3.1. Characterization of monomer and polymer
Figure 3 shows the FT-IR spectra of the synthesized compounds (Sb and PSb). In the monomer spectrum (Sb), the broad and middle peak at 3327 cm−1 formed as a result of the overlap of the stress vibrations of -OH groups in the structure of boric acid and naphthol. The weak shoulder peaks forming in the range 2968-2921 cm−1 near the -OH peak indicated the C-H stretching vibrations of the aromatic rings of the monomer. The presence of the characteristic sharp peaks at 1611 cm−1 and 1579 cm−1 was identified as the symmetric vibrations of the schiff base group C=N and aromatic C=C. The O–H bending and C–O stretching vibrations were found around 1379 cm−1 and 1207 cm−1, respectively .
The spectrum of polymer showed a very wide band in the range of 3700 cm−1 to 1900 cm−1 as a result of -OH and aromatic C-H stretching vibrations. This is because there are too many repeating units in the structure of the polymer and the groups belonging to these units vibrate at the same time. The increased number of groups in the polymer structure caused the peaks to overlap and their number to decrease. Broad peaks formed in other groups in the structure. The peak observed between 1500 cm−1 and 1700 cm−1 formed as a result of overlapping of C=N and C=C stretching vibrations .
The single peak at 7.94 ppm proved the presence of the proton of the imine group (-CH = N–). Singlet and doublet peaks occurring in the range of 7.81-7.14 ppm were caused by naphthol and furan ring protons (Figure 4). The naphthol proton peak formed very weakly in the spectrum of the polymer. This is because -OH group contributes in polymerization more than the benzene ring (Figure 5). In the spectra of PSb, the two singlet peaks at 10.47 and 10.06 ppm belonged to the -OH proton of boric acid. The imine proton of PSb was observed in the range of 8.12-7.84 ppm as multiplet. In addition, the multiple, doublet and singlet peaks in the range 7.53-5.86 ppm belonged to aromatic protons .
Gel permeation chromatography (GPC) was used to determine the molecular weight (Mn and Mw) and polydispersity index (PDI) of the polymer (PSb). The values obtained for PSb were Mn:4.938, Mw: 5.864, and PDI:1.18. According to the obtained results, PSb formed by bonding of about 20 monomers.
3.2. FT-IR spectroscopy
The FT-IR spectroscopy was used to determine the functional groups of the newly synthesized HAp samples with metal at different rates. The spectra of the examples are given in figure 6. In the samples synthesized using different rates of B, Cu and Zn, after calcination, peaks belonging to the PO4ˉ3, OHˉ and CO3-2 functional groups as well as the peaks of the oxide bonds of the added metals were observed. The most distinctive difference observed in the spectra of the HAp samples is that the intensity of the peaks and their positions changed in proportion to the change in metal rates in the samples. One of the common features of FT-IR spectra of HAp samples was that the vibration peak of the CO3-2 (ν3) group, which should be observed around 1450-1600 cm−1, did not form in any spectrum. But peaks around 2000 cm−1 indicated that there was CO3-2 absorbed in the HAp structure. In addition, very weak peaks around 870 cm−1 supported the presence of CO3-2 (ν2). This is because the presence of different metals in the structure changed the position of the CO3-2 group in the crystalline structure . Two types of carbonate substitution are possible in hydroxyapatite. The first is the direct substitution of OHˉ by CO3-2 (A-type substitution CO3-2 ↔ 2 OHˉ) and the second is charge compensation, PO43- substituting tetrahedral group with CO3-2 (B-type substitution). The proof of the formation of both A and B type carbonate substitutions in the synthesized novel metal- doped hydroxyapatites is seen in their spectra .
The wide and weak peaks observed at 3422 cm−1 in the HAp-B1 spectrum formed as a result of the stretching vibrations of the water absorbed by the sample. The peaks forming at 1985 and 1934 cm−1 indicated the presence of CO3-2 absorbed in the structure. Vibration peaks of the apatite phase of sample were seen in two bands. The first peak formed quite broadly by producing peaks of 979 cm−1 (ν1(PO4ˉ3)) and 1007-1118 cm−1 (ν3(PO4ˉ3)). The peaks of the second band occurred at 543 cm−1, 58 cm−1 6, 603 cm−1 (ν4(PO4ˉ3)) and 460 cm−1 (ν2(PO4ˉ3)). Also, peaks of B, Cu and Zn-oxygen bond formed at 433 cm−1, 396 cm−1, and 356 cm−1.
