Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene

Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene

Biomaterials xxx (2014) 1e11 Contents lists available at ScienceDirect Biomaterials journal homepage: www.elsevier.com/locate/biomaterials Mechanic...

4MB Sizes 0 Downloads 2 Views

Biomaterials xxx (2014) 1e11

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene Yan-Fei Huang a, Jia-Zhuang Xu a, Jian-Shu Li b, Ben-Xiang He c, Ling Xu a, **, Zhong-Ming Li a, * a State Key Laboratory of Polymer Materials Engineering, College of Polymer Science and Engineering and, Sichuan University, Chengdu 610065, People’s Republic of China b Department of Biomedical Engineering, College of Polymer Science and Engineering and, Sichuan University, Chengdu 610065, People’s Republic of China c Rehabilitation Centre, Sport Hospital Attached to Chengdu Sport University, 251 Wuhou Temple Street, Chengdu 610041, People’s Republic of China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 April 2014 Accepted 22 April 2014 Available online xxx

The low efficiency of fabrication of ultrahigh molecular weight polyethylene (UHMWPE)-based artificial knee joint implants is a bottleneck problem because of its extremely high melt viscosity. We prepared melt processable UHMWPE (MP-UHMWPE) by addition of 9.8 wt% ultralow molecular weight polyethylene (ULMWPE) as a flow accelerator. More importantly, an intense shear flow was applied during injection molding of MP-UHMWPE, which on one hand, promoted the self-diffusion of UHMWPE chains, thus effectively reducing the structural defects; on the other hand, increased the overall crystallinity and induced the formation of self-reinforcing superstructure, i.e., interlocked shish-kebabs and oriented lamellae. Aside from the good biocompatibility, and the superior fatigue and wear resistance to the compression-molded UHMWPE, the injection-molded MP-UHMWPE exhibits a noteworthy enhancement in tensile properties and impact strength, where the yield strength increases to 46.3  4.4 MPa with an increment of 128.0%, the ultimate tensile strength and Young’s modulus rise remarkably up to 65.5  5.0 MPa and 1248.7  45.3 MPa, respectively, and the impact strength reaches 90.6 kJ/m2. These results suggested such melt processed and self-reinforced UHMWPE parts hold a great application promise for use of knee joint implants, particularly for younger and more active patients. Our work sets up a new method to fabricate high-performance UHMWPE implants by tailoring the superstructure during thermoplastic processing. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Artificial joints Ultrahigh molecular weight polyethylene Melt processing Flow accelerator Self-reinforcement

1. Introduction Ultrahigh molecular weight polyethylene (UHMWPE) has a long successful clinical record as the load-bearing surface material in total joint replacement (TJR) due to its outstanding wear resistance, fatigue resistance as well as recognized biocompatibility [1]. Recently, a great progress has been made by Oral et al., who developed highly wear-resistant and antioxidative UHMWPE via combination of radiation crosslinking and antioxidant stabilization [2e8]. A crucial issue currently remaining unresolved is the low efficiency of fabrication of UHMWPE implants because of its extremely high melt viscosity. At temperature above its melting temperature, UHMWPE does not flow like normal thermoplastic

* Corresponding author. Tel.: þ86 28 8540 0211; fax: þ86 28 8540 5402. ** Corresponding author. Tel.: þ86 28 8540 5731; fax: þ86 28 8540 5402. E-mail addresses: [email protected] (L. Xu), [email protected] (Z.-M. Li).

polymers, making it practically impossible to process by conventional thermoplastic processing techniques, especially those highly productive processing operations, such as screw extrusion and injection molding. Nowadays, the processing of this biomaterial is basically limited to compression molding and ram extrusion, which, however, requires a long molding time at high temperature (e.g., about 200e300  C for about 3 h) and elevated pressures for the molecular chains to migrate across grain boundaries in order for UHMWPE particles to completely melt and finally consolidate [9]. There is a concern that such a long time for UHMWPE staying at high temperatures may increase the risk of molecular chain scission and oxidization, resulting in performance deterioration of joint implants. Another concern is that, due to the high entanglement density and the resulting extremely slow self-diffusion of UHMWPE chains [10,11], current processing methods inevitably leave either particle boundaries reflecting the powder flakes of as-polymerized UHMWPE resin [12e14], or white specks, or voids [15]. Such structural defects inside the components would serve as stress

http://dx.doi.org/10.1016/j.biomaterials.2014.04.077 0142-9612/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

