Materials Science and Engineering C 33 (2013) 499–506
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Morphological structure and characteristics of hydroxyapatite/β-cyclodextrin composite nanoparticles synthesized at different conditions Kyoung Dan Son, Young-Jin Kim ⁎ Department of Biomedical Engineering, Catholic University of Daegu, Gyeongsan 712-702, Republic of Korea
a r t i c l e
i n f o
Article history: Received 2 February 2012 Received in revised form 10 July 2012 Accepted 25 September 2012 Available online 29 September 2012 Keywords: Hydroxyapatite Nanoparticle β-Cyclodextrin Particle morphology Crystallinity
a b s t r a c t Hydroxyapatite (HA) nanoparticles were prepared simply in the presence of β-cyclodextrin (β-CD). Mixing sequence of ion precursors during the synthesis of HA greatly affected the morphological structure of nanoparticles. Ca–P showed only the sphere-like structure, however P–Ca exhibited the mixture of spherical and rod-like nanoparticles. The size of nanoparticles slightly decreased with increasing the content of β-CD. The HAs synthesized in the presence of β-CD agglomerated, leading to the formation of aggregates with a size of hundreds nanometer and narrow size distribution. FT-IR, XRD and XRF analyses confirmed that the HA nanoparticles could be synthesized with using β-CD, in which the Ca/P molar ratio was ranged from 1.72 to 1.70. The crystalline phase of these HA nanoparticles was similar to that of the stoichiometric HA. In addition, the content of β-CD contained in the products could influence the initial deposition rate of bone-like apatite on the surface of HA nanoparticles in simulated body fluid (SBF). © 2012 Elsevier B.V. All rights reserved.
1. Introduction Hydroxyapatite (HA) has been widely used in medical and dental applications as a material for damaged bones or teeth, important implant and scaffold material, and drug delivery agent, due to its biocompatibility and bioactivity as well as the similarity to the inorganic component of the hard tissues in natural bones [1–3]. So many kinds of applications require HA particles having more suitable properties for the use such as mechanical strength and thermal property. These properties usually depend on the particle size, particle size distribution, morphology and so on [4]. The morphology and dimensions of HA crystals in natural bone also affect its mechanical and thermal properties [5]. Therefore, in recent years significant research effort has been devoted to the morphology control method of HA particles. HA particles with various morphologies had been synthesized by means of solid-state reaction, sol–gel process, hydrothermal method and emulsion technique [4,6–8]. However, the common wet chemical precipitation method under atmospheric pressure was proved to be the most convenient way to prepare HA particles. In nature, the nucleation and growth of mineralized materials are often controlled by organic macromolecules such as proteins and polysaccharides. Bone and teeth consist of a small amount of organic matrix which manipulates the formation of apatite into distinct microstructures suitable for the mechanical forces [9]. A new development in biomaterials is the biomimetic synthesis of HA in polymer matrices to produce composites. Polymers with different molecular organizations ⁎ Corresponding author. Tel.: +82 53 850 3443; fax: +82 53 850 3292. E-mail address: [email protected] (Y.-J. Kim). 0928-4931/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2012.09.020
may be used as template to control the geometry of the apatite to mimic that found in bone. Polymers containing polar functional groups such as COOH and OH are useful components in the formation of nanosized HA because these ionizable side groups provide a greater affinity to positive Ca 2+ ions and the nucleation of HA crystals in the solution [10,11]. In other words, Ca 2+ ions easily dispersed in a solution at the molecular level because of the interaction within the polymer chain and could form the HA nanoparticles. β-Cyclodextrin (β-CD) is a doughnut-shaped cyclic oligosaccharide, which contains seven α-(1,4) linked glycosyl units in their macrocyclic structures [12]. It has a hydrophilic exterior and a hydrophobic internal cavity. β-CD can form non-covalent inclusion complex with a wide variety of hydrophobic compounds by intermolecular interactions such as electrostatic affinity and hydrophobic interaction, and complexation often alters the physicochemical and biological properties of guest molecules. The attractive property of β-CD is not only inclusion complexation with guest molecules but also many hydroxyl groups of the glucose units. These hydroxyl groups can promote the formation of intermolecular interaction between β-CD and hydrophilic molecules. With their high area to volume ratio, HA nanoparticles are expected to be excellent materials for biomedical applications, in which the morphology of nanoparticles have influence on their characteristics such as mechanical and thermal properties [4]. In the present study, an attempt was made to develop a new wet chemical precipitation method for the synthesis of HA nanoparticles with different morphology in the presence of β-CD as template. The prepared HA nanoparticles were systematically examined by considering their morphologies, compositions, chemical structures, crystalline phases and thermal properties. In addition, the ability of bone-like apatite formation on the surface of
500
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
synthesized HA nanoparticles by immersion in simulated body fluid (SBF) was evaluated. 2. Experimental 2.1. Materials Calcium nitrate tetrahydrate (Ca(NO3)2·4H2O), ammonium phosphate dibasic ((NH4)2HPO4), ammonium hydroxide solution (NH4OH) and β-cyclodextrin (β-CD) were purchased from Sigma–Aldrich Co. and used without further purification. Other reagents and solvents were commercially available and were used as received. 2.2. Synthesis of hydroxyapatite nanoparticles
The formation of bone-like apatite on the surface of synthesized HA nanoparticles was examined in simulated body fluid (SBF) with ion concentrations nearly equal to those of human blood plasma [14]. The solution was prepared by dissolving reagent grade NaCl, NaHCO3, KCl, NaHPO4·2H2O, MgCl2·6H2O, CaCl2·2H2O, Na2SO4 and (CH2OH)3CNH2 into deionized water and titrated with 1 M HCl to the pH of 7.4 at 37 °C. Immersion studies were performed by incubating the nanoparticles in 15 mL SBF solution in a Petri dish. All dishes were placed in a thermostatical shaking incubator (BioShaker MBR-022UP, Taitec Co.) at 37 °C. After 3, 5 and 7 days, the nanoparticles were gently washed with deionized water and dried in vacuo. 3. Results and discussion
A synthesis of hydroxyapatite (HA) nanoparticles is as follows. 10 mL of 0.1 M Ca(NO3)2·4H2O or 0.1 M (NH4)2HPO4 solution was first added dropwise to 60 mL of β-CD solutions with the contents of 0.05 (Ca–P5 or P–Ca5) and 0.1% (w/v) (Ca–P10 or P–Ca10), and then pH was adjusted to 10 by the addition of NH4OH. To this solution, a determined amount (Ca/P=1.67) of counterion aqueous solution (0.1 M (NH4)2HPO4 or 0.1 M Ca(NO3)2·4H2O solution) was added dropwise for 1 h. The mixture was stirred at 40 °C under air to induce the nucleation and growth of HA crystals in the β-CD matrix. After 24 h, the resultant HA nanoparticles were washed with distilled water several times and freeze dried. 2.3. Characterization of nanoparticles The morphologies of HA nanoparticles were observed by a field emission-scanning electronic microscope (FE-SEM, JSM-6335F, JEOL). Prior to SEM observation, all of the samples were coated with gold. For the same samples of FE-SEM, the spectrum of energy dispersive X-ray spectroscopy (EDX) was applied to analysis the elemental composition of HA nanoparticles. The nanoparticles were also observed by transmission electron microscopy (TEM, JEM-2010, JEOL). The particle size distribution was determined by the DLS technique using a Zetasizer Nano ZS (Malvern Instruments). UV-visible spectra were recorded on a Hitachi U-2900 spectrometer at 25 °C. The chemical composition of synthesized HA nanoparticles was determined by measuring the adsorption of the samples at 177.434 nm for P and 183.994 nm for Ca using an inductively coupled plasma-optical emission spectroscopy (ICP-OES, Varian 720-ES). FT-IR spectra of the samples were obtained with an ALPHA spectrometer (Brucker Optics) in the wavenumber range 400–4000 cm–1. X-ray diffraction (XRD) measurements were carried out to characterize the crystalline phase of HA nanoparticles with a Panalytical X-ray diffractometer X'Pert Pro with Cu Kα radiation at 40 kV/30 mA. The diffractograms were scanned in a 2θ range of 20–60° at a rate of 2°/min. From the XRD data, the crystallinity of the HA nanoparticles was calculated according to the following equation [13]: h i Crystallinity ð% Þ ¼ 1– V 112=300 =I300 100
2.4. Simulated body fluid immersion test
ð1Þ
where V112/300 is the intensity of the hollow between (112) and (300) peaks and I300 is the intensity of the (300) peak. Ca/P molar ratio of HAs was analyzed with X-ray fluorescence spectroscopy (XRF, ZSX Primus II, Rigaku). The thermal stability of particles was evaluated by thermogravimetric analysis (TGA, Q500, TA Instruments). The TGA measurements were carried out under nitrogen atmosphere at a heating rate of 5 °C/min from 30 to 800 °C, in which all of the samples were dried in vacuo at 100 °C for 48 h prior to the measurement.
