Influence of Magnesium Content on the Physico-Chemical Properties of Hydroxyapatite Electrochemically Deposited on a Nanostructured Titanium Surface

Abstract

The aim of this research was to obtain hydroxyapatite (HAp)-based coatings doped with different concentrations of Mg on a Ti nanostructured surface through electrochemical techniques and to evaluate the influence of Mg content on the properties of HAp. The undoped and doped HAp-based coatings were electrochemically deposited in galvanostatic pulsed mode on titania nanotubes with a diameter of ~72 nm, being designed to enhance the adhesion of the HAp coatings to the Ti substrate. The obtained materials were investigated by Scanning Electron Microscopy (SEM), Energy Dispersive Spectroscopy (EDS), X-Ray Diffraction (XRD), and Fourier-Transform Infra-Red spectroscopy (FTIR). The adhesion of the coatings to the substrate was also evaluated with the help of the “tape-test” and the micro-scratch test. The morphology (SEM) of all the coatings is made of very thin and narrow ribbon-like crystals, with some alterations with respect to the Mg amount in the coatings. Thus, a concentration of 1 mM of Mg in the electrolyte leads to wider and thicker ribbon-like crystals, while a concentration of 1.5 mM in the electrolyte generated a morphology that resembles the undoped HAp. Both phase composition (XRD) and chemical bonds (FTIR) analysis proved the formation of HAp in all coatings. Moreover, according to XRD, all coatings have a strong orientation toward the (002) plane. Irrespective of the Mg content, all coatings registered an average roughness between approx. 500 and 600 nm, while the coating thickness increased after addition of Mg, from a value of 9.6 μm, for the undoped HAp, to 11.3 μm and ~13.7 μm for H/Mg1 and H/Mg2, respectively. In terms of adhesion, it was shown that the coatings a H/Mg2 had a poorer adhesion when compared to H/Mg1 and the undoped HAp (H), which registered similar adhesion, indicating that a concentration of 1.5 mM of Mg in the electrolyte reduces the adhesion of the Hap-based coatings to the nanostructured surface. The obtained results indicated that Mg concentrations up to 1 mM in the electrolyte can enhance the properties of HAp-based coatings electrochemically deposited on a nanostructured surface, while even a slightly higher concentration of 1.5 mM can negatively impact the characteristics of HAp coatings.

1. Introduction

Metallic materials have been used to replace and reconstruct the skeletal functions of the human body for a very long time [1,2]. Due to their high strength, toughness, and durability, approximately 70–80% of the implantable materials are made of metallic biomaterials [3,4,5]. The need for metallic biomaterials with proper features and characteristics is rapidly growing as the world is progressively aging, and elderly people tend to present a higher risk of hard tissue failure [6,7].

Within this context, titanium (Ti) and its alloys are often the material of choice due to a favorable combination of properties such as excellent corrosion resistance which offers long-term stability, high biocompatibility, and reliable mechanical strength, with minimal long-term toxicity to the host [8,9]. Nevertheless, one weakness of the implantable metallic biomaterials is given by their no bio-function recognized as bioinert character. For metallic biomaterials such as the class of Ti-based alloys, imparting multifunctionality on its bioinert character is generally carried out by surface modification techniques, such as surface structuring and/or coating with bioactive ceramics and polymers [10,11].

Generally, the aim of the material’s coating is to protect or to enhance its functionality, and it is certainly not a new concept [12]. In medicine, the coatings field is rapidly diversifying, aiming to improve osseointegration and to tackle the bioinert character of the metallic implants [13,14].

Hydroxyapatite (HAp, Ca₁₀(PO₄)₆(OH)₂) is the main inorganic phase of the hard tissue with a Ca/P ratio of 1.67 [15,16,17] that has been extensively studied and used as biomaterial in the past decades [18]. The extensive research conducted in this period highlighted that HAp is brittle and not appropriate for load-bearing applications, such as large bone defects [19,20,21], but it can be successfully used for other types of treatments such as bone substitutes for small bone defects [22], as scaffolds along with other materials such as polymers [23,24], and also as coatings [25,26] to control and accelerate osseointegration [27,28].

The features of HAp such as osteointegration capacity, mechanical properties, and bioactivity can be enhanced through the addition of small amounts of elements that are found in natural bones and tooth mineral [29,30].

Magnesium (Mg) is a cation present in the human body, which despite its low concentration (between 0.5 and 1.5 wt.%) promotes bone cells activation and proliferation. It influences bone strength, mineralization, and growth [31,32], and it plays an important role in the changes in the bone matrix that determine bone fragility [33]. In terms of applicability, Mg can be used as pure metal and/or alloys [34,35], but also as a doping element [36,37] in HAp or other types of CaP [38].

According to Tampieri et al. [39] and Ren et al. [40], Mg²⁺ prefers the Ca (I) site of the hydroxyapatite structure, while Kannan et al. [41] have shown that Mg²⁺ is essentially accommodated in the Ca (II) site. Laurencin et. al. [42] found that at 10% Mg concentration, Mg ions will replace Ca in the Ca (II) site. Therefore, a possible explanation would be that the place of Mg substitution depends on the preparation method and on the amount of Mg, indicating that different methods lead to different results.