2

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11

concentration, which makes the local stress around the particles sharply increase. Given that conventional melt processing techniques are utilized, compounding force field can be provided to effectively mix polymer melt within a short time at a relatively low temperature, avoiding the aforementioned disadvantages. In this case, the melt fluidity of UHMWPE has to be greatly improved. In practical cases, the development of reducing the high viscosity of UHMWPE often involves the use of a low molecular weight solvent [16,17]. Such processes require significant quantities of organic solvents, which are difficult to recycle and remove, even potentially activate immunological responses in vivo, and thus has to be absolutely avoided in the biomedical applications. An alternative strategy includes blending UHMWPE with polypropylene [18], polyethylene glycol [19,20], and polylactic acid [21], filling with nano-inorganic fillers [22], and incorporating a small amount of processing aids (liquid crystalline polymer, fluoroelastomer, and stearates) [23,24]. This method has advantages for its solvent-free processing, but also has to be excluded for artificial joint materials due to either the bad biological compatibility of the additives, or their weak interfacial interaction with UHMWPE. In our previous work [25,26], we have well demonstrated that incorporation of low molecular weight polyethylene (LMWPE) into UHMWPE could be a possible approach for effective viscosity reduction, and particularly, such UHMWPE-based material could be melt processed by injection molding. Meantime, LMWPE has good biological compatibility [27e29] and interfacial affinity with UHMWPE due to their exactly same chemical and crystallographic structures. On basis of this fact, a melt processable LMWPE/ UHMWPE blend was prepared by melt mixing 60 wt% LMWPE (Mw ¼ 1.2  105 g mol1) with 40 wt% UHMWPE [25]. The mechanical properties, fatigue and wear resistance of the blend have been successfully guaranteed through the formation of shish-kebab self-reinforced superstructure by introducing shear flow during injection molding. The subsequent work approaches to increasing the UHMWPE content as much as possible for the maximum retention of excellent performance of UHMWPE, such as the good wear and fatigue resistance. However, the increased UHMWPE component inevitably causes poor fluidity. In the latest work, we proposed to improve the fluidity of the PE blend by cross-linking UHMWPE, thereby elevating the UHMWPE content [26]. Nonetheless, the UHMWPE loading was still no more than 50 wt%, above which the UHMWPE/LMWPE blend could not be melt processed. Therefore, further improvement on the processibility of UHMWPE is of great interest. In the present work, we employed ultralow molecular weight polyethylene (ULMWPE, Mw ¼ 2.8  104 g/mol) as a flow accelerator to dramatically promote UHMWPE molecular chains to slip, thus substantially reducing the melt viscosity of UHMWPE. While melt injection molding, a specific shear flow was applied to induce oriented lamellae and shish-kebab self-reinforcing superstructure. The mechanical properties, tribological properties and biocompatibility were evaluated. 2. Materials and methods 2.1. Materials UHMWPE was purchased from Samsung General Chemicals Corporation, Ltd., Korea, with an average viscosity molecular weight in the range of 5.0e6.0  106 g/ mol. ULMWPE was supplied by Petroleum & Chemical Corporation, China, whose weight-average molecular weight was 2.8  104 g/mol, with a melt flow rate of 45 g/ 10 min (0.325 kg, 190  C). 2.2. Sample preparation High-molecular-weight species (such as UHMWPE) were proved to be helpful for efficient formation of shish-kebabs in the entangled melt under a given flow condition [30e32], thus 2 wt % UHMWPE (much higher than the overlap concentration, c* w0.2 wt % [33]) was first mixed with ULMWPE by a solution blending

procedure to achieve molecular-level dispersion. The 2 wt% UHMWPE/98 wt% ULMWPE mixture was dissolved in xylene at 140  C by continuous stirring to obtain a uniform dispersion. The polymer was precipitated by pouring the solution into cold alcohol under continuous stirring. The extracted solid mixture was filtered and dried in a vacuum oven at 60  C for about 2 weeks to remove the solvent. Such drying conditions assured complete elimination of the residual solvent since there is no weight loss during heating the dried polymer from room temperature to 200  C by thermogravimetric analysis (not shown here for brevity, see Fig. S1 in the Supplementary Data). 10 wt% of the 2 wt% UHMWPE/98 wt% ULMWPE mixture obtained above (still named as ULMWPE for brevity), acting as a flow accelerator, was melt mixed with 90 wt% UHMWPE in a twin-screw extruder. The processing temperature profile was limited within 190e220  C from hopper to die, and the screw speed was held constantly at 150 rpm. Finally, the extruded pellets after dried were injection molded into dumbbell and rectangle samples in an injection temperature profile of 200e230  C from hopper to nozzle. At packing stage of an injection molding cycle, a controlled shear flow was continuously imposed on the melt in the mold cavity. The apparatus of such injection molding technology is shown in Supplementary Data (Fig. S2eS4). For comparison, compression molded UHMWPE with approximate dimensions of 10  10  4 cm3 were pressurized to 10 MPa and melted at 200  C for 2 h involving a slow cooling procedure as control sample. Sketch of the processing procedures is shown in Fig. 1. By virtue of ULMWPE as a flow accelerator, UHMWPE is factually converted to a melt processable polymer. For brevity, injection molded melt processable UHMWPE under shear flow refers to as MP-UHMWPE while the conventional compression molded neat UHMWPE is named CM-UHMWPE. The formulation of MP-UHMWPE is 90.2 wt% UHMWPE and 9.8 wt% ULMWPE, and CM-UHMWPE is 100 wt% UHMWPE. 2.3. Mechanical property testing Tensile test of the dumbbell samples (n ¼ 5) parallel to the shear flow direction were prepared according to ASTM-D 638. The dimension of tensile specimens is shown in Supplementary Data (Fig. S5). These samples were tested in an Instron universal test instrument model 5576 (USA) with a crosshead speed of 10 mm/min. The average values were reported with standard deviation. The notched Izod impact test were carried out according to the standard ASTM D256-06a at room temperature, the dimension of testing specimens were carefully machined to be 60 mm  10 mm  4 mm with a single V-notch of 2 mm depth. 2.4. Wear testing The sliding wear behavior of the samples was examined on M-200 wear tester. In order to simulate the in vivo environment, the tests were conducted with newborn (14 days or less) bovine serum (New Zealand) as lubricating medium. The sliding pair consisted of a tested specimens (MP-UHMWPE or CM-UHMWPE) block and a 45 steel (Ra ¼ 1 mm), in which the upper side was the steel ring and the lower side was the sample. The diameter of the counterpart steel ring was 40 mm. Under normal force of 160  0.2 N, the tests were carried out at a linear velocity of 0.43 m/s, with the axis rotation speed of 200 rpm. The block specimens were of size 30  7  4 mm3, according to the standard ASTM D1894. To ensure reproducibility, the tests were done more than three times for each sample with sliding time of 2 h. The friction coefficient f was calculate by f ¼ T/RP, where T and P are the friction torque (Nm) and the normal load (N), respectively. The test specimens were weighed every 0.2 million cycles (MC) up to 2.4 MC. The wear rate was calculated as the linear regression of weight loss versus number of cycles from 0.2 MC to 2.4 MC. Three test specimens of each sample were tested. 2.5. Fatigue strength testing The dumbbell specimens (n ¼ 3) were precracked at the notch using a razor blade following ASTM E-647. Fatigue crack propagation test was conducted at a sinusoidal load cycle frequency of 5 Hz and stress ratio of 0.1 in tension, which, obviously, would result in a sacrifice of ultimate tensile strength. After 20,000 cycles, the impaired ultimate tensile strength was measured as the same as reported in Section 2.3. Fatigue resistance is evaluated by the retention rate in ultimate tensile strength after fatigue test, and the higher of the retention rate suggests a better fatigue resistance. The retention rate was calculated by Fatigue strength ¼ UTS2/UTS1, where UTS1 represents ultimate tensile strength of the sample (MPa), UTS2 represents ultimate tensile strength after fatigue testing (MPa). 2.6. Morphology observation Polarizing optical microscopy (POM, Olympus BX51) was attempted to observe the distribution of ULMWPE into UHMWPE matrix, and the sample with the thickness of 20 mm was prepared using a microstone. Further structure analysis was carried out by scanning electron microscopy (SEM). The test specimens were cryogenically fractured in liquid nitrogen. In order to distinguish the crystalline structure clearly, the amorphous phase has been chemically etched by 0.5% solution of potassium permanganate in a mixture of concentrated sulfuric acid (98%) and concentrated nitric acid (65%). The etched surface was finally covered with a thin

