Investigating the mechanical behavior of hydroxyapatite-reduced graphene oxide nanocomposite under different loading rates

In this study, the hydroxyapatite (HA)-reduced graphene oxide (rGO) nanocomposite was investigated for its mechanical properties. The nanocomposite used in this study was made in two stages. The HA-rGO powders were first synthesized by hydrogen gas injected hydrothermal method, and then consolidated by spark plasma sintering. HA-rGO nanocomposite was subjected to Vickers indentation experiments with different loading rates. Various analyzes have been used in this study, including x-rays diffraction, field emission scanning electron microscopy, high-resolution transmission electron microscopy, fast fourier transform, and inverse fast fourier transform. The findings of this study showed that the HA in this nanocomposite was reinforced with rGO sheets coated with HA. As the loading rate increased, the slope of the curves in the elastic region was increased, indicating that the elastic modulus was increased. Also, the contact depth at higher loading rates was increased. Plastic deformation was higher at higher loading rates and the hardness had increased. As the loading rate increased from 300 mN to 1 N, the hardness and elastic modulus increased with more slope than when the loading rate changed from 1 N to 2 N. The presence of rGO sheets had partially controlled the HA brittleness.


Introduction
Applications of biomaterials are expanding rapidly because they are associated with human health. As a result, the advancement and improvement of medical materials and devices will always be growing. Medical materials are usually implanted in the body through surgery, which requires high precision in their design. Therefore, examination of the properties of biomaterials is among the most important medical challenges. In particular, materials used as implants must have biocompatibility properties [1,2]. Among the biomaterials known so far, HA and its composites have received much attention and among the HA composites, nanocomposites using carbon materials have shown more interesting properties. Research on carbon nanomaterials-HA nanocomposites is extremely expanding. Carbon nanomaterials include graphene sheets, graphene oxide (GO), carbon nanotubes, graphene nanoribbones, and nanodiamonds along with HA are the main phases of these nanocomposites [3][4][5][6][7]. Among the carbon nanomaterials, graphene has received more attention than others because of its unique properties such as good biocompatibility and excellent mechanical properties. Graphene plays a reinforcing role in graphene-HA nanocomposites. High specific surface area with honeycomb structure and thickness of about one carbon atom has made graphene excellent reinforcing properties [8][9][10][11]. Graphene has been used as a reinforcing phase to overcome the weakness of the mechanical properties of HA. HA is inherently brittle, has low fracture toughness and low wear resistance despite its excellent biocompatibility, bioactivity, non-toxicity and similarity to bone mineral composition. Therefore, reinforcing phases are added to eliminate mechanical weaknesses [12][13][14][15]. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
One of the synthesis methods for graphene-HA nanocomposites is the use of hydrothermal process. In this method, the rGO-HA nanostructured powders are first synthesized by hydrothermal method at high temperature and pressure [16][17][18][19]. These powders subsequently consolidated by various methods such as spark plasma sintering (SPS). The rGO sheets obtained in this method are the GO sheets that are reduced in the hydrothermal process and converted to rGO. Previous researches have shown that the addition of graphene sheets to HA improves the mechanical and biological properties of HA. Adding a small amount of graphene sheets (up to 1%) has significantly increased the mechanical characteristics of HA (compression strength and fracture toughness). The addition of graphene sheets has also improved the biological properties under in vitro conditions, although more research is needed under in vivo conditions. In all published reports, the mechanical behavior of these nanocomposites against the specified amount of loading has been investigated, but no report has been published on the mechanical behavior of these materials against load changes [20][21][22][23]. The mechanisms that lead to increased fracture toughness in these nanocomposites include graphene pull-out, crack deflection, graphene bridging, and inhibition of crack growth by graphene sheets [24,25]. One of the mechanical properties indicators of materials is their behavior against the variable loads applied to the materials at varying rates. Indentation (micro and nano) is one of the methods to obtain the mechanical properties (hardness, elastic modulus, and fracture toughness) of ceramics, which is performed by various indenters such as Vickers and Berkovich indenters [26][27][28][29][30][31][32][33].
