Materials Science and Engineering C 47 (2015) 407–412
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
A new sol–gel synthesis of 45S5 bioactive glass using an organic acid as catalyst J. Faure a,⁎, R. Drevet a,⁎, A. Lemelle a, N. Ben Jaber a, A. Tara a, H. El Btaouri b, H. Benhayoune a a b
Université de Reims Champagne-Ardenne, Laboratoire Ingénierie et Sciences des Matériaux, LISM EA 4695, 21 rue Clément ADER, 51685 REIMS Cedex 2, France Université de Reims Champagne-Ardenne UMR CNRS MEDyC, EA 7369, Campus Moulin de la Housse, 51687 REIMS Cedex 2, France
a r t i c l e
i n f o
Article history: Received 5 July 2014 Received in revised form 10 October 2014 Accepted 11 November 2014 Available online 13 November 2014 Keywords: Bioactive glass 45S5 Sol–gel Citric acid In vitro tests Bioactivity
a b s t r a c t In this paper a new sol–gel approach was explored for the synthesis of the 45S5 bioactive glass. We demonstrate that citric acid can be used instead of the usual nitric acid to catalyze the sol–gel reactions. The substitution of nitric acid by citric acid allows to reduce strongly the concentration of the acid solution necessary to catalyze the hydrolysis of silicon and phosphorus alkoxides. Two sol–gel powders with chemical compositions very close to that of the 45S5 were obtained by using either a 2 M nitric acid solution or either a 5 mM citric acid solution. These powders were characterized and compared to the commercial Bioglass®. The surface properties of the two bioglass powders were assessed by scanning electron microscopy (SEM) and by Brunauer–Emmett–Teller method (BET). The Fourier transformed infrared spectroscopy (FTIR) and the X-ray diffraction (XRD) revealed a partial crystallization associated to the formation of crystalline phases on the two sol–gel powders. The in vitro bioactivity was then studied at the key times during the first hours of immersion into acellular Simulated Body Fluid (SBF). After 4 h immersion into SBF we clearly demonstrate that the bioactivity level of the two sol–gel powders is similar and much higher than that of the commercial Bioglass®. This bioactivity improvement is associated to the increase of the porosity and the specific surface area of the powders synthesized by the sol–gel process. Moreover, the nitric acid is efficiently substituted by the citric acid to catalyze the sol–gel reactions without alteration of the bioactivity of the 45S5 bioactive glass. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Bioactive glasses are materials with osteoconductive and osteoproductive properties which confer them the ability to repair and replace diseased or damaged bone [1,2]. These properties result from their progressive dissolution in physiological medium, where the release of calcium, phosphate, and sodium ions will undergo the formation of an apatite layer that in turn will create a strong bond with the surrounding bone tissues [2–4]. Several compositions of bioactive glasses present a great interest such as sodium calcium phosphosilicate, borosilicates or 13–93 coumpounds [5,6]. Among them Larry Hench firstly developed the famous 45S5 Bioglass®, in the quaternary system SiO2–CaO–Na2O–P2O5, with an oxide composition that allows it to bond both to hard and soft tissues (class A bioactivity). Therefore it has been used since decades in many medical devices employed for orthopedic and dental treatments [7,8]. Traditionally this commercial bioactive glass is produced by the conventional melting-quenching process which requires very high temperatures and greatly limits the porosity and specific surface of the powder finally obtained. Meanwhile, ⁎ Corresponding authors. E-mail addresses: [email protected] (J. Faure), [email protected] (R. Drevet).
http://dx.doi.org/10.1016/j.msec.2014.11.045 0928-4931/© 2014 Elsevier B.V. All rights reserved.
increasing research efforts are being invested to develop the sol–gel process which leads to bioactive glass powders of high purity and homogeneity. The sol–gel process also significantly expands the range of the chemical composition of bioactive glasses as compared to the traditional melting-quenching method [9,10]. These advantages make the sol–gel method very efficient for the synthesis of the 45S5 bioactive glass. Even though the sol–gel process is said to proceed under mild conditions, most protocols advocate the use of strong acid (nitric acid and hydrochloric acid principally) or bases (ammonium hydroxide) at elevated concentrations (up to 1 M) to catalyze the sol–gel reactions [11]. Therefore it is suitable to find ways to make the process cleaner avoiding the usually extreme pH or high temperature conditions of most protocols. In this context, several recent papers report on the synthesis of bioactive glass nanoparticles using different organic acids at very low concentrations (typically in the range of 0.01 M) to catalyze the chemical reactions involved during the process [12–15]. In these works, citric acid solutions are commonly used to catalyze the hydrolysis of the silicon alkoxide (TEOS: tetraethylorthosilicate). This idea comes from finding inspiration in nature, therefore it is considered to be a biomimetic inspiration. Indeed, an interesting advancement of the sol–gel chemistry is the understanding of the biosilicification in diatoms and glass sponges [16,17]. These organisms accumulate silicic acid Si(OH)4 from their environment and, through the actions of proteins
408
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412
Table 1 Experimental conditions of the Sol–gel process. SiO2 precursor Na2O precursor CaO precursor P2O5 precursor Reaction temperature Catalyst
Tetraethylorthosilicate (TEOS): Si(OC2H5)4 Sodium nitrate: NaNO3 Calcium nitrate tetrahydrate: Ca(NO3)2·4H2O Triethylphosphate (TEP): PO(C2H5)3 20 °C; 35 °C; 50 °C Nitric acid (HNO3): 0.5 M and 2 M Citric acid (C6H8O7): 0.5 mM, 5 mM and 50 mM
and other macromolecules, form complex silica structures [17,18]. The elucidation of the mechanisms underlying the biosilicification mechanisms has led to the in vitro synthesis of silica under physiological conditions using proteins, polypeptides, amino acids and mainly citric acid to catalyze the sol–gel reactions [19–21]. Considering the specific composition of bioactive glasses and in particular its calcium and phosphate content, it may be necessary to draw inspiration from bone morphogenesis. Consequently, citric acid is here used to trigger the sol–gel reactions. Thus the main objective of this work concerns the ability to succeed the synthesis of the sol–gel derived 45S5 bioactive glass by using a low concentrated citric acid solution to catalyze the hydrolysis reaction. Firstly, we determine the optimal sol–gel parameters to reach the 45S5 chemical composition by using either the conventional nitric acid or either the innovative citric acid as catalyst. In the second part we compare the physico-chemical properties and the in vitro bioactivity of the two optimal sol–gel powders with that of the commercial 45S5 Bioglass®. 2. Materials and methods 2.1. Sol–gel synthesis The sol–gel synthesis of the bioactive glasses proceeded according to a conventional protocol for the synthesis of 45S5 bioglass [22,23] using the following chemical precursors tetraethyl orthosilicate Si(OC2H5)4 (TEOS), triethyl phosphate PO(C2H5)3 (TEP), calcium nitrate tetrahydrate Ca(NO3)2,4H2O and sodium nitrate NaNO3. Two different aqueous acid solutions prepared with nitric acid (HNO3) or with citric acid (C6H8O7) were used to catalyze the hydrolysis reaction. The hydrolysis and condensation reactions were performed within a thermostated reactor to control the reaction temperature. The different elaboration parameters tested in this work are listed in Table 1 and the flowchart of the sol–gel 45S5 fabrication process is presented in Fig. 1. The molar ratio of TEOS, TEP, NaNO3 and Ca(NO3)2·4H2O were designed according to the molar ratio of SiO2, P2O5, Na2O and CaO in 45S5. To achieve a clear sol the molar ratio between the aqueous acid solution and the four chemical precursors was set to 10. Firstly, the acidic solution (26 mL) was magnetically stirred
Preparation of aqueous acid solutions
