Z Kardiol 90: Suppl 3, III/86 III/91 (2001) Steinkopff Verlag 2001

T. Kokubo H.-M. Kim M. Kawashita T. Nakamura

Process of calcification on artificial materials

Summary CaO, SiO2-based glasses form the bonelike apatite on their surfaces in an acellular simulated body uid (SBF) with ion concentrations nearly equal to those of the human blood plasma. The apatite formation of the former

Prof. Dr. Tadashi Kokubo ( ) Dr. H.-M. Kim Dr. M. Kawashita Department of Material Chemistry Graduate School of Engineering Kyoto University Sakyo-ku, Kyoto 606-8501, Japan E-mail: kokubo@sung7.kuic.kyoto-u.ac.jp T. Nakamura, MD Department of Orthopaedic Surgery Graduate School of Medicine Kyoto University Sakyo-ku, Kyoto 606-8507, Japan

glasses is attributed to the catalytic effect of the Si-OH groups, which are formed on their surfaces in SBF, for the apatite nucleation. The gels of SiO2, TiO2, ZrO2, Ta2O5, and Nb2O5 formed the apatite on their surfaces in SBF, but Al2O3 gel did not. This indicates that the Ti-OH, Zr-OH, Ta-OH, and Nb-OH groups besides the Si-OH groups are also effective for the apatite nucleation, but Al-OH groups are not effective. Apatite formation on self-assembled monolayer terminated with various functional groups in SBF showed that COOH and H2PO4 groups are also effective for the apatite nucleation. All these groups are negatively charged around pH 7.40. Their apatite nucleating ability is varied with their arrangements. Among the Ti-OH groups, those in anatase structure are most

effective for the apatite nucleation. Transmission electron microscope attached with energy dispersive Xray spectrometer showed that these functional groups induce the apatite nucleation not directly, but through formation of their calcium compounds and subsequent formation of amorphous calcium phosphate with low Ca/P atomic ratios. Key words Apatite nucleation functional groups Si-OH group Ti-OH group amorphous calcium phosphate Abbreviations SBF = Simulated body uid; TF-XRD = Thin-lm X-ray diffraction; SEM = Scanning electron microscope; TEM-EDX = Transmission electron microscope attached with energy dispersive X-ray spectrometer.

Introduction
Calcication on the vascular wall is considered to be similar to deposition of the apatite in the bone. In the eld of bone-repairing materials, it has been revealed that some ceramics, such as Bioglass (1), sintered hydroxyapatite (2) and glass-ceramic A-W (3), spontaneously bond to the bone in the living body. The key issue in this bonding of the ceramics to the bone is the formation of bonelike apatite layer on their surfaces in the living body (4). Once the bonelike apatite layer is formed on their surfaces,

osteoblast is activated (5) to preferentially proliferate and differentiate on this apatite layer (6). Consequently, the neighboring bone can come into direct contact with the apatite layer on the surface of the ceramics. When this occurs, a tight chemical bond is formed between the bone apatite and the surface apatite in order to reduce the interface energy between them (7). In order to develop new kinds of bone-bonding materials, it is important to reveal the kind of materials and how they form the bonelike apatite layer on their surfaces in the living body. Fortunately, the formation of the bonelike apatite layer on the bone-bonding ceramics in

T. Kokubo et al. Calcification on material

III/87

the living body can be reproduced in an acellular SBF with ion concentrations (Na+ 142.0, K+ 5.0, Mg2+ 1.5, Ca2+ 2.5, Mg2+ 1.5, Cl 148.8, HCO3 4.2, HPO42 1.0, SO42 0.5 mM, and pH 7.40) nearly equal to those (Na+ 142.0, K+ 5.0, Mg2+ 1.5, Ca2+ 2.5, Mg2+ 1.5, Cl 103.0, HCO3 27.0, HPO42 1.0, SO42 0.5 mM, and pH 7.20 7.40) of the human blood plasma (8). The problems described above can be investigated in SBF. In the following, the results of the studies in SBF on the problems described above will be reviewed. These fundamental ndings might provide some valuable information on the calcification of the vascular wall.

