A Study to Evaluate the Bioactivity Behavior and Electrical Properties of Hydroxyapatite/Ag2O-Borosilicate Glass Nanocomposites for Biomedical Applications

In this study, nanocomposites with different contents of borosilicate glass (BG) and carbonated hydroxyapatite (CHA) were mixed, ground and sintered at 750 °C. In order to examine their phase composition, molecular structure and microstructure, x-ray diffraction (XRD) technique, Fourier transform infrared (FTIR) spectroscopy and scanning electron microscopy (SEM), respectively were used. Moreover, the DC electrical conductivity and physical and mechanical properties of the prepared nanocomposites were measured. In addition, the in vitro bioactivity of the sintered samples was evaluated using XRD and SEM. Unexpectedly; the results indicated that the successive increase in BG contents promoted the partial decomposition of CHA at this lower sintering temperature. Also, it was responsible for the enhanced bioactivity behavior along with giving CHA better mechanical properties whereby microhardness, compressive strength, and Young’s, elastic, bulk and shear moduli were improved even 50, 40, 85, 81.81, 78.5 and 77.27%, respectively. In addition, the density of these nanocomposites was enhanced to 31.03%. However, the electrical conductivity of the examined samples exhibited an opposite trend where it decreased by 87.3% with the increase of the BG content to 32 wt%. According to the results obtained, the prepared samples are suitable for use in various biomedical applications.


Introduction
The continuous development of new prosthetics is in great demand as a result of the tremendous improvements in medicine that have increased the life span of people all over the world [1]. In this regard, this goal can be achieved by many biomaterials such as various metals and their alloys, bioactive glasses, glass-ceramics, calcium phosphates and calcium silicates. In the past few decades, biomaterials have been considered as a research hot spot because they require many factors that cannot be easily met. Among these important factors are its electrical properties as they are very useful in relieving pain and improving quality of life. On the other hand, others are related to their promising biological traits such as bioactivity and biocompatibility [2]. In addition, the ability of the biomaterial to promote cell proliferation along with facilitating bone induction is highly required. It should be noted that these remarkable properties are strongly related to the characteristics of the implant surface [3]. On the basis of these facts, the researchers believe that bioactive materials such as bioglasses/ calcium phosphate compounds are potential candidates for achieving this goal [4].
Hydroxyapatite (HA; Ca 10 (PO 4 ) 6 (OH) 2 ) is a member of the calcium phosphate family and is found naturally in human and animal bones. Therefore, it has many amazing properties such as high surface area, fair bioactivity, excellent biocompatibility, etc. Moreover, its biological and surface properties can be improved by incorporating carbonate (CO 3 ) 2− ions into its crystal structure forming carbonated hydroxyapatite (CHA) and preparing it at the nano-scale range, respectively [5,6]. Although these remarkable properties, HA has several drawbacks such as low mechanical properties and relatively slow bone bonding ability. Therefore, to solve these undesirable defects, adding some elements or ceramics can be a useful solution to counteract these problems [7][8][9].
Bioglasses (BGs) have attracted a lot of interest in the past few decades due to their ability to gradually dissolve when implanted into the human body resulting in a controlled release of certain beneficial ions that enhance the ability to bone bonding [1]. BGs were first discovered by Hench et al. [10] in the 1970s and ever since, intensive efforts to discover new types of BGs have continued to be ideal candidates for the purpose of bone replacement. It is worth to mention that the major benefit of these glasses is their ability to induce the body's own repair process called "osteostimulation" [11,12]. One of the main groups of these glasses is the borosilicate group which deserves the attention of many researchers as a successful scaffold material. The reason for such interest is its faster absorption rate compared to silicate glass [13].
Extensive research has been conducted to study HA/polymers [14][15][16] or HA/ceramics composites [17][18][19][20]. However, at least to the authors' knowledge, the study of HA/ borosilicate bioglass nanocomposites still needs great attention; especially with measuring their electrical properties. In our attempt to contribute to highlighting of these important nanocomposites, the effect of adding low contents of borosilicate, up to 20%, to HA was studied, and it was found that this addition was effective in improving its mechanical and bioactivity [21]. In continuation of these efforts, the present work is concerned with adding glass, with relatively higher contents of up to 32%, to HA and raising the sintering temperature to 750 °C and studying the effect of these factors on the in vitro bioactivity, mechanical and electrical properties of these nanocomposites.

