Materials Characterization 173 (2021) 110911

Contents lists available at ScienceDirect

Materials Characterization
journal homepage: www.elsevier.com/locate/matchar

Correlative experimental and theoretical characterization of transition

metal doped hydroxyapatite nanoparticles fabricated by
hydrothermal method
Anastasia V. Sadetskaya, Natalia P. Bobrysheva, Mikhail G. Osmolowsky,
Olga M. Osmolovskaya, Mikhail A. Voznesenskiy *
St. Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg, 199034, Russia

A R T I C L E I N F O A B S T R A C T

Keywords: Hydroxyapatite nanoparticles (HAp NPs) bare and doped with cobalt, nickel or copper were obtained via hy­
Hydroxiapatite drothermal synthesis method and fully characterized by X-ray diffraction, Raman spectroscopy, Fourier-
Hydrothermal sinthesis transform infrared spectroscopy, transmission electron microscopy techniques. It was shown that doping
Doped nanoparticles
impact the band gap. The density function theory (DFT) calculations were employed to investigate the electrical
Band gap
DFT calculation
properties of the hydroxyapatite with usage of functionals in local density approximation (LDA) and generalized
gradient approximation (GGA). Two approaches to quantum-chemical calculations within DFT were considered.
The densities of states were calculated for as-prepared samples and study of correlation between the calculated
and experimental values of band gap was carried out. Revealed patterns provide a new viewpoint on the choice
of dopants for biocompatible pigments and scaffold materials with semiconductor properties.

1. Introduction biocompatible implants, which induce the tissue growth by electrical

stimulation [9]. It was reported, that doping do not impact biocom­
Hydroxyapatite (Ca10(PO4)6(OH)2, HAp), the main inorganic patibility of HAp (lack of toxicity and efficient operation over the life
component of bone tissue, along with collagen – the main organic cycle) [10,11], and in some cases even enhance it [12]. Nowadays a
component – form a bone matrix. Namely due to ability to form com­ great attention is paid to the production of doped HAp NPs with addi­
posite with collagen and high biodegradation and biocompatibility, HAp tional functionality such as structural (doped with Ag [13], Zn [14]),
has been widely employed in cosmetic industry and bone tissue engi­ magnetic (doped with Fe [15,16]), and electrical (doped with Sr [17],
neering [1–3]. Dopants introduction results in variation of material Mg [18]) properties.
properties, enhancing number of applications [4]. HAp nanoparticles (NPs) for commercial application are synthesized
HAp is a dielectric material with a wide band gap, which value differs by using simple scalable wet chemistry methods: precipitation and hy­
from 3.9 to 5.3 eV according to different studies [5,6]. Not only in the drothermal synthesis [19–23].
HAp case doping with transition metal can lead to decrease of the band Precipitation method allows to obtain HAp NPs with low crystallinity
gap of dielectric materials or wide-gap semiconductors down to 2–3 eV doped with Zn [24], Cu [25], Ag [26], Sr and Mg [27,28]. A common
[7]. That leads to the appearance of optical absorbance in a Vis region peculiarity of these works (except Ag) is that dopants were chosen with
(change in color of the material), and provides new electrical properties. the same oxidation state as calcium. All studies report that doping led to
The change in crystallinity degree and crystallite size provided by cal­ decrease of average particles and crystallite sizes and changing of NPs
cium atoms substitution by dopant atoms could lead to the significant shape (from needle- and rod-like to spherical morphology). NPs length
change of bioresorpsion [8]. Thus, doping became a possibility to con­ varied in range from 15 to 600 nm and width in range from 4 to 100 nm,
trol these key factors for bio-applications. The first one could be used in depending on precipitation conditions.
cosmetic industry in production of biocompatible multifunctional pig­ Hydrothermal method implies continuous holding of previously
ments, while the second can be used in biomedicine as a base for prepared suspension in sealed autoclave under temperature in the range

* Corresponding author.
E-mail address: [email protected] (M.A. Voznesenskiy).

https://doi.org/10.1016/j.matchar.2021.110911
Received 25 June 2020; Received in revised form 10 December 2020; Accepted 13 January 2021
Available online 16 January 2021
1044-5803/© 2021 Elsevier Inc. All rights reserved.
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

120–290 ◦ C for 2–72 h. For HAp NPs synthesis in the average the tem­ calculation data with experimental results. To avoid crystal distortion
perature equals to 200 ◦ C and duration is around 12 h. Such synthetic the series of works was started with dopant oxidation state taken to be
procedure is not energy efficient and hard to scale to an industrial equal to 2+. All samples were characterized with X-ray diffraction
production, but allows to obtain a highly crystallized product. Hydro­ (XRD), Raman spectroscopy, Fourier-transform infrared spectroscopy
thermal method allows to obtain HAp NPs doped with Mg [29], Ti [30], (FTIR), transmission electron microscopy (TEM), including high reso­
Sr [31], Co. [32] All studies also report a decrease of average particles lution technique (HRTEM) and selected area electron diffraction
and crystallite sizes with doping, while NPs shape commonly were (SAED), scanning electron microscopy (SEM), including energy-
needle- or rod-like shaped. NPs length varied in the range from 41 to dispersive X-ray spectroscopy (EDX) with elemental mapping, specific
200 nm and width in the range from 11 to 100 nm depending on syn­ surface area (SSA) measurement techniques, the thermogravimetric
thesis conditions. The aspect ratios were higher than for precipitation analysis (TGA) and differential scanning calorimetry (DSC) analysis.
method. We also apply the original approach [52], using obtained from XRD
The chemical peculiarities of doping process could lead to the for­ data crystal lattice parameters, as input parameters in DFT calculations
mation of not only HAp phases, but also the others calcium phosphates, with further verification via experimentally measured band gap. The
as, for example, indicated in Refs. 27, 33, 34, and. The authors associate data obtained are using to reveal key factors influencing the material
it with the distortion effect of dopant atom on the environment and this optical and thermal properties, at the point of view of prediction a
is also can be the reason of decrease of average particles and crystallite dopant nature to achieve the required properties.
sizes. Moreover, NPs morphology depends on the way doping is per­
formed. In accordance with Refs. [7,35–42] size and shape of semi­ 2. Experimental
conductors can affect the band gap and consequently optical properties.
So it can be resumed that synthetic procedure plays an important role 2.1. Materials and methods
and the dependence of product properties on dopant nature and syn­
thesis condition is not straightforward. All chemicals were of analytical grade and used without further
The band gap value, providing the optical and electrical properties, purification. HAp NPs have been successfully synthesized via the hy­
can be determined experimentally from absorbance spectra, but the drothermal method using Ca(NO3)2∙4H2O (NevaReaktiv, Russia),
number of articles for doped HAp is not numerous [43,44]. Doping H3PO4 (85%, NevaReaktiv, Russia), NH4OH (30%, NevaReaktiv, Russia)
result in displacement of the atoms in the crystal cell, the severity of and salts of dopant atoms (Co(NO3)2∙6H2O (ReaChem, Russia),
which is related with the substitution position of the dopant atoms, and NiCl2∙6H2O (NevaReaktiv, Russia), CuCl2∙2H2O (Vekton, Russia) for
results in variation of electronic structure. doped NPs. Dopant concentration for all dopants was 10 at.%.
In the HAp crystal unit cell there are two distinct Ca sites, marked as Ca(NO3)2∙4H2O, dopant salt and H3PO4 in molar ratio Ca/P = 1.67
Ca1 and Ca2, where dopant substitution may occur [45]. In the case of along with 75 mL of distilled water were added in a beaker (100 mL)
bulk material, the position of substitution could be determined via equipped with pH meter and magnetic stirrer. pH of reaction medium
crystallographic data [46]. For nanoobject the applicability of this was adjusted to 10 by adding NH4OH (1:1). Suspensions were trans­
method is limited to definition of cell parameters. ferred into a Teflon-lined autoclave with stainless steel shell and main­
Last decade density functional theory (DFT) calculations have tained for 2 h at 180 ◦ C. The autoclave was cooled naturally to room
become an essential tool of the material science. However usually there temperature. The obtained NPs were collected by centrifugation (8000
is a notable, around 1.5–2% difference between experimental and ×g, 3 min) using Sigma 2–16P, washed by distilled water three times and
calculated data for lattice parameters, which cause dramatic difference dried in air at 80 ◦ C in a laboratory oven.
in experimental and calculated band gap values, which does not allow to
predict electronic-related properties [6,7]. 2.2. Characterization of as-prepared samples
For example, in the study [45] DFT calculations were done for Zn and
Mg doped HAp NPs. Authors found out, that Ca2 position of substitution Phase characterization of prepared HAp NPs was carried out by XRD
is more favorable. Unfortunately, in this work there is no comparison using Bruker «D2 Phaser» diffractometer with conditions: Cu-Kα radia­
with experimental data, as well as in the study of Si doped HAp [47]. tion λ = 0.15406 nm at 30 kv and 10 mA within the range from 10 to
Same substitution position was determined in the work [48] for HAp and 100◦ .
fluorapatite (FAp) doped with Cd, Zn, and Cd and Zn simultaneously. In The refinement of the lattice parameters of NPs was calculated from
this case 1.5% difference between calculated and experimental results the obtained results by Pawley method using Topas 4.2 software (Bruker
for lattice parameters was reported. In the study of Sr and Mg substi­ AXS, Germany) [46]. The average crystallites sizes were evaluated from
tution in HAp [49] deviation reached 2%, however authors pointed out the XRD results according to Scherrer formula [53].
the correlation between lattice parameters and atomic radii of dopants. FT-IR spectra were recorded on IRAffinity-1 in the range of
In the study of simultaneous two dopant (Zn and Si) incorporation in 4000–400 cm− 1.
HAp structure [50] comparison revealed the greatest among all the cases The molecular vibrations of the different HAp groups were examined
2.4% deviation. In the work Ref. 51 authors calculated DOS (Density of from 200 cm− 1 to 4000 cm− 1 using a Raman spectroscopy (Senterra
States), projected DOS (PDOS), band structure and determined band gap (Bruker) microscope). The spectra were recorded using 25 mW semi-
for Mn, Fe and Co doped HAp. the deviation was up to 1.5%. Finally, in conductor laser (AlGaAs continuous wave laser) with wavelength 488
the work Ref. 33 the HAp and tricalcium phosphate (TCP) mixture nm and 785 nm; the number of repetitions was three.
doped with Sr was studied and authors came to conclusion, that dopants The optical transmittance and reflectance of the obtained samples
occupy Ca1 position. were measured with a UV/Vis spectrometer (UV-2550, Lambda 1050) in
As we can see, there is no comprehensive comparison of experi­ a wavelength range of 200 to 800 nm. The experimental band gap values
mental and calculated data of lattice parameters, band structure is of obtained NPs were determined by Tauc method [54].
almost not studied, authors usually note the approaches and exchange Transmission electron microscopy (TEM, JEOLS JEM 107, Japan),
functionals they use, but the choice is not justified, there are no attempts high-resolution TEM (HR-TEM), selective area electron diffraction
to predict properties and to give recommendation on dopant choice. (SAED) (TM-1000, Hitachi, Japan), scanning electron microscopy (SEM,
In the present work, synthesis of HAp, bare and doped with Co, Cu Zeiss Merlin, Germany) were used to evaluate the morphology of syn­
and Ni atoms was reported. Synthesis procedure allowed us to obtain thesized HAp NPs. The composition of the as-prepared nanoparticles
HAp NPs with similar shape and size. It was done in order to eliminate was determined using an energy-dispersive X-ray spectrometer with
the impact of morphology and create optimal condition to compare elemental mapping (EDX; Oxford Instruments INCAx-act).

