1 Introduction

Ti and Ti alloys are important metallic biomaterials as well as stainless steels and Co-Cr alloys because of their excellent properties such as high specific strength, high corrosion resistance, and low allergenicity [1, 2]. Since they can be directly connected to living bone at an optical microscopic level, i.e., osseointegration [3, 4], they have been used as substitutes for hard tissues such as in the stems of artificial hip joints and in dental implants, where implantation in bones for the long-term is expected. However, a relatively long time is required for establishing osseointegration, and fixation between Ti implants and bones can be influenced by the state of the bones and by the implant/bone interfacial area. Surface modification is a promising way to improve bone compatibility of Ti implants [5], while leaving bulk mechanical properties intact.

Surface modification used to improve bone compatibility of Ti implants is conducted from the point of view of the morphology and phase/composition of their surfaces [6–8]. The aim of modifying surface morphology is to increase adhesion between bones and implants by an anchorage effect, while the purpose of phase/composition modification is to form either an apatite coating, or a non-apatite coating that enhances the formation of apatite [7].

We have reported amorphous calcium phosphate coating using RF magnetron sputtering [8–10], and TiO2 coating using thermal oxidation [11–13], as surface modifications of Ti and Ti alloys for biomedical applications. In this chapter, first, the current status of TiO2 coating of Ti and Ti alloys is reviewed, and then our recent work on preparation and evaluation of photocatalytic activity of TiO2 layers on Ti and Ti alloys formed by a two-step thermal oxidation process is described.

2 TiO2 Layers on Ti and Ti Alloys for Biomedical Applications

TiO2 layers on Ti and Ti alloys are reported to be effective for improving biological performance such as biomimetic growth of apatite [14], initial adhesion of osteoblast-like cells [15], bone-bonding ability [16], and bone growth [17]; in fact, Ti implants coated with a porous TiO2 layer through anodic oxidation are clinically used [18]. It is known that TiO2 can exhibit photocatalytic activity [19, 20]. Photo-induced superhydrophilicity and photocatalytic oxidation of organic compounds are closely related to biological phenomena on the TiO2 surface such as cell response, removal of hydrocarbons, and antibacterial properties [21–23].

TiO2 has three polymorphic phases at atmospheric pressure: rutile, anatase, and brookite. Rutile is a thermodynamically stable phase on the macroscale, with the stability of these phases depending on particle size. Rutile and anatase are the most stable phase for particles above 35 nm and below 11 nm, respectively, and brookite has been found to be the most stable for nanoparticles in the 11–35 nm range [20, 24]. Anatase is considered to possess excellent bone compatibility [14, 16, 25], and the photocatalytic activities of anatase [19, 26], anatase + rutile composite [27], and anatase + brookite composite [28] are reported to be high, although the precise reason for different photocatalytic activities has not been elucidated in detail [20].

Many processes for preparing TiO2 layers on Ti and Ti alloys have been investigated including chemical vapor deposition [29], physical vapor deposition [30], anodic oxidation/micro arc oxidation (MAO) [31, 32], and sol–gel [33] methods. Thermal oxidation, which is based on the reaction between an oxidizing gas and Ti at elevated temperatures, is a simple and low-cost method to prepare TiO2 layers on Ti with excellent adherence and high crystallinity, and can be applied to substrates with a complex geometry.

The thermal oxidation of commercially pure (CP) Ti in air and oxygen has been reported since the 1950s [34]. Recently, the oxide layer on Ti and/or the oxygen-dissolved and hardened layer of Ti prepared by thermal oxidation have been utilized for improving corrosion and wear resistance [35–37]. Browne and Gregson [38] showed that the air oxidation treatment at 673 K for 2.7 ks for Ti-6Al-4V implants reduced metal ion dissolution into bovine serum, particularly in the early stages.

The major product obtained in the thermal oxidation of Ti and Ti alloys has been reported to be the thermodynamically stable rutile [37]. A few reports show formation of the anatase phase in thermal oxidation of Ti and Ti alloys [39, 40]. Borgioli et al. [39] reported the formation of rutile + anatase in the oxide film on a Ti–6Al–4V alloy after glow-discharge processing in air at a total gas pressure of 0.01 atm. Lee and Park [40], using oxidation in a wet oxygen atmosphere for 10.8 ks at 683 K in combination with post-annealing in air at 773 K, showed formation of a Ti5O7 + anatase layer on a magnetron-sputtered Ti thin film. In these reports, however, the main oxidation products were rutile and Ti5O7 phases, namely, not an anatase-rich TiO2 layer.

