Materials Science and Engineering C 36 (2014) 20–24
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Investigation of the sample preparation and curing treatment effects on mechanical properties and bioactivity of silica rich metakaolin geopolymer M. Catauro ⁎, F. Bollino, F. Papale, G. Lamanna Department of Industrial and Information Engineering, Second University of Naples, Via Roma 29, 81031 Aversa, Italy
a r t i c l e
i n f o
Article history: Received 4 July 2013 Received in revised form 28 October 2013 Accepted 16 November 2013 Available online 2 December 2013 Keywords: Geopolymers Molar ratio Si/Al = 31 Bioactivity Compressive strength
a b s t r a c t In many biomedical applications both the biological and mechanical behaviours of implants are of relevant interest; in the orthopaedic field, for example, favourable bioactivity and biocompatibility capabilities are necessary, but at the same time the mechanical characteristics of the implants must be such as to allow one to support the body weight. In the present work, the authors have examined the application of geopolymers with composition H24AlK7Si31O79 and ratio Si/Al = 31 to be used in biomedical field, considering two different preparation methods: one of the activators (KOH) has been added as pellets in the potassium silicate solution, in the other as a water solution with 8 M concentration. Moreover, a different water content was used and only some of the synthesized samples were heat treated. The chemical and microstructural characterizations of those materials have been carried out by Fourier transform infrared spectroscopy (FT-IR) and scanning electron microscopy (SEM). Subsequently, the effects of the adopted preparation on the mechanical and biological properties have been studied: compressive strength tests have demonstrated that more fragile specimens were obtained when KOH was added as a solution. The bioactivity was successfully evaluated with the soaking of the samples in a simulated body fluid (SBF) for 3 weeks. The formation of a layer of hydroxyapatite on the surface of the materials has been shown both by SEM micrographs and EDS analyses. © 2013 Published by Elsevier B.V.
1. Introduction The search for new biocompatible materials for biomedical applications has been constantly expanding in the recent decades. The nature of biomaterials can be organic, inorganic or hybrid, depending on the function they have to perform in the body. Generally in orthopaedics and odontoiatric fields inorganic or hybrid materials are preferred for their mechanical characteristics and have proposed several types of materials such as bioglass [1–6], calcium phosphates [7,8] and aragonite [9]. Materials that have recently attracted interest for hard tissue prostheses or bone fillers are the geopolymers, synthetic aluminosilicates. The term “geopolymer” was used by J. Davidovits [10] to identify a family of amorphous aluminosilicates activated by alkali or alkali silicates. They have been mainly suggested to replace Portland cement as environmentally friendly building materials [11,12] due to the lower energy required in their preparation and the substantial reduction in CO2 emissions [13,14]. Geopolymers are formed by a dissolution and precipitation process using aluminosilicates as precursor materials, such as thermally treated kaolin and fly-ash, in alkaline media and silicate solutions [15]. In those media, the formation of monomers \OSi(OH)3 and Al(OH)− 4 occurs and the following reactions with alkali ions and ⁎ Corresponding author. Tel.: +39 0815010360; fax: +39 0815010204. E-mail address: [email protected] (M. Catauro). 0928-4931/$ – see front matter © 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.msec.2013.11.026
water results in the formation of dimers and longer chains [16]. The final product consists of an amorphous or semi-crystalline three dimensional network of tetravalent silica and alumina species with the negative charge compensating single valent cations, e.g. Na+ and K+ and with good mechanical properties [17–19]. The water used in the synthesis is expelled during the reaction to form the pores after drying [15]. Geopolymers with different microstructures can be formed by varying the aluminosilicate precursor type [15], the stoichiometric ratio of Si/Al [20,21], the alkali hydroxide and water contents [22] as well as the reaction temperature, pressure [15], the setting and curing procedure [23]. The use of geopolymers in the biomaterials field, however, is limited by several potential problems: the high alkalinity of the materials can cause cell death, and a certain toxicity can be induced by the possible leaching of aluminium from the implant [24]. Oudanesse et al. [25], on the other hand, demonstrated that amorphous geopolymers of the potassium-poly(sialate)-nanopolymer type, obtained with molar ratios Si:Al = 31 and K2O:SiO2 = 0.54, give excellent results in terms of biological compatibility. The aim of the present work is to investigate the effects of sample preparation method on the mechanical properties and bioactivity of silica rich metakaolin geopolymers (with molar ratio Si:Al = 31). A particular attention has been devoted to the role of the activation procedure, water content and temperature.
