Journal of Nanoparticle Research

, Volume 11, Issue 3, pp 691–699

In situ processing and properties of nanostructured hydroxyapatite/alginate composite

Authors

    • China National Academy of Nanotechnology & Engineering
  • Yue Li
    • Department of Chemical Machinery, School of Chemical EngineeringDalian University of Technology
  • Chunzhong Li
    • Key Laboratory for Ultrafine Materials of Ministry of EducationEast China University of Science and Technology
Research Paper

DOI: 10.1007/s11051-008-9431-y

Cite this article as:
Wang, L., Li, Y. & Li, C. J Nanopart Res (2009) 11: 691. doi:10.1007/s11051-008-9431-y

Abstract

A series of hydroxyapatite/alginate (HA/Alg) nanocomposites with alginate amounts varying from 10 to 40 wt% were prepared through in situ hybridization technique. The inorganic phase in the composites was carbonate-substituted HA with low crystallinity. The crystallinity of HA decreased with the increase of alginate content. HA crystallites were needle-like in shape with a typical size of 20 to 50 nm in length and 5 nm in width. FT-IR spectroscopy indicated that the chemical interaction occurred between the mineral phase and the polymer matrix. As compared to pure HA without alginate, the composites showed more homogeneous microstructures, where HA nanocrystals were well embedded in alginate matrix. Among all the samples, the composite containing 30 wt% alginate exhibited a highly ordered three-dimensional network, similar to natural bone’s microstructure.

Keywords

HydroxyapatiteAlginateNanocompositeIn situ hybridizationMicrostructureNanocrystalsNanomaterials

Introduction

Recently, dramatic increases in aging population and osseous defects originating from sports and traffic-related trauma, infections, and skeletal diseases have significantly stimulated the exploitation of artificial bone graft materials (Bigi et al. 1998; Murugan and Ramakrishna 2004). An ideal bone substitute should be able to provide bony bonding with surrounding living tissues, to serve as a scaffold with sufficient mechanical strength and to stimulate growth and formation of new bone tissues (Zhao et al. 2002; Tampieri et al. 2005). Besides, it should be biodegradable so that it can degrade gradually and ultimately be replaced by newly formed bone tissues. Directing efforts towards the development of desirable synthetic scaffold materials in view of the specific challenges posed by rapid advances in bone tissue engineering will undoubtedly be rewarding.

Natural bone has been well documented as a complex tissue comprising approximately 70 wt% of hydroxyapatite (HA) and 30 wt% of collagen matrix, in which nanosized HA particles are embedded within organic matrix in a well-organized manner (Bigi et al. 2002; Wang 2003). Natural bone’s ubiquitous hybrid composition and inimitable highly ordered hierarchical structure have presented us valuable clues to achieve intelligent scaffold materials. Therefore, increasing interest has been focused on the development of novel bio-inspired HA/polymer composites attempting to create devices capable of reproducing the complex hierarchical structure of native bone (Dujardin and Mann 2002; Mann and Ozin 1996; Stupp and Braun 1997). Several natural polymers with good biodegradability and bioactivity are commonly used, such as collagen (Du et al. 2000; Kikuchi et al. 2001; Zhang et al. 2003a, b), gelatin (Bigi et al. 1998; Chang et al. 2003), chitosan (Kong et al. 2005; Murugan and Ramakrishna 2004; Yamaguchi et al. 2001), silk fibroin (Furuzono et al. 2000; Wang et al. 2002, 2004), alginate (Lin and Yeh 2004; Maruyama et al. 1995; Ribeiro et al. 2004; Sivakumar and Rao 2003; Teng et al. 2006), and chondroitin sulfate (Rhee and Tanaka 2002).

