High strength and bioactive hydroxyapatite nano-particles reinforced ultrahigh molecular weight polyethylene

Liming Fang1, Ping Gao2, Yang Leng1* 1

Department of Mechanical Engineering and 2 Department of Chemical Engineering Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong

Abstract Nano-sized hydroxyapatite (HA) particles reinforced UHMWPE composite was developed for biomedical applications. The biocomposite with HA volume fraction of 0.5 was successfully processed by combined swelling/twin-screw extrusion, compression molding, and then hot drawing at 100 ºC. SEM and EDAX characterization revealed that HA nanoparticles were homogeneously dispersed in UHMWPE. WAXD showed that hot drawing effectively oriented the UHMWPE chains along the drawing direction. The composite exhibited tensile strength of 100 ± 22 MPa after hot drawing, which was comparable to that of cortical bone. The composite also exhibited great ability of inducting calcium phosphate precipitates on its surface in simulated body fluid, which was widely accepted as indication of bioactivity.

Keywords A. Nano-structures; B. Directional orientation; B. Mechanical properties; D. Extrusion

*

Corresponding author: Yang Leng. Email: [email protected]; Fax: (852)23581543

1. Introduction Biocomposite materials have been widely promoted as possible orthopaedic biomaterials for their tailorable mechanical and biological properties, but to date have found few successful commercial applications, due to the many challenging problems presented by their design, fabrication and testing [1, 2]. Ever since Bonfield et al [3] pioneered the hydroxyapatite (HA) reinforced high density polyethylene (HDPE), a great number of polymer matrix biocomposites have been proposed [2, 4]. However, the low mechanical properties of these composites limited their applications as non-load bearing implants [4].

In order to exploit the advantages of biocomposites, various methods were explored. Stronger matrices, such as poly(ether-ether-ketone) (PEEK) [5] and poly(methy1 methacrylate) (PMMA) [6], were used to improve the stiffness and strength, but the brittleness of the glassy polymers were not favorable for a bone substitute. Filler surface coated with coupling agents or/and polymer graft treatment were employed to improve the interface adhesion, but produced marginal enhancement [7], while the bioactivity of the filler might lose after treatment. Hydrostatic extrusion [8] or shear controlled orientation in injection moulding (SCORIM) [9] was also tried to align polymer chains, the stiffness was significantly enhanced but the yield strength was not improved much. Wang et al. [10] and Roeder et al. [11] studied the effects of particle size and shape, and found that smaller particle size and larger aspect ratio increased the composite mechanical properties, but still much lower than cortical bone due to the micron-sized reinforcement.

Essentially, bone is a natural biocomposite with a hierarchical organization, which is made

up of collagen fibrils reinforced by nano-sized mineral crystals (hydroxyapatite) at the subnanostructure scale [12-14]. The mineral crystals bond to the well aligned collagen fibers with a preferred orientation parallel to the longitudinal axis of bone, which contributes to the amazing mechanical properties of bone [12]. Therefore, a synthetic biocomposite with mechanical properties comparable to natural bone should exhibit a microstructure mimic natural bone.

Recently, we developed a hydroxyapatite reinforced ultrahigh molecular weight polyethylene (HA/UHMWPE) nanocomposite to combine the high stiffness of HA and the excellent toughness of UHMWPE [4, 15]. Preliminary results showed that HA nanoparticles were dispersed homogeneously and mechanical-locked intimately to the UHMWPE matrix by twin-screw extrusion and swelling treatment. The nanocomposite exhibited a significant enhancement of stiffness compared with both pure UHMWPE and other biocomposites with the same filler content [4]. However, the yield strength of this composite was not improved because of the low strength of UHMWPE. Since the composite ductility was extremely high and the strength of UHMWPE could be enhanced by hot drawing [16, 17], the as-processed nanocomposite was hot drawn to further increase its yield strength. In this paper, the relationship of processing, microstructure and properties, including mechanical and bioactivity properties, of the hot drawn HA/UHMWPE biocomposite were investigated. 2. Experimental 2.1 Materials and processing The UHMWPE was HiFax 1900 reactor powder (Basell Ltd, USA) with a weight average

molecular weight of 6×106 g/mol and an average size of ~200 μm. The HA powder (National Engineering Research Center for Biomaterials, China) was wet synthesized with an average aggregate size of ~2 μm. Both were the same as those used in previous work [15].