M. Catauro et al. / Materials Science and Engineering C 36 (2014) 20–24
21
Table 1 Silica rich formulation and details of activation. Label
Activation
Consolidation
Thermal treatment
GEO GEO' GEO-TA GEO-TA'
KOH pellets directly added to K-silicate (water added = 35 ml) KOH solution 8 M (water added = 25 ml) KOH pellets directly added to K-silicate (water added = 35 ml) KOH solution 8 M (water added = 25 ml)
3 days in a closed plastic bag
None
1 day in polyethylene container prior thermal treatment
65 °C in polyethylene for 180 min
Geopolymers with that formulation were already proposed in literature as bony filling [9] or as bioactive materials capable to form bonelike minerals by interaction with blood plasma [24–26], but the effect of their preparation on the mechanical and biological properties has not yet been investigated extensively. In the following, two preparation methods are described: in one of them solid KOH (pellets) has been added as alkaline activator, while an 8 M KOH solution has been used in the second one. Moreover, different water contents were used in the two methods. Subsequently, some of the synthesized samples were heat treated. The effect of that parameter variation on the materials structure has been investigated. FTIR and SEM have allowed the detection of a different microstructure which is responsible of a different behaviour in terms of the mechanical and biological response of those materials. 2. Materials and methods 2.1. Geopolymers preparation Metakaolin (MK) was used as the principal source of aluminosilicate because it improves mechanical strength and reduces the transport of water and salts in the final product. Metakaolin, with a molar ratio Si/ Al = 2.10, was prepared by calcining kaolinitic clay at 700 °C in muffle oven for 3 h. In order to reduce the consequences of the presence of non-bonded Al atoms in the final product, an increase of silica was operated by SiO2 additions to MK. Thus the percentage of the Al present in the final formulation decreased to Si/Al = 31, but, at the same time, the effect of 3D bonded silica chains increased the network connectivity reducing the Al release [9,25,26]. Following Oudadesse et al. [9,25,26], geopolymers with theoretical formula H24AlK7Si31O79 are proposed here. The raw materials used in addition to the previously mentioned metakaolin were KOH, K2SiO3 and SiO2. Laboratory grade KOH pellets and distilled water were used to prepare 8 M alkaline solution (4.45 g of solution in 1.225 g of MK) or were directly added to the potassium silicate solution (1.995 g
in 7.164 mol). Potassium silicate solution (grade N, RM3,0 from Ingessil s.r.l., Verona, Italy) with a molar ratio of SiO2/K2O equal to 3.10 was used. In both procedures, 1.225 g MK was added to 14.655 g of completely amorphous SiO2 which was produced hydrothermally from natural silicates. Thermal cycles for 3D silica network consolidation were applied, as reported in Table 1, following literature indications [9,25,26]. In Fig. 1 only GEO-TA sample structure is shown, as all samples are quite similar. 2.2. Chemical and microstructural characterization To identify the chemical structure of the obtained geopolymers, Fourier transform infrared (FTIR) transmittance spectra were recorded in the 400–4000 cm− 1 region using a Prestige 21 Shimatzu system, equipped with a DTGS KBr (Deuterated Tryglycine Sulphate with potassium bromide windows) detector, with a 2 cm−1 resolution (45 scans). KBr pelletised discs containing 2 mg of sample and 198 mg KBr were prepared. FTIR spectra were elaborated by Prestige software (IRsolution). Simple microstructural analysis was carried out by Scanning Electron Microscopy (SEM, QUANTA 200, FEI Europe Company, the Netherlands) and with Energy Dispersion Spectroscopy (EDS). 2.3. Mechanical properties The compressive strength was evaluated on a universal testing machine with 250 kN load capacity (Zwick Rolell Z250). Prior to the tests, the surfaces of the specimens were opportunely polished to avoid the need of capping. The specimens were given a cubic shape (40 × 40 × 40 mm), centred in the compression-testing machine and loaded up to failure. A quasi static compressive test was performed, controlling the position of the moving crossbeam (velocity 0.1 mm/min). The presented results are to be considered as the average value of seven specimens. 2.4. Bioactivity test In order to study their bioactivity, three discs per each composition were soaked in a simulated body fluid (SBF) solution with an ion concentration nearly equal to that in human blood plasma (Table 2), at 37 °C and in a polystyrene bottle, as required for bioactivity tests
Table 2 Simulated body fluid (SBF) ionic concentration (mM).
Fig. 1. Optical micrograph of GEO-TA bulk sample.
Ion
Ions concentration (mM) Human blood plasma
SBF
Na+ K+ Mg2+ Ca2+ Cl− HCO− 3 HPO2− 4 SO2− 4 pH
142.0 5.0 1.5 2.5 103.0 27.0 1.0 0.5 7.2–7.4
142.0 5.0 1.5 2.5 147.0 4.2 1.0 0.5 7.4
22
M. Catauro et al. / Materials Science and Engineering C 36 (2014) 20–24
Fig. 2. FTIR spectra of (A) GEO, (B) GEO-TA, (C) GEO' and (D) GEO-TA'.
Table 3 IR characteristic bands and their interpretation. Absorption bands
Interpretation
3450 and 1650 cm−1 1080 cm−1 460 cm−1 798 cm−1 960 cm−1 600–800 cm−1 1420 cm−1
hydration water Si\O stretching vibration Si\O bending vibration Al\O stretching vibration Al\OH stretching vibration Al\O\Si vibrations K2CO3
in vitro [27]. The SBF solution was prepared by dissolving reagent grade chemicals NaCl, NaHCO3, KCl, MgCl2·6H2O, CaCl2, Na2HPO4, and Na2SO4 (Sigma-Aldrich) in ultra-pure water and buffered at pH 7.4 using HEPES sodium salt (C8H18N2O4SNa) (Sigma-Aldrich) and 1 M NaOH solution to adjust the pH. In those tests the ratio between the total surface (ST) of the material in contact with the SBF solution and the volume of such solution (VSBF) influence the formation of a hydroxyapatite layer. In our case a constant ratio of ST/VSBF = 10 mm2/ml was used. After an immersion period of 21 days, the materials were removed from the SBF, gently washed with ultra-pure water, and dried in a dessicator. The ability to form an apatite layer on their surface was studied by SEM and EDS analyses. Moreover, SBF solution was analysed by inductively coupled plasma optical emission spectrometer (ICP-OES Shimadzu mod. ICPE 9000) to detect the possible Al release by geopolymers after 21 days in contact with samples.
Fig. 4. EDS spectra of (A) non-reacted and (B) reacted areas.
3. Results and discussion 3.1. Chemical and microstructural characterization FTIR spectra (Fig. 2) of potassium based geopolymers show the typical alluminosilicate vibrations (see Table 3) which are attributed to internal vibrations of Si\O\Si and Si\O\Al. The bands at about 1080 cm− 1 and at about 460 cm− 1 are due to Si\O stretching and bending vibrations respectively. The shoulder at about 960 cm−1 indicates the Al (VI)\OH bonds' presence. Moreover, the bands between 600 and 800 cm−1 are due to Al\O\Si vibrations [28]: in particular, the bands at 798 cm−1 are due to Al(IV)\O stretching vibrations. In all geopolymers FTIR spectra, an approximate relationship between the frequency of the absorption bands and the ratio of Si:Al in the aluminosilicate framework was observed by Milkey [29]: the higher the Al inclusions, the lower the wavelength. Also other features are present in all spectra: two absorption bands at about 3450 and 1650 cm− 1 resulting from hydration water and a weak band at about 1420 cm−1 which indicates the potassium carbonate presence on the surface of this sample. That is due to the passage of water through the bulk during
Fig. 3. SEM micrographs of GEO (A) and GEO' (B) soon after preparation. Circles evidence non reacted metakaolin area.