in the thermostated reactor at the desired temperature and TEOS (11.6 mL) and TEP (1 mL) were added dropwise to the solution and stirred until a clear solution was obtained. Next, the NaNO3 powder (4.66 g) was slowly added in the stirred solution until its complete dissolution. Finally the Ca(NO3)2,4H2O powder (7.15 g) was added slowly to the solution stirred during 1 h to result in a transparent sol. Therefore the sol starts to transform into gel through polycondensation reactions. The gel was then kept at 60 °C for 12 h and finally dried at 200 °C and 700 °C for 5 h and 2 h respectively. After the drying step the gel was manually crushed to obtain a fine powder. Pellets of the obtained powders were then prepared with a manual press for the characterization of the elemental composition and for the bioactivity tests. In order to observe the specific properties of the two bioglasses synthesized by the sol–gel process, they were systematically compared to commercial Bioglass® synthesized by the classical melting-quenching process. 2.2. Physico-chemical and structural characterizations 2.2.1. Scanning electron microscopy and X-ray microanalysis (SEM–EDXS) The morphology of the synthesized powders was observed by SEM using a LaB6 electron microscope (JEOL JSM-5400LV) operating at 0–30 kV. This microscope is associated to an energy dispersive EDAX spectrometer equipped with an ultra-thin window Si(Li) detector cooled with liquid nitrogen. This spectrometer is associated with an acquisition and quantification system GENESIS (Eloïse SARL, France) based on the ZAF method used to determine the elemental concentrations of the analyzed sample [24,25]. All the X-ray microanalysis results presented in this work were obtained according to the following method: acquisition and quantification of four spectra on four flat areas of the pellet selected by SEM. All the X-ray spectra were obtained with energy of 15 keV and an acquisition time of 100 s. The final values of elemental concentrations were obtained after calculation of mean values and dispersions. 2.2.2. Fourier transformed infrared spectroscopy (FTIR) Structural characterizations of the powders were performed by FTIR spectroscopy. The FTIR spectra were obtained in reflexion mode with a FTIR imaging system (Spotlight, Perkin–Elmer, Courtaboeuf, France) coupled to a spectrometer (Spectrum 300, Perkin–Elmer) in the 400– 2000 cm−1 range with a spectral resolution of 4 cm−1. 2.2.3. X-ray diffraction (XRD) The crystallinity of the powders was studied by XRD with a Bruker D8 Advance diffractometer using a monochromatic copper radiation (CuKα) of wavelength λ = 0.15406 nm. The diffractograms were recorded in steps of 0.04° with an acquisition time of 12 s in a range of
Addition of TEOS and TEP
Stirring
Clear sol
Addition of NaNO3
Dissolution
Powder
12 h at 60°C 5 h at 200°C 2 h at 700°C
Drying Gel
Fig. 1. Flowchart of the sol–gel process.
Aging
Addition of Ca(NO3)2
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412 Table 2 Ionic concentration of human blood plasma and SBF solution used for the evaluation of in vitro bioactivity. Ionic concentration (mmol·L−1)
Human blood plasma Simulated body fluid (SBF)
Na+
K+
Mg2+
Ca2+ Cl−
HCO− 3
HPO2− 4
SO2− 4
142.0 142.0
5.0 5.0
1.5 1.5
2.5 2.5
27.0 4.2
1.0 1.0
0.5 0.5
103.0 147.8
was analyzed by SEM–EDXS for morphological and chemical composition analysis. 3. Results and discussion 3.1. Optimization of the experimental conditions of the sol–gel process
diffraction angles 2θ between 10 and 60°. The crystalline phases were then identified by the powder diffraction files of the International Centre for Diffraction Data (ICDD). 2.2.4. Determination of the specific surface area by BET method Nitrogen adsorption measurements were carried out using a Micromeritics ASAP 2010 instrument at 77 K. The Brunauer–Emmett– Teller (BET) method was applied to derive the surface area from physisorption isotherm data. The isotherm was constructed point-bypoint by the admission and withdrawal of known amounts of gas, with adequate time allowed for equilibration at each relative pressure (P/P0, where P is the equilibrium vapor pressure and P0 is the saturation vapor pressure). The BET method is based on the determination of the monolayer capacity, i.e. the nitrogen amount corresponding to the adsorption of a complete monolayer (Vmono). Before analysis, the samples were maintained overnight at 170 °C in an oven. Prior to adsorption, they were out-gassed under vacuum at 250 °C for 1 h. The surface area is determined from a set of 10 experimental points of the linear range of the BET plot (0.05 b P/P0 b 0.3). The surface area (SBET) is calculated according to:
SBET ¼
409
Vmono NA A VM
where Vmono is the monolayer capacity (cm3·g−1) at T = 273 K and P = 101,325 Pa, NA is the Avogadro constant, A is the molecular crosssectional area (for nitrogen, A = 0.162 nm2 at 77 K) and VM is the molar volume (22414 cm3·mol−1). 2.3. In vitro bioactivity tests To evaluate the bioactivity of the synthesized powders, in vitro tests were performed in Simulated Body Fluid (SBF). This solution is acellular, protein free with a pH of 7.4 buffered with Tris–hydroxymethylaminomethane (TRIS). Table 2 shows the composition of SBF compared to human blood plasma. In vitro bioactivity tests were performed by soaking a pellet into SBF for two periods (1 h and 4 h) at 37 °C that correspond to the key times of bioactivity for bioactive glasses. The volume of SBF used during the test is calculated according to the procedure described by Kokubo and Takadama [26]: SBF volume (mL) = 10% pellet mass (mg). After the bioactivity test the pellet was extracted from the solution and immersed into acetone for 10 s to remove SBF and stop surface reactions. After drying, the pellet surface
In the first part of this work we aim to determine the optimal experimental conditions to reach the 45S5 composition by varying parameters such as reaction temperature and concentration of the aqueous acid solutions used to catalyze the sol–gel reactions. Nitric acid has been tested as a catalyst by combining three temperatures (20 °C, 35 °C and 50 °C) and two concentrations (0.5 M and 2 M). Table 3 presents the results obtained by quantitative EDXS that provides the elemental compositions of the powders synthesized with these process parameters. The optimal parameters, which lead to the best composition very close to the theoretical one, are observed at 35 °C and 2 M (this sol–gel powder was named sol–gel1). The citric acid has been also tested as a catalyst by combining two temperatures (20 °C and 35 °C) and three concentrations (0.5 mM, 5 mM and 50 mM). Table 4 presents the results obtained by quantitative EDXS that provides the elemental compositions of the powders synthesized with these experimental parameters. In this case, the optimal parameters that provide the best chemical composition are observed at 20 °C and 5 mM (this sol–gel powder was named sol–gel2). The comparison of the two synthesized powders (sol–gel1 and sol–gel2) shows that a lack of silicon and phosphorus is observed in both cases that can be attributed to the incomplete hydrolysis of the precursors and to the incomplete condensation due to unfavorable pH conditions. We can also observe a slight improvement of the chemical composition of the bioglass by using citric acid (sol–gel2). This behavior is probably due to the ability of citric acid to chelate calcium and sodium ions with its three carboxylic groups [13, 14,27]. Therefore we can conclude that the use of citric acid instead of nitric acid to catalyze the hydrolysis of TEOS and TEP leads to the sol– gel synthesis of the 45S5 bioactive glass with the expected chemical composition. 3.2. Structural characterization of the bioglasses The FTIR spectra of the two synthesized sol–gel glasses are presented in Fig. 2 and compared with the spectrum of the commercial Bioglass®. Both spectra of sol–gel1 and sol–gel2 reveal vibration bands similar to that of Bioglass®. The two broad bands at 930 cm−1 and 1039 cm−1 correspond to silicate adsorption bands which are respectively Si–O–Si stretching of non-bridging oxygen atoms and Si–O–Si asymmetric stretching of bridging oxygen atoms within the silicate tetrahedron. Silicate bands for Si–O–Si bending mode are also observable at 478 cm−1 and 503 cm−1 [28,29]. The peak at around 602 cm−1 and 880 cm−1 are attributed to P–O bending of PO3− 4 groups [28,29]. The weak bands observed at around 1450 cm−1 are related to the presence of residual carbonate groups from the precursors [29]. Thus, the FTIR spectra obtained from sol–gel1 and sol–gel2 are very similar and clearly reveal the glassy structure of the two sol–gel powders. In addition the comparison with
Table 3 Elemental compositions of sol–gel powders synthesized with nitric acid as catalyst (the bold line correspond to the optimal powder named sol-gel1).