When the spectra of HAp-B2 and HAp-B1 were compared, it was observed that the peaks of the phosphate group formed more sharply. In addition, the locations of the stretching vibration peaks of both absorbed water and carbonate shifted and formed at 3007 cm−1 and 2005 cm−1. The addition of metal to the hydroxy apatite structure changed the location and density of the peaks. The peaks of ν3(PO4ˉ3), ν1(PO4ˉ3), ν4(PO4ˉ3) and ν2(PO4ˉ3) stretching vibrations were (1123 cm−1 and 1027 cm−1), 970 cm−1, (602 cm−1 and 552 cm−1) and 459, respectively. A weak peak of the carbonate group was observed at 880 cm−1. Metal-oxygen bond peaks formed at 378 cm−1, 366 cm−1 and 359 cm−1.
PO4ˉ3 peaks in the HAp-B3 spectrum formed more wider compared to the spectra of other samples. While peaks at (1123 cm−1, 1011 cm−1) and 970 cm−1 formed as a result of ν3(PO4ˉ3) and ν1(PO4ˉ3) stretching vibrations, peaks at (606 cm−1, 587 cm−1, 550 cm−1) and 459 cm−1 formed as a result of ν4 (PO4ˉ3) and ν2(PO4ˉ3) stretching vibrations. The absorbed water and carbonate peaks occurred at 2966 cm−1 and 2004 cm−1 in the spectrum. The other carbonate (ν2(CO3-2)) peak appeared in a very weak form at 864 cm−1. However, the peaks formed at 359 cm−1, 366 cm−1 and 355 cm−1 as a result of metal-oxygen bond vibration. Spectra of HAp-B4 and HAp-B5 formed peaks with similar intensities. Especially PO4ˉ3 stretching vibration peaks were very similar to each other. The peaks of ν3(PO4ˉ3) and ν1(PO4ˉ3) formed in the range of 1123-968 cm−1 for HAp-B4 and 1124-968 cm−1 for HAp-B5. The ν4(PO4ˉ3) stretching vibration peaks formed at 591 and 551 cm−1 in both samples, respectively, while the peak of ν2(PO4ˉ3)vibration did not occur. Another peak not seen in the spectra belonged to ν2 (CO3-2). The peaks of absorbed water and carbonate formed at 3075 cm−1 for HAp-B4, 2934 cm−1 for HAp-B5 and 1995 cm−1 for HAp-B4, and 1985 cm−1 for HAp-B5, respectively. Metal-oxygen bond vibration peaks formed in the range of 387-359 cm−1.
3.3. X-ray diffraction
Figure 7- Figure 9 are XRD patterns of the synthesized powders. According to the XRD results, it was observed that all the XRD models had characteristic peaks that were consistent with International Center for Diffraction Data (ICDD) files that are valid for calcium phosphates. As a result of the XRD analysis, 2θ angle values were seen that are characteristic for HAp, β-TCP, B4C, B6O, CuO and ZnO in the structure of HAp that is produced. As seen in the XRD patterns, main phase of the samples was β-TCP. Changes were observed in the 2θ values and intensities according to the addition rates of copper and zinc. The intensity and model of relevant peaks are expected to be changed according to the relative composition rate of HA/β-TCP and the metal addition amount.
XRD pattern of HAp/B1 showed peaks that appeared quite crystalline in the 10-90 degree range, proving that the structure formed in HAp and β-TCP. XRD pattern of HAp/B1 (HA/β-TCP) shows main peaks corresponding to HAp (JCPDS no. 09-0432) and β-TCP (JCPDS no. 09-169) in accordance to ICDD standard. The major diffraction peaks identified for HAp were in agreement with the standard JCPDS as a hydroxyapatite. Major sharp and intense peaks were observed at 2θ values of 25.8°, 31.9° and 32.5°. For β-TCP, the main peaks were indexed at 25.8°, 26.7°, 29.8°, 31.1°, 32.6°, and 34.5°.
The reason for the HAp/B1 powder to peak up to 90 degrees in the XRD model was the use of boron containing polymer in the production of hydroxyapatite. The presence of boron in the composition of the HAp/B1 powder changed the peak intensities of HAp and β-TCP. In XRD patterns, the peaks observed at 2θ = 14.2°, 23.4°, 33.6°, 37.8°, and 41.9° and all observed peaks between 70° and 90° were characteristic peaks of B6O (JCPDS file No: 01-080-2254). In addition, the peaks forming at 2θ = 23.4°, 35.6°, 61.3°, 62°, and 66.9° prove the existence of B4C (JCPDS file No: 00-035-0798).
