Contents lists available at ScienceDirect

Journal of the Mechanical Behavior of Biomedical Materials

journal homepage: www.elsevier.com/locate/jmbbm

On the synthesis of nanostructured akermanite scaffolds via space holdermethod: The effect of the spacer size on the porosity and mechanicalproperties

Aliakbar Najafinezhad, Majid Abdellahi⁎, Sepideh Nasiri-Harchegani, Ali Soheily,Mohammadsaleh Khezri, Hamid Ghayour

Advanced Materials Research Center, Department of Materials Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran

A R T I C L E I N F O

Keywords:BioceramicsNanoparticlesAkermaniteScaffold

A B S T R A C T

In the present study, for the first time, the space holder method was used to prepare akermanite scaffolds withhigh porous structures, high interconnectivity, and high compressive strength, while the role of different spacersizes on the akermanite scaffold properties was also evaluated. The results showed that the increase in the NaClparticle size which was used as spacer leads to an increase of the pore size and interconnectivity and a decreaseof compressive strength. When the size of the spacer was 420–600 µm and more than 600 µm, a total porosity of82 and 83% and a compressive strength of 0.86 and 0.82 MPa were obtained, respectively. These values arehigher than those reported in previously studies and provide a great potential for akermanite to be used as bonesubstitute in tissue engineering. The in vitro bioactivity of the obtained akermanite scaffolds was alsoinvestigated.

1. Introduction

Akermanite (Ca2MgSi2O7), has attracted a special attention as aresult of several factors such as good mechanical properties, consider-able bioactivity, favorable ion release, and moderate degradation(Huang et al., 2009). There is limited work on the synthesis andcharacterization of akermanite bioceramics both in bulk form(Choudhary et al., 2015; Wu and Chang, 2004; Najafinezhad et al.2017) and as scaffolds (Wu et al., 2006; Han et al., 2014). This isdespite the fact that highly porous scaffold, can play an important rolein bone oxygenation and angiogenesis (Vallet-Reg, 2006). Pores withthe size of 150–900μm, allow for nutrient supply and waste removal ofthe cells grown on the scaffold (Roohani Esfahani et al., 2008). Theporous scaffolds require three dimensional interconnected porousstructures (Wu et al., 2007) which can supply the adequate space forcell migration, adhesion, proliferation, and differentiation (Wu et al.,2007).

Gel casting, polymeric sponge, rapid prototyping and laser sinteringare widely used to produce scaffold structures (Ghomi et al., 2016).Ribeiro et al. (2011) also introduced a low cost technique named “spaceholder” (SH) with a better control on the pore size and morphology,and the capability of near net shape manufacturing. Regarding thismethod, Ghomi et al. (2016) showed that the parameters including the

spacer content and compaction pressures are among the most im-portant parameters which affect the size and morphology of the poresand hence the ability of apatite formation. However, the effect of thespacer size on the pores size and morphology is still unknown. In thepresent study, the effect of spacer size on the pores’ size andmorphology was evaluated along with the compressive strength ofthe synthesized scaffolds.

2. Materials and methods

MgO, SiO2 and eggshell (as the calcium source) powders were usedas the starting materials. The mixture of raw materials was then ballmilled for 6 h, using a planetary ball mill. The amount of each rawmaterials is presented in Table 1.

The ball to powder weight ratio was 10:1 and the speed of the vialwas 400 rpm. Following that, the resultant powder was exposed to atemperature increasing to 900 °C (with a heating rate of 5° per min)with a holding time of 3 h to produce akermanite nanopowder. The SHmethod was then used to obtain akermanite scaffolds, during which anappropriate weight ratio of akermanite nanopowder to NaCl particles(as the spacer) with different sizes (250–300, 300–420, 420–600 and> 600 µm) was selected. The powder was then pressed at 100 MPa in acylindrical form with a size of 10 mm×14 mm (diameter×thickness)

http://dx.doi.org/10.1016/j.jmbbm.2017.01.002Received 29 November 2016; Received in revised form 26 December 2016; Accepted 2 January 2017

⁎ Corresponding author.E-mail address: Abdellahi@Pmt.iaun.ac.ir (M. Abdellahi).