Preparation of Bioactive Glass (BG) Samples
Glass system has been prepared by melting the starting materials (SiO 2 , H 2 BO 3 , Na 2 CO 3 , CaCO 3 and AgNO 3 ) in a platinum crucible at 1100 °C for 2 h in air. In order to ensure a homogenous mixing of all reactants and get a bubble-free sample, the molten liquid was stirred to ensure homogenous mixing of all constituents. Then, the batches were poured in a mold with the desired shape and instantly put in another muffle to be annealed at about 380-400 °C for 1 h. Subsequently, the muffle was switched off and the temperature decreased to room temperature with a rate of 25 °C/h. The nominal composition of the prepared glass system is tabulated in Table 1.

Preparation of CHA Nanopowders
The CHA nanopowders were prepared with the help of Refs. [22,23] and they were characterized by the use of XRD technique and FTIR spectroscopy. Briefly, a high-energy ball mill was used using calcium carbonate (CaCO 3 ) and calcium hydrogen phosphate dihydrate (CaHPO 4 ·2H 2 O) as raw materials for the preparation of CHA nanopowders having in mind that the success of the mechanochemical synthesis method for producing CHA depends on many parameters like the ball-to-powder ratio (BPR) and milling time because they enhance the occurrence of the reaction. To better elucidate the morphology of the as-synthesized CHA nanopowders, scanning electron microscopy-coupled with energy dispersive spectroscopy (SEM-EDS, type Quanta FEG250) with an acceleration voltage of 30 kV and a magnification × 10 up to × 300,000 was used.

Preparation of CHA/BG Nanocomposites Powders
The CHA and BG, based on their weight%, represented in Table 2, were blended by a high-energy ball mill for 10 h with BPR equal to 1:2 and the diameter of balls was 10 mm. Subsequently, these mixtures were milled for 5 h in a high-energy ball mill operating at 400 rpm as a rotation speed and BPR = 20:1. Noteworthy, the milling has been performed under dry condition.

Physical Properties
Based on the recent articles of Taha et al. [24,25], the milled powders were pressed into pellets of 16 mm in diameter and 4 mm in length using hydraulic press at 50 MPa. Archimedes' method (ASTM: B962-13) was employed for determining bulk density and apparent porosity of the sintered samples at 750 °C.

Mechanical Properties
According to the references [26][27][28], the microhardness and compressive strength of the examined samples were measured. Additionally, Poisson's ratio and the values of elastic moduli; Young, elastic, bulk and shear modulus were calculated according to the reference [29].

Electrical Properties
The DC electrical conductivity of the prepared nanocomposites was measured at room temperature using a Keithley 616.

Characterization of the Sintered Samples
The phases formed in the nanocomposite samples result from sintering process were identified using X-ray diffraction (XRD) technique, "Philips PW 1373" X-ray powder diffractometer with CuK-Ni filtered radiation. In order to confirm the chemical composition of the prepared nanocomposites, Fourier transform infrared spectroscopy (Vertex 80, Bruker, Germany) was used. The FTIR absorption spectra of the tested nanocomposites were collected immediately by an attenuated total reflection (ATR) unit, at room temperature, in the wavenumber range 4000-400 cm −1 , 60 scans with a resolution 4 cm −1 . Moreover, SEM was also used to examine the microstructure of these samples.

The Evaluation of the In Vitro Bioactivity of Sintered Nanocomposites
The in vitro bioactivity of the resulting nanocomposites was assessed by soaking the sintered nanocomposites for 10 days in a simulated body fluid (SBF) prepared according to the recipe described by Kokubo et al. [30,31] while maintaining the ratio of glass grains to volume of solution = 0.01 g/ ml [32]. Then, the soaked samples were subjected to XRD and SEM to examine the HA layer formed on their surfaces.

Physical Properties
The densification behavior of the material is closely related to the porosity present in the sintered sample bearing in mind that the presence of CHA nano-sized powders is effective in improving the densification behavior by closing the pores [33,34]. Moreover, the glass contents in the nanocomposites along with the selected sintering temperature are expected to have an important role in enhancing the condensation manner [35,36]. On this basis, the bulk density and apparent porosity of the sintered nanocomposites were measured as shown in Fig. 1a, b. From this figure, it is possible to see that successive increases in BG content positively affected the bulk density. On the contrary, this increase in BG content reduces the porosity of the sintered nanocomposites. As previously discussed in references [5,20], good densification behavior depends on the sintering temperature while carrying out the sintering process through three stages. First, the compaction of powders facilitates contact formation. Second, the particles are closely bonded due to the creation of "necks" between them. Notably, these nicks are formed when the sintering temperature reaches two-thirds of the melting point. Finally, the particles are completely bound, cannot be seen individually and the remaining porosity is closed.

