Multi-dimensional hydroxyapatite (HAp) nanocluster architectures fabricated via Nafion-assisted biomineralization

Abstract

Multi-dimensional hydroxyapatite (HAp) nanoclusters are fabricated for the first time via a membrane assisted fabrication utilising a Nafion N-117 membrane. The effect of biomineralization time upon the resulting morphology is investigated in detail with the likely crystal growth mechanism proposed. This work opens up the new possibility for preparing HAp with other specific morphological alternatives and is of great value for applications in a plethora of fields.

---

Introduction

Materials with self-assembled structures are expected to have many interesting properties different from that of either individual nanocrystals or bulk materials.^1,2 They are of particular interest both fundamentally and in a range of applications such as when utilised as catalysts, sensors, and energy-storage and well as providing useful sites for adsorption of target molecules and used in purification process.^3–6 However, a key challenge in order for their successful application into these identified applications is their assembly into desired large architectures, often termed as a “bottom-up” nanofabrication process.^7–9

One typical example is the fabrication of hydroxyapatite (Ca₅(PO₄)₃OH, HAp) crystals with complex structures and well-defined morphologies. It is of great significance for their potential uses such as implants for hard tissues, or carriers for genes, enzymes and proteins.¹⁰ It has also been reported that the morphologies and microstructures of HAp greatly influence their chemical and biological properties. Consequently researchers are focused on fabrication methodologies that will produce well defined HAp morphologies with approaches such as electro-precipitation methods,¹¹ ultrasonic irradiation,¹² precipitation,¹³ and hydrothermal synthesis.¹⁴ However, these approaches have their drawbacks, sometimes using toxic chemicals and required high energy which increases the production costs and with the desired morphology difficult to control. Inspired by nature, biomineralization provides a simple strategy in order to construct complex nanoarchitectures.¹⁵ It has been found that the process of biomineralization and assembly can promote the development of a variety of approaches simulating the recognition and nucleation capabilities in biomolecules for inorganic material synthesis.¹⁶ Additionally, a number of advantages should be offered by biomimetic synthesis approaches over conventional inorganic synthesis routes:¹⁷ they usually work in ‘mild’ chemical conditions, which usually imply less consumption of energy and lower release of environmentally unfriendly chemicals.¹⁸

Originating from nature, the process of biomineralization is controlled by templates, such as a specific membrane, colloid, biomolecule or artificial polymers. In prior work, we reported the crystal growth of flower-like HAp agglomerates on egg-shell and bamboo membranes¹⁹ where these membranes provide nucleation sites for crystalline solids where crystal growth is subject to the continuous addition of ions.¹⁷ Consequently we proposed the strategy that sea cucumber-like HAp agglomerates constructed with the assistance of a Nafion N-117 cation-exchange membrane progressed in a self-designed crystallizer and was applied for a ultra-highly sensitive detection of heavy metal ions.²⁰ This facile approach produced self-assembled HAp with unique morphology, enlarged surface areas and multi-adsorbing sites, which allowed the sensitive quantification of heavy metal ions (as low as 10⁻¹² M). To our knowledge, our prior work represents the first on Nafion-assisted assembly of HAp with multi-dimensional architecture.²¹ However, insights into the fabrication mechanism were missing.

In this paper we investigate the fabrication mechanism of multi-dimensional HAp nanocluster architectures fabricated via Nafion-assisted biomineralization.

Experimental

Materials and methods

The fabrication process of self-assembled involves taking 0.5 M Ca(NO₃)₂ and 0.3 M NH₄H₂PO₄ (pH was adjusted to be 11 by 0.1 M NaOH beforehand) solutions were injected into the two cylindrical polypropylene containers respectively, and separated by the tightly fixed Nafion N-117 cation-exchange membrane (DuPont, USA). Then the crystallization process was carried out at 50 °C without stirring. After holding for 1–7 days, the membrane was recovered after gently rinsed with deionized water for a few times, and then dried in air at 40 °C for 12 h. All the chemical reagents were purchased from Alfa Aesar and used as received without any further purification. All aqueous solutions were prepared with ultra-pure water (18 MΩ) from a Milli-Q Plus system (Millipore).