On the other hand, it is known that anatase is formed as a main product in thermal oxidation of TiC [41, 42] and TiN [43]. Shabalin et al. [42] presented an oxidation model of TiC, in which incorporation of carbon in TiO2 stabilized the anatase phase. Meanwhile, Kao et al. [44] observed the reversible transformation between TiO having an NaCl-type structure and anatase. They pointed out the similarity between their structures: TiC and TiN exhibit NaCl type structure with a lattice constant close to TiO; the anatase formation on TiC and TiN in thermal oxidation would be related to this NaCl-type structure.

Based on the information on anatase formation on TiC and TiN under thermal oxidation, we proposed and investigated a two-step thermal oxidation process in which an anatase-rich TiO2 layer is formed on Ti and Ti alloys [11–13]. The preparation and photocatalytic evaluation of TiO2 layers formed on Ti and Ti alloys by two-step thermal oxidation are described in Sects. 6.3 and 6.4, respectively.

3 Preparation of Anatase-Rich TiO2 Layer on Ti and Ti Alloys

Figure 6.1 schematically shows the process of two-step thermal oxidation. This process consists of treatment in CO-containing atmospheres such as Ar-CO [11] and N2-CO [12] gas mixtures (first step) and subsequent treatment in air (second step). A Ti(C,O) or Ti(C,N,O) phase is formed on the Ti and Ti alloys in the first step and converted to TiO2 through air oxidation in the second step.

Fig. 6.1
figure 1

Schematic of the two-step thermal oxidation process

Full size image

The α-2θ XRD patterns (α = 0.3°, Cu Kα) of the reaction layer on CP Ti, Ti-25mass%Mo (Ti-25Mo) alloy, and Ti-25mass%Nb (Ti-25Nb) alloy after the first-step treatment in Ar-1%CO at 1,073 K for 3.6 ks are shown in Fig. 6.2. Reflections that are located close to but at a slightly higher angle than those of TiC are observed. It is known that oxygen substitution in a carbon site decreases the lattice parameter of TiC [42]; in fact, the reflections of the reaction layer are located between those of TiC and TiO as shown in Fig. 6.2. In addition, chemical composition analysis by X-ray photoelectron spectroscopy (XPS) revealed the presence of oxygen in the reaction layer as well as carbon and Ti [11]. From these results, the phase of the reaction layer is considered to be Ti(C,O). In the case of using an N2-CO gas atmosphere in the first step, a Ti(C,N,O) reaction layer was formed [12]. Figure 6.3a, b depict potential diagrams of Ti-C-O and Ti-C-N-O systems respectively, at 1,100 K [11, 12]. The chemical composition of Ti(C,O) and Ti(C,N,O) phases was arbitrarily chosen as TiC0.5O0.5 because of the lack of reliable thermodynamic data for these phases. The relationship between carbon activity (a C) and oxygen partial pressure (\( {P}_{O_2} \)) suggests that the TiC0.5O0.5 phase is thermodynamically stable at a CO partial pressure (P CO) of 0.01 atm, which corresponds to Ar-1%CO and N2-1%CO.

Fig. 6.2
figure 2

α-2θ XRD patterns of the reaction layers on CP Ti, Ti-25Mo alloy, and Ti-25Nb alloy after first-step treatment in Ar-1%CO at 1,073 K for 3.6 ks

Full size image
Fig. 6.3
figure 3

Potential diagrams for the (a) Ti-C-O system and (b) Ti-C-N-O system (N2 pressure: 0.1 MPa) at 1,100 K [11, 12]

Full size image

Figure 6.4 shows cross-sectional SEM images of the Ti(C,O) and Ti(C,N,O) layers, which were formed in Ar-1%CO and N2-1%CO, respectively. From the images, it was confirmed that the films were dense and uniform.

Fig. 6.4
figure 4

Cross sections of (a) Ti(C,O) and (b) Ti(C,N,O) layers on CP Ti after the treatments in Ar-1%CO and N2-1%CO, respectively [11, 12]