Mechanical Properties
Generally, the sintered glass samples show better mechanical properties compared to those of the original glass provided the sintering process is carried out under suitable conditions [37]. Therefore, microhardness, compressive strength, Young's modulus, elastic modulus, bulk modulus, shear modulus and Poisson's ratio of all sintered nanocomposites were measured and represented in Figs. 2, 3, 4. All the mechanical properties of the examined nanocomposites exhibit marked increases with successive increases in BG content. As will be explained in Sect. 3.2.1, although CHA was partially decomposed giving β-tricalcium phosphate (β-TCP; Ca 3 (PO 4 ) 2 ) which in turn leads to a marked decrease in the mechanical properties of the sintered samples, while the effect was largely offset by the existence of grains in the nano-scale range which considerably improves the mechanical properties. Also, the decrease in the apparent porosity of the sintered nanocomposites up to 78.94% after adding 32 wt% BG could contribute to giving the tested samples better compactness and mechanical properties. Furthermore, according to our recent articles [22,23], the measured hardness value of HA is 3.3 GPa which is much lower than that of similar borosilicate glass compositions which in some cases reaches 16 GPa [38]. Accordingly, the addition of BG is very helpful in improving the microhardness and group of elastic moduli. The results obtained are supported by Refs. [39][40][41]. Most importantly, understanding the relationships between porosity and mechanical properties is essential to take advantage of the porosity to improve the bioactivity behavior of the prepared samples without sacrificing their mechanical properties [22].

Electrical Properties of the Sintered Nanocomposites
Generally, electrical stimulation is known to enhance the rate of bone growth and thus reduce healing time [42][43][44].  Based on this principle, it is important to study the electrical response of the prepared nanocomposites. In this regard, the DC conductivity of all prepared nanocomposites was measured as shown in Fig. 5. As can be seen from this figure, the successive addition of glass, up to 32 wt%, to CHA was responsible for a significant decrease in the DC conductivity from 1.34 × 10 −8 to 1.7 × 10 −9 S/m. Before interpreting the obtained results, it is useful for the reader to clarify the possible mechanism of DC conductivity of the prepared nanocomposites. In general, in glass samples, the increase in DC conductivity due to the presence of Ag 2 O can be attributed to the successive jumping of Ag + ions from one non-bridging oxygen (NBO) to another [45]. However, sintering led to the transformation of the glass into glass-ceramic and the formation of crystalline phases, as will be discussed in the XRD results, which contributes to reducing the effect of Ag 2 O on the DC conduction of nanocomposites compared to amorphous glass which is one of the reasons for obtaining this result. Another possible reason for this result is that the borosilicate glass has a high electrical resistance, i.e. 10 11 -10 13 Ωm, thus, 10 wt% of Ag 2 O content is insufficient to significantly improve the electrical conductivity [46]. Although the conduction mechanism of HA is not clear, the proposed conductive species are ions such as protons (H + ), oxide (O 2 ) and hydroxyl (OH − ) [47]. Some published articles excluded the effect of Ca 2+ and PO 4 3− ions and stated that the ionic conductivity of HA is mainly due to the presence of OH − ions within the c-axis columns and thus, it is considered a one-dimensional anionic conductor. This assumption was confirmed by electrolysis measurements which revealed that OH − ions are considered charge carriers [48][49][50][51]. The results obtained are in a good agreement with those reported in Refs. [2,48].