T = 20 °C T = 35 °C T = 50 °C
Element (at.%)
O
Si
Na
Ca
P
Theoretical
55.3
16.3
17.1
9.5
1.8
0.5 M 2M 0.5 M 2M 0.5 M 2M
58.2 56.5 55.4 56.7 56.4 55.6
± ± ± ± ± ±
0.3 0.5 0.4 0.3 0.2 0.5
11.7 13.7 14.0 15.2 13.4 14.5
± ± ± ± ± ±
0.3 0.2 0.3 0.2 0.1 0.3
15.3 13.5 14.5 15.3 14.0 15.6
± ± ± ± ± ±
0.2 0.3 0.2 0.3 0.3 0.6
13.6 14.8 14.5 11.6 14.8 13.0
± ± ± ± ± ±
0.2 0.4 0.3 0.3 0.3 0.2
1.2 1.5 1.6 1.2 1.4 1.3
± ± ± ± ± ±
0.2 0.2 0.2 0.1 0.2 0.2
410
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412
Table 4 Elemental compositions of sol–gel powders synthesized with citric acid as catalyst (the bold line correspond to the optimal powder named sol-gel2).
T = 20 °C
T = 35 °C
Element (at.%)
O
Si
Na
Ca
P
Theoretical
55.3
16.3
17.1
9.5
1.8
0.5 mM 5 mM 50 mM 0.5 mM 5 mM 50 mM
55.8 54.5 55.4 55.8 54.2 55.6
± ± ± ± ± ±
0.3 0.5 0.3 0.3 0.3 0.2
14.1 15.9 14.0 13.4 15.5 13.5
the commercial Bioglass® indicates that a partial crystallization of the two sol–gel glasses occurred during the process. The structural characterization has been completed with XRD studies. The Bioglass® powder appears to be essentially amorphous due to the melting–quenching synthesis process. However the XRD patterns reveal few crystalline peaks probably due to the speed of the quenching step [30]. We can also observe that the diffractograms of sol–gel1 and sol–gel2 are very similar (Fig. 3) with a more important crystallization than that of Bioglass®. Indeed there are several crystalline sodium calcium silicate phases inside the amorphous structure of the bioactive glass. Among them, combeite Na2Ca2Si3O9 (ICDD PDF #22.1455) is the most interesting one as it is known to influence the bioactivity. The formation of the combeite phase is associated to the heat treatment performed at 700 °C during the drying step used for sol–gel synthesis [31]. An accurate comparison of the diffractograms reveals that the formation of the Na2Ca2Si3O9 phase is more pronounced in sol–gel2. This observation is characteristic of the bioactive glasses synthesized by the sol–gel process in comparison with those synthesized by the usual melting– quenching process [32]. This may impact the behavior of the bioactive glasses in contact with physiological environment as the chemical properties of the material, such as solubility, are then modified [30–32].
± ± ± ± ± ±
0.3 0.3 0.3 0.2 0.2 0.1
12.5 16.7 15.4 13.8 17.2 14.4
± ± ± ± ± ±
0.3 0.3 0.3 0.4 0.2 0.8
16.3 11.3 14.1 15.6 11.5 14.8
± ± ± ± ± ±
0.6 0.4 0.2 0.6 0.3 0.7
1.3 1.6 1.1 1.4 1.6 1.7
± ± ± ± ± ±
0.2 0.2 0.1 0.2 0.2 0.2
provides very interesting information. The grain surface of the two sol– gel powders is highly rough and porous whereas for the Bioglass® powder the grain surface appears to be extremely smooth. In addition the roughness and porosity observed on the grain surface of sol–gel2 is higher than that of sol–gel1. These SEM observations can be related to the results obtained from the BET method which indicate that the specific surface area of the two sol–gel powders is higher than that of the Bioglass® powder, particularly for sol–gel2. The very low roughness and porosity of the Bioglass® powder is due to the melting–quenching process that uses an extremely quick cooling of the liquid phase during the quenching step [30]. However, the original observation highlighted in the present work concerns the specific effect of citric acid which increases the roughness and the porosity of the sol–gel powder. Indeed, the surface of the sol–gel2 grain powder is highly porous and made of nanoscale cylinders. This specific observation indicates that the use of citric acid as a catalyst of the hydrolysis reaction modifies the surface morphology of the grain powder at a nanometer scale. Obviously, these results highlight differences between the exchange surfaces of the powders that could lead to various behaviors in physiological environment.
3.4. In vitro bioactivity test of the bioglasses 3.3. Characterization of the specific surface area of the bioglasses The results obtained from the BET method indicate that the specific surface area of the materials clearly depends on the elaboration process and also on the experimental conditions used to produce the powders. Indeed, the specific surface area of sol–gel1, sol–gel2 and Bioglass® powders are respectively 0.6 m2·g− 1, 0.9 m2·g− 1 and 0.4 m2·g− 1. The SEM micrographs of Fig. 4 show that the three powders are constituted by grains with diverse sizes. The biggest grains are about 100 μm in size and the smallest ones are less than 10 μm. The sol–gel2 powder seems to get more small grains than the sol–gel1 powder which itself seems to have more small grains than the Bioglass® powder. The high magnification micrographs carried out on one big grain of each powder
SEM micrographs and EDXS spectra of the samples obtained from Bioglass®, sol–gel1 and sol–gel2 powders before and after immersion in SBF solution are shown in Fig. 5. After 1 h of immersion, the SEM micrographs reveal that no significant change occurred on the three samples. After 4 h of immersion significant changes of the surface morphology and of the chemical composition are observed. For the Bioglass® sample a slight dissociation of the pellet surface during immersion in the SBF solution can be observed without any evidence of its mineralization. The associated EDXS spectrum demonstrates that the main change in the surface chemical composition is due to a significant decrease in sodium concentration indicating that the bioactivity reactions are just starting during the test [1]. For the sol–gel1 and sol–gel2 samples, the SEM micrographs highlight an important mineralization of their surfaces after
Fig. 2. FTIR spectra of the sol–gel glasses and of the commercial Bioglass®.
Fig. 3. X-ray diffractograms of the sol–gel glasses and of the commercial Bioglass®.