What kind of materials form the apatite

In order to investigate what kind of materials form the bonelike apatite layer on their surfaces in the living body, various kinds of glasses with different compositions in the simple ternary system CaO-SiO2-P2O5 were prepared by a conventional melt-quenching technique. They were soaked in SBF at 36.5 C for various periods. Apatite formation on their surfaces was examined by TF-XRD and SEM (9). The bonelike apatite layer was formed on the surfaces of CaO, SiO2-based glasses within 28 days in SBF, but not on those of CaO, P2O5-based glasses, contrary to the conventional expectation, as shown in Fig. 1. The former glasses released mainly the calcium ion into SBF, whereas the latter glasses released mainly the phosphate ion. Both the released ions increased almost equally the ionic activity product of the apatite in SBF. Despite this, only the CaO, SiO2-based glasses formed the apatite on their surfaces. This indicates that their surfaces provided sites effective for the apatite nucleation. When the CaO, SiO2based glasses are soaked in SBF, they release the calcium ions from their surfaces via exchange with H3O+ ion in SBF, and the silica networks of the glasses are hydrated. As a result, a lot of Si-OH groups are formed on their surfaces. These Si-OH groups might be responsible for the apatite nucleation.
Fig. 1 Apatite formation on the surfaces of CaO-SiO2-P2O5 glasses in SBF within 28 days.

Functional groups effective for apatite nucleation

In order to confirm the catalytic effect of the Si-OH groups for the apatite nucleation, a pure silica gel was prepared by hydrolysis and polycondensation of tetraethoxysilane, and soaked in SBF at 36.5 C. Apatite formation on its surface was examined by TF-XRD and SEM. The silica gel formed the apatite on its surface in SBF within 14 days, as shown in Fig. 2 (10). This indicates that Si-OH groups abundant on the surface of the gel are effective for the apatite nucleation.

In order to investigate other kinds of functional groups effective for the apatite nucleation, various kinds of metal oxide gels were prepared by hydrolysis and polycondensation of metal alkoxides, and soaked in SBF. Gels of TiO2 (11), ZrO2 (12), Ta2O5 (13), and Nb2O5 (14) besides the SiO2 gel also formed the apatite on their surfaces in SBF within 14 days, as shown in Fig. 2, but Al2O3 gel did not (11). This indicates that Ti-OH, Zr-OH, Ta-OH, and Nb-OH groups abundant on their surfaces of the gels are also effective for the apatite nucleation, but Al-OH group is not effective. All the functional groups effective for the apatite nucleation described above are negatively charged around pH 7.40, but the Al-OH group is positively charged at the same pH. It should be noted here, however, that all the Si-OH groups are not equally effective for the apatite nucleation. A silica gel prepared by hydrolysis and polycondensation of tetraethoxysilane in an aqueous solution containing polyethylene glycol formed the apatite on its surface in SBF within 14 days, but those prepared in pure water and in aqueous solution containing polyacrylic acid did not (15). This indicates that the apatite nucleating ability of the Si-OH group is varied with the arrangement of the Si-OH groups. Even the latter silica gels formed the apatite on their surfaces in SBF added with the excessive Ca2+ ion (16). Titania gels prepared by hydrolysis and polycondensation of tetraisopropyl titanate in an aqueous solution containing diethanolamine take an amorphous, anatase, and rutile structure at 500, 600, and 800 C, respectively. Among them, the gel in amorphous structure did not form the apatite on its surface within 14 days, whereas that in anatase structure formed a lot of the apatite and that in rutile structure formed only a small number of the apatite (17), as shown in Fig. 3. This indicates that Ti-OH

III/88

Zeitschrift fr Kardiologie, Band 90, Supplement 3 (2001) Steinkopff Verlag 2001

Fig. 2 SEM photographs of apatites formed on gels of SiO2, TiO2, ZrO2, Ta2O5, and Nb2O5 in SBF within 14 days.

Fig. 3 SEM photographs of apatites formed on titania gels in amorphous, anatase, and rutile structures in SBF within 14 days.

groups in anatase structure are most effective for the apatite nucleation, but those in amorphous structure are not effective. Tanahashi et al. prepared self-assembled monolayers terminated with various functional groups on a gold lm deposited on a slide glass and soaked in SBF (18). Those terminated with COOH and PO4H2 formed the apatite on their surfaces, but those terminated with CONH2, OH, NH2, and CH3 groups did not. The former functional groups form negatively charged COO and PO42 groups in an aqueous solution. Not only functional groups combined with a solid substrate but also those combined with ions dissolved in an aqueous solution can induce the apatite nucleation when they sit on a solid substrate. For example, when a solid substrate was faced to a CaO, SiO2-based glass (4.6MgO, 44.7CaO, 34.0SiO2, 16.2P2O5, 0.5CaF2 mass%) in SBF, a lot of apatite nuclei were formed on the solid substrate, as shown in Fig. 4, since silicate ions terminated with Si-OH groups were dissolved from the glass and adsorbed on the surface of the solid to induce the apatite nucleation in situ (19).

Fig. 4 Atomic force microscopic picture of apatite nuclei formed on a poly(ethylene terephthalate) faced to a CaO, SiO2-based glass in SBF.

