Effect of the Fluorine Substitution for –OH Group on the Luminescence Property of Eu³⁺ Doped Hydroxyapatite

Abstract

In this study, different fluoridated hydroxyapatite doped with Eu³⁺ ion nanoparticles were prepared by the hydrothermal method. The relationship between luminescence enhancement of Eu³⁺ ions and a fluorine substitution ratio for hydroxyl group in hydroxyapatite was discussed. Moreover, the effect of fluorine substitution for a hydroxyl group on phase composition, crystallinity, and crystal size was studied. Phase composition and chemical structures were identified by X-ray diffraction (XRD) and Fourier Transform Infrared (FT-IR) Spectroscopy analyses. Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM) patterns were performed to analyze the morphology and particle size. X-ray Photoelectron Spectroscopy (XPS) patterns were observed to analyze fluorine substitution for the hydroxyl group and chemical state of Eu³⁺ ions in fluoridated hydroxyapatite. The results of these experiments indicated that the samples with a different fluorine substitution ratio were prepared successfully by maintaining the apatite structure. With an increasing fluorine substitution ratio, the morphology maintained a rod-like structure but the aspect ratio tended to decrease. XPS patterns displayed that the fluorine replaced the hydroxyl group and brought environmental variation. The fluorine ions could affect the crystal field environment and promote luminescence conversion. There was a linear relationship between the fluorine substitution ratio and luminescence enhancement.

1. Introduction

Hydroxyapatite (Ca₁₀(PO₄)₆(OH)₂), which is also known as HAp, is the main inorganic ingredient of hard tissues similar to its mineral constituents [1,2,3], which have outstanding biocompatibility and bioactivity [4,5]. Meanwhile, HAp crystals are good host material for doping of rare earth (RE) ions, and the doped HAp such as Eu:HAp [3], Eu/Tb:HAp [6], and Eu/Dy:HAp [1] display unique luminescence properties under excitation of a specific wavelength. Moreover, the RE ions doped HAp nanoparticles have better features as luminescence imaging agents including low toxicity, photostability, thermal and chemical stability, and real-time imaging [4,7,8,9]. Their luminescence imaging ability is affected by a degree of crystallinity, size distribution, and dispersion of nanoparticles as well as crystal field contributions [1,5,10].

Recently, more attention has been put by the researchers toward the luminescence performance of RE ions in natural as well as synthesized apatites. The main focus for the study of RE ions doped with HAp has been the enhancement of luminescence with previous results showing the possibility of improving luminescence performance of RE elements doped HAp by increasing the RE content [11], calcination [3], and co-doping of RE ions [12,13,14]. However, they are accompanied by certain issues like quenching effect, oversize, severe aggregation, and toxicity. Fluoride ions (F⁻) are usually used for replacing the hydroxyl (–OH) group of HAp to improve some physicochemical properties such as solubility, crystallinity, and resistance to acid corrosion [15]. Moreover, the Eu³⁺ ions in fluorapatite (Ca₁₀(PO₄)₆F₂, FAp) emits stronger luminescence than that in HAp, which can be attributed to the effect of F^- ions on the crystal field environment. The –OH group in the crystal lattice has a quenching effect on luminescence of rare earth ions. The F^- ions can generate lower vibration energy and, thereby, promote luminescence conversion [16]. However, the relationship between the fluorine substitution ratio for the –OH group and luminescence of Eu³⁺ ions has not been revealed in detail.

Herein, we used a hydrothermal synthetic route to prepare fluoridated hydroxyapatite doped with Eu³⁺ ions ((Ca, Eu)₅(PO₄)₃(OH)_1−xF_x), Eu:F_xH_1−xAp,). The change of luminescence was observed with varying fluorine substitution ratio (from 0 to 1) for the –OH group, which aims to reveal the relationship between the fluorine substitution ratio and luminescence intensity. Moreover, the effect of the fluorine substitution ratio for the –OH group on phase composition, crystallinity, crystal size, and chemical state of Eu³⁺ was studied by X-ray diffraction (XRD), Fourier Transform Infrared (FT-IR) Spectroscopy, Scanning Electron Microscopy (SEM), High Resolution Transmission Electron Microscope (HRTEM), Energy Dispersive Spectrometer (EDS), and X-ray Photoelectron Spectroscopy (XPS).