3.1. Morphology and particle size distribution of nanoparticles The application of hydroxyapatite (HA) in an artificial implant is limited because of its fragility and poor mechanical properties [15]. Attempts have been made to improve the mechanical and biological performance of HA-based nanocomposites. Generally polymers have good toughness and flexibility whereas their bioactivity is low when compared with bioactive ceramics [16]. HA reinforced in a polymer matrix has many viable clinical uses including prosthetic bone cement, joint replacement and dental implants as well as filling of bone defects [17–19]. Moreover, the biological properties of these nanoparticles would be expected to be strongly influenced by both their intrinsic (chemical compositions and crystalline structures) and extrinsic properties (particle size and morphology). Therefore, a great deal of attention has been focused on the synthesis of HA particles with a controlled morphology [11,20,21]. In the present study, the HA nanoparticles were prepared with different concentration of β-cyclodextrin (β-CD) and synthesis procedure (mixing sequence of ion precursors). The contents of β-CD in the solutions during the synthesis of HA nanoparticles were 0.05 (Ca–P5 or P–Ca5) and 0.1% (w/v) (Ca–P10 or P–Ca10). Ca–P means that the HA nanoparticles were synthesized by mixing previously prepared β-CD solutions containing Ca 2 + ion precursors (Ca(NO3)2·4H2O) with PO43– ion precursors ((NH4)2HPO4). On the other hand, P–Ca nanoparticles were prepared by the reaction of β-CD solutions containing PO43– ion precursors with Ca 2 + ion precursors. Fig. 1 shows the morphological structure of HA nanoparticles. All of the resulting nanoparticles exhibited spherical shape with a size of below 100 nm and P–Ca showed higher particle size than that of Ca–P. The size of nanoparticles slightly decreased with increasing the content of β-CD. The structure of β-CD provides a molecule shape like a segment of hollow cone with an exterior hydrophilic surface and interior electron-rich hydrophobic cavity, which is capable of forming stable and supramolecular structures with various molecules including organometallic compounds [22]. However, even though the size of single particle was below 100 nm, the aggregate formation of HAs synthesized in the presence of β-CD was observed with a size of hundreds nanometer as shown in Fig. 1. The HA nanoparticles were also characterized with a TEM to confirm the morphological structure and aggregation behavior (Fig. 2). Ca–P showed the sphere-like structure, which was in good agreement with the result of SEM observation. It was previously reported that polymers containing polar functional groups such as COOH and OH have been found to be useful for the nucleation of HA crystals in the solution because these ionizable side groups provide a site for the selective adhesion of Ca 2+ ions [10,11]. β-CD is a polar and water soluble oligomer with OH side group, which can be cross-linked with each other through the hydrogen bonding interaction and inclines to form the sphere-like structure [23]. Upon ionization, the negatively charged surface groups can provide binding sites for the Ca2+ ions present in
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
501
Fig. 1. SEM micrographs of HA nanoparticles synthesized with different content of β-CD and mixing sequence of ion precursor: (a) 0.05% (w/v) (Ca–P5), (b) 0.1% (w/v) (Ca–P10), (c) 0.05% (w/v) (P–Ca5) and (d) 0.1% (w/v) (P–Ca10).
the solution. As the Ca2+ ions accumulate on the surface, the surface gains an overall positive charge. These positively charged surface sites will combine with negatively charged PO43– ions initiating the nucleation and growth of the spherical HA nanoparticles. The increase of β-CD content leads to the formation of larger number of nuclei and therefore smaller HA crystallite size [10]. In the mean time, P–Ca exhibited the mixture of spherical and rod-like nanoparticles. Since (NH4)2HPO4 takes the mixed form of HPO42– and PO43– at pH = 10, the element P in the β-CD solution should be in the mixed form of HPO42– and PO43–, in which the oxygen atoms of HPO42– and PO43– can accept protons from the OH side groups of β-CD [24]. Thus these ions can be entirely inserted between the hollow cone of β-CD and form the complexes with β-CD through the multiple hydrogen bonds [25]. These complexes can strongly interact with the Ca2+ ions to nucleate the HA nanoparticles. As a result, the initial nucleation is preferentially triggered at the positions of hydroxyl groups, and the rod-like structure may be partially produced. Moreover, P–Ca is probably synthesized through the development of coordination anions Ca–P6O24 which are able to build nanowires or rod-like nanoparticles [20]. The formation of aggregates was also clearly observed in all the samples. The aggregation of HA nanoparticles should be restrained for convenient use. Therefore, the suspensions of HA nanoparticles were prepared for the determination of aggregate particle size. As shown in Fig. 3, not merely the concentration of β-CD but the method for the synthesis of HA, mixing sequence of ion precursors, affected the average
particle size of aggregates. The average particle size slightly decreased with increasing the β-CD concentration from 525 ±97 nm (Ca–P5) to 480 ± 91 nm (Ca–P10). In the case of P–Ca, the average particle size enormously reduced from 836 ± 60 nm to 580 ±71 nm with the increase of β-CD content from 0.05% (w/v) to 0.1% (w/v). Furthermore, the average particle size of P–Ca was higher than that of Ca–P as mentioned in the result of SEM observation. It was concluded that the β-CD content and the synthesis procedure can affect the aggregation behavior of nanoparticles. That is to say, the particle with smaller single particle size gives the smaller aggregate particle size. 3.2. Chemical structure and crystalline phase FT-IR analysis was carried out for identifying the functional groups present in the HA nanoparticles, which in turn provided information about the constitution and phase composition of the products synthesized with different content of β-CD and synthesis procedure. As shown in Fig. 4, all of the samples synthesized in the presence of β-CD exhibited characteristic absorption bands for the vibrational modes of PO43– appeared at around 1090, 1016, 959, 596 and 557 cm–1, and the bands at 3340 and 1640 cm–1 associated with OH of β-CD and absorbed H2O [23]. In addition, the band observed at 1420 cm–1 was attributed to the substitution of CO23– ions in the place of PO43– ions and confirmed the substitution of CO23– in apatite structure, which was stronger in P–Ca nanoparticles [2]. These CO23– ions were formed by the reaction of CO2
502
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
Fig. 2. TEM micrographs of HA nanoparticles synthesized with different content of β-CD and mixing sequence of ion precursor: (a) 0.05% (w/v) (Ca–P5), (b) 0.1% (w/v) (Ca–P10), (c) 0.05% (w/v) (P–Ca5) and (d) 0.1% (w/v) (P–Ca10).