Journal of the mechanical behavior of biomedical materials 69 (2017) 242–248

Available online 03 January 20171751-6161/ © 2017 Elsevier Ltd. All rights reserved.

MARK

and sintered in a kiln to 1200 °C at a rate of 5 °/min for 4 h. Thisprocess was continued by removing NaCl particles by soaking thesintered samples in deionized water for 24 h. The interconnected andtrue porosity were estimated using the method presented in Ref. Ghomiet al. (2016), which was based on the Archimedes principle. FTIR(Bomen MB 100), SEM (JSM/JEOL-6360), TEM (Hyundai, 100 keV),simultaneous thermal analyzer (SDTQ-600) were used for gatheringand analyzing the results. The compressive strength of samples wasdetermined using a Hounsfield H50KS universal testing machine, witha crosshead speed of 5 mm/min. Changes of ion concentrations of theSBF solution as a function of soaking time were determined byinductively coupled plasma-mass spectroscopy (Varian,USA). SBFsolution (for in vitro bioactivity tests) was prepared in laboratory withthe ionic concentration nearly similar to the human blood plasmaaccording to the Kokubo method. The sintered akermanite scaffoldswere soaked in the SBF solution for 28 days for in vitro bioactivity tests.Simultaneously the elemental compositions of the samples are ana-lyzed using Energy Dispersive Spectroscopy (EDS, Varian, USA).

3. Results and discussion

Fig. 1 shows an overview of the synthesis and characterizationanalyses of the pure akermanite nanopowder (Khajeh Sharafabadiet al., 2017). As shown in Fig. 1a, the O-Ca-O bending modes at411 cm−1, the O-Mg-O bending modes at 473 cm−1, the Ca=O band at589 cm−1, the O-Si-O bands at 641 cm−1 and 680 cm−1, the Si-O

stretching modes at 848 cm−1, 932 cm−1 and 981 cm−1 and thesymmetric stretching at 1021 cm−1 (which is assigned to the Si-O-Sivibration) are all consistent with the FTIR pattern of the pureakermanite (Choudhary et al., 2015; Khajeh Sharafabadi et al.,2017). On the other hand, the XRD result is in accordance with thatof the pure akermanite with the reference code 01-047-0087 (Fig. 1b).As can be seen in Fig. 1c which is related to the thermal analysis of the6 h milled powder, an endothermic peak starts at around 563 °C and isassociated with a mass loss of 1.9 mg (9.5–7.6 mg) which is related tothe transition of alpha to beta SiO2 (Kowalski, 2010). The secondendothermic peak at around 716 °C is associated with a mass lossranging from 7.6 to 7.3 mg which is related to the thermal decomposi-tion of CaCO3 to CaO (Halikia et al., 2001; Khajeh Sharafabadi et al.,2017; Kazemi et al., 2017). The exothermic peak at 900 °C is related tothe formation of akermanite. TEM nanograph of the akermanitesample is shown in the Fig. 1d. As can be seen, the powder particleshave spherical shapes with sizes less than 100 nm. It should be notedthat, osteoblasts are attached better to nano-biocomposites in compar-ison with their micro-biocomposite counterparts. Improved cytophili-city of nano-bioceramics in comparison with micro-grain ones has beenpreviously reported (Wang and Shaw, 2007). Sun et al. (2007) showedthat nanobioceramics facilitate the formation of periodontal ligamentcell regeneration through the reconstruction of alveolar bone. Thesereports confirm the importance of manufacturing bio-ceramics innanoscale structures. During the fabrication of scaffolds by using theSH method, various spacers were examined (Arifvianto and Zhou,2014). The use of NaCl as the spacer is preferred because of its lowcost, rapid dissolution in water and more precise control of the processvariables (Aida et al., 2016). Fig. 2 shows the cross-sectional morphol-ogy of the prepared akermanite scaffolds with different particle sizes ofNaCl. As can be seen, by increasing the spacer size, the pore size alsoincreases. This event is more perceptible when the spacer size increasesfrom 250–300 µm (Fig. 2a) to 300–420 µm (Fig. 2b) and from 300–420 µm to 420–600 µm (Fig. 2c). However, during the increase of the

Table 1The amount of each raw material used in this work.

