J. Am. Ceram. Soc.

, 94 [12] 4459–4462 (2011)

DOI: 10.1111/j.1551-2916.2011.04716.x
© 2011 The American Ceramic Society

Journal
The System Ca3(PO4)2–Ca2SiO4: The Sub-System Ca2SiO4–7CaOP2O52SiO2
Virtudes Rubio,‡ Miguel A. de la Casa-Lillo,‡ Salvador De Aza,§,** and Piedad N. De Aza,‡,†
‡
Instituto de Bioingenierı́a, Universidad Miguel Hernández, Avda. de la Universidad s/n, 03202, Elche (Alicante), Spain
§
Instituto de Cerámica y Vidrio, CSIC, C/Kelsen, 5, 28049, Madrid, Spain

A study of the sub-system Ca2SiO4–7CaOP2O52SiO2 within CaCO3 were heated in a platinum crucible at 1500°C for 3 h.
the system Ca3(PO4)2–Ca2SiO4 was conducted as a prelimin- Then was liquid nitrogen quenched by rapid withdrawal
ary step toward obtaining new biomaterials with controlled from the furnace. The obtained material was ground and
microstructures. Phase composition of the resulting ceramics characterized using XRD.
was studied using X-ray diffraction and scanning electron C2S was obtained using solid state reaction-sintering, start-
microscopy with attached wavelength dispersive spectroscopy. ing from an appropriate mixture of calcium carbonate,
The results showed that the sub-system presents an invariant (CaCO3 >99.0 wt%; Sigma) with an average particle size
eutectoid point at 518°C ± 6°C with a composition of 31 wt% <50 lm, and silicon oxide (SiO2 >99.7 wt%; Strem Chemicals
Ca3(PO4)2 and 69 wt% Ca2SiO4, and typical eutectoid micro- Inc, Newburyport, MA), with an average particle size
structure of lamellae morphology. These results are in dis- <60 lm. As a preliminary step, SiO2 was wet ground in a
agreement with the previous reported data, which locate the laboratory mixing miller (MM301-Retsch) using PSZ-zirco-
invariant eutectoid point at 500 ± 20°C with a composition of nia balls and isopropyl alcohol as suspension media. The
36.75 wt% Ca3(PO4)2 and 63.25 wt% Ca2SiO4. powder was then dried and sieved to <50 lm. The average
particle size of the final SiO2 used was 44.5 lm as measured
using laser diffraction (Mastersizer2000E device-Malvern).
I. Introduction
Next, the desired proportions of the constituents were

