In vitro evaluation of hydroxyapatite reinforced polyhydroxybutyrate ...

Materials Science and Engineering C 20 (2002) 101 – 109 www.elsevier.com/locate/msec

In vitro evaluation of hydroxyapatite reinforced polyhydroxybutyrate composite J. Ni, M. Wang* School of Mechanical and Production Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore

Abstract Particulate hydroxyapatite (HA) was incorporated into polyhydroxybutyrate (PHB) to form a bioactive and biodegradable composite for applications in hard tissue replacement and regeneration. HA/PHB composite containing 10, 20, and 30 vol.% of HA was made for in vitro evaluation. In vitro studies were conducted using an acellular simulated body fluid (SBF). Composite specimens were immersed in SBF at 37 jC for various periods of time prior to surface analysis and mechanical testing. Results obtained from scanning electron microscopic (SEM) examination, thin film X-ray diffraction (TF-XRD) analysis, and Fourier transform infrared (FTIR) spectroscopy showed that a layer of bonelike apatite formed within a short period on HA/PHB composite after its immersion in SBF, demonstrating high in vitro bioactivity of the composite. The bioactivity and mechanical properties of the composite could be changed by varying the amount of HA in the composite. Dynamic mechanical analysis (DMA) revealed that the storage modulus (EV) of HA/PHB composite increased initially with immersion time in SBF, due to apatite formation on composite surface and decreased after prolonged immersion in SBF, indicating degradation of the composite in a simulated body environment. HA/PHB composite thus has the potential for its intended applications. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Hydroxyapatite; Polyhydroxybutyrate; Composite; In vitro behaviour; Bioactivity; Biodegradation

1. Introduction The original concept of using a bioceramic to reinforce a polymer was introduced by Bonfield et al. [1] in the early 1980s. Various particulate bioceramics as the reinforcement (and also as the bioactive phase) have been incorporated into biocompatible and ‘‘bio-stable’’ polymers to produce bone substitutes [1– 6], considering the fact that human cortical bone at the ultrastructure level is a composite consisting of nanometer-size apatite crystals and collagen fibrils [7]. The mechanical as well as biological performance of bioactive ceramic/polymer composites can be controlled through using different particulate bioceramics for the composites and also through varying the amounts of bioceramic particles in the composite [8]. Hydroxyapatite (HA) reinforced high-density polyethylene (HDPE), the first bioceramic/polymer composite, has been used clinically for orbital floor reconstruction and orbital volume augmentation [9], and in middle ear implants [10]. In recent years, emphasis in biomaterial engineering has moved from materials that remain stable in the biological *

Corresponding author. Tel.: +65-790-5151; fax: +65-791-1859. E-mail address: [email protected] (M. Wang).

environment to materials that can in some way alter their properties (i.e., ‘‘biodegrade’’) in response to the cellular and extracellular environment. Biodegradable materials are designed to degrade gradually in the body and will be replaced eventually by newly formed tissues. After being implanted in the body, a biodegradable bone substitute material will have gradual decreases in strength and stiffness over a clinically determined optimal period. As bone repairs itself, the external load will be transferred from the biodegrading implant to bone. This approach provides the best biomaterials solution to tissue replacement and regeneration, if requirements for the initial strength and stiffness and other short-term performance can be met. Following this philosophy and using established composite technology, biodegradable composites consisting of particulate bioceramics and polyhydroxybutyrate (PHB), which is a natural polymer and can degrade in the human body environment, were manufactured for potential applications in hard tissue repair [11,12]. These composites should be fully characterised and assessed before they can be used clinically. In vivo studies are essential for biomaterials that are currently developed. However, results obtained from in vivo experiments are often difficult to interpret due to the complexity of various cellular responses. On the other hand, in

0928-4931/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 8 - 4 9 3 1 ( 0 2 ) 0 0 0 1 9 - X

