Titanium Bioactive Surface Formation Via Alkali and Heat Treatments for Rapid Osseointegration

Titanium and its alloys are widely used as implant materials and many studies to accelerate the osseointegration have been performed. This work aims to evaluate the formation of a bioactive surface in commercially pure titanium (cp-Ti) grade 4 after alkali (AT) and heat treatments at 600 °C (AHT600) and 900 °C (AHT900). Characterization techniques were SEM, AFM, Raman, TF-XRD, wettability, nanoindentation and indentation adhesion. Additionally, SBF soaking tests were performed to evaluate apatite growth and showed that alkali and heat treatment accelerates apatite growth. The AT samples formed sodium hydrogen titanate (1 μm thick), and AHT600 and AHT900 formed sodium titanate (1 μm thick), while rutile TiO2 increased with temperature, reaching up to 5 μm thick and the surface changed from slightly hydrophilic to fully hydrophilic. Roughness and surface area increased, especially in AHT900. The hardness of the surface layer was significantly increased by the heat treatment.


Introduction
Commercially pure titanium (cp-Ti) and its alloys have been widely using as dental implant biomaterials due to its biocompatibility and good mechanical and corrosion resistance [1][2][3][4][5] . On the other hand, due to the bioinert properties of titanium, the bone formation rate (osseointegration) is very slow in the body 6 . Properties such as surface chemistry, texture, and wettability play a key role to accelerate the osseointegration of dental implants 7 . Hydrophilicity can alter bone cell adhesion, proliferation, and differentiation properties 8 . Increased surface roughness may induce strong bone-titanium interaction by increasing the bone-implant contact area 9 . In addition, the chemical characteristics of the altered surface can induce bone binding and affinity by promoting hydroxyapatite precipitation from body fluid 10,11 .
In recent decades, several surface treatment studies have been conducted to improve osseointegration, such as plasma-spraying, sandblasting, and chemical treatments 12,13 . Among chemical treatments, alkali and heat treatment have been widely studied due to the simplicity of the method and ease of process control. Alkali and heat treatment consist of soaking the sample in a NaOH solution between 2 and 10 M and temperature between 60 and 100 °C for 24 hours to obtain a layer of sodium hydrogen titanate. Since this layer needs to be dehydrated and thickened, heat treatment is usually carried out between 600 and 800 °C to obtain a mechanically stable layer of sodium titanate. Previous studies have reported that Ti metal undergoing the alkali and heat treatment process has the potential to spontaneously form a bone apatite layer on its surface when soaked in SBF 14,15 .
There are no studies in the literature reporting alkali and heat treatment at 900 °C with an in-depth evaluation of the structural changes occurring on the surface and cross-section of the material. In this context, the objective of this work is to evaluate the formation of a bioactive surface in cp-Ti grade 4 after alkali and heat treatment at 600 °C and 900 °C.

Alkali and heat treatments
A cp-Ti grade 4 ingot was machined into disks (12.7 mm in diameter and 3.0 mm thick) and used as substrate without treatment (WT). All samples were cleaned with acetone in an ultrasonic cleaner for a period of 20 min and dried at 40 °C. The samples used for alkali treatment (AT) were then soaked in an aqueous 5M NaOH solution at 60 °C under agitation for a period of 24 h. After being removed from the solution, these samples were gently rinsed with demineralized water and dried at 40 °C. Two groups of these alkali treated samples were subsequently heated to different conditions: 600 °C (AHT600) and 900 °C (AHT900), at a heating rate of 5 °C/min, kept at this temperature for an hour without atmosphere control and then let cool naturally in the furnace. *e-mail: adriano.reis@unesp.br

In vitro test
Treated (AT, AHT600, and AHT900) and untreated (WT) samples were soaked in 17 ml of SBF. The SBF preparation and sample soaking was performed according to Kokubo and Takadama 16 . The SBF ion concentration is nearly equal to those of human blood plasma at 36.5 °C (Table 1). After soaking in the SBF for 30 days at 36.5 °C, the samples were removed from the SBF, gently rinsed with demineralized water and dried at 40 °C.