2
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Fig. 1. Research methodology (left), HAp crystal lattice (center) and chart of dopant atoms sorted by increase of radius/electronegativity values (right).

Brunauer–Emmett–Teller (BET) specific surface area (SSA) mea­ performed in air from room temperature to 1200 ◦ C with a heating rate
surement was performed by a surface area analyzer (Micromeritics of 10 ◦ C/min using SETSYS - 1750 CS Evol., France.
ASAP-2020MP).
The thermogravimetric analysis (TGA) and differential scanning
calorimetry (DSC) analysis of the as-synthesized HAp NPs were carried 2.3. Computational method
out to determine thermal properties of the samples. The test was
The dopant positions in HAp structure were computed within DFT

Fig. 2. (A) XRD spectra (bar chart – Ca10(PO4)6(OH)2 (ICDD card # 01–073-6113); grey cube – Ni(OH)2 (ICDD card #01–075-6920); dark grey circle – CoCo2O4
(ICDD card #01–080-1533)), (B) Raman spectra, (C and D) FTIR spectra of bare and doped HAp samples.

3
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Table 1
NPs experimental parameters summary: band gap, crystallite size, lattice parameters a and c, relative change of lattice parameters Δa and Δc, average NP sizes ob­
tained from TEM images, SSА data.
Sample Band gap dXRD Lattice parameters (Å) (%) TEM (nm) SSA, m2/g
(nm)
a c Δa Δc l d

HAp 4.64 27.04 9.4257 6.8875 – – 58.3 ± 1.1 26.3 ± 0.4 45

Co-HAp 1.62 / 2.50 13.79 9.4157 6.8673 − 0.106 − 0.293 53.2 ± 0.6 22.0 ± 0.2 76
Ni-HAp 4.71 16.14 9.4251 6.8785 − 0.006 − 0.131 53.6 ± 1.0 26.7 ± 0.4 70
Cu-HAp 3.64 13.52 9.4375 6.8746 0.125 − 0.187 40.1 ± 1.6 19.7 ± 0.2 86

method, using pseudopotentials and a plane wave basis in ABINIT It’s quite usual within DFT calculations, to take lattice parameters for
software package [55]. The conducted geometrical optimization for 44- a bare bulk crystal, to construct a cell (or supercell) with dopant atoms
atom Ca10(PO4)6(OH)2 cells with P63/m symmetry was performed using being placed in the substitution positions and then perform quantum
the Broyden-Fletcher-Goldfarb-Shanno (BFGS) minimization algorithm. calculations with a geometry optimization procedure, which includes
For exchange and correlations, our DFT calculations employed both the not only optimization of atom positions, but of lattice parameters as
local density approximation (LDA) and the generalized gradient well. A disadvantage of this approach is that a bulk optimized lattice
approximation (GGA). Doping was done by replacing one of the cell Ca parameters will be obtained. All experimental data show, that lattice
atoms by the dopant atom in two dissimilar Ca atom positions (marked parameters in NPs are different from those in the bulk. And as one
below as Ca1 and Ca2). An energy cutoff of 400 eV was kept for all the knows, a relatively small variation in lattice parameters will result in
structures of HAp. noticeable changes in DOS and band gap.
Two different calculation approaches were tested. For the first one In our calculations, we applied a modified approach, using experi­
the 44-atom primitive unit cell with data from Springer Materials [56] mental XRD data for lattice parameters as an input data for quantum
was used, where cell parameters were unfixed and atom positions were calculations. Verification of calculation results was performed via
optimized. In the second approach, the same unit cell was used, but cell comparison of the calculated band gap values with the band gap values
parameters were fixed and taken equal to those obtained from experi­ obtained from experimental absorption spectra. Further data analysis
mental data [52] and atom positions were also optimized. allowed to establish relations between dopant properties and properties
Drawings were obtained in Vesta software package [57]. Also, Vesta of doped particles.
was used to generate coordinates of the atoms in simulation cells.
3.2. Impact of dopant nature on morphological and functional parameters
3. Results and discussion
of as-prepared HAp NPs