In a previous published study, rGO (1.5%wt)-HA nanocomposite was prepared by high pressure hydrothermal process followed by the spark plasma sintering method. The results showed that the interfaces between the rGO sheets and the HA were coherent. Also, indentation analysis with nanoindentation method showed that the mechanical properties of this nanocomposite were significantly increased compared to pure HA [34]. The main purpose of this study is to investigate the mechanical behavior of HA-rGO nanocomposites against the load changes. For this purpose, HA-rGO powders were first synthesized by a hydrothermal method and then consolidated by SPS. The fabricated nanocomposite was subjected to a Vickers indentation experiments with different loading rates. Various analyzes have been used in this study, including x-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), field emission scanning electron microscopy (FESEM), high-resolution transmission electron microscopy (HRTEM), fast fourier transform (FFT) and inverse fast fourier transform (IFFT). The findings of this study could be useful for applications of this nanocomposite.

Preparation of GO
GO was synthesized by the Hummers' method. Briefly, 2 g of graphite powders was added to 46 mL of H 2 SO 4 (98%) and was stirred for 1 h. Then, 6 g of KMnO 4 was added gradually and was stirred for 2 h at 35°C. Then, 92 mL of water was added slowly. The mixture was stirred for 1 h and diluted with 280 mL of water. Then, 10 mL of H 2 O 2 (35%) was added dropwise. The resulting GO suspension was left overnight. The remaining GO suspension was washed by centrifugation, with HCl and acetone, and water. Finally, the precipitation of GO was dried in a vacuum oven at 60°C for 24 h [23, 35].

Preparation of HA-rGO nanocomposite
The solution containing calcium ions (4.7 grams of calcium nitrate tetrahydrate in 120 mL of water) was added dropwise to a 20 mL stirred suspension of GO (HA/1.5% rGO) with stirring continued for 1 h. The solution containing phosphate ions (1.56 grams of diammonium hydrogenphosphate in 80 mL of water) was dropwise added to the solution and the pH was adjusted to 11 with ammonium solution. Then, the resulting solution was poured into the Teflon vessel and transferred to the hydrothermal autoclave (5 h at 180°C by injection of hydrogen gas at 10 bar). The resulting powders were dried at oven for 24 h at 60°C. Then, the synthesized powders were consolidated by SPS method (950 o C, 50 MPa). The specifications of the chemicals and hydrothermal system used in this study have been published in previous reports [34]. Instrumented microindentation experiments were conducted on the polished surfaces of samples using Grindosonic tester with a Vickers tip at a maximum load of 300 mN, 1 N, and 2 N (ramp dwell time of 10 s). Figure 1 shows the time-temperature-displacement-sintering speed diagram for SPS process, indentation affected zone, loaddisplacement diagram, and load-time diagram. According to these diagrams ( figure 1(a)), the displacement value is ultimately about 2 mm. The distortions observed in the 900°C temperature range correspond to the critical temperature of the sintering. After sintering, nine tests were performed at different locations of each sample by the Vickers indentation technique ( figure 2(b)). Elastic modulus and hardness were calculated from the load-displacement curves (figure 2(c)) using Oliver-Pharr method [36].   Table 1 shows the specifications for the analytical methods and devices [34,37]. A diamond nail was used to evaluate the nanocomposite's resistance to scratch. Also, ImageJ, CasaXPS, Origin pro 2016, and Diamond 3.2 softwares were used in this study. Figure 2 shows the XRD and XPS analyzes for GO and rGO synthesized under similar conditions. To investigate the effect of hydrothermal conditions on the reduction of GO, some rGO powders was prepared under the same conditions as HA-rGO. The XRD patterns (figure 2(a)) showed that GO had a peak in the range of 2θ≈10 deg. As a result of the reduction, this peak disappeared and a new peak appeared in the range of 2θ≈26 deg. Also, the XPS patterns of GO, rGO, the presence of marked bands (O-C=O, C-O, C=O, and sp 2 C), and the specific displacements in these patterns (figures 2(b)-(d)) showed that GO was well reduced under hydrothermal conditions [19,38]. Figure 3 shows the SEM image of consolidated sample, FESEM and TEM images of HA-rGO powders and FESEM images of HA-rGO nanocomposite after SPS. In figure 3(a), the large black spot corresponded to the rGO sheets that their surfaces were covered with HA nanorods. Figures 3(b) and (c) are the images of the initial