Titanium and titanium alloys have been widely used in dentistry and orthopedics due to their excellent biocompatibility and physical properties. Although titanium is biologically inert, it reacts with oxygen in the air and spontaneously forms a dense and stable oxide layer with a thickness of 1.5 to 10 nm [1], which has a low level of electronic conductivity [2], great thermodynamic stability [3], and a low tendency toward ionformation in aqueous environments [4]. Thus, the biocompatibility of titanium and its high corrosion resistance are attributed to this surface oxide film.

Although the "osseointegration" phenomenon was first described by Branemark et al. [5], integration between titanium and tissue depends on mainly morphological connection rather than chemical bonding [6]. Albrektsson et al. [7] observed that the bone-implant interface consisted of a fibrous tissue-free zone with a 20 to 40 nm thick proteoglycan coat immediately adjacent to the titanium oxide (TiO₂) surface of the implant. Thus, many studies have been conducted to improve the bioactivity of the titanium surface. It has been reported that the activity of TiO₂ is influenced by the crystal structure, surface area, size distribution, porosity, and surface hydroxyl density [8,9].

In its native form, the TiO₂ layer has poor mechanical properties and is easily fractured under fretting and sliding wear conditions [10]. Thus, various techniques such as the sol-gel method, anodizing, and hydrothermal methods have been applied to change the layer's characteristics [11]. However, the sol-gel and anodizing methods have disadvantages such as cost and time needed. On the other hand, titanium with isothermal treatment has been found to be desirable for biomedical applications due to its controllable oxide layer thickness, good uniformity, and conformability over large areas at low cost. Furthermore, TiO₂ with a rutile structure was typically formed by thermal treatment [8]. Studies have reported that a thick oxide layer could enhance the mechanical and biological properties of titanium [12,13]. Surface topography, crystalline structure, and roughness could be changed by thermal treatment, and the transition temperatures ranged from 400℃ to 1,000℃ according to the kind of titanium used [9,14,15]. Furthermore, surface topography and roughness can affect the surface energy [16].

During implantation of titanium implants, titanium contact with blood originated from bone marrow and surrounding tissue. Because blood-surface interactions are influenced by surface energy [17], the change in the surface composition and the increase in the specific surface area could be attributed to the high surface energy. Furthermore, cell spreading increases on the substratum with higher surface energy in both the presence and absence of serum proteins [18].

The morphology of cells grown on titanium can be informative of their interactions with the surface, especially in the early phase of culture, i.e., attachment and adhesion [19]. Some studies have reported that cell attachment to material and cell proliferation is sensitive to the micro-topography of the titanium surface [20,21].

Thermal treatment is a relatively simple method for surface modification of titanium to enhance its surface characteristics and bioactivity. Therefore, the aim of this study was to evaluate the surface characteristics and the response of osteoblasts on titanium (Ti-S) discs with different thermal treatments.

The surface morphology and topography of control and thermally treated Ti-S are shown in Figs 1 and 2. There was a tendency toward an increase in the mean average surface roughness (Ra) as the treatment temperature increased (Table 2). The most noticeable surface change was observed in group III. In group III, most of the crystallites tended to be larger and the surface topography had a more spiked shape than the other groups. As shown in Fig. 3, the cross-sectioned image also showed that the thickness of the TiO₂ layer increased as the temperature increased.

Fig. 4 shows the XRD patterns of the samples. For the control, group I, and group II samples, the XRD patterns are almost the same, presenting only the peaks of titanium. For group III, there are additional peaks belonging to rutile TiO₂. This indicates that titanium was further oxidized in group III and formed oxide layers with a rutile structure instead of anatase.

The contact angles and surface energies are shown in Fig. 5 and Table 3. The water contact angle of the samples gradually decreased as the temperature increased. Although there was no statistical difference between group I and group II, group III had a statistically significantly lower contact angle than the other groups. In contrast, the surface energies (dyne/cm) significantly increased as the temperature increased.