present in the atmosphere with OH– ions of reaction medium. Hence P– Ca nanoparticles possibly contain the higher amount of CaCO3 as a mixture with HA. Moreover, the HA nanoparticles showed absorption bands assigned to OH– of HA at 3570 and 630 cm–1. These data suggest that the HA nanoparticles could be simply synthesized by the use of β-CD as template, which were organic–inorganic composites produced
by the complexation of inorganic HA with β-CD. Additionally, the chemical structure of products was a little changed with the synthesis procedure. To verify the complexation behavior between inorganic HA and β-CD, the change of UV absorption spectra was determined using 1% (w/v) solution of all samples (Fig. 5). Commercial HA nanoparticles without β-CD
Fig. 3. Particle size distribution of (a) Ca–P5, (b) Ca–P10, (c) P–Ca5 and (d) P–Ca10 nanoparticles.
Fig. 4. FT-IR spectra of (a) commercial HA, (b) Ca–P5, (c) Ca–P10, (d) P–Ca5 and (e) P–Ca10 nanoparticles.
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
503
Table 1 Crystallinity and chemical composition of the HA nanoparticles synthesized with different content of β-CD and mixing sequence of ion precursor. Sample
Commercial HA Ca–P5 Ca–P10 P–Ca5 P–Ca10
Fig. 5. UV-visible spectra of commercial and synthesized HA nanoparticles.
revealed specific absorption peak at around 206 nm. However, the synthesized HA nanoparticles, Ca–P10 and P–Ca10, showed amplified intensity of absorption, and specific absorption peak due to HA shifted to 209 nm. This result can prove the complexation behavior through the intermolecular interaction and the formation of composites consisted of inorganic HA and β-CD. Moreover, the composite formation between HA and β-CD in the resultant HA nanoparticles was also confirmed by the EDX analysis. As shown in Fig. 6, the four characteristic peaks ascribed to carbon, oxygen, phosphorous and calcium atoms were observed at 0.26, 0.52, 2.02 and 3.68 keV, respectively. The peak intensity assigned to carbon atom increased with the increment of β-CD content, however, the peaks due to phosphorous and calcium atoms reduced. To assess the chemical composition of HA nanoparticles, 10 mg of each samples was dissolved in 10 mL of 1 N HCl solution, and Ca or P element was measured with an ICP–OES. The synthesized HA nanoparticles contained less amount of inorganic constituent compared with commercial HA (Table 1). In addition, this inorganic constituent amount reduced slightly with increasing the content of β-CD. This clearly explains on the presence of β-CD in the synthesized HA nanoparticles. The crystalline phases of the HA nanoparticles were investigated by means of XRD (Fig. 7). Commercial HA nanoparticles exhibited the diffraction peaks at around 25.9°, 28.3°, 29.1°, 31.8°, 32.2°, 32.8°, 34.2°, 39.9°, 42.1°, 43.9°, 45.5°, 46.7°, 48.2°, 49.6°, 50.5°, 51.3°, 52.2° and 53.3°, which were imputed to the (002), (102), (210), (211), (112), (300), (202), (310), (311), (113), (203), (222), (312), (213), (321), (400), (402) and (004) planes [2,22]. This XRD pattern was exactly
Fig. 6. EDX spectra of (a) Ca–P5, (b) Ca–P10, (c) P–Ca5 and (d) P–Ca10 nanoparticles.
Crystallinity (%)
55.1 13.4 12.5 7.7 7.6
Ca (mM/g)
12.60 10.18 10.03 10.13 9.76
P (mM/g)
7.75 5.91 5.78 5.88 5.65
Ca/P molar ratio ICP-OES
XRF
1.63 1.72 1.74 1.72 1.73
1.67 1.72 1.72 1.72 1.70
matched with the structural data of stoichiometric HA described in the Powder Diffraction File (JCPDS 09-432). In addition, the XRD patterns of synthesized HA nanoparticles showed the peaks attributed to the HA crystalline phase at around 25.8°, 29.1°, 31.9°, 32.4°, 32.9°, 34.1°, 39.9°, 43.9°, 46.7°, 49.6° and 53.3°, which reflected characteristic of the (002), (210), (211), (112), (300), (202), (310), (113), (222), (213) and (004) planes. However, all the peaks were broad diffraction peaks indicating a poorly crystallized HA phase. The diffraction peaks became wider and less intense with increasing the concentration of β-CD. Additionally mixing sequence of ion precursors also influenced these diffraction peaks, leading that P–Ca showed more broad diffraction
Fig. 7. X-ray diffraction patterns of (a) commercial HA, (b) Ca–P5, (c) Ca–P10, (d) P–Ca5 and (e) P–Ca10 nanoparticles.
Fig. 8. TGA curves of (a) commercial HA, (b) Ca–P5, (c) Ca–P10, (d) P–Ca5 and (e) P–Ca10 nanoparticles.
504 K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
Fig. 9. SEM micrographs of (a) Ca–P5, (b) Ca–P10, (c) P–Ca5 and (d) P–Ca10 nanoparticles after 3, 5 and 7 days of immersion in SBF.