2. Materials and Methods

2.1. Preparation of Eu:F_xH_1−xAp Nanoparticles

The nanocrystalline Eu:F_xH_1−xAp materials were obtained using the following hydrothermal method. First, CaCl₂·2H₂O, Na₂HPO₄·12H₂O, Eu(NO₃)₃·6H₂O, and NaF were dispersed individually in deionized water. Na₂HPO₄ and NaF powders were dissolved in deionized water to get solution A (PO₄³⁻: 0.24 mol/L, F⁻: 0.08x mol/L). CaCl₂·2H₂O and Eu(NO₃)₃·6H₂O powders were dissolved in deionized water to get solution B (Ca²⁺: 0.393 mol/L, Eu³⁺: 0.008 mol/L). Second, 15 ml of solution A was poured into 15 ml of solution B under vigorous stirring to get the suspension. The pH value of the suspension was adjusted to about 9.0~9.5 by NH₃ solution. Subsequently, the suspension was moved to a sealed stainless-steel autoclave, and maintained at a stirring condition at 120 °C for 60 min. After being naturally cooled at room temperature, the resulting milky matter was separated by centrifugation at 2000 rpm for 5 min, and washed three times using deionized water. Lastly, it was freeze dried for 20 hours to get the powder sample. The molar ratios of (Ca + Eu)/P and Eu/ (Ca + Eu) were 1.67 and 0.02, respectively. The amount of F^- was adjusted to obtain a different fluorine substitution ratio for the –OH group depending on the x value in Eu:F_xH_1−xAp (x = 0, 0.1, 0.3, 0.5, 0.7, 0.8, and 1).

2.2. Characterization of Eu:F_xH_1-xAp Nanoparticles

The phase composition, crystalline structure, and the crystallinity of the Eu:F_xH_1−xAp were explored by XRD (D/Max-ⅢA, Rigaku Co., Tokyo, Japan). The 2θ range of the XRD pattern was collected from 10° to 70° with a scanning speed of 0.02°/s.

The degree of crystallinity Xc for Eu:F_xH_1−xAp was calculated by the following equation.

$${\mathrm{X}}_{\mathrm{c}}=\left[1-\frac{{\mathrm{v}}_{112/300}}{{\mathrm{I}}_{300}}\right]$$

(1)

where ${\mathrm{v}}_{112/300}$ is the hollow intensity between the planes (112) and (300) reflection, and ${\mathrm{I}}_{300}$is the intensity of the plane (300) diffraction peak [1,17]. Moreover, crystallite sizes $\mathrm{D}$ of Eu: HAp were concluded by using Scherer’s Equation.

$$\mathrm{D}=\frac{\mathrm{K}·\mathsf{\lambda }}{{\mathrm{FWHM}}_{002}·\cos \mathsf{\theta }}$$

(2)

where the shape factor K is 0.9, wavelength$\lambda \mathrm{is} 0.15418$, ${\mathrm{FWHM}}_{002}$ express the full width at half maximum for reflection of the plane (002), and θ stands for the angle of a certain signal [1,3,18]. FT-IR (Nexus, Thermo Nicolet, USA) was exploited to analyze the material’s vibrational modes of the characteristic band. The spectra were recorded between 4000 cm⁻¹ to 400 cm⁻¹. The Eu:F_xH_1−xAp powders (7.5 mg) were dispersed in alcohol (5 ml) and then the suspension was acquired by ultrasonic cleaning. The morphology and the average particle size of Eu:F_xH_1−xAp were evaluated by SEM (JSM-5610LV, JEOL, Japan) and HRTEM (JEM-2100 F, JEOL, Japan). XPS (ESCALAB 250Xi/ESCALAB 250Xi) was used to identify the composition and chemical states of the elements. The luminescence spectra of Eu:F_xH_1−xAp was recorded using a 970CRT Luminescence Spectrophotometer (Shanghai Sanco, China). An emission spectrum (or excitation spectrum) was recorded by fastening the excitation wavelength (or emission wavelength) while scanning the determined wavelength region of the spectrofluorimeter [19].