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412
411
Fig. 4. SEM micrographs of the sol–gel glasses and of the commercial Bioglass® (low and high magnification).
exposure to SBF associated to the formation of a calcium phosphate deposit [26,32]. This surface mineralization is confirmed by the EDXS spectra which clearly show a strong increase of the intensities of the peaks of calcium and phosphorus. Thus, the in vitro bioactivity test of the two sol–gel samples clearly indicates that the bioactivity level of the powders obtained by the sol–gel synthesis is much higher than that of the commercial Bioglass®. It is usually established that the combeite phase decreases the bioactivity reaction kinetic of the bioactive glasses when they are immersed in physiological environment by transforming it into amorphous calcium phosphate [31]. However, in this work, the bioactivity of the sol–gel glasses appears to be higher than the bioactivity of the Bioglass®. Indeed, the grains of sol–gel glasses reveal highly rough surfaces and porosity that provide high exchange surface in physiological environment. Thus the exchange surface of the sol–gel bioglass is more important than the exchange surface of the Bioglass® synthesized by the melting–quenching process. Indeed, the BET measurements confirmed that the specific surface area of the two sol–gel glasses is higher than that of the Bioglass® [33,34]. On the other hand, the comparison between sol–gel1 and sol–gel2 indicates that the bioactivity level of the two sol–gel powders seems to be very similar. Therefore the use of citric acid instead of nitric acid as catalyst during sol–gel synthesis allows to obtain the bioactive glass 45S5 without alteration of its bioactivity.
4. Conclusions In this work we demonstrate that the 45S5 bioactive glass powder can be synthesized by the sol–gel process with a very low concentrated citric acid solution instead the usual highly concentrated nitric acid solution to catalyze the hydrolysis reaction. Firstly, we determine the optimal synthesis experimental conditions to reach the 45S5 elemental composition by varying two parameters, the reaction temperature and the concentration of the acid solutions. The best chemical composition of the bioactive glass has been obtained either with nitric acid at 2 M and 35 °C or either with citric acid at 5 mM and room temperature. The use of citric acid as catalyst for the sol–gel synthesis of a 45S5 powder extremely decreases the acid solution concentration necessary to catalyze the hydrolysis reactions of TEOS and TEP. We then compare the two optimal sol–gel powders with the commercial Bioglass®. The structural characterizations highlight a partial crystallization of the amorphous sol–gel powders which are made of several sodium calcium silicate phases including the combeite in particular. The formation of this highly bioactive crystal phase is associated to the last drying step at 700 °C during the sol–gel process. Finally we compare the in vitro bioactivity of the powders by using immersion tests into SBF solution. We clearly demonstrate that the bioactivity level of the two sol–gel
412
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412
Fig. 5. SEM–EDXS analysis of the sol–gel glasses and of the commercial Bioglass® as a function of the immersion time into SBF.
powders is much higher than that of the commercial Bioglass®. Moreover, the bioactivity of the bioactive glasses is mainly influenced by the specific surface area of the synthesized powder rather than the presence of the combeite crystalline phase inside the glass. Thus the use of citric acid instead of nitric acid as catalyst during the sol–gel synthesis allows to obtain the bioactive glass 45S5 without altering its bioactivity. The development of this new process is very attractive as it uses conditions suitable to add and encapsulate biomolecules (proteins, growth factor, drugs, etc.) that could help speed up the bone growth formation which is not only possible in most protocols owing to the usually extreme pH or high temperature conditions. A perspective of this work will then consist in the study of longer immersion time in physiological medium and observations of cell behavior in contact with these prosthetic surfaces.
[6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
Acknowledgment The authors sincerely thank Professor Aldo R. Boccaccini from the University of Erlangen-Nuremberg and the Imperial College of London for kindly providing the commercial Bioglass® studied in this work. They also express their gratitude to Professor Hassan Oudadesse from the University of Rennes 1 for the BET measurements.
[23] [24] [25] [26] [27] [28] [29]
References [1] [2] [3] [4] [5]
L.L. Hench, J. Am. Ceram. Soc. 81 (1998) 1705–1728. V.J. Shirtliff, L.L. Hench, J. Mater. Sci. 38 (2003) 4697–4707. L.L. Hench, Ceram. Int. 22 (1996) 493–507. L.L. Hench, J. Eur. Ceram. Soc. 29 (2009) 1257–1265. Y. Gu, G. Wang, X. Zhang, Y. Zhang, C. Zhang, X. Liu, M.N. Rahaman, W. Huang, H. Pan, Mater. Sci. Eng. C 36 (2014) 294–300.
[30] [31] [32] [33] [34]
X. Liu, M.N. Rahaman, G.E. Hilmas, B.S. Bal, Acta Biomater. 9 (2013) 7025–7034. L.L. Hench, J. Mater. Sci. Mater. Med. 17 (2006) 967–978. J.R. Jones, E. Gentleman, J. Polak, Elements 3 (2007) 393–399. J. Zhong, D.C. Greenspan, J. Biomed. Mater. Res. 53 (2000) 694–701. D. Arcos, M. Vallet-Regi, Acta Biomater. 6 (2010) 2874–2888. D. Avnir, T. Coradin, O. Lev, J. Livage, J. Mater. Chem. 16 (2006) 1013–1030. B. Lei, X. Chen, Y.H. Koh, J. Sol-Gel Sci. Technol. 58 (2011) 656–663. G.M. Luz, J.F. Mano, Nanotechnology 22 (2011) 494014. Z. Hong, A. Liu, L. Chen, X. Chen, X. Jing, J. Non-Cryst. Solids 335 (2009) 368–372. B. Lei, X. Chen, Y. Wang, N. Zhao, C. Du, L. Fang, Biomed. Mater. 5 (2010) 054103. D.J. Belton, O. Deschaume, C.C. Perry, FEBS J. 279 (2012) 1710–1720. S.V. Patwardhan, Chem. Commun. 47 (2011) 7567–7582. H.C. Schröder, X. Wang, W. Tremel, H. Ushijima, W.E. Müller, Nat. Prod. Rep. 25 (2008) 455–474. T. Coradin, J. Livage, Colloids Surf. B: Biointerfaces 21 (2001) 329–336. D. Belton, G. Paine, S.V. Patwardhan, C.C. Perry, J. Mater. Chem. 14 (2004) 2231–2241. T. Mizutani, H. Nagase, N. Fujiwara, H. Ogoshi, Bull. Chem. Soc. Jpn. 71 (1998) 2017–2022. I. Cacciotti, M. Lombardi, A. Bianco, A. Ravaglioli, L. Montanaro, J. Mater. Sci. Mater. Med. 23 (2012) 1849–1866. H. Pirayesh, J.A. Nychka, J. Am. Ceram. Soc. 96 (2013) 1643–1650. N. Dumelie, H. Benhayoune, G. Balossier, J. Phys. D. Appl. Phys. 40 (2007) 2124–2131. H. Benhayoune, J. Phys. D. Appl. Phys. 35 (2002) 1526–1531. T. Kokubo, H. Takadama, Biomaterials 27 (2006) 2907–2915. P.B. Wagh, A. Venkateswara Rao, D. Haranath, J. Porous. Mater. 4 (1997) 295–301. H.A. Elbatal, M.A. Azooz, E.M.A. Khalil, A. Soltan Monem, Y.M. Hamdy, Mater. Chem. Phys. 80 (2003) 599–609. A. Lucas-Girot, F.Z. Mezahi, M. Mami, H. Oudadesse, A. Harabi, M. le Floch, J. NonCryst. Solids 357 (2011) 3322–3327. L. Lefebvre, J. Chevalier, L. Grémillard, R. Zenati, G. Thollet, D. Bernache-Assollant, A. Govin, Acta Mater. 55 (2007) 3305–3313. Q.Z. Chen, I.D. Thompson, A.R. Boccaccini, Biomaterials 27 (2006) 2414–2425. R.L. Siqueira, O. Peitl, E.D. Zanotto, Mater. Sci. Eng. C 31 (2011) 983–991. I. Izquierdo-Barba, A.J. Salinas, M. Vallet-Regi, J. Biomed. Mater. Res. 47 (1999) 243–250. R. Li, A.E. Clark, L.L. Hench, J. Appl. Biomater. 2 (1991) 231–239.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
A new sol–gel synthesis of 45S5 bioactive glass using an organic acid as catalyst J. Faure a,⁎, R. Drevet a,⁎, A. Lemelle a, N. Ben Jaber a, A. Tara a, H. El Btaouri b, H. Benhayoune a a b
Université de Reims Champagne-Ardenne, Laboratoire Ingénierie et Sciences des Matériaux, LISM EA 4695, 21 rue Clément ADER, 51685 REIMS Cedex 2, France Université de Reims Champagne-Ardenne UMR CNRS MEDyC, EA 7369, Campus Moulin de la Housse, 51687 REIMS Cedex 2, France
a r t i c l e
i n f o
Article history: Received 5 July 2014 Received in revised form 10 October 2014 Accepted 11 November 2014 Available online 13 November 2014 Keywords: Bioactive glass 45S5 Sol–gel Citric acid In vitro tests Bioactivity