Alginate (Alg) is a natural acidic polysaccharide derived from brown sea algae (Sivakumar and Rao 2003; Lin and Yeh 2004; Teng et al. 2006). It is a linear anionic copolymer containing (1-4)-linked β-d-mannuronic acid (M units) and (1-4)-linked α-l-guluronic acid (G units) residues. Alginate is highly biocompatible, biodegradable, hydrophilic, and relatively economical and has sufficient adhesiveness, plasticity, and hemostatic property. Therefore, alginate has been widely utilized in wound dressing, surgical or dental impression, and tissue engineering. Several reports in literature addressed the preparation of HA/Alg composite scaffolds as bone filler via different techniques such as phase separation, droplet extrusion, or dispersion polymerization (Lin and Yeh 2004; Maruyama et al. 1995; Ribeiro et al. 2004; Sivakumar and Rao 2003; Teng et al. 2006). Maruyama et al. reported that a mixture of HA powder and saline solution of sodium alginate showed good osteoconductivity to bridge the gap between implants and cortical bone without any adverse reaction (Maruyama et al. 1995). A HA/Alg composite with a highly interconnected porous structure and considerable mechanical properties was prepared by Lin and Yeh which displayed appreciable osteoblastic cell attachment (Lin and Yeh 2004). Sotome et al. and Zhang et al. demonstrated that the incorporation of alginate into HA/collagen composite improved the mechanical properties and promoted active tissue invasion and bone formation in vitro (Sotome et al. 2004; Zhang et al. 2003a).

A common feature of the above synthesis procedure lies in that the conventional mechanical mixing of as-prepared HA granules with Ca2+ crosslinked alginate was involved. However, lack of physicochemical homogeneity and adequate interfacial bonding between the mineral and polymer phases still remains an issue of concern over HA/Alg composites obtained by this method (Yamaguchi et al. 2001; Sailaja et al. 2003). As we know, natural bone apatite is formed in a physiological environment in the presence of collagen, Ca2+, PO43−, and other mineral ions (Gibson and Bonfield 2002; Liou et al. 2005). Inspired by this naturally occurring biomineralization, recently, material researchers have exploited in situ hybridization aimed at producing chemically and biologically improved HA-polymer composites as bone substitute materials. The advantage of this method over the mechanical mixing is that the binding strength at the inorganic–organic interface can be enhanced by the established chemical interaction between HA and the polymer matrix. However, little attention has been given to the in situ synthesis of the HA/Alg composite.

The present study was designed to establish the feasibility of in situ synthesis of HA/Alg composite with chemical and structural homogeneity, and to gain more insight into the microstructure and physicochemical properties of the composite. In order to reach these scopes, a series of HA/Alg composites with different alginate contents were prepared via in situ hybridization. Moreover, the investigation into the influences of alginate content on the crystallographic properties and morphology of HA crystallites, the chemical interaction between HA and alginate, as well as the microstructure of the composites was carried out. This work is expected to provide a framework for the design and generation of HA-based composite materials for bone repair and remodeling.

Experimental procedure

Preparation of HA/Alg composites

Ca(OH)2, CaCl2, and 85 wt% H3PO4 of analytical grade were used without further purification. Sodium alginate (with a low viscosity of 0.25 Pas for its 2 wt% solution at 25 °C) was purchased from Sigma-Aldrich Inc., USA.

Table 1 lists the weight ratios of HA/Alg in the final products, the amount of each reagent initially added, and the corresponding notations for the obtained samples. The molar ratio of Ca/P in the starting mixture was set as 1.67, equal to that of stoichiometric HA. An alginate gel solution (3% w/v) was prepared by dissolving alginate powder into distilled water. The alginate solution was added dropwise into a suspension of Ca(OH)2 (70 g, 7.37 wt%) under vigorous agitation. A solution of H3PO4 was obtained by mixing H3PO4 (4.82 g, 85 wt%) with 50 mL distilled water, and then was added in drops into the Alg/Ca(OH)2 suspension. The reaction temperature was kept at 25 °C and the mixture was stirred continuously at 700 rpm for 6 h. Then, a solution of CaCl2 (10 mL, 0.1 mol L−1) was slowly added, supplying Ca2+ to crosslink alginate. After half an hour of crosslinkage, the gel-like mixture was centrifuged and water-washed alternately for three cycles to harvest the precipitates. The precipitates were vacuum-dried at 50 °C for 48 h and subsequently ground into fine powders using an agate mortar. Meanwhile, pure HA without alginate involved was prepared as a control sample by the same procedure.
Table 1

Notations for the samples and the amount of each starting material used

Sample

HA/Alginate weight ratio

Weight of Ca(OH)2 (g)

Weight of 85 wt% H3PO4 (g)

Weight of sodium alginate (g)

HA

 

