPed samples for microstructural observations. They were then mechanically polished using emery papers up to 2000 grade and washed using ethanol. Further, the samples were polished using diamond paste (1–3 μm grit size) with the help of disk polishing machine (Binpol-VTD, Chennai Metco, India). Then the polished samples were etched with a solution comprised of 5 ml acetic acid, 5 g picric acid, 10 ml water and 100 ml ethyl alcohol for 20 to 60 s. After polishing and before etching, the samples were cleaned ultrasonically in ethanol to remove any residues generated from polishing. The microstructural observations were carried out using optical microscope (Vertimet-CP, Chennai Metco, India), scanning electron microscope (SEM, FEI Quanta 200, Netherlands) operated at 30 kV. Energy dispersive X-ray (EDS) mapping was also carried out to confirm the distribution of calcium and phosphorous in the FSP-Mg–nHA sample. For TEM
Martens et al. [26]
observations, samples were cut from the center of the processed zone and thin slices were cut from the samples using an automated slow speed saw. Then they were mechanically thinned to a thickness of 100 μm using fine graded emery papers. Then the thin disks were subjected to electro-polishing using a twin-jet electro-polishing facility using an electrolyte (mixture of 1% perchloric acid and 99% ethanol) until a hole was formed. The adjacent areas of these holes having a thickness less than 100 nm were examined by transmission electron microscope (TEM, Philips CM12, Holland) operated at 120 kV. Samples of weight 25 mg were cut from the stir zone as well as from the unprocessed regions and dissolved in 2% HNO3 aqueous solution. The elemental composition of the solution was analyzed by ICP-AES method to assess the possible dissolution of iron from the FSP tool into the magnesium sheets during the process. For wettability studies, all the samples were mechanically polished using emery papers up to 2000 grade, cleaned with ethanol, dried and the contact angles were measured (Easy DROP, KRUSS, Germany) using distilled water as the solvent at five different locations under ambient conditions. 2.3. In vitro bioactivity Specimens of size 10 × 10 × 5 mm were cut from the central part of the stirred zone and dried at 60 °C for 2 h after ultrasonic cleaning in ethanol. Super saturated concentrations (SBF 5 ×) have been used in the present study to accelerate the mineralization and to quickly assess the role of mineral phases on the degradation behavior of the samples. The samples were immersed in SBF 5 × and kept in a constant water bath at a temperature of 37 °C for 72 h to study the bioactivity. The ion concentrations of SBF 5 × are shown in Table 2 and compared with that of SBF as reported by Kokubo et al. [28]. The reagent grade chemicals (NaCl, NaHCO3, KCl, K2HPO4, MgCl2·6H2O, CaCl2 and Na2SO4 (Merck, India.)) with an appropriate weight were dissolved in deionized water as per the recommended sequence. Then the solution was buffered at pH = 7.4 with tris-hydroxymethyl aminomethane (TRIS) and appropriate amount of 1 M HCl at 37 °C [28]. Each sample was immersed in 50 ml of SBF 5× solution (the ratio of the SBF volume to the sample apparent surface area is more than 1:10). After different Table 2 Ion concentrations of the SBF 5×. Ion
Na+ K+ Mg2+ Ca2+ Cl− HCO− 3 HPO2− 4 2− SO4 pH
Ion concentrations (mM) Blood plasma
SBF
SBF 5×
142 5 1.5 2.5 103 27 1.0 0.5 7.2–7.4
142 5 1.5 2.5 147.8 4.2 1.0 0.5 7.4
710 25 7.5 12.5 739 21 5.0 2.5 7.4
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
intervals of time, the samples were removed from the SBF and gently rinsed with de-ionized water. The deposited phases on the surfaces of the samples after immersion were characterized by X-ray powder diffractometer (D8 DISCOVER, Bruker, USA) with Cu Kα radiation (λ = 1.54 Å) at a scanning rate of 1 step/s and a step size of 0.1°/step. SEM and EDS analysis of the immersed samples were also carried out.
317
3. Results 3.1. Microstructure
Samples of size 10 × 10 × 1 mm were exposed to rat skeletal muscle (L6) cells (National Centre for Cell Sciences, Pune, India) for 72 h to estimate the cytotoxicity quantitatively using MTT [3-(4,5dimethylthiazole-2-yl)-2,5-diphenyl tetrazolium bromide] colorimetric assay [30]. The methodology for carrying cytotoxicity and cell adhesion studies was reported in our earlier paper [31]. The experiments were carried in triplicate and the cytotoxicity was expressed as percent cell viability with respect to the control (polystyrene tissue culture plates). The attachment and proliferation of the cells were observed using SEM.