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11

3

Fig. 1. Sketch of the processing procedures.

layer of gold and observed by an SEM instrument (Inspect F, FEI, Finland) operating at 20 kV. 2.7. WAXD And SAXS measurements Wide-angle X-ray diffraction (WAXD) and small-angle X-ray scattering (SAXS) measurements were used to characterize the crystalline structure and molecular

orientation distribution along the sample width direction. WAXD and SAXS measurements were carried out at the Advanced Polymers Beamline (X27C, wavelength l ¼ 1.371 Å) in the National Synchrotron Light Source (NSLS), Brookhaven National Laboratory (BNL), USA. A MAR CCD X-ray detector (MARUSA) was employed for collection of 2D-WAXD and 2D-SAXS images (a resolution of 1024  1024 pixels, pixel size ¼ 158 mm). For evaluation of molecular orientation, the orientation parameter of MP-UHMWPE and CM-UHMWPE samples was calculated

Fig. 2. The yield strength, ultimate tensile strength, Young’s modulus, Izod impact strength, fatigue strength and elongation at break (from left to right) of MP-UHMWPE and CMUHMWPE.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

4

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11 to 180  C at a constant heating rate of 10  C/min under a nitrogen atmosphere, and the sample weight was about 3e5 mg. Crystallinity of the samples (n ¼ 3 each) was determined by integrating the enthalpy peak from 40 to 160  C, and normalizing it with the enthalpy of melting of 100% crystalline polyethylene, 291 J/g. 2.9. Biocompatibility evaluation 2.9.1. Cell viability (MTT assay) The cytotoxicity of MP-UHMWPE and CM-UHMWPE samples was evaluated via methyl thiazolyl tetrazolium (MTT) assay in L929 cell line. The cells (8  104 cells/ well) were cultured in Dulbecco’s modified eagle medium (DMEM), supplemented with 10% fetal bovine serum (FBS), 100 units/mL of penicillin and 100 mg/mL of streptomycin at 37  C, under an atmosphere of 5% CO2 and 95% relative humidity for 2 days. Then, 20 mL of MTT solution (5 mg/mL) was added to each well and incubated for 5 h at 37  C. After that, the solution was removed, and 400 mL/well dimethyl sulfoxide (DMSO) was added to dissolve the formed formazan crystals. After 30 min, the DMSO solution was transferred to a 96-well plate with 150 mL/well. Finally, the absorbance at a wavelength of 492 nm was measured using a microplate reader (KTH 360). The cell viability was calculated from [A]test/[A]control  100%, where [A]test and [A]control are denoted as absorbance of the sample and control wells, respectively. For each sample, the final absorbance was the average of those measured from three wells in parallel.

Fig. 3. Friction coefficient vs. the distance traveled during wear for (a) MP-UHMWPE and (b) CM-UHMWPE samples. The inset presents the wear rates of MP-UHMWPE and CM-UHMWPE.

mathematically using Picken’s method from the (110) reflection (2q ¼ 19.2 ) of WAXD for PE [34]. In order to investigate the lamellar structure in MP-UHMWPE and CM-UHMWPE samples quantitatively, the long period was calculated using the Bragg equation, L ¼ 2p/q*, where L is long period, and q* represents the peak position in the scattering curves [35].

2.9.2. Cell morphology The 2 day’s fluorescent images were obtained by a Fluorescence Microscope (Olympus IX71). The DMEM medium was removed and rinsed several times with phosphate buffer solution (PBS). Subsequently, the L929 cells on samples stained with Lysotracker Green DND-26 (250 nmol/L, 0.2 mL/well) were incubated in the dark at room temperature for 20 min. Then the medium was removed and washed twice with PBS. The cells were fixed at a 4% formaldehyde solution in the dark at 4  C for 20 min. Finally, the sample was washed twice with PBS and taken out, and then imaged by a Fluorescence Microscope. 2.10. Statistic analysis

2.8. Differential scanning calorimetry (DSC) measurement The melting behavior of the specimen was determined by a PerkineElmer diamond-II differential scanning calorimetry (DSC). The data were collected from 40

In the following studies in which there were at least three samples, statistical analysis was performed using a Student’s t-test for two-tailed distributions with unequal variance. Significance was assigned to p < 0.05.