In the XRD pattern of HAp/B2, 54 peaks proving that the main structure consisted of HAp and β-TCP formed in the range of 2θ=10°-60°. Furthermore, new peaks formed indicating the presence of metal oxides in this range, while the number of peaks beyond 60° proved that the presence of copper, zinc, and boron oxides decreased. The peak having the highest intensity and proving the presence of HAp/β-TCP occurred at 2θ = 31.2° and 27.9° adjacent to the nearby peaks. The peak observed at 35.2° position with (111) indicated that the HAp/B2 contained copper oxide as crystalline, which was in a good agreement with JCPDS card number 45-0937. In addition, the peaks observed at 56.9°, 59.8° 62.8 and 66.3° belonged to copper oxide. Also, the peaks of XRD pattern at 2θ = 31.7°, 47.5°, 56.9°, 62.8°, 68.8°, and 76.7° proved the presence of ZnO (JCPDS file No: 36-1451). The peaks of B6O were observed at 2θ = 14.3°, 23.4°, 33.6°, 37.8°, 41.9°, 54.7°, 56.6°, 65.6° . Also, the peaks of B4C observed at 2θ = 22°, 35.7°.
The XRD result of HAp/B3 showed that the crystal structure of HAp/B3 powder was slightly different from both HAp/B1 and HAp/B2. The change in the rates of copper and zinc in the powder structure caused the formation of a more crystalline structure. The range of peaks was 10-100, unlike the other two powder samples. Especially, the intensity of the peaks after 60 degrees increased. The top five peaks with the highest density (2θ = 27.9°, 31.2°, 31.5°, 34.5°, and 35.3°) clearly showed the presence of HAp and β-TCP. The peaks forming at 2θ = 14.2°, 33.6°, 41.9°, 54.7°, 56.4°, 65.6° belonged to B6O and the peaks forming at 2θ = 22.1°, 35.8°, and 37.5° belonged to B4C. Further, peaks of copper and zinc oxides formed at 2θ = 32.6°, 36.7°, 38.8°, 47.0°, 48.6 °, 53.2°, 56.9°, 57.6°, and 61.0°.
Based on the XRD result of HAp/B4 and HAp/B5, the peak intensities decreased and increased due to the change in metal content. Especially, the intensity of HAp/B5 peaks decreased. Peaks proving the presence of HAp, β-TCP, and metal oxides (B, Cu, Zn) formed in the XRD data of both compounds.
The average particle size of HAp/B1-5 powders was calculated using the "Scherrer" equation (Eq.1)
D = (kλ)/( 𝛽D cosθ) (1)
D is Crystal size (nm), λ is the wavelength of CuKα1 radiation (1.5406 Å), β is the full width at half maximum for the diffraction peak under consideration, θ is half value of the diffraction angle of the most severe peak (2θ/2), and k is dimensionless shape factor. The particle size was calculated for highest intensity peak of HAp/B1, HAp/B2, HAp/B3, HAp/B4 and HAp/B5, and the average of particle sizes was found to be 88 nm, 69 nm, 95 nm, 84 nm, and 70 nm, respectively.
On the other hand, the fraction of the crystalline phase of hydroxyapatite in the samples (Xc) was calculated using the (Eq.2)
Xc = 100x((Ib-Va/b)/Ib)(2)
where Xc is the fraction of crystalline phase, Ib is the highest intensity of diffraction peak, and Va/b is the intensity of the trough between the lowest and highest diffraction peaks. The specific surface area was calculated with the formula(Eq.3)
Ssp = 6 x 103 / d x ρth (3)
where d is the average particle diameter and the theoretical density of the hydroxyapatite is ρth (HAP) = 3.16 g/cm3 for spherical particles. Table 1 shows the calculated fraction of the crystalline phase and specific surface area values. The crystal phase fraction of metal-doped hydroxyapatite powders is quite high as expected. The highest crystal phase fraction belonged to HAp/B2 (99%) and the lowest crystal phase fraction belonged to HAp/B5 (93%). The specific area of metal-doped powders was between 19.98 and 27.51 m2/g and different specific area values were obtained from each other due to the varying metal rates in the powders.
Scanning electron microscopy images were used to show the surface morphology of the metal-doped HAp powders obtained and EDX spectra were taken for quantitative element analysis of the imaged regions. SEM images revealed that all powder samples had a different agglomerated spherical and granular morphology, due to the difference in metal content in the structure even when calcined at the same temperature. Differentiation of the morphology of the powders increased especially with the addition of Cu and Zn elements to their structures.
Macrostructure and microstructure SEM images obtained at different magnifications showed that HAp/B1 and HAp/B2 were quite different from HAp/B3, HAp/B4 and HAp/B5 in terms of roughness, pore size, geometry, and total porosity. Powder samples were compact, also had both micro and macroporosity and were not interconnected.
If the powder samples were compared in terms of porosity, it was observed that the best porous structure formed in HAp/B1 and HAp/B2 (Figure 10, 11). The presence of pores of different sizes in the morphology of these samples indicated that they can initiate an inflammatory process similar to the early phase of fracture healing when applied to the tissue, as in almost all artificial bone tissue ceramics of mineral origin, depending on their porosity (17).