The luminescence enhancement relative ratio (R) was calculated by the following equation.

$$\mathrm{R}=\frac{\mathrm{B}-\mathrm{A}}{\mathrm{A}} ·100\%$$

(3)

where A is the emission peak’s intensity of Eu:HAp and B is the emission peak’s intensity of Eu:F_xH_1−xAp.

The luminescence lifetime and quantum yield were measured by FLS980 Series of Fluorescence Spectrometers. The average luminescence lifetime (${\mathsf{\tau }}_{\left(\mathrm{avg}\right)}$) was calculated by the following equation.

$${\mathsf{\tau }}_{\left(\mathrm{avg}\right)}=\frac{{\mathrm{A}}_1{\mathsf{\tau }}_1^2+{\mathrm{A}}_2{\mathsf{\tau }}_2^2}{{\mathrm{A}}_1{\mathsf{\tau }}_1+{\mathrm{A}}_2{\mathsf{\tau }}_2}$$

(4)

where ${\mathrm{A}}_1$ and ${\mathrm{A}}_2$ are the exponential factors. ${\mathsf{\tau }}_1$ and ${\mathsf{\tau }}_2$ are luminescence lifetime values of Eu: F_xH_1-−xAp.

3. Results

3.1. Phase Composition and Chemical Structure

Figure 1A shows the phase composition. Crystallite crystal and the degree of crystallinity of samples have different fluorine substitution ratios. By comparing the XRD patterns, it was observed that no additional diffraction peaks appeared in the case of Eu:F_xH_1−xAp nanoparticles except for characteristic diffraction peaks of FAp (PDF # 75-0915) and HAp (PDF # 74-0566), which indicates no impurity phase. Moreover, with the gradual increase of a fluorine substitution ratio, the diffraction angle and relative intensity varied slightly. Considering that the F⁻ ions (1.28 Å) are smaller than the –OH group (1.37 Å), with an increasing fluorine substitution ratio, there would be natural shrinkage of the lattice leading to a slight shift of the diffraction signals to higher degrees. The calculated crystallite sizes assigned to (002) diffraction peaks with a different fluorine substitution ratio, respectively, also indicated that the crystallite sizes tended to reduce (

Table A1. Structural parameters and degree of crystallinity of the Eu:F_xH_1−xAp.

Samples (Eu:FxH1−xAp, x)	D(002) (nm)	V × 103 (nm3)	Xc (%)
0	39.4	529.0	38.73%
0.1	39.8	528.6	35.46%
0.3	40.6	528.0	28.87%
0.5	35.7	527.9	20.87%
0.7	36.5	523.7	21.49%
0.8	32.8	522.2	43.11%
1	33.7	522.5	55.56%

). Moreover, the crystallinity also decreased with the increase of the fluorine substitution ratio from 0 to 0.5. When the fluorine substitution ratio was above 0.5, the crystallinity gradually grew with an increasing substitution content (Table A1). In addition, the crystallinity of Eu:FAp was as high as 55.56%, which was higher than the crystallinity of Eu:HAp (38.73%).

