Effect of conventional and microwave heating on the structural and physicochemical properties of silver-doped hydroxyapatite nanopowders

Abstract

This paper discussed the influence of duration and modes of heating on structural and physicochemical properties of silver-doped hydroxyapatite (Ag-HAP) nanopowders (NPs). Conventional electric furnace and microwave (MW) heating modes were employed during the synthesis. MW mode of synthesis was relatively an efficient method as it prepared monolithic HAP NP in just one minute of heating. With the increase in duration of heating in both modes of heating; lattice parameters, crystal size, lattice strain, Ca/P ratio, and degree of crystallinity of HAP phase increased. The morphology of particles was rod-shaped having aspect ratio between 2 and 3. EDX confirmed the presence of Ag and corroborated the formation of apatite. The hydrodynamic diameter of Ag-HAP NPs was significantly bigger than the particle sizes calculated using XRD, FESEM, and TEM. Thus, an overall examination concluded MW as an efficient mode of synthesis, able to produce Ag-HAP NPs in a possible minimum time.

Keywords

Hydroxyapatite silver microwave conventional electric furnace duration of heating Rietveld nanorods strain

Introduction

Success and long-term survival of orthopaedic and dental implants in the human body depend on the prevention of infections which usually encountered following implant placement after surgery [1]. Among the various antibiotic ionic metals such as Cu²⁺, Zn²⁺, Ag⁺, and so forth, Ag⁺ exhibits a broad range of antimicrobial ability and possess many advantages such as excellent biocompatibility and low toxicity towards mammalian cells [2]. All common bacterial strains involved in implant-related infections are eliminated by Ag [3].

Furthermore, synthetic hydroxyapatite (HAP, Ca₁₀(PO₄)₆(OH)₂) has been a most promising ceramic material widely used as a bone graft, a drug carrier, and prosthetic coatings because of its excellent biocompatibility, bioactivity, and osteoconductivity. Also, cation exchange rate of HAP is very high with Ag ions [4], and hence it is easy to incorporate Ag⁺ into HAP crystal structure. Among the various synthesis techniques, the ion-exchange route is a contemporary and widely employed method for doping of Ag⁺ into HAP crystal structure [1]. With this approach, Ag⁺ will be directly inserted into the lattice structure of HAP [5] to produce stable silver-doped hydroxyapatite (Ag-HAP) compounds [6]. The process parameters of an ion-exchange route such as temperature, duration, and modes of the heating i.e. conventional electric furnace (CEF) and microwave (MW) significantly influence the properties of resultant Ag-HAP NPs. For instance, Jadalannagari et al. [7] and Costescu et al. [8] synthesised Ag-HAP NPs using CEF at 100°C for 48 and 72 hours, respectively. Furthermore, Popa et al. [9] and Chung et al. [10] aged the Ag-HAP sol in CEF at 40°C for 60 minutes and 80°C for 16 hours, respectively, to synthesise Ag-HAP NPs. Similarly, Singh et al. [11] heat-treated the Ag-HAP sols at 80°C for 2 hours in CEF to synthesise Ag-HAP NPs. On the other hand, Iqbal et al. [12] employed MW and irradiated the Ag-HAP sol for 10 minutes at 800 W and synthesised Ag-HAP NPs. Similarly, MW assisted refluxing of Ag-HAP sol was carried out at 600 W for seven minutes to synthesise Ag-HAP NPs as reported by Iqbal et al. [13]. In an another study by Iqbal et al. [14], MW refluxing of Ag-Zn-doped HAP sol at 800 W for 10 minutes was carried out. Rameshbabu et al. [15] irradiated Ag-HAP sols at 800 W for 30 minutes and synthesised Ag-HAP NPs. Therefore, in the light of the cited and other reviewed literature, it was evident that the authors have employed different durations in conventional and MW modes of heating to synthesise Ag-HAP NPs. Often, it makes difficult to select the optimal duration of heating during CEF and MW modes of heating to synthesise Ag-HAP NPs.

With an aim to overcome this limitation, the present study comprehensively examined the influence of duration and mode of heating on structural and physicochemical properties of Ag-HAP NPs. A wide range of duration of heating in both CEF and MW modes of heating were chosen to synthesise Ag-HAP NPs. The synthesis was carried out using an atomic ratio of Ag/(Ag + Ca) at 3% and (Ca + Ag)/P molar ratio of 1.67. The resultant NPs were characterised using XRD, FESEM, TEM, EDX, and DLS techniques. The crystallography of NPs was studied using Rietveld refinement method. Therefore, this paper provides an insight in choosing the optimal duration of heating in CEF and MW modes to synthesise Ag-HAP NPs for this composition ratio.