Surface and cross-section characterization
The surface of the samples were examined by using a scanning electron microscope (SEM-FEG) Tescan Mira 3 with a coupled Energy Dispersive X-ray Spectroscopy (EDS, X-MAX -Oxford); optical profiler Veeco Wyko NT1100.
The surface wettability was evaluated by contact angle measurements. The contact angles of the samples were assessed by dropping 10 µL of distilled water on specimens and measured using an image-based goniometer (Theta Lite Biolin Scientific).
Surface and cross-section hardness was measured by Nano Indentation Tester (NHT2) from Anton Paar with 10 mN of load during 10 seconds. The adherence of the formed layer was evaluated by using the VDI3198 indentation test 17 that enables the evaluation of the film's adhesion, cohesion, and brittleness. This test was performed using a Rockwell C diamond tip with a curvature radius of 200 μm and 150 kg load. This indentation corresponds to a contact pressure of around 18 GPa.
The cross-section of the samples was evaluated using a field emission scanning electron microscopy (FEG-SEM, MIRA3 -TESCAN) and a coupled Energy Dispersive X-ray Spectroscopy (EDS, X-MAX -Oxford) to analyze the elements and its distribution. Figure 1 shows the difference in the surface chemical composition of Ti before and after treatments, and it can be observed that the treatments altered the surface chemistry. The major surface compound changed from titanium (98% in WT) to oxygen (60%, 62%, and 68%, respectively for AT, AHT600, and AHT900). Moreover, in the treated samples the presence of approximately 3% sodium was observed.

Surface Chemistry
These chemical changes on the surface are due to Ti reactions with NaOH. The mechanism of nanostructures formation in titanium can be considered a simultaneous process of dissolution and precipitation. At high NaOH concentrations, general titanium dissolution occurs and sodium titanate precipitation occurs. The general dissolution reaction of Ti in aqueous alkaline solution to form a titanate anion can be represented by 18 : The titanate anions produced on dissolution may react with Na + cations in aqueous NaOH solution to produce sodium titanate. Equation 2 presents the general precipitation reaction of sodium titanate 18 : Sodium titanate formation was confirmed by TF-XRD and Raman analysis. Figure 2a shows the TF-XRD and Figure 2b shows the surface Raman profiles of Ti metal control and treated samples. From Figure 2 it can be seen that sodium hydrogen titanate (Na x H 2-x Ti 3 O 7 ) was formed on the metal surface by NaOH treatment (AT), and then was transformed into sodium titanate (Na 2 Ti 6 O 13 ), rutile and anatase by subsequent heat treatments (AHT600 and AHT900). In the AHT900 group, very sharp rutile phase peaks are observed, as a consequence of the treatment temperature (900 °C), increasing the reaction kinetics and oxygen diffusion on the titanium surface.
SEM images and EDX line profile of a cross-section of the surface layer of treated samples showed that the sodium and oxygen content gradually decreased, while Ti content increased near to a depth of 1 µm. It means that sodium hydrogen titanate (AT) and sodium titanate (AHT600 and AHT900) have around 1 µm depth.  On the other hand, AHT900 showed a profile that oxygen content didn't decrease totally at 1 µm depth, but it was kept as high as Ti until around 6 µm when then finally the oxygen content was reduced and Ti was increasing according to the substrate. The greater depth of oxygen content for AHT900 is due to its greater oxygen diffusion, which is proportional to the applied temperature. Figure 3 shows an SEM image and EDX line profile of a cross-section of the surface layer of AHT900. Figure 4 shows the changes in the surface topography of the Ti samples with respect to alkali and heat treatments.