XRD Analysis
It is known that heat treatment of prepared glass samples is responsible for converting them into corresponding  glass-ceramics. Importantly, the amount of the residual glass composition, the resulting crystalline phases and their amounts depend on the properties of the glass as well as the conditions of the heat treatment process [38]. Because of these facts, XRD technique is favorable for determining the phases produced as a result of heat treatment. Therefore, XRD patterns were recorded for all sintered nanocomposites specimens as represented in Fig. 6. A precise analysis of the obtained XRD patterns indicates that there are distinct peaks of CHA (JCPDS No. 19-0272) only for the HG0 sample. However, the HG1 sample exhibits an inconsiderable decrease in the intensity of some of characteristic peaks of CHA. This decrease could be attributed to the addition of 4 wt% of BG, at the expense of CHA, or a very slight degeneration of CHA and formation of β-tricalcium phosphate (β-TCP) that falls below the detection limit of the XRD instrument. For the HG2 sample, the distinct peaks belonging to the CHA appear next to those of the β-TCP phase identified according to (JCPDS No. 03-0691). Interestingly, the characteristic hump of the glasses that results from lack of lattice periodicity is not present in all tested samples, indicating that the specified sintering temperature effectively crystallized these glasses to their corresponding glass-ceramics. Of note, these results obtained are inconsistent with those stated in Ref. [23] which discussed that CHA nanopowders prepared by the mechanochemical synthesis method possesses high thermal stability even after exposure to 1000 °C. These results can be explained in terms of the presence of BG that promotes the decomposition of CHA at this lower temperature giving the characteristics peaks of the above phases due to the expulsion of some OH − groups. For the HG3 sample, the CHA peaks show significant decreases in their intensity coupled with considerable increases in the intensity of the peaks characteristic of β-TCP phase due to an increased influence of BG on the incidence of partial decomposition of CHA. Also, three new phases; namely calcium borate (CaB 2 O 4 ; JCPDS No. 76-0747), sodium borate (NaBO 2 ; JCPDS No. 12-0492) and calcium sodium borate (CaNaB 5 O 9 ; JCPDS No. 37-0827) are clearly shown. Finally, the new phase representing the crystalline silver oxide phase (Ag 2 O; JCPDS No. 76-1393) appears in the HG4 sample along with the above mentioned phases. It should be noted that Ag 2 O appeared in the XRD patterns only in the last sample containing 32 wt% of glass. The reason for this result is that the composition of the given glass system contains 10 wt% of Ag 2 O, so when glass was added to the HA with lower contents, i.e. 4, 8 and 16 wt%, the content of Ag 2 O in the prepared nanocomposites was 0.4, 0.8 and 1.6% which cannot be detected by the XRD instrument. By similarity, the characteristic XRD peaks belonging to SiO 2 did not appear at all in all investigating samples due to that the SiO 2 content in the selected glass system was only 5 wt%. Therefore, when a glass sample was added to HA with the contents used in this manuscript, i.e. 4, 8, 16 and 32 wt%, the SiO 2 content in the as-prepared nanocomposites was 0.2, 0.4, 0.8 and 1.6% which fall under the detection limit of the XRD instrument.
Most of the published crystallization research [52][53][54] has focused on silicate bioactive glasses and very few ones have been devoted to borate and phosphate glasses. In this regard, this study is an effort to contribute to the interest in studying the effect of sintering temperature on crystallization of borosilicate bioactive glasses. As described in Ref. [55], the presence of metal oxides such as Ag 2 O along with the heat treatment process is a major reason for the phase separation and the subsequent crystallization process.

Structural Analysis Using FTIR Spectroscopy
Generally, FTIR spectroscopy is a valuable technique for identifying basic structural groups in prepared samples. Therefore, the FTIR absorption spectra of all as-prepared nanocomposites were collected as shown in Fig. 7 while they were interpreted with the help of literature [5,[56][57][58][59]. Careful examination of the FTIR spectra reveals the following: (1) The characteristic bands belonging to CHA at 1630, 1460, 1420, 1100, 1035, 980, 870, 630, 605, 560 and 470 cm −1 are clearly visible in the HG0 sample. As expected, no other bands in this spectrum have been seen.
(2) In the HG1 sample, the aforementioned bands showed noticeable decrease in their intensity due to the slight decomposition that has occurred in the CHA molecule It is important to note that no vibrations were detected for the crystalline phases belonging to SiO 2 as its contribution to the formation of the prepared glass is about 5 wt%. Hence, the addition of this glass by 32 wt% is insufficient to appear.

Microstructural Analysis Using SEM
In general, the different biomedical applications of materials are strongly determined by their morphology [18]. Based on this fact, SEM coupled with EDS was employed to examine the morphology of the as-prepared CHA nanopowders as shown in Fig. 8a, b. Moreover, SEM was also used to investigate the microstructure of the HG0, HG2 and HG4 sintered samples as illustrated in Fig. 9a-c. As can be seen from Fig. 8a, the synthesized CHA is mainly composed of spherical particles with a large amount of agglomeration because the mechanochemically prepared materials usually show high agglomeration [18]. It is important to note that the presence of this agglomeration in the synthesized apatite sample is beneficial, according to the literature, as it allows the circulation of body fluids when this sample is applied as a coating on the implant [60]. The elemental analysis of the prepared CHA, as shown in Fig. 8b, reveals the existence of Ca, P, O and C peaks. On the contrary, the sintered CHA sample, i.e. HG0 appears to contain spherical particles with a number of needle-shaped particles with uniform distribution. This recorded change in morphology may be attributed to the sintering process of CHA at 750 ºC which agrees, to some extent, with those reported in Refs. [61,62]. Figure 9b shows that the needle-shaped particles disappeared as a result of the addition of BG to CHA at 8 wt% with partial decomposition that occurred in the CHA particles in agreement with the XRD results. It should be noted that the particles can easily be seen individually due to the high porosity. This observation is highly supported by the density results discussed in Sect. 3.1.1. By increasing the BG content up to 32 wt%, Fig. 9c, the porosity is greatly reduced and thus, the particles are no longer seen individually. These observations are well consistent with the results for XRD, density and porosity.

