Materials Transactions, Vol. 50, No. 5 (2009) pp. 1046 to 1049
Special Issue on Nano-Materials Science for Atomic Scale Modification
#2009 The Japan Institute of Metals
Synthesis and Characterization of Silicon-Doped Hydroxyapatite
Kentaro Nakata1; * , Takashi Kubo1; *, Chiya Numako2 , Takamasa Onoki1 and Atsushi Nakahira1;3
1
Department of Materials Science, Graduate School of Engineering, Osaka Prefecture University, Sakai 599-8531, Japan
Department of Natural system, Faculty of Integrated Science, Tokushima University, Tokushima 770-8502, Japan
3
Institute of Materials Research, Osaka Center, Tohoku University, Sakai 599-8531, Japan
2
Silicon was doped to hydroxyapatite by hydrothermal techniques for higher biocompatibility. Products contained tetra-ethyl-orthosilicate
(TEOS) as a silicon source in the range of 0 to 15 mass%. In order to evaluate bioactivity of silicon-doped hydroxyapatite, the samples were
soaked in simulated body fluid (SBF). Silicon doped samples showed faster apatite forming ability than the undoped samples. The samples were
examined by transmission electron microscopy (TEM), X-ray diffraction patterns (XRD), Fourier transform infrared spectroscopy (FTIR) and
X-ray absorption fine structure (XAFS). There results indicated that SiO4 4 ion substituted PO4 3 ion site in apatite structures. And it was found
that appropriate TEOS doping ratio was 10 mass% for superior biocompatibility due to amorphous SiO2 segregation in the 15 mass% TEOS
doped samples. [doi:10.2320/matertrans.MC200808]
(Received November 4, 2008; Accepted January 6, 2009; Published February 18, 2009)
Keywords: hydroxyapatite, silicon, hydrothermal, doping, X-ray absorption fine structure
1.
Introduction
Hydroxyapatite (HAp;[Ca10 (PO4 )6 (OH)2 ]) has some advantages as a biomaterial due to its chemical similarity to the
inorganic component of the bone and tooth, compared to
other ceramics, such as Bioglass and A-W glass.1) For this
reason hydroxyapatite is utilized as a biomaterial to fill bone
defects, for bioactive coating for metallic prostheses.2)
However, stoichiometric hydroxyapatite has relatively low
ability to associate with existing bone, remaining as a
permanent fixture susceptible to long term failure.3) In
general, the mineral in bone is non-stoichiometric, but
exhibits variable deficiencies in Ca, P and OH. Especially,
the elements such as Si, Mg, Sr and Na are contained to
biological apatite, and they play important roles in the
biochemistry of bone and tooth.4) There substitutions can
affect the surface structure and charge of hydroxyapatite,
leading to influence on the material in biological environments. In vitro and in vivo studies of performed by Carlisle
indicated the importance of the silicon on bone formation and
calcification.5) In particular, their study has shown that silicon
was localized in active growth areas, such as the osteoid, of
the young bone of mice and rats. And it is noted that silicon is
essential to the growth and development of biological tissue
such as bone, teeth and some invertebrate skeletons. In
addition, it has been observed that the doping of Si to
hydroxyapatite results in an improvement of the bioactive
behavior and advance of bone formation in vivo.6) Some
authors were assuming that Si or SiO4 4 , replaces P or
PO4 3 , with the subsequent loss of charge equilibrium.7) In
this sense, a one of potential method for improving the
bioactivity of hydroxyapatite is the doping of silicon.8)
However, it is important that this doping of the silicon does
not result in thermal instability of the silicon doped
hydroxyapatite as this effect upon sintering would result in
the decomposition of the Si-HAp to undesirable second
phase.