powders, as can be seen the rGO sheets were coated with HA particles. HA has positive charge in some crystalline planes due to the presence of calcium sites. Therefore, a strong electrostatic interaction between HA and GO sheets was formed. In addition, a hydrogen interaction between hydroxyl groups from HA and functional groups containing oxygen from rGO planes such as carboxyl, hydroxyl, carbonyl, and epoxy groups occurs, and a good adhesion between HA and graphene may arise. GO basal planes were more covered with epoxy and hydroxyl groups, while carboxyl groups were located at the edges. The charge of the GO surface was negative in terms of functional groups. The structure of GO affected the formation of HA and induced epitaxial and directional growth. Also, as anchoring sites, oxygen-containing groups induced the formation and bonding of particles to the surface of GO. In the chemical precipitation stage, when adding calcium nitrate to a solution  containing GO, Ca 2+ cations were absorbed and bounded to GO by electrostatic interactions with hydroxyl groups or ion exchange with carboxyl groups and played the role of the primary sites for nucleation and growth of HA particles. Calcium ions can react in situ with phosphate ions and form calcium phosphate nanoparticles through electrovalence reactions. In this process, the solution was strongly stirred and the rGO layers were homogeneously dispersed, then the same nucleation and the controlled growth occurred. The particles collided several times before connecting to the rGO sheets, which is why a compact structure was formed that had a rodshaped morphology. The frequency of the collision with the diffusion rate of the growth units, and the reaction rate was determined by the thermodynamic conditions, which depended on the heat transfer rate of the accumulated particles. Most research suggested that HA precipitation occurs more at negative charged sites than positive charged sites, and among the negatively charged groups, phosphate groups have the strongest nucleating for HA [17,18]. These findings indicated that the HA in this nanocomposite was reinforced with rGO sheets coated with HA. Figure 3(d) also show that the HA nanorods were well sintered. Due to the pressure applied at high temperatures, the morphology of the powders had changed slightly. Due to the coating of rGO sheets with HA and the proper bonding of the HA particles in the sintering process, it was expected that the twophase bonding (interface) was performed well [34,39]. Figure 4 shows the XRD and XPS analyzes of the HA-rGO nanocomposite. The XRD pattern of this nanocomposite was consistent with that of pure HA (JCPDS 09-0432) ( figure 4(a)). Accordingly, there was no phase other than HA in this nanocomposite, in other words, the HA present in this nanocomposite had not been decomposed at sintering temperature. rGO peaks in this pattern were covered by HA peaks and if it had been oxidized during the sintering operation, it should have seen a long peak at 2θ≈10 deg. According to these patterns, the 2θ≈26, 32, 32. And in the case of HA nanorods synthesized by the hydrothermal method, the (002) planes were preferable to the others (c axis) [40][41][42][43][44]. According to the XPS pattern ( figure 4(b)), P 2p and Ca 2p signals confirmed the presence of HA phase after consolidating process (SPS, 950 o C). In other words, HA phase had retained its composition after high temperature sintering and had not been decomposed. According to figures 4(c) and (d), the carboxyl groups of rGO surface still remained the same as the GO, due to the strong electrostatic interaction between carboxyl groups located on the surface of rGO and calcium phosphate [34]. Figure 5 shows the load-displacement, time-displacement, load-hardness, and load-elastic modulus diagrams. Figure 5(a) shows that as the loading rate increased; the slope of the curves in the elastic region was increased, indicating that the elastic modulus was increased. Also, the contact depth at higher loading rates was increased according to figure 5(b). These graphs show that plastic deformation was higher at higher loading rates. As a result, the hardness had increased. According to figures 5(c), (d) (in both graphs) as the loading rate increased from 300 mN to 1 N, the hardness increased with more slope than when the loading rate changes from 1 N to 2 N. As the number of created cracks and their length increased at higher loading rates, the slope of these diagrams decreased at higher loads. In other words, at low loading rates, the energy was spent on plastic and elastic deformation. But at high loading rates, the energy was