The composite processing details were presented in our previous publication [4]. Briefly, HA and UHMWPE powder mixture with HA volume fraction 0.5 (VHA = 0.5) was compounded by twin-screw extrusion using paraffin oil as the swelling agent. After the oil was removed by hot press and extraction, the extrudate was compression molded. Then, it was cut into rectangular strips with the dimension ~ 2 mm × 40 mm × 5 mm (thickness × length × width). Hot drawing was carried out using Instron 5500 (Instron Co., USA) equipped with a temperature control chamber. The specimens were hot drawn at 100 ºC with a cross head speed 50 mm/min to a draw ratio of 15 (DR = 15).

2.2 Microstructure characterization The composite microstructure was characterized by scanning electron microscope (SEM 6300F, JEOL, Japan). Three samples were prepared: 1) as-processed top surface; 2) cryogenically-fractured cross-section surface; and 3) as-drawn film surface. Element analysis and map scanning were carried out using SEM 6300 (JEOL, Japan) equipped with an energy dispersive analysis X-ray system (EDAX, Oxford Instruments, UK). The orientation of polymer chains was examined by wide-angle X-ray diffractometer (WAXD, Bruker SMART APEX, Germany) using Mo Kα radiation (λ = 0.71073 Å). The beam was perpendicular to the specimen draw axis, that is, perpendicular to the UHMWPE

orthorhombic crystal c axis, and the data was collected in the a, b plane (flat display). The specimen was rotating around the drawing direction through 360º during exposure.

2.3 Tensile testing Tensile tests were performed on an Advanced Rheometric Expansion Systems (ARES, TA Instrument, USA) using a thin film fixture. Hot drawn film was mounted on a paper tab so that the specimen was gripped firmly and aligned axially. The specimens were conditioned at 80 ºC in a vacuum oven overnight. The film width was measured using a profile projector (PJ311, Mitutoyo, Japan) with a resolution of 1 μm, while the thickness was measured using a micrometer (resolution 1 μm). Each specimen was measured 6 times at different positions and the average value was used in calculation of the mechanical properties. The distance between the grips was 20 mm. The accuracy of the load cell and extension were 10-6 N and 1 μm, respectively. Seven specimens were tested at the cross head speed of 50 mm/min.

2.4 In vitro evaluation HA sintered block, UHMWPE compression molded sheet, and hot drawn HA/UHMWPE composite film were immersed in revised simulated body fluid (R-SBF [18]) at 37 ºC for certain periods (2, 4, 10, 24 hours and 7 days). After SBF immersion, the sample surface was examined by SEM (6300 JEOL, Japan) and thin film X-ray diffractometer (TF-XRD, X'pert Pro, PANalytical, USA). The TF-XRD measurement was performed in the range of 2-60º in 2θ using Cu Kα (λ = 1.5405 Å, 40 kV, 40 mA) radiation as the source with a step

size of 0.05º at a rate of 0.01º/second.

3. Results 3.1 Microstructure SEM micrographs of the HA/UHMWPE composite as-processed top surface were shown in Fig. 1. Low magnification micrograph (Fig. 1a) revealed fairly uniform dispersion of HA in the UHMWPE matrix. High magnification micrograph (Fig. 1b) showed that the initial HA aggregates were effectively broken down to nano-sized particles and embedded in the interspaces among the UHMWPE fibrils. The SEM micrograph of the cryogenically fractured cross-section surface (Fig. 2) confirmed that the overall dispersion of HA nano particles was homogeneous. An X-ray energy dispersive spectrum (Fig. 3a) revealed the Calcium (Ca), Phosphorus (P) elements and the Ca/P atomic ratio was 1.7, which was very close to the theoretical value 1.67 of HA (Ca10(PO4)6(OH)2). Elemental mapping (Fig. 3b) of selected area further indicated the homogeneous dispersion of HA nano particles.

The SEM micrograph (Fig. 4a) revealed that the UHMWPE chains were well aligned along the drawing direction. The orientation of the UHMWPE chains was further confirmed by the WAXD pattern (Fig. 4b). The diffraction ring of UHMWPE orthorhombic crystal planes (200) became an arc due to the c-axis of UHMWPE crystals was oriented along the drawing direction and the normal of (200) planes was perpendicular to the c-axis. Therefore, the orientation angle between UHMWPE chains (c-axis) and the drawing direction was measured from the half-width of the (200) arc [19], which was around 17º in the drawn composite, while the angle was about 10º in the pure UHMWPE with the same draw ratio

(Fig. 4b, inset).