3.1. Research methodology, chemical and computational experiments

Characterization data of all as-prepared HAp NPs are given in Fig. 2.
design
The phase composition of as-prepared samples was determined by
XRD analysis. The XRD patterns of bare and doped HAp NPs are pre­
The Scheme illustrating the principle of research methodology is
sented in Fig. 2A and ESM Fig. 5. For all the samples the observed
given in Fig. 1.
diffraction peaks can be indexed to HAp phase (ICDD card # 01–073-
Optical and electric properties of dielectric and semiconductor ma­
6113) with hexagonal crystal structure (space group P63/m, 176) along
terials are determined by their band gap. Therefore, in order to obtain
with two low intensity impurity peaks corresponding to β-Ni(OH)2 for
material with required properties it is crucial to know how the structural
Ni-HAp sample (marked as grey square, ICDD card # 01–075-6920,
parameters and consequently the band gap are related to the synthesis
space group P3m1,164), and CoCo2O4 for the Co-HAp sample (marked
procedure. In most cases experimental data for band gap are not suffi­
as black circle, ICDD card # 01–080-1533, space group Fd3m,227).
cient to understand this relation, thus computer simulation is a powerful
Impurities will be discussed below.
tool to obtain energy spectrum and see the correlation between synthesis
The information about crystal structure of as-prepared samples was
procedure, structural parameters and the band gap. One of the methods
obtained from Raman spectra (Fig. 2B). The analysis of Raman spectra
intended to calculate DOS and band structure are DFT based quantum
revealed that all samples demonstrate typical HAp vibrations [58,59].
calculations, widely used to study semiconductor properties.
The PO3− 2 − 1 3− 4
4 ν vibrations (429 and 446–448 cm ), PO4 ν vibrations
An important issue of DFT calculation is a methodology selection, e. − 1 3− 3 − 1 3− 1
(588 and 609 cm ), PO4 ν (1044–1049 cm ) and PO4 ν vibrations
g. exchange functional or correction, which allows to obtain calculation
(960 cm− 1) are seen. Furthermore, a weak CO2− 3 vibrations were
results close to experimental data. Depending on formulation of a
observed at the frequencies around 1075–1082 cm− 1 (stretching ν1
problem, two approaches are possible 1) experiment plays a major role,
mode of CO2− − 1
3 ) and 3572 cm . Co-HAp sample was the only one, who
while computer calculation is used to better understand obtained data,
demonstrated the OH− group vibrations at 3572 cm− 1 and also the low
2) computer calculation plays a major role, and experiment is needed to
intensity peak at 690 cm− 1 which was associated with Co2O2− 4 vibrations
verify calculated results. In both cases, speaking of the subject of the
[60]. No other samples revealed any additional crystalline phase.
article, one needs to understand how a dopant nature impact NPs
The information on the nature of obtained samples was provided by
properties.
FTIR spectroscopy technique. The FTIR spectra of as-prepared bare and
Since morphological parameters of NPs among other factors depend
doped HAp are shown in Figs. 2С and 2D. For all the samples the
on dopant nature, our choice of synthesis conditions was based on the
characteristic bands for HAp were observed [13,61]. The positions
similarity of obtained particles sizes and shapes. In this case, depen­
around 475, 564, 604, 635 and 1032 cm− 1 were assigned as stretching
dence of electronic properties on particle sizes and shapes can be elim­
and bending of phosphate in HAp structure. The broad absorption band
inated and dependence of electronic properties on dopant nature can be
around 1642–1713 for Co-HAp and Ni-HAp and 3434–3573 cm− 1 cor­
studied. Dopants were especially chosen with the same oxidation state as
responds to the OH− stretching mode of adsorbed water and the defor­
one of the substituted atom and are characterized by different radiuses,
mation band of water molecule. The characteristic absorption peaks of
electronegativity and number of unpaired electron. This allows us to
CO2−3 groups are at 875 and 1389–1489 cm
− 1
related to the air carbon
study how mentioned dopant atom nature manifest in the properties of
dioxide. It should be noted that doping led to a significant shift in po­
NPs.
sitions of characteristic bands of OH− and CO2− 3 groups (up to 7 cm ).
− 1

4
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Fig. 3. TEM images (top row) and size distributions (length and thickness, bottom rows) of as-prepared samples. Color code from left to right: light grey fill –
undoped (bare) HAp, dark grey fill – doped with Co2+, cyan – with Cu2+, dark yellow – with Ni2+. (For interpretation of the references to color in this figure legend,
the reader is referred to the web version of this article.)

Fig. 4. (A) Optical transmittance spectra and (B) reflectance spectra for as-prepared samples; (C) Tauc plot calculated from optical transmittance spectra.

5
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Fig. 5. In the top: Dependence of relative change Δc on doping atom radius, dependence of relative change Δa on stability constant, dependence of band gap on
electronegativity. In the bottom: Sketch of bare and cobalt, nickel and copper doped HAp NPs. Crystallites are shown in scale in dark grey.

The presence of dopant atoms was determined by splitting peaks in absorption peak around 420 nm, thus it is grey [68]. Since, Co-HAp
range 540–590 and 620–700 cm− 1 and the new peaks were shown at sample the absorbance spectrum demonstrate continuous absorption
576 and 668 cm− 1 in Co-HAp and 573 cm− 1 in Cu-HAp samples corre­ in a large range of wavelength, thus, we defined two values of energy
spondingly, which associated with M-O bonding. In addition, the peaks gap, for both parts of the spectrum and gave them in the table separated
at 420 cm− 1 in Ni-HAp and 500 cm− 1 in Cu-HAp were associated with by slash. A peak observed for Ni-HAp is not related to β‑nickel hydrox­
M-O bonding [62–66]. No additional peaks related to Ni(OH)2 and ide, which maximum is less than 400 nm [69] and as well as smooth
CoCo2O4 were detected. All observed peaks are given in ESM Table 1. growth of absorption at wavelength higher than 600 nm, seen also in the
Fig. 3. shows TEM images of bare and doped HAp and corresponding case of copper, is caused by dopant incorporation in HAp structure.
size distributions. As it can be seen, all NPs are nanoscale and have a rod- By comparison of the reflectance and transmittance spectra (Fig. 4A,
like shape with aspect ratio of approximately 2. Size distribution was B) we have concluded that the peaks observed for all samples except Cu-
obtained from these images via measuring a hundred of different NPs HAp in range 200–800 nm are absorbance of light with the material, and
within ImageJ program. Average lengths and thicknesses are given in not the reflectance of material surface, because if the low uniform
Table 1. SSA values of as-prepared samples also are given in Table 1 and reflectance. The maximum reflectance among the samples is achieved
they are in good agreement with TEM results for NPs sizes. for Cu-HAp, which is in agreement with the bright color of the powder
Any particles with different shapes can be seen in TEM images, which (Fig. 1). This observation allows to recommend Cu-HAp sample for
indicate absence of impurity phases. Also, presence of a shell, which is application as biocompatible pigment.
especially obvious in the case of Co and Ni doped samples, should be Crystallites sizes and lattice parameters obtained from XRD data, and
noted, which will be discussed further. relative change of lattice parameters Δa and Δc comparing to bare HAp
The low (scale bar: 100 nm) and higher resolution (scale bar: 20 nm) lattice parameters, calculated via the formula ΔX = (X−X0X0 )∙100%, where
SEM images for all samples (ESM Fig. 1) as well as HR-TEM images (ESM X0 – lattice parameter of bare HAp, X – lattice parameter of doped HAp,
Fig. 2) confirm the TEM results, given in Fig. 3. SEM-EDX analysis shows are also given in the Table 1.
that the dopant content in HAp is slightly less than expected after syn­ The impact of the dopant nature on nanoparticles thermal properties
thetic procedure conducted, but within a margin of error. SEM elemental was studied. Thermal properties are important for materials used for
mapping results confirm the even dopant distribution in HAp structure scaffold since production includes thermal treatment. ESM Fig. 4 pre­
(ESM Fig. 3). sents TGA/DTG, DSC curves and the schematic illustration of thermal
A standard approach to experimental measure of a band gap is to transformations for all samples. The XRD results after thermal treatment
calculate it from absorption spectra (Fig. 4A) [67]. Method is based on (ESM Fig. 5) shows that the phase composition changes to the mixture of
linear approximation of modified absorption coefficient dependence on HAp, TCP (tricalcium phosphate) and dopant oxide. All thermal trans­
energy of probing radiation next to the fundamental border of absorp­ formations are denoted in accordance with Ref. 34. The dopant presence
tion, known as Tauc plot [54]. Absorption spectra and corresponding in HAp structure expectedly leads to decrease in thermal stability, which
Tauc plots are given in the Fig. 4C. Band gap values are given in Table 1. is more expressed for Co-HAp and less expressed for Cu-HAp sample.
Obtained spectra are in a good agreement with sample colors (See Note that the temperature of HAp nanoparticles partial conversion
Fig. 1). Bare HAp do not absorb VIS, thus it is white, Co-HAp partially for doped samples linearly depend on calculated total energy of the
absorb almost in the whole VIS, thus it is black, Cu-HAp shows an ab­ lattice (ESM Fig. 6.) We never met this fact in the literature, despite its
sorption minima around 500 nm, thus it is blue and Ni-HAp shows an obviousness. The lower is the total energy of the lattice, the higher is the