In terms of biocompatibility and bioactivity characteristics, research studies have highlighted that Mg addition into HAp has a beneficial effect [38,43,44,45]. Chirica et al. [46] investigated the physico-chemical properties of HAp doped with various elements, including Mg, and showed that Mg favors the initial cell adhesion. Chen et al. [47] reported that Mg²⁺ ions in HAp enhance cell attachment, proliferation, and differentiation of cells to an osteoblast phenotype, as well as the expression of osteogenesis-related markers. These findings are also supported by the results of He et al. [48], who have shown that Mg-substituted HAp enhances the differentiation of stem cells into osteoblasts. In a study carried out by Jiao et al. [49] regarding the effect of Mg–HAp coatings with various Mg concentrations, it was shown that an Mg concentration up to 2 wt.% provided better biocompatibility. Thus, it can be stated that special attention must be taken regarding toward the proper selection of the Mg concentration since at a high concentration the HAp properties can deteriorate [50,51].

Hydroxyapatite-based coatings can be synthesized through different techniques such as a biomimetic method [52], plasma spraying [37,53], sputtering deposition [54,55], sol–gel process [56], laser deposition [57,58], and electrochemical deposition [59]. Among these techniques, electrochemical deposition (ED) appears to be more suitable to develop HAp-based coatings due to its simplicity and good control over the involved electrochemical parameters, which dictates the coatings properties and behavior. It is also more accessible in terms of costs and can be employed to coat complex shapes because it is not a line-of-sight technique [59,60,61].

Mg-substituted HAp is most often obtained by an aqueous precipitation method [62,63] and a hydrothermal method [64]. In our previous work [65], we have shown that Mg-doped HAp coatings can also be successfully obtained by pulsed galvanostatic mode on Ti6Al4V and that a small amount of Mg has a beneficial effect on the HAp properties. Among the drawbacks of the ED technique, when employed to obtain HAp-based coatings, there is unsatisfactory adhesion of the coating to the substrate. To overcome this vulnerability and to enhance the adhesion of the HAp-based coatings on the Ti substrate, an intermediate layer consisting of titania (TiO₂) nanotubes (NT) can be developed via anodic oxidation [66,67]. Growth of HAp inside the hollow nanotubes would give an anchoring effect to the electrodeposit and enhance the interfacial bond strength at the TiO₂ layer-HAp coating interface [68].

It is known that compared to bare Ti, the TiO₂ NT surface provides an optimum biological environment for bone tissue through the ingrowth of mineralized tissue into the pore space and consequently improved osseointegration [69,70,71,72]. The biocompatibility of the TiO₂ nanotube was confirmed by subcutaneous injection of the TiO₂ nanotube into rats. In this case, the TiO₂ nanotube did not cause chronic inflammation or fibrosis [73]. By analyzing the histological features and fluorochrome labeling changes after implantation, Wang et al. [74] showed a notable increase in bone implant contact and gene expression levels in the bone attached to TiO₂ NT, while von Willmowski et al. [75] reported that osteocalcin expression was increased the most in TiO₂ nanotubes with a diameter of 70 nm when different diameters (15–100 nm) were implanted in a pig skull.

In the past decades, four generations of electrolytes have been used to obtain TiO₂ NT on titanium and its alloys, of which the first three are fluorine-based [76]. Fluorine (F⁻) is the element responsible for the local etching of the Ti surface by using an inside-bottom approach. As a result, of this process, some residual fluorine can remain in the nanotubes, which can be tailored to enhance osseointegration but can also be cytotoxic [77,78]. The residual amount of fluorine in the nanotubes can be reduced by an annealing treatment, which is also used to enhance the nanotubes crystallinity [79,80].

On the influence of fluorine on the cell’s behavior, it can be said that fluorine can quickly enter cell membranes by diffusion, and it can essentially influence every stage of metabolism for humans [77,81]. Li et al. [82], showed that a content of 1.21 wt.% (1.88 at.%) fluorine in an implant surface can enhance bone fixation, while Isa et al. [83] showed that a fluoride-modified surface (titanium fluoride 0.5–3 wt.%) supports and promotes cell proliferation and differentiation better than non-fluoride surfaces. This is also supported by in vivo studies [84,85,86], which have shown that the modified biochemistry of implant surfaces may facilitate osseointegration and that the fluoride presence may be beneficial. Other studies have implied that most surfaces, including polystyrene and TiO₂ with fluorine on the surface for bone, endothelial, and smooth muscle cells, showed increased proliferation and differentiation rates up to a threshold higher than those surfaces without fluorine [84,87,88].

In terms of antibacterial efficiency, some reports have found that fluorine can enhance bacterial adhesion on the surface of biomaterials [89,90], while others have reported that fluorine shows antibacterial ability [91,92]. Studies also indicate that nanotubes with 60 or 80 nm in diameter decrease the number of live bacteria compared to smaller ones (20 or 40 nm) [79,80].

Thus, based on the literature, it can be assumed that fluorine is relatively biologically safe, and to some extent it can enhance osseointegration and offer antibacterial efficiency. Based on this, it can be assumed that to achieve a proper antibacterial efficiency and cellular response, the nanotubes diameter and the fluorine content must be properly selected.

Therefore, by coating the NT TiO₂ surface with a layer of HAp, the in vitro behavior of the Ti surface can be further increased. In a study performed by Zhang et al. [93], it was shown that the addition of a HAp-based coating on the NT TiO₂ surface was more favorable for cellular activities compared to bare Ti and TiO₂ nanotubes alone. This is in accordance with the study performed by Fathyunes et al. [94], in which it was shown that compared to the uncoated Ti, the HAp-based coatings electrochemically deposited on NT TiO₂ have enhanced the cell viability, indicating a proper biocompatibility.