102

J. Ni, M. Wang / Materials Science and Engineering C 20 (2002) 101–109

vitro studies can provide information that indicates the in vivo performance of materials [13,14]. A common feature of bioactive materials is that their surfaces develop a biologically active hydroxy carbonate apatite (HCA) layer after implantation, which is essential for establishing bonding with bone [15]. For a material to be bioactive in vivo, it must have the ability to induce HCA formation on its surface in vitro. A bioactive material designed to be degradable in the human body should also exhibit degradation in a simulated body environment. In this investigation, the surface structure of hydroxyapatite reinforced polyhydroxybutyrate (HA/PHB) composite after its immersion in a simulated body fluid (SBF) was studied. The mechanical performance of the composite was also evaluated before and after immersion in SBF.

of HA were embedded in an acrylic resin, and ground and polished for examining their cross-sections. Rectangular specimens of the composite with dimensions of 2051 mm were also immersed in SBF at 37 jC for 2 weeks, 1, 2, and 4 months, respectively. Mechanical properties of these rectangular specimens were evaluated using dynamic mechanical analysis (DMA) after their preset immersion time had been reached.

2. Materials and methods

2.4. Thin-film X-ray diffraction analysis (TF-XRD)

2.1. Composite and test specimens

TF-XRD analysis of immersed specimens was performed on a Rigaku DMAX 22001 X-ray diffractometer with a thinfilm analyzer. Cu Ka radiation was used for the diffraction with a voltage of 40 kV and a current of 30 mA. Specimens were aligned at 1j to the incident beam. A step size of 0.02j (2h) and a scan speed of 3j/min were used and the diffraction data were collected from 20j to 50j (2h).

Commercially available HA (Taihei Chemicals, Japan) and PHB (ICI, UK) were raw materials for making the composite, with both HA and PHB being used in their asreceived state without further treatment. The raw materials were fully characterised prior to composite production. The median particle size of as-received HA was 24.5 Am. X-ray diffraction (XRD) analysis of the ceramic powder showed that it was phase-pure HA. XRD patterns of PHB only exhibited peaks of the HB unit. (PHB is a semicrystalline polymer and, hence, shows diffraction peaks.) Manufacture of the HA/PHB composite followed a standardised procedure, which consisted of compounding, milling, and compression molding [12]. Composite containing 0, 10, 20, and 30 vol.% of HA was used in this investigation. Test specimens were cut from compression molded plates, ground, and polished using up to #1000 sandpaper to remove any defects. They were then cleaned in an ultrasonic bath containing distilled water and dried in an oven before they were used in various tests. 2.2. Immersion in simulated body fluid (SBF) An acellular simulated body fluid (SBF) was used for in vitro experiments. The SBF was prepared by dissolving reagent-grade chemicals of NaCl, NaHCO3, KCl, K2HPO4 3H2O, MgCl26H2O, CaCl22H2O, Na2SO4, and (CH2 OH)3CNH2 into distilled and deionised water and buffered with HCl to pH 7.4 at 37 jC. It had ion concentrations that were nearly the same as those in human blood plasma [16]. Specimens with a surface area of 1520 mm were immersed in SBF at 37 jC for various periods of time. 100 ml of SBF was used for each specimen. Changes of the surface structure of immersed specimens were analysed after the specimens had been removed from SBF, washed with distilled water, and dried. Furthermore, after drying, some immersed specimens of the composite containing 20 vol.%

2.3. Scanning electron microscopy (SEM) Before SEM examination, the surface and polished crosssections of immersed specimens were coated with a thin layer of platinum. These specimens were then examined using a JEOL JSM5600LV SEM at an accelerating voltage of 20 kV.

2.5. Fourier transform infrared spectroscopy (FTIR) FTIR spectra of mineral crystals formed in vitro were obtained using a Perkin Elmer System 2000 Fourier transform infrared spectrometer. A small amount of the mineral crystals was gathered from the surface of each immersed specimen, milled with KBr and pressed into a transparent film for FTIR analysis. FTIR spectra were collected over the range of 4000– 400 cm1. 2.6. Dynamic mechanical analysis (DMA) Mechanical properties of HA/PHB composite before and after immersion in SBF were evaluated using a Perkin Elmer DMA 7 system. A 15 mm knife edge, three-point bending platform with a 5 mm probe tip was used. DMA tests were performed in a temperature range of 30 –100 jC and at a heating rate of 4 jC/min. Assuming that the physiological frequency was 1 Hz, all tests were conducted at this frequency. From DMA tests, three properties, namely, storage modulus (EV), loss modulus (EW), and loss tangent (tand), could be determined. For polymers and their composites, the following equations exist [17]: E ¼ EV þ iEW tand ¼ EW=EV

ð1Þ ð2Þ

where E is the dynamic modulus. The storage modulus represents the capability of a material to store mechanical energy and resist deformation. The higher the storage modulus, the stiffer the material.