a b s t r a c t In this paper a new sol–gel approach was explored for the synthesis of the 45S5 bioactive glass. We demonstrate that citric acid can be used instead of the usual nitric acid to catalyze the sol–gel reactions. The substitution of nitric acid by citric acid allows to reduce strongly the concentration of the acid solution necessary to catalyze the hydrolysis of silicon and phosphorus alkoxides. Two sol–gel powders with chemical compositions very close to that of the 45S5 were obtained by using either a 2 M nitric acid solution or either a 5 mM citric acid solution. These powders were characterized and compared to the commercial Bioglass®. The surface properties of the two bioglass powders were assessed by scanning electron microscopy (SEM) and by Brunauer–Emmett–Teller method (BET). The Fourier transformed infrared spectroscopy (FTIR) and the X-ray diffraction (XRD) revealed a partial crystallization associated to the formation of crystalline phases on the two sol–gel powders. The in vitro bioactivity was then studied at the key times during the first hours of immersion into acellular Simulated Body Fluid (SBF). After 4 h immersion into SBF we clearly demonstrate that the bioactivity level of the two sol–gel powders is similar and much higher than that of the commercial Bioglass®. This bioactivity improvement is associated to the increase of the porosity and the specific surface area of the powders synthesized by the sol–gel process. Moreover, the nitric acid is efficiently substituted by the citric acid to catalyze the sol–gel reactions without alteration of the bioactivity of the 45S5 bioactive glass. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Bioactive glasses are materials with osteoconductive and osteoproductive properties which confer them the ability to repair and replace diseased or damaged bone [1,2]. These properties result from their progressive dissolution in physiological medium, where the release of calcium, phosphate, and sodium ions will undergo the formation of an apatite layer that in turn will create a strong bond with the surrounding bone tissues [2–4]. Several compositions of bioactive glasses present a great interest such as sodium calcium phosphosilicate, borosilicates or 13–93 coumpounds [5,6]. Among them Larry Hench firstly developed the famous 45S5 Bioglass®, in the quaternary system SiO2–CaO–Na2O–P2O5, with an oxide composition that allows it to bond both to hard and soft tissues (class A bioactivity). Therefore it has been used since decades in many medical devices employed for orthopedic and dental treatments [7,8]. Traditionally this commercial bioactive glass is produced by the conventional melting-quenching process which requires very high temperatures and greatly limits the porosity and specific surface of the powder finally obtained. Meanwhile, ⁎ Corresponding authors. E-mail addresses: [email protected] (J. Faure), [email protected] (R. Drevet).
http://dx.doi.org/10.1016/j.msec.2014.11.045 0928-4931/© 2014 Elsevier B.V. All rights reserved.
increasing research efforts are being invested to develop the sol–gel process which leads to bioactive glass powders of high purity and homogeneity. The sol–gel process also significantly expands the range of the chemical composition of bioactive glasses as compared to the traditional melting-quenching method [9,10]. These advantages make the sol–gel method very efficient for the synthesis of the 45S5 bioactive glass. Even though the sol–gel process is said to proceed under mild conditions, most protocols advocate the use of strong acid (nitric acid and hydrochloric acid principally) or bases (ammonium hydroxide) at elevated concentrations (up to 1 M) to catalyze the sol–gel reactions [11]. Therefore it is suitable to find ways to make the process cleaner avoiding the usually extreme pH or high temperature conditions of most protocols. In this context, several recent papers report on the synthesis of bioactive glass nanoparticles using different organic acids at very low concentrations (typically in the range of 0.01 M) to catalyze the chemical reactions involved during the process [12–15]. In these works, citric acid solutions are commonly used to catalyze the hydrolysis of the silicon alkoxide (TEOS: tetraethylorthosilicate). This idea comes from finding inspiration in nature, therefore it is considered to be a biomimetic inspiration. Indeed, an interesting advancement of the sol–gel chemistry is the understanding of the biosilicification in diatoms and glass sponges [16,17]. These organisms accumulate silicic acid Si(OH)4 from their environment and, through the actions of proteins
408
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412
Table 1 Experimental conditions of the Sol–gel process. SiO2 precursor Na2O precursor CaO precursor P2O5 precursor Reaction temperature Catalyst
Tetraethylorthosilicate (TEOS): Si(OC2H5)4 Sodium nitrate: NaNO3 Calcium nitrate tetrahydrate: Ca(NO3)2·4H2O Triethylphosphate (TEP): PO(C2H5)3 20 °C; 35 °C; 50 °C Nitric acid (HNO3): 0.5 M and 2 M Citric acid (C6H8O7): 0.5 mM, 5 mM and 50 mM
and other macromolecules, form complex silica structures [17,18]. The elucidation of the mechanisms underlying the biosilicification mechanisms has led to the in vitro synthesis of silica under physiological conditions using proteins, polypeptides, amino acids and mainly citric acid to catalyze the sol–gel reactions [19–21]. Considering the specific composition of bioactive glasses and in particular its calcium and phosphate content, it may be necessary to draw inspiration from bone morphogenesis. Consequently, citric acid is here used to trigger the sol–gel reactions. Thus the main objective of this work concerns the ability to succeed the synthesis of the sol–gel derived 45S5 bioactive glass by using a low concentrated citric acid solution to catalyze the hydrolysis reaction. Firstly, we determine the optimal sol–gel parameters to reach the 45S5 chemical composition by using either the conventional nitric acid or either the innovative citric acid as catalyst. In the second part we compare the physico-chemical properties and the in vitro bioactivity of the two optimal sol–gel powders with that of the commercial 45S5 Bioglass®. 2. Materials and methods 2.1. Sol–gel synthesis The sol–gel synthesis of the bioactive glasses proceeded according to a conventional protocol for the synthesis of 45S5 bioglass [22,23] using the following chemical precursors tetraethyl orthosilicate Si(OC2H5)4 (TEOS), triethyl phosphate PO(C2H5)3 (TEP), calcium nitrate tetrahydrate Ca(NO3)2,4H2O and sodium nitrate NaNO3. Two different aqueous acid solutions prepared with nitric acid (HNO3) or with citric acid (C6H8O7) were used to catalyze the hydrolysis reaction. The hydrolysis and condensation reactions were performed within a thermostated reactor to control the reaction temperature. The different elaboration parameters tested in this work are listed in Table 1 and the flowchart of the sol–gel 45S5 fabrication process is presented in Fig. 1. The molar ratio of TEOS, TEP, NaNO3 and Ca(NO3)2·4H2O were designed according to the molar ratio of SiO2, P2O5, Na2O and CaO in 45S5. To achieve a clear sol the molar ratio between the aqueous acid solution and the four chemical precursors was set to 10. Firstly, the acidic solution (26 mL) was magnetically stirred
Preparation of aqueous acid solutions