Characterization

The formed crystals were coated with gold and observed by using the NOVA NANOSEM 230 field emission scanning electron microscope (FESEM). Analysis of the novel structures were performed using JEOL JEM-2100F transmission electron microscopy (TEM), selected area electron diffraction (SAED) and high-resolution TEM (HRTEM). The phase structure was identified by a D/ruax 2550PC powder X-ray diffraction meter using Cu Kα (1.5406 Å) radiation (XRD).

Results and discussion

Morphology evolution of the crystals grown on the P-site of membrane

Fig. 1 presents the morphology and crystalline structure evolution of the crystals grown on the P-site of membrane (the site closed to the phosphate solution) after a duration of 1–7 days. The FESEM images clearly suggested that the 1 day crystallization sample (1D) is roughly sphere-like, consisting of nanorods, but promotionally assembled and laterally connected (Fig. 1a and b). The 4 day-crystallization sample (4D) differentiates to be smaller flowers, with the platelets orientationally consisting of nanorods with a length of 450–600 nm and a width of 50–65 nm (Fig. 1c and d). When the crystallization time was extended to 7 days, it was transformed to a sea cucumber-like morphology with diameter of 2.5–3.5 μm (Fig. 1e). The morphology of 7 day-crystallization sample (7D) might potentially be considered a random processes rather than a result of self-assembling of the nanorods during enough crystallization time, confirmed by TEM characterization, and the SAED pattern show obvious multi-crystalline electron diffraction rings that can be indexed to the 210, 211, 112, 300 and 310 planes of HAp (Fig. 1e and f). However, the interplanar spacing (d) values for the diffraction strongest 211, 002 and 310 peaks all declined when the duration time increased from 1 to 7 days, confirmed by XRD identification (PDF 09-0432). Moreover, the corresponding fractural dimensions (f_D) also decreased, calculated by classical Box-counting method.

Fig. 1 Morphology evolution of crystals formed on the P-side of membrane after holding for 1–7 days. (a–e) FESEM images of all the samples, (f, g) TEM and SAED characterization of 7 day-sample, (h) fractural dimension calculation results by box-counting, (i) relationship between holding time and interplanar spacing (d₍₂₁₁₎, d₍₀₀₂₎ and d₍₃₁₀₎).

By comparing the refined structural cell parameters of the three samples (Table 1) in the space group P63/m (176), it can be found that the parameters of 1 day crystallization HAp are a = 0.9343 nm, c = 0.6882 nm, and the calculated cell volume is 0.5203 nm³. When the crystallization extended to 4 days, the parameters correspondingly changed to 0.9348 nm and 0.6899 nm, and the calculated cell volume is 0.5221 nm³. It is clear that, after another 3 days' crystallization, there is a slight decrease in the c-axis length and an increase in the a-axis length, which is corresponding with the lattice distortion, suggested by the 2θ₂₁₁ value deviated toward high-angle. The refined structural cell parameters of a = 0.9360 nm and c = 0.6892 nm. These changes result in an increase of cell volume, corresponding to 0.5229 nm³. On the other hand, the crystallinity of obtained HAp significantly increased from 63.51% to 94.72% in the process mentioned above.

Table 1 The fitting results of XRD patterns in Fig. 2

Days	Crystallinity (%)	2θ211 (°)	a (nm)	c (nm)	Cell volume (nm^3)
1	63.51	31.681	0.9343	0.6882	0.5203
4	75.63	31.770	0.9348	0.6899	0.5221
7	94.72	31.842	0.9360	0.6892	0.5229

---

Morphology evolution of the crystal grown on the Ca-site of membrane

On the Ca-site of membrane (the site closed to calcium solution), the grown crystals are totally different from those on the P-site. The morphology evolution has been illustrated in Fig. 2. It is obviously observed that after being crystallized for 1 day, the crystals on the Ca-site comprise irregular prisms, with a size of 100–200 nm (Fig. 2a). When the crystallization time is extended to 4 days, these prisms transform to tetrahedrons, with distinct edges of 150–250 nm (Fig. 2b). Moreover, it is very interesting to notice that the products after 7 days' crystallization comprise triangular prisms and display typical lamellar growth characteristics. The tetrahedrons are 400–500 nm in size (Fig. 2c).