The objective of the current work consisted of obtaining a uniform coating layer of undoped and doped HAp with Mg in different concentrations on titania nanotubes through electrochemical deposition in a pulsed galvanostatic mode and in evaluating the influence of the Mg amount on the physico-chemical properties of HAp. To the best of our knowledge, only a few studies [66,95] have reported on Mg-doped HAp on titania nanotubes through electrochemical techniques. Both papers [66,95] reported that by using an HF-based electrolyte and a constant voltage of 20 V, titania nanotubes with diameters of 100 nm can be obtained, if the anodization time is 60 min [66], while by reducing the time to 30 min, nanotubes with a diameter of 60 nm can be obtained [95].

In terms of coatings, Yajing et al. [66] successfully obtained Mg-doped HAp on titania nanotubes from an electrolyte which contained 0.084 mol/L Mg, by applying a constant current density of −0.85 mA/cm² (galvanostatic technique), while the electrolyte temperature was set to 65 °C and the pH value was 4.2 (adjusted with ammonia solution). Their findings showed that in the proposed concentration, the morphology of the Mg–HAp coatings became plate-like instead of needle-like, as identified for the undoped HAp, while in terms of cell culture, the addition of Mg²⁺ had no cytotoxic effect, suggesting that the Mg–HAp coatings obtained on titania nanotubes showed potential in satisfying the requirements of biomedical applications. In the study conducted by Shahmohammadi et al. [95], the Ti-nanostructured surface was coated with calcium phosphates (CaP) co-doped with Mg and Zn in different concentrations, including only Mg-doped HAp and Zn-doped HAp, by using two stage deposition, namely a pulsed voltage technique (a potential of 0 V and −2 V was applied for 100 cycles), followed by a deposition under constant voltage [95]. Their findings showed that the dominant phases of the obtained CaP coatings were brushite and β-tricalcium phosphate and that the Mg–HAp coatings had the highest protein absorption, while the Zn–HAp coatings had the highest antibacterial effect [95].

The novelty of this study is the fact that even though the proposed electrochemical techniques, namely electrochemical deposition and anodic oxidation, are well known for obtaining undoped and doped HAp-based coating and nanostructured surfaces, few literature reports are available on the influence of Mg content on the properties of HAp electrochemically deposited on titania nanotubes. Thus, compared to previous reports [66,95], in this study, the coatings were obtained by the pulsed galvanostatic method, and by addressing and evaluating the influence of the Mg concentration on the physicochemical properties of HAp, with respect to the morphology, chemical and phasic composition, chemical bonds, layer thickness, roughness, and adhesion, a new insight can be revealed.

2. Materials and Methods

2.1. Sample Preparation

As raw material, commercially pure titanium specimens cut into discs of 20 mm diameter and 2 mm thickness (cp-Ti, grade 2, purchased from Bibus Metals AG, Essen, Germany) were used. The samples have been metallographically prepared on SiC papers of different grits (300–1200) and polished with alumina (Al₂O₃) slurry with a particle dimension of 1 μm. All polished samples had an average roughness of 82 nm (±5 nm). The samples were ultrasonically cleaned in acetone for 20 min and washed with ultrapure water (ASTM I), to eliminate the unwanted residues.

2.2. Biofunctionalization of cp-Ti

Before the electrochemical deposition of the HAp-based coatings doped with Mg in different concentrations, the cp-Ti surface was modified through anodic oxidation in 0.5 wt.% HF solution by applying a constant voltage of 20 V for 30 min with a DC power supply system (model N5771A, Keysight, Böblingen, Germany). The anodized cp-Ti surfaces were annealed at 450 °C for 2 h in air, using a heat treatment furnace (N17/HR model, Nabertherm, Lilienthal, Germany) [96].

The annealed nanostructured surfaces (NT) were further biofunctionalized with hydroxyapatite (HAp)-based coatings undoped and doped with Mg in different concentrations. The electrochemical deposition was carried out in a conventional three electrode electrochemical cell configured as follows. The working electrode (WE) was the cp-Ti nanostructured surface with an exposed area of ~2 cm², the reference electrode (RE) was a calomel electrode, and the counter electrode (CE) was a platinum foil with a surface area of approximately 1 cm². The deposition was controlled using a multichannel Potentiostat/Galvanostat (Parstat MC, Princeton Applied Research–Ametek, Oak Ridge, TN, USA) by applying the pulsed galvanostatic technique. A total of 1200 cycles were applied to obtain the HAp-based coatings undoped and doped with Mg. One cycle has an activation stage in which the applied current density (i_ON) was of −0.85 mA/cm² for a period (t_ON) of 1 s, followed by a relaxation stage in which the applied current density (i_OFF) was set to 0 mA/cm² for a period (t_OFF) of 1 s. During the electrochemical deposition, the electrolyte was maintained at a constant temperature of 75 °C (±0.5 °C), and it was continuously stirred at a speed of 50 rpm using a basic magnetic hotplate stirrer (KA RCT Basic Safety Control Hotplate/Stirrer and ETS-D6 Temp, IKA, Staufen, Germany).

