Bioactive Zirconia Composites

Bioactive Zirconia Composites Akemi A. Nogiwa-Valdeza, Dora A. Cortés-Hernández, José M. Almanza-Robles, Alejandra Chávez-Valdez Centro de Investigación y Estudios Avanzados IPN Unidad Saltillo Carretera Saltillo-Monterrey km 13 Apdo. Postal 663 C.P. 25000, Saltillo, Coahuila, México

Keywords: Zr02-Alzo3composites, bioactivity, biomimetic process, simulated body fluid.

Abstract. Zirconia-alumina composites with additions of a CaO-Si02 glass are prepared by uniaxial pressing and sintering. In order to promote bioactivity, the composites are biomimetically treated. The effect of immersion time in simulated body fluids (SBF) and that of the presence of a wollastonite powder bed, as a calcium ion provider, on the apatite forming ability are investigated. The influence of replacing the simulated body fluids each 7-day-period for a more concentrated solution is also studied. A bonelike apatite layer is observed after 21 days of immersion when the SBF is renewed, whether the bed of wollastonite powder is present or not. However, a thicker layer is formed by using wollastonite and the agglomerates of the apatite layer are finer on the composites containing CaO-Si02 glass. Introduction

Zirconia and alumina ceramics are used as biomaterials for orthopedic and dental applications due to their appropriate mechanical and tribological properties. A disadvantage of these materials is that they do not exhibit a strong chemical bond with living tissue when implanted [ l , 21. On the other side, bioactive materials, such as wollastonite and BioglassB, bond chemically with bone by forming a biologically active bonelike apatite layer between the implant and the living tissue [3]. However, these materials have poor mechanical properties that limit their use for wider applications [4,51. It has been found that the addition of 20 vol % of alumina to a stabilized zirconia ceramic improves its flexura1 strength and fracture toughness. Moreover, it prevents the low-temperature degradation of zirconia, which is particularly harmful for materials that must be autoclave-sterilized [6]. On the other hand, the presence of Si-OH groups on the surface of a bioceramic induces apatite nucleation when immersed in simulated body fluids [7, S]. Then, it is expected that, by the addition of small amounts of a CaO-Si02 glass, silicon remains distributed on the surface and acts later as an apatite nucleation site. On the basis of these reported findings, the composition of the materials under study is determined. Biomimetic processes have been widely studied for growing a bonelike apatite layer on different substrates by immersion in simulated body fluids. In the case of bioactive CaO and Si02-based glasses, calcium ions released into the solution increase the ionic activity product of the apatite in the fluid and the Si-OH groups formed on the surface provide sites for apatite nucleation. After the apatite nuclei are formed, they grow spontaneously since the solution is already saturated with respect to apatite [9, 101. This research leaded to the study of inei-t biomaterials biomimetically treated in the presence of a CaO, SiO2-based glass [ l l , 121. In this case, calcium and silicate ions dissolved from the bioactive glass induced nucleation of apatite not only on the glass but also on the surface of materials placed nearby. In the present work, the effect of the experimental conditions of a biomimetic process on the growth of an apatite layer on zirconia-alumina composites with additions of a CaO-Si02 glass was studied.

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any rneans without the written perrnission of the publisher: Trans Tech Publications Ltd, Switzerland, www.tti~.net.(ID: 148.247.22.104-1 1105106.15:31:48)