Figure 1B shows FT-IR spectra of Eu:F_xH_1−xAp samples. In the FT-IR spectrum of Eu:HAp samples without F^- ions substitution, the vibrational peaks of PO₄^3- group in the apatite structure was distinguished, including ν₁ (at 962 cm⁻¹), ν₂ (at 472 cm⁻¹), ν₃ (at 1034 cm⁻¹ and 1107 cm⁻¹), and ν₄ (at 564 cm⁻¹ and 602 cm⁻¹) [20,21]. Hydroxyapatite is a kind of apatite crystal whose general molecular formula is Me₁₀(XO₄)₆Y₂. In this study, the Me is a divalent cation, XO₄ is the negative trivalent group, and Y is a negative monovalent ion [22]. For RE-doped apatite with a different fluorine substitution ratio, RE ions will occupy the position of Me and Y will be replaced by F⁻ ions. Accordingly, the position and the intensity of the vibrational peak of the –OH group in the spectrum changes with the fluorine substitution ratio. When the F^- ions enter into the lattice of HAp, the original H-O-H is gradually converted into the F-O-H structure, which is attributed to the hydrogen bond formed by the F^- ions (Figure 1C). This will generate a new chemical bond accompanied by a new vibration mode (F-O-H) and causes the decrease of the H-O-H vibration [23]. Due to higher electronegativity of F^- ions compared to O^2- ions, a new vibration (F-F-OH-F) will also be detected at about 746 cm⁻¹ (Figure 1D). Results showed that the absorption peaks of the –OH group was decreased at 3569 cm⁻¹ and 631 cm⁻¹ with an increase of the fluorine substitution ratio. The absorption peaks at 3540 cm⁻¹ (F-O-H) and 746 cm⁻¹ (F-F-OH-F) appeared and the intensity was enhanced. This indicated that the F^- ions replaced the –OH group and entered into the apatite crystal lattice in the case of Eu:F_xH_1−xAp samples. The hygroscopic property is related to the surface chemical property [24]. For example, the –OH group makes it easy to form hydrogen bonds with water molecules, which make the water molecules absorb on the surface of materials. This is conducive toward improving the hygroscopic property [24,25]. The chemical property of the apatite surface was changed when the –OH group was replaced by F⁻ ions. The –OH group on the surface of HAp easily forms hydrogen bonds with water and absorbs on the surface of HAp. Therefore, HAp has a certain hygroscopic property. After replacing the –OH group, the hygroscopic property exhibited by hydrogen bonding is weakened. Therefore, the hygroscopic property is reduced after F⁻ ions replace the –OH group.

3.2. Morphology Observation

Figure A1 shows the SEM images of as-dried Eu:F_xH_1−xAp powders with a different fluorine substitution ratio. The morphology was mainly rod-like, and the length of the rod-like particles tended to decrease with the increase of the fluorine substitution ratio. The EDS spectrum further showed the signals of Ca, Eu, O, P, and F in Eu:F_xH_1−xAp (x = 1) (Figure A2). The calculated molar ratios of Eu/(Ca + Eu), (Ca + Eu)/P, and F/(Ca + Eu) were 0.02 ± 0.01, 1.69 ± 0.05, and 0.24 ± 0.04, respectively, which were consistent with the setting values (

Table A2. Calculated elemental molar ratios of Eu:FAp.

Molar Ratio	Calculated Value	Observed Value
Eu/(Ca + Eu)	0.02	0.02 ± 0.01
(Ca + Eu)/P	1.67	1.69 ± 0.05
F/(Ca + Eu)	0.2	0.24 ± 0.04

).

TEM observations showed that samples consisted of short rod-like nanoparticles (Figure 2). The average length and diameter of the nanoparticles were obtained by collecting data of more than 100 particles (

Table A3. Particle sizes of the Eu:F_xH_1−xAp with different fluorine substitution ratio.

Samples (Eu:FxH1−xAp, x)	Length (nm)	Diameter (nm)	Aspect Ratio (%)
0	77.70 ± 2.88	15.43 ± 0.26	5.04
0.3	54.37 ± 1.78	14.09 ± 0.30	3.86
0.5	47.19 ± 2.55	14.31 ± 0.15	3.30
1	43.36 ± 2.02	14.07 ± 0.49	3.08

). With the increase of the fluorine substitution ratio from 0 to 1, the length of the particles was decreased significantly from 77.7 ± 2.9 nm to 43.4 ± 2.0 nm. However, the diameter of the particles varied little (between 15.4 nm and 14.1 nm). This led to the decrease of the aspect ratio of rod-like particles from about 5 to about 3. Furthermore, the lattice planes of (002), (300) were observed in the HRTEM images (Figure 2), which were assigned to an apatite crystal structure. With the increase of the fluorine substitution ratio, the interfacial spacing of (002) decreased slightly, which was consistent with the change of length of rod-like particles. In addition, the capillary phenomenon between particles will make the water molecules absorbed on the particles [26]. However, the capillary and mesoporous structures were not observed in the microstructure of the Eu:F_xH_1−xAp synthesized by the hydrothermal method. Therefore, the hygroscopic property was not improved with fluorine substitution.