Mechanism of apatite nucleation

In order to investigate mechanism of apatite nucleation induced by Si-OH groups, collodion lms supported by titanium grids were faced to the CaO, SiO2-based glass

T. Kokubo et al. Calcification on material Fig. 5 TEM-EDX pictures of a collodion film supported by a Ti grid faced to a CaO, SiO2-based glass in SBF for 12 h (a) and 2 days (b).

III/89

taking a distance of 0.5 mm in SBF at 36.5 C. They were observed under TEM-EDX after soaking in SBF for various periods (20). Silicate ions dissolved from the glass were attached on the surface of the collodion lm within 6 h, and combined with calcium ions in SBF to form an amorphous calcium silicate within 12 h, as shown in Fig. 5 (a). These calcium ions combined with the phosphate ions in SBF to form an amorphous calcium phosphate with Ca/P atomic ratio of 1.53 within 2 days, as shown in Fig. 5 (b). This amorphous calcium phosphate transformed into a crystalline apatite with Ca/P ratio of 1.64, which is nearly equal to that of the bone apatite, within 4 days. This apatite contained small amounts of Na+, Mg2+, and Cl ions similar to the bone apatite.

In order to investigate the mechanism of apatite nucleation induced by Ti-OH groups, titanium grids were previously subjected to a 5.0M-NaOH treatment at 60 C for 24 h and, subsequently, to a heat treatment at 600 C for 1 h to form an amorphous sodium titanate on their surfaces (21). They were soaked in SBF at 36.5 C and observed under TEM-EDX after soaking for various periods. The titanium grids released the Na+ ions from the surface sodium titanate layer via exchange with H3O+ ions in SBF to form Ti-OH groups on their surfaces within 30 min. The Ti-OH groups combined with the calcium ions in SBF to form an amorphous calcium titanate immediately after they were formed, as shown in Fig. 6 (a) (22). These calcium ions combined with the phos-

III/90

Zeitschrift fr Kardiologie, Band 90, Supplement 3 (2001) Steinkopff Verlag 2001

Fig. 6 TEM-EDX pictures of a Ti grid which was subjected to NaOH and heat treatments, and then soaked in SBF for 30 min (a) and 3 days (b).

phate ions in SBF to form an amorphous calcium phosphate with Ca/P ratio of 1.40 within 1.5 days. This amorphous calcium phosphate transformed into a crystalline apatite with Ca/P ratio of 1.62 within 3 days, as shown in Fig. 6 (b). Its Ca/P ratio increased up to 1.65, which is nearly equal to that of the bone mineral, within 5 days. The resultant apatite contained small amounts of Mg2+ ion, similar to the bone mineral.

These results indicate that both the Si-OH and Ti-OH groups induce the apatite nucleation not directly, but through formation of their calcium compounds and an amorphous calcium phosphate with low Ca/P ratio. Similar process of apatite nucleation was observed also for Ta-OH groups (23).

T. Kokubo et al. Calcification on material

III/91

Conclusion
CaO, SiO2-based glasses formed the apatite on their surfaces in SBF, but CaO, P2O5-based glasses did not. The apatite formation on the former glasses is attributed to the catalytic effect of the Si-OH groups, which were formed on their surfaces in SBF, for the apatite nucleation. Gels of SiO2, TiO2, ZrO2, Ta2O5, and Nb2O5 formed the apatite on their surfaces in SBF. This indicates that Ti-OH, Zr-OH, Ta-OH, and Nb-OH groups besides the Si-OH groups are also effective for the apatite nucleation. They are all negatively charged at pH 7.40. Their apatite

nucleating ability is, however, varied with their arrangements. Among the Ti-OH groups, those in anatase structure is most effective for the apatite nucleation. These functional groups induce the apatite nucleation not directly, but through formation of their calcium compounds and subsequent formation of an amorphous calcium phosphate with low Ca/P atomic ratio.

Acknowledgments This study was supported by Grant-in-Aid 11490020 for Scientic Research, the Ministry of Education, Science, Sports, and Culture, Japan.