M. Catauro et al. / Materials Science and Engineering C 36 (2014) 20–24
23
Fig. 5. SEM micrographs of (A) GEO-TA and (B) GEO-TA' after 60 °C thermal treatment.
the drying of the sample, which generates a high concentration of potassium close to the surface, where it forms carbonates after the reaction with atmospheric CO2. In Fig. 2, comparing the spectra of heat treated samples with those untreated (i.e. curve a with b and c with d), an increase of the bands at about 798 cm−1, due to Al(IV)\O stretching vibrations, is observed. This phenomenon is mostly evident when KOH 8 M solution was used. That result suggests a greater advance of the geopolymerization reaction in the heat treated samples. Indeed, there is a progressive formation of Al(IV) species and a reduction of the number of Al(VI) groups (typical of the metakaolin) during the geopolymerization process [15,30]. Moreover, comparing curve a with c and b with d, it is evident that there are not many differences between the geopolymer matrixes obtained with the two said procedures but for what concerns the intensity of water bands' intensity which increases in the case of KOH pellets, because also the water content increases. FT-IR observations were confirmed by SEM/EDS analysis. The micrographs taken one day after preparation of geopolymers show some unreacted metakaolin area in untreated GEO' synthesized using chemical alkali activation with KOH 8 M solution added to K2SiO3 solution (Fig. 3B). That is confirmed by EDS analyses that show, in these zones, a very limited presence of K (Fig. 4A) while in the surrounding material, where K+ cations compensate the negative charge of tetravalent alumina species, the peak of potassium is well represented (Fig. 4B). Similar areas are present also in GEO (Fig. 3A) but in this case the platelet typical of metakaolin morphology is not so well defined, which means that the alkali attack has already begun. This can be explained by the higher presence of water used in the synthesis procedure
compared to GEO'. Indeed, the alkali activation induces the dissolution of the solid aluminosilicate source (metakaolin) by alkaline hydrolysis and the production of aluminate and silicate species (most likely in monomeric form) allowing the start of the geopolymerization by condensation of this monomer in solution [15]. Those areas disappear completely after the 65 °C-treatment when a more uniform microstructure is reached, as shown in SEM micrographs (see Fig. 5). It is evident that the procedure in which the 8 M KOH solution was used leads to a more bonded microstructure (see Fig. 5B) with respect to the case of KOH pellets (Fig. 5A). Probably, that is due to the lower water content which moves away from the materials during the evaporation process. Also EDS analysis confirms the presence of evenly distributed K over the entire surface (the spectra are similar to that shown in Fig. 4B, therefore they are not shown). 3.2. Mechanical properties The mechanical characterization was performed only on GEO-TA and GEO-TA' because the samples GEO and GEO' showed cracks on the surface after the consolidation in a closed plastic bag. The mechanical tests on GEO-TA and GEO-TA' were carried out after the 30 days complete curing from the preparation and the thermal treatment at 65 °C in a Teflon® container for 180 min. Fig. 6 shows one of the performed tests; Tables 4 and 5 show the mean values and standard deviation of the stress at fixed strain values. GEO-TA formulation mainly presents an elastic behaviour, a plateau value follows, which is placed at about 5–10% below the maximum strength (0.6 MPa) (see Fig. 6). The other tested formulation (GEOTA') shows the same behaviour with a maximum compressive strength of 2 MPa. The higher compression resistance of GEO-TA' in respect to
Table 4 GEO-TA compressive test. Strain [−]
Mean stress [MPa]
Stress Std dv [MPa]
0.021 0.041 0.067 0.085
0.479 0.563 0.562 0.551
0.022 0.026 0.025 0.025
Table 5 GEO-TA' compressive test.
Fig. 6. GEO-TA and GEO-TA' compressive strength.
Strain [−]
Mean stress [MPa]
Stress Std dv [MPa]
0.023 0.038 0.058 0.084
0.572 1.531 1.945 1.955
0.026 0.071 0.088 0.091
24
M. Catauro et al. / Materials Science and Engineering C 36 (2014) 20–24
Fig. 7. SEM micrographs of (A) samples after soak in SBF for 21 days, (B) of a hydroxyl-apatite crystal and (C) EDS spectra.