5.16

4.82

0.0

HA90/Alg

90/10

5.16

4.82

0.78

HA80/Alg

80/20

5.16

4.82

1.75

HA70/Alg

70/30

5.16

4.82

3.00

HA60/Alg

60/40

5.16

4.82

4.67

Characterization

Crystalline phase composition of the as-prepared powders was confirmed by X-ray diffractometer (XRD, RINT PC1 Rigaku Co.) with CuKα radiation. Fourier transform infrared (FT-IR) spectra were recorded using a spectrometer (Nicolet AVATR360). An enhanced laser diffraction granulometer (Beckman Coulter LS230) was used to analyze particle size distribution. Morphology and size dimensions of HA crystallites were examined by a transmission electron microscope (TEM, JEOL JEM-2100). Microstructural observations were performed on a scanning electron microscope (SEM, JEOL JSM-6700F).

Results and discussion

Crystalline phase composition

XRD patterns of the composites and pure HA are shown in Fig. 1. The diffraction peaks are well defined and assigned to monophase crystalline HA only, since no peaks from other calcium phosphate phases are detected. Notable line broadening and overlap of diffraction peaks are present in all the samples, particularly in HA/Alg composites. Diffraction peaks of (211), (300), and (202) can be discerned as three individual peaks in pure HA, whereas as a broader peak in the HA/Alg composites. It indicates that the precipitated HA crystals have small size and low crystallinity similar to natural bone mineral (Murugan and Ramakrishna 2004; Rhee and Tanaka 2002). The poor crystalline nature of the as-prepared HA is possibly attributed to the wet-chemical synthesis at room temperature and the subsequent low temperature drying rather than sintering. From HA to HA60/Alg, the peak intensity decreases and the peak tends to be broader, i.e., the full width at half maximum (FWHM) of the (002) peak shows an apparent increase. This can be taken as a sign for the decreased crystallinity of HA with an increase in alginate content. The findings demonstrate that the presence of alginate affects the HA crystal growth and therefore the crystallographic characteristics of HA.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-008-9431-y/MediaObjects/11051_2008_9431_Fig1_HTML.gif
Fig. 1

XRD patterns of pure HA and HA/Alg composites

Particle size distribution

The analysis results from particle size distribution curves are presented in Table 2. The mean particle size shows a general trend of increasing with alginate content, i.e., a nearly linear increase from HA to HA70/Alg, while a two-fold increase from HA70/Alg to HA60/Alg. There is no apparent change in the size distribution range for the samples with alginate content lower than 30 wt%, and the particles are falling in the range of 1.3 to 33.0 μm. With alginate content increased to 40 wt%, the size range (1.7–121.8 μm) becomes obviously wider. It implies that the addition of alginate in excessive amount over 30 wt% significantly intensifies the particle agglomeration. It is not easy to achieve uniform dispersion of HA particles into the polymer matrix due to the adhesive nature of alginate. Therefore, we suggest that the alginate content in the composites should not be higher than 30 wt%.
Table 2

Results of particle size distribution

Sample

Mean size (μm)

D(50%)a (μm)

D(10%)b (μm)

D(90%)b (μm)

HA

5.93

4.38

2.33

7.78

HA90/Alg

6.72

5.43

2.64

9.87

HA80/Alg

7.60

6.49

3.30

10.96

HA70/Alg

8.65

7.07

2.82

13.13

HA60/Alg

17.15

11.87

3.41

25.75

aD(50%) is the size at which volume fraction 50% of the sample is smaller and 50% is larger

bD(10%) or D(90%) is the size below which volume fraction 10% or 90% of the sample lies