Full size image

The phase fraction in TiO2 layers formed on CP Ti, Ti-25Mo alloy, and Ti-25Nb alloy at different second-step temperatures and holding times are summarized in Fig. 6.5a–c, respectively [13]. The first-step treatment was carried out in Ar-1%CO at 1,073 K for 3.6 ks, and the reaction layer was confirmed to be Ti(C,O) single phase after the first-step. The phase fractions of anatase and rutile in TiO2 layer were calculated using the equation given by Spurr and Myers [45]. The anatase-rich TiO2 layers were formed for second-step temperatures between 673 and 873 K. At a lower temperature of 573 K, single-phase anatase was produced, but the Ti(C,O) phase remained, indicating that the oxidation reaction from Ti(C,O) to TiO2 was not completed in the second step. On the other hand, thermodynamically stable rutile was a main phase in the TiO2 layers for the higher second-step temperatures of 973 and 1,073 K. At these higher temperatures, rutile single phase was detected on CP Ti, while anatase was detected as a minor phase on Ti-25Mo and Ti-25Nb alloys. Moreover, the anatase fraction in the TiO2 layer on Ti-25Mo and Ti-25Nb alloys was higher than on CP Ti at mid-level temperatures of 773 and 873 K. The formation window for anatase in two-step thermal oxidation of Ti alloys is wider than that of CP Ti.

Fig. 6.5
figure 5

Phase fraction of the reaction layer on (a) CP Ti, (b) Ti-25Mo alloy, and (c) Ti-25Nb alloy after the second-step treatment in air [13]. Grey and white parts in small circles show rutile and anatase fractions, respectively

Full size image

Anatase irreversibly transforms to rutile at high temperatures, and the larger the valence number and ionic radius of dopants in TiO2, the more suppressed the anatase-to-rutile transformation: the transformation is enhanced by relaxation of the large oxygen sublattice through the increased presence of oxygen vacancies [46]. Figure 6.6 shows the comprehensive valence/radius plot of the anatase-to-rutile transformation, categorizing TiO2 dopants as inhibiting or promoting [46]. From this figure, Mo and Nb are likely inhibiting dopants. The incorporation of Mo and Nb into the TiO2 layer during two-step thermal oxidation process may have resulted in the presence of anatase on the Ti alloys at higher second-step temperatures and the higher anatase fraction at mid-level temperatures.

Fig. 6.6
figure 6

Comprehensive valence/radius plot of anatase-to-rutile transformation, categorizing dopants as inhibiting or promoting [46]

Full size image

The formation of the Ti(C,O) or Ti(C,N,O) single phase during the first step and optimization of the second-step temperature are required for preparing an anatase-rich TiO2 layer on Ti and Ti alloys. We varied CO partial pressures in the first-step treatment between Ar- or N2-0.1%CO and 20%CO. The rutile phase tended to form to a greater degree in the first-step treatment under higher partial pressures (up to 20 %) of CO gas, because of its high oxidizing potential.

Figure 6.7a, b show a cross-sectional TEM image and an electron diffraction pattern, respectively, of the anatase + rutile TiO2 layer formed on CP Ti after a second-step treatment at 673 K that was preceded by a first-step treatment at 1,073 K in N2-1%CO [12]. Nanoscale crystallites of anatase and rutile are observed. The thickness of the TiO2 layer formed in the second-step at 873 K was much greater than that formed in the 573–773 K range [12]. This result suggests that the formation of the rutile phase at higher second-step temperatures is also caused by direct oxidation of metallic Ti after completion of oxidation of the Ti(C,O) or Ti(C,N,O) layer.

Fig. 6.7
figure 7

(a) Cross-sectional TEM image and (b) electron diffraction pattern of the TiO2 layer formed on CP Ti after two-step thermal oxidation [12]

Full size image

Bonding strength of the anatase-rich TiO2 layer to the CP Ti substrate was evaluated by a pulling test using an Al stud, and was greater than the strength of the epoxy glue (60–70 MPa) used for bonding between the TiO2 layer and the Al stud [14]. The high bonding strength is an advantage of the thermal oxidation process over wet processes such as anodic oxidation.

4 Evaluation of Photocatalytic Activity of TiO2 Layers Formed by Two-Step Thermal Oxidation

The photocatalytic activity of TiO2 layers prepared on CP Ti, Ti-25Mo alloy, and Ti-25Nb alloy was evaluated for water contact angle, decomposition of methylene blue (MB), and antibacterial effect under UV irradiation. Figure 6.8 shows the average water contact angle obtained for UV irradiation times of 3.6–7.2 ks with an irradiance of 1 mW ċ cm−2 as a function of anatase fraction (f A) in TiO2 layers [13]. The water contact angle decreased with increasing f A, and in particular, a water contact angle less than 5° was achieved on TiO2 layers for an f A higher than 0.6. It is confirmed that anatase is effective for expression of superhydrophilicity. Meanwhile, in the case of Ti-25Nb alloy, a low water contact angle was observed even on TiO2 layers with lower f A values such as 0.2 and 0.4. The effect of Nb doping on the photocatalytic activity of TiO2 was reported to be complex [47]. Further studies on the chemical state, concentration, and distribution of Nb in the TiO2 layer are required. The water contact angle increased again under dark condition after UV irradiation; however, the hydrophobization rate was reduced in TiO2 layers with high f A.