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
peaks. This is owing to the complex formation of HA with amorphous β-CD. Moreover, the isomorphous substitution of PO43– by CO23– derived from the absorption of CO2 in the air during preparation process of the HA nanoparticles affected the decrease of crystallinity [22]. The crystallinity of synthesized HA nanoparticles was calculated from the XRD data, which was lower than that of commercial HA and slightly decreased with the increase of β-CD content as shown in Table 1. As mentioned above, mixing sequence of ion precursors also affected the crystallinity which was 12.5–13.4% for Ca–P and 7.6–7.7% for P–Ca. This result is related to the complex formation HA with amorphous β-CD and the β-CD amount included in the composites. Previous study reported that the calcium phosphate (CaP) spheres synthesized with poly(acrylic acid) (PAA) exhibited only a broad diffraction peak, indicating that the CaP phase was poorly crystallized [11]. The CaPs subjected to thermal treatment at 550 °C for the elimination of amorphous polymer displayed strong peaks that corresponded well to those of stoichiometric HA crystal, owing to the increase in crystallinity. The Ca/P molar ratio of HA nanoparticles was determined by XRF, which was ranged from 1.70 to 1.72 and similar to the result of ICP-OES (Table 1). The Ca/P molar ratio was hardly touched with the β-CD content and mixing sequence of ion precursors. These results mean that HA can be easily fabricated by the use of β-CD. Sugar molecules are known to coordinate Ca2+ ions or PO43– ions in various configurations and this effect can be enhanced at the crystal surfaces of HA by the high spatial charge density and possibility of hydrogen bonding to the surface phosphate groups [26]. Afterwards, counterions (PO43– ions or Ca2+ ions) will accumulate at calcium or phosphate complexes. Mineral nuclei formed at these sites will gradually grow on the surface of β-CD, leading to the formation of HA crystal. Consequently, the use of β-CD as matrix is useful for the synthesis of HA. 3.3. Thermal stability Thermal property of the synthesized HA nanoparticles was investigated. It is well known that the thermal property of HA is strongly influenced by its Ca/P molar ratio and hence the decomposition temperature of HA can give an evidence to its Ca/P molar ratio qualitatively [2]. Fig. 8 shows the typical TGA curve where the amount of weight loss is plotted against the temperature. All of the samples synthesized in this study had very similar Ca/P molar ratio and thus exhibited the same pattern of weight loss comprised of three steps. The first step takes place between 30 °C and about 150 °C. This step mainly is assigned to the evaporation of absorbed water and the second step in temperature range of 200–450 °C is maybe imputed to the elimination of crystalline water and the decomposition of β-CD [23,27,28]. Additionally, the third step is probably the loss of constitution water of HA, which can change to oxyapatite (Ca10(PO4)6O). The total weight loss at 800 °C increased from 14.2% to 19.3% with the increment of β-CD content and the change of synthesis procedure. The decomposition temperature of β-CD is about 308 °C, resulting that the heightened weight loss is attributed to the decomposition of β-CD. The HA nanoparticles synthesized with β-CD contain some amount of β-CD even after repeated washing. Therefore, the mass fraction of β-CD included in the resulting HA nanoparticles was calculated from the TGA data, which was 6.8% for Ca–P5, 10.0% for Ca–P10, 9.3% for P– Ca5 and 11.9% for P–Ca10. This is a result of the complex formation via electrostatic interaction and hydrogen bonding between HA and β-CD. 3.4. Simulated body fluid immersion test The essential requirement for an artificial material to bond to living bone is the formation of bone-like apatite on its surface when implanted in the living body, and that this in vivo apatite formation can be reproduced in a simulated body fluid (SBF) with ion concentrations
505
nearly equal to those of human blood plasma [14]. This means that in vivo bone bioactivity of a material can be predicted from the apatite formation on its surface in SBF. After 3 days of immersion in SBF, the deposition of bone-like apatite was observed at some sites on the surface of Ca–P5 and P–Ca5 nanoparticles as shown in Fig. 9. After 5 and 7 days of mineralization in SBF, these nanoparticles were almost completely covered with apatite structures. On the other hand, the formation of bone-like apatite was hardly detected on the surface of Ca–P10 and P–Ca10 nanoparticles after 3 days of immersion in SBF. After 7 days of mineralization in SBF, the whole surfaces of Ca–P10 and P–Ca10 nanoparticles were covered by a layer of apatite. In addition, the aggregation of HA nanoparticles severely proceeded by immersion in SBF and the resulting particles exhibited cauliflower-shaped morphology. These results mean that the β-CD content included in the HA nanoparticles hugely affected the formation of bone-like apatite. The initial deposition rate of apatite on the surface of HA nanoparticles reduced with increasing the β-CD content. 4. Conclusions The HA nanoparticles have attracted a great deal of attention in biomedical applications. An increase in the surface area of HA leads to a greater proportion of its atoms or molecules being displayed on its surface rather than in the interior. As a result, it enhances the biological performances such as cell adhesion, proliferation and differentiation. In the present study, a novel and simple reaction for the preparation of HA nanoparticles was successfully developed by the use of β-CD as template. The β-CD content slightly affected the particle size of products. In addition, the synthesis procedure (mixing sequence of ion precursors) also influenced the particles size and morphological structure of nanoparticles. The HA nanoparticles synthesized in the presence of β-CD exhibited the aggregation behavior. The HA nanoparticles had very narrow size distribution, and the composition and crystalline phase were similar to those of the stoichiometric HA. However, the crystallinity of synthesized HA nanoparticles was low compared with that of commercial HA. These results are ascribed to the complexation behavior between inorganic HA and β-CD through electrostatic interaction and hydrogen bonding, causing the production of composites. It is concluded that this process offers a simple and potentially safe new approach for preparing the HA nanoparticles. Acknowledgements This research was financially supported by the Ministry of Knowledge Economy (MKE), Korea Institute for Advancement of Technology (KIAT) through the Inter-ER Cooperation Projects. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]
T.Y. Liu, S.Y. Chen, D.M. Liu, S.C. Liou, J. Control. Release 107 (2005) 112–121. B. Viswanath, N. Ravishankar, Biomaterials 29 (2008) 4855–4863. J.L. Xu, K.A. Khor, J.J. Sui, J.H. Zhang, W.N. Chen, Biomaterials 30 (2009) 5385–5391. S. Bose, S.K. Saha, Chem. Mater. 15 (2003) 4464–4469. H.M. Kim, C. Rey, M.J. Glimcher, J. Bone Miner. Res. 10 (1995) 1589–1601. G. Bezzi, G. Celotti, E. Landi, T.M.G. La Torretta, I. Sopyan, A. Tampieri, Mater. Chem. Phys. 78 (2003) 816–824. S. Pramanik, A.K. Agarwal, K.N. Rai, A. Garg, Ceram. Int. 33 (2007) 419–426. Y.X. Sun, G.S. Guo, D.L. Tao, Z.H. Wang, J. Phys. Chem. Solids 68 (2007) 373–377. S.I. Stupp, P.V. Braun, Science 277 (1997) 1242–1248. S. Mollazadeh, J. Javadpour, A. Khavandi, Ceram. Int. 33 (2007) 1579–1583. P. Wang, S.W. Yook, S.H. Jun, Y.L. Li, M. Kim, H.E. Kim, Y.H. Koh, Mater. Lett. 63 (2009) 1207–1209. H. Liu, C. Liu, X. Yang, S. Zeng, Y. Xiong, W. Xu, Anal. Chim. Acta 628 (2008) 87–94. Y.X. Pang, X. Bao, J. Eur. Ceram. Soc. 23 (2003) 1697–1704. T. Kokubo, Biomaterials 12 (1991) 155–163. M. Ohgaki, K. Yamashite, J. Am. Ceram. Soc. 86 (2003) 1440–1442. J. Huang, Y.W. Lin, X.W. Fu, S.M. Best, R.A. Brooks, N. Rushton, W. Bonfield, J. Mater. Sci. Mater. Med. 18 (2007) 2151–2157. S. Morita, K. Furuya, K. Ishihara, N. Nakabayashi, Biomaterials 19 (1998) 1601–1606.
506
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
[18] R.W. Arcis, A. Lopez-Macipe, M. Toledano, E. Osorio, R. Rodriguez-Clemente, J. Murtra, M.A. Fanovich, C.D. Pascual, Dent. Mater. 18 (2002) 49–57. [19] S. Haque, I. Rehman, J.A. Darr, Langmuir 23 (2007) 6671–6676. [20] P. Wang, C. Li, H. Gong, X. Jiang, H. Wang, K. Li, Powder Technol. 203 (2010) 315–321. [21] H. Zhang, M. Zhang, Mater. Chem. Phys. 126 (2011) 642–648. [22] F. Hapiot, S. Tilloy, E. Monflier, Chem. Rev. 106 (2006) 747–781. [23] X.F. Xiao, R.F. Liu, C.F. Qiu, D.C. Zhu, F. Liu, Mater. Sci. Eng. C 29 (2009) 785–790.
[24] A. Oyane, K. Onuma, A. Ito, H.M. Kim, T. Kokubo, T. Nakamura, J. Biomed. Mater. Res. 64A (2003) 339–348. [25] H. Luecke, F.A. Quiocho, Nature 347 (1990) 402–406. [26] D. Walsh, J.L. Kingston, B.R. Heywood, S. Mann, J. Cryst. Growth 133 (1993) 1–12. [27] Y. Dong, W. Dong, Y. Cao, Z. Han, Z. Ding, Catal. Today 175 (2011) 346–355. [28] Y. Yang, X.F. Li, J.B. Chen, H.L. Chen, X.M. Bao, Chem. Phys. Lett. 373 (2003) 22–27.