3.2 Mechanical properties Typical tensile test stress-strain curves of the hot drawn pure UHMWPE and HA/UHMWPE composite films were shown in Fig. 5. The mechanical properties of the composite, pure UHMWPE and cortical bone were summarized in Table 1. Compared with UHMWPE, the HA/UHMWPE composite exhibited a higher modulus but lower strength and ductility. The absolute value of Young’s modulus was underestimated because the extension was calculated from the cross head speed, but the relative trend was not affected, which clearly revealed the stiffening effect of HA nano-particles. The yield strength was measured from the upper yield point on the tensile curve, which was accurate with the high sensitive load transducer and careful measurement of specimen dimensions. The yield strength of the hot drawn HA/UHMWPE composite was comparable to that of cortical bone [12], which was much higher than that of other biocomposites [4]. The ductility was defined as the strain at fracture, which was better than that of cortical bone [12], although it was significantly reduced after hot drawing [4].

3.3 Bioactivity SEM micrographs of HA block, UHMWPE sheet, and hot drawn composite film after different immersion times in R-SBF were shown in Fig. 6, which displayed the precipitates nucleation and growth with immersion time on the composite surface. Fig. 6a showed that nucleates were present after only 2 hour immersion. After 4 hour immersion, the whole

surface had been covered by the precipitates (Fig. 6b), which grew to micron-sized granules within 10 hours (Fig. 6c). After one day immersion, coating layer of 1 μm thick was formed on the composite surface as seen from the edge on view of the cracks (Fig. 6d). The ability of precipitation on the composite surface was comparable to that of pure HA (Fig. 6e). However, there were not any precipitates formed on the pure UHMWPE surface after immersion in SBF for 7 days (Fig. 6f). The Ca-P layer on HA/UHMWPE continuously grew to about 10 μm thick in 7 day immersion as revealed by the cross-section SEM micrograph (Fig. 7a), which showed a flake-like morphology under higher magnifications (Fig. 7b).

TF-XRD patterns (Fig. 8) of the composite surface showed the intensity of UHMWPE crystal peaks were decreased gradually and disappeared eventually. The (110) peak of UHMWPE was still visible due to the substrate area exposed at the cracks. The strongest peak at 25.975º and the broad peaks in the range of 30-33º indicated the precipitation layer was Ca-P, either HA or octacalcium phosphate (OCP) [20]. However, the Ca-P layer was most likely to be OCP because it showed the typical flake-like feature of OCP formed in SBF immersion as shown in Fig. 6f [21].

4. Discussion 4.1 Mechanical properties The strength of UHMWPE film was significantly enhanced after hot drawing [4], which was attributed to the highly oriented UHMWPE fibrils (Fig. 4). Theoretically, the ultimate tensile strength of UHMWPE along the chain direction varies between 33 GPa and 66 GPa

based on a semi-empirical estimation [22] and a quantum mechanical calculation to rupture the C-C bonds [23]. Experimentally, however, it is difficult to achieve the theoretical strength because of the imperfect polymer chain alignment or orientation [16].

Note that the yield strength of HA/UHMWPE nanocomposite was lower than that of UHMWPE with the same draw ratio. The high content brittle HA made the composite fracture before the ultimate tensile strength of UHMWPE was reached. Furthermore, the orientation of UHMWPE chains were hindered by the HA particles during hot drawing (Fig. 4b). Thus the UHMWPE fibril strength in the composite was less than the unfilled UHMWPE. Consequently, this also explained the higher stiffness of the hot drawn composite compared with pure UHMWPE. As a result, the composite exhibited a stronger resistance to deformation at the same strain level. The fracture strain of the composite (> 3%) was significantly reduced after hot drawing, but still higher than cortical bone suggested it was acceptable as a bone substitute [4].

4.2 Bioactivity Bioactivity of orthopaedic materials was widely investigated by SBF immersion experiments, because such in vitro evaluation was simple and efficient, and has been demonstrated fair correlation with the in vivo test [18, 21, 24-26]. Faster nucleation, larger crystals, and stronger bonding of the bioactive Ca-P layer on the orthopaedic material surface were considered as better bioactivity. Note that the immersion time to cover the whole surface with Ca-P was more than one day for other biocomposites with filler volume fraction 0.4, such as HA/PEEK [24], apatite-wollastonite (AW)/HDPE [25] and

Bioglass®/HDPE [26]. However, the HA/UHMWPE nanocomposite surface was entirely covered by small nucleates in 4 hours, indicated its excellent bioactivity in vitro.