Fig. 4. Morphology and structure of MP-UHMWPE and CM-UHMWPE. (a1) POM image of MP-UHMWPE; SEM images of (a2) cryo-fracture surface of MP-UHMWPE; (a3) cryofracture surface of CM-UHMWPE; (b1) schematic diagram of outer and inner layers of the sample, the arrow above represents width direction; (b2) out layer of MP-UHMWPE, ULMWPE-rich region; (b3) out layer of MP-UHMWPE, UHMWPE-rich region; (b4) inner layer of MP-UHMWPE; (c) outer layer of CM-UHMWPE. The shear flow direction is parallel.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11

3. Results 3.1. Mechanical and tribological properties Fig. 2 shows the mechanical properties of MP-UHMWPE samples, herein, the results of CM-UHMWPE samples are also presented for comparison. In comparison to CM-UHMWPE, MPUHMWPE exhibits a dramatical improvement in yield strength, from 20.3  0.5 to 46.3  4.4 MPa (p ¼ 0.0037) with an increment of 128.0%. The ultimate tensile strength (p ¼ 0.0011) and Young’s modulus (p ¼ 0.0002) of MP-UHMWPE also increase remarkably from 41.6  1.4 MPa and 662.7  24.1 MPa for CM-UHMWPE up to 65.5  3.1 MPa and 1248.7  45.3 MPa for MP-UHMWPE, respectively. The impact toughness is also elevated as manifested by impact strength rising from 68.7  1.3 kJ/m2 for CM-UHMWPE to 90.6  1.6 kJ/m2 for MP-UHMWPE. Meanwhile, the retention rate of ultimate tensile strength of MP-UHMWPE samples after being fatigue tested is 77.3  7.1%, which outperforms CM-UHMWPE (53.4  6.0%) by 44.8%, demonstrating a superior (p ¼ 0.0001)

5

fatigue resistance to the normal UHMWPE sample. In addition, the elongation at break of MP-UHMWPE (83.9  3.1%) is reduced compared to CM-UHMWPE, but is still adequate for knee joint implants [36,37]. Fig. 3 shows the friction coefficient as a function of distance traveled during wear for MP-UHMWPE and CM-UHMWPE samples, respectively. The tests were performed in bovine calf serum to simulate the in vivo environment. The friction coefficients remain invariable as the distance exceeds 2800 m. The friction coefficient of MP-UHMWPE (0.15  0.03) is below that of CM-UHMWPE (0.18  0.03, p ¼ 0.0001). Fig. 3 also presents the wear rates of MP-UHMWPE and CM-UHMWPE (see the inset). MP-UHMWPE exhibits a notably dropped wear rate (7.1  0.3 mg/MC) in comparison with CM-UHMWPE (10.0  0.4 mg/MC; p ¼ 0.0079). Moreover, a compression molded 10 wt% ULMWPE/90 wt% UHMWPE blend was prepared to examine the effect of 10 wt% ULMWPE on the mechanical properties, and shows an ultimate tensile strength of 38.7  2.1 MPa, which is comparable to CMUHMWPE (p ¼ 0.667).

Fig. 4. (continued).

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

6

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11

3.2. Morphology Fig. 4 shows the POM and SEM micrographs of MP-UHMWPE and CM-UHMWPE samples. One can clearly observe that, in the MP-UHMWPE sample, the minor ULMWPE regions (ULMWPE-rich region) localize in the thin interfaces of neighboring UHMWPE matrix regions (UHMWPE-rich regions), which reminds of a typical segregated structure [38] (Fig. 4a1). The UHMWPE-rich and ULMWPE-rich regions are well consolidated without any structure defects (Fig. 4a2), while in the CM-UHMWPE sample (Fig. 4a3), explicit granule boundaries with some structure defects (voids and cracks) are very visible. To clearly observe the crystalline morphology, the amorphous phase in the cryo-fracture surface was etched away. For simplicity, we referred to the layer from surface to ca. 2.5 mm along the width direction as outer layer while the layer from ca. 2.5 mm to the core as inner layer, as shown in Fig. 4b1. As expected, numerous interlocked shish-kebab entities are observed in the ULMWPE-rich region in the outer layer of the MP-UHMWPE (see Fig. 4b2). The

physical origin of such self-reinforcing superstructure formation refers to our previous work [25,26], and is not described here for brevity. Interestingly, in the UHMWPE-rich region, large amounts of oriented lamellae appear in both outer and inner layers (Fig. 4b3, b4), which distinctly differ from the self-reinforced UHMWPE/ LMWPE blends reported in our previous work [25,26], where only isotropic lamellae of UHMWPE appeared even in the outer layer. The reason for oriented UHMWPE lamellae existing in the whole sample will be discussed in the Discussion section. As for CMUHMWPE, only randomly distributed crystalline lamellae are detected throughout the sample, without any traces of oriented crystals, and as an example, one SEM image from the outer layer is shown here (Fig. 4c). 3.3. Crystalline structure and molecular orientation Fig. 5 shows WAXD patterns of MP-UHMWPE and CMUHMWPE, in which two strong diffraction reflections from inner to outer circles represent the (110) and (200) crystal planes of

Fig. 5. WAXD patterns of (a) MP-UHMWPE and (b) CM-UHMWPE along the transverse direction, the shear flow direction is vertical; (c) degree of orientation of (A) MP-UHMWPE and (B) CM-UHMWPE along the transverse direction.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11

polyethylene, respectively. For MP-UHMWPE, the arc-like diffractions of (110) plane are clearly seen in all the positions (Fig. 5a), which is ascribed to the existence of oriented crystals. While for CM-UHMWPE, only isotropic diffraction circles appear (Fig. 5b). The orientation parameters were estimated from the intensity of (110) along azimuthal angle, as shown in Fig. 5c. The molecular chains are oriented in all the positions of MP-UHMWPE, and their degree of orientation reaches a rather high level in both outer and inner layers. In contrast, the crystalline structure of CM-UHMWPE is absolutely. Fig. 6 shows SAXS patterns of MP-UHMWPE and CM-UHMWPE. For MP-UHMWPE, equatorial streak and meridional scattering maxima appear in outer layer (Fig. 6a). The appearance of equatorial streak indicates the existence of shish structures and the emergence of meridional scattering maxima refers to the kebabs [32,39], which grow perpendicularly to the shish. It is worth noting that the inner layer also exhibits the meridional maximum apart from a relatively weak signal of equatorial streak, which is probably ascribed to the large quantity of oriented crystals in this layer (Fig. 4b4). As expected, CM-UHMWPE displays only isotropic