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Morphological structure and characteristics of hydroxyapatite/β-cyclodextrin composite nanoparticles synthesized at different conditions Kyoung Dan Son, Young-Jin Kim ⁎ Department of Biomedical Engineering, Catholic University of Daegu, Gyeongsan 712-702, Republic of Korea
a r t i c l e
i n f o
Article history: Received 2 February 2012 Received in revised form 10 July 2012 Accepted 25 September 2012 Available online 29 September 2012 Keywords: Hydroxyapatite Nanoparticle β-Cyclodextrin Particle morphology Crystallinity
a b s t r a c t Hydroxyapatite (HA) nanoparticles were prepared simply in the presence of β-cyclodextrin (β-CD). Mixing sequence of ion precursors during the synthesis of HA greatly affected the morphological structure of nanoparticles. Ca–P showed only the sphere-like structure, however P–Ca exhibited the mixture of spherical and rod-like nanoparticles. The size of nanoparticles slightly decreased with increasing the content of β-CD. The HAs synthesized in the presence of β-CD agglomerated, leading to the formation of aggregates with a size of hundreds nanometer and narrow size distribution. FT-IR, XRD and XRF analyses confirmed that the HA nanoparticles could be synthesized with using β-CD, in which the Ca/P molar ratio was ranged from 1.72 to 1.70. The crystalline phase of these HA nanoparticles was similar to that of the stoichiometric HA. In addition, the content of β-CD contained in the products could influence the initial deposition rate of bone-like apatite on the surface of HA nanoparticles in simulated body fluid (SBF). © 2012 Elsevier B.V. All rights reserved.
1. Introduction Hydroxyapatite (HA) has been widely used in medical and dental applications as a material for damaged bones or teeth, important implant and scaffold material, and drug delivery agent, due to its biocompatibility and bioactivity as well as the similarity to the inorganic component of the hard tissues in natural bones [1–3]. So many kinds of applications require HA particles having more suitable properties for the use such as mechanical strength and thermal property. These properties usually depend on the particle size, particle size distribution, morphology and so on [4]. The morphology and dimensions of HA crystals in natural bone also affect its mechanical and thermal properties [5]. Therefore, in recent years significant research effort has been devoted to the morphology control method of HA particles. HA particles with various morphologies had been synthesized by means of solid-state reaction, sol–gel process, hydrothermal method and emulsion technique [4,6–8]. However, the common wet chemical precipitation method under atmospheric pressure was proved to be the most convenient way to prepare HA particles. In nature, the nucleation and growth of mineralized materials are often controlled by organic macromolecules such as proteins and polysaccharides. Bone and teeth consist of a small amount of organic matrix which manipulates the formation of apatite into distinct microstructures suitable for the mechanical forces [9]. A new development in biomaterials is the biomimetic synthesis of HA in polymer matrices to produce composites. Polymers with different molecular organizations ⁎ Corresponding author. Tel.: +82 53 850 3443; fax: +82 53 850 3292. E-mail address: [email protected] (Y.-J. Kim). 0928-4931/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2012.09.020
may be used as template to control the geometry of the apatite to mimic that found in bone. Polymers containing polar functional groups such as COOH and OH are useful components in the formation of nanosized HA because these ionizable side groups provide a greater affinity to positive Ca 2+ ions and the nucleation of HA crystals in the solution [10,11]. In other words, Ca 2+ ions easily dispersed in a solution at the molecular level because of the interaction within the polymer chain and could form the HA nanoparticles. β-Cyclodextrin (β-CD) is a doughnut-shaped cyclic oligosaccharide, which contains seven α-(1,4) linked glycosyl units in their macrocyclic structures [12]. It has a hydrophilic exterior and a hydrophobic internal cavity. β-CD can form non-covalent inclusion complex with a wide variety of hydrophobic compounds by intermolecular interactions such as electrostatic affinity and hydrophobic interaction, and complexation often alters the physicochemical and biological properties of guest molecules. The attractive property of β-CD is not only inclusion complexation with guest molecules but also many hydroxyl groups of the glucose units. These hydroxyl groups can promote the formation of intermolecular interaction between β-CD and hydrophilic molecules. With their high area to volume ratio, HA nanoparticles are expected to be excellent materials for biomedical applications, in which the morphology of nanoparticles have influence on their characteristics such as mechanical and thermal properties [4]. In the present study, an attempt was made to develop a new wet chemical precipitation method for the synthesis of HA nanoparticles with different morphology in the presence of β-CD as template. The prepared HA nanoparticles were systematically examined by considering their morphologies, compositions, chemical structures, crystalline phases and thermal properties. In addition, the ability of bone-like apatite formation on the surface of
500
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
synthesized HA nanoparticles by immersion in simulated body fluid (SBF) was evaluated. 2. Experimental 2.1. Materials Calcium nitrate tetrahydrate (Ca(NO3)2·4H2O), ammonium phosphate dibasic ((NH4)2HPO4), ammonium hydroxide solution (NH4OH) and β-cyclodextrin (β-CD) were purchased from Sigma–Aldrich Co. and used without further purification. Other reagents and solvents were commercially available and were used as received. 2.2. Synthesis of hydroxyapatite nanoparticles
The formation of bone-like apatite on the surface of synthesized HA nanoparticles was examined in simulated body fluid (SBF) with ion concentrations nearly equal to those of human blood plasma [14]. The solution was prepared by dissolving reagent grade NaCl, NaHCO3, KCl, NaHPO4·2H2O, MgCl2·6H2O, CaCl2·2H2O, Na2SO4 and (CH2OH)3CNH2 into deionized water and titrated with 1 M HCl to the pH of 7.4 at 37 °C. Immersion studies were performed by incubating the nanoparticles in 15 mL SBF solution in a Petri dish. All dishes were placed in a thermostatical shaking incubator (BioShaker MBR-022UP, Taitec Co.) at 37 °C. After 3, 5 and 7 days, the nanoparticles were gently washed with deionized water and dried in vacuo. 3. Results and discussion
A synthesis of hydroxyapatite (HA) nanoparticles is as follows. 10 mL of 0.1 M Ca(NO3)2·4H2O or 0.1 M (NH4)2HPO4 solution was first added dropwise to 60 mL of β-CD solutions with the contents of 0.05 (Ca–P5 or P–Ca5) and 0.1% (w/v) (Ca–P10 or P–Ca10), and then pH was adjusted to 10 by the addition of NH4OH. To this solution, a determined amount (Ca/P=1.67) of counterion aqueous solution (0.1 M (NH4)2HPO4 or 0.1 M Ca(NO3)2·4H2O solution) was added dropwise for 1 h. The mixture was stirred at 40 °C under air to induce the nucleation and growth of HA crystals in the β-CD matrix. After 24 h, the resultant HA nanoparticles were washed with distilled water several times and freeze dried. 2.3. Characterization of nanoparticles The morphologies of HA nanoparticles were observed by a field emission-scanning electronic microscope (FE-SEM, JSM-6335F, JEOL). Prior to SEM observation, all of the samples were coated with gold. For the same samples of FE-SEM, the spectrum of energy dispersive X-ray spectroscopy (EDX) was applied to analysis the elemental composition of HA nanoparticles. The nanoparticles were also observed by transmission electron microscopy (TEM, JEM-2010, JEOL). The particle size distribution was determined by the DLS technique using a Zetasizer Nano ZS (Malvern Instruments). UV-visible spectra were recorded on a Hitachi U-2900 spectrometer at 25 °C. The chemical composition of synthesized HA nanoparticles was determined by measuring the adsorption of the samples at 177.434 nm for P and 183.994 nm for Ca using an inductively coupled plasma-optical emission spectroscopy (ICP-OES, Varian 720-ES). FT-IR spectra of the samples were obtained with an ALPHA spectrometer (Brucker Optics) in the wavenumber range 400–4000 cm–1. X-ray diffraction (XRD) measurements were carried out to characterize the crystalline phase of HA nanoparticles with a Panalytical X-ray diffractometer X'Pert Pro with Cu Kα radiation at 40 kV/30 mA. The diffractograms were scanned in a 2θ range of 20–60° at a rate of 2°/min. From the XRD data, the crystallinity of the HA nanoparticles was calculated according to the following equation [13]: h i Crystallinity ð% Þ ¼ 1– V 112=300 =I300 100
2.4. Simulated body fluid immersion test
ð1Þ
where V112/300 is the intensity of the hollow between (112) and (300) peaks and I300 is the intensity of the (300) peak. Ca/P molar ratio of HAs was analyzed with X-ray fluorescence spectroscopy (XRF, ZSX Primus II, Rigaku). The thermal stability of particles was evaluated by thermogravimetric analysis (TGA, Q500, TA Instruments). The TGA measurements were carried out under nitrogen atmosphere at a heating rate of 5 °C/min from 30 to 800 °C, in which all of the samples were dried in vacuo at 100 °C for 48 h prior to the measurement.