In the MTT assay, after 24 hours of incubation, the assessment of cell viability showed that the optical density of the control tended to be higher than the other groups, but there was no significant difference (Fig. 6). However, the ALP activity increased as the temperature increased, especially in group III (Fig. 7).

Osseointegration depends on the behaviors of cells at the bone-implant interface, such as their initial attachment, adhesion, and spreading [23]. These events can be affected by the surface characteristics of the implant, including chemical composition, wettability, and topography [20]. Theoretically, pure Ti surfaces exhibit high surface energy due to the oxide layer that grows spontaneously at room temperature [4]. Such oxide surfaces are known to be spontaneously nucleated calcium phosphate layers (apatite) in contact with simulated body fluid [8]. However, inorganic anions or organic hydrocarbon contaminants from the atmosphere were absorbed into surface with high energy within seconds. As a result, the surface chemical composition might change and the surface hydrophilicity decrease. Thus, many studies have been conducted to increase the wettability and surface energy. One study [14] reported that the anatase/rutile transition could occur at temperatures ranging from 400℃ to 1,000℃ and the transition could affect the surface morphology, chemical composition, crystal structure, and surface roughness [9], which could, in turn, increase the wetability and surface energy [8]. In this study, SEM and AFM examination also revealed that the crystal structure and surface topography changed as the treatment temperature increased. Although all of the tested samples showed differences among the groups, prominent changes could be seen in group III. This indicated that the anatase/rutile phase transition occurred only in group III, and XRD patterns confirmed this.

It has been reported [24] that as the treatment temperature is raised, the oxygen solubility of the titanium matrix is enhanced. Furthermore, the plane of rutile had the highest atomic density [25]. The higher density of titanium ions on the surfaces was responsible for more active sites. This indicated that a thicker oxide layer could be created by the rutile plane. Our cross-sectioned image also showed a significant increase in the oxide layer thickness in group III compared with group I and group II.

Some studies [8,17,18] have reported increased wettability due to high surface energy and enhanced interaction between the implant surface and biologic environment. Because the Gibb's function of the thermal-treated samples with higher surface energies should be greater, absorption and reaction occur more easily on such surfaces [8]. The present study showed a decrease in the contact angle and increase in the surface energy as the treatment temperature increased. Furthermore, group III showed a prominent difference in surface characteristics and physical properties from the other groups. This indicated that the phase transition of the oxide layer is another important factor in surface modification.

Actually, the mean Ra value of all of the groups involved a minimally rough surface and the differences are not substantial, but a surface topographical change was observed. Even though both the spiky surface and the undulating surface had the same Ra, the cells on the former surface would have influence on the behavior of cell adhesion than those on the latter [26]. It is generally known that as surface micro-roughness increases, osteoblast proliferation decreases, whereas as differentiation increases on such surfaces, the cells tend to increase production of factors like transforming growth factor-β1 that promote osteogenesis while decreasing osteoclastic activity [20]. Twenty-four hours is generally considered a key point in time in the analysis of short-term cellular interactions with biomaterials in vitro [19]. In this study, at 1 day, the MTT assay showed that the cell viability was higher in the untreated group; however, micro-roughness and ALP activity increased as the temperature increased, especially in group III. The ALP activity is used as a biomarker for expressing osteoblastic activity because ALP can mediate bone mineralization by decomposing phosphate compounds and stimulating the combination of phosphate and calcium in the extracellular matrix [27].

In this study, the surface characteristics of thermally treated titanium discs were observed. Consequently, the surface characteristics and bioactivities improved as the temperature increased. This indicates that surface modification by thermal treatment could be a useful method for medical and dental implants.

In the present study, the surface characteristics of Ti-S discs treated by different thermal conditions were evaluated. The results are as follows:

The micro-morphology, crystal structure, and cellular interactions of the TiO₂ layer were influenced by the treatment temperature. Thermal treatment at 750℃ (group III) showed greater improvement in surface characteristics compared with the other groups.