Titanium Coated with Graphene and Niobium Pentoxide for Biomaterial Applications

Graphene and niobium oxide are used in biomaterial coatings. In this work, commercially pure titanium (cp Ti) was coated with graphene oxide (GO), niobium pentoxide (Nb2O5), and a mixture of both materials (NbGO) by the electrochemical deposition method. The surface morphology, roughness, wettability, and degradation of coated and uncoated samples were analyzed by scanning electron microscopy, interferometry, and contact angle. The results showed that the specimens coated with NbGO (cp Ti-NbGO) showed the highest surface roughness (Ra = 0.64 μm) and were hydrophobic. The contact (θ) angle between water and the surface of uncoated specimens (cp Ti), coated with GO (cp Ti-GO), coated with a mixture with GO and Nb2O5) (cp Ti-NbGO), and coated with Nb2O5 were 50.74°, 44.35°, 55.86°, and 100.35°, respectively. The electrochemical corrosion tests showed that coating with graphene oxide increased the corrosion resistance and coating with Nb2O5 decreased the corrosion resistance. The negative effect of the effect of Nb2O5 coating in corrosion resistance compensated for the release of Nb2O5, which helps osseointegration, increasing cell viability, and proliferation of osteoblasts. The NbGO coating may be a good way to combine the bactericidal effect of graphene oxide with the osseointegration effect of Nb2O5.


Introduction
Titanium and its alloys are widely used as biomaterials. Its main feature is osseointegration since the titanium oxide flm on the surface enhances the adhesion of proteins and bone cells. Te wrought titanium 6-aluminum4-vanadium ELI alloy (Ti-6Al-4V ASTM F136), alpha plus beta titanium alloy forgings (ASTM F620), and wrought titanium 6-aluminum7-niobium alloy (ASTM F1295) have adequate mechanical properties for use as a biomaterial, but they release toxic ions (Al and V), decreasing its biological activity. Another problem observed is that among the diferent causes of failures of orthopedic prostheses and dental implants, the adhesion of bioflms and chronic infammatory processes deserves to be highlighted. In the case of implants, bioflm adhesion can also facilitate periimplantitis. Despite the various procedures adopted to minimize this problem, implant scraping and the use of antibiotics are the most efcient. Even so, the infammatory process can be recurrent, leading to implant loss. For this reason, surface treatment and chemical modifcations of titanium alloys for implant biomaterials application are important.
Graphene oxide (GO) has a bactericidal efect and does not inhibit cell growth. Te antibacterial activity of GO is based on the production of hydroxyl radicals, which attack the cell membranes of bacteria, causing their death. During the process, reactive oxygen species are generated, which lead to oxidative stress, exceeding the antioxidant defense capacity of the bacteria and causing damage to their lipids, proteins, and DNA. Te morphology of GO is formed by nanosheets with sharpened edges that break the bacterial membrane, causing the intracellular matrix to drain into the medium [1]. GO inhibits the growth of bacteria such as Escherichia coli and Streptococcus iniae. In high concentrations, GO inhibits the action of Gram-negative bacteria [2].
Among niobium oxides, niobium pentoxide (Nb 2 O 5 ) is the most stable thermodynamically [3]. Cell culture results showed that coating with Nb 2 O 5 enhances cellular viability and proliferation after 21 days. Te cells present good adhesion and are homogeneously distributed on the surface. Previous results showed that the lactate dehydrogenase activity was similar to the titanium surface with and without Nb 2 O 5 coating, indicating a similar number of cells.
It is extremely important to evaluate metallic implant corrosion resistance to avoid degradation and release of toxic ions. Te accelerated degradation causes problems in tissue, but in some cases, it can be benefcial [4]. Literature results report the corrosion of dental implants made with titanium alloy ASTM F136 in contact with saliva and fuoride, which can cause the release of ions that can induce toxicity in the organism and loosen the implant [4][5][6]. Coating titanium with GO may increase the corrosion resistance of dental implants.
Tis work aims to examine the properties of cp Ti coated with graphene oxide, niobium pentoxide, and the mixture of both materials for use in dental implants.

Materials and Methods
Commercially pure titanium grade 4 bars (cp Ti ASTM G4) were used in the present work. Te material was supplied by Conexão Sistema e Prótese (Arujá, SP-Brazil). Te samples were acid-etched with a mixture of sulfuric acid, hydrochloric acid, and distilled water and washed in an ultrasonic bath with distilled water before all testing.
Te deposition of graphene oxide was carried out by electrodeposition in an aqueous solution ( Figure 1). Te graphene oxide concentration was between 0.5 mg/mL and 0.7 mg/mL. During deposition, the working electrode (cathode) was a cp Ti sample, and the anode was a platinum plate. Te distance between the cathode and the anode was 1 cm. Te depositions of Nb 2 O 5 and a mixture of GO and deposition and Nb 2 O 5 followed similar routes. Table 1 shows the code used to identify the specimens.