All chemicals were purchased from Sigma Aldrich (Munich, Germany) and are of high purity. The electrolytes pH was adjusted to 5 by dropwise addition of 1 M NaOH, and it was purged with nitrogen gas (N₂) for 20 min to reduce the formation of CaCO₃ deposits. The deposition was carried out at a temperature of 75 °C, which was maintained constant with a heating plate (KA RCT Basic Safety Control Hotplate/Stirrer and ETS-D6 Temp, IKA, Staufen, Germany), while a magnetic stirrer was used to keep the concentration homogeneous. After the electrochemical deposition, the samples were removed from the electrolyte and gently rinsed with ultra-pure water. In

Table 1. Samples codification.

	Surface Treatment	Material	Sample Codification
1	Anodic oxidation	cp-Ti with nanostructured surface consisting in titania nanotubes	NT
2	Electrochemical deposition in pulsed galvanostatic mode	Functionalization of NT with undoped hydroxyapatite coatings	H
3		Functionalization of NT with Mg (concentration 1) doped hydroxyapatite coatings	H/Mg1
4		Functionalization of NT with Mg (concentration 2) doped hydroxyapatite coatings	H/Mg2

, the samples codification is presented, while in

Table 2. Main electrochemical parameters involved in the deposition process.

1 cycle	Activation	iON = −0.85 mA/cm2, tON = 1 s
	Relaxation	iOFF = 0.00 mA/cm2, tOFF = 1 s
Cycles number		1200
Temperature (°C)		75 °C

and

Table 3. Chemical composition of the electrolyte.

Samples	Chemical Composition (mM)			(Ca + Mg)/P Ratio
	Ca(NO3)2∙4H2O	NH4H2PO4	Mg(NO3)2∙6H2O
H	10	6 mM	0	(Ca + Mg)/P = 1.67
H/Mg1	9		1
H/Mg2	8.5		1.5

, the main electrochemical parameters and the chemical composition of the electrolytes employed in the deposition process are presented.

2.3. Characterization

The nanostructured surface of cp-Ti, consisting of titania nanotubes (NT), was analyzed in terms of morphology with a field-emission scanning electron microscope (Scienta Omicron NanoSAM Lab system, Oxford Instruments, High Wycombe, UK), while the surface morphology and elemental composition of the HAp coatings were investigated with a Scanning Electron Microscope equipped with an X-Ray Energy Dispersive Spectrometer (SEM-EDS, Phenom ProX, Phenom World, Eindhoven, The Netherlands).

The phase composition of the samples was studied by X-Ray diffraction (Rigaku SmartLab, Tokyo, Japan) using an X-ray source of CuKα (λ = 1.541 Å). Grazing incidence measurements were performed on the θ/2θ range of 20–80° with a step of 0.01°/min at an incident angle of 3°.

The average crystallite size for the (002) diffraction plane, which is the most intense diffraction peak in HAp coatings obtained by electrochemical deposition [97], was estimated by the Debye–Scherrer mathematical relation (Equation (1)):

$$L_{\left(002\right)}=\frac{k\cdot \lambda }{\beta \cdot cos{\theta }_{\left(002\right)}}$$

(1)

where k is the constant shape factor for HAp with hexagonal structure, which is equal to 0.9, β is the full width at half maximum (FWHM) of (002) diffraction plane, θ in degrees indicates the Braggs diffraction angle, and λ = 1.5406 Å is the wavelength of CuKα radiation.

In the literature, several types of methods can be used to estimate the crystallinity (χ_c) [98]. In the present work, the crystallinity was estimated using the following equation (Equation (2)) [99]:

$${\chi }_c={\left(\frac{K_A}{\beta }\right)}^3$$

(2)

where K_A is a constant found equal to 0.24 for HAp, and β is the FWHM of refection (002) in degrees.

To evaluate the structural differences induced by the addition of Mg into the HAp structure, the lattice parameters were calculated using Equation (3), which involves the relation between the lattice constants (a and c) and the interplanar spacing distance (d_hkl) for the hexagonal crystallographic system [100], as follows:

$$\frac{1}{d_{hkl}^2}=\frac{4\cdot \left(h^2+k^2+l^2\right)}{3a^2}+\frac{1}{c}$$

(3)

The chemical bonds were analyzed by Fourier Transform Infra-Red (FTIR) spectroscopy in attenuated total reflectance (ATR) mode. The FTIR spectra were registered in the 4000–6000 cm⁻¹ wavenumber range using a FT-IR Jasco 6300 (Jasco, Tokyo, Japan) with the ATR sampling accessory Pike MIR-acle (Pike Technologies, Madison, WI, USA).

A stylus profilometer (DEKTAK 150, Veeco Instruments, Plainview, NY, USA) was used to evaluate the surface roughness of the coatings, which was measured with over a length of 3 mm, while to determine the coating thickness, the level difference between the deposited film and the uncoated substrate was measured.

The adhesion of HAp coatings was investigated according to ASTM D 3359-17 standard (Standard Test Methods for Measuring Adhesion by Tape Test) [101], using an Elcometer 107 Cross Hatch Adhesion Tester kit (Ulmer, Aalen, Germany). Parallel cuts of 6 × 6 mm with gaps of 1 mm between them were made using a blade with cutting edges. Thus, a lattice pattern was generated. Over the lattice pattern indentation, the ASTM adhesive tape was placed. After 90 s, the tape was removed by pulling in a single smooth action at angle of 180° to the surface. The lattice pattern indentation was examined macroscopically and by scanning electron microscopy after the adhesion tests. In accordance with the ASTM D 3359-17 standard, the coating adhesion was assessed in terms of area removed, and it was classified in terms of percentages as follows: 5B—0%; 4B—≤5%; 3B—5–15%; 2B—15–35; 1B—35–65%; and 0B—≥65%.

