Formation of Biomimetic Porous Calcium Phosphate Coatings on ...

Page 1 of 12

Materials Science and Technology (MS&T) 2007 September 16-20, 2007, Detroit, Michigan · Copyright © 2007 MS&T'07® FUNDAMENTALS AND CHARACTERIZATION: Frontiers in Materials Science: Materials for Sports and Medicine Organized by James Earthman, Srinath Viswanathan Sports, Prosthetics and Biocompatibility

Formation of Biomimetic Porous Calcium Phosphate Coatings on Surfaces of Polyethylene/Zinc Stearate Blends Jaroslaw Drelich and Kevin G. Field Department of Materials Science and Engineering Michigan Technological University Houghton, MI 49931, USA Keywords: biomaterials, biomimetic deposition, calcium phosphate coating, polyethylene

Abstract Studies were undertaken investigating improvements to the biological interaction of polymeric implant materials through their coating with an osteoinductive calcium phosphate (CaP)-type film using biomimetic deposition technology. Past research indicates that CaP coatings on implant materials increase bone growth and remodeling rates as well as enhance the stability of the bone-implant interface. This is due to the highly biocompatible nature of CaP and its chemical similarities to natural bone mineral, which enables it to form chemical bonds with bone. Additionally, adding additives to polymers, which have the proper functionality to serve as nucleation sites for CaP coating growth, can potentially improve biocompatibility and long-term stability of implant devices. Our research efforts presented in this paper concentrated on adding zinc stearate to polyethylene. Important potential benefits of using polyethylene-stearate blends having CaP coatings include: increased surface porosity that can improve mechanical stability of the implant via enhanced osseointegration, improved rates and quality of bone-implant fusion, and enhanced soft tissue wound healing through stimulation of angiogenesis. Our results show that chemical immersion of polyethylene-zinc stearate blends in supersaturated calciumphosphate solutions elicits growth of porous CaP coatings, although at a very slow rate, ~0.1 μm/day.

Introduction Polyethylene is a commonly used material for bearing surfaces in total shoulder, knee and hip arthroplasty due to its elastic properties which are similar to biological tissue found in joint capsules. Most polyethylene components used in joint replacements are equipped with a metal component such as the tibial tray in total knee arthroplasty, or the acetabular cup in total hip arthroplasty. These metal components are required to stabilize the polyethylene bearing surface and act as the interface between the bone and the implant. Developing a method by which polyethylene components can bond directly to bone would eliminate the need for the metallic containment devices. Calcium phosphate (CaP) coatings on either metallic or polymeric implant devices are expected to improve the implant’s bone tissue compatibility and accelerate the healing process for patients who have experienced orthopedic surgery [1,2]. The addition of a CaP-based ceramic to the surface of implants results in three primary benefits to the implant materials: i) improvement in their biocompatibility;

187

Page 2 of 12

ii)

increase in implant-bone mechanical stability by providing a porous coating that allows osseointegration; and iii) improvement in the rate and quality of bone growth at the implant interface. Plasma sprayed CaP has been the most common form of bioactive coating used to enhance hard tissue integration with orthopedic implants [3]. Coatings produced by plasma deposition on implants with complex geometries (screws, interbody fusion cages) have not been as successful in terms of uniform surface coverage and coating thickness [1]. The temperatures used for plasma spray deposition are well above the decomposition temperature for most polymers and therefore, make this technique less attractive in coating polymeric implants. Biomimetic processes that rely on heterogeneous nucleation of CaP from ionic solutions have been explored by many research groups in recent years as attractive and cheaper alternative technologies in coating biomaterials [4-13]. In this process, the implant of appropriate surface characteristic is immersed in either a simulated body fluid or solutions saturated or supersaturated with calcium and phosphate ions at room or body fluid temperature. Although simulated body fluids which produce hydroxyapatite (HA) coatings were studied first, the deposition of CaP films from supersaturated solutions appears to dominate in recent years. The formation of ceramic coatings is accelerated from supersaturated solutions as compared to those nucleated from simulated body fluid. Additionally, recent studies indicate a higher dissolution rate of less stable forms of CaP, such as octacalcium phosphate, as compared to HA, making them more biocompatible coatings. One of the major concerns in implanting the biomimetic coating technology to polymers is the adhesion strength between the ceramic film and polymeric substrate. Due to the hydrophobic nature of polyethylene and a lack of polar chemical functionality in its composition, a CaP coating binds poorly to polyethylene [13]. Strong polymer-CaP adhesive bonding is needed to secure the structural integrity of implant devices during implantation and biointegration. Any unsuccessful integration of an implant device during the healing process, caused for example by delamination of coating from the polymer surface, will negatively impact the stability of the fusion site. Introduction of ceramic coatings to polyethylene surfaces poses the need for oxidation or functionalization of the polymer surface to improve the adhesion at the polyethylene-CaP interface [14]. Oxidation of polyethylene surfaces can be accomplished through plasma treatment, ultra-violet radiation, glow-discharge processing, and alkaline etching [13,14]. These treatments produce oxygen-based functional groups such as hydroxyl, carboxylic acid, carbonyl, aldehyde, ether, and ester groups, which enhance the polymer-coating adhesion, increase CaP deposition rate, and lead to improvements in biocompatibility and adhesion of cells. To understand the effect of organic functionality on nucleation and growth of CaP coatings, researchers have turned to systems with well-defined and organized molecular structures of controlled functionality. It was found that the growth rate of apatite coatings on SAMs with PO4H2 and COOH groups were substantially higher than for molecular layers with CONH2, OH, and NH2 groups [12,15,16]. There was practically no growth of the apatite structures on monolayers with methyl as the end functionality. These earlier studies [15,16] were conducted with simulated body fluids, which mimic the composition of blood plasma with inorganic ions [17,18], commonly used in the formation of the HA coating on biomaterials in the past. More recently, we demonstrated that the deposition of bioactive CaP coatings on molecular monolayers can be accelerated from supersaturated calcium phosphate solutions [12]. Our results