2CaCO3 (egg shell powder)+MgO+2SiO2→Ca2MgSi2O7+2CO2

Material CaCO3 MgO SiO2

Content (wt%) ~55 ~11 33

Fig. 1. a) FTIR; b) XRD; c) DTA and TG; d) TEM analyses pf the akermanite nanopowder (Khajeh Sharafabadi et al., 2017).

A. Najafinezhad et al. Journal of the mechanical behavior of biomedical materials 69 (2017) 242–248

243

spacer size from 420–600 µm to more than 600 µm (Fig. 2d), nosignificant difference in the size of pores is observed.

Another issue that should be mentioned here is the tendency of thescaffolds toward a uniform pore size distribution (800–1000 µm) andspherical pore morphology. This means that NaCl particles with thesize of 400 µm to the top have a tendency to maintain their originalspherical shape during the compaction and sintering processes.Regarding the NaCl spacer, Khodaei et al. (2015) showed that cellmorphology was circular and more regular than that of the ammoniumbicarbonate spacer.

Fig. 2e shows a higher magnification of Fig. 2d. The red circle in thisfigure confirms that there are many micropores (less than 10 µm) onthe surface of the macropores. Microporosity ( < 50 µm) is essential forimmediate protein and cell adhesion, cell migration, and osseointegra-

tion (Reyes et al., 2012). Higher pore sizes ( > 300 µm) are required forenhanced new bone formation, greater bone ingrowth, and formationof capillaries.

In addition to porosities with a size more than 300 µm, an idealscaffold should have an interconnected porous structure (Gerhardt andBoccaccini, 2010). This proves that the scaffold is highly permeable forcell seeding, tissue ingrowth, and vascularization. Fig. 3 shows anoverview of the porosity and compressive strength variations of theakermanite scaffold. As Fig. 3a shows, the interconnected porosity ofthe scaffold increases as a result of the increase of the spacer size. Infact, with increasing the spacer size, the diameter of the macroporesincreases, so that they share their edges and therefore the intercon-nectivity enhances. An interconnected porosity of 82% has beenachieved for the scaffold containing NaCl with the size more than

Fig. 2. The cross-sectional morphology of the akermanite scaffolds with different particle sizes of NaCl; a) 250–300 µm; b) 300–420 µm; c) 420–600 µm; d) > 600 µm; e) > 600 µmwith high magnification. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

A. Najafinezhad et al. Journal of the mechanical behavior of biomedical materials 69 (2017) 242–248

244

600 µm. A total porosity of 83% was also obtained for these scaffolds(Fig. 3b). This increasing trend in porosity provides a leading stage todecreasing the compressive strength, as shown in Fig. 3c. According toWu et al. (2006), a compressive strength of 0.79 MPa and a totalporosity of 82% were obtained for akermanite scaffolds prepared bypolymer sponge method. However, the present work reported a highercompressive strength of 0.82 MPa at a higher total porosity of 83%. Thelocation-related instantaneous changes of the compressive strength ofthe obtained scaffolds containing NaCl with the size of more than600 µm confirm that these types of scaffolds have a stable compressivestrength of 0.82 MPa (Fig. 3d). This ensures the homogeneity of thepowder mixture before the pressing process.

Apatite formation on the surface of the bioactive materials duringthe soaking in SBF solution, ensures the bone regeneration in the invivo conditions. In other words, the formation of apatite layer is anecessary precondition for the direct binding of bioactive materials tothe living bones. During the in vitro study, SBF solution was used todetermine the bioactivity of akermanite scaffolds. Fig. 4 shows the SEMimages of the optimized akermanite scaffolds (those which obtained bythe spacers with the size of 420–600 µm), after soaking in SBF solutionfor 28 days. Different magnifications (Fig. 4a-g) show that the apatitelayer has covered the surface of the scaffold.

Fig. 5a shows the EDS analysis of the apatite formed on the surfaceof the scaffold. The EXD pattern shows the absence of the silicon peak(akermanite is a silicate biomaterial) and the presence of the calcium,phosphorus and oxygen as major peaks and magnesium as minor peak.This confirms the deposition of apatite layer on the immersed surface,as it replaces silicon and reduces the concentration of magnesium. So,the EDS analysis demonstrates that within 28 days of immersion in theSBF solution the apatite layer has covered the entire surface ofakermanite samples.