References
1. Hench LL, Andersson H (1993) Bioactive glasses. In: Hench LL, Wilson J (eds) An Introduction to Bioceramics. World Scientic, Singapore: 4162 2. LeGeros RZ, LeGeros JP (1993) Dense hydroxyapatite. In: Hench LL, Wilson J (eds) An Introduction to Bioceramics. World Scientic, Singapore: 139180 3. Kokubo T (1993) A/W Glass-ceramic: processing and properties. In: Hench LL, Wilson J (eds) An Introduction to Bioceramics. World Scientific, Singapore: 7588 4. Neo M, Kotani S, Nakamura T, Yamamuro T, Ohtsuki C, Kokubo T, Bando Y (1992) A comparative study of ultrastructures of the interface between four kinds of surface-active ceramic and bone. J Biomed Mater Res 26: 14191432 5. Loty C, Sautier JM, Boulekbache H, Kokubo T, Kim HM, Forest W (2000) In vitro bone formation on a bonelike apatite layer prepared by a biomimetic process on a bioactive glass-ceramic. J Biomed Mater Res 49: 423434 6. Neo M, Nakamura T, Ohtsuki C, Kokubo T, Yamamuro T (1993) Apatite formation on three kinds of bioactive material at an early stage in vivo: a comparative study by transmission electron microscopy. J Biomed Mater Res 27: 9991006 7. Kokubo T, Ito S, Shigematsu M, Sakka S, Yamamuro T (1987) Bonding between bioactive glasses, glass-ceramics or ceramics in a simulated body fluid. Yogyo-Kyokai-Shi 95: 785791 8. Kokubo T, Kushitani H, Sakka S, Kitsugi T, Yamamuro T (1990) Solutions able to reproduce in vivo surface-structure changes in bioactive glass-ceramic A-W. J Biomed Mater Res 24: 721734 9. Ohtsuki C, Kokubo T, Yamamuro T (1992) Mechanism of apatite formation on CaO-SiO2-P2O5 glasses in a simulated body uid. J Non-Cryst Solids 143: 8492 10. Li P, Ohtsuki C, Kokubo T, Nakanishi K, Soga N, Nakamura T, Yamamuro T (1992) Apatite formation induced by silica gel in a simulated body uid. J Am Ceram Soc 75: 20942097 11. Li P, Ohtsuki C, Kokubo T, Nakanishi K, Soga N, Nakamura T, Yamamuro T (1992) A role of hydrated silica, titania and alumina in forming biologically active apatite on implant. Trans Fourth World Biomater Cong 4 12. Uchida M, Kim HM, Miyaji F, Kokubo T, Nakamura T (1998) Apatite-forming ability of zirconia gel in modied SBF solutions. In: LeGeros RZ, LeGeros JP (eds) Bioceramics Vol 11. World Scientific, Singapore: 7780 13. Miyazaki T, Kim HM, Miyaji F, Kokubo T, Nakamura T (1998) Apatite-forming ability of sodium tantalate gels. In: LeGeros RZ, LeGeros JP (eds) Bioceramics Vol 11. World Scientic, Singapore: 481484 14. Miyazaki T, Kim HM, Kokubo T, Kato H, Nakamaura T, Ohtsuki C (2000) Bonelike apatite formation on niobium oxide gel in a simulated body fluid. In: Giannini S, Moroni A (eds) Bioceramics Vol 13, Trans Tech Pub., Switzerland: 4346 15. Cho SB, Nakanishi K, Kokubo T, Soga N, Ohtsuki C, Nakamura T (1996) Apatite formation on silica gel in simulated body uid: its dependence on structure of silica gels prepared in different media. J Biomed Mater Res (Appl Biomater) 33: 145151 16. Cho SB, Miyaji F, Kokubo T, Nakanishi K, Soga N, Nakamura T (1996) Apatite formation on various silica gels in simulated body uid containing excessive calcium ion. J Ceram Soc Japan 104: 399404 17. Uchida M, Kim HM, Kokubo T, Nakamura T (1999) Apatite-forming ability of titania gels with different structures. In: Ohgushi H, Hastings H, Yoshikawa T (eds) Bioceramics Vol 12. World Scientic, Singapore: 149152 18. Tanahashi M, Matsuda T (1997) Surface functional group dependence on apatite formation on self-assembled monolayer in a simulated body uid. J Biomed Mater Res 134: 305315 19. Tanahashi M, Yao T, Kokubo T, Minoda M, Miyamoto T, Nakamura T, Yamamuro T (1995) Apatite coated on organic polymers by biomimetic process: improvement in its adhesion to substrate by glowdischarge treatment. J Biomed Mater Res 29: 349357 20. Takadama H, Kim HM, Miyaji F, Kokubo T, Nakamura T (2000) Mechanism of apatite formation induced by silanol groups TEM observation. J Ceram Soc Japan 108: 118121 21. Kim HM, Miyaji F, Kokubo T, Nakamura T (1997) Effect of heat treatment on apatite forming of Ti metal induced by alkali treatment. J Mater Sci: Mater Med 8: 341347 22. Takadama H, Kim HM, Kokubo T, Nakamura T (2000) TEM-EDX study of process of apatite formation on bioactive titanium metal in simulated body uid. In: Giannini S, Moroni A (eds) Bioceramics Vol 13, Trans Tech Pub., Switzerland: 5154 23. Miyazaki T, Kim HM, Kokubo T, Kato H, Nakamura T (2000) Mechanism of apatite formation on bioactive tantalum metal prepared by chemical treatment. Trans Sixth World Biomater Cong: 264