3.3. XPS

Figure 3A shows the XPS spectra of Eu:F_xH_1−xAp (x = 0, 0.3, 0.5, 0.8, and 1) samples. For the samples, the characteristic peaks of Ca 2p, P 2p, O 1s, and Eu 3d were all detected. Moreover, with the increase of the fluorine substitution ratio, the characteristic peak of F 1s was enhanced gradually, which indicates that the F⁻ ions entered into the apatite lattice (Figure 3B) [27]. When the F^- ions entered into the apatite lattice, the binding energy of the atom varied due to the change in chemical environments, corresponding to a shift of the peak position in the XPS spectrum. For example, the binding energy of Eu³⁺ increased from 1134.73 eV to 1135.33 eV while the x increased from 0 to 1. This indicated that the surrounding environment for Eu³⁺ ions was changed due to the fluorine substitution for the –OH group.

3.4. Luminescence Spectrum

The luminescence property of Eu:F_xH_1−xAp with a different fluorine substitution ratio was detected. Figure 4 shows the excitation spectra with the emission at 617 nm. Results showed the strongest peak at 394 nm and some smaller peaks at 364 nm, 381 nm, and 264 nm, respectively. The smaller peak at 264 nm derived from the Eu-O bonds belong to a process of energy transfer. The peaks at 394 nm, 364 nm, and 381 nm were due to the electric dipole forbidden intra-configurational f-f transitions [28]. Furthermore, with the increase of the fluorine substitution ratio, the intensity of excitation peaks was enhanced gradually.

With the excitation at 394 nm (Figure 5A), the emission spectra display the strongest peak at 617 nm (⁵D₀→⁷F₂) and three smaller peaks at 594 nm (⁵D₀→⁷F₁), 654 nm (⁵D₀→⁷F₃), and 699 nm (⁵D₀→⁷F₄). The luminescence signal at 617 nm was stronger than that at 590 nm. In the hydroxyapatite structure, two Ca sites can be occupied by Eu³⁺ ions. One is Ca (I) with C₃ symmetry surrounded by 9 oxygens. The other is Ca (II) with Cs symmetry surrounded by six oxygens and one –OH group [9,29]. Eu³⁺ ions located in Ca (I) site show the main emission at about 618 nm, and the main emission appears at about 573–578 nm for Eu³⁺ ions in Ca (II) site [29]. Therefore, in our study, the Eu³⁺ ions predominantly occupied the Ca (I) site (C₃ symmetry) in Eu:F_xH_1−xAp [30]. The luminescence intensity enhanced with the increase of the fluorine substitution ratio and the peak at 617 nm increased significantly. The luminescence enhancement relative ratio was further calculated using the emission peak at 617 nm (Figure 5B). Results displayed that the luminescence enhancement showed a near linearly increasing relationship with the increase of a fluorine substitution ratio. Under excitation at 264 nm and 394 nm, the emission peaks of Eu:F_xH_1−xAp nanoparticles were observed at 594 nm, 617 nm, 654 nm, and 699 nm ranging from 550 nm to 750 nm, where the transitions of Eu³⁺ ions were ⁵D₀→⁷F₁, ⁵D₀→⁷F₁, ⁵D₀→⁷F₃ and ⁵D₀→⁷F₄, respectively. The luminous mechanism may be explained by the energy level diagram (Figure 6). With the increase of a fluorine substitution ratio, the emission peak position of Eu:F_xH_1−xAp in luminescence spectra was not moved and the peak intensity was gradually enhanced. Especially, the relationship between the luminescence enhancement and fluorine substitution ratio displayed a nearly linear relationship. Accordingly, it can be expected that the fluorine content in Eu:F_xH_1−xAp can be estimated by determining the luminescence of Eu³⁺ ions.