194

Advanced Structural Materials II

Experimental Procedure Preparation of CaO-Si02 glass. A CaO-Si02 bioactive glass with a Caísi atomic ratio of 1 was prepared by mixing 37.51 wt % Si02 and 62.49 wt % C a c o 3 in a polyethylene bottle with alumina balls for 1 hour. The mixture was melted with an oxygen-butane torch in alumina crucibles. The melt was poured onto distilled water ice cubes to avoid surface crystallization. The glass was crushed in an alumina ball mil1 and sieved through a #400 mesh. Substrate preparation. Two different Partially Stabilized Zirconia (PSZ)-A1203 composites were prepared from magnesium oxide stabilized zirconia (Stanford Materials Corp., USA) and alumina (Sasol, USA). The obtained CaO-Si02 glass was added according to compositions shown in Table 1. Particle mean size of the zirconia, alumina and bioactive glass used for the substrate preparation was 0.5, 0.4 and 11.3 pm, respectively. The powder mixtures shown in Table 1 were ball-milled with alumina balls for 1 hour in acetone, except for alumina which was mixed with polyvinyl alcohol in water. The slurries were then dried and disk-shaped by uniaxial pressing at 100 MPa for 15 seconds. Composition Dispersant Additive PSZ - 20 vol % A Acetone none (PSZ-20vol%A)5wt%G Acetone none A: alumina, PSZ: partially stabilized zirconia, G: CaO-Si02 glass. Table 1. Mixtures utilized for the fabrication of samples by dye-pressing. The zirconia composites were sintered in air at 1550 "C for 2 hours. The heating rate used was 10 "C/min up to 1000 "C and 5 "C/min to the final temperature, and the cooling rate was 10 "C/min. Preparation of simulated body fluids. Two simulated body fluids, one with ion concentrations nearly equal to those of human blood plasma (SBF) and other 40 % more concentrated (1.4 SBF), were used (Table 2). The solutions were prepared dissolving reagent grade sodium chloride (NaCl), sodium hydrogen carbonate (NaHCO3), potassium chloride (KCl), dipotassium hydrogen phosphate (K2HP04.3H20), magnesium chloride hexahydrate (MgC12.6H20),calcium chloride dihydrate (CaC12.2H20)and sodium sulfate (Na2S04)in deionized water and buffered with tri(hydroxymethy1)-aminomethane and 1N HCl at 36.5 "C. Concentration (mM) SBF 1.4 SBF Blood plasma

5 2.5 148.8 4.2 142 1.5 7 3.5 2.1 208.32 5.88 198.9 27.0 1.5 103.0 142 5 2.5 Table 2. Ion concentrations of simulated body fluids

1 1.4 1

0.5 0.7 0.5

Immersion of the materials in simulated body fluids. The substrates were washed with acetone and deionized water, and then immersed in 150 mL of SBF or 1.4 SBF at 36.5 "C in an incubator. The immersion was performed following two different routes. The first one consisted in the immersion of the substrates in SBF for 7 days with or without the presence of a bed of wollastonite powder. After 7 days, SBF was replaced for 1.4 SBF and the wollastonite bed was removed in the correspondent cases. The solution was replaced for fresh 1.4 SBF after one week. The procedure described (re-immersion method) is shown in Fig. 1. The second route was a single immersion

Materials Science Forum Vol. 509

0-7 days lrnrnersion in SBF with wollastonite bed

lrnrnersion in SBF without wollastonite bed

195

8-14 days

15-21 days

SBF for 1 4 SBF and Of the wollastonite bed

Renewal of 1.4 SBF

'1

1

iSji Substrate Figure 1. Schematic representation of the re-immersion method. periods, the substrates were removed from the bottles, gently washed with deionized water and dried at room temperature.

Characterization of the Apatite layer. The surface of the biomimetically treated substrates was analyzed by X-ray diffraction (XRD) (X'Pert, Philips, Holland), then the substrates were carbon coated and observed under a scanning electron microscope (SEM) with an energy dispersive spectroscopy attachment (EDS) (JSM 6300, Jeol, Japan). Results and Discussion Figure 2 shows XRD pattems of the surfaces of composites studied. It can be observed that new peaks appear in the patterns of the materials treated for 21 days with weekly SBF replacement (reimmersion method). In those cases, for both composites, the new peaks that appeared are al1 ascribed to hydroxyapatite (HA). The effect of the presence of a wollastonite powder bed on the HA layer formation can be related to the intensity of the substrate peaks. When the wollastonite powder bed is present zirconia and alumina reflections are less intense than when it is absent. This can be easily noticed for zirconia (1 11), (111) and (020) reflections at 28, 31 and 34" 28 and the alumina (1 13) reflection at 43" 28. The broadening of the hydroxyapatite peaks in al1 cases may indicate that it has a defective structure. Composites with CaO-Si02 glass in their composition showed apparently less HA formation. The formation of a hydroxyapatite layer by the biomimetic process proposed was not observed after 7 days of immersion. Figure 3 shows SEM photographs of the HA layers formed on the surfaces of the zirconia composites immersed in the solutions for 21 days. A homogeneous HA layer was formed on both materials when treated following the re-immersion method whether the wollastonite bed was present or not. The size of the agglomerates was smaller for the layer formed on the glass containing composites (Figure 3b). This may indicate an increase of nucleation sites due to the presence of the bioactive glass. The HA formation phenomenon occurs due to the high saturation of the 1.4 SBF with respect to hydroxyapatite. It is possible that the formation of HA

Advanced Structural Materials II

C 20

l

25

35 40 2 8 degree

30

45

20

.

,

25

.

,

.

,

.

,

35 40 2 0 degree

30

.

,

45

Figure 2: XRD patterns of (a) re-immersion in SBF with wollastonite bed, (b) re-immersion in SBF, and (c) untreated zirconia composites.