Several methods of the synthesis of silicon doped
hydroxyapatite have been described. Ruys suggested the
*Graduate
Student, Osaka Prefecture University
use of a sol-gel procedure, however, these materials, in
addition to a hydroxyapatite phase, contained calcium
silicophosphate and either - or -tricalcium phosphate,
depending on the degree of silicon doping.9) Boyer et al.
conducted studies on the silicon-substituted hydroxyapatite
by solid state reaction, but in these cases the incorporation of
a secondary ion, like lanthanum or sulphate, was needed.10)
Gibson and Kim et al. synthesized silicon-containing
hydroxyapatite by using a wet-chemical method, however,
sintering at high temperature usually leads to bigger crystal
size.11) Hydrothermal technique usually gives materials with
a high degree of crystallinity and with a Ca/P ratio close to
stoichiometric hydroxyapatite.12) Kim et al. reported that a
porous silicon incorporated hydroxyapatite had been prepared by hydrothermal treatment and solvothermal treatment
of natural coral repeatedly.13,14) Our previous study on HAp
containing with SiO2 sol indicated the possibility doping of
SiO2 to HAp by the hydrothermal process.15–20) However, it
has not come out about state of silicon in the silicon doped
hydroxyapatite synthesized by hydrothermal technique. If
SiO2 segregated on surface of the sample, it might lead a
danger to public health such as blood coagulation, infection.
The purpose of the present study is to bring out about the state
and solubility limit of silicon which silicon-doped hydroxyapatite synthesized by hydrothermal technique with various
silicon contents. Additionally, chemical characterization of
the silicon-doped hydroxyapatite has been carried out.
2.
Experimental
Stoichiometric hydroxyapatite (HAp) and a series of
silicon-doped HAp (Si-HAp) were prepared by hydrothermal
method using Ca(OH)2 , H3 PO4 and Si(OCH2 CH3 )4 (TEOS)
as starting materials. The amount of reagents was calculated
on the assumption that silicon would substitute phosphorus.
The starting materials at the Ca/(P+Si) ratio of 1.67 were
mixed and at the time TEOS introduced 0, 5, 10, 15 mass%.
Ca(OH)2 , ion exchange water of 25 ml stirred and added
H3 PO4 , TEOS then hydrothermal treatment at 150 C for 4
days. The resulting products were filtered, dried at 50 C over
night in an oven.
Synthesis and Characterization of Silicon-Doped Hydroxyapatite
3.
Results and Discussion
The doping mechanism of silicon to HAp would describe
the substitution of a phosphate group for a silicate group, with
an appropriate mechanism for charge balance, is given in
eq. (1):
Ca10 (PO4 )6 (OH)2 þ xSiO4 4
) Ca10 (PO4 )6x (SiO4 )x (OH)2x þ xPO4 3 þ xOH ð1Þ
To compensate for the extra negative charge of the silicate
groups, some of the OH groups would be lost to retain
charge balance.
Figure 1 shows TEM images of the products prepared by
hydrothermal technique. As shown in Fig. 1, the pure HAp
was rod-shaped. However, Si doped samples became
spherical-shaped with increasing TEOS content. The value
of crystal size decreases with the TEOS addition. Figure 2
shows XRD pattern of the products prepared by hydrothermal
technique. The silicon doping did not appear to affect the
diffraction pattern of hydroxyapatite. The pattern of the all
samples appeared to be identical, without secondary phase,
such as TCP or SiO2 . All the diffraction peak of samples
Fig. 1 TEM images of synthesized HAp and Si-HAp. (a) HAp; (b)
5 mass% Si-HAp; (c) 10 mass% Si-HAp; (d) 15 mass% Si-HAp.
hydroxyapatite
(d)
Intensity/a.u.