Surface topography and morphology
The control sample (Figure 4a) showed machine marks and the treated samples showed peaks and valleys depending on variations in the treatment, being qualitatively more expressive in the AHT900 (Figure 4d). A quantitative analysis of the surface roughness was also performed based on the topographies of Figure 4. The average and standard deviations are shown in Figure 5. The quantitative analysis in Figure 5 corroborated with the qualitative analysis of Figure 4: According to Figure 5, there was a significant increase in total roughness on the microscale for AHT900 (32.3 ± 3.9 µm) compared to WT (7.0 ± 1.7 µm), AT (6.5 ± 1.7 µm), and AHT600 (6.6 ± 1.3 µm). A one-way ANOVA and Tukey analysis revealed a significant difference (p <0.05) between TQT900** and the other samples (WT*, AT*, and AHT600*), but without significant difference between WT*, AT*, and AHT600*. The literature also reported that the roughness didn't change, in similar conditions of treatment, when comparing WT with AHT600 [19][20][21] .   Figure 6 shows the evolution of surface area after treatment. The same effect presented by roughness could be observed in the increase of surface area. The peaks and valleys generated due to increased roughness provided a larger surface area than a completely flat surface. From Figure 6 it can be observed that there was a significant increase in surface area for AHT900 (299.3 ± 48.4%) compared to WT (5.3 ± 1.2%), AT (11.1 ± 4, 2%), and AHT600 (19.2 ± 3.4%).
One-way ANOVA and Tukey analysis revealed a significant difference (p <0.05) between the AHT900** and the other samples (WT*, AT*, and AHT600*). However, despite the trend of surface area growth presented in the results, the difference between the WT*, AT*, and AHT600* was not considered significant at a significance level of 0.05. By increasing the surface area it increases the area of contact between the metal and bone cells, improving the conditions for osseointegration 10,11 .
FEG-SEM and AFM analyzes were performed and it can be seen in Figures 7 and 8, respectively. From Figures 7 and 8 it can be seen that the morphology of the formed surface is homogeneous and without the presence of microcracks. It is possible to observe that WT (Figures 7a and 8a) has a relatively flat surface with the presence of machining marks. From Figures 7b-d and 8b-d were observed in the formation of a finer, porous nanoscale structure that formed on the surface of Ti after alkali treatment (Figure 7b), and this microstructure basically did not change after heat treatment at 600 °C (Figure 7c). However, after heat treatment at 900 °C (Figure 7d), a distinct nanometric needle-like structure is observed, that is related to the TiO 2 rutile phase growing, resulting in an increased surface area and roughness reported in the Figures 5 and 6. Figure 9 shows the contact angles of the samples before and after treatments, and the wettability of Ti was significantly changed from a partially hydrophilic surface in the WT group to a surface hydrophilic in the AT and AHT600 groups, and finally completely hydrophilic for the AHT900. Similar behavior is reported elsewere 19 . This change is probably due to the modification in texture (greater roughness and surface area) and chemical surface modification after alkali and heat treatments. By tailoring the functional groups available at the material surface, the surface wettability may also be altered. Hydrophilic surfaces can be constructed by the presence of polar functional groups. Sodium titanate is reported to have hydroxyl groups 22 , a strong polar group, and it may have influenced the wettability enhancement. Surface wettability of a material is of vital importance for protein adsorption and cell adhesion. When an implant material is placed inside a human body, wetting of the implant material is among the first and most important events that occur by physiological fluids. This further controls protein adsorption, followed by cell attachment to the implant surface. Thus, surface wettability is considered an important criterion that may dictate implant osseointegration 23 . Figure 10 shows the surface hardness from treated and untreated samples. The surface hardness reduced from 483±74 HV (WT) to 276±66 HV (AT), and it is related to the porous surface created by the alkali treatment. A slight surface hardening recovery can be observed in the AHT600 sample (297±43 HV), but the surface hardening increase is more evident in the AHT900 (551±101 HV). The hardness increase is related to the TiO 2 rutile phase formation, according to XRD and Raman analysis (Figure 2). Figure 11 shows the cross-section hardness profile. The analyses were made from 5 to 100 µm from the surface. AT sample did not change the hardness profile from the surface. AHT600 shows a higher hardness value (530 HV) at 5 µm, but after that, the results are similar to the substrate values. The higher hardness value at 5 µm in AHT600 is probably related to oxygen diffusion. AHT900 shows a hardness peak at around 15 µm (925 HV), and it's also related to oxygen diffusion that was more expressive when compared to AHT600, as shown by EDS profile from Figure 3. However, the substrate hardness of AHT900 (309 HV) was lower than AT (419 HV) and AHT600 (404HV). It is probably related to titanium annealing, reducing its hardness value. Figure 12 shows the indentation marks obtained on adhesion indentation test VDI3198 on (a) AT, (b) AHT600, and (c) AHT900. Both AT (Figure 12a) and AHT600 (Figure 12b) samples are classified as an acceptable failure due to and adequate interfacial adhesion. From Figure 12b (AHT600) is possible to observe some cracks around the indentation indicating a strong adherence but a brittle material, where the substrate piles up, but there is no indication of delamination. On the other hand, the AHT900 sample (Figure 12c) presented an unacceptable failure, where extended delamination at the vicinity of the imprint indicates poor interfacial adhesion, indicating a strongly adherent coating but also brittle one. The higher surface hardness from the AHT900 surface of 551 HV corroborates with this result.