188

Page 3 of 12

confirm earlier observations that the carboxylic acid groups attract and facilitate the formation of uniform and porous calcium carbonate coatings as compared to methyl and alcohol groups. Polymer surface functionality is difficult to control in treatments such as plasma treatment, ultra-violet radiation, glow-discharge processing, or alkaline etching. These treatments, besides being often quite expensive, can produce different functionalities on the outer surfaces of devices of a complex geometry as compared to their internal surfaces. For these reasons, we explore the use of polymeric blends, which contain functional chemicals with a tendency to migrate and stay at the surface of the polymer, enhancing its polarity. These chemicals promote a stronger binding of CaP coating to the polymer, similar to what is expected from organic functionality produced in surface treatments. In this communication, we present the results on biomimetic coating of polyethylene-zinc stearate blends with CaP films deposited from a supersaturated calcium phosphate solution. Zinc stearate is a commonly used mold releasing agent and lubricant in the processing of polymers, which, due to its hydrophobic nature, can be incorporated into a matrix of polyethylene. Possible benefits of adding zinc stearate to polyethylene include the following: a) carboxylate groups can serve as adsorption sites for calcium ions which are detrimental for nucleation and growth of CaP coating [12]; b) carboxylate groups can also enhance the strength of the polyethylene-CaP interfacial bonding [13]; c) zinc stearate provides zinc that can be beneficial in bone development processes through enhancement in osteoblast proliferation and osteoblastic cell stimulation [1921]; d) a possible reduction in friction for the bearing parts made of polyethylene-zinc stearate blends instead of polyethylene can be expected. Potential drawbacks of adding zinc stearate are losses in mechanical properties of polyethylene, but such losses will depend on the amount of chemicals blended with polyethylene. The changes in the mechanical properties of polyethylene can be avoided, or at least substantially reduced, if zinc stearate is only added to the surface region of polyethylene during, for example, extrusion of the implant device. This option, however, has not yet been studied and is not discussed in this communication.

Experimental Procedure Formulation of Polyethylene-Zinc Sterate Blends Initial samples of polyethylene-zinc stearate blends were made using a MINI-MAX extruder. However, our extruder mixed zinc stearate (ZnSt) with either high-density or lowdensity polyethylene (PE) poorly, producing samples of very inhomogeneous composition. Due to this problem, zinc stearate was dispersed manually in melted polyethylene. Specifically, a two weight percentage of zinc stearate (Merck Co.) was mixed with pellets of low-density polyethylene (Scientific Polymer Products, Inc.) in a beaker. The beaker was then heated untill the polymer pellets went into a melt. The melt was hand stirred for approximately thirty seconds to produce an homogeneous mixture. Pre-cleaned glass slides were preheated, and then the melts were transferred onto the heated slides. The melts were then sandwiched with another set of precleaned slides and allowed to cool. The samples were transferred to de-ionized water and allowed to soak for 1-5 minutes. The de-ionized water allowed for easier removal of the slides