The characterization of the microstructure analysis of HAp
and Si-HAp were performed by TEM (JEM2010/SP, JOEL,
Japan) with accelerating voltage of 200 kV. The phase
composition of HAp and Si-HAp were determined by XRD
(RINT 2100, Rigaku Co., Japan), using X-ray diffractometer
at 30 kV and 30 mA. Scans were run from 10 to 60 2 at a
speed of 2 /min and a step of 1 using Cu K radiation. And
the determination of HAp and Si-HAp deviations were
performed by FTIR (370DTGS/FT-IR, Nicolet-Avatar Co.,
Japan). The calcium and phosphorus contents were determined by ICP (ICPS-7000, Shimadzu Co., Japan). The in
vitro bioactivity of samples was investigated by soaking them
in simulated body fluid (SBF) for various days. The SBF
solution was prepared according to the procedure described
by Kokubo.21) The soaking experiment was carried out in a
case maintained at 36.5 C, the SBF solution was not
refreshed during the experiment. After the pre-selected
soaking time, the samples were dried. The microstructures
of the samples were observed by SEM (S-4500, Hitachi,
Japan) with accelerating voltage of 15 kV. The local structure
of samples was investigated by XANES spectra for Ca Kedge, P K-edge, and Si K-edge. Ca K-edge XAFS data for
this study was corrected by transmission mode using the
Si (111) double crystal monochromater at BL01B1 in the
SPring8. The data were collected with the ionization
chambers filled with gas (I0 chamber: He/N2 = 7/3, I
chamber: N2 ). For the XAFS measurements, the samples
were prepared as pellets with the thickness varied to obtain a
0.5-1 jump at the Ca K absorption edge. Cu metallic foil
was used for the energy calibration. P K-edge and Si K-edge
XANES spectra were recorded in total-electron yield mode
using InSb and KTP crystal monochromater at BL1A in
UVSOR. The data were collected with the ionization
chambers. XANES were analyzed by subtracting a linear
background computed by least-squares fitting. The analysis
of XANES data was conducted using the commercial
software ‘‘REX2000’’ (Rigaku Co. Ltd., Japan).
1047
(c)
(b)
(a)
10
20
30
40
2 θ /degree
50
60
Fig. 2 XRD patterns of synthesized HAp and Si-HAp. (a) HAp; (b)
5 mass% Si-HAp; (c) 10 mass% Si-HAp; (d) 15 mass% Si-HAp.
matched the standard for hydroxyapatite. However, these
reflections become slightly broad and less intense as the
TEOS addition. The reason could be the result of degradation
of crystallinity attributed to the formation of hydroxyl
vacancy caused by the substitution of PO4 3 by SiO4 4 .
It also could be caused by the decrease of crystal size.
This result suggested the substitution of silicon into the
hydroxyapatite lattice.
FTIR was evaluated to confirm the effect of the silicon
substitution on the different functional groups, such as
hydroxyl and phosphate groups of hydroxyapatite. Figure 3
shows the FTIR spectra of prepared samples. First, the bands
at 3571 cm1 and 631 cm1 corresponded to the hydroxyl
group stretching and vibrational modes, respectively. The
intense bands at 962 cm1 corresponded to P-O stretching
1048
K. Nakata, T. Kubo, C. Numako, T. Onoki and A. Nakahira
Transmittance/a.u.
(d)
(c)
(b)
800
(a)
603
962
631
3571
4000
3500
1500
1000
567
500
Wavenumbers / cm-1
Fig. 3 FTIR spectra of synthesized HAp and Si-HAp. (a) HAp; (b)
5 mass% Si-HAp; (c) 10 mass% Si-HAp; (d) 15 mass% Si-HAp.
Table 1
Fig. 4 SEM images of bulk of HAp and Si-HAp after immersion in SBF
for various periods. (a) HAp after 3day’s immersion; (b) 5 mass% Si-HAp
after 3 day’s immersion; (c) 10 mass% Si-HAp after 2 day’s immersion;
(d) 15 mass% Si-HAp after 1 day immersion.
The Ca/P rate of synthesized HAp and Si-HAp.
Samples
Ca/P
HAp
1.670
5 mass% Si-HAp
10 mass% Si-HAp
1.979
2.084
15 mass% Si-HAp
2.275
vibration modes, whereas the doublet at 603–567 cm1
corresponds to the O-P-O bending mode.22) Furthermore,
the broad band at 3500 cm1 was assigned to moisture in the
samples. The FTIR spectra of Si-HAp samples showed
several significant changes from the synthesized HAp. The
notable effect of silicon doped hydroxyapatite on FTIR was
the decrease in the hydroxyl stretching bands at 3571 cm1
and 631 cm1 and phosphate stretching bands at 962 cm1
with increasing TEOS content. Moreover, new band appearing at 800 cm1 only for Si-HAp, was assigned to Si-O-Si
vibration modes of SiO4 4 groups which have been polymerized.23) According to eq. (1), incorporation of silicon into
the hydroxyapatite structure (Ca10 (PO4 )6x (SiO4 )x (OH)2x )
reduces the amount of hydroxyl groups for compensating
for the extra negative charge of the silicate group. There
result suggest that incorporation of silicon into some of the
phosphorous sites in the lattice results in changes in the
bonding and symmetry of the phosphate groups, charge is
compensated by loss of OH . Table 1 lists Ca/P molar ratio
of prepared samples set up by ICP analysis. Ca/P molar ratio
increased with increasing TEOS content. It could be the
result of the decrease phosphate groups by addition of silicon.