Fig. 2 Morphology evolution of the crystal grown on the Ca-site of membrane.

These tetrahedrons were semi-quantitatively identified by EDAX-SEM (Table 2) where their composition can be inferred by considering the nature of a cation-exchange membrane. The relevant reaction formulas are listed as below. Theoretically, in the calcium solution:

Ca(NO₃)₂ → Ca²⁺ + 2NO₃⁻

and in the phosphate solution:

NH₄H₂PO₄ → NH₄⁺ + H₂PO₄⁻

H₂PO₄⁻ ↔ H⁺ + HPO₄²⁻

HPO₄²⁻ ↔ H⁺ + PO₄³⁻

Table 2 Wt% and at% of elements in the compounds grown on the Ca-site characterized by EDAX-SEM facility

Days	Wt% of elements				At% of elements
	O	Na	P	Ca	O	Na	P	Ca
1	52.30	—	14.74	32.95	3.27	—	0.48	0.82
4	29.69	1.73	22.68	45.89	1.86	0.08	0.73	1.15
7	37.90	3.09	18.07	40.94	2.37	0.13	0.58	1.02

---

However, in the phosphate solution with a pH value of 11, the H⁺ generated in the last two reactions would be neutralized by amounts of OH⁻, which was in favour of the produce of PO₄³⁻. It is reasonable that the NH₄⁺ ions hydrolysed immediately because no N element was detected by EDAX. However, it should be noticed that there are Na component in the 4 day- and 7 day-crystallization samples and the at% value increased with the holding time. It can be inferred that the Na⁺ ions in the pH modifier NaOH participated and played role in the crystallization. The tetrahedral shapes grown on the Ca-site of membrane can be considered as a kind of complex compound combined by Ca²⁺, Na⁺, and PO₄³⁻.

Possible crystal growth mechanism of sea cucumber-like HAp

Consequently, it can be assumed that the obtained multi-dimensional HAp crystals have a plate-like structure built upon nanorods. The key feature of their morphology is the specific assembly of these nanorods, which consists of the orientation of the crystallographic planes. This fact might be associated with the mechanism for the HAp synthesis under the control of Nafion membrane with specific structure. Some of the explanations related to this mechanism suggested an amorphous calcium phosphate (ACP)–octacalcium phosphate (OCP)–hydroxyapatite (HAp) transformation.²² In our previous reports, we also found this transforming process in terms of eggshell membrane- or bamboo membrane-assisted biomineralization conditions.²⁰ In this work, we have revealed the specific morphology and structure of synthesized HAp. In the existence of Nafion membrane, particles have their natural, thermodynamically most stable growth direction.²³ Although the XRD results in the present work did not clearly illustrate this transformation, it can still be considered as a very fast process (within 1 day), or directly generated HAp without this owing to the specific functional groups on Nafion template.

In view of the fact that the sea cucumber-like HAp agglomerates formed on the P-site of Nafion N-117 membrane and simplified the morphology as a “tube”, then two possible crystal growth mechanisms can be proposed:

One is “grown at the top of the tube”, and another is “grown at the bottom of the tube” (Fig. 3). It has been mentioned that there are ionized PO₄³⁻ and OH⁻ in NH₄H₂PO₄ solution. That is to say, for the “grown at the top of the tube” mechanism, Ca²⁺ transmitted through the cation-exchange membrane at first, then moved from the surface to the top of “tube”, and reacted with PO₄³⁻ and OH⁻. For the other mechanism, Ca²⁺ reacted with PO₄³⁻ and OH⁻ at the interface of membrane and the bottom of “tube”. The reaction formula is listed as below:

5Ca²⁺ + 3PO₄³⁻ + OH⁻ → Ca₅(PO₄)₃OH

However, it can be inferred that the prefer mechanism is “grown at the bottom of the tube”, as suggested from the FESEM and TEM characterization. Firstly, the sphere-like agglomerates formed on the membrane and then gradually differentiated to flower-like morphology. When the crystallization time further extended, these HAp nanorods are laterally “glued” during the biomineralization and grow towards the cylindrical water channels of Nafion but while no amorphous Hap is observed this is subjective. We note that a tracer method might be useful to help ascertain where the calcium ions migrate to unambiguously determine bottom or top growth formation. Meanwhile, according to Ostwald Ripening mechanism,²⁴ it can be assumed that the HAp microcrystallines will be consumed to form larger crystallines, and the sea cucumber-like structures are created as a result of this.

Fig. 3 (a) Possible crystal growth mechanism and (b) the microstructure of Nafion membrane.

Electrochemical behaviour in the adsorption process of Pb²⁺ and Cd²⁺ ions on the HAp modified carbon paste electrode

Fig. 4 compares the electrochemical behaviour of HAp modified carbon paste electrode, characterized by EIS determination. It can be obviously noticed that for all the three electrodes, pure carbon paste (CP) electrode and 1 d-HAp/7 d-HAp modified electrodes, the EIS patterns are different at a scan potential from −1 V to −0.4 V in 0.2 M PBS containing 100 nM Pb²⁺ and Cd²⁺ ions (Fig. 4a). It suggests that for CP and 7 d-HAp electrodes, the electrochemical behaviours are both diffusion limited. However, the semi-circle in the EIS pattern of 1 d-HAp electrode reveals that the electron transfer hardly occurred on its surface. Besides, the EIS tests at the stripping potential of Pb²⁺ (−0.8 V) and Cd²⁺ (−0.6 V) also confirmed the adsorption process of these two target ions preferred to take place on 7 d-HAp electrode than the others (Fig. 4b). Associated with crystalline lattice characterization results, this feature can be attribute to the larger cell volume that favours the crystalline occupation of Pb²⁺ and Cd²⁺ ions, and further verifies that the 7 d-HAp modified CP electrode is the most suitable for electrochemical sensing of Pb²⁺ and Cd²⁺, which corresponds to our previous research.

Fig. 4 Electrochemical behaviour in the adsorption process of Pb²⁺ and Cd²⁺ ions on the HAp modified carbon paste electrode under different potentials (vs. SCE) (a) at a scan potential from −1.0 V to −0.4 V and (b–d) respectively at −0.8 V, −0.6 V and −0.5 V, in 0.2 M PBS containing 100 nM Pb²⁺ and Cd²⁺ ions.

Conclusions

In the present work, we attempt to find out the crystal growth mechanism of multi-dimensional HAp materials by investigating the influence of crystallization time and principle role of Nafion cation-exchange membrane on the morphology and microstructure in detail. It was revealed that the major function of Nafion membrane is supposed to control the diffusion of aqueous cations and the crystal growth follows the “grown at the bottom of the tube” mechanism towards the cylindrical water channels in the membrane. In this process, both the interplanar spacing and fractal dimension values decreased with the duration time while the cell volume increased. At the same time, it was confirmed that the 7 day-HAp crystals with larger cell volume and smaller interplanar spacing were suitable for electrochemical sensing of Pb²⁺ and Cd²⁺, which is corresponding to the previous research. It has been believed that the proposed Nafion membrane assistant strategy would inspire us to design and construct other new materials with multi-dimensional structure, which are expected to equip more novel features.

Acknowledgements

This project was financially supported by General Financial Grant and Special Financial Grant from the China Postdoctoral Science Foundation (No. 2012M521431 and 2013T60720).