Materials and method

All chemicals were analytical reagents and used without further purification. All experiments were carried out in an open air. Calcium nitrate tetrahydrate (CNT, Merck, 99%), diammonium hydrogen orthophosphate (DAHP, Merck, 99%), and silver nitrate (AgNO₃, Merck, 99%) were employed as Ca, P, and Ag ion sources, respectively. Double distilled water (DDW) was used as a solvent. Ammonium hydroxide (NH₄OH, Merck, 25%) was used to improve the gelation. Figure 1 details the adopted synthesis protocol.

Briefly, calculated amounts of CNT, DAHP, and AgNO₃ reagents were dissolved in DDW (100 ml) to prepare Ca, P, and Ag ion precursors. Weights of CNT, DAHP, and AgNO₃ reagents were calculated using an atomic ratio of Ag/(Ag + Ca) at 3% and (Ca + Ag)/P molar ratio of 1.67. The AgNO₃ precursor was added drop-wise into CNT precursor and stirred vigorously using magnetic stirrer followed by the drop-wise addition of DAHP precursor to obtain Ag-HAP sol having a molar ratio of (Ca + Ag)/P equal to 1.67. NH₄OH was added into Ag-HAP sol till its pH increased to 11. After stirring for 10 minutes, resultant Ag-HAP sol was separately heated in different environments i.e. in CEF at 80°C and MW (800 W, 2.45 GHz) for selected durations. The resultant solutions were centrifuged at 4500 rpm for 5 minutes, washed with DDW until the odour of ammonia obliterated (typically takes 5–6 washings), followed by drying in an electric furnace at 60 ± 2°C for 24 hours. Dried powders were gently crushed using mortar and pestle. Nomenclatures of NPs, selected durations corresponding to CEF and MW modes of heating are given in Table 1.

Table 1

Nomenclature of NPs classified according to mode and duration of heating.

Nomenclature	C-0	C-1	C-4	C-8	C-24	M-1	M-10	M-20	M-30	M-40
Mode	CEF					MW
Time	0 h	1 h	4 h	8 h	24 h	1 min	10 min	20 min	30 min	40 min

Characterisation

XRD (Philips X'Pert 1710) was performed using CuKα radiation (λ = 1.54 Å, 2θ = 10–80°, step size 0.02°, time per step 20 s and scan speed 0.005° s⁻¹). Crystallographic information of NPs was obtained by full-pattern Rietveld refinement of XRD patterns using MAUD 2.7 software. Refinement of structural parameters (atomic and weight fractions, lattice parameters, occupancies, and thermal parameters) and microstructural parameters (particle size and lattice strain) of Ag-HAP NPs was carried out. Rietveld programme modelled the background by fifth-order polynomial function and shape of the peaks was studied using the Pseudo-Voigt algorithm. Least square refinement was used as a refinement weighting model. Refinement was carried out until close fit between experimental and calculated patterns was attained. The experimental diffraction patterns were simulated using HAP (JCPDS # 09-0432), β-TCP (09-0169), Monetite (01-071-1759), CaCO₃ (5-586), CaO (01-074-1226), Ag (00-002-1098), Ag₂O (00-041-1104) and Ag₃PO₄ (01-084-0194) phases.