Characterization
Te surface morphology of three samples from each group was investigated by scanning electron microscopy (Field Emission Gun FEI Quanta FEG 250, Hilesboro, Oregon USA) before and after coating.
Te sample wettability was measured by the contact angle method. Contact angle measurements were performed with deionized water, using the sessile drop technique. Te equipment used was the FTA 100 goniometer (First Ten Angstroms, Portsmouth, VG, USA).
Literature results showed that osseointegration is the main parameter for the success of dental implants, which means a protein and bone cell attachment on the implant surface [7]. Tis attachment is infuenced by the implant surface roughness. Among the parameters that characterize the surface roughness, the arithmetic average roughness (Ra), the average absolute value of the fve highest peaks and the fve lowest valleys over the evaluation length (Rz), and the root mean square roughness (RMS) were measured before and after the surface treatments. Te roughness measurements were performed using Zygo NewView 7100 optical interferometry (Zygo Corporation, Connecticut, the United States).
Two electrochemical analyses were carried out to evaluate the adhesion of the coating on sample surfaces and to analyze the infuence on corrosion resistance. Te frst electrochemical testing was the evolution of the open circuit potential (OCP), which was monitored for 3600 seconds. Te second performed testing was potentiodynamic polarization to obtain the polarization curves. For potentiodynamic polarization, a sweep rate of 0.001 V/s and potentials of −0.500 V to 1 V were used. Te electrolyte was a physiological solution containing 0.9% of NaCl at room temperature. Te equipment used was an AUTOLAB potentiostat/galvanostat (Metrohm Autolab B.V., 3526 KM Utrecht, Te Netherlands). Te cell used for the analyses consists of a calomel reference electrode, a platinum counter electrode, and the sample disc as the working electrode.
Te statistical analyses were performed by an ANOVA variance calculation followed by Tukey's test. For statistical analysis, PRISM GRAPH 9 (Graphstats Technologies Private Limited, Karnataka, India) software was used. Figure 2 shows the spectra obtained in the semiquantitative chemical analysis with EDS. Te surface composition did not change after graphene oxide deposition (Figure 2(b)). Figure 2(c) shows the EDS spectrum after niobium pentoxide deposition, and it is possible to observe the presence of Nb. After graphene and niobium pentoxide deposition on the surface, the EDS spectrum shows the presence of Nb, as well as C and Ti.    Figure 5 shows water droplets on titanium surfaces. Te surface properties infuence protein adhesion and cell proliferation. In the initial phase of contact of the biomaterial with the organism, there are physical interactions, chemical bonds between cells and surfaces, or, indirectly, through a change in the adsorption of conditioning molecules, proteins, for example [9].