Fig. 1 shows the optical macro- and micro-images of Mg and bright field TEM image of nHA powder used in the present study. Before FSP, the average grain size of annealed pure magnesium (Mg) was measured as ≈1500 μm. In the macroscopic image, the grains appeared with different contrast (white, gray and black) after etching. The grain size was varying from few hundreds of micrometers (Fig. 1(b)) to less than 2 mm (Fig. 1(a)). The nHA particles were of acicular morphology with 15–20 nm width and 60–80 nm length (Fig. 1(c)). The SEM images of FSPed samples observed at the cross section (Fig. 2(a) and (b)) clearly indicate that FSP results in grain refinement up to 3.5 μm in both the FSP Mg and FSP-Mg–nHA composite. Fig. 2(b) shows the SEM image of a selected area from FSP-Mg–nHA and Fig. 2(c) to (g) shows the elemental maps corresponding to the selected area obtained by EDS. These elemental maps confirm that the white particles appearing in the selected area are composed of calcium, phosphorous and oxygen which indicate nHA. Typical photograph of the stir zone of FSP-Mg–nHA sample is shown in Fig. 3(a). The optical macroscopic image (Fig. 3(b)) obtained at the cross section of FSP-Mg–nHA sample shows typical depth and width of the stir zone, thermo-mechanical affected zone (TMAZ) and heat affected zone (HAZ). The distribution of nHA, near the surface (Fig. 3(b)) and the accumulation of more nHA beneath the surface (Fig. 3(e)) can be observed from the SEM image obtained at the cross section of FSP-Mg–nHA sample. The EDS analysis (shown in Fig. 3(c) and (e)) corresponding to the white particles indicates the presence of calcium and phosphorous which confirms the incorporation of nHA into the matrix. TEM images of FSPed samples shown in Fig. 4(a) and (b) indicate the grain refinement after FSP in both the samples. The embedded nHA particles represented by white arrows and dislocations as indicated by black arrows can be seen within the grains (Fig. 4(b) and (c)). Based on the morphology and size of individual nHA crystals shown in Fig. 1, the small particles shown in Fig. 4(b) and (c) were identified to be nHA individual crystals. Also, the corresponding SAED pattern (Fig. 4(d)) was indexed and this confirms the presence of nHA along with Mg. Even though agglomerated nHA particles are observed in SEM, this could not be observed in TEM. This is because during the sample preparation for TEM observations, these large agglomerated particles may not remain within the thin disks. But some individual nHA crystals may stay within the thin foil as observed in the TEM images. The area that is observed in TEM is also very small compared to SEM.
2.6. Statistical analysis
3.2. Wettability
The experiments were carried in triplicate to evaluate the difference in cell viability and the statistical analysis was carried out using oneway ANOVA analysis. The percent cell viability of the samples was shown as the mean ± standard deviation and a value of p b 0.05 was considered to be statistically significant.
The Mg, FSP Mg and FSP-Mg–nHA samples show variation in contact angles (θ) (Fig. 5). The Mg sample surface was found to be hydrophilic with water contact angles of 76.7° (±1.8, N = 5). The contact angles for FSP Mg and FSP-Mg–nHA samples were measured as 63.1° (±4.7, N = 5) and 62.2° (± 3.4, N = 5) respectively. The surface energy Es
2.4. Corrosion behavior 2.4.1. Electrochemical test The corrosion behavior of the samples was studied by potentiodynamic polarization tests (Model K0235, EG&G Princeton Applied Research, USA) conducted in SBF 5 × solution at room temperature (30 °C). The samples were metallographically polished using emery sheets up to 2000 grade and ultrasonically cleaned with ethanol. The counter electrode was made of graphite and the saturated calomel electrode (SCE) was used as the reference electrode. About one square cm area of the sample (working electrode) was exposed to the solution for 30 min prior to the beginning of the experiments to establish open circuit potential. Then, with a scanning rate of 5 mV/s, the test was completed between the potentials −2.5 to 0.5 V. 2.4.2. Immersion test The degradation behavior of the samples was studied by measuring the weight loss after immersing them in SBF 5 × (in a volume of 500 ml) at 37 °C for 24, 48 and 72 h (weight loss = (weight before immersion − weight after removing the corrosion products) / surface area) [29]. The samples were removed from SBF 5× and cleaned using boiling solution of 180 g/l chromic acid to remove the surface corrosion products. pH change in the solution at different intervals of time was also measured (pH 700, Eutech instruments, Singapore) to monitor the change of ionic concentration during the degradation of the samples. 2.5. Cytotoxicity and cell adhesion
Fig. 1. a) Optical macro-image, b) optical micro-image of as annealed pure magnesium, and c) TEM image of nano-hydroxyapatite.