FTIR pattern as a tool to characterize of the absorption bands ofdifferent functional groups, is used to determine the apatite formation(Fig. 5b). In comparison with the FTIR pattern of the akermanite

samples before soaking process (Fig. 1a), O-Ca-O and O-Mg-O peakshave been replaced by the phosphate groups. This means that thecalcium and magnesium ions were hydrolyzed. The bending vibration(478 cm−1) and stretching vibration (971 and 1077 cm−1) of phosphategroup verify the mineralization of apatite on the immersed samplesurface.

Fig. 5c illustrates the changes of the ion concentration versus thesoaking time. As can be seen after 7 days of soaking, the phosphorouscontent quickly decreased as a result of the apatite formation duringthis period of time. Magnesium content also showed an increasingtrend, which confirms the resorbability of the prepared akermanitescaffolds. The dissolution of the Ca(II) ions from the sintered akerma-nite samples was also confirmed by increasing the Ca(II) content after14 days of soaking (the supersaturated SBF is obtained after 14 days ofsoaking due to the rapid increase of Ca ions content). According to theprevious researches (Ohtsuki et al., 1992), dissolution of the Ca(II) ionsfrom CaO-SiO2 glass produced silanol (Si-OH) groups on the surface asthe nucleation sites for apatite.

4. Conclusions

In this work, for the first time, the SH method was used to prepareakermanite scaffolds with a high porosity and compressive strength.The effect of the particle size of NaCl as spacer on the pore size,porosity and compressive strength of the scaffolds was also examined.According to the obtained results, nanostructured akermanite scaffoldsprepared in this work had a porosity more than 80%, and theircompressive strength was close to 0.9 MPa, i.e. about the strength ofthe spongy bone. The nano-sized structure and a large amount ofmicro/macrospores observed in the scaffolds can provide the condi-tions for high specific surface area and hence the increase of bioactivityand the release of ionic products. According to the ICP analysis, thephosphorous content quickly decreased as a result of the apatiteformation, after 7 days of soaking process. Magnesium content also

Fig. 3. a) Interconnected porosity; b) total porosity; c) compressive strength; d) location-related instantaneous changes of compressive strength of the akermanite scaffolds.

A. Najafinezhad et al. Journal of the mechanical behavior of biomedical materials 69 (2017) 242–248

245

Fig. 4. SEM image of akermanite scaffold (at different magnifications), after soaking in SBF solution for 28 days.

A. Najafinezhad et al. Journal of the mechanical behavior of biomedical materials 69 (2017) 242–248

246

showed an increasing trend in the SBF solution, which confirms theresorbability of the prepared akermanite scaffolds.

References

Huang, Y., Jin, X., Zhang, X., Sun, H., Tu, J., Tang, T., Chang, J., Dai, K., 2009. In vitroand in vivo evaluation of akermanite bioceramics for bone regeneration.Biomaterials 33, 5041–5048.

Choudhary, R., Koppala, S., Swamiappan, S., 2015. Bioactivity studies of calciummagnesium silicate prepared from eggshell waste by sol–gel combustion synthesis. J.Asian Ceram. Soc. 3, 173–177.

Wu, C., Chang, J., 2004. Synthesis and apatite-formation ability of akermanite. Mater.Lett. 58, 2415–2417.

Najafinezhad, A., Abdellahi, M., Ghayour, H., Soheily, A., Chami, A., Khandan, A., 2017.A comparative study on the synthesis mechanism, bioactivity and mechanicalproperties of three silicate bioceramics. Mater. Sci. Eng. C 72, 259–267.

Wu, C., Chang, J., Zhai, W., Ni, S., Wang, J., 2006. Porous akermanite scaffolds for bonetissue engineering: preparation, characterization, and in vitro studies. J. Biomed.Mater. Res. B Appl. Biomater. 78, 47–55.

Han, Z., Feng, P., Gao, C., Shen, Y., Shuai, C., Peng, S., 2014. Microstructure, mechanicalproperties and in vitro bioactivity of akermanite scaffolds fabricated by lasersintering. Biomed. Mater. Eng. 24, 2073–2080.

Vallet-Reg, M., 2006. Revisiting ceramics for medical applications. Dalton Trans. 44,5211–5220.