Complementary to this technique, a micro-scratch test was performed under similar conditions as the ones presented by Mokabber et al. [102]. The adhesion strength of the HAp-based coatings was performed with a TriboLab UMT tester, Bruker Ltd. (Billerica, MA, USA), using a diamond indenter with a radius of 0.2 mm, working in progressive load from 0.5 to 35 N, with a scratching speed of 1.7 mm/min over a length of 4 mm. The resulting scratches were analyzed by SEM, and the typical scratch curves obtained from the micro-scratch tests were plotted.

3. Results and Discussion

3.1. Morphology

Figure 1 depicts the nanostructured surface after annealing, which consists of hollow TiO₂ nanotubes, uniformly dispersed on the surface with an inner diameter of ~72 nm (±2 nm).

The morphology of the HAp-based coatings undoped and doped with Mg electrochemically deposited on the nanostructured surface present a morphology (Figure 2a–i) that consists of ribbon-like crystals that cover the entire surface exposed to the electrolyte during the electrochemical deposition.

The addition of Mg in different concentrations into the HAp-based coatings altered the dimensions of the ribbon-like crystals as follows. For the H/Mg1 coatings, the ribbon-like crystals became wider and thicker (Figure 2f), which by increment of the Mg content in the case of H/Mg2 coatings led to the formation of a coating which appears to be denser (Figure 2i) compared to other two types of coatings with crystals that tend to resemble those of the undoped HAp (Figure 2c). Thus, based on these morphological observations, it can be stated that the Mg concentration influences the morphology and the crystal density of the HAp-based coatings.

It is well-known that deposition of CaP species on metallic substrates occurs through the nucleation and growth mechanism. Thus, in electrochemical deposition, by applying electric current through electrodes, the concentration of hydroxyl ions increases in the vicinity of the cathode, leading to a local increase of pH whose result is the formation of calcium phosphate nuclei on the metallic cathode and grow to form HAp crystals [60,103].

In terms of the growth mechanism of the HAp-based coatings obtained by electrochemical deposition, Figure 3 shows that the coatings consist of two layers of different morphologies. The first layer, which was formed within the first period of the deposition, is more compact and is not very thick, being constituted from plate-like crystals, while the second one is made of ribbon-like crystals, highly elongated along the c-axis.

Mokabber et al. [60] showed that in the first minute of deposition, nano-plates without any specific direction are formed, which after 3 min grow preferentially on the b and c-axes, forming the plate-like morphology. By increasing the deposition time to 30 min., the plate-like morphology becomes highly elongated along the c-axis, forming ribbon-like crystal, indicating that that the crystal growth behavior of CaP species on metallic substrates is greatly affected by the supersaturation of the electrolyte and that it is a time-dependent process. Similar findings were also noted by dos Santos et al. [104], who showed that in the first 15 m plate-like crystals are formed, which tend to elongate merely along the c axis direction and grow perpendicular on the substrate.

These observations are also in good correlation with deposition cycles in which during the t_ON period, the process starts. Then, in the close vicinity of the working electrode, the concentration of the ion species is diminished, while in the relaxation period (t_OFF), the ions diffuse toward the electrode and recover the electrolyte concentration [59,105]. In addition, the preferred orientation of the HAp-based layer is also influenced by the substrate orientation [106] and by some of the electrochemical deposition conditions [103]. Thus, even a slight agitation of the electrolyte suggests that the deposition process is driven by the electric (Laplacian) and concentration (diffusion) fields [103,107].

Thus, it can be assumed that several parameters involved in the deposition process strongly influence the HAp layer properties, and a proper selection can lead to the development of materials with enhanced behavior and suitable characteristics.

3.2. Elemental Composition

The elemental composition of the coatings was obtained by EDS analysis, and the results are presented in Figure 4 as EDS spectra (a), element distribution on the investigated area (b), and elemental composition and (Ca + Mg)/P ratio (c). The EDS analysis was carried out in triplicate by using different specimens of each material, and the average values along with their standard deviation were taken into consideration.

The EDS spectra achieved on the HAp-based coatings have highlighted the presence of the main elements of the nanostructured surface (Ti and O) and in the coatings (Ca, P, O, and Mg). Moreover, the elemental composition carried out on the annealed nanostructured NT TiO₂ surface (results not shown) revealed the presence of fluorine (F⁻) in a concentration of 0.65 wt.% (0.92 at.%). According to the literature, a concentration between 0.5 at% and 3 at.% supports and promotes cell proliferation and differentiation [83]. The residual fluorine found on the NT TiO₂ was not further detected in the coatings. Nevertheless, it is worth mentioning that the concentration detected on the nanostructure surfaces is within acceptable limits (0.5–3 at.%).

Since Mg²⁺ ions are substituting the Ca²⁺ one in the HAp structure [44], the Mg amount was added to Ca, and the resulting quantity was divided to the P amount.

Thus, the addition of Mg into different concentrations slightly decreased the (Ca + Mg)/P ratio from a value of 1.59, obtained for the H coatings, to a value of 1.57 and 1.56, respectively, for H/Mg1 and H/Mg2 (Figure 4c). Nevertheless, it can be noted that all (Ca + Mg)/P ratios are smaller than that of stoichiometric HAp, indicating the formation of a Ca-deficient apatite.