Notes and references

1. S. Burazerovic, J. Gradinaru, J. Pierron and T. R. Ward, Angew. Chem., Int. Ed., 2007, 46, 5510.
2. M. H. Park, B. G. Choi and B. Jeong, Adv. Funct. Mater., 2012, 22, 5118.
3. J. Mosquera, S. Zarra and J. R. Nitschke, Angew. Chem., Int. Ed., 2014, 53, 1556.
4. J. Gong, Z. Zhou, X. Hu, M. Wong, K. Wong and Z. Du, ACS Appl. Mater. Interfaces, 2009, 1, 26.
5. M. Kim, G. H. Jeong, K. Y. Lee, K. Kwon and S. W. Han, J. Mater. Chem., 2008, 18, 2208.
6. Y. Qiao, Y. Wang, Z. Yang, Y. Lin and J. Huang, Chem. Mater., 2011, 23, 1182.
7. J. Eiblmeier, M. Kellermeier, M. Deng, L. Kienle, J. M. García Ruiz and W. Kunz, Chem. Mater., 2013, 25, 1842.
8. S. Yang and L. Gao, J. Am. Chem. Soc., 2006, 128, 9330.
9. Y. Zhu, W. Zhao, H. Chen and J. Shi, J. Phys. Chem. C, 2007, 111, 5281.
10. Z. Yang and C. Zhang, Appl. Surf. Sci., 2009, 255, 4569; C. Xu, D. He, L. Zeng and S. Luo, Colloids Surf., B, 2009, 73, 360.
11. S. D. Huelin, H. R. Baker, E. F. Merschrod S. and K. M. Poduska, Cryst. Growth Des., 2006, 6, 2634.
12. B. Palazzo, D. Walsh, M. Iafisco, E. Foresti, L. Bertinetti, G. Martra, C. L. Bianchi, G. Cappelletti and N. Rover, Acta Biomater., 2009, 5, 1241.
13. M. Jevtić, A. Radulović, N. Ignjatović, M. Mitrić and D. Uskokovic, Acta Biomater., 2009, 5, 208.
14. L. Cao, C. Zhang and J. Huang, Mater. Lett., 2005, 59, 1902.
15. R. R. Naik, S. J. Stringer, G. Agarwal, S. E. Jones and M. O. Stone, Nat. Mater., 2002, 1, 169.
16. S. Brown, M. Sarikaya and E. Johnson, J. Mol. Biol., 2000, 299, 725; J. Cha, G. D. Stucky, D. E. Morse and T. J. Deming, Nature, 2000, 403, 289.
17. G. Pezzotti, S. M. F. Asmus, L. P. Ferroni and S. Miki, J. Mater. Sci., 2002, 13, 783.
18. S. A. Hutchens, R. S. Benson, B. R. Evans, H. M. O' Neill and C. J. Rawn, Biomaterials, 2006, 27, 4661; R. Yoh, T. Matsumoto, J. Sasaki and T. Sohmura, J. Biomed. Mater. Res., Part A, 2008, 87, 222; E. Landi, A. Tampieri, M. Mattioli-Belmonte, G. Celotti, M. Sandri, A. Gigante, P. Fava and G. Biagini, J. Eur. Ceram. Soc., 2006, 26, 2593.
19. M. Takiguchi, K. Igarashi, M. Azuma and H. Ooshima, Cryst. Growth Des., 2006, 6, 1611; M. Takiguchi, K. Igarashi, M. Azuma and H. Ooshima, Cryst. Growth Des., 2006, 6, 2754.
20. Y. Zhang, Y. Liu, X. Ji, C. E. Banks and J. Song, Colloids Surf., B, 2011, 82, 490; Y. Zhang, Y. Liu and X. Ji, Mater. Lett., 2011, 65, 1982.
21. Y. Zhang, Y. Liu, X. Ji, C. E. Banks and W. Zhang, Chem. Commun., 2011, 47, 4126.
22. R. Xin, Y. Leng and N. Wang, J. Cryst. Growth, 2006, 289, 339; W. E. Brown, J. P. Smith, J. R. Lehr and W. A. Frazier, Nature, 1962, 196, 1050; M. S. Tung and W. E. Brown, Calcif. Tissue Int., 1983, 35, 783.
23. A. L. Rollet, O. Diat and G. Gebel, J. Phys. Chem. B, 2002, 106, 3033.
24. Y. Xu, J. Jiang, Y. Lu, R. Sun, J. Song, L. Ren and S. Yu, Cryst. Growth Des., 2008, 8, 3822.

---