6
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Fig. 6. Calculated DOS within a standard approach with springer handbook lattice parameters with LDA (top row) and GGA (down row) functionals. Dopant
electronegativity increase from left to right.

conversion temperature. The found dependency not only additionally formation. In order to a doped particle would form, the complex have to
confirm correctness of our approach, but also opens a way to predict NP decay, and considering the amount of the dopant (10 at.%), this appears
thermal properties via lattice total energy calculations. to be a limiting factor of the particle growth. Also, it results in a slow
Consider an impact of dopant parameters (radius, electronegativity diffusion-controlled growth and leads to decrease of the obtained par­
and stability constant) on morphological (cell sizes, crystallite size, ticles sizes.
particle size) and functional (band gap) properties of doped NPs. De­ A relative change Δa is the smallest and negative for cobalt with the
pendences are shown in Fig. 5. lowest stability constant value, and positive for copper with the highest
It’s not surprising, that lattice parameters change upon doping. A stability constant value. We establish a linear dependence of Δa on
relative change Δc is negative, so the lattice of the doped HAp is smaller stability constant.
in the Z-direction and the greater a dopant atom radius is, the smaller is All of the obtained particles are monocrystalline. If for a bare HAp
the lattice parameter c, which indicates that dopant atom occupy cal­ the size of the crystallite is equal to the size of the NP, in the case of
cium position and it is related with substitution position Ca2. doped particles the size of the crystallite is smaller than the size of the
It’s known that 3d element ions form complexes in solution. Ac­ NP, which indicates the presence of a shell on the NP surface. In the case
cording to Refs. [70, 71] and taking into account the composition of the of Cu-HAp it is thinner than in Ni-HAp and Co-HAp cases, which can be
reaction media, 3d element ions are in the form [M(NH3)x(H2O)6-x]2+. seen in TEM images, Fig. 3. The monocrystallinity of the obtained
Stability of this complex is increasing in the row cobalt – nickel – copper samples was confirmed with SAED results (ESM Fig. 2).
[72,73]. As it is known from works on theory of particle growth [70,73], Crystallites sizes get smaller upon doping, since the dopant intro­
the speed of complex decay influence doped HAp crystal lattice duction leads to lattice distortion, which leads to reduction of defect-free

7
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Table 2
Calculated lattice parameters from springer data*.
Sample HAp Co-HAp Ni-HAp Cu-HAp

Functional LDA GGA LDA GGA LDA GGA LDA GGA

a, Å 9.25377 9.56144 9.24112 9.53489 9.24165 9.55303 9.25424 9.55400

Δa, %** − 1.82 1.44 − 1.85 1.27 − 1.95 1.36 − 1.94 1.23
b, Å 9.25377 9.56144 9.25794 9.55501 9.26466 9.57262 9.27790 9.57748
Δb, % − 1.82 1.44 − 1.68 1.48 − 1.70 1.57 − 1.69 1.48
c, Å 6.74560 6.89001 6.70853 6.86375 6.70290 6.86130 6.71236 6.84786
Δc, % − 2.06 0.04 − 2.31 − 0.05 − 2.55 − 0.25 − 2.36 − 0.39
Band gap 5.22 5.18 2.09/2.30 2.59/2.31 0.96/4.02 0.79/3.12 4.08 3.99
Fermi energy, eV 3.81 3.81 6.87 6.19 5.80 5.18 5.01 4.42

*Input crystal cell parameters: a = b = 9.39830 Å, c = 6.86770 Å [56].

(X − X0 )
**Relative difference between calculated and experimental values according to the formula: ∆X = ∙100%, where X0 – experimental value, X – value obtained
X0
in the computer calculation.

length. The biggest change is observed for dopant with highest stability marked as Ca1 and Ca2 (see. Fig. 1), so at first, we studied what is the
constant value. substitution position of our dopants.
Considering, that parameter a of nickel hydroxide is multiple of the Using handbook data (Springer Materials) we calculated energy of
parameter a of HAp (precisely 3) and parameters a of cobalt oxide are the cell with two functionals (LDA and GGA) for two possible dopant
only 14% smaller than parameter a of HAp (which, according to Ref. 74 substitution positions. We found, that for both functionals the cell en­
allows a heteroepitaxial layer to form), that we can’t see any additional ergy is lower in the case of Ca2 dopant substitution position. Also, since
phase in TEM pictures and that average NP sizes are in a good agreement in a HAp cell there are 4 equivalent Ca1 and 6 equivalent Ca2 atoms, we
with SSA data, we assume that additional phase seen in XRD analysis are checked the equivalency and our calculations showed that the cell en­
forming a thin shell on the HAp NP surface. The lack of such shell in the ergy do not depend which of equivalent positions is taken. For further
case of copper doped NPs may be related to a lower growth rate due to calculations with both functionals we used dopant substitution positions
higher complex stability constant. Ca2, which coincides with literature data [51].
Finally, we established that band gap is in linear dependence on Data for cell sizes, band gap and Fermi energy obtained within
electronegativity of dopant, which characterize the ability of the atom standard approach, with the Springer crystallography data [56], after
attract a shared pair of electrons (electron density in case of the crystal) optimization of cell sizes and atom positions are given in Table 2. Col­
towards itself. It is a general believe, that the band gap depend on the umns in the table are arranged in order of increasing total number of
number of electrons subject to equal oxidation states. But, as we see, not electrons in the dopant atom (or in order of decreasing number of un­
the number of electrons, but electrons distribution plays a major role. In paired elections). Obtained data are in a good agreement with literature
chemistry it corresponds to electronegativity. So the width of band gap data for bare and cobalt doped HAp GGA calculations: parameter a in­
might be tuned based on electronegativity of the dopant. This fact needs crease from 9.4 up to 9.5–9.52, and parameter c iancrease from 6.85 up
to be further studied and checked on different host matrices. Based on all to 6.9 [51,77]. Since substitution of a doping atom resulted in a sym­
the obtained data, we can conclude, that a minor quantity of additional metry break, an a = b requirement was no longer imposed, but the
phase we see in XRD analysis are included in the surface layer of HAp difference between a and b is 10 times less than difference between
NPs and do not impact the band gap. calculated and experimental values and thus can be ignored.
Summarizing the section, we performed a successful hydrothermal The difference between calculated and experimental values is about
synthesis of bare and doped HAp NPs at lower temperatures (180 ◦ C vs. 2%, which is consistent with results obtained by other authors [49,50].
average 200 ◦ C), lower holding times (2 h vs average 12 h) and higher This is a quite big difference, considering a fact, that even a 0,5% change
dopant concentrations than were previously reported [75,76]. Obtained will result in significant changes in properties [6,7].
NPs possess a narrow size distribution, there is no signs of in­ In the case of LDA functional geometry optimization results in a
homogeneity of dopant distribution, doping leads to decrease of a band decrease of lattice parameters and gives values even smaller than
gap, dependencies of morphological and functional properties on dopant experimental ones, see Tables 1 and 2.
parameters were established. In the case of GGA functional geometry optimization results in an
increase of input value of lattice parameter a and, except the case of bare
HAp, decrease of input lattice parameter c. Comparison of calculation
3.3. Structural properties and band structures of bare and doped HAp
results with experimental data revealed that calculated values of lattice
NPs: standard approach
parameters a and b are greater than experimental ones and calculated
lattice parameters c, again except the case of bare HAp, are slightly
In this part of the work we studied the impact of calculation meth­
smaller than experimental, see Tables 1 and 2.
odology and exchange functionals on calculation results, which we
If we compare calculation results for doped samples with calculation
evaluated by the difference between calculated and experimentally
results for bare HAp, we can see the same patterns as in the experiment.
measured values of the band gap.
Both functionals give negative relative change of c, relative change of a
We used simple LDA and GGA functionals. The latter is known to give
(since a = b requirement is not valid in the calculation, so we consider an
good results in the study of doped HAp [33,50], LDA functional is not
average of a and b), is negative for cobalt and nickel and positive for
widely used for this purpose. Hybrid functionals are gaining popularity
copper.
but require noticeably more computational power.
DOS are given in the Fig. 6.
First, we used an approach, which is considered a standard and
Calculated values of band gap and Fermi Energy are given in Table 2.
consist in construction of supercell based on handbook sizes of host
DOS structure coincide with the literature data for bare [77] and cobalt
crystal lattice and geometry optimization of cell sizes and atom
doped HAp [38], however in the latter work a cobalt peak (see further
positions.
PDOS) is closer to the valence band by about 2 eV, which can be related
According to the literature, 3d elements in HAp substitute into Ca
with difference in lattice parameters (a = 9.52, с = 6.94, calculation was
positions [45,51]. In HAp there are two nonequivalent Ca positions