S INCE approximately the mid-1990s, silicon (Si) containing

and/or silicon-substituted calcium phosphates (Si–CaP)
have received increasing attention from researchers and, even
weighed out and thoroughly wet mixed in a mixing miller
with the same ZrO2 balls mentioned above as a grinding
media. After the milling process, the mixture was dried and
more recently, from clinicians.1–4 Silicon influences the bio- heated up to 1000°C for 24 h to remove the CO2. Then, this
logic activity of calcium phosphate (CaP) materials by modi- powder was cold isostatically pressed at 200 MPa and heat
fying material properties and by direct effects on the treated at 1525°C for 12 h, at a heating rate of 8.3°C/min,
physiologic processes in skeletal tissue.5–8 Compositions followed by a cooling rate of 5°C/min. The reaction-sintering
belonging to the system dicalcium silicate (Ca2SiO4 = C2S)- temperature was selected, bearing in mind, the information
tricalcium phosphate (Ca3(PO4)2 = TCP),9,10 able to release provided by the phase equilibrium diagram SiO2–CaO, evalu-
that element in situ, are promising candidates for preparing ated and reported by Eriksson et al.11 The obtained material
new ceramic bone implants. An exact knowledge of the phase was ground and characterized using XRD.
relationships in the system is required to design materials Mixtures with compositions show in Table I were selected
with controlled phase compositions and physical–chemical and and prepared for studying the sub-system C2S-A. First,
biologic properties. Thus, the sub-system 2CaOSiO2 (C2S)- TCP and C2S powders were ground to an average particle
7CaOP2O52SiO2 (A) was studied as a preliminary step toward size of 25 and 20 lm, respectively, and the desired propor-
obtaining new monophasic or biphasic bioceramics. Phase tions of each component were weighed in an analytical bal-
composition of the resulting ceramics was studied using X-Ray ance and thoroughly mixed in a manual agate mortar under
Diffraction (XRD) and Scanning Electron Microscopy (SEM) acetone. After drying, the samples were isostatically pressed
fitted with Wavelength Dispersive Spectrometer (WDS). in cylinders at 200 MPa. Pellets, obtained from the cylinders
were put into small platinum foil crucibles that were sus-
pended from a platinum wire in the hot zone of an electrical
II. Experimental Procedure
furnace with an electronic temperature controller (±1°C). All
The starting materials for the present study were TCP and the samples were heated up to 1550°C for a total period of
C2S previously synthesized. TCP was synthesized using solid time of 144 h (6 days), with quenching in liquid nitrogen,
state reaction from a stoichiometric mixture of calcium milling, pressing, and reheating every 24 h. A prolonged heat
hydrogen phosphate anhydrous (CaHPO4; Sigma, St. Louis, treatment was required to ensure that the equilibrium was
MO), and calcium carbonate (CaCO3; Sigma) with an aver- achieved. Finally, all the samples were studied using XRD
age particle size <50 lm. The mixture of CaHPO4 and and SEM–WDS to establish the high temperature region of
the subsystem. In addition, duplicate sample were heat trea-
ted in the same way, but cooled slowly inside the furnace up
P. W. Brown—contributing editor to the temperatures within 525°C–475°C range, which is
slightly above and below the eutectoid temperature, as
reported by Fix et al. (500 ± 20°C), and kept at these 24 h,
followed by quenching in liquid nitrogen.
Manuscript No. 29340. Received February 21, 2011; approved May 25, 2011. Next, the equilibrated compositions were removed from
**Fellow of the American Ceramic Society. In Memoriam: Prof. Salvador de Aza
passed away peacefully on 13 April 2011. Salvador was born on November 13, 1933. the platinum foils, embedded in an epoxy resin under vac-
He devoted his life to his family, friends, and his love of researching and teaching. As a uum, and progressively polished down to 0.1 lm diamond
researcher and educator, he gave his time, energy and love to his students and faculty
and research colleagues. His generosity and loving spirit will be missed by his family
paste and etched with acetic acid at a concentration of 0.5%
and friends. for 2 s. Then, they were gently cleaned in an ultrasonic bath
Part of this work was supported by CICYT under project no. MAT2006-12749- with distilled water, dried, and palladium coated for scanning
C02-02
†
Author to whom correspondence should be addressed. e-mail: [email protected]. electron microscope (SEM-Hitachi S-3500N, Ibaraki, Japan)
4459
4460 Journal of the American Ceramic Society—Rubio et al. Vol. 94, No. 12

Table I. Composition of the Samples observations. The chemical composition of crystalline grains
was qualitatively determined using a wavelength dispersive
Composition (wt%) spectroscopy (WDS) system coupled to the above-described
TCP C2S
electron microscope.
To evaluate the phase’s composition, XRD patterns were
3 97 obtained in a Bruker AXS D8-Advance X-ray Diffractometer
15 85 (Karksruhe, Germany) using kCuKa1 radiation (0.15418 nm)
20 80 and a secondary curved graphite monochromator. Data were
22 78 collected in the Bragg–Brentano (h/2h) vertical geometry (flat
25 75 reflection mode) between 20° and 50° (2h) at 0.05° steps,
30 70 counting 1.5 s per step. Samples were rotated at 30 rpm dur-
31 69 ing acquisition of patterns to improve averaging. The diffrac-
32 68 tometer optic was a system of primary Soller foils between
33 67 the X-ray tube and the fixed aperture slit. One scattered radi-
36.75 63.25 ation slit of 1 mm was placed after the sample, followed by a
45 55 system of secondary Soller slits, and a detector slit of
47 53 0.1 mm. The X-ray tube was operated at 40 kV at 30 mA.
Diffractograms of samples were compared with the data
provided by the Joint Committee on Powder Diffraction
Standards (JCPDS) database.