that of GEO-TA can be attributed to the more bonded microstructure observed in the SEM micrographs (Fig. 5B). 3.3. Bioactivity test The evaluation of apatite forming ability was carried out by SEM accompanied by EDS analysis. Fig. 7 shows SEM micrographs (panel A–B) of a sample soaked in SBF for 21 days. The characteristic apatite globular crystals are clearly visible on the sample surface, proving that these geopolymers are bioactive [31]. Elemental analysis (Panel C) confirms that the observed crystals are made of apatite. All samples show a similar bioactivity, independently from the preparation method and the heat treatment. That result is due to the presence of Al–OH groups on the surface of all samples, as the peaks at 960 cm−1in the FTIR spectra (Fig. 2) show. In literature [5,32] it is reported that the hydroxyl-apatite nucleation on silicate glasses and ceramics exposed to SBF, is promoted by hydroxyl groups on their surface and improves when the materials contain cations, e.g. Na+ or K+ ions. These materials release cations, via exchange with H3O+ ions in SBF to form Si–OH and Al–OH groups on their surface; this reaction causes a pH increase of SBF solution and, consequently, Si–OH and Al–OH groups are dissociated into negatively charged units Si\O− and Al\O−. These groups combine with Ca2 + ions present in the fluid imposing an increase of the positive charge on the surface. In addition Ca2+ ions combine with the negative charge of the phosphate ions to form an amorphous phosphate, which spontaneously transforms into hydroxyl-apatite [Ca10(PO4)6(OH)2]. EDS analysis confirms that the crystals observed in SEM micrographs consist of calcium phosphate. The ICP analyses of the SBF solutions where the samples were soaked have revealed that Al release was less than 0.1 mg/l, confirming that in geopolymers with molar ratio Si/Al = 31 the network connectivity increase and Al release is reduced. 4. Conclusions The study has presented the synthesis and characterization of silica rich metakaolin geopolymer activated with KOH/K2SiO3 using KOH as pellets or as 8 M water solution and different curing methods. The experimental data show that the activation procedure prevalently affects the materials structure and hence the mechanical properties of synthesized simples. Higher water content promotes the alluminosilicate source dissolution phase which occurs by alkaline hydrolysis. The thermal treatment improves the geopolymerization in both preparation methods probably because the removal of the water formed during condensation process is promoted. The water removal, however, causes the empty areas' formation. Therefore, the activation procedure that requires the addition of 8 M KOH solution to K-silicate
solution and the heat treatment at 65 °C leads to materials without unreacted metakaolin, with a more bonded microstructure and a higher compressive strength. On the other hand, the hydroxyl-apatite [Ca10(PO4)6(OH)2] formation was observed on the surface of all geopolymers after the immersion in a SBF solution for 21 days, independently from both the activation procedure and the heat treatment. References [1] M. Catauro, F. Bollino, J. Appl. Biomater. Funct. Mater. (Jun 22 2012), http://dx.doi.org/10. 5301/JABFM.2012.9256. [2] M. Catauro, F. Bollino, F. Papale, J. Biomed. Mater. Res. A (Jun 15 2013), http://dx.doi.org/10.1002/jbm.a.34836. [3] M. Catauro, F. Papale, G. Roviello, C. Ferone, F. Bollino, M. Trifuoggi, C. Aurilio, J. Biomed. Mater. Res. A (Oct 4 2013), http://dx.doi.org/10.1002/jbm.a.34978. [4] M. Catauro, M.G. Raucci, M.A. Continenza, A. Marotta, J. Mater. Sci. 39 (2004) 373–375. [5] L.L. Hench, J. Am. Ceram. Soc. 74 (1991) 1487–1510. [6] M.G. Raucci, K. Adesanya, L. Di Silvio, M. Catauro, L. Ambrosio, J. Biomed. Mater. Res. B Appl. Biomater. 92 (2010) 102–110. [7] H.-W. Kim, H.-E. Kim, J. Biomed. Mater. Res. B Appl. Biomater. 77B (2006) 323–328. [8] G.X. Ni, W.W. Lu, K.Y. Chiu, Z.Y. Li, D.Y.T. Fong, K.D.K. Luk, J. Biomed. Mater. Res. B Appl. Biomater. 77B (2006) 409–415. [9] H. Oudadesse, A.C. Derrien, A. Lucas-Girot, Eur. Phys. J. Appl. Phys. 31 (2005) 217–223. [10] J. Davidovits, J. Therm. Anal. Calorim. 37 (1991) 1633–1656. [11] P. Duxson, J.L. Provis, G.C. Lukey, J.S.J. van Deventer, Cem. Concr. Res. 37 (2007) 1590–1597. [12] C. Menna, D. Asprone, C. Ferone, F. Colangelo, A. Balsamo, A. Prota, R. Cioffi, G. Manfredi, Compos. Part B 45 (2013) 1667–1676. [13] C. Ferone, F. Colangelo, R. Cioffi, F. Montagnaro, L. Santoro, Adv. Appl. Ceram. 112 (2013) 184–189. [14] G. Habert, J.B. d'Espinose de Lacaillerie, N. Roussel, J. Clean. Prod. 19 (2011) 1229–1238. [15] P. Duxson, A. Fernández-Jiménez, J. Provis, G. Lukey, A. Palomo, J. van Deventer, J. Mater. Sci. 42 (2007) 2917–2933. [16] W.M. Hendricks, A.T. Bell, C.J. Radke, J. Phys. Chem. 95 (1991) 9513–9518. [17] M.R. Anseau, J.P. Leung, N. Sahai, T.W. Swaddle, Inorg. Chem. 44 (2005) 8023–8032. [18] M.R. North, T.W. Swaddle, Inorg. Chem. 39 (2000) 2661–2665. [19] C. Ferone, G. Roviello, F. Colangelo, R. Cioffi, O. Tarallo, Appl. Clay Sci. 73 (2013) 42–50. [20] C. Ferone, F. Colangelo, G. Roviello, D. Asprone, C. Menna, A. Balsamo, A. Prota, R. Cioffi, G. Manfredi, Materials 6 (2013) 1920–1939. [21] P. Duxson, J.L. Provis, G.C. Lukey, S.W. Mallicoat, W.M. Kriven, J.S.J. van Deventer, Colloids Surf. A Physicochem. Eng. Asp. 269 (2005) 47–58. [22] E. Jämstorp, M. Strømme, G. Frenning, J. Pharm. Sci. 100 (2011) 4338–4348. [23] I. Lancellotti, M. Catauro, C. Ponzoni, F. Bollino, C. Leonelli, J. Solid State Chem. 200 (2013) 341–348. [24] K. MacKenzie, N. Rahner, M. Smith, A. Wong, J. Mater. Sci. (2010) 999–1007(Springer Netherlands). [25] H. Oudadesse, A. Derrien, M. Lefloch, J. Davidovits, J. Mater. Sci. 42 (2007) 3092–3098. [26] H. Oudadesse, A.C. Derrien, M. Mami, S. Martin, G. Cathelineau, L. Yahia, Biomed. Mater. 2 (2007) S59–S64. [27] T. Kokubo, H. Takadama, Biomaterials 27 (2006) 2907–2915. [28] J.A. Gadsden, Infrared spectra of minerals and related inorganic compounds, Butterworths, London, 1975. [29] R.G. Milkey, Am. Mineral. 45 (1960) 990–1007. [30] P. Duxson, G.C. Lukey, F. Separovic, J.S.J. van Deventer, Ind. Eng. Chem. Res. 44 (2005) 832–839. [31] G. Sandeep, H. Varma, T. Kumary, S. Suresh Babu, J. Annie, Trends Biomater. Artif. Organs Int. J. 19 (2006) 99–107. [32] C. Ohtsuki, T. Kokubo, T. Yamamuro, J. Non-Cryst. Solids 143 (1992) 84–92.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Investigation of the sample preparation and curing treatment effects on mechanical properties and bioactivity of silica rich metakaolin geopolymer M. Catauro ⁎, F. Bollino, F. Papale, G. Lamanna Department of Industrial and Information Engineering, Second University of Naples, Via Roma 29, 81031 Aversa, Italy
a r t i c l e
i n f o
Article history: Received 4 July 2013 Received in revised form 28 October 2013 Accepted 16 November 2013 Available online 2 December 2013 Keywords: Geopolymers Molar ratio Si/Al = 31 Bioactivity Compressive strength
a b s t r a c t In many biomedical applications both the biological and mechanical behaviours of implants are of relevant interest; in the orthopaedic field, for example, favourable bioactivity and biocompatibility capabilities are necessary, but at the same time the mechanical characteristics of the implants must be such as to allow one to support the body weight. In the present work, the authors have examined the application of geopolymers with composition H24AlK7Si31O79 and ratio Si/Al = 31 to be used in biomedical field, considering two different preparation methods: one of the activators (KOH) has been added as pellets in the potassium silicate solution, in the other as a water solution with 8 M concentration. Moreover, a different water content was used and only some of the synthesized samples were heat treated. The chemical and microstructural characterizations of those materials have been carried out by Fourier transform infrared spectroscopy (FT-IR) and scanning electron microscopy (SEM). Subsequently, the effects of the adopted preparation on the mechanical and biological properties have been studied: compressive strength tests have demonstrated that more fragile specimens were obtained when KOH was added as a solution. The bioactivity was successfully evaluated with the soaking of the samples in a simulated body fluid (SBF) for 3 weeks. The formation of a layer of hydroxyapatite on the surface of the materials has been shown both by SEM micrographs and EDS analyses. © 2013 Published by Elsevier B.V.