7

diffraction circles in the whole sample (Fig. 6b). Based on the Lorentz-corrected intensity profiles of circularly integrated SAXS patterns (see Fig. 7), using Bragg’s law, the long spacing (L) is estimated and presented in Fig. 6c. The long period of both MPUHMWPE and CM-UHMWPE does not change considerably from different layers as a result of nearly the same thickness of polyethylene lamellae throughout the whole sample. However, the long period of MP-UHMWPE is less than that of CM-UHMWPE, which indicates existence of flow field leads to either denser packing of the lamellae or more nuclei. 3.4. Thermal behavior Fig. 8 shows the DSC melting curves of MP-UHMWPE and CMUHMWPE. In the case of MP-UHMWPE, the melting temperature is ca. 132.0  C in the whole sample (Fig. 8a). A second melting peak at 134.3  C exists in the DSC curves in outer layer (Fig. 8a), which can be ascribed to the melting of shishes [40]. For CM-UHMWPE, there is only one melting peak at around 132.0  C in both outer and inner layers (Fig. 8b). The crystallinity of MP-UHMWPE and

Fig. 6. SAXS patterns of (a) MP-UHMWPE and (b) CM-UHMWPE along the transverse direction, the shear flow direction is vertical; (c) long spacing (L) at different depths of (A) MPUHMWPE and (B) CM-UHMWPE samples.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

8

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11

region (Fig. 4b3), respectively. With such a unique superstructure combined with less structural defects (Fig. 4a2) and increased crystallinity (Fig. 8), the comprehensive performance of the MPUHMWPE samples was dramatically improved, of which, especially, the yield strength was considerably increased by 128.0% (Fig. 2; p ¼ 0.0037). Because failure of a polymer material is basically marked by yielding, the increased yield strength is of high significance which could endow the knee implant material with a higher ability to withstand external stress. These superior performances allow to consider MP-UHMWPE as a new generation UHMWPE material of total knee implants. In this part, we will try to reveal the mechanism behind the performance through microstructure evolution in MP-UHMWPE sample. First of all, the fluidity of MP-UHMWPE is successfully warranted by incorporation of ULMWPE (see Fig. S6). When heated above the melting temperature, the short ULMWPE chains tend to permeate between the UHMWPE granules like a lubricant, as schematically shown in Fig. 10a. In factual cases, MP-UHMWPE was successfully melt processed by injection molding. This is the first time to obtain an injection molded UHMWPE based material for knee implants. The processing technology and method used greatly affect the internal structure and morphology of a polymer, which then play a pivotal role in determining the performance of the final products. During injection molding, the application of the intense shear acting as a driving force could intensify self-diffusion of chain

Fig. 7. Lorentz-corrected SAXS intensity profiles of (a) MP-UHMWPE blend and (b) CM-UHMWPE along the transverse direction.

CM-UHMWPE is 62.5e65.0% and 55.9e56.7%, respectively, as shown in Fig. 8. 3.5. Biocompatibility The cytotoxicity of MP-UHMWPE and CM-UHMWPE was assessed by incubating fibroblast (L929) in culture medium. Fig. 9 shows representative fluorescent microscopy images of the cells incubated for 2 days. Morphological observation suggests that all cells are attached, spread, and proliferated nicely on the MPUHMWPE and CM-UHMWPE sheets (Fig. 9a and b) without considerably reduced in numbers, similar to those adhered and grown on a cell culture plate (Fig. 9c). Meanwhile, the MMT assay of cells (Fig. 9d) shows a slight increase in the cell numbers on MPUHMWPE sample compared with CM-UHMWPE one; more cells adhere and grow on MP-UHMWPE (p ¼ 0.11). These results indicate the biocompatibility of MP-UHMWPE is comparable to that of CMUHMWPE. 4. Discussion The above results demonstrated that, with the incorporation of 9.8 wt% ULMWPE acting as a flow accelerator, UHMWPE was successfully melt processed. By applying shear flow during injection molding, large quantities of shish-kebabs and orientated lamellae were formed in ULMWPE-rich region (Fig. 4b2) and UHMWPE-rich

Fig. 8. DSC curves of (a) MP-UHMWPE and (b) CM-UHMWPE from outer and inner layers.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11

9

Fig. 9. Fluorescent images of fibroblast cells cultured for 2 days on (a) MP-UHMWPE, (b) CM-UHMWPE, and on (c) a standard culture plate; (d) MMT results of the MP-UHMWPE and CM-UHMWPE showing the survival rate of fibroblast cultured for 2 days.

entanglements across the granule boundaries, leading to reduction of structural defects (Fig. 4a2) with a concomitant decrease of stress concentration in MP-UHMWPE. Aside from promoting intergranule diffusion, the strong shear force imposed on polymer melt also induces formation of oriented crystals, i.e. shish-kebabs and oriented lamellae etc. [41e44], which are fascinating selfreinforced features that together with less structural defects, can render improved mechanical properties. The formation of shishkebabs, as verified by our previous work [25,26], could be efficiently facilitated by the pre-additive 2 wt% UHMWPE acting as shear induced nucleation precursors. This is the reason that 2 wt% UHMWPE in respect to the ULMWPE content used was separately, uniformly mixed with ULMWPE in the current work. Moreover, the existence of 9.8 wt% ULMWPE could facilitate the formation of another type of oriented crystals, i.e., oriented lamellae. While some ULMWPE tends to locate between the UHMWPE granules to form the ULMWPE-rich region, a portion of ULMWPE is liable to penetrate into UHMWPE granules to create