III. Results and Discussion

Figure 1 shows the X-ray diffraction patterns of the synthe-
sized powders of TCP and C2S. The obtained TCP correspond

Table II. Phases Identified Using XRD for Each

Composition After Treatment in Solid Stated at the Selected
Temperatures and Plateaus Times
Holding conditions

Sample wt Temperature Time General

% TCP (°C) (h) Phases present characteristic

Fig. 1. X-ray diffraction patterns of the synthesized materials: TCP 3 1550 144 Rss Solid, white
[all the peaks correspond to a-TCP] C2S [all the peaks correspond to 15 1550 144 Rss Solid, white
γ-C2S]. 15 525 24 a′C2Sss Solid, white
15 512 24 a′C2Sss Solid, white
20 512 24 a′C2Sss Solid, white
20 500 24 a′C2Sss Solid, white
22 512 24 a′C2Sss + Ass Solid, white
22 500 24 a′C2Sss + Ass Solid, white
25 525 24 a′C2Sss + Rss Solid, white
25 500 24 a′C2Sss + Ass Solid, white
30 550 24 a′C2Sss + Rss Solid, white
30 525 24 a′C2Sss + Rss Solid, white
30 512 24 a′C2Sss + Ass Solid, white
30 500 24 a′C2Sss + Ass Solid, white
31 1550 144 Rss Solid, white
31 550 24 Rss Solid, white
31 525 24 Rss Solid, white
31 512 24 a′C2Sss + Ass Solid, white
31 500 24 a′C2Sss + Ass Solid, white
31 475 24 a′C2Sss + Ass Solid, white
32 1550 144 Rss Solid, white
32 550 24 Rss + Ass Solid, white
32 525 24 Rss + Ass Solid, white
32 512 24 a′C2Sss + Ass Solid, white
32 500 24 a′C2Sss + Ass Solid, white
33 525 24 Rss + Ass Solid, white
33 500 24 a′C2Sss + Ass Solid, white
*36.75 1550 144 Rss Solid, white
*36.75 525 24 Rss + Ass Solid, white
*36.75 500 24 a′C2Sss + Ass Solid, white
*36.75 475 24 a′C2Sss + Ass Solid, white
45 1550 144 Rss Powder, white
45 512 24 a′C2Sss + Ass + Rss Powder, white
45 500 24 a′C2Sss + Ass + Rss Powder, white
Fig. 2. Redrawn of the sub-system C2S-A within the system C2S- 47 512 24 a′C2Sss + Ass + Rss Powder, white
TCP [C2Sss = dicalcium silicate solid solution; Ass = 7CaOP2O52SiO2
47 500 24 a′C2Sss + Ass + Rss Powder, white
solid solution; R = (a-C2S-a′TCP) solid solution; TCP = tricalcium
phosphate]. *, fix invariant point.
December 2011 The Sub-System Ca2SiO4–7CaOP2O52SiO2 4461

f (a)

(b)

Fig. 4. SEM images of sample 31 wt% TCP/69 wt% C2S after

heating at 500°C/24 h and quenched.

Fig. 5. SEM image and WDS microanalysis of the sample 31 wt%

TCP/69 wt% C2S after heating at 1550°C/144 h and quenched.

Fig. 3. X-ray diffraction patterns of samples 30 wt% TCP/70 wt% treatment at the selected temperatures and plateaus time are
C2S (a and b), 31 wt% TCP/69 wt% C2S (c and d) and 36.75 wt% summarized in Table II. The sub-system C2S-A that was
TCP/63.25 wt% C2S (e and f), after heating at 512°C for 24 h (a, c,
and e), and at 525°C for 24 h (b, d, and f) [ = a′C2S; = A; = R].
plotted from these results is shown in Fig. 2. The markings
in the diagram near the eutectoid point indicate the composi-
tions tested in this study.
to the a polymorph, according to the JCPDS card no. 09-0348, Figure 3 shows the X-ray patterns of samples 30 wt%
without any trace of other polymorphic forms. On the other TCP/70 wt% C2S, 31 wt% TCP/69 wt% C2S, and 36.75 wt
hand, the C2S correspond to JCPDS card no. 87-1257 that % TCP/63.25 wt% C2S after the samples were heat treated
belongs to the γ-C2S polymorphic form. of all of them at 512°C and at 525°C for 24 h. After the
The experimental results showing the phase coexisting heating at 512°C, slightly below the eutectoid point, all
in equilibrium for each composition after the solid state the samples were composed of a mixture of a′-C2Sss + Ass.
4462 Journal of the American Ceramic Society—Rubio et al. Vol. 94, No. 12