The bioactivity of biocomposite was controlled by the surface area of bioactive filler exposed to the physiological environment. Smaller filler size and better dispersion are favorable for Ca-P induction because more bioactive surfaces can be exposed to the solution and the nucleation areas are larger. Therefore, the nucleation and growth of Ca-P was uniform and fast on the surface of homogenously dispersed HA nano-particles reinforced UHMWPE composite, but the Ca-P nucleated initially on the agglomerated micro-sized bioactive filler rich part and then spread to the bioinert polymer rich area of other biocomposites surfaces.

5. Conclusions HA/UHMWPE nanocomposite with HA volume fraction of 0.5 was successfully developed. The HA nano-particles were homogeneously dispersed in the UHMWPE matrix and formed an inter-penetrated network structure. The UHMWPE fibrils in the composite were highly oriented along the hot drawing direction. The hot drawn composite with draw ratio of 15 exhibited yield strength of 100 ± 22 MPa, which was comparable to that of cortical bone. The biocomposite demonstrated an excellent ability of inducing calcium phosphate formation in simulated body fluid, which was promising to be used for load bearing bone substitutes.

Acknowledgements

This project was funded by the Research Grants Council of Hong Kong (Grant No. HKUST6244/00P), and the Hong Kong University of Science and Technology through the research grants for High Impact Areas (HIA01/02.EG8 and HIA03/04.EG03). We thank Basell Ltd., USA for providing UHMWPE and the Engineering Research Center in Biomaterials at the Sichuan University in China for providing HA particles.

Reference: 1. Evans SL, Gregson PJ. Composite technology in load-bearing orthopaedic implants. Biomaterials 1998;19(15):1329-1342. 2. Ramakrishna S, Huang ZM, Kumar GV, Batchelor AW, Mayer J, editors. An introduction to biocomposites. London: Imperial College Press, 2004. 3. Bonfield W, Grynpas MD, Tully AE, Bowman J, Abram J. Hydroxyapatite reinforced polyethylene - a mechanically compatible implant material for bone replacement. Biomaterials 1981;2(3):185-186. 4. Fang LM, Leng Y, Gao P. Processing and mechanical properties of HA/UHMWPE nanocomposites. Biomaterials In Press. 5. Abu Bakar MS, Cheang P, Khor KA. Mechanical properties of injection molded hydroxyapatite-polyetheretherketone biocomposites. Compos Sci Technol 2003;63(34):421-425. 6. Cheang P, Khor KA. Effect of particulate morphology on the tensile behaviour of polymer-hydroxyapatite composites. Mater Sci Eng A-Struct Mater Prop Microstruct Process 2003;345(1-2):47-54. 7. Wang M, Bonfield W. Chemically coupled hydroxyapatite-polyethylene composites: structure and properties. Biomaterials 2001;22(11):1311-1320. 8. Wang M, Ladizesky NH, Tanner KE, Ward IM, Bonfield W. Hydrostatically extruded HAPEXTM. J Mater Sci 2000;35(4):1023-1030. 9. Sousa RA, Reis RL, Cunha AM, Bevis MJ. Processing and properties of bone-analogue biodegradable and bioinert polymeric composites. Compos Sci Technol 2003;63(34):389-402. 10. Wang M, Joseph R, Bonfield W. Hydroxyapatite-polyethylene composites for bone substitution: effects of ceramic particle size and morphology. Biomaterials 1998;19(24):2357-2366.