318
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Fig. 2. SEM images observed at the cross section of a) FSP Mg and b) FSP-Mg–nHA (white arrows indicate large particles of nHA accumulated) and EDS mappings of Mg–nHA sample showing the distribution of individual elements: c) magnesium, d) calcium, e) phosphorous, and f) oxygen and g) the corresponding merged images.
has been calculated from the contact angles using the following equation [32]: Es ¼ Evl cosθ
ð1Þ
where Evl is the surface energy between water and air under ambient condition (i.e., 72.8 mJ/m2 at 20 °C) for pure water and θ is the static contact angle. The FSP Mg and FSP-Mg–nHA samples were found to have higher surface energy (32.84 ± 4.7 and 33.89 ± 3.8 respectively)
Fig. 3. a) Typical photograph of stir zone of FSP-Mg–nHA sample at the surface, b) optical macroscopic image of FSP-Mg–nHA sample at the cross section, c) SEM image of FSP-Mg–nHA observed at the cross section (near the surface) in high magnification, d) cross section in low magnification and e) beneath the surface in high magnification showing agglomerated nHA.
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
319
Fig. 4. TEM images of a) FSP Mg and b) FSP-Mg–nHA. c) Magnified image of FSP-Mg–nHA composite (white arrows indicate individual nHA crystals and black arrows indicate the dislocations) and d) the corresponding SAED pattern.
than Mg sample (16.73 ± 2.2) (Fig. 5). Surface energies are nearly identical for FSP Mg and FSP-Mg–nHA. 3.3. In vitro bioactivity Fig. 6 shows the SEM images of the samples immersed in SBF 5× and the corresponding EDS analysis of the chemical composition of the phases. Actually, EDS analysis does not give exact atomic ratio of the phases quantitatively formed on the surfaces. But a relative elemental composition in atomic percentages that is obtained from EDS analysis
Fig. 5. Surface energies of the samples calculated from the water contact angles (typical photographs of the water droplets on the surface of the samples are shown in the graph).
can be used to compare the level of mineralization on the samples immersed in SBF under the same conditions after different intervals of time. Mg shows wide degraded area with cracks over the surface (Fig. 6(a)) after 24 h of immersion. The magnified image (Fig. 6(d)) shows the deposited white phases. The chemical composition of the precipitates over Mg sample indicates the presence of Mg, Al, Zn, O, P, Ca, and Cl elements. The elements, Mg, Al, and Zn are originally from the sample and the Ca, P and Cl are from the SBF. FSP Mg has the deposition of white precipitates (Fig. 6(b)) and the magnified image shows Mg(OH)2 flakes along with precipitates in the shape of tiny clusters (Fig. 6(e)). The surface of FSP-Mg–nHA sample was covered by white precipitates of spherical morphology after 24 h of immersion (Fig. 6(c) and (f)). The presence of Ca and P was found to be more on FSPed samples and the effect was more on FSP-Mg–nHA sample. The XRD analysis (Fig. 7) confirms these spheres as hydroxyapatite (HA). As the immersion time increased to 48 h and 72 h, the surface of Mg was observed to be degraded and covered with discontinuous flakes of Mg(OH)2 (Fig. 6(g)). Also, from the corresponding EDS analysis, the presence of Ca and P indicate the deposition of apatite after 72 h. A thick and dense apatite formation was observed on many locations for FSP-Mg– nHA sample (Fig. 6(i)). However, the entire sample surface was not covered with apatite like coating within 72 h. But on FSP Mg sample, a combination of different morphologies appeared (Fig. 6(h)), which may be due to the presence of both Mg(OH)2 and HA. Fig. 8 shows the relative comparison of Ca, P and Cl elements from the EDS analysis on the surface of the samples after 24 h, 48 h and 72 h of immersion. The chlorine content was observed to be low as the deposition of Ca and P increased on both the FSPed samples. After 72 h of immersion, the Ca/P atomic ratio was calculated as 0.64, 0.95 and 1.35 for Mg, FSP Mg and Mg– nHA samples respectively. Peaks corresponding to (002), (121), (030), (130), (222), (123), (231) and (004) planes of HA were identified from the XRD patterns respectively (Fig. 7). These peaks are prominent for FSP-Mg–nHA.