8
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Fig. 7. On the left: Fermi energy dependence on the number of unpaired electrons. On the right: comparison of experimental (left bars) and calculated (right bars)
band gap values with relative errors. Dopants are arranged in the order of increase of electronegativity.

Fig. 8. Calculated PDOS obtained within standard approach, with geometry optimization of atom positions and lattice parameters with LDA functional. In the
bottom - band structure. Dash lines display the Fermi energy levels for each sample.

performed in VASP with GGA functional) [51,77]. the band gap on electronegativity, this is not the case for calculation
For both functionals Fermi energy, as expected, linearly depends on results. Calculated band gap values increase with decrease of the num­
the number of unpaired electrons: the greater their number, the closer to ber of unpaired electrons.
the conduction band is Fermi level, see Fig. 7. Calculated PDOS and band structures, obtained with LDA functional,
Calculated band gaps differ from experimentally obtained values by since computed values are closer to experimental values, are given in
at least 10%, see Fig. 7. Experimental data showed linear dependence of Fig. 8.

9
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Fig. 9. Calculated DOS for as-prepared samples with geometry optimization (lattice parameters are taken from experimental data and are fixed) with LDA (A) and
GGA (B) functionals.

Obtained data confirmed, that dopants are electron donors, since a direct.
new peak in DOS plot lays between Fermi level and valence band [78]. It was revealed that in doped semiconductors the indirect band gap
Dopants were intentionally selected to be d-elements in order to do not narrowing is mainly caused by impurity state, while the direct band gap
provide defects in lattice structure. narrowing is a result of lattice distortion induced by the dopant atom
From the band structure for all samples the minimum of conduction [81].
band is occurring at Gamma point. For the bare HAp sample the band
gap is direct, the minima of conduction band as well as maxima of 3.4. Structural properties and band structures of bare and doped HAp
valence band are positioned on a same point, thus the transition of NPs: original approach
electron easily happens without wasting incident energy [79].
As for doped samples the maxima of valence band occurs between H As was already said, within our original approach we use lattice
and A points, which means that the band gap is indirect, e.i. the electron parameters obtained from XRD experimental data as input for DFT
transition occur by wasting considerable amount of incident energy. calculation, in the geometry optimization the change of lattice param­
In the case of Ni-HAp, there is a high peak in the middle of the band eters was prohibited. Within our approach, in a similar way it was
gap, so, if we will add up two gap values on both sides of the peak, we described earlier, we established that Ca2 substitution position provide
can achieve an agreement with experimental data. lower energy, also equivalence of all Ca2 positions in the cell was
There is no consent about the band gap in the literature. In the work verified. Obtained DOS are given in Fig. 9.
[77] authors obtained indirect band gap for bare HAp, meanwhile in the DOS structures appears to be similar to the DOS structures obtained
work [80], which gives a reference to previous work direct band gaps within standard approach, there are two band gaps in the case of cobalt
were obtained both for bare and doped HAp. In the work [51] authors do and nickel. In the nickel case, in the optical transmittance spectra we see
not say anything about the band gap, but according to the pictures it is absorbance at 420 nm, which is roughly 3 eV, as well as slight increase of

10
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

Fig. 10. On the left: Fermi energy dependence on the number of unpaired electrons. On the right: comparison of experimental (left bars) and calculated (right bars)
band gap values with relative errors. Original approach. Dopants are arranged in the order of increase of electronegativity.

Fig. 11. Calculated PDOS for as-prepared samples with geometry optimization of atom positions and LDA functional, band structure. Dash lines display the Fermi
energy levels for each sample.

absorbance at wavelength higher than 600 nm, and this is in agreement are even smaller. Calculation results within our approach are in better
with calculated spectra. The sum of the band gaps coincide with the agreement with experimental data than results within standard
value of the band gap, determined from optical transmittance spectra. approach.
Similarly to previous calculation, for both functionals it appears that As for GGA functional, deviations of band gap values form experi­
Fermi energy is linearly dependent on the number of unpaired electrons, mental results are greater than 10%. So we can conclude, that LDA
Fig. 10. functional gave better results than GGA within both approaches.
It can be seen, that deviations of both calculated band gap values Another important issue is that calculated band gap as well as
from experimental band gap values in the case of cobalt are very close experimental one depends on electronegativity. We can resume, that
and are of the order of 10%. Deviations in the case of nickel and copper among other things being equal, the main factor is electronegativity,

11
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

which should be taken into account when choosing dopant, at least in the work reported in this paper.
the HAp case, in order to obtain samples with predetermined properties.
PDOS and band structures obtained with LDA functional are given in Acknowledgments
Fig. 11. Samples are arranged in the order of decrease of the number of
unpaired electrons. Scientific research was performed at the Centre for X-ray Diffraction
In this case, we see that bare HAp is a wide zone p-type semi­ Studies, Center for Chemical Analysis and Materials Research, Centre for
conductor, since Fermi level lays in the top of valence band. Optical and Laser Materials Research, Computing Centre of the Research
Doped samples represent n-type semiconductors since dopant levels Park of Saint Petersburg State University.
possess high intensity, are donor levels and lay between Fermi level and
valence band. Appendix A. Supplementary data
An interesting fact is that in the case of cobalt and copper we see
appearance of acceptor levels close to the conduction band. Supplementary data to this article can be found online at https://doi.
org/10.1016/j.matchar.2021.110911.
4. Conclusion