Furthermore, all Si-HAp samples were confirmed the
existence of silicon. These results also show that PO4 3
tetrahedrons are replaced by SiO4 4 tetrahedrons in the
hydroxyapatite structure.
The compactions of synthesized samples were soaked in
SBF solution to evaluate their bioactivity. Figure 4 shows
SEM images of the samples after immersion in SBF for
various periods. There were observed the notable precipitation of bone-like apatite on the surface of 5 mass% Si-HAp
after 3 days immersion, 10 mass% Si-HAp after 2 days
immersion and 15 mass% Si-HAp after 1 day immersion.
However, silicon undoped sample only have no change after
soaking SBF for any periods. From this result, it found that
the biocompatibility of samples was improved by TEOS
addition and the formation of bone like apatite more
accelerated with increasing TEOS content. And it also
suggested that the substitution of silicon for HAp can affect
the surface structure, surface charge, and solubility of HAp,
which can, result in changes in the biological performance in
vitro. The characterization of surface charge and adsorption
for silicon doped HAp is now under investigation.
The local structure of near Ca, P, Si atoms were evaluated
by XAFS to circumstantially examine the state of doping
silicon to HAp. Figure 5(A) shows the first derivative of
functions to Ca K-edge XANES spectra for Si-HAp and
commercial HAp as a reference sample. The spectra of
synthesized all samples were very close to the peak position
of commercial HAp. This result indicated that the local
structure of near Ca atoms in all Si-HAp were similar to HAp
regardless of TEOS content. Figure 5(B) shows the first
derivative of functions of P K-edge XANES spectra for SiHAp and commercial HAp. The energy position of adsorption edge of silicon doping HAp shifted to high energy side.
As previously mentioned, FTIR showed that PO4 3 groups
were replaced by SiO4 4 groups in the hydroxyapatite
structure with the TEOS addition, charge balance was
maintained by loss of OH . This reaction could be caused
the change of electron state which the local structure of near
P atoms, leading to shift of the energy position of adsorption
edge. Besides, Fig. 5(C) shows the first derivative of
functions of Si K-edge for silicon doping HAp and
Synthesis and Characterization of Silicon-Doped Hydroxyapatite
(A)
(B)
(c)
(b)
(a)
Derivatives/a.u.
(d)
4030
(C)
(e)
Derivatives/a.u.
Derivatives/a.u.
(e)
(d)
(c)
(b)
4070
2140
(d)
(c)
(b)
(a)
(a)
4040 4050 4060
Energy/eV
1049
2150
2160
Energy/eV
2170
1840
1850
Energy/eV
Fig. 5 The first-derivative functions of (A): the Ca K edge XANES spectra for (a) commercial HAp; (b) HAp; (c) 5 mass% Si-HAp;
(d) 10 mass% Si-HAp; (e) 15 mass% Si-HAp;, (B): the P K edge XANES spectra shown in (A), (C): the Si K edge XANES for
(a) 5 mass% Si-HAp; (b) 10 mass% Si-HAp; (c) 15 mass% Si-HAp and (d) commercial amorphous SiO2 .
commercial amorphous SiO2 as a reference. The energy
position of adsorption edge of Si K-edge of 5 and 10 mass%
Si-HAp were agreed, however, 15 mass% Si-HAp sifted to
amorphous SiO2 side. From this result, it suggested that a
doping TEOS for from 5 to 10 mass% Si-HAp were
substituted to phosphate groups in the hydroxyapatite
structure, and for 15 mass% Si-HAp in part TEOS was
segregated as amorphous SiO2 . Therefore, it could be caused
the increase of biocompatibility for 15 mass% Si-HAp by
both segregated amorphous SiO2 and doping SiO2 into
HAp, leading to the good bioactivity of bioactive glasses and
glass-ceramics.
4.