Crystal size ( ) of HAP was calculated using Scherrer formula (Equation 1): (1) where X_s is in nanometre, λ is the wavelength of the X-ray beam, β is the broadening of diffraction line at half of its maximum intensity in radians, and is the Bragg's diffraction angle ( ). The FWHM of (0 0 2), (2 1 1), (3 0 0), (2 0 2), and (3 1 0) diffraction peaks of HAP were used to calculate the crystal size. The degree of crystallinity (X_c) of NPs was calculated according to Equation (2) [16]. Crystal size and lattice strain in HAP were calculated using Williamson–Hall-ISM model (Equation 3) and Williamson–Hall-ASM model (Equation 4). (2) (3) (4) where is the intensity of (3 0 0) diffraction peak, is the intensity of hollow between (1 1 2) and (3 0 0) diffraction peaks of HAP phase, K is the shape factor, ε is lattice strain, σ is lattice stress and is Young's modulus in the direction normal to the set of (h k l) crystal lattice planes. Morphology and elemental composition of NPs were examined using FESEM/EDX (JEOL) operated at 20 kV. Samples were gold coated prior to FESEM/ EDX testing. The efficiency of doping of Ag was calculated according to Equation (5). TEM (Hitachi, 7500) with resolution of 0.2 nm, operating at an accelerating voltage of 80 kV was used for observing particle size and morphology. The NPs were ultrasonically dispersed in ethanol to form a dilute suspension and then a drop of suspension was dropped on carbon-coated copper grid of 300 mesh for observation using TEM. Particle size and aspect ratio were determined by Image-J software [17]. The hydrodynamic diameter of NPs was determined by dynamic light scattering (DLS) technique (Malvern Zetasizer Nano ZS-90, UK). One milligram of NP was dispersed in 10 ml ethanol and sonicated for 30 minutes. One millilitre of the supernatant was then removed and used for DLS measurements. (5) where η is the efficiency, C_f is the final concentration of Ag in the NP and C_t is the theoretical concentration of Ag in the NP.

Results and discussion

Phase structure

XRD patterns of Ag-HAP NPs synthesised in CEF (C-0, C-1, C-4, C-8, and C-24) and MW (M-1, M-10, M-20, M-30, and M-40) are shown in Figure 2. Rietveld refinement suggested the presence of characteristic peaks (0 0 2), (2 1 1), (3 0 0), (2 0 2), and (3 1 0) of HAP phase in synthesised NPs. In addition to HAP, monetite phase was also present in C-0 NP only as shown in Figure 2(a). Peaks of pure or bonded phases of Ag were not observed in the XRD pattern of any NP [8]. In addition, impurities such as CaCO₃ and CaO were also absent. Thus, the qualitative-phase analysis suggested single-phase apatitic constitution of C-1, C-4, C-8, C-24, M-1, M-10, M-20, M-30, and M-40 NPs. Furthermore, the presence of broader peaks suggested the low crystallinity of NPs [15]. Good match regarding peak positions and peak intensities was observed on comparing the experimental XRD patterns with the standard JCPDS files of constituent phases.

Ciobanu et al. [4,18] and Singh et al. [11] also observed only apatitic phase in Ag-HAP NPs. The absence of peaks about Ag or its bonded phases was either due to the successful substitution of Ag⁺ at the sites of Ca²⁺ without altering the HAP crystal structure [8] or was probably due to the doping of small amount of Ag (Ca_10-xAg _x (PO₄)₆(OH)_2-x, x = 0.03). Incorporation of Ag⁺ into apatite was assumed to be based on Equation (6) as suggested by Lim et al. [19]. Formation of bioactive monetite (CaHPO₄) phase in C-0 NP indicated the onset of phase transformation process [20] towards the evolution of HAP. Negligible time of ripening of C-0 NP caused the incomplete nucleation of HAP and thus formed an intermediate phase of CaHPO₄. The proposed chemical reaction to form CaHPO₄ is simplified in Equation (7). Furthermore, the transformation from CaHPO₄ to HAP completed within 1 hour, which was much shorter than the duration of 4 hours reported elsewhere [20,21]. (6) (7)

Crystallographic structure

For reference, experimental and calculated XRD profiles of C-1 and M-1 NPs are shown in Figure 3, together with the difference between both profiles (with an aim to save space, profiles of all other NPs are not shown in this paper). The goodness of fit values (σ, R_wp, R_exp) about the refinement of each NP are given in Table 2. The goodness of fit parameter (σ) was calculated using weighted profile R-factor (R_wp) and expected R-factor (R_exp) parameters using Equation (8). Obtained (σ) values were relatively small i.e. less than 4, which was considered as acceptable according to the fundamental principle of the goodness of fit [22]. Various crystallographic parameters such as wt-% of constituent phase(s), lattice parameters, crystal size (Rietveld, Scherrer, WH-ISM, and WH-ASM), lattice strain (Rietveld and WH-ISM), and Ca/P ratio (at.-% and wt-%) of all NPs were calculated. (8) Figure 4 shows the quantitative-phase analysis of all NPs. Only C-0 NP was biphasic composed of 95% of CaHPO₄ and 5% of HAP, whereas, all other NPs were monolithic HAP (100%) NPs.

Table 2

Goodness of fit values pertinent to Rietveld refinement.