the UHMWPE-rich region (Fig. 10b), in which the chain mobility is intensified [45]. Upon shear flow, UHMWPE molecular chains themselves could, to some extent, slip and align along the flow direction. The orientation-induced crystallization occurs therein, thereby giving rise to large amounts of oriented lamellae (Fig. 4b3, b4). In fact, these oriented lamellae interlock each other, appearing in the form of interlocked oriented lamellae (Fig. 4b3), which has never been reported before. The origin of interlocked lamellae formation can be ascribed to the special processing conditions. Before frozen (or crystallized), the molecules are continuously subjected to intense reciprocating shear flow which can promote the adjacent UHMWPE lamellae to inset each other (Fig. 4b3). The interlocked lamellae, which can prevent the slippage of molecular chains and bear higher external stress, should be also responsible for the significantly increased yield strength (46.3  4.4 MPa) and ultimate tensile strength (65.5  3.1 MPa) of MP-UHMWPE as compared to those of conventional UHMWPE (ca. 20 MPa [46e48] and 40e55 MPa [7,48e51], respectively)

Fig. 10. Schematic diagram of (a) ULMWPE and UHMWPE phase distribution in MP-UHMWPE, and (b) local crystal structures of MP-UHMWPE.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

10

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11

reported in other literatures. This is encouraging because highly improved tensile strength holds a great application promise for expanding the use of implant material to younger and more active patients. We also noticed that a difference existed between the impact strength of CM-UHMWPE (68.7  1.3 kJ/m2) and the data reported by others (ca. 120 kJ/m2), which may be ascribed to the different materials used, and more importantly, the different test method [47,52,53]. It is worth noting that the lamellae orientation appears in the inner layer of MP-UHMWPE (Fig. 4b4, 5, and 6) as well, which is seldom reported before. Although the shear stress also exists in the inner layer, the slow solidifying rate leaves enough time for the relaxation of extended PE chains, and finally PE melt usually crystallizes into isotropic lamellae other than oriented ones. What caused the change of crystal structure in the inner layer? One reasonable explanation seems to be the tremendous difference in melt viscosity between two types of polyethylene, which leads to a shear amplification effect [54], and thereby enhances the molecular orientation and preserves more oriented crystals (Fig. 5c). In addition to mechanical properties, fatigue resistance and tribological properties are also improved (Figs. 2 and 3), which are of great benefit to prolong the life span of implant material. Fatigue resistance is closely related to crystallinity [46], thereby it is logic that MP-UHMWPE with significantly increased crystallinity (Fig. 8, p ¼ 0.0046) exhibits much higher fatigue strength (p ¼ 0.0001) than CM-UHMWPE (Fig. 2). It is generally accepted that the decreased ductility is helpful for wear reduction [55], and the reduced plastic deformation on contact surfaces would result in decreased friction coefficient [56]. The oriented crystals that possess the rigid structure in MP-UHMWPE improve its stiffness, and thus could efficaciously contribute to wear resistance [25] (Fig. 3). Furthermore, large amount of oriented crystals are distributed in the outer layer, which serves as the articular surface of the implant, providing decreased ductility and plasticity [25], and consequently help reducing the wear coefficient (Fig. 3). These oriented crystals substantially paralleling to the direction of motion for total knee implants would bring great performance enhancement. Meanwhile in the transverse direction, the interlocked crystalline state could enhance the interfacial adhesion between separate orientated superstructure, as demonstrated in our previous work [25], and is indeed efficacious to provide a tendency to homogenize properties of the sample in all direction. As a consequence, the mechanical and tribological properties would not deteriorate in the orthogonal direction. The biocompatibility of MP-UHMWPE plays a crucial role in its application to joint implants. Inspiringly, MP-UHMWPE shows comparable biocompatibility to CM-UHMWPE (Fig. 9; p ¼ 0.11). This is because polyethylene, such as ULMWPE, has good biological compatibility [27e29]. 5. Conclusions We have successfully fabricated melt processable and selfreinforced UHMWPE with the incorporation of 9.8 wt% ULMWPE acting as a flow accelerator, and its injection molded parts show significantly improved tensile strength, fatigue resistance as well as wear resistance. The structural defects were reduced, and large amount of interlocked shish-kebabs and orientated lamellaes self-reinforced superstructure were formed through injection molding process, which together with the increased crystallinity, contribute to the improved performance. We believe the melt processed and self-reinforced UHMWPE parts hold a great application promise for use of knee joint implants, particularly for younger and more active patients.