1. Introduction The search for new biocompatible materials for biomedical applications has been constantly expanding in the recent decades. The nature of biomaterials can be organic, inorganic or hybrid, depending on the function they have to perform in the body. Generally in orthopaedics and odontoiatric fields inorganic or hybrid materials are preferred for their mechanical characteristics and have proposed several types of materials such as bioglass [1–6], calcium phosphates [7,8] and aragonite [9]. Materials that have recently attracted interest for hard tissue prostheses or bone fillers are the geopolymers, synthetic aluminosilicates. The term “geopolymer” was used by J. Davidovits [10] to identify a family of amorphous aluminosilicates activated by alkali or alkali silicates. They have been mainly suggested to replace Portland cement as environmentally friendly building materials [11,12] due to the lower energy required in their preparation and the substantial reduction in CO2 emissions [13,14]. Geopolymers are formed by a dissolution and precipitation process using aluminosilicates as precursor materials, such as thermally treated kaolin and fly-ash, in alkaline media and silicate solutions [15]. In those media, the formation of monomers \OSi(OH)3 and Al(OH)− 4 occurs and the following reactions with alkali ions and ⁎ Corresponding author. Tel.: +39 0815010360; fax: +39 0815010204. E-mail address: [email protected] (M. Catauro). 0928-4931/$ – see front matter © 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.msec.2013.11.026
water results in the formation of dimers and longer chains [16]. The final product consists of an amorphous or semi-crystalline three dimensional network of tetravalent silica and alumina species with the negative charge compensating single valent cations, e.g. Na+ and K+ and with good mechanical properties [17–19]. The water used in the synthesis is expelled during the reaction to form the pores after drying [15]. Geopolymers with different microstructures can be formed by varying the aluminosilicate precursor type [15], the stoichiometric ratio of Si/Al [20,21], the alkali hydroxide and water contents [22] as well as the reaction temperature, pressure [15], the setting and curing procedure [23]. The use of geopolymers in the biomaterials field, however, is limited by several potential problems: the high alkalinity of the materials can cause cell death, and a certain toxicity can be induced by the possible leaching of aluminium from the implant [24]. Oudanesse et al. [25], on the other hand, demonstrated that amorphous geopolymers of the potassium-poly(sialate)-nanopolymer type, obtained with molar ratios Si:Al = 31 and K2O:SiO2 = 0.54, give excellent results in terms of biological compatibility. The aim of the present work is to investigate the effects of sample preparation method on the mechanical properties and bioactivity of silica rich metakaolin geopolymers (with molar ratio Si:Al = 31). A particular attention has been devoted to the role of the activation procedure, water content and temperature.