In vitro test: apatite formation
Apatite formation was detected by SEM after being soaked in SBF for 30 days in the AT, AHT600, and AHT900 samples, but it was not detected in the WT sample. Figure 13 shows the SEM image of the surface of the AT sample, where it can be seen spherical particles that start to cover the entire surface. The apatite formation was more evident at the AT sample. Figure 14 shows the difference in surface chemical regarding Na, P, and Ca ions composition after being soaked in SBF for 30 days of WT, AT, AHT600, and AHT900 samples. Na, P, and Ca ions were chosen since it is an indication of apatite similar to bone mineral phase growing.
The WT sample didn't show Na + since it was not exposed to the alkali treatment. In addition to that, Ca and P ions were not detected in the WT sample, confirming that apatite didn't growth after 30 days of SBF soaking, corroborating with the SEM images. On the other hand, Na, P, and Ca ions were detected in the AT, AHT600, and AHT900 samples.
Ca and P ions were reduced with increasing temperature of heat treatment, and, at the same time, Na ions was increasing    with the heat treatment. This is because during the immersion in SBF, Na ions release from sodium hydrogen titanate (AT) and from sodium titanate (AHT600 and AHT900) to react with a hydroxyl group (negatively charged) in the SBF.      Consequently, the surface attracts the positively charged Ca + ion which can be defined as a possible nucleation area for bone-like apatite formation. The calcium and phosphate ions are consumed since apatite nuclei are formed. The apatite growing on the substrates is directly affected by heat treatment.
The apatite rate formation decreases with heat treatment, especially at 900°C. It is associated with the Na ion release rate decrease from the densified sodium titanate after heat treatment (AHT600 and AHT900) [24][25][26] .
Although AT samples forms a dense apatite after being soaked into SBF when compared with AHT600 and AHT900, sodium hydrogen titanate is too unstable for practical use. The more stable sodium titanate layer formed by the heat treatment is preferred (AHT600 and AHT900), from a practical point of view 27 .
According to in vivo tests reported in the literature, titanium alkali and heat-treated surfaces bonds directly to the bone, but titanium surface that was only alkali-treated does not. The reason for that is the unstable reactive surface layer of alkali-treated titanium 28 .

Conclusions
Both alkali (AT) and alkaline heat treatment (AHT600 and AHT900) proved to be efficient to accelerate osseointegration due to the formation of sodium hydrogen titanate and sodium titanate, thus increased wettability, roughness, and surface area.
Alkali and heat treatments AHT600 and AHT900 showed to increase the mechanical resistance on the surface, but AHT900 showed high delamination after the indentation, besides reducing the mechanical resistance of the substrate due to annealing. The surface hardness and mechanical resistance of the top surface can be used to help define for which types of implants this superficial treatment is most suitable. Therefore, among the treatments tested, AHT600 proves to be the most suitable treatment to accelerate the osseointegration of cp-Ti grade 4.
The AHT900 demonstrated a huge increase in roughness, surface area and additionally are totally hydrophilic. Future in vivo experiments are recommended to compare AHT600 and AHT900 and evaluate the long-term osseointegration, in order to improve the results.