320
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Fig. 6. SEM morphologies of the samples after in vitro bioactivity test: a) Mg after 24 h of immersion, b) FSP Mg after 24 h of immersion, c) FSP-Mg–nHA after 24 h of immersion, d), e) and f) are the magnified images of the corresponding samples, g) Mg after 72 h of immersion, h) FSP Mg after 72 h of immersion and i) FSP-Mg–nHA after 72 h of immersion.
3.4. Corrosion behavior 3.4.1. Electrochemical test Fig. 9(a) and (b) shows the potentiodynamic polarization curves of the samples before and after 72 h of immersion in SBF 5 × solution and the corresponding parameters are listed in Table 4. Compared to Mg sample, the corrosion potentials (Ecorr) of FSPed samples shifted towards more positive potential and the corrosion current density (Icorr) was lower than that of the Mg sample which indicates improved corrosion resistance for both the FSPed samples. Similar trend was observed for the samples after 72 h of immersion in SBF 5×.
Fig. 7. XRD patterns of the samples after immersing in SBF 5× for 72 h.
3.4.2. Immersion test The weight loss after different intervals of time for all the samples is shown in Fig. 9(c). The Mg sample undergoes more weight loss compared to FSP Mg and FSP-Mg–nHA composite, while the composite showed lower values than FSP Mg. From the pH measurements (Fig. 9(d)), determined for every 12 h, it can be understood that the increment was more for Mg compared to the other samples. The pH was increased and the rate of pH change was reduced due to the biomineralization on all of the samples as the immersion time increased to
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
321
Fig. 8. Elemental compositions of the samples obtained by EDS analysis after in vitro bioactivity test.
72 h. Both the FSP Mg and FSP-Mg–nHA have similar pH variation as the immersion time increased to 72 h. 3.5. Cytotoxicity and cell adhesion Fig. 10 shows the percentage viability of the L6 cells exposed to the samples for 72 h. All the materials showed negligible toxicity. Based on the statistical analysis (p b 0.05), there is no significant difference between Mg, FSP Mg and FSP-Mg–nHa samples with respect to cell viability. Fig. 11 shows the morphologies of the adhered L6 cells on the sample surfaces. Better adhesion of interconnected L6 cells on the FSPMg–nHA surface can be clearly observed when compared to other samples. 4. Discussion Melting of magnesium does not occur during FSP (i.e. temperature reached during the process is lower than 650 °C, which is the melting temperature of magnesium) [16,23]. Also, as reported by Rameshbabu et al. [27], nHA used in the present study is stable at this temperature. The crystallite size of nHA measured by using Scherrer's formula is around 32 nm. Agglomeration of nHA particles and more amount of nHA accumulation beneath the surface (Fig. 3) are due to the high surface energy associated with nanoparticles and the difference in the material flow in thickness direction during the process. Even though, the width and the depth of the groove produced on the workpiece before FSP to fill nHA powder is 1 mm and 2 mm respectively, the width of HA distribution will be nearly the same as to the width of the stir zone due to the characteristic of the material flow during FSP [16]. As the
FSP tool moves in the transverse direction, the material captured in the swirl zone beneath the tool pin undergoes a vortex flow around the pin from the front to the rear in the extrusion zone and this result in a non-uniform depth of the processed zone [33,34]. The crosssectional micrograph also shows skewed depth of the stir zone (Fig. 3(b)). The EDS analysis near the surface region as well as far below the surface at the boundary of the stir zone confirms the presence of HA particles (Fig. 3(d) and (e)). Hence, it can be concluded that the effective region of HA distribution in Mg after FSP is not limited to 1 mm width but distributed throughout the stir zone of around 12–15 mm width and 2–2.5 mm depth. TEM observations (Fig. 4) revealed that nHA was distributed within the grain as individual crystals. This is due to entrapped nHA crystals during the evolution of new grains due to dynamic recrystallization in FSP. Hence, from SEM and TEM observations, it can be concluded that the distribution of nHA in Mg matrix happened at different length scales varying from well dispersed crystals within the grains to agglomerates in the Mg matrix. It is evident from the ICP-AES results (Table 3) that the dissolution of Fe from FSP tool into the workpiece after FSP is almost negligible. The value of 0.003 wt.% Fe is well within the limit of 0.015 wt.% Fe which can lead to abnormal galvanic corrosion [35]. Wettability influences the quality of protein adsorption and cell adhesion on the implant surface [36]. In the present study, the effect of the surface roughness was negligible on the wettability as all the samples were mechanically polished using emery papers to induce the same level of roughness before performing the measurements. Therefore, the variation in contact angles is due to the difference in grain sizes only. Nearly similar surface energies were found for FSP Mg and FSPMg–nHA. These observations, indicate that the FSP induced grain
322
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Fig. 9. Corrosion behavior of the samples estimated by electrochemical test and immersion test: a) potentiodynamic polarization curves of the samples before immersion test, b) potentiodynamic polarization curves of the samples after 72 h of immersion in SBF 5×, c) weight loss comparison and d) pH change during the immersion test.