Chemical state analysis

FT-IR spectra of alginate before crosslingage, pure HA, and HA/Alg composites are illustrated in Fig. 2. The bands at 1,615 and 1,418 cm−1 are assigned to asymmetrical and symmetrical stretching modes of COO in alginate, respectively (Fig. 2a) (Ribeiro et al. 2004; Teng et al. 2006). In pure HA (Fig. 2b), the bands at 1,091, 1,034, 962, 603 and 565 cm−1 correspond to different vibrational modes of PO4 group in HA (Rhee and Tanaka 2002; Yamaguchi et al. 2001). The bands at 1,452, 1,422, and 874 cm−1 are derived from carbonate ions, indicative of the precipitated HA containing carbonate ions (Gibson and Bonfield 2002). No carbonate source was introduced into the starting materials, and all the samples were prepared in an atmospheric environment. It is reasonable to infer that the carbonate ions incorporated into HA must arise from carbon dioxide gas present in air. These typical bands from HA are also witnessed in the composites (Fig. 2c–f). As compared to pure HA, the bands from PO4 group in the composites become wider and weaker, e.g., the shoulder bands at 962 and 1,091 cm−1 cannot be detected in HA90/Alg. This might be closely related to the decreased crystallinity by the effect of alginate. In Fig. 2b, the band at 1,635 cm−1 is attributed to H2O group in HA. This band appears separately as a sharper band at 1,622–1,630 cm−1 in the composites with different amounts of alginate, possibly ascribed to the overlap of the H2O group of HA and COO (at 1,615 cm−1) of alginate. The band tends to be stronger with alginate increase in content from 10 to 30 wt%. From alginate to the composites, there are notable band shifts toward higher wave numbers found in the asymmetrical stretching mode of COO, i.e., a blue shift of 9 cm−1 from alginate to HA70/Alg. The blue shifts imply that there exists chemical interaction between the mineral phase and the organic matrix, most likely encouraged by the chemical bonding between Ca2+ and the negative charged carboxyl group in alginate (Kikuchi et al. 2001; Teng et al. 2006). The symmetrical stretching mode of COO is not observed in the composites. Due to the influence of COO of alginate, the bands at 1,422 and 1,452 cm−1 in the composites are slightly stronger than those in pure HA. The FT-IR results confirm that alginate has been incorporated into the composites.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-008-9431-y/MediaObjects/11051_2008_9431_Fig2_HTML.gif
Fig. 2

FT-IR spectra of (a) alginate, (b) pure HA, (c) HA90/Alg, (d) HA80/Alg, (e) HA70/Alg, and (f) HA60/Alg

Morphology of HA crystallite

TEM images of all the samples are shown in Fig. 3. In pure HA and the composites, the elongated HA crystallites are envisioned as needle-like in shape with a typical size of 20 to 50 nm in length and about 5 nm in width. In pure HA, HA crystallites are in relatively well-dispersed state. The images of the crystallites in the composites are not as clear as those in pure HA. The separate HA crystallites are not easily discerned in the composites due to the severe agglomeration and the amorphous phases appearing as obscure shadows somewhere. The amorphous phases are attributed to alginate, transient amorphous calcium phosphates and poorly crystalline HA. The areas of the obscure shadows tend to be larger with an increase in alginate content. It demonstrates that the crystallinity of HA is lowered by the introduction of alginate, consistent with the above XRD result. HA nanocrystals are spontaneously aggregated together to form bundles. There is an apparent tendency in the composites that the aggregation is preferentially along the c-axis. This phenomenon is thought to be closely associated with the preferential self-assembly of HA crystallites along the c-axis, as reported in several other studies (Kikuchi et al. 2001; Rhee and Tanaka 2002; Yamaguchi et al. 2001). HA90/Alg and HA80/Alg show similar TEM images where the crystallites are 20 to 30 nm long, shorter than those about 50 nm long in HA70/Alg and HA60/Alg. It suggests that the addition of larger amount of alginate (higher than 30 wt%) promotes the epitaxial growth of HA crystals along c-axis.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-008-9431-y/MediaObjects/11051_2008_9431_Fig3_HTML.jpg
Fig. 3

TEM images of pure HA, HA90/Alg, HA80/Alg, HA70/Alg, and HA60/Alg

Alginate is crosslinked with Ca2+ to form Ca–alginate complex with an “egg-box” structure through the ionic interaction between Ca2+ and the carboxyl group of G units located on the polymer chain (Lin and Yeh 2004; Teng et al. 2006). The Ca–alginate complexes are thought to affect the nucleation and growth of HA crystals. In the composites with low alginate content of 10 to 20 wt%, the influence of the alginate matrix is not very significant and thus the morphology of HA crystallites is not greatly changed. With alginate content increased to 30–40 wt%, larger number of Ca–alginate complexes with long polymer chains are formed, which in turn intensifies the steric hindrance of the adjacent complex macromolecules. The extended direction of the polymer chains of alginate macromolecules may be parallel to the elongated direction, ca. c-axis of HA crystallites. Consequently, HA crystallites show an apparent tendency to elongate and aggregate preferentially along c-axis.