3.1. Morphology and particle size distribution of nanoparticles The application of hydroxyapatite (HA) in an artificial implant is limited because of its fragility and poor mechanical properties [15]. Attempts have been made to improve the mechanical and biological performance of HA-based nanocomposites. Generally polymers have good toughness and flexibility whereas their bioactivity is low when compared with bioactive ceramics [16]. HA reinforced in a polymer matrix has many viable clinical uses including prosthetic bone cement, joint replacement and dental implants as well as filling of bone defects [17–19]. Moreover, the biological properties of these nanoparticles would be expected to be strongly influenced by both their intrinsic (chemical compositions and crystalline structures) and extrinsic properties (particle size and morphology). Therefore, a great deal of attention has been focused on the synthesis of HA particles with a controlled morphology [11,20,21]. In the present study, the HA nanoparticles were prepared with different concentration of β-cyclodextrin (β-CD) and synthesis procedure (mixing sequence of ion precursors). The contents of β-CD in the solutions during the synthesis of HA nanoparticles were 0.05 (Ca–P5 or P–Ca5) and 0.1% (w/v) (Ca–P10 or P–Ca10). Ca–P means that the HA nanoparticles were synthesized by mixing previously prepared β-CD solutions containing Ca 2 + ion precursors (Ca(NO3)2·4H2O) with PO43– ion precursors ((NH4)2HPO4). On the other hand, P–Ca nanoparticles were prepared by the reaction of β-CD solutions containing PO43– ion precursors with Ca 2 + ion precursors. Fig. 1 shows the morphological structure of HA nanoparticles. All of the resulting nanoparticles exhibited spherical shape with a size of below 100 nm and P–Ca showed higher particle size than that of Ca–P. The size of nanoparticles slightly decreased with increasing the content of β-CD. The structure of β-CD provides a molecule shape like a segment of hollow cone with an exterior hydrophilic surface and interior electron-rich hydrophobic cavity, which is capable of forming stable and supramolecular structures with various molecules including organometallic compounds [22]. However, even though the size of single particle was below 100 nm, the aggregate formation of HAs synthesized in the presence of β-CD was observed with a size of hundreds nanometer as shown in Fig. 1. The HA nanoparticles were also characterized with a TEM to confirm the morphological structure and aggregation behavior (Fig. 2). Ca–P showed the sphere-like structure, which was in good agreement with the result of SEM observation. It was previously reported that polymers containing polar functional groups such as COOH and OH have been found to be useful for the nucleation of HA crystals in the solution because these ionizable side groups provide a site for the selective adhesion of Ca 2+ ions [10,11]. β-CD is a polar and water soluble oligomer with OH side group, which can be cross-linked with each other through the hydrogen bonding interaction and inclines to form the sphere-like structure [23]. Upon ionization, the negatively charged surface groups can provide binding sites for the Ca2+ ions present in
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
501
Fig. 1. SEM micrographs of HA nanoparticles synthesized with different content of β-CD and mixing sequence of ion precursor: (a) 0.05% (w/v) (Ca–P5), (b) 0.1% (w/v) (Ca–P10), (c) 0.05% (w/v) (P–Ca5) and (d) 0.1% (w/v) (P–Ca10).
the solution. As the Ca2+ ions accumulate on the surface, the surface gains an overall positive charge. These positively charged surface sites will combine with negatively charged PO43– ions initiating the nucleation and growth of the spherical HA nanoparticles. The increase of β-CD content leads to the formation of larger number of nuclei and therefore smaller HA crystallite size [10]. In the mean time, P–Ca exhibited the mixture of spherical and rod-like nanoparticles. Since (NH4)2HPO4 takes the mixed form of HPO42– and PO43– at pH = 10, the element P in the β-CD solution should be in the mixed form of HPO42– and PO43–, in which the oxygen atoms of HPO42– and PO43– can accept protons from the OH side groups of β-CD [24]. Thus these ions can be entirely inserted between the hollow cone of β-CD and form the complexes with β-CD through the multiple hydrogen bonds [25]. These complexes can strongly interact with the Ca2+ ions to nucleate the HA nanoparticles. As a result, the initial nucleation is preferentially triggered at the positions of hydroxyl groups, and the rod-like structure may be partially produced. Moreover, P–Ca is probably synthesized through the development of coordination anions Ca–P6O24 which are able to build nanowires or rod-like nanoparticles [20]. The formation of aggregates was also clearly observed in all the samples. The aggregation of HA nanoparticles should be restrained for convenient use. Therefore, the suspensions of HA nanoparticles were prepared for the determination of aggregate particle size. As shown in Fig. 3, not merely the concentration of β-CD but the method for the synthesis of HA, mixing sequence of ion precursors, affected the average
particle size of aggregates. The average particle size slightly decreased with increasing the β-CD concentration from 525 ±97 nm (Ca–P5) to 480 ± 91 nm (Ca–P10). In the case of P–Ca, the average particle size enormously reduced from 836 ± 60 nm to 580 ±71 nm with the increase of β-CD content from 0.05% (w/v) to 0.1% (w/v). Furthermore, the average particle size of P–Ca was higher than that of Ca–P as mentioned in the result of SEM observation. It was concluded that the β-CD content and the synthesis procedure can affect the aggregation behavior of nanoparticles. That is to say, the particle with smaller single particle size gives the smaller aggregate particle size. 3.2. Chemical structure and crystalline phase FT-IR analysis was carried out for identifying the functional groups present in the HA nanoparticles, which in turn provided information about the constitution and phase composition of the products synthesized with different content of β-CD and synthesis procedure. As shown in Fig. 4, all of the samples synthesized in the presence of β-CD exhibited characteristic absorption bands for the vibrational modes of PO43– appeared at around 1090, 1016, 959, 596 and 557 cm–1, and the bands at 3340 and 1640 cm–1 associated with OH of β-CD and absorbed H2O [23]. In addition, the band observed at 1420 cm–1 was attributed to the substitution of CO23– ions in the place of PO43– ions and confirmed the substitution of CO23– in apatite structure, which was stronger in P–Ca nanoparticles [2]. These CO23– ions were formed by the reaction of CO2
502
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
Fig. 2. TEM micrographs of HA nanoparticles synthesized with different content of β-CD and mixing sequence of ion precursor: (a) 0.05% (w/v) (Ca–P5), (b) 0.1% (w/v) (Ca–P10), (c) 0.05% (w/v) (P–Ca5) and (d) 0.1% (w/v) (P–Ca10).
present in the atmosphere with OH– ions of reaction medium. Hence P– Ca nanoparticles possibly contain the higher amount of CaCO3 as a mixture with HA. Moreover, the HA nanoparticles showed absorption bands assigned to OH– of HA at 3570 and 630 cm–1. These data suggest that the HA nanoparticles could be simply synthesized by the use of β-CD as template, which were organic–inorganic composites produced
by the complexation of inorganic HA with β-CD. Additionally, the chemical structure of products was a little changed with the synthesis procedure. To verify the complexation behavior between inorganic HA and β-CD, the change of UV absorption spectra was determined using 1% (w/v) solution of all samples (Fig. 5). Commercial HA nanoparticles without β-CD
Fig. 3. Particle size distribution of (a) Ca–P5, (b) Ca–P10, (c) P–Ca5 and (d) P–Ca10 nanoparticles.
