Ceramics International 30 (2004) 1781–1785

Sol–gel derived bioglass as a coating material

for porous alumina scaffolds
Jianli Liu, Xigeng Miao∗
School of Materials Engineering, Nanyang Technological University, Nanyang Avenue, Singapore 639798, Singapore

Received 12 November 2003; received in revised form 11 December 2003; accepted 22 December 2003

Available online 6 May 2004

Abstract

Bioglasses are important bioactive materials and have been used for the repair and reconstruction of diseased bone tissues, as they exhibit
direct bonding with human bone tissues. However, bioglasses have low mechanical properties and thus are usually used as coatings on strong
substrates. A sol–gel method was used in this study to produce the 58S bioglass powder with the composition of 58 mol% SiO2 –38 mol%
CaO–4 mol% P2 O5 . The bioglass powder was then pressed into discs and sintered at 500, 800, 1000, and 1200 ◦ C, respectively. The apparent
density and the Vicker’s hardness of the sintered bioglass were tested using a densometer and a Vicker’s hardness tester. The thermal
properties of the bioglass were also studied using differential scanning calorimetry (DSC) and dilatometry. The crystallization of the bioglass
was examined using X-ray diffraction (XRD) and scanning electron microscopy (SEM). The sintered bioglass discs were then immersed in the
simulated body fluid (SBF) solution. The bioglass surface after the immersion was examined using field emission scanning electron microscopy
(FE-SEM). The formation of apatite layer was confirmed using Fourier transform infrared spectroscopy (FTIR). The apparent density and the
hardness of the bioglass reached 3.12 g/cm3 at 1000 ◦ C and 1.05 GPa at 1200 ◦ C, respectively. The formation of hydroxyl-carbonate apatite
layer confirmed the bioactivity of the bioglass. It was also found that at sintering temperatures above 800 ◦ C, crystalline phase CaSiO3 was
formed. With the formation of CaSiO3 , the bioactivity of the bioglass started to decrease.
© 2004 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Keywords: Sol–gel; Bioglass; Crystallinity; Bioactivity; Coating

1. Introduction ceramic particles and fill the micropores during the sintering
process [4,5]. More importantly, one advantage of bioactive
Over the last decade or so, several kinds of bioceramics glasses over other bioactive ceramics is the possibility of
have been studied for treating damaged and diseased bones controlling a range of chemical properties and the rate of
[1,2]. Namely, they are bioinert ceramics (e.g., alumina and bonding to tissues. It is also possible to design glasses with
zirconia), bioactive ceramics (e.g., bioglasses and hydroxya- properties specific to particular clinical applications [7].
patite) and biodegredable ceramics (e.g., tri-calcium phos- A few methods have been developed to prepare the bio-
phate and bone cement). Among these bioceramics, bioglass glasses. Melting method is the traditional one for glass
and glass-ceramics have attracted more and more attentions. preparation [8,9]. It is simple and suitable for massive
As a kind of bioactive ceramics, the primary characteristic production. However, during the high temperature stage,
of bioglasses is their rapid rate of surface reaction, which the volatile component P2 O5 tends to escape. In recent
leads to direct attachment to bone by a chemical bond [3]. years, sol–gel method has attracted many studies [10]. The
The excellent bioactivity but low mechanical properties re- sol–gel process has the advantage of low reaction tem-
quires the bioglass to be used as coating for surface modifi- perature. Compared with the melting method, the sol–gel
cation and bioactive phase in composites [4–6]. As a glass derived glass powder has homogenous composition in par-
phase, bioglass also has relatively low softening temperature. ticles. A group of sol–gel derived bioglasses in the system
Therefore, bioglass can be used as sintering aid to bond the of SiO2 –CaO–P2 O5 has been widely studied, such as 55S,
58S, 60S, etc. Their excellent bioactivity offers great poten-
∗ Corresponding author. Tel.: +65-67904260; fax: +65-67909081.
tial for hard tissue surgery [10]. In the present paper, 58S
E-mail address: [email protected] (X. Miao). sol–gel derived bioglass was chosen and the properties of

0272-8842/$30.00 © 2004 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
doi:10.1016/j.ceramint.2003.12.120
1782 J. Liu, X. Miao / Ceramics International 30 (2004) 1781–1785

58S bioglass as a coating material were studied. More im- 0.5

portantly, the influence of the crystallization of the bioglass
0.0
on the mechanical properties and the bioactivity were also EXO
studied. -0.5
DSC (mW/mg)