Table III. d values for the Six Strongest Lines of the Phase R Quenched at 1550°C After 144 h, Together
with the d values of the Pure a′-TCP (JCPDS Card no. 89-8960) and a-C2S (JCPDS Card no. 87-1260)
for Comparison Purposes. Compositions are in wt%
2CaO·SiO2 100 97 85 69 68 63.25 0
3CaO·P2O5 0 3 15 31 32 36.75 100
Temp. (ºC) 1525 1550 1550 1550 1550 1550 1447
d I d I d I d I d I d I d I
3.78 18 3.84 11 3.89 4 3.90 32 3.89 22 3.98 41
3.67 4 3.55 17 3.46 19 3.85 50
2.92 61 2.73 74 2.74 42 2.82 58 2.83 85 2.83 100 2.96 100
2.77 100 2.70 100 2.70 100 2.70 100 2.70 100 2.69 91 2.68 68
2.28 9 2.19 21 2.20 18 2.21 6 2.22 22 2.22 27 2.22 17
2.01 28 1.93 10 1.95 3 1.95 33 1.95 27 1.99 43

closed pores. WDS confirmed the presence of phosphorous,

calcium, and silicon.
The sample 45 wt% TCP/55 wt% C2S after heating at
1550°C/144 h (Fig. 6) produced powder composed of Rss,
Ass, and a′-C2Sss phases.