Fig. 4. FT-IR spectra of (a) commercial HA, (b) Ca–P5, (c) Ca–P10, (d) P–Ca5 and (e) P–Ca10 nanoparticles.
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
503
Table 1 Crystallinity and chemical composition of the HA nanoparticles synthesized with different content of β-CD and mixing sequence of ion precursor. Sample
Commercial HA Ca–P5 Ca–P10 P–Ca5 P–Ca10
Fig. 5. UV-visible spectra of commercial and synthesized HA nanoparticles.
revealed specific absorption peak at around 206 nm. However, the synthesized HA nanoparticles, Ca–P10 and P–Ca10, showed amplified intensity of absorption, and specific absorption peak due to HA shifted to 209 nm. This result can prove the complexation behavior through the intermolecular interaction and the formation of composites consisted of inorganic HA and β-CD. Moreover, the composite formation between HA and β-CD in the resultant HA nanoparticles was also confirmed by the EDX analysis. As shown in Fig. 6, the four characteristic peaks ascribed to carbon, oxygen, phosphorous and calcium atoms were observed at 0.26, 0.52, 2.02 and 3.68 keV, respectively. The peak intensity assigned to carbon atom increased with the increment of β-CD content, however, the peaks due to phosphorous and calcium atoms reduced. To assess the chemical composition of HA nanoparticles, 10 mg of each samples was dissolved in 10 mL of 1 N HCl solution, and Ca or P element was measured with an ICP–OES. The synthesized HA nanoparticles contained less amount of inorganic constituent compared with commercial HA (Table 1). In addition, this inorganic constituent amount reduced slightly with increasing the content of β-CD. This clearly explains on the presence of β-CD in the synthesized HA nanoparticles. The crystalline phases of the HA nanoparticles were investigated by means of XRD (Fig. 7). Commercial HA nanoparticles exhibited the diffraction peaks at around 25.9°, 28.3°, 29.1°, 31.8°, 32.2°, 32.8°, 34.2°, 39.9°, 42.1°, 43.9°, 45.5°, 46.7°, 48.2°, 49.6°, 50.5°, 51.3°, 52.2° and 53.3°, which were imputed to the (002), (102), (210), (211), (112), (300), (202), (310), (311), (113), (203), (222), (312), (213), (321), (400), (402) and (004) planes [2,22]. This XRD pattern was exactly
Fig. 6. EDX spectra of (a) Ca–P5, (b) Ca–P10, (c) P–Ca5 and (d) P–Ca10 nanoparticles.
Crystallinity (%)
55.1 13.4 12.5 7.7 7.6
Ca (mM/g)
12.60 10.18 10.03 10.13 9.76
P (mM/g)
7.75 5.91 5.78 5.88 5.65
Ca/P molar ratio ICP-OES
XRF
1.63 1.72 1.74 1.72 1.73
1.67 1.72 1.72 1.72 1.70
matched with the structural data of stoichiometric HA described in the Powder Diffraction File (JCPDS 09-432). In addition, the XRD patterns of synthesized HA nanoparticles showed the peaks attributed to the HA crystalline phase at around 25.8°, 29.1°, 31.9°, 32.4°, 32.9°, 34.1°, 39.9°, 43.9°, 46.7°, 49.6° and 53.3°, which reflected characteristic of the (002), (210), (211), (112), (300), (202), (310), (113), (222), (213) and (004) planes. However, all the peaks were broad diffraction peaks indicating a poorly crystallized HA phase. The diffraction peaks became wider and less intense with increasing the concentration of β-CD. Additionally mixing sequence of ion precursors also influenced these diffraction peaks, leading that P–Ca showed more broad diffraction
Fig. 7. X-ray diffraction patterns of (a) commercial HA, (b) Ca–P5, (c) Ca–P10, (d) P–Ca5 and (e) P–Ca10 nanoparticles.
Fig. 8. TGA curves of (a) commercial HA, (b) Ca–P5, (c) Ca–P10, (d) P–Ca5 and (e) P–Ca10 nanoparticles.
504 K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
Fig. 9. SEM micrographs of (a) Ca–P5, (b) Ca–P10, (c) P–Ca5 and (d) P–Ca10 nanoparticles after 3, 5 and 7 days of immersion in SBF.
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
peaks. This is owing to the complex formation of HA with amorphous β-CD. Moreover, the isomorphous substitution of PO43– by CO23– derived from the absorption of CO2 in the air during preparation process of the HA nanoparticles affected the decrease of crystallinity [22]. The crystallinity of synthesized HA nanoparticles was calculated from the XRD data, which was lower than that of commercial HA and slightly decreased with the increase of β-CD content as shown in Table 1. As mentioned above, mixing sequence of ion precursors also affected the crystallinity which was 12.5–13.4% for Ca–P and 7.6–7.7% for P–Ca. This result is related to the complex formation HA with amorphous β-CD and the β-CD amount included in the composites. Previous study reported that the calcium phosphate (CaP) spheres synthesized with poly(acrylic acid) (PAA) exhibited only a broad diffraction peak, indicating that the CaP phase was poorly crystallized [11]. The CaPs subjected to thermal treatment at 550 °C for the elimination of amorphous polymer displayed strong peaks that corresponded well to those of stoichiometric HA crystal, owing to the increase in crystallinity. The Ca/P molar ratio of HA nanoparticles was determined by XRF, which was ranged from 1.70 to 1.72 and similar to the result of ICP-OES (Table 1). The Ca/P molar ratio was hardly touched with the β-CD content and mixing sequence of ion precursors. These results mean that HA can be easily fabricated by the use of β-CD. Sugar molecules are known to coordinate Ca2+ ions or PO43– ions in various configurations and this effect can be enhanced at the crystal surfaces of HA by the high spatial charge density and possibility of hydrogen bonding to the surface phosphate groups [26]. Afterwards, counterions (PO43– ions or Ca2+ ions) will accumulate at calcium or phosphate complexes. Mineral nuclei formed at these sites will gradually grow on the surface of β-CD, leading to the formation of HA crystal. Consequently, the use of β-CD as matrix is useful for the synthesis of HA. 3.3. Thermal stability Thermal property of the synthesized HA nanoparticles was investigated. It is well known that the thermal property of HA is strongly influenced by its Ca/P molar ratio and hence the decomposition temperature of HA can give an evidence to its Ca/P molar ratio qualitatively [2]. Fig. 8 shows the typical TGA curve where the amount of weight loss is plotted against the temperature. All of the samples synthesized in this study had very similar Ca/P molar ratio and thus exhibited the same pattern of weight loss comprised of three steps. The first step takes place between 30 °C and about 150 °C. This step mainly is assigned to the evaporation of