refinement causes the increased surface energy in FSPed samples as fine grain structure exhibits higher surface energy (due to availability of higher surface area) than coarse grained one [37,38]. From the in vitro bioactivity test, it can be understood that the localized corrosion was the reason behind the irregular pits along with large cracks of wide degraded area on the surface of Mg sample (Fig. 6(a)). A higher level of Cl observed along with the other elements is due to the formation of MgCl2 phase [6]. The results indicate the excellent bioactivity for FSPed samples, particularly for FSP-Mg–nHA sample. The nHA particles introduced into magnesium act as nucleation sites to initiate the crystallization of HA from the SBF [39]. The accelerated heterogeneous apatite crystallization in SBF achieved due to: (i) the presence of apatite nuclei by introducing nHA into the surface that avoided the necessity of apatite crystal nucleation and (ii) a surface with an optimum interfacial energy to initiate the nucleation [40,41]. The higher value of relative Ca/P ratio close to that of HA indicates the highly Table 3 Elemental composition of the samples in atomic percentage after and before FSP obtained from ICP-AES. Element
Mg Al Fe Mn Zn Ca P
As received (%)
99.99 0.0028 0.002 0.0008 0.0012 – –
After FSP (%) FSP Mg
Mg–nHA
99.98 0.002 0.001 0.0016 0.0016 – –
96.47 0.0018 0.003 0.00065 0.0007 2.184 1.336
bioactive nature of FSP-Mg–nHA composite compared to other. However, the low Ca/P ratios than the stoichiometric HA (1.67) for both the FSPed samples immersed for 72 h suggest the formation of magnesium phosphate as reported by Wang et al. [6] and in our earlier study [42]. Interestingly, FSP Mg shows peaks corresponding to magnesium phosphate, but not FSP-Mg–nHA. This may be due to the early formation of the apatite on FSP-Mg–nHA that reduced the formation of other phosphorous containing compounds. Also, the presence of lower phosphorous from the EDS analysis on FSP-Mg–nHA compared with FSP Mg (Fig. 8) after 72 h, suggest the insufficient quantity of precipitated magnesium phosphate on FSP-Mg–nHA sample, thus the corresponding peaks do not appear in XRD analysis. Reduced corrosion rate in the FSP-Mg–nHA compared to the other samples is due to the development of a quick passive layer [43] and reduced intensity of galvanic couple between grain interior and grain boundary as the grain size becomes smaller [44]. The improved corrosion behavior after bioactivity test and lower weight loss in immersion test for FSP-Mg–nHA composite among all of the samples are due to
Table 4 Electrochemical parameters of the samples obtained from the electrochemical test. Sample
Before/after biomineralization study (72 h)
Ecorr (V)
Icorr (A/cm2)
Mg FSP Mg Mg–nHA Mg FSP Mg Mg–nHA
Before
−1.88 −1.61 −1.57 −1.46 −1.38 −1.23
8.705 2.79 0.674 1.028 1.03 0.18
After
× × × × × ×
10−3 10−3 10−3 10−4 10−4 10−4
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
Fig. 10. Cytotoxicity of the samples towards rat skeletal muscle (L6) cells expressed as cell viability using MTT assay (the percentage cell viability in each case was normalized to control. The control cells are the cells grown on standard polystyrene tissue culture plates at the same condition as that of other samples).