in the thermostated reactor at the desired temperature and TEOS (11.6 mL) and TEP (1 mL) were added dropwise to the solution and stirred until a clear solution was obtained. Next, the NaNO3 powder (4.66 g) was slowly added in the stirred solution until its complete dissolution. Finally the Ca(NO3)2,4H2O powder (7.15 g) was added slowly to the solution stirred during 1 h to result in a transparent sol. Therefore the sol starts to transform into gel through polycondensation reactions. The gel was then kept at 60 °C for 12 h and finally dried at 200 °C and 700 °C for 5 h and 2 h respectively. After the drying step the gel was manually crushed to obtain a fine powder. Pellets of the obtained powders were then prepared with a manual press for the characterization of the elemental composition and for the bioactivity tests. In order to observe the specific properties of the two bioglasses synthesized by the sol–gel process, they were systematically compared to commercial Bioglass® synthesized by the classical melting-quenching process. 2.2. Physico-chemical and structural characterizations 2.2.1. Scanning electron microscopy and X-ray microanalysis (SEM–EDXS) The morphology of the synthesized powders was observed by SEM using a LaB6 electron microscope (JEOL JSM-5400LV) operating at 0–30 kV. This microscope is associated to an energy dispersive EDAX spectrometer equipped with an ultra-thin window Si(Li) detector cooled with liquid nitrogen. This spectrometer is associated with an acquisition and quantification system GENESIS (Eloïse SARL, France) based on the ZAF method used to determine the elemental concentrations of the analyzed sample [24,25]. All the X-ray microanalysis results presented in this work were obtained according to the following method: acquisition and quantification of four spectra on four flat areas of the pellet selected by SEM. All the X-ray spectra were obtained with energy of 15 keV and an acquisition time of 100 s. The final values of elemental concentrations were obtained after calculation of mean values and dispersions. 2.2.2. Fourier transformed infrared spectroscopy (FTIR) Structural characterizations of the powders were performed by FTIR spectroscopy. The FTIR spectra were obtained in reflexion mode with a FTIR imaging system (Spotlight, Perkin–Elmer, Courtaboeuf, France) coupled to a spectrometer (Spectrum 300, Perkin–Elmer) in the 400– 2000 cm−1 range with a spectral resolution of 4 cm−1. 2.2.3. X-ray diffraction (XRD) The crystallinity of the powders was studied by XRD with a Bruker D8 Advance diffractometer using a monochromatic copper radiation (CuKα) of wavelength λ = 0.15406 nm. The diffractograms were recorded in steps of 0.04° with an acquisition time of 12 s in a range of
Addition of TEOS and TEP
Stirring
Clear sol
Addition of NaNO3
Dissolution
Powder
12 h at 60°C 5 h at 200°C 2 h at 700°C
Drying Gel
Fig. 1. Flowchart of the sol–gel process.
Aging
Addition of Ca(NO3)2
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412 Table 2 Ionic concentration of human blood plasma and SBF solution used for the evaluation of in vitro bioactivity. Ionic concentration (mmol·L−1)
Human blood plasma Simulated body fluid (SBF)
Na+
K+
Mg2+
Ca2+ Cl−
HCO− 3
HPO2− 4
SO2− 4
142.0 142.0
5.0 5.0
1.5 1.5
2.5 2.5
27.0 4.2
1.0 1.0
0.5 0.5
103.0 147.8
was analyzed by SEM–EDXS for morphological and chemical composition analysis. 3. Results and discussion 3.1. Optimization of the experimental conditions of the sol–gel process
diffraction angles 2θ between 10 and 60°. The crystalline phases were then identified by the powder diffraction files of the International Centre for Diffraction Data (ICDD). 2.2.4. Determination of the specific surface area by BET method Nitrogen adsorption measurements were carried out using a Micromeritics ASAP 2010 instrument at 77 K. The Brunauer–Emmett– Teller (BET) method was applied to derive the surface area from physisorption isotherm data. The isotherm was constructed point-bypoint by the admission and withdrawal of known amounts of gas, with adequate time allowed for equilibration at each relative pressure (P/P0, where P is the equilibrium vapor pressure and P0 is the saturation vapor pressure). The BET method is based on the determination of the monolayer capacity, i.e. the nitrogen amount corresponding to the adsorption of a complete monolayer (Vmono). Before analysis, the samples were maintained overnight at 170 °C in an oven. Prior to adsorption, they were out-gassed under vacuum at 250 °C for 1 h. The surface area is determined from a set of 10 experimental points of the linear range of the BET plot (0.05 b P/P0 b 0.3). The surface area (SBET) is calculated according to:
SBET ¼
409
Vmono NA A VM
where Vmono is the monolayer capacity (cm3·g−1) at T = 273 K and P = 101,325 Pa, NA is the Avogadro constant, A is the molecular crosssectional area (for nitrogen, A = 0.162 nm2 at 77 K) and VM is the molar volume (22414 cm3·mol−1). 2.3. In vitro bioactivity tests To evaluate the bioactivity of the synthesized powders, in vitro tests were performed in Simulated Body Fluid (SBF). This solution is acellular, protein free with a pH of 7.4 buffered with Tris–hydroxymethylaminomethane (TRIS). Table 2 shows the composition of SBF compared to human blood plasma. In vitro bioactivity tests were performed by soaking a pellet into SBF for two periods (1 h and 4 h) at 37 °C that correspond to the key times of bioactivity for bioactive glasses. The volume of SBF used during the test is calculated according to the procedure described by Kokubo and Takadama [26]: SBF volume (mL) = 10% pellet mass (mg). After the bioactivity test the pellet was extracted from the solution and immersed into acetone for 10 s to remove SBF and stop surface reactions. After drying, the pellet surface
In the first part of this work we aim to determine the optimal experimental conditions to reach the 45S5 composition by varying parameters such as reaction temperature and concentration of the aqueous acid solutions used to catalyze the sol–gel reactions. Nitric acid has been tested as a catalyst by combining three temperatures (20 °C, 35 °C and 50 °C) and two concentrations (0.5 M and 2 M). Table 3 presents the results obtained by quantitative EDXS that provides the elemental compositions of the powders synthesized with these process parameters. The optimal parameters, which lead to the best composition very close to the theoretical one, are observed at 35 °C and 2 M (this sol–gel powder was named sol–gel1). The citric acid has been also tested as a catalyst by combining two temperatures (20 °C and 35 °C) and three concentrations (0.5 mM, 5 mM and 50 mM). Table 4 presents the results obtained by quantitative EDXS that provides the elemental compositions of the powders synthesized with these experimental parameters. In this case, the optimal parameters that provide the best chemical composition are observed at 20 °C and 5 mM (this sol–gel powder was named sol–gel2). The comparison of the two synthesized powders (sol–gel1 and sol–gel2) shows that a lack of silicon and phosphorus is observed in both cases that can be attributed to the incomplete hydrolysis of the precursors and to the incomplete condensation due to unfavorable pH conditions. We can also observe a slight improvement of the chemical composition of the bioglass by using citric acid (sol–gel2). This behavior is probably due to the ability of citric acid to chelate calcium and sodium ions with its three carboxylic groups [13, 14,27]. Therefore we can conclude that the use of citric acid instead of nitric acid to catalyze the hydrolysis of TEOS and TEP leads to the sol– gel synthesis of the 45S5 bioactive glass with the expected chemical composition. 3.2. Structural characterization of the bioglasses The FTIR spectra of the two synthesized sol–gel glasses are presented in Fig. 2 and compared with the spectrum of the commercial Bioglass®. Both spectra of sol–gel1 and sol–gel2 reveal vibration bands similar to that of Bioglass®. The two broad bands at 930 cm−1 and 1039 cm−1 correspond to silicate adsorption bands which are respectively Si–O–Si stretching of non-bridging oxygen atoms and Si–O–Si asymmetric stretching of bridging oxygen atoms within the silicate tetrahedron. Silicate bands for Si–O–Si bending mode are also observable at 478 cm−1 and 503 cm−1 [28,29]. The peak at around 602 cm−1 and 880 cm−1 are attributed to P–O bending of PO3− 4 groups [28,29]. The weak bands observed at around 1450 cm−1 are related to the presence of residual carbonate groups from the precursors [29]. Thus, the FTIR spectra obtained from sol–gel1 and sol–gel2 are very similar and clearly reveal the glassy structure of the two sol–gel powders. In addition the comparison with
Table 3 Elemental compositions of sol–gel powders synthesized with nitric acid as catalyst (the bold line correspond to the optimal powder named sol-gel1).