Results
Cell attachment and expansion are lower on hydrophobic surfaces than those on hydrophilic ones. Moderately hydrophilic surfaces promote cell attachment [10]. Ti with a hydrophilic surface (contact angle <90°) has a lower time for osseointegration than Ti with a hydrophobic surface (contact angle >90°) [11]. Te wettability results of all analyzed samples are adequate based on data reported in the literature. Table 2 shows that the cp Ti-Nb specimens have a contact angle of 100.35°, which is typical of hydrophobic behavior. Tree groups of specimens (cp Ti, cp Ti-GO, and cp Ti-NBGO) have a contact angle lower than 90°, which is hydrophilic behavior. Figure 6 shows the comparative analysis of the Tukey test of wettability data. It can be seen that there is no statistically signifcant diference among the hydrophilic groups, and there is a signifcant diference between the hydrophilic groups and the Ti-NB group.
Roughness and wettability are important parameters to evaluate the success of dental implants. Hydrophobic surfaces decrease the primary interactions with cells in an  International Journal of Biomaterials 5 aqueous environment. It is usually reported that biomaterial surfaces with moderate hydrophilicity improve cell growth and material biocompatibility [10]. Graphene oxide, obtained by the oxidation of graphite fakes, has hydrophilic characteristics, good dispersion, and compatibility with various polymeric matrices [12,13] Titanium, by itself, is normally hydrophilic. However, most available commercial implants after surface treatment are hydrophobic [14].
Te analysis of the simultaneous infuence of surface wettability, roughness, and chemical composition on osseointegration is complex. Surface treatments modify the wettability of implants, making them hydrophilic or hydrophobic. Some proteins interact with hydrophilic surfaces, and others interact with hydrophobic ones. Smooth surfaces are suitable for interaction with fbroblasts, and rougher surfaces are suitable for interaction with osteoblasts. As for the wettability, it is possible to estimate the relationship between the surface energy and the contact angle. Tere is an inversely proportional relationship between the two properties, regardless of whether the samples have micrometer (Ti G4) or submicrometer (Ti Hard) grain size [15].
Cell adhesion and scattering occur on surfaces with moderate wettability and contact angles between 50°and 80°. Te fxation and proliferation of osteoblasts increase as the surface wettability increases [16,17]. Te implant osseointegration rate (BIC: bone implant contact) with commercially SLActive surface (hydrophilic, 48.3%) is higher than for implants with hydrophobic surfaces of SLA implants, 32.4% [18]. Cell adhesion and propagation occur on surfaces with moderate wettability and contact angles of 50°-80°.
Classifying surfaces as hydrophilic or hydrophobic is not enough to determine a good relationship between the material surface properties and cellular adhesion and to analyze the possibility of implant osseointegration [19]. Te infuence of wettability must take roughness into account since there is a correlation between roughness and wettability. Te contact angle increases as roughness parameters (Ra, PV, and RMS) increase in a nonlinear way [15]. Figure 7 shows the surface morphology of the specimens determined by interferometry, and Table 3 shows the surface roughness parameters of the samples analyzed in this study. Te specimens had Ra roughness values similar to those reported in the literature [20]. For uncoated samples, Ra � 0.52 ± 0.06 μm. Samples coated with niobium pentoxide    Te Ra parameter is most used in the literature to characterize the surface roughness of dental materials and applies to most manufacturing processes. Among the disadvantages of using the Ra value, we highlight the fact that this parameter only quantifes the average roughness. It makes no distinction between peaks and valleys and does not defne the shape of the irregularities. For this reason, in the present work, the roughness measurements were complemented by the RMS and Rz parameters. Te RMS value corresponds to the square root of the mean of the squares of the ordinates, that is, the root means square deviation. Te Rz parameter corresponds to the arithmetic mean of fve values of the distances between peaks and valleys. Tese parameters were quantifed to analyze the infuence of coatings on surface roughness, which act directly on the cell's adsorption of micrometric surface characteristics [21].
Te cell interaction with the implant surface is infuenced by topography at macroscopic levels and roughness at the microscopic level. Increasing the surface area of the implant increases the number of possible sites for cell attachment, facilitates tissue growth, and increases mechanical stability. However, this is not a general rule. Te level of roughness must be controlled because the cells need anchorage points on the implant surface to initiate proliferation and ensure mechanical stability. When the surface has a roughness much smaller than the cells' sizes, there may be an absence of adhesion sites. On the other hand, if the implant has a larger quantity of small peaks or valleys, the surfaces may be smooth and the cells, likewise, cannot be fxed [22].
Dental implant surfaces with Ra between 0.8 and 2.0 μm have better osseointegration and mechanical stability than smooth or rougher surfaces. Implants with surface roughness between 0.8 and 1.5 μm and with nanoroughness have the best success rates [23]. Possibly, the combination of micrometric roughness and the presence of graphene nanoparticles can improve dental implant osseointegration because cell adhesion on surfaces is directly related to chemical afnity and roughness. On dental implant titanium surfaces with irregularities larger than the size of the cell itself (osteoblast), there is lower adhesion. Tis happens due to the inability of the cell to establish a sufcient contact area with the substrate [24] Some cellular properties such as adhesion, morphological changes, proliferation, and differentiation are afected by the chemical composition, roughness, surface energy, and wettability of titanium [25]. On a smooth surface, polar groups increase wettability and nonpolar hydrophobic groups decrease it [26].
In vivo tests with rabbits showed that machined implants without surface treatment have Ra � 0.32 ± 0.03 μm and a removal torque of 62.08 N·cm. Implants with Ra � 1.51 μm are removed with a torque of 66.56 N·cm. Te implant with greater Ra (1.0 μm) needs a torque of 76.45 N·cm to be removed. Tese results show that osteoblasts have a better afnity for surfaces with medium roughness (Ra � 1.0 μm) [20].