103

3. Results 3.1. Scanning electron microscopy (SEM) Mineral crystals were observed to grow on all HA/PHB composite specimens only after 1-day immersion in SBF. Fig. 1a shows the crystals formed on the surface of the composite containing 10 vol.% of HA. The crystals grew locally on the specimen surface following the contour of the specimen. After 3 days immersion, mineral crystals covered nearly the whole surface of the specimen, but the mineral layer formed was very thin and, hence, the original contour of the specimen could still be seen (Fig. 1b). After 7 days immersion, the mineral layer grew thicker and the surface of the mineral layer was considerably less rough (Fig. 1c). The morphology of the mineral layer formed after immersion in SBF for 2, 4, and 8 weeks was similar to that of the mineral layer formed after immersion in SBF for 7 days. In addition, nucleation of new mineral particles could be seen on the mineral layer already formed after 4 weeks immersion in SBF. The formation and growth of mineral crystals on HA/ PHB composite of other compositions was similar to that on 10 vol.% HA/PHB (Figs. 2 and 3), but the mineral crystals grew more rapidly on 20 and 30 vol.% HA/PHB than on 10 vol.% HA/PHB. For 20 vol.% HA/PHB, a thin and uniform mineral layer was formed after 3 days immersion (Fig. 2b) and new mineral particles were already present after 7 days immersion (Fig. 2c). The size of these new near-spherical crystals was around 3 – 4 Am. For 30% HA/PHB, the mineral layer was formed only after 1 day immersion (Fig. 3a) and more near-spherical crystals were observed on the specimen after 7 days immersion (Fig. 3c). These observations indicated that the mineral layer was formed layer by layer on the surface of the composite and that the composite containing larger amount of HA had greater ability to induce the formation of minerals in vitro. Fig. 4 shows at a high magnification the morphology of the mineral particles, revealing that the mineral particles were composed of small flake-like crystallites, which is similar in morphology to crystals formed on other biomaterials [16]. A cross-section of the mineral layer formed in vitro is shown in Fig. 5, displaying a compact structure. Mineral crystals were also found on unfilled PHB after its immersion in SBF for 14 days (Fig. 6a), but the growth of these crystals was very slow. After 28 days immersion, the mineral layer formed was still very thin and rough (Fig. 6b). Only after 56 days immersion was a uniform mineral layer observed (Fig. 6c). 3.2. Thin-film X-ray diffraction (TF-XRD) TF-XRD was used to determine the phase(s) of mineral crystals (or layer) formed on the composite surface. Fig. 7 shows TF-XRD patterns of 10, 20, and 30 vol.% HA/PHB composite before and after immersion in SBF. For both 10 and 20 vol.% HA/PHB composite, apatite peaks were first

Fig. 1. SEM micrographs of 10 vol.% HA/PHB composite after immersion in SBF for (a) 1 day, (b) 3 days, (c) 7 days.

observed after 7 days immersion, with one peak at 26j (2h) and another peak between 31j and 33j (Fig. 7a and b). The broad apatite peaks indicated low crystallinity of apatite formed in vitro at this early stage of immersion. The HA

104


and PHB peaks, which appeared in the XRD patterns of HA/ PHB composite before immersion in SBF, were suppressed in XRD patterns of the composite after 7 days immersion. The intensity of apatite peaks increased gradually with

immersion time, indicating the growth of an apatite layer on the composite surface in SBF and another peak at 49j appeared after 28 days immersion. For 30 vol.% HA/PHB composite, apatite peaks appeared in XRD patterns only




105

Fig. 4. Morphology of mineral crystals formed in vitro.