M. Catauro et al. / Materials Science and Engineering C 36 (2014) 20–24
21
Table 1 Silica rich formulation and details of activation. Label
Activation
Consolidation
Thermal treatment
GEO GEO' GEO-TA GEO-TA'
KOH pellets directly added to K-silicate (water added = 35 ml) KOH solution 8 M (water added = 25 ml) KOH pellets directly added to K-silicate (water added = 35 ml) KOH solution 8 M (water added = 25 ml)
3 days in a closed plastic bag
None
1 day in polyethylene container prior thermal treatment
65 °C in polyethylene for 180 min
Geopolymers with that formulation were already proposed in literature as bony filling [9] or as bioactive materials capable to form bonelike minerals by interaction with blood plasma [24–26], but the effect of their preparation on the mechanical and biological properties has not yet been investigated extensively. In the following, two preparation methods are described: in one of them solid KOH (pellets) has been added as alkaline activator, while an 8 M KOH solution has been used in the second one. Moreover, different water contents were used in the two methods. Subsequently, some of the synthesized samples were heat treated. The effect of that parameter variation on the materials structure has been investigated. FTIR and SEM have allowed the detection of a different microstructure which is responsible of a different behaviour in terms of the mechanical and biological response of those materials. 2. Materials and methods 2.1. Geopolymers preparation Metakaolin (MK) was used as the principal source of aluminosilicate because it improves mechanical strength and reduces the transport of water and salts in the final product. Metakaolin, with a molar ratio Si/ Al = 2.10, was prepared by calcining kaolinitic clay at 700 °C in muffle oven for 3 h. In order to reduce the consequences of the presence of non-bonded Al atoms in the final product, an increase of silica was operated by SiO2 additions to MK. Thus the percentage of the Al present in the final formulation decreased to Si/Al = 31, but, at the same time, the effect of 3D bonded silica chains increased the network connectivity reducing the Al release [9,25,26]. Following Oudadesse et al. [9,25,26], geopolymers with theoretical formula H24AlK7Si31O79 are proposed here. The raw materials used in addition to the previously mentioned metakaolin were KOH, K2SiO3 and SiO2. Laboratory grade KOH pellets and distilled water were used to prepare 8 M alkaline solution (4.45 g of solution in 1.225 g of MK) or were directly added to the potassium silicate solution (1.995 g
in 7.164 mol). Potassium silicate solution (grade N, RM3,0 from Ingessil s.r.l., Verona, Italy) with a molar ratio of SiO2/K2O equal to 3.10 was used. In both procedures, 1.225 g MK was added to 14.655 g of completely amorphous SiO2 which was produced hydrothermally from natural silicates. Thermal cycles for 3D silica network consolidation were applied, as reported in Table 1, following literature indications [9,25,26]. In Fig. 1 only GEO-TA sample structure is shown, as all samples are quite similar. 2.2. Chemical and microstructural characterization To identify the chemical structure of the obtained geopolymers, Fourier transform infrared (FTIR) transmittance spectra were recorded in the 400–4000 cm− 1 region using a Prestige 21 Shimatzu system, equipped with a DTGS KBr (Deuterated Tryglycine Sulphate with potassium bromide windows) detector, with a 2 cm−1 resolution (45 scans). KBr pelletised discs containing 2 mg of sample and 198 mg KBr were prepared. FTIR spectra were elaborated by Prestige software (IRsolution). Simple microstructural analysis was carried out by Scanning Electron Microscopy (SEM, QUANTA 200, FEI Europe Company, the Netherlands) and with Energy Dispersion Spectroscopy (EDS). 2.3. Mechanical properties The compressive strength was evaluated on a universal testing machine with 250 kN load capacity (Zwick Rolell Z250). Prior to the tests, the surfaces of the specimens were opportunely polished to avoid the need of capping. The specimens were given a cubic shape (40 × 40 × 40 mm), centred in the compression-testing machine and loaded up to failure. A quasi static compressive test was performed, controlling the position of the moving crossbeam (velocity 0.1 mm/min). The presented results are to be considered as the average value of seven specimens. 2.4. Bioactivity test In order to study their bioactivity, three discs per each composition were soaked in a simulated body fluid (SBF) solution with an ion concentration nearly equal to that in human blood plasma (Table 2), at 37 °C and in a polystyrene bottle, as required for bioactivity tests
Table 2 Simulated body fluid (SBF) ionic concentration (mM).
Fig. 1. Optical micrograph of GEO-TA bulk sample.
Ion
Ions concentration (mM) Human blood plasma
SBF
Na+ K+ Mg2+ Ca2+ Cl− HCO− 3 HPO2− 4 SO2− 4 pH
142.0 5.0 1.5 2.5 103.0 27.0 1.0 0.5 7.2–7.4
142.0 5.0 1.5 2.5 147.0 4.2 1.0 0.5 7.4
22
M. Catauro et al. / Materials Science and Engineering C 36 (2014) 20–24
Fig. 2. FTIR spectra of (A) GEO, (B) GEO-TA, (C) GEO' and (D) GEO-TA'.
Table 3 IR characteristic bands and their interpretation. Absorption bands
Interpretation
3450 and 1650 cm−1 1080 cm−1 460 cm−1 798 cm−1 960 cm−1 600–800 cm−1 1420 cm−1
hydration water Si\O stretching vibration Si\O bending vibration Al\O stretching vibration Al\OH stretching vibration Al\O\Si vibrations K2CO3
in vitro [27]. The SBF solution was prepared by dissolving reagent grade chemicals NaCl, NaHCO3, KCl, MgCl2·6H2O, CaCl2, Na2HPO4, and Na2SO4 (Sigma-Aldrich) in ultra-pure water and buffered at pH 7.4 using HEPES sodium salt (C8H18N2O4SNa) (Sigma-Aldrich) and 1 M NaOH solution to adjust the pH. In those tests the ratio between the total surface (ST) of the material in contact with the SBF solution and the volume of such solution (VSBF) influence the formation of a hydroxyapatite layer. In our case a constant ratio of ST/VSBF = 10 mm2/ml was used. After an immersion period of 21 days, the materials were removed from the SBF, gently washed with ultra-pure water, and dried in a dessicator. The ability to form an apatite layer on their surface was studied by SEM and EDS analyses. Moreover, SBF solution was analysed by inductively coupled plasma optical emission spectrometer (ICP-OES Shimadzu mod. ICPE 9000) to detect the possible Al release by geopolymers after 21 days in contact with samples.
Fig. 4. EDS spectra of (A) non-reacted and (B) reacted areas.
3. Results and discussion 3.1. Chemical and microstructural characterization FTIR spectra (Fig. 2) of potassium based geopolymers show the typical alluminosilicate vibrations (see Table 3) which are attributed to internal vibrations of Si\O\Si and Si\O\Al. The bands at about 1080 cm− 1 and at about 460 cm− 1 are due to Si\O stretching and bending vibrations respectively. The shoulder at about 960 cm−1 indicates the Al (VI)\OH bonds' presence. Moreover, the bands between 600 and 800 cm−1 are due to Al\O\Si vibrations [28]: in particular, the bands at 798 cm−1 are due to Al(IV)\O stretching vibrations. In all geopolymers FTIR spectra, an approximate relationship between the frequency of the absorption bands and the ratio of Si:Al in the aluminosilicate framework was observed by Milkey [29]: the higher the Al inclusions, the lower the wavelength. Also other features are present in all spectra: two absorption bands at about 3450 and 1650 cm− 1 resulting from hydration water and a weak band at about 1420 cm−1 which indicates the potassium carbonate presence on the surface of this sample. That is due to the passage of water through the bulk during
Fig. 3. SEM micrographs of GEO (A) and GEO' (B) soon after preparation. Circles evidence non reacted metakaolin area.