the enhanced biomineralization that reduced further degradation in the aggressive environment. Magnesium hydroxide (Mg(OH)2) is formed during the degradation of magnesium, and pH of the SBF solution increases due to the consumption of hydrogen ions and release of OH− ions as explained by Wang et al. [6,45]. Release of lower amounts of OH− ions due to lower degradation has decreased the rate of pH change in FSPed samples as
323
observed in the present study. For non-degradable metals such as commercially pure titanium, the pH change in SBF solution is negligible. But for the materials like magnesium, which degrade in SBF by releasing OH− ions; pH is increased and precipitation is initiated. As the pH increment was rapid in Mg compared with the FSPed samples, mineralization due to pH change should be more for the Mg sample. On the contrary, it is observed that the FSPed samples show more mineralization compared with Mg sample in spite of showing much lower increase in pH. Therefore, pH change alone does not increase the mineralization. Wettability plays a more important role in the nucleation and crystal growth of apatite during mineralization as explained by Das et al. [46]. Also, nHA in the Mg matrix facilitates heterogeneous nucleation and increases the rate of apatite crystal formation and growth. Hence the results indicate that while the pH variation has an influence on precipitation, the rapid mineralization achieved in FSPed samples at lower pH compared with Mg sample is due to high wettability and the presence of nHA in the Mg matrix. The MTT assay studies showed the negligible effect of all of the samples on cell viability. The FSP Mg has marginally reduced viability after 72 h. But statistically, no significant difference was observed for all the samples (p b 0.05). FSP Mg and FSP-Mg–nHA samples have shown better cell–cell connectivity and cell–material interactions compared to Mg. Among the three, FSP-Mg–nHA shows improved attachment and proliferation of the cells to cover wider surface area. The reasons behind the better adhesion of the cells on the FSP-Mg–nHA sample compared to other samples are high surface energy, embedded nHA particles and lower degradation rate. Surfaces with high wettability promote adsorption of proteins and cell adhesion to develop strong bond between the material and the tissue [47]. The HA embedded into the FSP-Mg– nHA surface enhances the osseointegration [12]. Lower degradation favors better cell adhesion and growth [8]. It is difficult for cells to
Fig. 11. SEM morphologies of the L6 cells on the surface of the samples incubated for 24 h: a) control (polystyrene tissue culture plate), b) Mg, c) FSP-Mg and d) FSP-Mg–nHA.
324
B. Ratna Sunil et al. / Materials Science and Engineering C 39 (2014) 315–324
proliferate on rapidly degrading surface as observed on Mg (Fig. 11(b)). Better corrosion resistance observed for the FSP-Mg–nHA favors the cell adhesion compared to that of Mg. The above studies demonstrate that FSP can be used to fabricate biodegradable Mg–nHA composite for temporary orthopedic applications like locking plates, bone plates and fixtures. Moreover, as the process is a solid state technique; the problems of stability of the dispersing phase that is generally encountered in melting and sintering methods can be avoided in developing new composites. However, the technique involves few machining operations like machining of groove before FSP and cutting and machining to prepare the components after processing. The economical feasibility of the technique can be addressed by designing the experiments with multiple passes with a single setup to increase the productivity using optimized tool geometry, especially the shoulder diameter, pin profile and the length. 