Microstructure of HA/Alg composite

SEM micrographs of all the samples are displayed in Fig. 4. Pure HA exhibits a loose discrete and inhomogeneous microstructure, where circular or elliptical island-like particles of 50 to 100 nm in size are randomly distributed. The particles are loosely bound to each other by weak electrostatic attractive forces. As shown in TEM images, the primitive HA crystals are 5 nm wide and 20–30 nm long. It thereby can be deduced that the island-like particles are formed owing to the primary or secondary aggregation of the primitive HA nanocrystals. In the composite, such aggregation is also observed, but the circular or elliptical contours of the particles cannot be discerned as clearly as in pure HA.
https://static-content.springer.com/image/art%3A10.1007%2Fs11051-008-9431-y/MediaObjects/11051_2008_9431_Fig4_HTML.jpg
Fig. 4

SEM images of (a) pure HA, (b) HA90/Alg, (c) HA80/Alg, (d) HA70/Alg, and (e) HA60/Alg. Original magnification: ×30,000

It is observed in HA/Alg composites that HA aggregates are embedded in the alginate matrix. The interfacial boundary between the mineral and organic phases is hardly distinguished, indicating the appreciable miscibility between them. The microstructure tends to be more homogeneous and compact with the increase of alginate content. In pure HA, HA90/Alg, and HA80/Alg, hollows with irregular shapes and variable sizes ranging from 50 nm to several hundred nanometers are observed among those large block aggregates, while fewer hollows are observed in HA70/Alg and HA60/Alg. Alginate acts as a binder for HA nanoparticles, and hence its content is of importance to determine the ultimate microstructure of the HA/Alg composite. Alginate content higher than 30 wt% is sufficient to permit complete fusion of HA crystallites into the alginate matrix. As reported, there is a problem always encountered with HA-based implant is that HA particles easily migrate from the implanted sites into surrounding tissues when mixed with patient’s body fluid and blood (Murugan and Ramakrishna 2004; Yamaguchi et al. 2001). The migration causes damage to healthy tissues and thus the therapeutic efficacy of the implant is not satisfactory. Bearing this in mind, the complete fusion of HA nanoparticles into the alginate matrix is beneficial for particulate immobilization.

In HA70/Alg, higher level of interpenetration between the HA nanoparticles and the alginate matrix is achieved so that a hierarchical three-dimensional network is built up. HA60/Alg shows a compact and homogeneous microstructure with smoother surface and uniform particle distribution. Among all the samples, HA70/Alg bears a very close microstructural resemblance to natural bone, and thereby, can be considered as an ideal candidate for healing bone defects.

Conclusions

HA/Alg nanocomposites with different HA/Alg weight ratios were synthesized by in situ hybridization. The inorganic component in the composite is identified as poorly crystalline HA containing carbonate ions, similar to biological apatite. The primary HA crystallites are needle-like and 20–50 nm long and around 5 nm wide. The crystallographic features of HA are affected by the involvement of alginate, i.e., the crystallinity of HA decreases and the crystallites tend to elongate along the c-axis upon increasing the alginate content. As confirmed by FT-IR spectroscopy, the notable blue shifts in the carboxyl group from alginate to the composites indicate that the chemical interaction takes place between the inorganic and polymeric constituents in the composites. The chemical interaction is assumed to be mediated by the chemical bonding between Ca2+ and the carboxyl group in alginate. The composites present a more compact microstructure than pure HA. The complete fusion of HA crystallites into the polymer matrix is achieved in the composites with the alginate content of 30 to 40 wt%, facilitating the immobilization of HA nanoparticles upon implantation. A three-dimensional network with a higher degree of structural organization is built up in the composite with HA/Alg weight ratio of 70/30. The innovative in situ processing might offer a promising future for HA/Alg composite materials in bone regeneration and implant fixation.

Acknowledgments

This work was supported by the national high technology research and development program of China (2006AA03Z358), and the Special Projects for Nanotechnology of Shanghai (0652 nm034).

Copyright information

© Springer Science+Business Media B.V. 2008