XRD Analysis
It is well known that immersion of bioactive substances in SBF solution allows them to react with ionic concentrations equivalent to human blood plasma, forming hydroxyapatite layer on their surfaces. Therefore, all sintered nanocomposites were dipped in SBF solution for 10 days and then subjected again to XRD to confirm the formation of the required apatite layer as shown in Fig. 10. It is possible to see from this figure that all the tested nanocomposites show good bioactivity considering that the increased BG content is responsible for the increased bioactivity of the nanocomposites. Therefore, the best bioactivity among all samples is the HG4 sample. This finding is attributed to that this sample contains 32 wt% of bioactive glass which possesses the  best bioactivity index over all types of bioceramics including CHA [63,64]. Unfortunately, the decomposition that occurs in the CHA molecules and the formation of β-TCP due to the increased BG content reduces the in vitro bioactivity of the examined samples due to the rapid dissolution rate of the β-TCP phase in the SBF solution. This interpretation is highly supported by the absence of this phase in the figure which is consistent with Refs. [18,29].
It is important to emphasize that although it is easy to test the bioactivity with an SBF solution of any biomaterial, it does not always give true results. In other words, it may give false positive/negative ones. This can be attributed to the fact that SBF consists only of inorganic ions present in human plasma at approximately the same concentrations, and thus, one can expect that this test does not simulate the dynamic conditions of living tissues. For instance, the TCP phase is usually unable to form an HA layer on its surface after soaking in SBF solution despite its great ability to osseointegrate under in vivo conditions. Based on this information, even samples showing lower bioactivity behavior may exhibit excellent bioactivity when tested under in vivo conditions [19,38].

Microstructural characterization by SEM
In this work, SEM was used to examine the apatite layer formed on the surface of the HG4 sample after it was immersed in SBF solution for 10 days as illustrated in Fig. 11. From this image, one can see that increasing the BG content to 32 wt% is a major reason for the formation of a relatively dense apatite layer on its surface. The reader can note that the dependence of this study on determining the formation of the apatite layer on XRD rather than the SEM technique. The reason for this reliance is its ability to detect this desired layer throughout the entire sample, which gives more accurate results, and therefore, it provides us with the information required to choose the sample with the best bioactivity among all. On the contrary, SEM is useful for giving readers only the visual note.

Conclusions
In this study, a borosilicate glass (BG) sample containing the composition: 5SiO2-40B2O3-20Na2O-25CaO-10A g2O (wt%) was prepared by conventional melt-quenching method. On the other hand, carbonated hydroxyapatite (CHA) nanopowders were prepared following the mechanochemical synthesis method described in our recent articles. Then, different contents of this glass sample, i.e. 0, 4, 8, 16 and 32 wt% have been added to the CHA nanopowders with the help of mechanical alloying method, consolidated and sintered at 750 °C. The physical, mechanical and electrical properties of the sintered nanocomposites were measured in terms of bulk density, apparent porosity, microhardness, compressive strength, Poisson's ration, and Young's, elastic, bulk, shear moduli, and DC conductivity. Moreover, the prepared samples were characterized using different tools such as Fourier transform infrared (FTIR) spectroscopy, X-ray diffraction (XRD) technique and scanning electron microscopy (SEM). Also, the bioactivity behavior of the nanocomposites after soaking in SBF solution for 10 days was evaluated with the help of XRD and SEM techniques. Analysis of the XRD patterns clarified that the chosen sintering temperature was sufficient to induce complete crystallization in the selected glass system giving calcium borate (CaB 2 O 4 ), sodium tetraborate (Na 2 B 4 O 7 ), crystalline silver oxide (Ag 2 O) and β-tricalcium phosphate (Ca 3 (PO 4 ) 2 ) phases beside CHA (Ca 10 (PO 4 ) 6 (OH) 2 ). It should be noted that both the bioactivity and mechanical properties were improved with the increase of BG content in the examined samples. In contrast, DC conductivity was negatively affected by this increase in BG content.