11. Roeder RK, Sproul MM, Turner CH. Hydroxyapatite whiskers provide improved mechanical properties in reinforced polymer composites. J Biomed Mater Res Part A 2003;67A(3):801-812. 12. Ji BH, Gao HJ. Mechanical properties of nanostructure of biological materials. J Mech Phys Solids 2004;52(9):1963-1990. 13. Rho JY, Kuhn-Spearing L, Zioupos P. Mechanical properties and the hierarchical structure of bone. Med Eng Phys 1998;20(2):92-102. 14. Weiner S, Wagner HD. The material bone: structure mechanical function relations. Annu Rev Mater Sci 1998;28:271-298. 15. Fang LM, Leng Y, Gao P. Processing of hydroxyapatite reinforced ultrahigh molecular weight polyethylene for biomedical applications. Biomaterials 2005;26(17):3471-3478. 16. Ruan S, Gao P, Yu TX. Ultra-strong gel-spun UHMWPE fibers reinforced using multiwalled carbon nanotubes. Polymer 2006;47(5):1604-1611. 17. Smith P, Lemstra PJ, Kalb B, Pennings AJ. Ultrahigh-strength polyethylene filaments by solution spinning and hot drawing. Polymer Bulletin 1979;1(11):733-736. 18. Kokubo T, Takadama H. How useful is SBF in predicting in vivo bone bioactivity? Biomaterials 2006;27(15):2907-2915. 19. Richard S. Stein FHN. The x-ray diffraction, birefringence, and infrared dichroism of stretched polyethylene. J Polymer Sci 1956;21(99):381-396. 20. Elliott JC, editor. Structure and chemistry of the apatites and other calcium orthophosphates. Amsterdam: Elsevier, 1994. 21. Xin R, Leng Y, Chen J, Zhang Q. A comparative study of calcium phosphate formation on bioceramics in vitro and in vivo. Biomaterials 2005;26(33):6477-6486. 22. He T. An estimate of the strength of polymers. Polymer 1986;27(2):253-255. 23. Crist B, Ratner MA, Brower AL, Sabin JR. Ab initio calculations of the deformation of polyethylene. J Appl Phys 1979;50(10):6047-6051. 24. Yu S, Hariram KP, Kumar R, Cheang P, Aik KK. In vitro apatite formation and its growth kinetics on hydroxyapatite/polyetheretherketone biocomposites. Biomaterials 2005;26(15):2343-2352. 25. Rea SM, Best SM, Bonfield W. Bioactivity of ceramic-polymer composites with varied composition and surface topography. J Mater Sci Mater Med 2004;15(9):997-1005. 26. Huang J, Di Silvio L, Wang M, Rehman I, Ohtsuki C, Bonfield W. Evaluation of in vitro bioactivity and biocompatibility of Bioglass®-reinforced polyethylene composite. J Mater Sci Mater Med 1997;8(12):809-813.

a

b

Fig.1 FANG LM et al

Fig.2 FANG LM et al

a

b

Fig. 3 FANG LM et al

a

b

HA (211, 112, 300)

PE (110)

HA (002)

PE (110)

PE (200)

Fig. 4 FANG LM et al

PE (200)

160 140

Pure UHMWPE (DR = 15) HA/UHMWPE (VHA = 0.5, DR = 15)

stress (MPa)

120 100 80 60 40 20 0 0.00

0.01

0.02

0.03

0.04

strain (mm/mm)

Fig. 5 FANG LM et al

0.05

0.06

0.07

a

b

c

d

e

f

Fig. 6 FANG LM et al

a

b

Fig. 7 FANG LM et al

Normalized Intensity (arbitrary units)

Substrate Soaking 7 days HA in substrate PE in substrate Ca-P in coating

HA: (002) OCP: (002), (102)

orthorhombic (110)

orthorhombic (110)

HA or OCP?

orthorhombic (200) HA: (211), (112), (300)

monoclinic (001)

0

10

20

30

2θ (degree) Fig. 8 FANG LM et al

40

50

60

Figure captions:

Figure 1 SEM micrographs of the HA/UHMWPE nanocomposite: (a) low magnification micrograph showing global HA dispersion; (b) High magnification micrograph showing the entanglement of HA participles and UHMWPE fibrils.

Figure 2 SEM micrograph of cryogenically fractured surface of the HA/UHMWPE nanocomposite

Figure 3 X-ray energy dispersive spectra of the HA/UHMWPE nanocomposite. (a) spectrum; (b) Calcium map.

Figure 4 Microstructure of the hot drawn HA/UHMWPE nanocomposite films. (a) SEM micrograph; (b) WAXD pattern (inset: UHMWPE). Solid arrow line: drawing direction.

Figure 5 Typical tensile stress/strain curves of the hot drawn UHMWPE and HA/UHMWPE nanocomposite.

Figure 6

SEM micrographs of the HA, UHMWPE, and the HA/UHMWPE surfaces after SBF immersion. a) HA/UHMWPE 2 hours; b) HA/UHMWPE 4 hours; c) HA/UHMWPE 10 hours; d) HA/UHMWPE 24 hours; e) HA 10 hours; f) UHMWPE 7 days.

Figure 7 SEM micrographs of HA/UHMWPE immersed in SBF for 7 days: a) cross-section of CaP layer; b) high magnification of Ca-P layer surface.

Figure 8 TF-XRD patterns of the HA/UHMWPE composite surface before and after SBF immersion (7 days).

Table 1 Mechanical properties of hot drawn UHMWPE and HA/UHMWPE composite Young’s modulus (GPa)

Yield strength (MPa)

Fracture strain (%)

UHMWPE

2.0 ± 0.3

153 ± 11

8.4 ± 1.7

HA/UHMWPE

4.1 ± 1.7

100 ± 22

3.6 ± 0.5

Cortical bone [12-14]

7-30

50-150

1-3