T = 20 °C T = 35 °C T = 50 °C
Element (at.%)
O
Si
Na
Ca
P
Theoretical
55.3
16.3
17.1
9.5
1.8
0.5 M 2M 0.5 M 2M 0.5 M 2M
58.2 56.5 55.4 56.7 56.4 55.6
± ± ± ± ± ±
0.3 0.5 0.4 0.3 0.2 0.5
11.7 13.7 14.0 15.2 13.4 14.5
± ± ± ± ± ±
0.3 0.2 0.3 0.2 0.1 0.3
15.3 13.5 14.5 15.3 14.0 15.6
± ± ± ± ± ±
0.2 0.3 0.2 0.3 0.3 0.6
13.6 14.8 14.5 11.6 14.8 13.0
± ± ± ± ± ±
0.2 0.4 0.3 0.3 0.3 0.2
1.2 1.5 1.6 1.2 1.4 1.3
± ± ± ± ± ±
0.2 0.2 0.2 0.1 0.2 0.2
410
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412
Table 4 Elemental compositions of sol–gel powders synthesized with citric acid as catalyst (the bold line correspond to the optimal powder named sol-gel2).
T = 20 °C
T = 35 °C
Element (at.%)
O
Si
Na
Ca
P
Theoretical
55.3
16.3
17.1
9.5
1.8
0.5 mM 5 mM 50 mM 0.5 mM 5 mM 50 mM
55.8 54.5 55.4 55.8 54.2 55.6
± ± ± ± ± ±
0.3 0.5 0.3 0.3 0.3 0.2
14.1 15.9 14.0 13.4 15.5 13.5
the commercial Bioglass® indicates that a partial crystallization of the two sol–gel glasses occurred during the process. The structural characterization has been completed with XRD studies. The Bioglass® powder appears to be essentially amorphous due to the melting–quenching synthesis process. However the XRD patterns reveal few crystalline peaks probably due to the speed of the quenching step [30]. We can also observe that the diffractograms of sol–gel1 and sol–gel2 are very similar (Fig. 3) with a more important crystallization than that of Bioglass®. Indeed there are several crystalline sodium calcium silicate phases inside the amorphous structure of the bioactive glass. Among them, combeite Na2Ca2Si3O9 (ICDD PDF #22.1455) is the most interesting one as it is known to influence the bioactivity. The formation of the combeite phase is associated to the heat treatment performed at 700 °C during the drying step used for sol–gel synthesis [31]. An accurate comparison of the diffractograms reveals that the formation of the Na2Ca2Si3O9 phase is more pronounced in sol–gel2. This observation is characteristic of the bioactive glasses synthesized by the sol–gel process in comparison with those synthesized by the usual melting– quenching process [32]. This may impact the behavior of the bioactive glasses in contact with physiological environment as the chemical properties of the material, such as solubility, are then modified [30–32].
± ± ± ± ± ±
0.3 0.3 0.3 0.2 0.2 0.1
12.5 16.7 15.4 13.8 17.2 14.4
± ± ± ± ± ±
0.3 0.3 0.3 0.4 0.2 0.8
16.3 11.3 14.1 15.6 11.5 14.8
± ± ± ± ± ±
0.6 0.4 0.2 0.6 0.3 0.7
1.3 1.6 1.1 1.4 1.6 1.7
± ± ± ± ± ±
0.2 0.2 0.1 0.2 0.2 0.2
provides very interesting information. The grain surface of the two sol– gel powders is highly rough and porous whereas for the Bioglass® powder the grain surface appears to be extremely smooth. In addition the roughness and porosity observed on the grain surface of sol–gel2 is higher than that of sol–gel1. These SEM observations can be related to the results obtained from the BET method which indicate that the specific surface area of the two sol–gel powders is higher than that of the Bioglass® powder, particularly for sol–gel2. The very low roughness and porosity of the Bioglass® powder is due to the melting–quenching process that uses an extremely quick cooling of the liquid phase during the quenching step [30]. However, the original observation highlighted in the present work concerns the specific effect of citric acid which increases the roughness and the porosity of the sol–gel powder. Indeed, the surface of the sol–gel2 grain powder is highly porous and made of nanoscale cylinders. This specific observation indicates that the use of citric acid as a catalyst of the hydrolysis reaction modifies the surface morphology of the grain powder at a nanometer scale. Obviously, these results highlight differences between the exchange surfaces of the powders that could lead to various behaviors in physiological environment.
3.4. In vitro bioactivity test of the bioglasses 3.3. Characterization of the specific surface area of the bioglasses The results obtained from the BET method indicate that the specific surface area of the materials clearly depends on the elaboration process and also on the experimental conditions used to produce the powders. Indeed, the specific surface area of sol–gel1, sol–gel2 and Bioglass® powders are respectively 0.6 m2·g− 1, 0.9 m2·g− 1 and 0.4 m2·g− 1. The SEM micrographs of Fig. 4 show that the three powders are constituted by grains with diverse sizes. The biggest grains are about 100 μm in size and the smallest ones are less than 10 μm. The sol–gel2 powder seems to get more small grains than the sol–gel1 powder which itself seems to have more small grains than the Bioglass® powder. The high magnification micrographs carried out on one big grain of each powder
SEM micrographs and EDXS spectra of the samples obtained from Bioglass®, sol–gel1 and sol–gel2 powders before and after immersion in SBF solution are shown in Fig. 5. After 1 h of immersion, the SEM micrographs reveal that no significant change occurred on the three samples. After 4 h of immersion significant changes of the surface morphology and of the chemical composition are observed. For the Bioglass® sample a slight dissociation of the pellet surface during immersion in the SBF solution can be observed without any evidence of its mineralization. The associated EDXS spectrum demonstrates that the main change in the surface chemical composition is due to a significant decrease in sodium concentration indicating that the bioactivity reactions are just starting during the test [1]. For the sol–gel1 and sol–gel2 samples, the SEM micrographs highlight an important mineralization of their surfaces after
Fig. 2. FTIR spectra of the sol–gel glasses and of the commercial Bioglass®.
Fig. 3. X-ray diffractograms of the sol–gel glasses and of the commercial Bioglass®.