Acknowledgments The authors gratefully acknowledge the financial support from the Program of National Natural Science of China (Grants 51121001, 51033004, 51273114, 51227801) and Science & Technology Department of Sichuan Province (Grant 2013TD0013). We are heavily indebted to Prof. Benjamin S. Hsiao and Miss Yan-Hui Chen for the help of synchrotron X-ray scattering measurement in the Brookhaven National Laboratory (USA). We also would like to express sincere thanks to the Shanghai Synchrotron Radiation Facility (SSRF, Shanghai, China) for the kind help on WAXD and SAXS measurements. Appendix A. Supplementary data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.biomaterials.2014.04.077. References [1] Kurtz SM, Muratoglu OK, Evans M, Edidin AA. Advances in the processing, sterilization, and crosslinking of ultra-high molecular weight polyethylene for total joint arthroplasty. Biomaterials 1999;20:1659e88. [2] Oral E, Wannomae KK, Hawkins N, Harris WH, Muratoglu OK. a-Tocopheroldoped irradiated UHMWPE for high fatigue resistance and low wear. Biomaterials 2004;25:5515e22. [3] Oral E, Wannomae KK, Rowell SL, Muratoglu OK. Diffusion of vitamin E in ultra-high molecular weight polyethylene. Biomaterials 2007;28:5225e37. [4] Oral E, Christensen SD, Malhi AS, Wannomae KK, Muratoglu OK. Wear resistance and mechanical properties of highly cross-linked, ultrahigh-molecular weight polyethylene doped with vitamin E. J Arthroplasty 2006;21:580e91. [5] Kurtz SM. UHMWPE biomaterials handbook: ultra high molecular weight polyethylene in total joint replacement and medical devices. New York: Elsevier Inc; 2009. [6] Lerf R, Zurbrügg D, Delfosse D. Use of vitamin E to protect cross-linked UHMWPE from oxidation. Biomaterials 2010;31:3643e8. [7] Gijsman P, Smelt HJ, Schumann D. Hindered amine light stabilizers: an alternative for radiation cross-linked UHMwPE implants. Biomaterials 2010;31:6685e91. [8] Fu J, Shen J, Gao G, Hou R, Xu Y, Cong Y, et al. Natural polyphenol-stabilised highly crosslinked UHMWPE with high mechanical properties and low wear for joint implants. J Mater Chem B 2013;1:4727e35. [9] Muratoglu OK, Bragdon CR, O’Connor DO, Jasty M, Harris WH, Gul R, et al. Unified wear model for highly crosslinked ultra-high molecular weight polyethylenes (UHMWPE). Biomaterials 1999;20:1463e70. [10] Wu JJ, Buckley CP, O’Connor JJ. Mechanical integrity of compression-moulded ultra-high molecular weight polyethylene: effects of varying process conditions. Biomaterials 2002;23:3773e83. [11] Buckley CP, Wu J, Haughie DW. The integrity of welded interfaces in ultra high molecular weight polyethylene: part 1emodel. Biomaterials 2006;27:3178e 86. [12] Wrona M, Mayor MB, Collier JP, Jensen RE. The correlation between fusion defects and damage in tibial polyethylene bearings. Clin Orthop Relat Res 1994;299:92e103. [13] Landy MM, Walker PS. Wear of ultra-high-molecular-weight polyethylene components of 90 retrieved knee prostheses. J Arthroplasty 1988;3:S73e85. [14] Gul RM, McGarry FJ. Processing of ultra-high molecular weight polyethylene by hot isostatic pressing, and the effect of processing parameters on its microstructure. Polym Eng Sci 2004;44:1848e57. [15] James SP, Lee KR, Beauregard GP, Rentfrow ED, McLaughlin JR. Clinical wear of 63 ultrahigh molecular weight polyethylene acetabular components: effect of starting resin and forming method. J Biomed Mater Res 1999;48:374e84. [16] Smith P, Lemstra PJ. Ultra-high-strength polyethylene filaments by solution spinning/drawing. J Mater Sci 1980;15:505e14. [17] Zwijnenburg A, Pennings AJ. Longitudinal growth of polymer crystals from flowing solutions III. Polyethylene crystals in Couette flow. Coll Polym Sci 1976;254:868e81. [18] Lee EM, Oh YS, Ha HS, Kim BK. Rheological properties of UHMWPE/iPP blends. Polym Adv Tech 2009;20:1121e6. [19] Xie M, Li H. Viscosity reduction and disentanglement in ultrahigh molecular weight polyethylene melt: effect of blending with polypropylene and poly (ethylene glycol). Eur Polym J 2007;43:3480e7. [20] Alcantar NA, Aydil ES, Israelachvili JN. Polyethylene glycol-coated biocompatible surfaces. J Biomed Mater Res 2000;51:343e51. [21] Li X, Zou W, Jiang T. Preparation of UHMWPE/PLA blends. China Synth Resin Plast 2008;25:11e5. [22] Wang X, Wu Q, Qi Z. Unusual rheology behaviour of ultra high molecular weight polyethylene/kaolin composites prepared via polymerization-filling. Polym Int 2003;52:1078e82.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077

Y.-F. Huang et al. / Biomaterials xxx (2014) 1e11 [23] Aiello R, La Mantia FP. On the improvement of the processability of UHMWHDPE by adding a liquid crystalline polymer and a fluoroelastomer. Macromol Mater Eng 2001;286:176e8. [24] Utsumi M, Nagata K, Suzuki M, Mori A, Sakuramoto I, Torigoe Y, et al. Effects of calcium stearate addition of ultrahigh molecular weight polyethylene in direct compression molding. J Appl Polym Sci 2003;87:1602e9. [25] Xu L, Chen C, Zhong GJ, Lei J, Xu JZ, Hsiao BS, et al. Tuning the superstructure of ultrahigh-molecular-weight polyethylene/low-molecular-weight polyethylene blend for artificial joint application. ACS Appl Mater Interfaces 2012;4:1521e9. [26] Huang YF, Xu JZ, Xu JY, Zhang ZC, Hsiao BS, Xu L, et al. Self-seinforced polyethylene blend for artificial joint application. J Mater Chem B 2014;2: 971e80. [27] Alothman OY, Almajhdi FN, Fouad H. Effect of gamma radiation and accelerated aging on the mechanical and thermal behavior of HDPE/HA nanocomposites for bone tissue regeneration. Biomed Eng Online 2013;12:95. [28] Fouad H, Elleithy R. High density polyethylene/graphite nano-composites for total hip joint replacements: processing and in vitro characterization. J Mech Behav Biomed Mater 2011;4:1376e83. [29] Zhil’tsova TV, Oliveira MSA, Ferreira JAF. Relative influence of injection molding processing conditions on HDPE acetabular cups dimensional stability. J Mater Proc Tech 2009;209:3894e904. [30] Zuo F, Keum JK, Yang L, Somani RH, Hsiao BS. Thermal stability of shearinduced shish-kebab precursor structure from high molecular weight polyethylene chains. Macromolecules 2006;39:2209e18. [31] Seki M, Thurman DW, Oberhauser JP, Kornfield JA. Shear-mediated crystallization of isotactic polypropylene: the role of long chain-long chain overlap. Macromolecules 2002;35:2583e94. [32] Somani RH, Hsiao BS, Nogales A, Srinivas S, Tsou AH, Sics I, et al. Structure development during shear flow-induced crystallization of i-PP: in-situ smallangle X-ray scattering study. Macromolecules 2000;33:9385e94. [33] Keum JK, Zuo F, Hsiao BS. Formation and stability of shear-induced shishkebab structure in highly entangled melts of UHMWPE/HDPE blends. Macromolecules 2008;41:4766e76. [34] Picken SJ, Aerts J, Visser R, Northolt MG. Structure and rheology of aramid solutions: X-ray scattering measurements. Macromolecules 1990;23: 3849e54. [35] Mykhaylyk OO, Chambon P, Graham RS, Fairclough JPA, Olmsted PD, Ryan AJ. The specific work of flow as a criterion for orientation in polymer crystallization. Macromolecules 2008;41:1901e4. [36] Oonishi H, Kuno M, Tsuji E, Fujisawa A. The optimum dose of gamma radiation-heavy doses to low wear polyethylene in total hip prostheses. J Mater Sci Mater Med 1997;8:11e8. [37] Oonishi H, Kadoya Y, Masuda S. Gamma-irradiated cross-linked polyethylene in total hip replacementsdanalysis of retrieved sockets after long-term implantation. J Biomed Mater Res 2001;58:167e71. [38] Kyrylyuk AV, Hermant MC, Schilling T, Klumperman B, Koning CE, van der Schoot P. Controlling electrical percolation in multicomponent carbon nanotube dispersions. Nat Nanotechnol 2011;6:364e9. [39] Yang L, Somani RH, Sics I, Hsiao BS, Kolb R, Fruitwala H, et al. Shear-induced crystallization precursor studies in model polyethylene blends by in-situ rheo-SAXS and rheo-WAXD. Macromolecules 2004;37:4845e59.