	NPs
Parameter	C-0	C-1	C-4	C-8	C-24	M-1	M-10	M-20	M-30	M-40
σ	2.77	1.60	1.51	2.71	1.68	1.68	1.51	1.95	1.85	1.73
R _wp	0.21	0.13	0.12	0.22	0.13	0.13	0.12	0.15	0.14	0.13
R _exp	0.075	0.081	0.079	0.081	0.077	0.077	0.079	0.076	0.075	0.075

Figure 5 shows the variation in lattice parameters of constituent HAP phase corresponding to synthesised NPs. Results indicated that except C-0 NP, lattice parameters of HAP in all other Ag-HAP NPs were close to the lattice parameters of stoichiometric HAP (a = 9.418 Å and c = 6.884 Å). Furthermore, the linear fitting model suggested that HAP lattice parameters along both a-axis and c-axis increased with the increase in duration of heating in both CEF and MW heating environments. Also, lattice dimensions of HAP crystals in MW synthesised NPs were relatively bigger than NPs synthesised in CEF.

Figure 6 shows the sizes of HAP crystal determined using Rietveld, Scherrer, WH-ISM, and WH-ASM models. Primarily, HAP crystals were nanodimensional in size varied between 10 to 40 nm as predicted by operating models. Furthermore, the linear fitting model suggested that size of HAP crystal increased with the increase in duration of heating in both CEF and MW modes of heating. Iqbal et al. [13] reported crystal size range of 18–33 nm for MW synthesised HAP/Ag (0.05 < x < 0.2) NPs. Similarly, Pang et al. [16] reported crystal size up to 50 nm for Ag-HAP NPs.

Figure 7 compares the lattice strain of HAP crystals as predicted using Rietveld and WH-ISM models corresponding to synthesised NPs. HAP lattice structure in all Ag-HAP NPs was subjected to tensile strain as predicted by both operating models. Also, HAP lattice strain increased with the increase in duration of heating in both CEF and MW modes of heating. This increase in lattice strain corroborated the previous finding of an increase in HAP crystal size (Rietveld and WH-ISM), considering the contribution of lattice strain behind the growth of crystal size.

Figure 8 shows the variations in Ca/P ratio (at.-% and wt-%) corresponding to synthesised NPs. Except for C-0 NP, resultant at.-% Ca/P and wt-% Ca/P ratio in all Ag-HAP NPs were close to the theoretical value of 1.67 and 2.15, respectively, thus further confirmed the formation of HAP phase in synthesised NPs. Furthermore, the linear fitting model suggested the increasing trend of Ca/P ratio (both at.-% and wt-%) with an increase in duration of heating in both CEF and MW modes of heating.

Figure 9 shows the degree of crystallinity of synthesised Ag-HAP NPs. Results suggested that degree of crystallinity increased with the increase in duration of heating in both modes of heating. As-prepared Ag-HAP NPs were moderately crystalline having a degree of crystallinity less than 30%. Pang et al. [16] reported an increase in crystallinity from 13% to 44% with the increase in ripening time of Ag-HAP NPs. The observed increase in crystallinity as a function of the duration of heating can be explained by the mechanism of crystal growth (Figure 6) in solution during heating [16].

Morphological and elemental structure

The morphology of C-1, C-24, and M-1, M-30 NPs are shown in Figures 10 and 11, respectively. All Ag-HAP NPs were white, bouncy, and free-flowing powders. NPs were highly agglomerated [16] and constituted grains of irregular shapes and sizes as shown in Figures 10(a,c) and 11(a,c). These irregularly shaped grains consisted of numerous nanodimensional elongated and spherical-shaped particles of vivid sizes in C-1, C-24, and M-1, M-30 NPs as shown in Figures 10(b,d) and 11(b,d), respectively. Agglomeration is an intrinsic property of NPs owing to their high surface energy. Average grain size was measured to be 69 ± 15 nm for C-1 NP, 63 ± 14 nm for C-24 NP, 75 ± 15 nm for M-1 NP, and 56 ± 14 nm for M-30 NP. Particles having an average size of 85 and 72 nm for MW synthesised Ag-HAP (x = 0.03) and Ag-HAP (x = 0.05) NPs, respectively were reported by Iqbal et al. [12,13].