M. Catauro et al. / Materials Science and Engineering C 36 (2014) 20–24
23
Fig. 5. SEM micrographs of (A) GEO-TA and (B) GEO-TA' after 60 °C thermal treatment.
the drying of the sample, which generates a high concentration of potassium close to the surface, where it forms carbonates after the reaction with atmospheric CO2. In Fig. 2, comparing the spectra of heat treated samples with those untreated (i.e. curve a with b and c with d), an increase of the bands at about 798 cm−1, due to Al(IV)\O stretching vibrations, is observed. This phenomenon is mostly evident when KOH 8 M solution was used. That result suggests a greater advance of the geopolymerization reaction in the heat treated samples. Indeed, there is a progressive formation of Al(IV) species and a reduction of the number of Al(VI) groups (typical of the metakaolin) during the geopolymerization process [15,30]. Moreover, comparing curve a with c and b with d, it is evident that there are not many differences between the geopolymer matrixes obtained with the two said procedures but for what concerns the intensity of water bands' intensity which increases in the case of KOH pellets, because also the water content increases. FT-IR observations were confirmed by SEM/EDS analysis. The micrographs taken one day after preparation of geopolymers show some unreacted metakaolin area in untreated GEO' synthesized using chemical alkali activation with KOH 8 M solution added to K2SiO3 solution (Fig. 3B). That is confirmed by EDS analyses that show, in these zones, a very limited presence of K (Fig. 4A) while in the surrounding material, where K+ cations compensate the negative charge of tetravalent alumina species, the peak of potassium is well represented (Fig. 4B). Similar areas are present also in GEO (Fig. 3A) but in this case the platelet typical of metakaolin morphology is not so well defined, which means that the alkali attack has already begun. This can be explained by the higher presence of water used in the synthesis procedure
compared to GEO'. Indeed, the alkali activation induces the dissolution of the solid aluminosilicate source (metakaolin) by alkaline hydrolysis and the production of aluminate and silicate species (most likely in monomeric form) allowing the start of the geopolymerization by condensation of this monomer in solution [15]. Those areas disappear completely after the 65 °C-treatment when a more uniform microstructure is reached, as shown in SEM micrographs (see Fig. 5). It is evident that the procedure in which the 8 M KOH solution was used leads to a more bonded microstructure (see Fig. 5B) with respect to the case of KOH pellets (Fig. 5A). Probably, that is due to the lower water content which moves away from the materials during the evaporation process. Also EDS analysis confirms the presence of evenly distributed K over the entire surface (the spectra are similar to that shown in Fig. 4B, therefore they are not shown). 3.2. Mechanical properties The mechanical characterization was performed only on GEO-TA and GEO-TA' because the samples GEO and GEO' showed cracks on the surface after the consolidation in a closed plastic bag. The mechanical tests on GEO-TA and GEO-TA' were carried out after the 30 days complete curing from the preparation and the thermal treatment at 65 °C in a Teflon® container for 180 min. Fig. 6 shows one of the performed tests; Tables 4 and 5 show the mean values and standard deviation of the stress at fixed strain values. GEO-TA formulation mainly presents an elastic behaviour, a plateau value follows, which is placed at about 5–10% below the maximum strength (0.6 MPa) (see Fig. 6). The other tested formulation (GEOTA') shows the same behaviour with a maximum compressive strength of 2 MPa. The higher compression resistance of GEO-TA' in respect to
Table 4 GEO-TA compressive test. Strain [−]
Mean stress [MPa]
Stress Std dv [MPa]
0.021 0.041 0.067 0.085
0.479 0.563 0.562 0.551
0.022 0.026 0.025 0.025
Table 5 GEO-TA' compressive test.
Fig. 6. GEO-TA and GEO-TA' compressive strength.
Strain [−]
Mean stress [MPa]
Stress Std dv [MPa]
0.023 0.038 0.058 0.084
0.572 1.531 1.945 1.955
0.026 0.071 0.088 0.091
24
M. Catauro et al. / Materials Science and Engineering C 36 (2014) 20–24
Fig. 7. SEM micrographs of (A) samples after soak in SBF for 21 days, (B) of a hydroxyl-apatite crystal and (C) EDS spectra.