Conclusion
Silicon doping hydroxyapatite powder were prepared
by hydrothermal method using Ca(OH)2 , H3 PO4 and
Si(OCH2 CH3 )4 (TEOS) as reagents. The analysis of TEM,
XRD, FTIR and ICP shows that the substitution of silicate
groups for the phosphate groups causes some OH loss to
maintain the charge balance. Furthermore, the synthesized
samples were soaked in SBF solution to evaluate their
bioactivity. The silicon doping HAp presented superior
bioactivity with increasing silicon content. However, the
analysis of XAFS suggested that 15 mass% Si-HAp segregated amorphous SiO2 . Therefore, 5 and 10 mass% Si-HAp
could be applied for biomaterial that had superior biocompatibility than 15 mass% Si-HAp for living body.
Acknowledgment
XAFS measurements were carried out at SPring8
(2007A1332 and 2006A1294) and at UVSOR (16-506).
This work was partly supported by Grant-in-Aid for
Scientific Research from Japan Society for the Promotion
of Science (JSPS) (No. 20047011), Grant-in-Aid for Scien-
tific Research on Priority Areas ‘‘Nano-materials Science for
Atomic Scale Modification’’.
REFERENCES
1) S. H. Maxian, J. P. Zawaddsky and M. G. Dunn: J. Biomed. Mater.
RES. 28 (1994) 1311.
2) E. L. Solla, J. P. Borrajo, P. Gonzalez, J. Serra, S. Chiussi, B. Leon and
J. Garcia Lopez: Appl. Surf. Sci. 253 (2007) 8282.
3) E. Schepers, M. Declercq, P. Ducheyne and R. Kempeneers: J. Oral
Rehabil. 18 (1991) 439.
4) S. Sprio, A. Tampieri, E. Landi, M. Sandri, S. Martorana, G. Celotti and
G. Logroscino: Mater. Sci. Eng. 28 (2008) 179.
5) E. M. Carlisle: Science 167 (1970) 279.
6) A. E. Porter, N. Patel, J. N. Skepper, S. M. Best and W. Bonfield:
Biomater. 25 (2004) 3307.
7) D. Arcos, J. R. Carvajal and M. V. Regi: Chem. Mater. Res. 4 (1994)
422.
8) R. Z. LeGeros: Nature 206 (1965) 279.
9) A. J. Ruys: J. Aust. Ceram. Soc. 29 (1993) 77.
10) L. Boyer, J. Carpena and J. L. Lacout: Sol. Sta. Ion. 95 (1997) 121.
11) I. R. Gibson, S. M. Best, W. Bonfield and J. Biomed: Mater. Res. 4
(1997) 422.
12) X. L. Tang, X. F. Xiao and R. F. Liu: Mater. Lett. 59 (2005) 3843.
13) Y. H. Kim, H. Song and D. H. Riu: Current Appl. Phys. 5 (2005) 538.
14) Y. Tanizawa and T. Suzuki: Phos. Res. Bull. 4 (1994) 87.
15) A. Nakahira, C. Karatani, S. Konishi, F. Nishimura, S. Takeda, S.
Nishijima and T. Watanabe: Zairyo 52 (2003) 4205.
16) A. Nakahira, C. Karatani and S. Nishida: Phos. Res. Bull. 17 (2004)
148–152.
17) M. Tamai, T. Yamamoto, H. Aritani and A. Nakahira: Phos. Res. Bull.
17 (2004) 69–74.
18) A. Nakahira, T. Okajima, T. Honma, S. Yoshioka and I. Tanaka: Chem.
Let. 35 (2006) 856–857.
19) A. Nakahira, S. Nakamura and M. Horimoto: IEEE Trans. Magn. 43
(2007) 2465–2467.
20) A. Nakahira, M. Horimoto, S. Nakamura, S. Ishihara, H. Nagata, T.
Kubo and C. Karatani: J. Ion Exc. 18 (2007) 306–309.
21) T. Kokubo: J. Non-Cryst. Solids 120 (1990) 138.
22) T. Leventouri, C. E. Bunaciu and V. Perdikatsis: Biomater. 24 (2003)
4205.
23) I. Rehman and W. Bonfield: J. Mater. Sci. Mater. Med. 8 (1997) 1.