The emissions of Eu³⁺ excited at 394 nm could be attributed to the two transition modes including electric dipole transition (ED transition) and a magnetic dipole transition (MD transition) [19,31]. The majority of them belonged to ED transition, and some transitions such as the ⁵D₀→⁷F₁ transition belonged to the MD transition. The ⁵D₀→⁷F₂ transition was greatly affected by the surrounding symmetry of the Eu³⁺ ions and properties of the host materials, which was much stronger than the other ED transition. This is also called “hypersensitive transition” [19,32,33]. The ⁵D₀→⁷F₁ MD transition is largely independent of the environment of the Eu³⁺ ions and its intensity is often used as a calibration for comparing luminescence spectra [19]. Furthermore, the transition intensity ratio of ⁵D₀→⁷F₂ (I_R) to ⁵D₀→⁷F₁ (I₀) was calculated to reveal the change of the environment of Eu³⁺ ions. Results showed that the value of I_R/I₀ was increased with the increase of fluorine substitution ratio, which indicates that the surrounding environment of Eu³⁺ ions became more asymmetric. It could be inferred that the fluorine substitution for the –OH group changed the environment of Eu³⁺ ions and caused the rise of asymmetry. In addition, the fluorine ions have lower vibrational energy than the –OH group. This will be favorable for Eu³⁺ ions’ luminescence transition [5]. Therefore, in the matrix of apatite, the fluorine substitution for the –OH group could decrease the luminescence quenching of Eu³⁺ ions, which is caused by the –OH group and promotes the luminescence conversion.

Figure 7 shows the decay curves of Eu:HAp and Eu:FAp. Under 394-nm excitation, the average luminescence lifetime of Eu:HAp and Eu:FAp were 727 μs and 1016 μs. The absolute quantum yield of Eu:HAp and Eu:FAp were 1.51% and 8.96%, respectively. The results showed that the luminescence lifetime and quantum yield increased when the –OH group was replaced by F^- ions.

With the excitation at 264 nm (Figure 8A), the emission spectra display four peaks similar to those excited at 394 nm (594 nm, 617 nm, 654 nm, and 699 nm). With the increase of the fluorine substitution ratio, the luminescence intensity was gradually enhanced. For example, the luminescence enhancement of peak at 617 nm increased significantly (Figure 8B). In addition, the relative ratio also showed a nearly linear increasing relationship with the increase of the fluorine substitution ratio. The emission peaks were determined by the charge transfer of the Eu-O bond, which is also called the charge transfer band (CTB) (Figure 6) [19,26]. The O^2- ions absorb energy and transfer to Eu³⁺ ions through the Eu-O bond. In our study, the fluorine substitution for the –OH group caused the shrinkage of the crystal lattice. It could be inferred that the distance of Eu-O bond was decreased correspondingly, which was beneficial for the energy transfer of CTB. Accordingly, the fluorine substitution for the –OH group promoted the process of the energy level transition of Eu³⁺ ions as well as the luminescence.

4. Conclusions

The Eu:F_xH_1-xAp nanoparticles were synthesized by the hydrothermal method with a different fluorine substitution ratio. With the increase of the fluorine substitution ratio, the luminescence property of Eu:F_xH_1−xAp could be enhanced. The absorption peak intensity of the hydroxyl group gradually decreased with the increase of the fluorine substitution ratio, which indicates that the hydroxyl group was successfully substituted by fluorine ions. The crystal structure of the Eu:F_xH_1−xAp was still composed of a single crystal phase where the crystallinity degree was better than HAp while x was greater than 0.5. When x was increased to 1, the crystallinity degree of Eu:F_xH_1−xAp was the largest, and the maximum was 55.56%. The morphology of Eu:F_xH_1−xAp was mainly a rod-like structure and the length tended to decrease with the increase of the fluorine substitution ratio. The fluorine affected the symmetry of the structure and, thus, changed the surrounding environment in which Eu³⁺ ions are located. Furthermore, regarding the fluorine substitution for the –OH group, the quenching effect caused by the –OH group was reduced. In addition, the fluorine ions could produce lower vibration energy and promote luminescence conversion. The fluorine could promote luminescence conversion and luminescence of Eu³⁺ ions would gradually improve with the increase of the fluorine substitution for the –OH group. There was almost a linear increase when the fluorine substitution ratio increased.