absorbed water and the second step in temperature range of 200–450 °C is maybe imputed to the elimination of crystalline water and the decomposition of β-CD [23,27,28]. Additionally, the third step is probably the loss of constitution water of HA, which can change to oxyapatite (Ca10(PO4)6O). The total weight loss at 800 °C increased from 14.2% to 19.3% with the increment of β-CD content and the change of synthesis procedure. The decomposition temperature of β-CD is about 308 °C, resulting that the heightened weight loss is attributed to the decomposition of β-CD. The HA nanoparticles synthesized with β-CD contain some amount of β-CD even after repeated washing. Therefore, the mass fraction of β-CD included in the resulting HA nanoparticles was calculated from the TGA data, which was 6.8% for Ca–P5, 10.0% for Ca–P10, 9.3% for P– Ca5 and 11.9% for P–Ca10. This is a result of the complex formation via electrostatic interaction and hydrogen bonding between HA and β-CD. 3.4. Simulated body fluid immersion test The essential requirement for an artificial material to bond to living bone is the formation of bone-like apatite on its surface when implanted in the living body, and that this in vivo apatite formation can be reproduced in a simulated body fluid (SBF) with ion concentrations
505
nearly equal to those of human blood plasma [14]. This means that in vivo bone bioactivity of a material can be predicted from the apatite formation on its surface in SBF. After 3 days of immersion in SBF, the deposition of bone-like apatite was observed at some sites on the surface of Ca–P5 and P–Ca5 nanoparticles as shown in Fig. 9. After 5 and 7 days of mineralization in SBF, these nanoparticles were almost completely covered with apatite structures. On the other hand, the formation of bone-like apatite was hardly detected on the surface of Ca–P10 and P–Ca10 nanoparticles after 3 days of immersion in SBF. After 7 days of mineralization in SBF, the whole surfaces of Ca–P10 and P–Ca10 nanoparticles were covered by a layer of apatite. In addition, the aggregation of HA nanoparticles severely proceeded by immersion in SBF and the resulting particles exhibited cauliflower-shaped morphology. These results mean that the β-CD content included in the HA nanoparticles hugely affected the formation of bone-like apatite. The initial deposition rate of apatite on the surface of HA nanoparticles reduced with increasing the β-CD content. 4. Conclusions The HA nanoparticles have attracted a great deal of attention in biomedical applications. An increase in the surface area of HA leads to a greater proportion of its atoms or molecules being displayed on its surface rather than in the interior. As a result, it enhances the biological performances such as cell adhesion, proliferation and differentiation. In the present study, a novel and simple reaction for the preparation of HA nanoparticles was successfully developed by the use of β-CD as template. The β-CD content slightly affected the particle size of products. In addition, the synthesis procedure (mixing sequence of ion precursors) also influenced the particles size and morphological structure of nanoparticles. The HA nanoparticles synthesized in the presence of β-CD exhibited the aggregation behavior. The HA nanoparticles had very narrow size distribution, and the composition and crystalline phase were similar to those of the stoichiometric HA. However, the crystallinity of synthesized HA nanoparticles was low compared with that of commercial HA. These results are ascribed to the complexation behavior between inorganic HA and β-CD through electrostatic interaction and hydrogen bonding, causing the production of composites. It is concluded that this process offers a simple and potentially safe new approach for preparing the HA nanoparticles. Acknowledgements This research was financially supported by the Ministry of Knowledge Economy (MKE), Korea Institute for Advancement of Technology (KIAT) through the Inter-ER Cooperation Projects. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]
T.Y. Liu, S.Y. Chen, D.M. Liu, S.C. Liou, J. Control. Release 107 (2005) 112–121. B. Viswanath, N. Ravishankar, Biomaterials 29 (2008) 4855–4863. J.L. Xu, K.A. Khor, J.J. Sui, J.H. Zhang, W.N. Chen, Biomaterials 30 (2009) 5385–5391. S. Bose, S.K. Saha, Chem. Mater. 15 (2003) 4464–4469. H.M. Kim, C. Rey, M.J. Glimcher, J. Bone Miner. Res. 10 (1995) 1589–1601. G. Bezzi, G. Celotti, E. Landi, T.M.G. La Torretta, I. Sopyan, A. Tampieri, Mater. Chem. Phys. 78 (2003) 816–824. S. Pramanik, A.K. Agarwal, K.N. Rai, A. Garg, Ceram. Int. 33 (2007) 419–426. Y.X. Sun, G.S. Guo, D.L. Tao, Z.H. Wang, J. Phys. Chem. Solids 68 (2007) 373–377. S.I. Stupp, P.V. Braun, Science 277 (1997) 1242–1248. S. Mollazadeh, J. Javadpour, A. Khavandi, Ceram. Int. 33 (2007) 1579–1583. P. Wang, S.W. Yook, S.H. Jun, Y.L. Li, M. Kim, H.E. Kim, Y.H. Koh, Mater. Lett. 63 (2009) 1207–1209. H. Liu, C. Liu, X. Yang, S. Zeng, Y. Xiong, W. Xu, Anal. Chim. Acta 628 (2008) 87–94. Y.X. Pang, X. Bao, J. Eur. Ceram. Soc. 23 (2003) 1697–1704. T. Kokubo, Biomaterials 12 (1991) 155–163. M. Ohgaki, K. Yamashite, J. Am. Ceram. Soc. 86 (2003) 1440–1442. J. Huang, Y.W. Lin, X.W. Fu, S.M. Best, R.A. Brooks, N. Rushton, W. Bonfield, J. Mater. Sci. Mater. Med. 18 (2007) 2151–2157. S. Morita, K. Furuya, K. Ishihara, N. Nakabayashi, Biomaterials 19 (1998) 1601–1606.
506
K.D. Son, Y.-J. Kim / Materials Science and Engineering C 33 (2013) 499–506
[18] R.W. Arcis, A. Lopez-Macipe, M. Toledano, E. Osorio, R. Rodriguez-Clemente, J. Murtra, M.A. Fanovich, C.D. Pascual, Dent. Mater. 18 (2002) 49–57. [19] S. Haque, I. Rehman, J.A. Darr, Langmuir 23 (2007) 6671–6676. [20] P. Wang, C. Li, H. Gong, X. Jiang, H. Wang, K. Li, Powder Technol. 203 (2010) 315–321. [21] H. Zhang, M. Zhang, Mater. Chem. Phys. 126 (2011) 642–648. [22] F. Hapiot, S. Tilloy, E. Monflier, Chem. Rev. 106 (2006) 747–781. [23] X.F. Xiao, R.F. Liu, C.F. Qiu, D.C. Zhu, F. Liu, Mater. Sci. Eng. C 29 (2009) 785–790.
[24] A. Oyane, K. Onuma, A. Ito, H.M. Kim, T. Kokubo, T. Nakamura, J. Biomed. Mater. Res. 64A (2003) 339–348. [25] H. Luecke, F.A. Quiocho, Nature 347 (1990) 402–406. [26] D. Walsh, J.L. Kingston, B.R. Heywood, S. Mann, J. Cryst. Growth 133 (1993) 1–12. [27] Y. Dong, W. Dong, Y. Cao, Z. Han, Z. Ding, Catal. Today 175 (2011) 346–355. [28] Y. Yang, X.F. Li, J.B. Chen, H.L. Chen, X.M. Bao, Chem. Phys. Lett. 373 (2003) 22–27.