-1.0

2. Experimental procedure -1.5

Sol–gel derived 58S bioglass, with the composition of -2.0

58 mol% SiO2 –38 mol% CaO–4 mol% P2 O5 was used in the -2.5
study. The glass was prepared by hydrolysis and polycon-
densation of tetraethyl orthosilicate (TEOS), triethyl phos- -3.0
200 400 600 800 1000
phate (TEP) and Ca(NO3 )2 ·4H2 O. Solution HCl (1N) was o
Temperature ( C)
used to catalyze the TEOS and TEP hydrolysis, with a mo-
lar ratio of (HCl + H2 O)/(TEOS + TEP) = 8. Each reactant Fig. 1. DSC curve for the bioglass powder sample.
was consecutively added under continuous stirring. The so-
lution was then aged in an oven at 60 ◦ C for 48 h. The dried
The coating was observed on a cross-section specimen using
gel was then calcined at 700 ◦ C for 2 h, followed by 4 h ball
SEM.
milling to obtain the glass powder. To study the thermal
properties of 58S bioglass, differential scanning calorime-
try (DSC) (Thermal Analysis DSC 404C) measurement was
carried out on the glass powder sample up to 1000 ◦ C at 3. Results and discussion
heating rate of 5 ◦ C/min. Thermal expansion measurement
was carried out using a dilatometer (Unitherm, Anter Corp.). 3.1. Thermal property analysis
The sample for the dilatometry was produced via uniaxially
Fig. 1 is the DSC graph of the 58S bioglass powder at
die-pressing of the bioglass powder at 250 MPa, followed
the heating rate of 5 ◦ C/min. A weak endothermic peak can
by sintering at 1200 ◦ C for 1 h.
be noticed on the DSC curve at the range of 277–302 ◦ C,
The samples for other studies were prepared by uniaxial
followed by an exothermic peak at the range of 848–903 ◦ C.
pressing of the bioglass powder at 250 MPa in a steel die
The endothermic peak corresponded to the glass transition
to produce disks of 13 mm in diameter and 3 mm in thick-
temperature of the bioglass, whereas the exothermic peak
ness. Disks were then fired at 500, 800, 1000, and 1200 ◦ C,
showed crystallization at the temperature of 848–903 ◦ C.
respectively, using a heating rate of 5 ◦ C/min. The set
Fig. 2 shows the thermal expansion behavior of the 58S
temperature was kept for 2 h and then the furnace was nat-
bioglass. It is clear that the sample expanded below 600 ◦ C.
urally cooled down. The apparent density and the Vicker’s
Form the expansion curve, a coefficient of thermal expan-
hardness of the sintered bioglass disks were tested using a
sion (CTE) of 0.73 × 10−6 ◦ C−1 can be calculated. With
densometer and a Vicker’s hardness tester. The crystalliza-
the increase of temperature, the sample began to shrink, re-
tion of the bioglass was examined using X-ray diffraction
sulting from the softening of the 58S bioglass. Therefore,
(XRD) (Lab XRD-6000 Shimadzu). The morphologies of
the softening temperature of the bioglass can be regarded as
the crystals in bioglass samples were observed using scan-
600 ◦ C.
ning electron microscopy (SEM) (SEM JEOL-JSM 5410).
The sintered bioglass discs were separately immersed into
the simulated body fluid (SBF) solution for 1 h, 6 h, 24 h, 3
days, and 7 days. The bioglass surface after the immersion
was observed using field emission scanning electron mi-
croscopy (FE-SEM) (FESEM, JSM 6340). The formation
Expansion (% )

of the apatite layer was analyzed using Fourier transform

infrared spectroscopy (FTIR) (Perkin-Elmer S2000 FTIR
spectrometer).
The ultimate purpose of the study was to use 58S bio-
glass as a coating material. Therefore, 58S bioglass coating
was applied on alumina substrate via a dip coating method.
The alumina specimen was formed via slip casting, fol-
lowed by sintering at 1550 ◦ C for 2 h. Then the alumina 200 400 600 800 1000 1200
o
specimen was dipped into a 50 wt.% 58S bioglass suspen- Temperature ( C)
sion for 5 min. After drying and firing at 1200 ◦ C for 1 h, Fig. 2. Thermal expansion curve of 58S bioglass sample sintered at
a bioglass coating was formed on the alumina specimen. 1200 ◦ C for 1 h.
 J. Liu, X. Miao / Ceramics International 30 (2004) 1781–1785 1783

1200 3.2

1000 3.1
Apparent density (g/cm )
3

800 o 3.0
Intensity (cps)

1200 C

600 2.9
o
1000 C
400 2.8
o
800 C
200 2.7
raw powder
0
2.6
20 30 40 50 60 70 80
400 500 600 700 800 900 1000 1100 1200 1300
2 Theta o
Tem perature ( C)
Fig. 3. XRD patterns of the 58S bioglass sintered at various temperatures.
Fig. 5. Influence of sintering temperature on the density of the 58S
bioglass samples.