In terms of element distribution, all the identified elements (Ca, P, and Mg) are uniformly distributed on the investigated areas, with no visible agglomerations, proving a homogeneous coating.

3.3. Phase Composition

In Figure 5a, the X-ray diffractograms of the undoped and Mg-doped HAp-based coatings are presented along with the computed values of the lattice parameters, the crystallite dimension, and the crystallinity (Figure 5b).

The XRD diffractogram of the HAp-based coatings shows the main diffraction peaks related to the HAp phase according to ICDD card #09-0432. In all the coatings, the presence and the formation of hydroxyapatite as the main phase were found. Two other small diffraction peaks identified at 2θ of approximately 24° and 30°, corresponding to (111) and (120) planes, were attributed to the monetite phase according to ICDD card #1-071-1759, indicating the presence of a secondary phase.

The diffraction peaks corresponding to pure Ti (ICDD card #44-1294) as bulk substrate material and the nanostructured surface made of TiO₂ nanotubes with its two phases, namely anatase (ICDD card #21-1272) and rutile (ICDD card #21-1276), were also identified, suggesting that the coatings are to some extent porous due to the presence of the ribbon-like crystals. The diffractogram of the NT surface is available in Ref. [96].

Irrespective of the absence or presence of Mg within the coatings and/or its amount, all diffractograms present the same allure and intensities, with just a few differences, namely the intensity of the peaks associated to the monetite phase which decreased after Mg addition and the peak of HAp identified at ~26°.

The diffraction peak of HAp corresponding to (002) the plane at 2θ of ~26° is the most intense and illustrates that the obtained coatings have a preferential orientation toward the c-axis which is in good agreement with the SEM images.

According to the literature [108,109], the change in crystal growth direction could be due to the incorporation of mineral ions and the increase in their content which may favor the growth of crystals in the (002) direction, which is perpendicular to the substrate surface (along the c-axis). Moreover, if the crystals were all perfectly aligned with the c-axis normal to the substrate, the (002) diffraction peak would be the only one visible. The fact that other HAp peaks are present in diffractograms (Figure 5a) indicates that some crystals are oriented in other directions [110].

Thus, to identify the preferential orientation of the crystallites along a crystal plane (hkl) in the samples, the texture coefficient TC_(hkl) of each peak was calculated following the relation [111,112]:

$$T{C}_{\left(hkl\right)}=\frac{\frac{I_{hkl}}{I_{0,hkl}}}{\frac{1}{N}{{\displaystyle \sum }}_0^n\frac{I_{hkl}}{I_{0,hkl}}}$$

(4)

where TC_(hkl) is the texture coefficient of plane (hkl), I_hkl is the measured intensity, I_0,hkl values are the intensity taken from the ICDD powder diffraction file of hydroxyapatite (ICDD #09-0432), and N is the number of peaks taken into consideration.

Thus, the (002) diffraction peak and three strongest diffraction peaks of HAp ((211), (112) and (213)) were selected; therefore n = 4. A random orientation would lead to a TC_(hkl) value of 1, whereas a complete dominance of one peak would return a value of 4, which represents the number of diffraction peaks taken into consideration.

The degree of orientation of each coating was also analyzed from the standard deviation σ of all TC_(hkl) values, being calculated based on the following relation [112]:

$$\sigma =\sqrt{\frac{1}{N}\sum {\left(T{C}_{\left(hkl\right)}-1\right)}^2}$$

(5)

The value of σ is an indicator of the orientation degree, and it can be used to compare different samples.

Table 4. Texture coefficients and degree of preferred orientation for the undoped and Mg-doped HAp coatings.

Coating	TC(hkl)				Degree of Preferred Orientation σ
	TC(002)	TC(211)	TC(112)	TC(213)
H	2.96	0.21	0.35	0.48	1.14
H/Mg1	3.23	0.15	0.25	0.36	1.29
H/Mg2	3.48	0.10	0.17	0.26	1.43

presents the values for the texture coefficient and degree of the preferred orientation.

Thus, for a complete randomized sample, the TC_(hkl) will be 1, and the preferred orientation σ will be 0, while for a fully aligned material, the TC_(hkl) will be 4 and the preferred orientation σ will be 1.73.

By comparing the values of the texture coefficient of all samples, it can be noted that compared to the undoped HAp, the addition and the increment of Mg concentration have led to the increment of the TC_(hkl) for the (002) plane and to the decrement of the TC_(hkl) for (211), (112) and (213) planes. Thus, the data presented in Table 4 indicate that the preferred orientation is in the (002) direction for all samples, the highest values being obtained for the H/Mg2 samples, followed by the H/Mg1 and H.

Another change that involves the diffraction peak at 26° consists of shifting the HAp main XRD peak toward the right, denoting the inclusion of the doping elements in the HAp structure. This is available when the ionic radius of the doping element is smaller than the one of Ca²⁺, which is also the case for Mg²⁺. By increasing the Mg amount, the XRD peaks became more intense, and some of the peak’s position shifted to the right, indicating lattice contraction caused by the Mg substitution. This effect could be explained by the increment of the crystallite size and lattice changes associated with Mg substitution in the HAp structure. As can be observed in Figure 5b(i,ii), compared to H coatings, the addition of Mg led to a slight decrement of the a parameter and an increment of the c parameter along with a higher crystallite dimension.