International Journal of Biomaterials
Considering that RMS is the standard deviation of Ra, it can be said that the values of Ra and RMS are complementary. Figure 8 shows the comparison of the RMS values among the groups. All groups showed a signifcant diference relative to the cp Ti-NBGO samples, and there is no signifcant diference among the other groups. To evaluate the coating adhesion and samples corrosion resistance, an open circuit potential (OCP) was monitored, and a polarization curve was obtained. Te OCP is the equilibrium potential between the sample surface and the solution. Figure 9 shows the OCP for cp Ti, cp Ti-NBGO, cp Ti-GO, and cp Ti-NB in a physiological solution containing 0.9% NaCl. It is observed that the potential of cp Ti is around −0.200 mV. When the samples were coated with GO, a displacement of the potential to positive values occurs, indicating that the corrosion increased. Tis indicative was also observed by Li et al. 2018 [27] in their analyses. Te niobium oxide showed the opposite behavior since there is a shift to more negative values. Tis shift indicates a decrease in corrosion resistance. Figure 10 shows the anodic potentiodynamic polarization curves for coated and uncoated samples. Table 4 shows the values of open circuit potential, corrosion potential, and current density. It is observed that the two conditions with the presence of niobium were the ones that presented the highest value of current and corrosion potentials, which corroborates what was observed in the curves. Figure 10 shows the polarization curves. All curves have a passive region. After niobium pentoxide deposition on the Ti surface, the current density increased in the anodic region, which means that the corrosion process increased. Tis behavior is due to the presence of a discontinuous flm, which causes localized pits and increases the corrosion process. After graphene flm coating, it is possible the corrosion resistance increases. Tis behavior may be associated with the formation of a protective layer that prevents the oxidation of titanium. Tis protection related to graphene flm has been observed by diferent authors concerning other metallic alloys such as copper alloys and steels [27].
Te deposition of graphene on metal surfaces forms a protective barrier that increases the corrosion resistance [27] and the dissolution process of diferent alloys. Te graphene deposition efciency against corrosion is on the order of 100 times higher than other processes. However, the deposition of large amounts of graphene induces galvanic corrosion [28]. Te increased corrosion resistance and microbiological properties of GO are favorable for biomaterial applications. Te flm defects decrease the corrosion resistance and increase the ion release. Table 4 shows the values of open circuit potential, corrosion potential, and current density. It is observed that the two conditions with the presence of niobium were the ones that presented the highest value of current and corrosion potentials, which corroborates with the potentiostatic curves. More negative OCP values are related to higher current values. Terefore, lower OCP values and higher current values are associated with lower corrosion resistance.
Tis decrease in corrosion resistance after the deposition of Nb 2 O 5 may be associated with the lower stability of niobium oxide than titanium oxide, which is confrmed by the Ellingham diagram [29]. However, this Nb 2 O 5 release is the desired efect since it may help osseointegration, increase in cell viability, and the proliferation of osteoblasts [30].
Based on the Ellingham diagram [28], it is possible to explain the decrease in corrosion resistance after the deposition of Nb 2 O 5 . Tis behavior may be associated with International Journal of Biomaterials lower niobium oxide stability than titanium oxide. However, this Nb 2 O 5 release is the desired efect since it may help osseointegration, increase in cell viability, and osteoblast proliferation [29].

Conclusion
Based on experimental results, it is possible to conclude that (a) Wettability measurements showed that the Ti samples coated with graphene oxide (cp Ti-GO) had lower surface energy (12.94 ± 9.33 mJ/m 2 ) than without coating cp Ti (51.61 ± 4.49 mJ/m 2 ), cp Ti-NB (40.80 ± 13.47 mJ/m 2 ), and cp Ti-NBGO (46.10 ± 6.38 mJ/m 2 ). Tese surface energy values are adequate for protein and cell adhesion. (b) Roughness measurements showed that uncoated samples had Ra � 0.52 ± 0.06 μm, niobia coated samples had Ra � 0.44 ± 0.03 μm, graphene oxide coated samples had Ra � 0.47 ± 0.04 μm, and samples coated with both niobia and graphene oxide had Ra � 0.64 ± 0.47. Statistical analysis of the roughness parameters showed no statistically signifcant difference among samples from the frst three groups and a signifcant diference with the last group. (c) Te anodic potentiodynamic polarization curves show that the graphene oxide coating increases corrosion resistance, the Nb 2 O 5 coating decreases corrosion resistance, and the coating with both materials may be benefcial since the release of Nb 2 O 5 helps in osseointegration, increasing cell viability, and proliferation of osteoblasts

Data Availability
Te data used to support the fndings of this study are included in the article.

Conflicts of Interest
Te authors declare that they have no conficts of interest. Table 4: Open potential circuit (OCP), corrosion potential (E corr ), and current density (i corr ).

10
International Journal of Biomaterials