3.2. Crystallization
sintered glass samples had relatively low density at low tem-
The crystallization of the 58S bioglass is shown in Fig. 3.
perature due to the incomplete sintering of the samples. At
It can be seen that the starting 58S bioglass powder was
1000 ◦ C the bioglass reached the maximum density. With the
amorphous, proving the sol–gel method could prepare pure
further increase of the temperature, the density decreased. It
glasses. As the temperature increased up to 800 ◦ C, only a
can be explained as the result of crystallization of the bio-
broad peak existed in the pattern, showing that the bioglass
glass. During the crystallization, residual pores were left be-
powder still kept the amorphous state. With further increase
hind, resulting in the decrease of density. On the other hand,
of the temperature, sharp peaks appeared at 1000 ◦ C, in-
the Vicker’s hardness of the 58S bioglass is shown in Fig. 6.
dicating crystallization occurred in the 58S bioglass. The
The Vicker’s hardness increased with the increase of sinter-
peaks matched the pattern of wollastonite CaSiO3 (JCPDS
ing temperature. This was because the increase of the sin-
#27-0088). The CaSiO3 peaks became much sharper at
tering temperature corresponded to the increased degree of
1200 ◦ C, corresponding to further crystallization. The wol-
crystallization, that is, more wollastonite CaSiO3 existed in
lastonite CaSiO3 crystals were also observed under SEM.
the glass matrix. It is known that the wollastonite phase has
Fig. 4 shows the wollastonite CaSiO3 platelets existing in
higher hardness than that of the amorphous glass. Therefore,
the bioglass sample sintered at 1200 ◦ C. The wollastonite
after higher temperature sintering, more amount of wollas-
CaSiO3 platelets had 3 ␮m in thickness.
tonite caused the increase of hardness of the 58S bioglass.

3.3. Density and Vicker’s hardness 3.4. Bioactivity

Fig. 5 shows the apparent density of the 58S bioglass sin- To study the effect of crystallization on the bioactivity
tered at 500, 800, 1000 and 1200 ◦ C. It can be seen that the of 58S bioglass, samples sintered at 800 and 1000 ◦ C were

1.1

1.0
Vickers hardness (GPa)

0.9

0.8

0.7

0.6

0.5

0.4
400 500 600 700 800 900 1000110012001300
o
Temperature ( C)

Fig. 4. SEM micrograph of CaSiO3 crystals in glass sample sintered at Fig. 6. Influence of sintering temperature on the Vicker’s hardness of the
1200 ◦ C. 58S bioglass samples.
1784 J. Liu, X. Miao / Ceramics International 30 (2004) 1781–1785

Fig. 7. SEM micrographs of the 58S bioglass sintered at 800 ◦ C and after Fig. 8. SEM micrographs of the 58S bioglass sintered at 1000 ◦ C and
the immersion in SBF for 1 h (a), and 7 days (b). after the immersion in SBF for 24 h (a), and 7 days (b).

studied, corresponding to amorphous and crystallized sam- phosphate peak (P–O stretching peak) could be seen at
ples. Figs. 7 and 8 show the micrographs of the 58S bioglass 1000–1220 cm−1 , indicating the formation of amorphous
samples sintered at 800 and 1000 ◦ C after immersion in the CaO–P2 O5 rich layer. Carbonate bands (C–O stretching
SBF for various times. The micrographs show the formation peaks) were presented at 2350–2400 and 2500–2600 cm−1 ,
of apatite layer on the bioglass. From Fig. 7, it can be seen indicating the formation of carbonated apatite. In addition,
that the pure 58S bioglass exhibited excellent bioactivity.
Apatite particles were observed on the surface of bioglass
90
sample after 1 h immersion, as shown in Fig. 7a. The ap-
atite crystals were round in shape and had particle sizes of 85
0.1 ␮m or so. After 7-day immersion, a fairly dense apatite
layer was formed on the sample surface. 80
% Transm ittance

P-O (bending)
Fig. 8 shows the micrographs of the 58S bioglass sintered 75
at 1000 ◦ C after immersion in SBF for various times. These
samples showed much lower bioactivity than those sintered 70
at 800 ◦ C. No apatite could be observed on the surface until 65 C-O (streching)
after 24 h of immersion. After 7 days, a porous apatite layer
was formed on the sample surface, in contrast to the dense 60
P-O (stretching)
apatite layer formed on the glass sample sintered at 800 ◦ C.
55
The decreased bioactivity of the 58S bioglass sample was 2000 1500 1000 500
due to the existing CaSiO3 . -1
cm
The apatite layers on the 58S bioglass samples were fur-
ther analyzed using FTIR, as shown in Fig. 9. Well-defined Fig. 9. FTIR curve of the apatite layer formed on the 58S bioglass sample.
 J. Liu, X. Miao / Ceramics International 30 (2004) 1781–1785 1785