11

[40] Yang J, Wang C, Wang K, Zhang Q, Chen F, Du R, et al. Direct formation of nanohybrid shish-kebab in the injection molded bar of polyethylene/multiwalled carbon nanotubes composite. Macromolecules 2009;42:7016e23. [41] Kimata S, Sakurai T, Nozue Y, Kasahara T, Yamaguchi N, Karino T, et al. Molecular basis of the shish-kebab morphology in polymer crystallization. Science 2007;316:1014e7. [42] Chen YH, Zhong GJ, Wang Y, Li ZM, Li L. Unusual tuning of mechanical properties of isotactic polypropylene using counteraction of shear flow and bnucleating agent on b-form nucleation. Macromolecules 2009;42:4343e8. [43] Koppi KA, Tirrell M, Bates FS, Almdal K, Colby RH. Lamellae orientation in dynamically sheared diblock copolymer melts. J Phys II 1992;2:1941e59. [44] Pujari S, Keaton MA, Chaikin PM, Register RA. Alignment of perpendicular lamellae in block copolymer thin films by shearing. Soft Matter 2012;8:5358e 63. [45] Zuo JD, Zhu YM, Liu SM, Jiang ZJ, Zhao JQ. Preparation of HDPE/UHMWPE/ MMWPE blends by two-step processing way and properties of blown films. Polym Bull 2007;58:711e22. [46] Oral E, Malhi AS, Muratoglu OK. Mechanisms of decrease in fatigue crack propagation resistance in irradiated and melted UHMWPE. Biomaterials 2006;27:917e25. [47] Fu J, Ghali BW, Lozynsky AJ, Oral E, Muratoglu OK. Ultra high molecular weight polyethylene with improved plasticity and toughness by high temperature melting. Polymer 2010;51:2721e31. [48] Xiong L, Xiong DS. The influence of irradiation dose on mechanical properties and wear resistance of molded and extruded ultra high molecular weight polyethylene. J Mech Behav Biomed Mater 2012;9:73e82. [49] Oral E, Godleski Beckos CA, Lozynsky AJ, Malhi AS, Muratoglu OK. Improved resistance to wear and fatigue fracture in high pressure crystallized vitamin Econtaining ultra-high molecular weight polyethylene. Biomaterials 2009;30: 1870e80. [50] Fang L, Leng Y, Gao P. Processing and mechanical properties of HA/UHMWPE nanocomposites. Biomaterials 2006;27:3701e7. [51] Kurtz SM, Villarraga ML, Herr MP, Bergström JS, Rimnac CM, Edidin AA. Thermomechanical behavior of virgin and highly crosslinked ultra-high molecular weight polyethylene used in total joint replacements. Biomaterials 2002;23:3681e97. [52] Oral E, Ghali BW, Rowell SL, Micheli BR, Lozynsky AJ, Muratoglu OK. A surface crosslinked UHMWPE stabilized by vitamin E with low wear and high fatigue strength. Biomaterials 2010;31:7051e60. [53] Oral E, Neils AL, Lyons C, Fung M, Doshi B, Muratoglu OK. Surface cross-linked UHMWPE can enable the use of larger femoral heads in total joints. J Orthop Res 2013;31:59e66. [54] Xu L, Huang YF, Xu JZ, Ji X, Li ZM. Improved performance balance of polyethylene by simultaneously forming oriented crystals and blending ultrahighmolecular-weight polyethylene. RSC Adv 2014;4:1512e20. [55] Muratoglu OK, Bragdon CR, O’Connor DO, Jasty M, Harris WH. A novel method of cross-linking ultra-high-molecular-weight polyethylene to improve wear, reduce oxidation, and retain mechanical properties: recipient of the 1999 HAP Paul Award. J Arthroplasty 2001;16:149e60. [56] Peng QY, Cong PH, Liu XJ, Liu TX, Huang S, Li TS. The preparation of PVDF/clay nanocomposites and the investigation of their tribological properties. Wear 2009;266:713e20.

Please cite this article in press as: Huang Y-F, et al., Mechanical properties and biocompatibility of melt processed, self-reinforced ultrahigh molecular weight polyethylene, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.04.077