5. Conclusions Friction stir processing was adapted successfully to achieve grain refinement and to fabricate Mg–nHA metal matrix composite. The nHA particles were found to be dispersed in pure Mg after friction stir processing. Wettability was observed to be increased substantially for the composite due to the induced fine grain structure. It has been found that fine grain structure and incorporated nHA particles have a pronounced effect on biomineralization in SBF. FSP-Mg–nHA composite showed excellent bioactivity compared to Mg and owing to the early formation of mineral phases over the surface, the localized degradation due to pitting was also controlled. Furthermore, the FSP-Mg–nHA composite was nontoxic to L6 cells and showed superior cell adhesion which is promising for hard tissue applications. Acknowledgments Authors would like to thank Dr K Prasad Rao, IIT Madras for providing the NRB supported FSP facility. References [1] M.P. Staiger, A.M. Pietak, J. Huadmai, G. Dias, Magnesium and its alloys as orthopedic biomaterials: a review, Biomaterials 27 (2006) 1728–1734. [2] F. Witte, The history of biodegradable magnesium implants: a review, Acta Biomater. 6 (2010) 1680–1692. [3] G.L. Song, Control of biodegradation of biocompatible magnesium alloys, Corros. Sci. 49 (2007) 1696–1701. [4] F. Witte, J. Fischer, J. Nellesen, H.-A. Crostack, V. Kaese, A. Pisch, B. Felix, W. Henning, In vitro and in vivo corrosion measurements of magnesium alloys, Biomaterials 27 (2006) 1013–1018. [5] F. Witte, N. Hort, C. Vogt, S. Cohen, K.U. Kainer, R. Willumeit, F. Feyerabend, Degradable biomaterials based on magnesium corrosion, Curr. Opin. Solid State Mater. Sci. 12 (2008) 63–72. [6] Y. Wang, M. Wei, J. Gao, J. Hu, Y. Zhang, Corrosion process of pure magnesium in simulated body fluid, Mater. Lett. 62 (2008) 2181–2184. [7] E.D. McBride, Magnesium screw and nail transfixion in fractures, South. Med. J. 31 (50) (1938) 508–515. [8] K. Sigrid, J.G. Brunner, B. Fabry, S. Virtanen, Control of magnesium corrosion and biocompatibility with biomimetic coatings, J. Biomed. Mater. Res. B 96B (2011) 84–90. [9] H. Hornberger, S. Virtanen, A.R. Boccaccini, Biomedical coatings on magnesium alloys — a review, Acta Biomater. 8 (2012) 2442–2455. [10] H. Wang, Y. Estrin, Z. Zúberová, Bio-corrosion of a magnesium alloy with different processing histories, Mater. Lett. 62 (2008) 2476–2479. [11] S. Shaylin, G.J. Dias, Calcium phosphate coatings on magnesium alloys for biomedical applications: a review, Acta Biomater. 8 (2012) 20–30. [12] H. Wang, Y. Estrin, H.M. Fu, G.L. Song, Z. Zúberová, The effect of pre-processing and grain structure on the bio-corrosion and fatigue resistance of magnesium alloy AZ31, Adv. Eng. Mater. 9 (2007) 967–972. [13] W. Suchank, M. Yoshimura, Processing and properties of hydroxyapatite-based biomaterials for use as hard tissue replacement implants, J. Mater. Res. 13 (1998) 94–117. [14] X. Gu, W. Zhou, Y. Zheng, L. Dong, Y. Xi, D. Chai, Microstructure, mechanical property, bio-corrosion and cytotoxicity evaluations of Mg/HA composites, Mater. Sci. Eng. C 30 (2010) 827–832. [15] F. Witte, F. Feyerabend, P. Maier, J. Fischer, M. Stormer, C. Blawert, W. Dietzel, N. Hort, Biodegradable magnesium–hydroxyapatite metal matrix composites, Biomaterials 28 (2007) 2163–2174.
[16] R.S. Mishra, Z.Y. Ma, Friction stir welding and processing, Mater. Sci. Eng. R 50 (2005) 1–78. [17] R.S. Mishra, M.W. Mahoney, S.X. McFadden, N.A. Mara, A.K. Mukherjee, High strain rate super plasticity in a friction stir processed 7075 Al alloy, Scripta Mater. 42 (2000) 163–168. [18] R.S. Mishra, Z.Y. Ma, I. Charit, Friction stir processing: a novel technique for fabrication of surface composite, Mater. Sci. Eng. A 341 (2003) 307–310. [19] W.-B. Lee, C.-Y. Lee, M.-K. Kim, J.-I. Yoon, Y.-J. Kim, Y.-M. Yoen, S.-B. Jung, Microstructures and wear property of friction stir welded AZ91Mg/SiC particle reinforced composite, Compos. Sci. Technol. 66 (2006) 1513–1520. [20] C.J. Lee, J.C. Huang, P.J. Hsieh, Mg based nano-composites fabricated by friction stir processing, Scripta Mater. 54 (2006) 1415–1420. [21] Y. Morisada, H. Fujii, T. Nagaoka, K. Nogi, M. Fukusumi, Fullerene/A5083 composites fabricated by material flow during friction stir processing, Compos. Part A 38 (2007) 2097–2101. [22] T.-J. Chen, Z.-M. Zhu, Y.-D. Li, Y. Ma, Y. Hao, Friction stir processing of thixoformed AZ91D magnesium alloy and fabrication of Al-rich surface, Trans. Nonferr. Met. Soc. 20 (2010) 34–42. [23] P. Asadi, G. Faraji, M.K. Besharati, Producing of AZ91/SiC composite by friction stir processing, Int. J. Adv. Manuf. Technol. 