spent on elastic deformation, plastic deformation, nucleation and growth of created cracks. The hardness and elastic modulus of this composite (hydrothermal process in 5 h) at 1 N loading was equal to the properties of pure HA under similar conditions and the hydrothermal process in 10 h [37]. This indicated that the presence of rGO sheets had a great influence on the mechanical properties because of the lower crystallinity of the powders at low synthesis times. But the effect of rGO was more important in increasing fracture toughness [17,45]. Figures 6 and 7, and schematic 1 show the HRTEM images of the HA-rGO nanocomposite, the FFT and IFFT analysis along with the schematic images of rGO sheet, rGO cross section, and HA nanorod crystal planes. In the previous studies, the interface between two phases was investigated and it was found that the interface was coherent. Figure 6 provides further details of the HRTEM images. In these analyzes, the honeycomb structure of the rGO sheet is well characterized (figures 6(b), (e)). As is evident, these rGO sheets were well positioned between the HA particles and there was a good connection between the two phases [34]. By comparing the d-spacing of the planes and the properties shown on the schematic images with the atomic images in figures 6 and 7, and schematic 1 a clearer understanding is obtained from the interface between the two phases described before. This kind of coherent interface between the two phases had increased the fracture toughness of HA-rGO nanocomposite. In other words, when the interface between the two phases is coherent, there is a stronger connection between them and the resulting composite strength increases [19].

Results and discussion
It is well known in figure 7 that the rGO sheets and the HA particles were fully integrated ( figure 7(a)) [34]. In some places, the rGO sheets were interconnected and form a three-dimensional scaffold structure ( figure 7(e)). The angle of the (211) planes was about 27 degrees with the rGO surface ( figure 7(c)). On the other hand, the angle between (211) and (300) planes in HA was about 27 degrees. Therefore, the rGO surface was tangible to the HA (300) planes [46,47]. By comparing FFT analyzes (figure 7(h)), it is found that rGO cross section analysis was galactic. Figure 8 shows the FESEM images of cracks, the mechanisms for increasing fracture toughness, and the FESEM images of scratch evaluation. Due to the inherent brittleness of HA, cracks in it grow rapidly and lead to failure. The presence of rGO sheets in HA-rGO nanocomposites prevented the growth of these cracks and increased the fracture toughness. The most common mechanisms that increased fracture toughness were rGO sheets bridging, rGO pull-out, crack branching, and crack deflection [17]. In other words, rGO sheets need more energy to be removed due to their high specific surface area. This increased in energy to be removed from the HA  phase increased the fracture toughness of HA-rGO nanocomposites [19]. Some of these mechanisms are illustrated in figure 8. In figure 8(a), the Vickers affected zone is known. This section was related to high loads > 2 N. In high loads such destruction occurred but they were much smaller than pure HA, due to the increased nanocomposite fracture toughness. The scratch test results (figures 8(e)-(g)) showed that the presence of rGO sheets had partially controlled the HA brittleness. Folding and wrinkling of rGO sheets had improved the weakness of HA brittleness. The mechanism that improved the brittleness involved the transfer of force that caused the HA particles to disintegrate into rGO sheets. Due to the high specific surface area of rGO sheets, this Schematic 1. The schematic images of graphene sheet, rGO cross section, and HA nanorod crystal planes force was compensated by rGO folding or transferring force to other particles, and the brittleness was somewhat improved.

Conclusions
The findings of this study showed that the HA in this nanocomposite was reinforced with rGO sheets coated with HA. As the loading rate increased, the slope of the curves in the elastic region was increased, indicating that the elastic modulus was increased. Also, the contact depth at higher loading rates was increased. Plastic deformation was higher at higher loading rates and the hardness had increased. As the loading rate increased from 300 mN to 1 N, the hardness and elastic modulus increased with more slope than when the loading rate changed from 1 N to 2 N. The presence of rGO sheets had partially controlled the HA brittleness. The findings of this study could be useful for applications of this nanocomposite.