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412
411
Fig. 4. SEM micrographs of the sol–gel glasses and of the commercial Bioglass® (low and high magnification).
exposure to SBF associated to the formation of a calcium phosphate deposit [26,32]. This surface mineralization is confirmed by the EDXS spectra which clearly show a strong increase of the intensities of the peaks of calcium and phosphorus. Thus, the in vitro bioactivity test of the two sol–gel samples clearly indicates that the bioactivity level of the powders obtained by the sol–gel synthesis is much higher than that of the commercial Bioglass®. It is usually established that the combeite phase decreases the bioactivity reaction kinetic of the bioactive glasses when they are immersed in physiological environment by transforming it into amorphous calcium phosphate [31]. However, in this work, the bioactivity of the sol–gel glasses appears to be higher than the bioactivity of the Bioglass®. Indeed, the grains of sol–gel glasses reveal highly rough surfaces and porosity that provide high exchange surface in physiological environment. Thus the exchange surface of the sol–gel bioglass is more important than the exchange surface of the Bioglass® synthesized by the melting–quenching process. Indeed, the BET measurements confirmed that the specific surface area of the two sol–gel glasses is higher than that of the Bioglass® [33,34]. On the other hand, the comparison between sol–gel1 and sol–gel2 indicates that the bioactivity level of the two sol–gel powders seems to be very similar. Therefore the use of citric acid instead of nitric acid as catalyst during sol–gel synthesis allows to obtain the bioactive glass 45S5 without alteration of its bioactivity.
4. Conclusions In this work we demonstrate that the 45S5 bioactive glass powder can be synthesized by the sol–gel process with a very low concentrated citric acid solution instead the usual highly concentrated nitric acid solution to catalyze the hydrolysis reaction. Firstly, we determine the optimal synthesis experimental conditions to reach the 45S5 elemental composition by varying two parameters, the reaction temperature and the concentration of the acid solutions. The best chemical composition of the bioactive glass has been obtained either with nitric acid at 2 M and 35 °C or either with citric acid at 5 mM and room temperature. The use of citric acid as catalyst for the sol–gel synthesis of a 45S5 powder extremely decreases the acid solution concentration necessary to catalyze the hydrolysis reactions of TEOS and TEP. We then compare the two optimal sol–gel powders with the commercial Bioglass®. The structural characterizations highlight a partial crystallization of the amorphous sol–gel powders which are made of several sodium calcium silicate phases including the combeite in particular. The formation of this highly bioactive crystal phase is associated to the last drying step at 700 °C during the sol–gel process. Finally we compare the in vitro bioactivity of the powders by using immersion tests into SBF solution. We clearly demonstrate that the bioactivity level of the two sol–gel
412
J. Faure et al. / Materials Science and Engineering C 47 (2015) 407–412
Fig. 5. SEM–EDXS analysis of the sol–gel glasses and of the commercial Bioglass® as a function of the immersion time into SBF.
powders is much higher than that of the commercial Bioglass®. Moreover, the bioactivity of the bioactive glasses is mainly influenced by the specific surface area of the synthesized powder rather than the presence of the combeite crystalline phase inside the glass. Thus the use of citric acid instead of nitric acid as catalyst during the sol–gel synthesis allows to obtain the bioactive glass 45S5 without altering its bioactivity. The development of this new process is very attractive as it uses conditions suitable to add and encapsulate biomolecules (proteins, growth factor, drugs, etc.) that could help speed up the bone growth formation which is not only possible in most protocols owing to the usually extreme pH or high temperature conditions. A perspective of this work will then consist in the study of longer immersion time in physiological medium and observations of cell behavior in contact with these prosthetic surfaces.
[6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
Acknowledgment The authors sincerely thank Professor Aldo R. Boccaccini from the University of Erlangen-Nuremberg and the Imperial College of London for kindly providing the commercial Bioglass® studied in this work. They also express their gratitude to Professor Hassan Oudadesse from the University of Rennes 1 for the BET measurements.
[23] [24] [25] [26] [27] [28] [29]
References [1] [2] [3] [4] [5]
L.L. Hench, J. Am. Ceram. Soc. 81 (1998) 1705–1728. V.J. Shirtliff, L.L. Hench, J. Mater. Sci. 38 (2003) 4697–4707. L.L. Hench, Ceram. Int. 22 (1996) 493–507. L.L. Hench, J. Eur. Ceram. Soc. 29 (2009) 1257–1265. Y. Gu, G. Wang, X. Zhang, Y. Zhang, C. Zhang, X. Liu, M.N. Rahaman, W. Huang, H. Pan, Mater. Sci. Eng. C 36 (2014) 294–300.
[30] [31] [32] [33] [34]
X. Liu, M.N. Rahaman, G.E. Hilmas, B.S. Bal, Acta Biomater. 9 (2013) 7025–7034. L.L. Hench, J. Mater. Sci. Mater. Med. 17 (2006) 967–978. J.R. Jones, E. Gentleman, J. Polak, Elements 3 (2007) 393–399. J. Zhong, D.C. Greenspan, J. Biomed. Mater. Res. 53 (2000) 694–701. D. Arcos, M. Vallet-Regi, Acta Biomater. 6 (2010) 2874–2888. D. Avnir, T. Coradin, O. Lev, J. Livage, J. Mater. Chem. 16 (2006) 1013–1030. B. Lei, X. Chen, Y.H. Koh, J. Sol-Gel Sci. Technol. 58 (2011) 656–663. G.M. Luz, J.F. Mano, Nanotechnology 22 (2011) 494014. Z. Hong, A. Liu, L. Chen, X. Chen, X. Jing, J. Non-Cryst. Solids 335 (2009) 368–372. B. Lei, X. Chen, Y. Wang, N. Zhao, C. Du, L. Fang, Biomed. Mater. 5 (2010) 054103. D.J. Belton, O. Deschaume, C.C. Perry, FEBS J. 279 (2012) 1710–1720. S.V. Patwardhan, Chem. Commun. 47 (2011) 7567–7582. H.C. Schröder, X. Wang, W. Tremel, H. Ushijima, W.E. Müller, Nat. Prod. Rep. 25 (2008) 455–474. T. Coradin, J. Livage, Colloids Surf. B: Biointerfaces 21 (2001) 329–336. D. Belton, G. Paine, S.V. Patwardhan, C.C. Perry, J. Mater. Chem. 14 (2004) 2231–2241. T. Mizutani, H. Nagase, N. Fujiwara, H. Ogoshi, Bull. Chem. Soc. Jpn. 71 (1998) 2017–2022. I. Cacciotti, M. Lombardi, A. Bianco, A. Ravaglioli, L. Montanaro, J. Mater. Sci. Mater. Med. 23 (2012) 1849–1866. H. Pirayesh, J.A. Nychka, J. Am. Ceram. Soc. 96 (2013) 1643–1650. N. Dumelie, H. Benhayoune, G. Balossier, J. Phys. D. Appl. Phys. 40 (2007) 2124–2131. H. Benhayoune, J. Phys. D. Appl. Phys. 35 (2002) 1526–1531. T. Kokubo, H. Takadama, Biomaterials 27 (2006) 2907–2915. P.B. Wagh, A. Venkateswara Rao, D. Haranath, J. Porous. Mater. 4 (1997) 295–301. H.A. Elbatal, M.A. Azooz, E.M.A. Khalil, A. Soltan Monem, Y.M. Hamdy, Mater. Chem. Phys. 80 (2003) 599–609. A. Lucas-Girot, F.Z. Mezahi, M. Mami, H. Oudadesse, A. Harabi, M. le Floch, J. NonCryst. Solids 357 (2011) 3322–3327. L. Lefebvre, J. Chevalier, L. Grémillard, R. Zenati, G. Thollet, D. Bernache-Assollant, A. Govin, Acta Mater. 55 (2007) 3305–3313. Q.Z. Chen, I.D. Thompson, A.R. Boccaccini, Biomaterials 27 (2006) 2414–2425. R.L. Siqueira, O. Peitl, E.D. Zanotto, Mater. Sci. Eng. C 31 (2011) 983–991. I. Izquierdo-Barba, A.J. Salinas, M. Vallet-Regi, J. Biomed. Mater. Res. 47 (1999) 243–250. R. Li, A.E. Clark, L.L. Hench, J. Appl. Biomater. 2 (1991) 231–239.