According to the ISO 13779-3 standard [113], the percentage of crystallinity should be a minimum of 45% for HAp. Thus, it can be observed that the coatings with Mg, codified as H/Mg1 and H/Mg2, are slightly above 45%, reaching values of approx. 47% and 49%, respectively, while the H coating registered a crystallinity of only ~21%. Overall, it was noted that the addition of Mg has a beneficial effect on the crystallinity of HAp-based coatings.

Compared to our previous research in which HAp-doped Mg was electrochemically deposited on a Ti6Al4V alloy [65], in the current research, the electrolyte was slightly adjusted by reducing the concentration of all chemical reagents involved by 5%, coupled with the use of pure Ti with a nanostructured surface as substrate instead of Ti6Al4V, could be the cause that led to higher crystallinity. Thus, it can be assumed that even the slightest change of the electrochemical parameters can affect the characteristics of the deposited materials. Nonetheless, further analyses in this direction are necessary to validate this assumption.

3.4. Chemical Bonds

Figure 6 depicts the FTIR spectra of the undoped and Mg-doped HAp coatings. FTIR spectra are dominated by internal (PO₄)³⁻ modes, which typically show a strong molecular character with respect to their vibrational properties [114]. Nonetheless, some variations of the absorption bands, which may indicate the substitutions of Mg ions into the HAp structure, are visible.

The spectrum of HAp shows intense peaks between 1120 and 1020 cm⁻¹, which were assigned to the ν₃ asymmetric stretching mode of (PO₄)³⁻ [115], while the weak peak at 960 cm⁻¹ can be assigned to the ν₁ asymmetric stretching mode of (PO4)³⁻ [65].

Compared with undoped HAp, a decrement in the intensity of the phosphate groups (PO₄)³⁻ was associated with the addition and increment of magnesium content. The observed FTIR spectral changes can also be related to the destabilization of the HAp structure by the incorporation of Mg²⁺ ions.

Nonetheless, the OH⁻ stretching vibration identified at 630 cm⁻¹ is unique and indicates that the HAp-based coatings are crystallized [115]. The adsorbed water generates a broad band in the range of 3000–3500 cm⁻¹ [116] and at about 1650 cm⁻¹ [115,117].

Thus, all developed coatings present the peaks assigned to PO₄³⁻ and OH⁻, which are considered specific to HAp, proving the formation of the HAp phase.

3.5. Surface Roughness

Despite the high number of studies, the results in the literature demonstrate difficulties in deciding the optimum value of surface roughness for better osseointegration and decreased bacterial adhesion [118,119,120].

In Figure 7, specific profiles of the nanostructured surface (NT), uncoated and coated with HAp-based coatings, are presented, while in

Table 5. Main roughness parameters and coating thickness.

	NT	H	H/Mg1	H/Mg2
Ra (nm)	132.2 (±9.1)	535.6 (±22.8)	582.6 (±34.6)	493.5 (±61.4)
Rq(nm)	154.5 (±12.9)	689.8 (±49.3)	741.9 (±49.4)	644.7 (±59.6)
Coating Thickness (μm)	-	9.6 (±0.9)	11.3 (±0.6)	13.7 (±1.1)

the values of the main roughness parameters along with the coating thicknesses are given.

A comparison of the average roughness (Ra) parameter for all materials indicates that roughness ranged between 490 and 590 nm, with the smallest value being registered for the coatings H/Mg2, showing that high concentrations of Mg (i.e., 1.5 mM Mg in the electrolyte) tend to smoothen the surface. The largest Ra value, of 583 nm was observed for the H/Mg1 coatings, while the H coatings had an Ra of 535 nm. Thus, based on the results of the Ra parameter, it can be noted that the amount of Mg has a direct influence on the coatings roughness.

The coatings thickness was measured, and it was noted that the smallest value of 9.6 μm (±0.9) was obtained for the undoped H followed by H/Mg1 coatings with a thickness of 11.3 μm (±0.6), while the largest one was attributed to the H/Mg2 coating of 13.7 μm (±1.1).

Thus, it can be said that the addition of Mg into HAp increases the deposition rate, leading to higher thicknesses, while in terms of roughness, a higher concentration of Mg leads to a decrement of the average roughness.

3.6. Adhesion

One of the main disadvantages of the HAp-based coatings obtained by electrochemical deposition is their adhesion to the substrate. Thus, coating adhesion to the substrate is very important for the implantable medical devices, and at times, it can be quite challenging to achieve.

The SEM images of lattice pattern indentation after adhesion tests were achieved on areas of 897 × 897 μm², presented in Figure 8. Furthermore, the SEM images were analyzed by Image J software to quantify the delaminated areas in percentages.

After detaching the scotch tape from the coating surface, a very thin layer of the coating remained on the tape but without affecting the integrity of coatings which is in good agreement with other studies [121].

A comparison of the lattice patterns highlighted that the best adhesion was noted for the H and H/Mg1 coatings, which had very small areas detached (0.31% and 0.44%, respectively), while the H/Mg2 coatings showed the lowest adhesion with large detachments/flakes from their surfaces, which falls into category 3B (5–15% delamination area) with an area of 8% that was detached from the surface. However, it can be observed that on the detached areas of the H/Mg2 sample, a layer of apatite that resembles the bottom layer of HAp remained on the substrate, indicating that the coating was not entirely detached from the surface. Thus, the results of adhesion tests showed that Mg–HAp coatings obtained from an electrolyte that contains at least 1.5 mM Mg can alter the adhesion of the coatings to the substrate.