Fig. 5. Cross-sectional view of the apatite layer formed on 20 vol.% HA/ PHB composite after 28 days immersion in SBF.

after the specimen had been immersed in SBF for 3 days (Fig. 7c). Again, HA and PHB peaks of constituent materials of the composite were suppressed by apatite peaks in the XRD pattern after 3 days immersion and the intensity of apatite peaks in XRD patterns increased gradually with immersion time. 3.3. Fourier transform infrared spectroscopy (FTIR) FTIR was used to detect the presence of functional groups such as phosphate and carbonate groups, thus providing more information on the apatite formed on the surface of composite after its immersion in SBF. Fig. 8 displays FTIR spectra of apatite formed in vitro on 20 vol.% HA/PHB composite. There was no major difference between the spectrum obtained at 7 days immersion time (Fig. 8a) and the spectrum obtained at 14 days or longer immersion time (Fig. 8b). The characteristic absorption bands of phosphate appearing at 565, 604, and 962 cm1

Fig. 6. SEM micrographs of unfilled PHB after immersion in SBF for (a) 14 days, (b) 28 days, (c) 56 days.

were observed for all immersed specimens. The FTIR spectra of apatite formed in vitro also had a strong absorption band at 873 cm1 which corresponded to the vibration mode of carbonate, indicating the incorporation of carbonate groups in the structure of apatite. Other carbonate peaks at

106


Fig. 7. TF-XRD patterns of HA/PHB composite before and after immersion in SBF: (a) 10 vol.% HA/PHB, (b) 20 vol.% HA/PHB, (c) 30 vol.% HA/PHB.


107

3.4. Dynamic mechanical analysis (DMA) The storage modulus was determined for HA/PHB composite before and after immersion in SBF. The repeatability of measurements from three specimens of each composition was good. Therefore, only averaged data were plotted without error bars in order to present results clearly (Fig. 9). DMA results showed that the storage modulus of HA/PHB composite increased initially with an increase in immersion time. After immersion in SBF, the storage modulus of the composite became higher than that of the composite before immersion, with a larger increase in modulus for composite having a higher HA content. The storage modulus of unfilled PHB did not change within the first month of immersion. It increased only slightly after 2 months immersion. With prolonged immersion in SBF (i.e., beyond 2 months), both unfilled PHB and HA/PHB composite exhibited decreases in storage modulus.

4. Discussion

Fig. 8. FTIR spectra of apatite formed on 20 vol.% HA/PHB in SBF: (a) 7 days immersion, (b) 28 days immersion.

1415 and 1454 cm1 were observed as well. These results suggested that the apatite formed on the surface of composite in SBF was carbonated apatite, which is similar in composition and structure to bone apatite [18].

SEM examination and TF-XRD and FTIR analyses showed the formation and growth of a layer of carbonated apatite on the surface of HA/PHB composite as well as unfilled PHB after their immersion in SBF. This apatite is bone-like apatite, which was also found to form on various bioactive materials when they were immersed in SBF [16,18 –20]. The in vitro bioactivity of HA/PHB composite has thus been demonstrated. Apatite was found to form on HA/PHB composite within 1 day in SBF. Such a fast rate of apatite formation indicates that high in vitro bioactivity can be obtained for the composite. It was also observed that the number of nucleation sites of apatite crystals was proportional to the HA content of the composite and that the apatite layer grew more rapidly on composite containing greater amount of HA. The rate of formation and growth of an apatite layer

Fig. 9. Variation of storage modulus with immersion time in SBF for HA/PHB composite.

108


on the HA/PHB composite increased in the following order: 10 vol:% HA=PHB < 20 vol:% HA=PHB < 30 vol:% HA=PHB