TEM micrographs suggested rod-shaped particle morphology of C-1, C-24, M-1, and M-30 NPs as shown in Figures 12–13. Average length (parallel to c-axis) and width (perpendicular to c-axis) of particles were measured to be 34 ± 11 and 11 ± 2 nm for C-1 NP, 29 ± 7 and 13 ± 4 nm for C-24 NP, 26 ± 8 and 13 ± 3 nm for M-1 NP, 32 ± 14 and 12 ± 4 nm for M-30 NP. The particle sizes measured using TEM were close to the results of crystal sizes predicted using XRD (Figure 6). Miranda et al. [3] reported globular-shaped Ag-HAP particles of vivid sizes varied between 10 and 50 nm. Ciobanu et al. [4] reported particle size of ∼40 nm for Ag-HAP (x = 0.2) NPs.

Qualitative and quantitative elemental constitution of M-30 NP is shown in Figure 14. The presence of peaks about Ca, P, and O elements corroborated the XRD-phase analysis of apatitic formation in M-30 NP. The Ag element was also present and thus supported the formation of Ag-HAP NP. Average Ca/P atomic ratio of 1.11 and average Ca/P weight ratio of 1.43 showed some disparity corresponding to similar results of 1.66 and 2.13 as calculated using XRD. Similarly, average wt-% of Ag was detected to be 2.51% of the theoretically calculated doping value of 3%, and hence the doping efficiency (η) of Ag was calculated to be ∼84%. This value indicated that only a small percentage of Ag lost during washing process of the NPs [3].

Hydrodynamic diameter

The average hydrodynamic diameter of all Ag-HAP NPs is given in Table 3. These hydrodynamic diameters were significantly bigger than particle sizes measured using FESEM and TEM (Figures 10–13). Hydrodynamic diameter typically includes the solvation layers due to which hydrodynamic diameter are always greater than the sizes estimated using FESEM and TEM. It was also anticipated that the particles were dispersed as agglomerates in suspension form. Agglomeration is a common characteristic in NPs owing to their high surface energy.

Table 3

Hydrodynamic diameter of Ag-HAP NPs.

	NPs
Parameter	C-0	C-1	C-4	C-8	C-24	M-1	M-10	M-20	M-30	M-40
Diameter (nm)	457 ± 27	432 ± 20	319 ± 27	332 ± 19	361 ± 26	103 ± 15	169 ± 28	211 ± 13	229 ± 21	248 ± 13

Conclusion

Ag-HAP NPs were successfully synthesised using different durations of heating in CEF and MW modes of heating. Qualitative- and quantitative-phase analyses suggested that 1 hour of heating in CEF, whereas only 1 minute of heating in MW produced the monolithic Ag-HAP NPs. The crystallographic study indicated that with the increase in duration of heating in CEF and MW environments; lattice parameters, crystal size, lattice strain, Ca/P ratio, and degree of crystallinity of constituent HAP phase increased. The morphological evaluation suggested the highly agglomerated state of Ag-HAP NPs constituted of elongated nanodimensional particles. Further examination using TEM indicated the rod-shaped particle morphology of Ag-HAP NPs. The particle sizes measured using TEM were compared to be close to the crystal sizes determined using XRD analysis. EDX confirmed the presence of Ag element along with the formation of apatite given the presence of Ca, P, and O elements. The hydrodynamic diameter of NPs was significantly bigger than the particle sizes measured using FESEM, TEM, and XRD analysis. The overall examination of evaluated properties concluded MW as an efficient mode of synthesis, able to produce Ag-HAP NPs in a minimum possible time.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Singh

, Singh

, Verma

, . Structural analysis of silver doped hydroxyapatite nanopowders by Rietveld refinement. Trans Indian Inst Met. 2016;70(8):1973–1980. doi: 10.1007/s12666-016-1019-0

Park

, Lee

, Bang

., Optimal synthesis and characterization of Ag nanofluids by electrical explosion of wires in liquids. Nanoscale Res Lett. 2011;6(1):223–233. doi: 10.1186/1556-276X-6-223

Miranda

, Fernandez

, Diaz

., Silver-hydroxyapatite nanocomposites as bactericidal and fungicidal materials. Int J Mater Res. 2010;101(1):122–127. doi: 10.3139/146.110256

Ciobanu

, Massuyeau

, Constantin

., Structural and physical properties of antibacterial Ag-doped nano-hydroxyapatite synthesized at 100°C. Nanoscale Res Lett. 2011;6(1):613–620. doi: 10.1186/1556-276X-6-613

, Liao

, Xu

., Preparation and antibacterial effect of silver-hydroxyapatite/titania nanocomposite thin film on titanium. Appl Surf Sci. 2008;255(2):435–438. doi: 10.1016/j.apsusc.2008.06.083

Melaiye

, Youngs

WJ.