Roohani Esfahani, S.I., Tavangarian, F., Emadi, R., 2008. Nanostructured bioactive glasscoating on porous hydroxyapatite scaffold for strength enhancement. Mater. Lett. 62,3428–3430.

Wu, C.H., Chang, J., Zhai, W., Ni, S., 2007. A novel bioactive porous bredigite(Ca7MgSi4O16) scaffold with biomimetic apatite layer for bone tissue engineering. J.Mater. Sci. Mater. Med. 18, 857–864.

Ghomi, H., Emadi, R., Javanmard, S.H., 2016. Fabrication and characterization ofnanostructure diopside scaffolds using the space holder method: effect of differentspace holders and compaction pressures. Mater. Des. 91, 193–200.

Ribeiro, G.B.M., Trommer, R.M., dos Santos, L.A., Bergmann, C.P., 2011. Novel methodto produce β-TCP scaffolds. Mater. Lett. 65, 275–277.

Khajeh Sharafabadi, A., Abdellahi, M., Kazemi, A.H., Khandan, A., Ozada, N., 2017. Anovel and economical route for synthesizing akermanite (Ca2MgSi2O7) nano-bioceramic. Mater. Sci. Eng. C 71, 1072–1078.

Kowalski, J. St, 2010. Thermal aspects of temperature transformations in silica sand.Arch. Foundry Eng. 10, 111–114.

Fig. 5. a) EDS analysis of the apatite formed on the surface of akermanite scaffold; b) FTIR pattern of the akermanite samples after soaking in SBF; c) The changes of the ionconcentration versus the soaking time.

A. Najafinezhad et al. Journal of the mechanical behavior of biomedical materials 69 (2017) 242–248

247

Halikia, I., Zoumpoulakis, L., Christodoulou, E., Prattis, D., 2001. Kinetic study of thethermal decomposition of calcium carbonate by isothermal methods of analysis. Eur.J. Miner. Process. Environ. Prot. (1), 89–102.

Kazemi, A., Abdellahi, M., Khajeh Sharafabadi, A., Khandan, A., Ozada, N., 2017. Studyof in vitro bioactivity and mechanical properties of diopsidenano-bioceramicsynthesized by a facile method using eggshell as raw material. Mater. Sci. Eng. C 71,604–610.

Wang, J., Shaw, L.L., 2007. Morphology‐enhanced low‐temperature sintering ofnanocrystalline hydroxyapatite. Adv. Mater. 19, 2364–2369.

Sun, W., Chu, C., Wang, J., Zhao, H., 2007. Comparison of periodontal ligament cellsresponses to dense and nanophase hydroxyapatite. J. Mater. Sci. Mater. Med. 18,677–683.

Arifvianto, B., Zhou, J., 2014. Fabrication of metallic biomedical scaffolds with the spaceholder method: a review. Materials 7, 3588–3622.

Aida, S.F., Hijrah, M., Amirah, A., Zuhailawati, H., Anasyida, A.S., 2016. Effect of NaCl asa space holder in producing open cell A356 aluminium foam by gravity die castingprocess. Proc. Chem. 19, 234–240.

Khodaei, M., Meratian, M., Savabi, O., 2015. Effect of spacer type and cold compactionpressure on structural and mechanical properties of porous titanium scaffold.Powder. Metall. 58, 152–159.

Reyes, E.A.A., Patin, C.A.L., Diaz, B.J., Lefebvre, L.P., 2012. Structural characterizationand mechanical evaluation of bioactive glass 45S5 foams obtained by a powder technology approach. J. Am. Ceram. Soc. 95, 3776–3780.

Gerhardt, C., Boccaccini, A.R., 2010. Bioactive glass and glass-ceramic scaffolds for bonetissue engineering. Materials 3, 3867–3910.

Amirjani,A., Hafezi, M., Zamanian, A., Yasaee, M., Osman, N.A.A., 2016. Synthesis of nanostructured sphene and optimization of the mechanical properties of its scaffold viaresponse surface methodology. J. Adv. Mater. Proc. 4, 56–62.

A. Najafinezhad et al. Journal of the mechanical behavior of biomedical materials 69 (2017) 242–248

248