189

Page 4 of 12

from the sample. Samples were placed in an L-C oven at 75°C for 15-20 minutes to aid in drying. Biomimetic Deposition of Porous CaP Films Before growing the CaP films, samples of polymers were pretreated with Ca(OH)2 solution. The solution was prepared by slowly dissolving Ca(OH)2 (Fisher Scientific) in deionized water in a Pyrex dish under constant magnetic stirring and heat. Ca(OH)2 was continuously added until the bulk solution became opaque. The solution was transferred to an LC-oven with an operating temperature of 75 °C. The solution was allowed to warm up before the samples were immersed in the solution. Samples were held in the vertical position by painted metal clips and allowed to remain in the solution for 20 minutes. Afterward, the treatment samples were rinsed with de-ionized water and allowed to air dry for 25 minutes. In order to deposit CaP on the surface of the samples, a supersaturated calcium phosphate solution (SCPS) was used. The solution was prepared by mixing 5.6 mM CaCl2⋅2H2O (Fisher Scientific) and 3.34 mM NaH2PO4⋅H2O (Fisher Scientific) under constant magnetic stirring in 1:1 volumetric ratio. In order to elevate the solution pH to 7.4, a Tris buffer (tris(hydroxymethyl) -aminomethane, Aldrich) was added to the solution. After preparation of the solution, the SCPS was transferred to a Pyrex dish. Samples mounted in painted metal clips were lowered into the SCPS solution and placed so that the samples were in a vertical position. Samples were immersed in the solution for 1 to14 days. The solution was refreshed after 1,3, 5, 7, 9, and 10 days. Samples were removed on their corresponding day by pipetting the solution from the Pyrex dish and then rinsed with de-ionized water. Drying of the samples took place in an L-C oven at 75°C for at least 10 minutes. Characterization of CaP Coating CaP-coated samples were imaged using the JEOL JSM-6400 scanning electron microscope with an accelerating voltage of 10 kV. CaP coated samples were sputter-coated with a thin Au layer prior to imaging to reduce charging effects. The presence of CaP coatings on the PE substrate was validated using Fourier transform infrared spectroscopy (FTIR) using a Mattson Instruments Genesis II FTIR spectrometer run in a diffuse reflection mode. 1024 scans were performed at a resolution of 4.0 cm-1. The PE-2wt%ZnSt sample undergoing the 14 day immersion and the same sample of polymer without coating were characterized by X-ray diffraction technique (XRD). A Scintag XDS 2000 Diffractometer was employed using Cu Kα radiation (λ=1.54 Ǻ) and a step size of 0.03 resulting in a scan duration of 12 hours for both samples.

Results and Discussion To avoid damaging polymers with poorly controlled chemical and/or plasma treatments, we have explored the formulation and testing of mixtures of polymers with additives that have the functionality necessary to serve as nucleation sites for the CaP coating. Mixing the polymers with additives is not new and has been practiced in non-biomedical applications of polymers for many years. For example, plasticizers are added to improve flexibility to inherently rigid polymers, which causes the processing temperature of the polymer to be reduced. Our recent efforts concentrate on adding carboxylates that can be compatible with biopolymers and have a

190

Page 5 of 12

strong affinity towards calcium. Carboxylate groups interact with calcium ions and CaP precipitates from supersaturated solutions during the biomimetic coating process. Such interactions promote nucleation and growth of a CaP coating with a strong polymer-coating adhesion at a higher rate than observed for unmodified polymers [12,13]. The results of nucleation and growth of CaP on self-assembled templates conducted in our laboratory [12] and by others [15,22] clearly corroborate such expectations. A number of trial-and-error tests were conducted in an early stage of this study. For example, from 3 wt% ZnSt, and the samples fractured under relatively weak forces. For this reason, the tests were limited to blends that contained 2 wt.% ZnSt. Mechanical properties of samples were not examined in this study.

Day 1

Day 2

Day 5

Day 7

Figure 1. Low magnification SEM images of the PE-2wt%ZnSt samples coated with CaP film. The size of each image is ~1.1 x 0.8 cm.

191

Page 6 of 12

Previous studies with metallic implant materials demonstrated that pre-calcification treatment aided in the nucleation and rate of growth of the CaP coating [11]. Also, in this study, we observed enhanced biomimetic deposition of CaP films and a more uniform structure when PE-ZnSt blends were treated with calcium hydroxide solutions at elevated temperatures. The precalcification treatment of blends was carried out at 75oC and could be even more efficient at temperatures higher than 75oC. However, low-molecular weight polyethylene (melting temperature: 117oC) used in this study, though a very convenient polymer for preparation of samples of desired shape and size, softens at >80oC causing the samples to deform during the pre-calcification treatment. We speculate here that our simple pre-calcification treatment promoted an ion exchange process in such a way that some of the zinc ions located at the polymer surface were replaced by calcium ions. Calcium carbonate serves as a nucleation site for the formation of CaP deposits whereas the growth of CaP appears to be inhibited on zinc carboxylate.