Silver and its application as an antimicrobial agent. Expert Opin Ther Pat. 2005;15(2):125–130. doi: 10.1517/13543776.15.2.125

Jadalannagari

, Deshmukh

, Ramanan

., Antimicrobial activity of hemocompatible silver doped hydroxyapatite nanoparticles synthesized by modified sol-gel technique. Appl Nanosci. 2014;4(2):133–141. doi: 10.1007/s13204-013-0197-x

Costescu

, Ciobanu

, Iconaru

., Fabrication, characterization, and antimicrobial activity, evaluation of low silver concentrations in silver-doped hydroxyapatite nanoparticles. J Nanomater. 2013;5:1–9. doi: 10.1155/2013/194854

Popa

, Ciobanu

, Voicu

., Influence of thermal treatment on the antimicrobial activity of silver-doped biological apatite. Nanoscale Res Lett. 2015;10(1):502–512. doi: 10.1186/s11671-015-1211-x

10.

Chung

, Hsieh

, Huang

., Anti-microbial hydroxyapatite particles synthesized by a sol-gel route. J Sol-Gel Sci Technol. 2005;33(2):229–239. doi: 10.1007/s10971-005-5618-1

11.

Singh

, Dubey

, Kumar

., In vitro biocompatibility and antimicrobial activity of wet chemically prepared Ca_10-xAg_x(PO₄)₆(OH)₂ (0.0≤x≤0.5) hydroxyapatites. Mater Sci Eng, C. 2011;31(7):1320–1329. doi: 10.1016/j.msec.2011.04.015

12.

Iqbal

, Kadir

MRA

, Malek

NANN

., Rapid microwave assisted synthesis and characterization of nanosized silver-doped hydroxyapatite with antibacterial properties. Mater Lett. 2012;89:118–122. doi: 10.1016/j.matlet.2012.08.057

13.

Iqbal

, Kadir

MRA

, Malek

NANN

., Characterization and antibacterial properties of stable silver substituted hydroxyapatite nanoparticles synthesized through surfactant assisted microwave process. Mater Res Bull. 2013;48(9):3172–3177. doi: 10.1016/j.materresbull.2013.04.068

14.

Iqbal

, Kadir

MRA

, Mahmood

., Characterization, antibacterial and in vitro compatibility of zinc-silver doped hydroxyapatite nanoparticles prepared through microwave synthesis. Ceram Int. 2014;40(3):4507–4513. doi: 10.1016/j.ceramint.2013.08.125

15.

Rameshbabu

, Sampath Kumar

, Prabhakar

., Antibacterial nanosized silver substituted hydroxyapatite: synthesis and characterization. J Biomed Mater Res. 2007;80A(3):581–591. doi: 10.1002/jbm.a.30958

16.

Pang

, Bao

Influence of temperature, ripening time and calcination on the morphology and crystallinity of hydroxyapatite nanoparticles. J Eur Ceram Soc. 2003;23(10):1697–1704. doi: 10.1016/S0955-2219(02)00413-2

17.

Kothapalli

, Wei

, Legeros

., Influence of temperature and aging time on HA synthesized by the hydrothermal method. J Mater Sci-Mater Med. 2005;16(5):441–446. doi: 10.1007/s10856-005-6984-5

18.

Ciobanu

, Iconaru

, Chiriuc

., Synthesis and antimicrobial activity of silver-doped hydroxyapatite nanopowders. BioMed Res Int. 2013;10.

19.

Lim

, Chang

, Thian

ES.

Development of nanosized silver-substituted apatite for biomedical applications: a review. Nanomed Nanotechnol. 2005;11:1331–1344. doi: 10.1016/j.nano.2015.03.016

20.

, Zhu

, Chang

Monetite formed in mixed solvents of water and ethylene glycol and its transformation to hydroxyapatite. J Phys Chem B. 2006;110(29):14226–14230. doi: 10.1021/jp061738r

21.

Da Silva

, Lima

JHC

, Soares

., Transformation of monetite to hydroxyapatite in bioactive coatings on titanium. Surf Coat Technol. 2001;137(2):270–276. doi: 10.1016/S0257-8972(00)01125-7

22.

, Kisi

, Gray

EMA.

Modelling dislocation-induced anisotropic line broadening in Rietveld refinements using a Voigt function. II. Application to neutron powder diffraction data. J Appl Crystallogr. 1998;31(3):363–368. doi: 10.1107/S0021889897012181