ð3Þ

Therefore, the composite should have a high volume percentage of HA if a high degree of bioactivity is required. Apatite was also found to form on unfilled PHB after immersion in SBF for 14 days, but by introducing HA particles into the PHB matrix, a much higher degree of bioactivity can be obtained and controlled for the composite, which is an advantage of the composite approach. The incorporation of HA particles into PHB has also resulted in a composite having higher stiffness than PHB. It is shown in the present investigation that the storage modulus of HA/PHB composite increased with an increase in the HA volume percentage. Nazhat et al. [21] have demonstrated that the storage modulus of a bioceramic/ polymer composite is closely related to its Young’s modulus. Therefore, as PHB has a relatively high Young’s modulus among all biocompatible polymers, it can be deduced that HA/PHB composite containing high percentages of HA should have Young’s modulus values that are within the range for cortical bone. PHB is well known as a biocompatible and biodegradable thermoplastic polymer [22,23]. It was found in the present investigation that the storage moduli of PHB and its composite increased with immersion time (up to 2 months) in SBF. These increases can be attributed to the apatite layer formed on their surfaces. The apatite layer could be more than 10 Am thick after 28 days immersion in SBF and it was a dense layer (Fig. 5). This apatite layer separated composite from SBF and thus prevented the solution from attacking the matrix polymer. It therefore protected the composite against in vitro degradation at the initial stage of immersion. The stiff apatite layer may have also enhanced the storage modulus of the composite by physically being an additional component of the composite. This phenomenon is similar to observations made by Zhang and Ma [24] on the apatite/ poly(L-lactic acid) composite which exhibited an increase in compressive modulus after the immersion in SBF for 2 months. With prolonged immersion in SBF (i.e., beyond 2 months), storage moduli of PHB and HA/PHB composite decreased gradually, indicating degradation of these materials in a simulated body environment. It is believed that at this stage of immersion in SBF, liquid had penetrated the apatite layer and diffused into the composite, thus causing the matrix polymer to degrade.

5. Conclusions A biologically active apatite layer forms within a short period on HA/PHB composite after its immersion in SBF, demonstrating high in vitro bioactivity of the composite. The

bioactivity and mechanical properties of the composite can be tailored by varying the HA volume percentage in the composite. The storage modulus of the composite increases initially with immersion time in SBF, which is due to the formation of the apatite layer on composite surface and decreases after prolonged immersion in SBF, displaying degradation of the composite in a simulated body environment. HA/PHB composite has the potential as a bioactive and biodegradable material for applications in hard tissue replacement and regeneration.

Acknowledgements J. Ni thanks Nanyang Technological University (NTU) for providing a research studentship, which enabled her to conduct the work reported. Assistance provided by fellow researchers and technical staff in the School of MPE, NTU is gratefully acknowledged.

References [1] W. Bonfield, M.D. Grynpas, A.E. Tully, J. Bowman, J. Abram, Hydroxyapatite reinforced polyethylene — a mechanically compatible implant material for bone replacement, Biomaterials 2 (1981) 185 – 186. [2] M. Wang, W. Bonfield, L.L. Hench, BioglassR/high density polyethylene composite as a new soft tissue bonding material, in: J. Wilson, L.L. Hench, D. Greenspan (Eds.), Bioceramics, vol. 8, Pergamon, Oxford, 1995, pp. 383 – 388. [3] R.L. Orefice, G.P. LaTorre, J.K. West, L.L. Hench, Processing and characterization of bioactive polysulfone-BioglassR composites, in: J. Wilson, L.L. Hench, D. Greenspan (Eds.), Bioceramics, vol. 8, Pergamon, Oxford, 1995, pp. 409 – 414. [4] M. Wang, T. Kokubo, W. Bonfield, A-W glass – ceramic reinforced polyethylene composite for medical applications, in: T. Kokubo, T. Nakamura, F. Miyaji (Eds.), Bioceramics, vol. 9, Pergamon, Oxford, 1996, pp. 387 – 390. [5] R.L. Reis, I.B. Leonor, M.T. Rego, A.M. Cunha, M.H. Ferdandez, R.N. Correia, Stiff and bioactive composites based on starch, polyethylene and SiO2 – 3CaOP2O5 – MgO glasses and glass – ceramics, in: R.Z. LeGeros, J.P. LeGeros (Eds.), Bioceramics, vol. 11, World Scientific, Singapore, 1998, pp. 169 – 172. [6] M. Wang, C.Y. Yue, B. Chua, L.C. Kan, Hydroxyapatite reinforced polysulfone as a new biomaterial for tissue replacement, in: H. Ogushi, G.W. Hastings, T. Yoshikawa (Eds.), Bioceramics, vol. 12, World Scientific, Singapore, 1999, pp. 401 – 404. [7] J.B. Park, Biomaterials: An Introduction, Plenum, New York, 1979. [8] W. Bonfield, M. Wang, K.E. Tanner, Interfaces in analogue biomaterials, Acta Materialia 46 (1998) 2509 – 2518. [9] R.N. Downs, S. Vardy, K.E. Tanner, W. Bonfield, Hydroxyapatite-polyethylene composite in orbital surgery, in: W. Bonfield, G.W. Hastings, K.E. Tanner (Eds.), Bioceramics, vol. 4, Butterworth, Oxford, 1991, pp. 239 – 246. [10] R.E. Swain, M. Wang, B. Beale, W. Bonfield, HAPEXk for otologic applications, Biomedical Engineering: Applications, Basis and Communications 11 (1999) 315 – 320. [11] M. Wang, J. Weng, C.H. Goh, J. Ni, C.X. Wang, Developing tricalcium phosphate/polyhydroxybutyrate composite as a new biodegradable material for clinical applications, in: S. Giannini, A. Moroni (Eds.), Bioceramics, vol. 13, Trans Tech Publications, Zurich, 2000, pp. 741 – 744.