C B

A

C

B

A

Figure 2. SEM images of the CaP coating produced on the PE-2wt.% ZnSt sample from a supersaturated calcium phosphate solution. The upper picture (~56 x 43 μm) shows the sample area on which three different thicknesses of CaP coating can be identified. This variation in coating thickness indicates that the CaP structure evolves either gradually in x and y directions through formation of one film over another one or the film expands non-uniformly in x, y and z direction. The pictures at the bottom (~11 x 8 μm) marked A to C are the areas of samples holding coating of different thickness, with image A of the thickest CaP film and C of the thinnest CaP film.

192

Page 7 of 12

In our biomimetic deposition process, the PE-ZnSt samples, after pre-calcification treatment, were immersed in an aqueous solution supersaturated with calcium and phosphate ions at room temperature for one to fourteen days. The CaP coatings were imaged with scanning electron microscopy (Figures 1 and 2). The brighter regions of the sample surface in Figure 1 are those coated with a well-developed rose-like CaP structure (shown in Figure 2) whereas darker regions are those coated with only very thin amorphous CaP “precursor film.” The thickness of the rose-like coatings varied from a fraction of a micron to a few microns. The precursor film was thinner than 40-50 nm; the exact thickness could not be determined due to the limited resolution of the SEM instrument used in this study.

Figure 3. Sample of PE-2wt%ZnSt with non-uniformly growing CaP coating after 3 days of immersion time. The size of the image is ~28 x 9 μm.

Although the mechanism of the CaP film growth was not explored in detail within this study, the SEM images of the coating captured at various stages of the CaP deposition process suggest that: i) a solid amorphous film of CaP is formed first, ii) small “crystals” evolve from the amorphous film, and iii) “crystals” grow in x, y and z direction forming a porous structure of CaP. Figure 2 shows the SEM image of the PE-2wt%ZnSt sample with non-uniform CaP coating. Each of the three marked regions represents coatings with a different degree of development and thickness. The first layer (picture A) is the least developed CaP film, called a “precursor film” in this communication. This film is predominantly of amorphous character with some crystals growing out/on this film. Pictures B and C show well developed rose-like CaP structures that cover the amorphous base. The coating shown in picture C is much thicker than the coating in picture B. Further, the pictures in Figures 1 and 2 clearly show that the CaP coatings grew both laterally and vertically. These expansions, however, were non-uniform as can be deduced from the images in Figures 1 to 2. The thickness of coating varied significantly but only in the early stages of the biomimetic deposition process as shown by the cross sections of the coatings in Figure 3. This variation in CaP coating thickness usually reduced with time. Figure 4 shows the SEM images of the CaP coating produced in two weeks of the biomimetic deposition process.

193

Page 8 of 12

The coating still remained porous, and the CaP layer became compact and uniform after two weeks of soaking.

Day 14

Figure 4. SEM images of the CaP coating after two weeks of biomimetic deposition. The size of images is from top to bottom: ~38 x 29 μm, 11 x 9 μm, and 356 x 14 μm.

194

Page 9 of 12

As expected, coverage of the sample surface with the CaP coating increased with a prolonged immersion time (Figure 1). This process, however, was slower than observed in our previous studies with metallic substrata [11]. The complete surface coverage of the PE2wt%ZnSt samples with a relatively uniform porous CaP film required at least five days as compared to three days for metallic surfaces. Also, the thickness of the film grew at a much slower rate. It took 7 to 14 days to grow the coating with a thickness of ~1 μm. Films with a thickness of 30-40 microns were deposited on titanium alloy samples in three days [11]. Cracks that appeared in the CaP coating after its drying indicate the brittle nature of the CaP coating (Figure 4). Delamination of the coating is also visible in Figure 4, which indicates a weak adhesion between the CaP coating and polymer. Coating delamination was observed for selected samples and their sections, suggesting (again) the heterogeneous nature of the PE2wt%ZnSt samples, with surface sites of stronger and weaker affinity to CaP. Compositional characterization of the CaP coating was carried out using FTIR and XRD techniques. Figure 5 shows the FTIR spectra of four samples that were soaked in a supersaturated solution of calcium and phosphate ions for 1, 3, 7 and 14 days. Carbonate absorption bands are located at 872 cm-1 (ν2), 1416 cm-1 (ν3) and 1462 cm-1 (ν3) whereas phosphate absorption peaks are at 564 cm-1 (ν4), 961 cm-1 (ν1) and 1038 cm-1 (ν3). These peaks are weak and broad for the sample soaked for one day, indicative of a small amount of deposited CaP and/or its amorphous character. As shown in Figure 5, both intensity and resolution of the absorption peaks increase for samples with an increasing amount of deposited CaP and presumably increasing crystallinity of the coating. 1 1.1 1.2 Day 14