References
Hydroxyapatite (HAp) is a promising biodegradable and biocom­
patible material, with a large area of possible applications. We synthe­ [1] A. Kumar, S. Kargozar, F. Baino, S.S. Han, Additive manufacturing methods for
sized HAp bare and doped with cobalt, nickel and copper, we showed producing hydroxyapatite and hydroxyapatite-based composite scaffolds: a review,
that doping leads to decrease of the band gap, which influence electrical Front. Mater. 6 (1) (2019) 1–20, https://doi.org/10.3389/fmats.2019.00313.
[2] A. Grumezescu, Nanobiomaterials in Galenic Formulations and Cosmetics:
properties of the material as well as its color. Applications of Nanobiomaterials, Elsevier, 2016.
It is known that properties are dependent on size and shape of [3] S.V. Dorozhkin, Biocomposites and hybrid biomaterials based on calcium
nanoparticles (NPs), which in their turn depend on synthesis conditions. orthophosphates, Biomatter 1 (1) (2011) 3–56, https://doi.org/10.4161/
biom.1.1.16782.
We performed synthesis of bare and doped NPs with similar morphology
[4] S. Zhang, Hydroxyapatite Coatings for Biomedical Applications, CRC Press, 2013.
to eliminate its influence and study dependence of the material prop­ [5] N.T.T. Linh, P.D. Tuan, N. Van Dzung, The shifts of band gap and binding energies
erties on dopant nature. Prediction of the functional properties is a of Titania/hydroxyapatite material, J. Compos. 1 (2014) (2014) 1–5, https://doi.
org/10.1155/2014/283034.
crucial problem, since it can withdraw the necessity to synthesize a large
[6] V.S. Bystrov, et al., Oxygen vacancies, the optical band gap (Eg) and Photocatalysis
set of samples. of hydroxyapatite: comparing Modelling with measured data, Appl. Catal. B
We carried out quantum chemical calculations based on DFT within Environ. 196 (2016) 100–107, https://doi.org/10.1016/j.apcatb.2016.05.014.
a standard approach, performing an optimization procedure of the [7] H. Pan, Y. Zhang, Y. Hu, H. Xie, Effect of Cobalt Doping on Optical, Magnetic and
Photocatalytic Properties of ZnO Nanoparticles, Optik (Stuttg) 208 (2020), https://
crystal cell sizes and atom positions. Deviation of obtained in calculation doi.org/10.1016/j.ijleo.2020.164560, 164560.
band gaps from experimental data was greater than 10%, dependencies [8] I.A. Neacsu, A.E. Stoica, B.S. Vasile, E. Andronescu, Luminescent hydroxyapatite
of the band gap on dopant electronegativity seen in the experiment were doped with rare earth elements for biomedical applications, Nanomaterials 9 (2)
(2019) 239, https://doi.org/10.3390/nano9020239.
not reproduced. [9] D. Lew, A. Marino, The effect of electrical stimulation on bone formation around
While within a suggested approach, based on usage of crystal cell hydroxyapatite implants placed on the rabbit mandible, J. Oral Maxillofac. Surg.
parameters obtained from XRD data and optimization of atom positions 49 (7) (1991) 735–739, https://doi.org/10.1016/S0278-2391(10)80238-1.
[10] Ž. Radovanović, B. Jokić, D. Veljović, S. Dimitrijević, V. Kojić, R. Petrović,
only, deviation of calculated band gap values from experimental data D. Janaćković, Antimicrobial activity and biocompatibility of Ag+− and Cu2+-
was 10% and less, dependence of the band gap on dopant electronega­ doped biphasic hydroxyapatite/α-Tricalcium phosphate obtained from
tivity was reproduced. hydrothermally synthesized Ag+− and Cu2+-doped hydroxyapatite, Appl. Surf.
Sci. 307 (2014) 513–519, https://doi.org/10.1016/j.apsusc.2014.04.066.
Results obtained with LDA functional were in better agreement with [11] M.H. Santos, P. Valerio, A.M. Goes, M.F. Leite, L.G.D. Heneine, H.S. Mansur,
experimental data that results obtained with GGA functional and the Biocompatibility evaluation of hydroxyapatite/collagen Nanocomposites doped
higher is the electronegativity, the smaller is the deviation. Higher with Zn+2, Biomed. Mater. 2 (2) (2007) 135–141, https://doi.org/10.1088/1748-
6041/2/2/012.
dopant atom electronegativity results in electron density redistribution
[12] A. Vladescu, et al., Mechanical properties and biocompatibility of the sputtered Ti
and better results in local density approximation. doped hydroxyapatite, J. Mech. Behav. Biomed. Mater. 63 (2016) 314–325,
In the synthesis procedure there are chemical processes that influ­ https://doi.org/10.1016/j.jmbbm.2016.06.025.
ence atomic positions, which can result in a cell configuration not [13] C.S. Ciobanu, et al., Structural properties of silver doped hydroxyapatite and their
biocompatibility, Mater. Sci. Eng. C 33 (3) (2013) 1395–1402, https://doi.org/
optimal from energy point of view. That’s the reason why calculation 10.1016/j.msec.2012.12.042.
based on optimization of handbook data fail to account for this effects. [14] C.L. Popa, A. Deniaud, I. Michaud-Soret, R. Guégan, M. Motelica-Heino, D. Predoi,
Comparison of the band gap before and after optimization give a 0.5 eV Structural and biological assessment of zinc doped hydroxyapatite nanoparticles,
J. Nanomater. 2016 (2016) 1–10, https://doi.org/10.1155/2016/1062878.
difference. In order to describe and clarify experimental results, it is [15] K.-H. Zuo, Y.-P. Zeng, D. Jiang, Synthesis and magnetic property of Iron ions-doped
suitable to use lattice parameters obtained from XRD data. hydroxyapatite, J. Nanosci. Nanotechnol. 12 (9) (2012) 7096–7100, https://doi.
Calculations also revealed linear dependence of partial conversion org/10.1166/jnn.2012.6578.
[16] O. Kaygili, S.V. Dorozhkin, T. Ates, A.A. Al-Ghamdi, F. Yakuphanoglu, Dielectric
temperature on lattice total energy, showing the possibility to predict properties of Fe doped hydroxyapatite prepared by sol–gel method, Ceram. Int. 40
the dopant impact on NP stability. (7) (2014) 9395–9402, https://doi.org/10.1016/j.ceramint.2014.02.009.
Among factors influencing the material properties, the main is [17] K.P. Tank, B.V. Jogiya, D.K. Kanchan, M.J. Joshi, Dielectric properties of pure and
strontium doped Nano-hydroxyapatite, Solid State Phenom. 209 (2013) 151–155,
electronegativity of the dopant, which should be taken into account
https://doi.org/10.4028/www.scientific.net/SSP.209.151.
while choosing the dopant, at least in the case of HAp. [18] M. Ibrahim, S. Sallam, The Dielectric Study of Calcium Hydroxylapatite Doped by
Magnesium Ions, Egypt. J. Biomed. Eng. Biophys 19 (1) (2018) 1–11, https://doi.
org/10.21608/ejbbe.2018.3208.1016.
Data availability
[19] F. Mohandes, M. Salavati-Niasari, Freeze-Drying Synthesis Characterization and in
Vitro Bioactivity of Chitosan/Graphene Oxide/Hydroxyapatite Nanocomposite,
The data that support the findings of this study are available from the RSC Adv 4 (49) (2014) 25993, https://doi.org/10.1039/c4ra03534h.
corresponding author upon reasonable request. [20] F. Mohandes, M. Salavati-Niasari, M. Fathi, Z. Fereshteh, Hydroxyapatite
Nanocrystals: simple preparation, characterization and formation mechanism,
Mater. Sci. Eng. C 45 (2014) 29–36, https://doi.org/10.1016/j.msec.2014.08.058.
Declaration of Competing Interest [21] F. Mohandes, M. Salavati-Niasari, Simple morphology-controlled fabrication of
hydroxyapatite nanostructures with the aid of new organic modifiers, Chem. Eng.
J. 252 (2014) 173–184, https://doi.org/10.1016/j.cej.2014.05.026.
The authors declare that they have no known competing financial [22] F. Mohandes, M. Salavati-Niasari, In vitro comparative study of pure
interests or personal relationships that could have appeared to influence hydroxyapatite Nanorods and novel polyethylene glycol/Graphene oxide/