J. Ni, M. Wang / Materials Science and Engineering C 20 (2002) 101–109 [12] M. Wang, C.X. Wang, J. Weng, J. Ni, P.Y. Quek, Production and evaluation of biodegradable hydroxyapatite/polyhydroxybutyrate composite, Proceedings of the 10th International Conference on Biomedical Engineering, Singapore, 2000, pp. 545 – 546. [13] T. Kitsugi, T. Yamamuro, T. Nakamura, T. Kokubo, Bone bonding behavior of MgO – CaO – SiO2 – P2O3 – CaF2 glass (mother glass of AW-glass – ceramics), Journal of Biomedical Materials Research 23 (1989) 631 – 648. [14] C.J. Kirkpatrick, C. Mittermayer, Theoretical and practical aspects of testing potential biomaterials in vitro, Journal of Materials Science: Materials in Medicine 1 (1990) 9 – 13. [15] L.L. Hench, Bioceramics: from concept to clinic, Journal of the American Ceramic Society 74 (1991) 1487 – 1510. [16] T. Kokubo, H. Kushitani, S. Sakka, T. Kitsugi, T. Yamamuro, Solutions able to reproduce in vivo surface-structure changes in bioactive glass – ceramic A-W, Journal of Biomedical Materials Research 24 (1990) 721 – 734. [17] I.M. Ward, Mechanical Properties of Solid Polymers, Wiley, New York, 1985. [18] P. Li, C. Ohtsuki, T. Kokubo, K. Nakanishi, N. Soga, K. de Groot, The role of hydrated silica, titania, and alumina in inducing apatite on implants, Journal of Biomedical Materials Research 28 (1994) 7 – 15.

109

[19] J. Weng, Q. Liu, J.G.C. Wolke, X. Zhang, K. de Groot, Formation and characteristics of apatite layer on plasma-sprayed hydroxyapatite coatings in simulated body fluid, Biomaterials 18 (1997) 1027 – 1035. [20] J. Weng, M. Wang, In vitro formation of bone-like apatite on the surface of solution-cast poorly crystallized hydroxyapatite/chitin composite, Key Engineering Materials 192 – 195 (2001) 657 – 660. [21] S.N. Nazhat, R. Joseph, M. Wang, R. Smith, K.E. Tanner, W. Bonfield, Dynamic mechanical characterization of hydroxyapatite reinforced polyethylene: effect of particle size, Journal of Materials Science: Materials in Medicine 11 (2000) 621 – 628. [22] M. Canetti, M. Urso, P. Sadocco, Influence of the morphology and of the supermolecular structure on the enzymatic degradation of bacterial poly(3-hydroxybutyrate), Polymer 40 (1999) 2587 – 2594. [23] N. Yoshie, K. Nakasato, M. Fujiwara, K. Kasuya, H. Abe, Y. Doi, Y. Inoue, Effect of low molecular weight additives on enzymatic degradation of poly(h-hydroxybutyrate), Polymer 41 (2000) 3227 – 3234. [24] R. Zhang, P.X. Ma, Porous poly(L-lactic acid)/apatite composites created by biomimetic process, Journal of Biomedical Materials Research 45 (1999) 285 – 293.