(a)

Fig. 3 . SEM photographs and EDS spectra of (a) PSZ-20%A, and (b) (PSZ-20%A) 5% G composites soaked for 21 days in simulated body fluids by using the re-immersion method. Scale bar: 1 0 p .

Materials Science Forum Vol. 509

197

nuclei occurs during the first stage of the biomimetic process, when the material is immersed in SBF. Afterwards, during the second stage, the HA saturated solution 1.4 SBF provides enough calcium and phosphate ions to promote the spontaneous growing of nuclei and subsequent formation of a homogeneous hydroxyapatite layer. The single-immersed composites did not show any ceramic material formation on their surface after 21 days. The EDS spectra shown in Figure 3 correspond to the analysis within an area of approximately 1 mm2 of the ceramic coating of the substrates. These analyses prove that the ceramic material formed on the surface of both composites is a calcium and phosphorus compound which has demonstrated to be hydroxyapatite by X-ray diffraction. However, the CaíP ratio of the material deposited is not exactly the same as in pure hydroxyapatite (Ca/P=1.67), but it is still within the range of bone apatites (1.2-1.6) (Table 3). The peaks of sodium and calcium in the spectrum might be present due to a slight precipitation of salts from the SBF. With wollastonite bed

Without wollastonite bed

(PSZ-20%A)5%G 1.5 1 1.69 Table 3. CaíP ratio of the material deposited on substrates immersed for 21 days. Figure 4 shows the behavior of pH of the solutions during the first 7 days of immersion. It can be observed that pH increases with time, reaching higher values when the wollastonite powder bed is present due to dissolution of calcium ions into the solution. This behavior could be explained supposing that during this period of time an interaction among the ions in the solution and those released from the substrate occurs generating this change in pH [12]. It is possible that the formation of hydroxyapatite nuclei on the composites could lead to this increase in pH.

With wollastonite Without wollastonite bed

O

50

1-

100

150

200

250

300

Time (hours)

--

P S Z - 2 0 % ~ +( P S Z - ~ O % A ) ~ % G ~ -

--

Figure 4. Behavior of pH during the first stage of the immersion processes (7 days).

Conclusions

A homogeneous hydroxyapatite coating was obtained by a biomimetic process after immersion of the zirconia composites studied for 21 days when the SBF was replaced for 1.4 SBF each 7-daysperiod (re-immersion process). The morphology of the apatite layer closely resembled the biomimetic coatings of the existing bioactive systems. The thickness of the coating was increased when a wollastonite powder bed was placed during the first stage of the re-immersion process. The glass containing composites showed finer HA agglomerates.

Advanced Structural Materials II

References

[ l ] C. Piconi and G. Maccauro: Biomaterials Vol. 20 (1999), p. 1. [2] L. Hench and J. Wilson: An lntroduction to Bioceramics (World Scientific Publishing Co., Singapore, 1993) [3] L. Hench: J. Am. Ceram. Soc. Vol. 81 (1998), p. 1705. [4] T. Kokubo: Bioactive Glass Ceramics: Properties and Applications, (Proc. of Biointeractions 90, UK, 1990) [5] T. Kokubo: J. Ceram. Soc. Jpn. Vol. 10 (1991), p. 965. [6] D.J. Kim, M.H. Lee, D.Y. Lee and J.S. Han: J. of Biomedical Mater. Res. Vol. 35 (2000), p.438.

i

[7] M. Uchida, H.M. Kim, T. Kokubo, M. Nawa, T. Asano, K. Tanaka and T. Nakamura: J. Biomedical Mater. Res. Vol. 60 (2002), p. 277. [S] P. Li, C. Ohtsuki, T. Kokubo, K. Nakanishi, N. Soga and K. de Groot: J. Biomedical Mater. Res. Vol. 85 (1994), p. 7. [9] Y. Ebisawa, T. Kokubo, K. Ohura and T. Yamamuro: J. Mater. Sci.: Mater. in Medicine Vol. 1 (1990), p. 239. [lo] T. Kokubo, H. Kushitani, C. Ohtuski, S. Sakka and T. Yamamuro: J. Mater. Sci.: Mater. Medicine Vol. 4 (1993), p. 1. [ l l ] K. Hata, T. Kokubo, T. Nakamura and T. Yamamuro: J. Amer. Ceram. Soc. 78 (1995), p. 1049. [12] Y. Abe, T. Kokubo and T. Yamamuro: J. Mater. Sci.: Mater. in Medicine Vol. 1 (1990), p. 233.