1.3

Day 7

1.5

Day 3

Absorbance

1.4

1.6

CO3 2-

PO4 3-

CO3

2-

Day 1

1.7 1.8 1.9 2 1500

1400

1300

1200

1100

1000

900

Wavenumber, cm -1

Figure 5. FTIR spectra for samples after different time of coating.

195

800

700

600

500

Page 10 of 12

XRD spectra for uncoated and coated polymers are shown in Figure 6. The CaP film with a thickness of 1-1.5 μm produced only small CaP peaks at 26, 33 and 48 degrees (2θ). These peak locations are identical to those reported in our previous studies for CaP coatings on titanium alloy samples [11]. The broad character of the peaks is also indicative of the amorphous CaP solids.

2750 250

2500

200 CPS

2250 2000

150 100

CPS

1750

50

1500

0

PE

1250

5

10 15 20 25 30 35 40 45 50 55 60 65 70 Deg.

PE with coating

1000 750 500 250 0 5

10

15

20

25

30

35

40

45

50

55

60

65

70

Deg.

Figure 6. XRD spectra for PE-2wt%ZnSt sample without and with CaP coating.

Summary We explored a new, simple, and inexpensive approach to fabricate biopolymers with enhanced nucleation and adhesion to a CaP coating. To avoid damaging implant materials with poorly controlled chemical and/or plasma treatments, we formulated and tested mixtures of polyethylene with zinc stearate additive, which has the functionality necessary to nucleate formation of CaP structures on the polymer surface. To enhance growth of CaP coatings on surfaces of the formulated polyethylene-zinc stereate blends, the blends were first soaked in Ca(OH)2 solutions at 75oC. Porous, compact and uniform CaP coatings were formed biomimetically from supersaturated solutions of calcium and phosphate ions on pre-calcificated blends after 10-14 days of soaking. The CaP coatings grew at a rate of approx. 0.1 μm/day and preceded through the following steps: i) deposition of a thin solid amorphous film, ii) nucleation and growth of nano-sized “crystals” on the amorphous film, and iii) lateral and vertical expansion of porous CaP coating.

196

Page 11 of 12

Acknowledgments KGF would like to express appreciation for the financial support received from the Department of Materials Science and Engineering of MTU through the McArthur Research Internship program. The authors would like to thank Edward Laitila for XRD analysis.