12
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

hydroxyapatite Nanocomposite, J. Nanopart. Res. 16 (9) (2014) 2604, https://doi. [45] S. Aryal, K. Matsunaga, W. Ching, Ab initio simulation of elastic and mechanical
org/10.1007/s11051-014-2604-y. properties of Zn- and mg-doped hydroxyapatite (HAP), J. Mech. Behav. Biomed.
[23] F. Mohandes, M. Salavati-Niasari, Particle size and shape modification of Mater. 47 (2015) 135–146, https://doi.org/10.1016/j.jmbbm.2015.03.018.
hydroxyapatite nanostructures synthesized via a Complexing agent-assisted route, [46] A. Le Bail, Whole powder pattern decomposition methods and applications: a
Mater. Sci. Eng. C 40 (2014) 288–298, https://doi.org/10.1016/j. retrospection, Powder Diffract. 20 (4) (2005) 316–326, https://doi.org/10.1154/
msec.2014.04.008. 1.2135315.
[24] E.A. Ofudje, A.I. Adeogun, M.A. Idowu, S.O. Kareem, Synthesis and [47] I.Y. Grubova, M.A. Surmeneva, S. Huygh, R.A. Surmenev, E.C. Neyts, Effects of
Characterization of Zn-Doped Hydroxyapatite: Scaffold Application, Antibacterial silicon doping on strengthening adhesion at the Interface of the
and Bioactivity Studies, Heliyon 5 (5) (2019), https://doi.org/10.1016/j. hydroxyapatite–titanium biocomposite: a first-principles study, Comput. Mater.
heliyon.2019.e01716 e01716. Sci. 159 (2019) 228–234, https://doi.org/10.1016/j.commatsci.2018.12.026.
[25] K. Poovendran, K.S.J. Wilson, M.S. Revathy, A. Ayeshamariam, K. Kaviyarasu, [48] T. Tamm, M. Peld, Computational study of Cation substitutions in Apatites, J. Solid
Functionalization effect of HAp with copper (cu) having excellent dielectric State Chem. 179 (5) (2006) 1581–1587, https://doi.org/10.1016/j.
applications, Surfaces and Interfaces 19 (2020) 100474, https://doi.org/10.1016/ jssc.2006.02.012.
j.surfin.2020.100474. [49] F.E. Imrie, M. Corno, P. Ugliengo, I.R. Gibson, Computational studies of
[26] C.S. Ciobanu, F. Massuyeau, L.V. Constantin, D. Predoi, Structural and physical magnesium and strontium substitution in hydroxyapatite, Key Eng. Mater.
properties of antibacterial Ag-doped Nano-hydroxyapatite synthesized at 100◦ C, 529–530 (2012) 123–128, https://doi.org/10.4028/www.scientific.net/KEM.529-
Nanoscale Res. Lett. 6 (1) (2011) 1–8, https://doi.org/10.1186/1556-276X-6-613. 530.123.
[27] T. Nagyné-Kovács, et al., Synthesis and characterization of Sr and mg-doped [50] R.J. Friederichs, H.F. Chappell, D. V Shepherd, and S.M. Best, Synthesis
hydroxyapatite by a simple precipitation method, Ceram. Int. 44 (18) (2018) Characterization and Modelling of Zinc and Silicate Co-Substituted
22976–22982, https://doi.org/10.1016/j.ceramint.2018.09.096. Hydroxyapatite, J. R. Soc. Interface 12, 108 (2015) 20150190. doi:https://doi.
[28] X. Yuan, B. Zhu, G. Tong, Y. Su, X. Zhu, Wet-chemical synthesis of mg-doped org/10.1098/rsif.2015.0190.
hydroxyapatite nanoparticles by step reaction and ion exchange processes, [51] M.E. Zilm, L. Chen, V. Sharma, A. McDannald, M. Jain, R. Ramprasad, M. Wei,
J. Mater. Chem. B 1 (47) (2013) 6551–6559, https://doi.org/10.1039/c3tb21315c. Hydroxyapatite substituted by transition metals: experiment and theory, Phys.
[29] M. Bricha, Y. Belmamouni, E.M. Essassi, J.M.F. Ferreira, K. El Mabrouk, Chem. Chem. Phys. 18 (24) (2016) 16457–16465, https://doi.org/10.1039/
Hydrothermal synthesis and appraisal of mg-doped hydroxyapatite Nanopowders, C6CP00474A.
J. Biomater. Tissue Eng. 3 (5) (2013) 570–580, https://doi.org/10.1166/ [52] I.E. Kolesnikov, D.S. Kolokolov, M.A. Kurochkin, M.A. Voznesenskiy, M.
jbt.2013.1120. G. Osmolowsky, E. Lähderanta, O.M. Osmolovskaya, Morphology and doping
[30] M.M.A. Rafique, Hydrothermal processing of phase pure and doped hydroxyapatite concentration effect on the luminescence properties of SnO2:Eu3+ nanoparticles,
and its characterization, J. Encapsulation Adsorpt. Sci. 08 (01) (2018) 18–37, J. Alloys Compd. 822 (2020) 153640, https://doi.org/10.1016/j.
https://doi.org/10.4236/jeas.2018.81002. jallcom.2020.153640.
[31] T. Nagyné-Kovács, B. Mészáros, M. Molnár, M. Tolner, I.E. Lukács, I.M. Szilágyi, [53] A.L. Patterson, The Scherrer formula for X-ray particle size determination, Phys.
G. Pokol, Hydrothermal synthesis of Sr-doped hydroxyapatite and its antibacterial Rev. 56 (10) (1939) 978–982, https://doi.org/10.1103/PhysRev.56.978.
activity, Period. Polytech. Chem. Eng. 64 (1) (2019) 54–60, https://doi.org/ [54] J. Tauc, Optical properties and electronic structure of amorphous Ge and Si, Mater.
10.3311/PPch.14062. Res. Bull. 3 (1) (1968) 37–46, https://doi.org/10.1016/0025-5408(68)90023-8.
[32] Z. Stojanović, L. Veselinović, S. Marković, N. Ignjatović, D. Uskoković, [55] X. Gonze, et al., ABINIT: first-principles approach to material and Nanosystem
Hydrothermal synthesis of Nanosized pure and cobalt-exchanged hydroxyapatite, properties, Comput. Phys. Commun. 180 (12) (2009) 2582–2615, https://doi.org/
Mater. Manuf. Process. 24 (10–11) (2009) 1096–1103, https://doi.org/10.1080/ 10.1016/j.cpc.2009.07.007.
10426910903032113. [56] P. Villars, K. Cenzual, Ca5(PO4)3(OH) (Ca5[PO4]3[OH] Rt) Crystal Structure:
[33] A. Bootchanont, W. Sailuam, S. Sutikulsombat, L. Temprom, Synchrotron X-ray Datasheet from `PAULING FILE Multinaries Edition – 2012′ in Springer Materials.
absorption spectroscopy study of local structure in strontium-doped Https://Materials.Springer.Com/Isp/Crystallographic/Docs/Sd_1626099, 2020.
hydroxyapatite, Ceram. Int. 43 (14) (2017) 11023–11027, https://doi.org/ [57] K. Momma, F. Izumi, VESTA 3 for three-dimensional visualization of crystal,
10.1016/j.ceramint.2017.05.144. volumetric and morphology data, J. Appl. Crystallogr. 44 (6) (2011) 1272–1276,
[34] E.A. Syukkalova, A.V. Sadetskaya, N.D. Demidova, N.P. Bobrysheva, M. https://doi.org/10.1107/S0021889811038970.
G. Osmolowsky, M.A. Voznesenskiy, O.M. Osmolovskaya, The effect of reaction [58] C.C. Silva, A.S.B. Sombra, Raman spectroscopy measurements of hydroxyapatite
medium and hydrothermal synthesis conditions on morphological parameters and obtained by mechanical alloying, J. Phys. Chem. Solids 65 (5) (2004) 1031–1033,
thermal behavior of calcium phosphate nanoparticles, Ceram. Int. 47 (2) (2021) https://doi.org/10.1016/j.jpcs.2003.10.071.
2809–2821, https://doi.org/10.1016/j.ceramint.2020.09.135. [59] V. Nosenko, et al., Structure of biocompatible coatings produced from
[35] M. Singh, M. Goyal, K. Devlal, Size and shape effects on the band gap of hydroxyapatite nanoparticles by detonation spraying, Nanoscale Res. Lett. 10 (1)
semiconductor compound Nanomaterials, J. Taibah Univ. Sci. 12 (4) (2018) (2015) 464, https://doi.org/10.1186/s11671-015-1160-4.
470–475, https://doi.org/10.1080/16583655.2018.1473946. [60] C. Julien, S. Gastro-Garcia, Lithiated Cobaltates for lithium-ion BatteriesStructure,
[36] Y. Wang, T. Jiang, D. Meng, J. Yang, Y. Li, Q. Ma, J. Han, Fabrication of morphology and electrochemistry of oxides grown by solid-state reaction, wet
nanostructured CuO films by Electrodeposition and their Photocatalytic properties, chemistry and film deposition, J. Power Sources 97–98 (2001) 290–293, https://
Appl. Surf. Sci. 317 (2014) 414–421, https://doi.org/10.1016/j. doi.org/10.1016/S0378-7753(01)00676-0.
apsusc.2014.08.144. [61] R. Cai, H. Wang, M. Cao, L. Hao, L. Zhai, S. Jiang, X. Li, Synthesis and antimicrobial
[37] Y. Wang, X. Yang, Q. Ma, J. Kong, H. Jia, Z. Wang, M. Yu, Preparation of flower- activity of Mesoporous Hydroxylapatite/zinc oxide Nanofibers, Mater. Des. 87
like CdS with SDBS as surfactant by hydrothermal method and its optical (2015) 17–24, https://doi.org/10.1016/j.matdes.2015.08.004.
properties, Appl. Surf. Sci. 340 (2015) 18–24, https://doi.org/10.1016/j. [62] M. Ghashghaee, M. Fallah, A. Rabiee, Superhydrophobic Nanocomposite coatings
apsusc.2015.02.104. of poly(methyl methacrylate) and stearic acid grafted CuO nanoparticles with
[38] Y. Wang, Q. Ma, H. Jia, Z. Wang, One-step solution synthesis and formation Photocatalytic activity, Prog. Org. Coatings 136 (2019) 105270, https://doi.org/
mechanism of flower-like ZnO and its structural and optical characterization, 10.1016/j.porgcoat.2019.105270.
Ceram. Int. 42 (9) (2016) 10751–10757, https://doi.org/10.1016/j. [63] R. Sathyamoorthy, K. Mageshwari, Synthesis of hierarchical CuO microspheres:
ceramint.2016.03.200. Photocatalytic and antibacterial activities, Phys. E Low-Dimensional Syst.
[39] Y. Wang, D. Wang, B. Yan, Y. Chen, C. Song, Fabrication of diverse CuO Nanostructures 47 (2013) 157–161, https://doi.org/10.1016/j.
nanostructures via hydrothermal method and their Photocatalytic properties, physe.2012.10.019.
J. Mater. Sci. Mater. Electron. 27 (7) (2016) 6918–6924, https://doi.org/10.1007/ [64] A.J. Haider, R. Al Anbari, H.M. Sami, M.J. Haider, Enhance Preparation and
s10854-016-4645-8. Characterization of Nickel-Oxide as Self-Cleaning Surfaces, Energy Procedia 157
[40] X. Yang, Y. Wang, Z. Wang, X. Lv, H. Jia, J. Kong, M. Yu, Preparation of CdS/TiO2 (2018) 1328–1342, https://doi.org/10.1016/j.egypro.2018.11.298.
nanotube arrays and the enhanced Photocatalytic property, Ceram. Int. 42 (6) [65] T. Tsoncheva, L. Ivanova, C. Minchev, M. Fröba, Cobalt-modified Mesoporous
(2016) 7192–7202, https://doi.org/10.1016/j.ceramint.2016.01.109. MgO, ZrO2, and CeO2 oxides as catalysts for methanol decomposition, J. Colloid
[41] X. Yang, Z. Wang, X. Lv, Y. Wang, H. Jia, Enhanced Photocatalytic activity of Zn- Interface Sci. 333 (1) (2009) 277–284, https://doi.org/10.1016/j.
doped dendritic-like CdS structures synthesized by hydrothermal synthesis, jcis.2008.12.070.
J. Photochem. Photobiol. A Chem. 329 (2016) 175–181, https://doi.org/10.1016/ [66] N. Osakoo, et al., Development and characterization of silica supported cobalt
j.jphotochem.2016.07.005. oxides for ethanol oxidation using different preparation methods, Radiat. Phys.
[42] S. Ahmadian-Fard-Fini, D. Ghanbari, M. Salavati-Niasari, Photoluminescence Chem. 171 (2020) 108718, https://doi.org/10.1016/j.radphyschem.2020.108718.
carbon dot as a sensor for detecting of pseudomonas Aeruginosa Bacteria: [67] R. Bhatt, I. Bhaumik, S. Ganesamoorthy, A.K. Karnal, M.K. Swami, H.S. Patel, P.
hydrothermal synthesis of magnetic hollow NiFe2O4 -carbon dots Nanocomposite K. Gupta, Urbach tail and Bandgap analysis in near stoichiometric LiNbO3 crystals,
material, Compos. Part B Eng. 161 (2019) 564–577, https://doi.org/10.1016/j. Phys. Status Solidi 209 (1) (2012) 176–180, https://doi.org/10.1002/
compositesb.2018.12.131. pssa.201127361.
[43] A. Mariappan, P. Pandi, N. Balasubramanian, R.R. Palanichamy, K. Neyvasagam, [68] S.K. L, Synthesis, Characterization and Optical Properties of Rare Earth –
Structural, optical and antimicrobial activity of copper and zinc doped Transition Metal Based Environmentally Friendly Red and Yellow Pigments, NIIST,
hydroxyapatite Nanopowders using sol-gel method, Mech. Mater. Sci. Eng. J. 9 2012.
(2017) 1, https://doi.org/10.2412/mmse.1.46.162. [69] D.S. Hall, D.J. Lockwood, C. Bock, B.R. MacDougall, Nickel hydroxides and related
[44] M. Eddya, B. Tbib, K. El-Hami, High Photocatalytic Activity of Hydroxyapatite materials: a review of their structures, synthesis and properties, Proc. R. Soc. A
Biodegradable Semiconductor for Solar Panels and Environment Protection, Incert. Math. Phys. Eng. Sci. 471 (2174) (2015) 20140792, https://doi.org/10.1098/
fiabilité des systèmes multiphysiques 2 (1) (2018) 1–8, https://doi.org/10.21494/ rspa.2014.0792.
ISTE.OP.2018.0262.