Fig. 6.8
figure 8

Variation in water contact angle with anatase fraction (f A) of the TiO2 layer under UV irradiation [13]

Full size image

Figure 6.9 shows the variation in concentration of MB with UV irradiation time on TiO2 layers formed on CP Ti, Ti-25Mo alloy, and Ti-25Nb alloy. The values of f A in the TiO2 layers were controlled by varying the second-step temperature between 673 and 1,073 K. The rate constants for degradation of MB can be expressed by the gradients of the lines in Fig. 6.9. The anatase-containing TiO2 layers exhibited higher decomposition rates compared to the rutile single-phase TiO2 layers. The maximum rate constant was obtained at the f A value of 0.78. Bickley et al. [48] proposed a synergetic effect between anatase and rutile in order to explain a greater photocatalytic activity of anatase + rutile + amorphous TiO2 particles. Su et al. [49] reported that porous TiO2 films with an f A value of 0.6 exhibited optimal performance of photocatalytic activity and suggested a synergetic effect on photocatalytic activity: electrons excited in rutile can migrate to the conduction band of anatase, thereby effectively suppressing recombination of electrons and holes [49].

Fig. 6.9
figure 9

Degradation of methylene blue under UV irradiation on the TiO2 layer with different anatase fractions (f A) formed on CP Ti

Full size image

Antibacterial activities of a CP Ti plate coated with anatase single-phase TiO2 layer by two-step thermal oxidation (Anatase-coated, 10 × 10 × 1 mm) and an as-polished CP Ti plate (Non-coated, 10 × 10 × 1 mm) were evaluated using gram-positive E. coli (DH 5α). All specimens were ultrasonically cleaned and sterilized in ethanol for 0.6 ks before the antibacterial tests. Solution (0.1 mL) containing the bacteria at a concentration of 107 CFU ċ mL−1 diluted using 1/500 nutrient broth (NB) was dropped onto the specimen in a 24-well plate. The specimen was exposed to UV with an irradiance of 0.25 mW ċ cm−2 at 298 ± 5 K in a dark room. After 10.8 ks incubation, the dropped bacterial solution was washed out from the specimen using 4.9 mL of phosphate buffered saline (PBS). The washed-out solution (0.1 mL) including bacteria was inoculated onto a standard NB agar culture plate (ϕ = 90 mm). The number of colonies resulting from the growth of viable bacteria was counted after incubation for 64.8 ks at 310 K, and the number of viable bacteria (N sp) was calculated. The same protocol was conducted with the well plate without specimen and the number of viable bacteria (N well) was also calculated. A percentage of viable bacteria (P v) was evaluated using a following equation.

$$ {P}_{\mathrm{v}}={N}_{\mathrm{sp}}/{N}_{\mathrm{well}}\times 100 $$
(6.1)

Significant differences were statistically evaluated using Student’s t-test.

Figure 6.10 shows the percentage of viable bacteria for the Non-coated and Anatase-coated specimens. The percentage of viable bacteria for Anatase-coated was significantly lower than for Non-coated. This result indicates that the anatase layer on Ti formed by two-step thermal oxidation is useful to improve the antibacterial activity of Ti implants. Many research groups have reported antibacterial activity of an anatase layer on Ti formed by anodic oxidation [50–52]. We have showed significant antibacterial activity of an anatase layer on Ti formed by a dry process: two-step thermal oxidation.

Fig. 6.10
figure 10

Percentage of viable bacteria of CP Ti plate coated with anatase single-phase TiO2 layer (Anatase-coated) and as-polished CP Ti plate (Non-coated). (**p < 0.01)

Full size image

5 Summary

TiO2 layers formed by thermal oxidation can improve the biological properties of Ti through their photocatalytic activity. Research and development of TiO2 coatings on Ti implants for hard tissue replacement is continuing. In applications of TiO2-coated Ti implants, it would be preferable if the photocatalytic response of TiO2 layers were to visible light. Theoretical and experimental studies are needed to further improve photocatalytic activity and clarify the detailed mechanism of photocatalytic activity of TiO2 layers on Ti, which would relate to phase fraction, defect structure, and dopants. In particular, precise microscopic analyses of the structure and composition of TiO2 thin layers on Ti are needed to aid in understanding their photocatalytic properties.