Complementary to the tape test, we have also performed a micro-scratch test, which showed that the initial penetration of the coating starts within the first mm of the scratch length (Figure 9).

Based on the signal captured by the acoustic emission sensor, the critical load (Lc) was estimated. The highest Lc was registered for the undoped HAp with a value of 4.33 N, closely followed by the H/Mg1 sample with an Lc of 4.15 N, while the smallest value of 2.30 N was obtained for the H/Mg2 sample which is in agreement with the results obtained from the tape test and presented in Figure 8.

Moreover, the elemental composition after the micro-scratch tests was investigated in the center of the scratch and at its end (Figure 10) to identify if some traces of the coatings were still present.

The results presented in

Table 6. Chemical composition within the scratches imprint of all investigated coatings.

Sample	EDS Mapping on Area inside the Scratch	Concentration (at. %)
		Ca	P	O	Ti
H	Area 1 *	2.43	2.23	41.45	Balance
	Area 2 **	1.91	1.78	39.38	Balance
H/Mg1	Area 1 *	2.23	2.02	40.39	Balance
	Area 2 **	1.91	1.84	38.16	Balance
H/Mg2	Area 1 *	2.01	1.9	39.6	Balance
	Area 2 **	1.76	1.68	38.88	Balance

* Area 1—in the middle of the scratch blue square in Figure 10; ** Area 2—at the end of the scratch, dark orange square in Figure 10.

highlight that the main elements of hydroxyapatite (Ca, P and O) are found in all the investigated areas. However, oxygen is also detected from the titanium dioxide, which occurs instantly on the surface of titanium.

It can also be seen that for all tested HAp coatings, the amount of calcium and phosphorous content decreases at the end compared with the middle of the scratch imprint. The presence of CaP traces within the scratch imprint is beneficial in the case of some damage of the coatings due to friction with hard tissue or careless handling of medical devices.

4. Conclusions

The findings of the present study can be summarized as follows:

• Hydroxyapatite-based coatings undoped and doped with Mg in different concentrations were successfully deposited by electrochemical deposition in pulsed galvanostatic mode on a nanostructured surface made of titania nanotubes with a diameter of ~72 nm (±2);
• In terms of morphology, the present study showed that the undoped HAp and the HAp coatings obtained with a concentration of 1.5 mM of Mg in the electrolyte present similar morphologies, which are made of very thin and narrow ribbon-like crystals, while the morphology of the HAp coatings obtained with a concentration of 1 mM of Mg in the electrolyte, are made of wider and thicker ribbon-like crystals, suggesting that the growth mechanism of the crystals is influenced by the Mg concentration within the electrolyte used to obtain the coatings;
• EDS analysis revealed that by increasing the Mg concentration in the electrolyte from 1 mM to 1.5 mM, the quantity of Mg identified in the coatings is 0.44 at.% (0.31 wt.%) and 0.54 at.% (0.37 wt.%) which are within the values of Mg found in human hard bone tissue (0.30–0.72 wt.%); also, all elements (Ca, P, and Mg) are uniformly distributed within the coatings, indicating that the pulsed galvanostatic method is a reliable deposition technique;
• Compared with the undoped HAp, which registered a thickness of 9.6 μm, addition and increment of the Mg concentration in the electrolyte from 1 mM to 1.5 mM, led to thickness of 11.3 and 13.7 μm, indicating a strong association between the Mg amount and the deposition rate. Thus, a concentration of 1.5 mM of Mg in the electrolyte increases the deposition rate, while with a decrement of the Mg concentration, to a value of 1 mM, a less thick coating can be obtained. Based on this, it can be said that the coating thickness can also be controlled through a careful selection of the doping element concentration within the electrolyte. Of course, this should also be associated and correlated with other parameters involved in the electrochemical deposition, such as the deposition technique, pH value, deposition temperature, and so on to design a coating with enhanced and tunable features;
• The phase composition and chemical bonds analysis confirmed that all coatings consist of the HAp phase and that Mg²⁺ substituted some of the Ca²⁺ in the HAp structure. Moreover, the XRD analysis showed that the coatings have a preferred orientation toward the c-axis direction and that the addition of Mg in the proposed concentrations of 1 mM and 1.5 mM in the electrolyte can enhance the crystallinity of the HAp-based coatings;
• With respect to the Mg concentration in the electrolyte, it was observed that compared to the undoped HAp, the use of 1.5 mM Mg leads to an average roughness of 493 nm, while the use of a concentration of 1 mM Mg leads to an average roughness of 582 nm, suggesting that higher concentrations of Mg tend to smoothen the surface;
• The “tape test” and the micro-scratch tests highlighted that the adhesion of the HAp-based coatings is in strong correlation with the Mg concentration, indicating that an increment of the Mg can affect the adhesion of the coating even if the coatings are obtained on a nanostructured surface, which is known to enhance the adhesion of the coatings obtained by electrochemical techniques;
• moreover, the adhesion tests also showed that even if the coatings are being scratched, a small amount of coating will remain on the metallic substrate.

Overall, it can be said that the results of the current investigation indicate that the Mg concentration influences the physico-chemical properties of HAp-based coatings, and a concentration of 1 mM of Mg in the electrolyte used to obtain the coatings can maintain and/or even enhance the properties and behavior of HAp-based coatings. The coatings obtained with the addition of 1 mM of Mg, will be investigated in terms of electrochemical behavior, cell viability and proliferation, biomineralization ability, and degradation to establish if the proposed coatings are suitable for biomedical applications.