The similarity in the morphology of HAp/B3, HAp/B4 and HAp/B5 powders is that they created smooth structures with little pore structure (Figure 12, 13, 14). A layered structure formed on the surface of all three powder samples. In addition, there were particles independent of the matrix structure and differently crystallized depression in their surfaces. The main reason for this formation was formation of metal-oxygen bond in the hydroxy apatite structure, which is degraded by the effect of high temperature, regardless of the difference in metal rates. This bond indicated the presence of both metal oxides and metal-phosphate bonds.
Although the data obtained from the EDX analysis showed that the expected elements (Ca, P, O, C, B, Cu, Zn, K) were present in the prepared powder samples (Figure 15, 16). The data revealed that the Ca:P ratio was 1.86, 2, 1.90, 1.71 and 1.87 for HAp/B1, HAp/B2, HAp/B3, HAp/B4 and HAp/B5, respectively, which was not very close to the ideal value of 1.67 normally associated with HAP. When HAp is exposed to high temperatures (900-1200 οC), there is degradation at the Ca/P mol ratio in its structure. In particular, adding different metal ions to the hydroxyapatite structure causes this ratio to be different than expected. The deviations of HA from the Ca/P ratio of 1.67 may cause the formation of other calcium phosphate phases (alpha, beta-tricalcium phosphate, tetra calcium phosphate) during these processes.
In addition, Apatite has a chemical structure that allows it to be replaced with other ions. Due to this feature, the displacements in Ca+2, PO4-3 or OHˉ groups in its structure cause changes on the properties of its substance. EDX analysis results showed that the addition of B, Cu and Zn ions at different rates into the structures of newly prepared hydroxyapatite powders caused a different morphological structure than expected. This difference in powder structure is an expected situation and show parallelism with the main purpose of this study.
Figure 17 shows the UV-vis spectra of the novel HAp powders. Due to similar element contents, the samples formed absorption bands in same regions. The absorption bands were formed by the interaction of calcium phosphate and metal oxides with photons. When metals interact with photons, they can form different electron transitions. These transitions are usually ligand-metal charge transfer (LMCT) and d-d transitions. These transitions can include the metal ion itself or its ligand or ions binding to the metal ion. All synthesized HAp powders did not show absorption in the spectral range of 400-700 nm. This is because the samples had a d-d transition instead of the LMCT transition. While HAp/B1, HAp/B2, HAp/B5 formed bands at 330 nm HAp/B3 and HAp/B4 formed bands at 331 and 332 nm. In addition, two more peaks were observed at 304 and 301 nm for HAp/B1, at 305 and 273 nm for HAp/B2 and at 305 and 296 nm for HAp/B5. These peaks seen in the spectrum formed as a result of charge transfer between oxygen and metal ions.
The thermal properties of the prepared metal doped HAp powders were determined by thermogravimetric analysis (TGA) and differential thermal analysis (DTA) method in a nitrogen atmosphere heated at 20 °C/min. in the range of 0 °C-1150 °C . Figure 18 and figure 19 show the TGA and DTA thermograms of prepared powders. From the TGA result of the powders, it can be seen that there was no phase transformation as a result of heating, thus indicated that the novel metal-doped hydroxyapatites had thermal stability at high temperatures.
The observed initial mass loss in the region of 20-200 °C can be due to the dehydroxylation of hydroxyapatite powders. There was no significant loss of mass for all powders during the thermal decomposition process. The main reason for this was B, Cu and Zn compounds in the structure of powders. Metals have high temperature resistance. Especially boron is very stable against temperature changes and today boron is used in the production of many materials that require thermal resistance, especially glass and ceramics. According to TGA data, the total mass loss of HAp/B1-HAp/B5 between 20-1150 °C was 0.446%, %.433%, 2.426%, 3.235%, and 1.184%, respectively. The first mass loss seen in the region of 200-700 °C is due to the removal of lattice bound water . In this temperature range, the highest mass loss occurred at h4, while the lowest mass loss occurred at HAp/B2. Mass loss rates from small to large were 0.087%, 0.133%, 0.294%, 0.907%, and 1.237% for HAp/B2, HAp/B1, HAp/B5, HAp/B3, and HAp/B4, respectively.
According to the mass loss rates between 700-1150 °C, the order from small to large was HAp/B1, HAp/B2, HAp/B5, HAp/B3 and HAp/B4 (0.100%, 0.385%, 0.522%, 0.622%, and 0.975%). The reason for the loss of mass in this region is the conversion of HPO42− to pyrophosphate (P2O74−) in HAp powders.