58S bioglass softened at 600 ◦ C and crystallized at the

temperature range of 850–900 ◦ C. The formation of the
crystalline wollastonite phase limited the densification of
the bioglass. However, the crystalline wollastonite phase
increased the hardness of the 58S bioglass. The forma-
tion of the crystalline wollastonite phase also decreased
the bioactivity of the 58S bioglass. A dense apatite layer
was formed on the highly bioactive amorphous bioglass,
whereas incomplete or porous apatite layer was formed
on the bioglass subjected to crystallization. Finally, the
58S bioglass as a coating could attach to the alumina
substrate.

Acknowledgements
Fig. 10. SEM micrograph of the 58S bioglass coating on the alumina
substrate.
The authors would like to acknowledge the financial sup-
port from the Nanyang Technological University (AcRF
P–O bending peak at 550–620 cm−1 was also observed in
RG26/01).
the FTIR spectra.

3.5. Bioglass coating on the alumina substrate

References
Fig. 10 shows the 58S bioglass coating on the alumina
[1] L.L. Hench, J. Wilson, An Introduction to Bioceramics, World Sci-
substrate. The 58S bioglass coating was 200–300 ␮m thick entific, Singapore, 1993, pp. 3–5.
and was firmly attached to the alumina substrate. Generally [2] L.L. Hench, Bioceramics: from concept to clinic, J. Am. Ceram.
speaking, to obtain a good coating, the CTE of the substrate Soc. 74 (7) (1991) 1487.
and the coating materials must be matched. Otherwise, ther- [3] L.L. Hench, Bioactive glass and glass-ceramics: a perspective, in:
T. Yamamuro, L.L. Hench, J. Wilson (Eds.), Handbook of Bioactive
mal stresses would exist between the substrate and coating
Ceramics, CRC Press, Boca Raton, FL, 1990, p. 7.
to cause cracks or even peeling of the coating layer. In this [4] M.A. Lopes, J.C. Knowles, J.D. Santos, Structural insights of
study, the CTEs of 58S bioglass and alumina were measured glass-reinforced hydroxyapatite composites by Rietveld refinement,
as 0.73 × 10−6 and 9.47 × 10−6 ◦ C−1 . Although the CTEs Biomaterials 21 (2000) 1905.
were quite different, the resultant thermal stresses could be [5] J.D. Santos, L.J. Jha, F.J. Monteiro, In vitro calcium phosphate for-
mation on SiO2 –Na2 O–CaO–P2 O5 glass reinforced hydroxyapatite
relaxed due to the softening of the glass. In addition, due to
composite: a study by XPS analysis, J. Mater. Sci. Mater. Med. 7
the smaller CTE of the bioglass, the stresses in the coating (1996) 181.
could be compressive in nature. As a result, 58S bioglass [6] G. Gabbi, A. Cacchioli, B. Locardi, E. Guadagnino, Bioactive glass
could form a good bonding with the alumina substrate. coating: physicochemical aspects and biological findings, Biomate-
From the current study, it can be seen that the 58S bio- rials 16 (1995) 515.
glass coating would partially crystallized at 1200 ◦ C. The [7] H.A. Elbatal, M.A. Azooz, E.M.A. Khalil, A.S. Monem, Y.M.
Hamdy, Characterization of some bioglass-ceramics, Mater. Chem.
crystallized 58S bioglass coating would form a porous ap- Phys. 80 (2003) 599.
atite layer in vivo. The porous apatite layer would not only [8] K. Franks, I. Abrahams, G. Georgious, J.C. Knowles, Investigation of
provide composition affinity with the bone tissue, but also thermal parameters and crystallisation in a ternary P2 O5 –CaO–Na2 O
induce the bone tissue ingrowth. Thus, the 58S bioglass coat- based glass system, Biomaterials 22 (2001) 497.
[9] J.D. Santos, P.L. Silva, J.C. Knowles, S. Talal, F.J. Monteriro, Re-
ing not only could bond to the alumina substrate but also
inforcement of hydroxyapitite by adding P2 O5 –CaO glasses with
would bond to the bone tissue. Na2 O, K2 O and MgO, J. Mater. Sci. Mater. Med. 7 (1996)
187.
[10] R.M. Vallet-Regı́, Static and dynamic in vitro study of a sol–gel
4. Conclusions glass bioactivity, Biomaterials 22 (2001) 2301.

In this study, the sol–gel derived 58S bioglass was stud-

ied for use as a coating material. It was found that the