that of GEO-TA can be attributed to the more bonded microstructure observed in the SEM micrographs (Fig. 5B). 3.3. Bioactivity test The evaluation of apatite forming ability was carried out by SEM accompanied by EDS analysis. Fig. 7 shows SEM micrographs (panel A–B) of a sample soaked in SBF for 21 days. The characteristic apatite globular crystals are clearly visible on the sample surface, proving that these geopolymers are bioactive [31]. Elemental analysis (Panel C) confirms that the observed crystals are made of apatite. All samples show a similar bioactivity, independently from the preparation method and the heat treatment. That result is due to the presence of Al–OH groups on the surface of all samples, as the peaks at 960 cm−1in the FTIR spectra (Fig. 2) show. In literature [5,32] it is reported that the hydroxyl-apatite nucleation on silicate glasses and ceramics exposed to SBF, is promoted by hydroxyl groups on their surface and improves when the materials contain cations, e.g. Na+ or K+ ions. These materials release cations, via exchange with H3O+ ions in SBF to form Si–OH and Al–OH groups on their surface; this reaction causes a pH increase of SBF solution and, consequently, Si–OH and Al–OH groups are dissociated into negatively charged units Si\O− and Al\O−. These groups combine with Ca2 + ions present in the fluid imposing an increase of the positive charge on the surface. In addition Ca2+ ions combine with the negative charge of the phosphate ions to form an amorphous phosphate, which spontaneously transforms into hydroxyl-apatite [Ca10(PO4)6(OH)2]. EDS analysis confirms that the crystals observed in SEM micrographs consist of calcium phosphate. The ICP analyses of the SBF solutions where the samples were soaked have revealed that Al release was less than 0.1 mg/l, confirming that in geopolymers with molar ratio Si/Al = 31 the network connectivity increase and Al release is reduced. 4. Conclusions The study has presented the synthesis and characterization of silica rich metakaolin geopolymer activated with KOH/K2SiO3 using KOH as pellets or as 8 M water solution and different curing methods. The experimental data show that the activation procedure prevalently affects the materials structure and hence the mechanical properties of synthesized simples. Higher water content promotes the alluminosilicate source dissolution phase which occurs by alkaline hydrolysis. The thermal treatment improves the geopolymerization in both preparation methods probably because the removal of the water formed during condensation process is promoted. The water removal, however, causes the empty areas' formation. Therefore, the activation procedure that requires the addition of 8 M KOH solution to K-silicate
solution and the heat treatment at 65 °C leads to materials without unreacted metakaolin, with a more bonded microstructure and a higher compressive strength. On the other hand, the hydroxyl-apatite [Ca10(PO4)6(OH)2] formation was observed on the surface of all geopolymers after the immersion in a SBF solution for 21 days, independently from both the activation procedure and the heat treatment. References [1] M. Catauro, F. Bollino, J. Appl. Biomater. Funct. Mater. (Jun 22 2012), http://dx.doi.org/10. 5301/JABFM.2012.9256. [2] M. Catauro, F. Bollino, F. Papale, J. Biomed. Mater. Res. A (Jun 15 2013), http://dx.doi.org/10.1002/jbm.a.34836. [3] M. Catauro, F. Papale, G. Roviello, C. Ferone, F. Bollino, M. Trifuoggi, C. Aurilio, J. Biomed. Mater. Res. A (Oct 4 2013), http://dx.doi.org/10.1002/jbm.a.34978. [4] M. Catauro, M.G. Raucci, M.A. Continenza, A. Marotta, J. Mater. Sci. 39 (2004) 373–375. [5] L.L. Hench, J. Am. Ceram. Soc. 74 (1991) 1487–1510. [6] M.G. Raucci, K. Adesanya, L. Di Silvio, M. Catauro, L. Ambrosio, J. Biomed. Mater. Res. B Appl. Biomater. 92 (2010) 102–110. [7] H.-W. Kim, H.-E. Kim, J. Biomed. Mater. Res. B Appl. Biomater. 77B (2006) 323–328. [8] G.X. Ni, W.W. Lu, K.Y. Chiu, Z.Y. Li, D.Y.T. Fong, K.D.K. Luk, J. Biomed. Mater. Res. B Appl. Biomater. 77B (2006) 409–415. [9] H. Oudadesse, A.C. Derrien, A. Lucas-Girot, Eur. Phys. J. Appl. Phys. 31 (2005) 217–223. [10] J. Davidovits, J. Therm. Anal. Calorim. 37 (1991) 1633–1656. [11] P. Duxson, J.L. Provis, G.C. Lukey, J.S.J. van Deventer, Cem. Concr. Res. 37 (2007) 1590–1597. [12] C. Menna, D. Asprone, C. Ferone, F. Colangelo, A. Balsamo, A. Prota, R. Cioffi, G. Manfredi, Compos. Part B 45 (2013) 1667–1676. [13] C. Ferone, F. Colangelo, R. Cioffi, F. Montagnaro, L. Santoro, Adv. Appl. Ceram. 112 (2013) 184–189. [14] G. Habert, J.B. d'Espinose de Lacaillerie, N. Roussel, J. Clean. Prod. 19 (2011) 1229–1238. [15] P. Duxson, A. Fernández-Jiménez, J. Provis, G. Lukey, A. Palomo, J. van Deventer, J. Mater. Sci. 42 (2007) 2917–2933. [16] W.M. Hendricks, A.T. Bell, C.J. Radke, J. Phys. Chem. 95 (1991) 9513–9518. [17] M.R. Anseau, J.P. Leung, N. Sahai, T.W. Swaddle, Inorg. Chem. 44 (2005) 8023–8032. [18] M.R. North, T.W. Swaddle, Inorg. Chem. 39 (2000) 2661–2665. [19] C. Ferone, G. Roviello, F. Colangelo, R. Cioffi, O. Tarallo, Appl. Clay Sci. 73 (2013) 42–50. [20] C. Ferone, F. Colangelo, G. Roviello, D. Asprone, C. Menna, A. Balsamo, A. Prota, R. Cioffi, G. Manfredi, Materials 6 (2013) 1920–1939. [21] P. Duxson, J.L. Provis, G.C. Lukey, S.W. Mallicoat, W.M. Kriven, J.S.J. van Deventer, Colloids Surf. A Physicochem. Eng. Asp. 269 (2005) 47–58. [22] E. Jämstorp, M. Strømme, G. Frenning, J. Pharm. Sci. 100 (2011) 4338–4348. [23] I. Lancellotti, M. Catauro, C. Ponzoni, F. Bollino, C. Leonelli, J. Solid State Chem. 200 (2013) 341–348. [24] K. MacKenzie, N. Rahner, M. Smith, A. Wong, J. Mater. Sci. (2010) 999–1007(Springer Netherlands). [25] H. Oudadesse, A. Derrien, M. Lefloch, J. Davidovits, J. Mater. Sci. 42 (2007) 3092–3098. [26] H. Oudadesse, A.C. Derrien, M. Mami, S. Martin, G. Cathelineau, L. Yahia, Biomed. Mater. 2 (2007) S59–S64. [27] T. Kokubo, H. Takadama, Biomaterials 27 (2006) 2907–2915. [28] J.A. Gadsden, Infrared spectra of minerals and related inorganic compounds, Butterworths, London, 1975. [29] R.G. Milkey, Am. Mineral. 45 (1960) 990–1007. [30] P. Duxson, G.C. Lukey, F. Separovic, J.S.J. van Deventer, Ind. Eng. Chem. Res. 44 (2005) 832–839. [31] G. Sandeep, H. Varma, T. Kumary, S. Suresh Babu, J. Annie, Trends Biomater. Artif. Organs Int. J. 19 (2006) 99–107. [32] C. Ohtsuki, T. Kokubo, T. Yamamuro, J. Non-Cryst. Solids 143 (1992) 84–92.