51 (2010) 247–260. [24] M. Azizieh, A.H. Kokabi, P. Abachi, Effect of rotational speed and probe profile on microstructure and hardnessof AZ31/Al2O3 nanocomposites fabricated by friction stir processing, Mater. Des. 32 (2011) 2034–2041. [25] P. Asadi, G. Faraji, A. Masoumi, M.K. Besharatigivi, Experimental investigation of magnesium-base nanocomposite produced by friction stir processing: effects of particle types and number of friction stir processing passes, Metall. Mater. Trans. A 42A (2011) 2820–2832. [26] A. Mertens, A. Simar, H.-M. Montrieux, J. Halleux, F. Delannay, J. Lecomte-Beckers, Friction stir processing of magnesium matrix composites reinforced with carbon fibres: influence of the matrix characteristics and of the processing parameters on microstructural developments, in: W.J. Poole, K.U. Kainer (Eds.), Mg 2012. Proceedings of the 9th International Conference on Magnesium Alloys and Their Applications, July 8–12 2012, pp. 845–850, (Vancouver (Canada), http://hdl.handle.net/2268/ 120134 ). [27] N. Rameshbabu, K. Prasad Rao, T.S. Sampath Kumar, Accelerated microwave processing of nanocrystalline hydroxyapatite, J. Mater. Sci. 40 (2005) 6319–6323. [28] T. Kokubo, H. Takadama, How useful is SBF in predicting in vivo bone bioactivity? Biomaterials 27 (2006) 2907–2915. [29] L. Xu, E. Zhang, D. Yin, S. Zeng, K. Yang, In vitro corrosion behaviour of Mg alloys in a phosphate buffered solution for bone implant application, J. Mater. Sci. Mater. Med. 19 (2008) 1017–1025. [30] T. Mosmann, Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays, J. Immunol. Methods 65 (1983) 55–63. [31] B. Ratna Sunil, Arun Anil Kumar, T.S. Sampath Kumar, Uday Chakkingal, Role of biomineralization on the degradation of fine grained AZ31 magnesium alloy processed by groove pressing, Mater. Sci. Eng. C 33 (2013) 1607–1615. [32] R.N. Wenzel, Resistance of solid surfaces to wetting by water, Ind. Eng. Chem. 28 (1936) 988–994. [33] W.J. Arbegast, Modeling friction stir joining as a metalworking process, in: Z. Jin, A. Beaudoin, T.A. Bieler, B. Radhakrishnan (Eds.), Hot deformation of Aluminum Alloys III, TMS, Warrendale, PA, USA, 2003, pp. 313–327. [34] W.J. Arbegast, A flow-partitioned deformation zone model for defect formation during friction stir welding, Scripta Mater. 58 (2008) 372–376. [35] G.L. Song, A. Atrens, Corrosion mechanisms of magnesium alloys, Adv. Eng. Mater. 1 (1999) 11–33. [36] C.J. Wilson, R.E. Clegg, D.I. Leavensley, M.J. Pearcy, Mediation of biomaterial–cell interactions by adsorbed proteins: a review, Tissue Eng. 11 (2005) 1–18. [37] M. Hosokawa, K. Nogi, M. Naito, T. Yokoyama, Nanoparticle Technology Handbook, Elsevier, 2007. 19–21. [38] R. Chatopadhyay, Surface Wear: Analysis, Treatment, and Prevention, ASM International, Ohio, USA, 2001. 1–51. [39] B. Marc, L. Jacques, Can bioactivity be tested in vitro with SBF solution? Biomaterials 302 (2009) 175–2179. [40] J.A. Juhasz, S.M. Best, A.D. Auffret, W. Bonfield, Biological control of apatite growth in simulated body fluid and human blood serum, J. Mater. Sci. Mater. Med. 19 (2008) 1823–1829. [41] H.M. Kim, T. Himeno, M. Kawashita, T. Kokubo, T. Nakamura, The mechanism of biomineralization of bone-like apatite on synthetic hydroxyapatite: an in vitro assessment, J. R. Soc. Interface 1 (2004) 17–22. [42] B. Ratna Sunil, T.S. Sampath Kumar, Uday Chakkingal, Bioactive grain refined magnesium by friction stir processing, Mater. Sci. Forum 710 (2012) 264–269. [43] G.B. Hamu, D. Eliezer, L. Wagner, The relation between severe plastic deformation microstructure and corrosion behavior of AZ31 magnesium alloy, J. Alloys Compd. 468 (2009) 222–229. [44] G.R. Argade, S.K. Panigrahi, R.S. Mishra, Effects of grain size on the corrosion resistance of wrought magnesium alloys containing neodymium, Corros. Sci. 58 (2012) 145–151. [45] Y. Wang, M. Wei, J. Gao, Improve corrosion resistance of magnesium in simulated body fluid by dicalcium phosphate dihydrate coating, Mater. Sci. Eng. C 29 (2009) 1311–1316. [46] K. Das, S. Bose, A. Bandyopadhyay, Surface modifications and cell–materials interactions with anodized Ti, Acta Biomater. 3 (2007) 573–585. [47] L. Xiaomei, Y.L. Jung, J.D. Henry, D. Ravi, M.M. Andrea, A.V. Erwin, Influence of substratum surface chemistry/energy and topography on the human fetal osteoblastic cell line hFOB 1.19: phenotypic and genotypic responses observed in vitro, Biomaterials 28 (2007) 4535–4550