IV. Conclusions
The sub-system C2S-A within the binary system C2S-TCP,
presents an invariant eutectoid point at 518 ± 6°C with a
composition of 31 wt% TCP and 69 wt% C2S.
At the temperature of the invariant point, the maximum
solid solution of C2S in A is less than ~0.39 wt% and that of
TCP in C2S is ~21 wt%.
In the light of the new findings, new bioceramics with a
Fig. 6. XRD patterns of the sample 45 wt% TCP/55 wt% C2S after
eutectoid microstructure (lamellae morphology) can be
heating at 1550°C/144 h and quenched [ = a′C2S; = A; = R]. designed in the sub-system C2S-A using an appropriate heat
treatment.
A material with a eutectoid microstructure could be very
However, after the treatment at 525°C, slightly above the
useful in getting implants with a porosity suitable for bone
eutectoid point, samples containing 30 and 36.75 wt% TCP
growth without reducing the mechanical properties of the
were composed of R + a′-C2Sss and R + Ass, respectively,
material. SBF in vitro test could help to determine the bio-
whereas the 31 wt% TCP sample was made of one phase R.
activity–biodegradability of the eutectoid material.
These results indicate the location of the eutectoid point at
31 wt% TCP. As all the phases are solid solutions, the dif- References
fraction peaks are slightly displaced with respect to the corre-
1
sponding JCPDS cards. The displacement range for Ass S. D. Langstaff, M. Sayer, T. J. N. Smith, and S. M. Pugh, “Resorbable
phase was 0.3° and for a′-C2Sss phase was 0.1°. Bioceramics Based on Stabilized Calcium Phosphates”; “Part II: Evaluation of
Biological Response,” Biomaterials, 2, 5–50 (2001).
The microstructure in Fig. 4, which corresponds to the 2
N. Patel, I. R. Gibson, K. A. Hing, S. M. Best, P. A. Revell, and W. A.
sample 31 wt% TCP/69 wt% C2S, shows a typical eutectoid Bonfield, “A Comparative Study on the In Vivo Behaviour of Hydroxyapatite
composition of lamellae morphology confirming the location and Silicon Substituted Hydroxyapatite Granules,” J. Mater. Sci. Mater.
of the eutectoid point. The sample was made up of very thin Med., 13, 1199–206 (2002).
3
A. K. Lynn, T. Nakamura, N. Patel, A. E. Porter, A. C. Renouf, P. R.
eutectoid platelets that constituted of a′-C2Sss and Ass phases. Laity, S. M. Best, R. E. Cameron, Y. Shimizu, and W. Bonfield, “Composi-
The two phases were so intimately mixed and thin that was tion-Controlled Nanocomposites of Apatite and Collagen Incorporating Sili-
impossible to distinguish between Ass and a′-C2Sss by WDS. con as an Osseopromotive Agent,” J. Biomed. Mater. Res., 74A, 447–53
These results clearly disagree with those previously pub- (2005).
4
J. Huang, S. N. Jayasinghe, S. M. Best, M. J. Edirisinghe, R. A. Brooks,
lished by Fix et al.,10 who reported that the system shows an N. Rushton, and W. Bonfield, “Novel Deposition of Nano-Sized Silicon
invariant eutectoid point at 500 ± 20°C with a composition Substituted Hydroxyapatite by Electrostatic Spraying,” J. Mater. Sci. Mater.
of 36.75 wt% TCP and 63.25 wt% C2S. Med., 16, 1137–42 (2005).
5
In the composition range between 3 and 44 wt% of TCP, A. M. Pietak and M. Sayer, “Functional Atomic Force Microscopy Investi-
gation of Osteopontin Affinity for Silicon Stabilized Tricalcium Phosphate
the high temperature phase R was obtained at room temper- Bioceramic Surfaces,” Biomaterials, 27, 3–14 (2006).
ature as a mestastable phase. The relevant XRD patterns 6
S. V. Dorozhkin, “In Vitro Mineralization of Silicon Containing Calcium
were essentially of the same type as the XRD spectrum that Phosphate Bioceramics,” J. Am. Ceram. Soc., 90 [1] 244–9 (2007).
7
is shown in Fig. 3(d). They are characterized by two strong A. M. Minarelli-Gaspar, S. Saska, R. G. Carrodeguas, A. H. De Aza,
P. Pena, P. N. De Aza, and S. De Aza, “Biological Response to Wollastonite
diffraction lines whose separation decreases systematically as Doped a-Tricalcium Phosphate Implants in Hard and Soft Tissues in Rats,”
the solid solution R become richer in C2S. The d values for Key. Eng. Mater., 396–8, 7–10 (2009).
8
the six strongest lines of the phase R, recorded from the sam- I. M. Martinez, P. A. Velasquez, and P. N. De Aza, “Synthesis and Stabil-
ples quenched from 1550°C after 144 h, are given in ity of a-Tricalcium Phosphate Doped with Dicalcium Silicate in the System
Ca3(PO4)2-Ca2SiO4,” Mater. Charact., 61, 761–7 (2010).
Table III together with the d values of the pure a′-TCP 9
R. W. Nurse, J. H. Welch, and W. Gutt, “High-Temperature Equilibria in
(JCPDS card no. 89-8960) and a-C2S (JCPDS card no. 87- the System Dicalcium Silicate-Tricalcium Phosphate,” J. Chem. Soc., 1077–83
1260) for comparison purposes. (1959).
10
Figure 5 shows the microstructure of the sample 31 wt% W. Fix, H. Heymann, and R. Heinke, “Subsolidus Relations in the System
2CaO·SiO2-3CaO·P2O5,” J. Am. Ceram. Soc., 52 [6] 346–7 (1969).
TCP/69 wt% C2S after heating at 1550°C/144 h with 11
G. Eriksson, P. Wu, M. Blander, and A. D. Pelton, “Critical Evaluation
quenching, milling, pressing, and reheating every 24 h. The and Optimisation of the Thermodynamic Properties and Phase Diagrams of
material presents only one phase with irregular grains and the MnO-SiO2 and CaO–SiO2 Systems,” Can. Metall. Q, 33, 13–21 (1994). h