13
A.V. Sadetskaya et al. Materials Characterization 173 (2021) 110911

[70] T. Sugimoto, Growth, Monodispersed Part, Elsevier, 2019, pp. 95–124, https://doi. [77] L.A. Avakyan, E.V. Paramonova, J. Coutinho, S. Öberg, V.S. Bystrov, L.A. Bugaev,
org/10.1016/B978-0-444-62749-0.00002-8. Optoelectronics and defect levels in hydroxyapatite by first-principles, J. Chem.
[71] T. Saito, Inorganic Chemistry, CreateSpace, 2014. Phys. 148 (15) (2018) 154706, https://doi.org/10.1063/1.5025329.
[72] B. Averill, P. Eldredge, General Chemistry: Principles, Patterns, and Applications, [78] U. Dehlinger, Solid state physics, Z. Phys. Chem. 9 (1956) 139, https://doi.org/
Saylor Academy, 2011. 10.1524/zpch.1956.9.1_2.139.
[73] J.E. House, Composition and Stability of Complexes, Elsevier, Inorg. Chem, 2020, [79] M. Rizwan, M. Anwar, Z. Usman, M. Shakil, S.S.A. Gillani, H.B. Jin, C.B. Cao,
pp. 745–768, https://doi.org/10.1016/B978-0-12-814369-8.00019-4. U. Mushtaq, Implementation of magnesium doping in SrTiO3 for correlating
[74] M.A. Herman, W. Richter, H. Sitter, Heteroepitaxy; Growth Phenomena, Epitaxy, electronic, structural and optical properties: a DFT study, Chin. J. Phys. 62 (2019)
2004, pp. 389–421, https://doi.org/10.1007/978-3-662-07064-2_14. 388–394, https://doi.org/10.1016/j.cjph.2019.09.036.
[75] A.J. Nathanael, D. Mangalaraj, S.I. Hong, Y. Masuda, Synthesis and in-depth [80] B.K. Mahmood, O. Kaygili, N. Bulut, S.V. Dorozhkin, T. Ates, et al., Effects of
analysis of highly ordered yttrium doped hydroxyapatite nanorods prepared by strontium - erbium co-doping on the structural properties of hydroxyapatite: an
hydrothermal method and its mechanical analysis, Mater. Charact. 62 (12) (2011) experimental and theoretical study, Ceram. Int. 46 (10) (2020) 16354–16363,
1109–1115, https://doi.org/10.1016/j.matchar.2011.09.008. https://doi.org/10.1016/j.ceramint.2020.03.194.
[76] M. Sato, A. Nakahira, Influence of Fe addition to hydroxyapatite by hydrothermal [81] Z.-H. Dai, Y.-P. Xie, Y.-C. Qian, L.-J. Hu, X.-D. Li, H.-T. Ma, Effects of P, as, and Sb
process, J. Ceram. Soc. Japan 121 (1415) (2013) 559–562, https://doi.org/ Heavy Doping on Band Gap Narrowing of Germanium as Light-Emitting Materials,
10.2109/jcersj2.121.559. arXiv Condens, Matter. Mater. Sci. http://arxiv.org/abs/1706.04050, 2017.

14