References 1. A.L. Oliveira and R.L. Reis, Bonelike apatite coatings nucleated on biodegradable polymers as a way to induce bone mineralization: current developments and future trends, In: Bone Tissue Engineering, J.O. Hollinger, T.A. Einhorn, B.A. Doll, and Ch. Sfeir (Eds), CRC Press, Boca Raton 2005, pp. 205-221. 2. R.J. Narayan, Hydroxyapatite/diamondlike carbon nanocomposites: a novel surface modification to extend orthopedic prosthesis lifetimes, J. Mater. Res. 20(9), 22882295(2005). 3. H. Liang, B. Shi, A. Fairchild, and T. Cala, Applications of plasma coatings in artificial joints: an overview, Vacuum 73, 317-326(2004). 4. S. Jalota, S.B. Bhaduri, and A.C. Tas, Effect of carbonate content and buffer type on calcium phosphate formation in SBF solutions, J. Mater. Sci: Mater. Med. 17, 697707(2006). 5. S.H. Barbanti, A.R. Santos, C.A. Zavaglia, and E.A. Duek, Porous and dense poly(l-lactic acid) and poly(D,L-lactic acid-co-glycolic acid) scaffolds: in vitro degradation in culture medium and osteoblasts culture, J. Mater. Sci: Mater. Med. 15, 1315-1321 (2004). 6. Q. Liu, J. Ding, F.K. Mante, S.L. Wunder, and G.R. Baran, The role of surface functional groups in calcium phosphate nucleation on titanium foil: a self-assembled monolayer technique, Biomaterials 23, 3103 –3111 (2002). 7. M. Tanahashi, T. Yao, T. Kokubo, M. Minoda, T. Miyamota, T. Nakamura, and T. Yamamuro, Apatite coated on organic polymers by biomimetic process: improvement of its adhesion to substrate by NaOH treatment, J. Biomater. B5(4), 339-347(1994). 8. R. Rohanizade, R.Z. LeGeros, M. Harsono, and A. Bendavid, Adherent apatite coating on titanium substrate using chemical deposition, J. Biomed. Mater. Res. A72(4), 428-438 (2005). 9. G.J. Liu, F. Miyaji, T. Kokubo, H. Takadama, T. Nakamura, and A. Murakami, Apatiteorganic polymer composites prepared by a biomimetic process: improvement of the adhesion of the apatite layer to the substrate by ultraviolet irradiation, J. Mater. Sci. Mater. Med. 9, 285-290(1998). 10. M. Tanahashi, T. Yao, T. Kokubo, M. Minoda, T. Miyamota, T. Nakamura, and T. Yamamuro, Apatite coated on polymers by biomimetic process: improvement in its adhesion to substrate by glow-discharge treatment, J. Biomed. Mater. Res. 29, 349357(1995). 11. K.C. Baker, M.A. Anderson, S.A. Oehlke, A.I. Astashkina, D.C. Haikio, J. Drelich, and S.W. Donahue, Growth, characterization and biocompatibility of bone-like calcium phosphate layers biomimetically deposited on metallic substrata, Mater. Sci. Eng. C26, 1351-1360(2006).

197

Page 12 of 12

12. K.C. Baker and J. Drelich, Biomimetic formation of nano-crystalline calcium-phosphate films on self-assembled monolayers of alkanethiols: effects of end functionality and deposition methodology, in Functional Fillers and Nanoscale Minerals, J.J. Kellar (Ed.), Society for Mining, Metallurgy, and Exploration, Inc., Littleton, CO, pp.199-216(2006). 13. K.C. Baker, J. Drelich, I. Miskioglu, R. Israel, and H.N. Herkowitz, Effect of polyethylene pretreatments on the biomimetic deposition and adhesion of calcium phosphate films, Acta Biomater. 3, 391-401(2007). 14. I. Pashkuleva and R.L. Reis, Surface activation and modification – a way for improving the biocompatibility of degradable biomaterials, In: Bone Tissue Engineering, J.O. Hollinger, T.A. Einhorn, B.A. Doll, and Ch. Sfeir (Eds), CRC Press, Boca Raton 2005, pp. 429-454. 15. M. Tanahashi and T. Matsuda, Surface functional group dependence on apatite formation on self-assembled monolayers in a simulated body fluid, J. Biomed. Mater. Res. 34, 305315(1997). 16. K. Sato, Y. Kumagai, and J. Tanaka, Apatite formation on organic monolayers in simulated body environment, J. Biomed. Mater. Res. 50, 16-20 (2000). 17. T. Kokubo, M. Minoda, T. Miyamoto, T. Nakamura, and T. Yamamuro, Apatite coating on organic polymers by a biomimetic process, J. Am. Cer. Soc. 77, 2805-2808 (1994). 18. X. Lu and Y. Leng, Theoretical analysis of calcium phosphate precipitation in simulated body fluid, Biomaterials 26, 1097-1108(2005). 19. M. Ikeuchi, A. Ito, Y. Dohi, H. Ohgushi, H. Shimaoka, K. Yonemasu, T. Tateishi, Osteogenic differentiation of cultured rat and human bone marrow cells on the surface of zinc-releasing calcium phosphate ceramics, J. Biomed. Mater. Res. A 67, 11151122(2003). 20. A. Ito, K. Ojima, H. Naito, N. Ichinose, and T. Tateishi, Preparation, solubility, and cytocompatibility of zinc-releasing calcium phosphate ceramics, J. Biomed Mater. Res. 50, 178-183(2000). 21. M. Yamaguchi, H. Oishi, Y. Suketa, Stimulatory effect of zinc on bone formation in tissue culture, Biochem Pharmacol 36, 4007-4012(1987). 22. B.J. Tarasevich, C.C. Chusuei, and D.L. Allara, Nucleation and growth of calcium phosphate from physiological solutions onto self-assembled templates by a solutionformed nucleus mechanism, J. Phys. Chem. B107, 10367-10377(2003).

198