Tribology and in-vitro biological characterization of TiO2 addition on ceria stabilized zirconia-toughened alumina (CSZTA) composite

ABSTRACT The effect of TiO2 addition on tribology and in-vitro biological characteristics of ceria-stabilized zirconia-toughened alumina (CSZTA-TiO2) ceramic composites were examined and presented in this study. Through the powder metallurgy route, the TiO2 added to CSZTA samples were synthesized and sintered (air environment). Additionally, the pin on the disc machine was used to examine the wear characteristics of sintered samples. Adding TiO2 to CSZTA enhances the tribological properties compared to pure ceria-stabilized zirconia-toughened alumina (CSZTA). In addition, the samples (CSZTA and CSZTA-TiO2) were subjected to aging. There is no monoclinic-phase transformation (no degradation) in the CSZTA-TiO2 sample after 100 h of testing, confirming its great resistance to LTD. The bioactivity of developed CSZTA and CSZTA-TiO2 samples was studied using simulated body fluid (SBF). After chemical treatment, it was shown that the composite would create an apatite layer that resembled bone when soaked in a simulated body fluid with ion concentrations equal to human blood plasma. These results suggest that it may generate apatite within a live organism and connect to the bone via the apatite layer.


Introduction
Alumina ceramics exhibit superior properties due to possession of low density, high hardness, high stiffness, good wear resistance, strong biological properties, corrosion resistance, and thermal and chemical inertnesses, which make them extensively used in orthopedic load-bearing applications [1][2][3]. Few studies have reported that alumina exhibits low fracture toughness, which limits its applicability [4][5][6]. After that, it is strengthened by the incorporation of tetragonal zirconia (often yttria) as a second phase into alumina and results in the formation of zirconia-toughened alumina (ZTA) [7,8]. The major concern with yttria-stabilized zirconia (YSZ) is its degradation at room temperature, known as low-temperature degradation (LTD) [9]. Low-temperature degradation in YSZ ceramics occurs during the phase transformation of zirconia from tetragonal (T) to monoclinic (M) phase in the presence of water vapor without any applied stress [10]. The LTD is a progressive, water-driven process that begins on the surface and penetrates deeper, resulting in surface roughening and micro cracking [11][12][13][14]. Watanabe et al. [15] and Tsukuma et al. [16] reported that adding a stabilizer and decreasing grain size in YSZ ceramics reduced LTD but couldn't eliminate it. Since ceria is isovalent to zirconia, it produces fewer defects (anionic vacancies) during zirconia atom replacement (Zr +4 ) than aliovalent yttria (Y +3 ), which improves LTD [17][18][19]. However, ceria stabilized zirconia (CSZ) exhibits poor sinterability due to the absence of defects caused by anionic vacancies, requiring a higher temperature to attain maximum density. The poor mechanical characteristics of CSZ result from its larger grain size (3-5 µm), which is caused by an increase in temperature [16]. The addition of TiO 2 decreased the sintering temperature and enhanced the mechanical properties of CSZTA [20][21][22][23] In the load-bearing application, the material for orthopedic implants must have high wear resistance, stiffness, toughness, biocompatibility, and chemical stability. It is well established that the wear resistance of ceramic material is determined mainly by its hardness, influenced by composition, grain size, and porosity [24]. In this context, grain refinement seems to be the essential mechanical property of polycrystalline ceramics that improves wear resistance. Cho et al. found that decreasing grain size from 20 to 4 µm increased the sliding wear resistance of Al 2 O 3 ceramics [25,26]. The improved tribological properties of the ZTA system were examined under dry sliding conditions with the addition of metal oxides [27,28]. In addition, the two-step sintering on the CSZTA−4TiO 2 system reduced the grain size, which improved the hardness, toughness, and phase change from the T-M phase and the LTD [29,30].
Bioinert CSZTA ceramics couldn't directly bond with living bone [31,32]. It must be bioactive and bond with the host's live bone by generating a bone-like apatite layer on its surface to replace bone in load-bearing applications [31]. Masaki Uchida et al. discovered bioactivity in zirconia ceramics by chemically treating them with HCl, H 2 SO 4 , and H 3 SO 4 [33,34]. In addition, the biological properties of TiO 2 ceramics would make an ideal implant material [35]. As a result, a ceramic made by combining CSZTA and TiO 2 would possess the best qualities for implant application. It is the reason that TiO 2 tends to absorb water at the surface and form titanium hydroxide groups. According to reports, the basic Ti-OH groups stimulate apatite nucleation and crystallization in SBF. Many titanium hydroxide groups on the surface of titanium oxides produced by several methods, including the sol-gel process and treatments of metallic titanium with H 2 O 2 or alkali. These hydroxide groups develop the apatite, which resembles bone [36][37][38].
Based on the extensive literature, it is found that limited studies have been conducted on tribological behavior, LTD, and bioactivity of titania-added ceria stabilized zirconia-toughened alumina (CSZTA-TiO 2 ). The present study, CSZTA-TiO 2 ceramics are compared to undoped CSZTA ceramics in terms of tribological characteristics, aging resistance, and bioactivity.

Sample preparation
The co-precipitation method was used to prepare the CSZTA and CSZTA-TiO 2 nanopowders, as mentioned in the previous article [20]. Al (NO 3 ) 3 .9 H 2 O, ZrO (NO 3 ) 2 . xH 2 O, Ce (NO 3 ) 3 . 6 H 2 O and TiCl 4 (Loba Chemicals, India) precursors mixed with distilled water at the appropriate weight percentages. The hydroxides were obtained by adding ammonium hydroxide (SRL, India). Complete precipitation required a pH level of 9-10. After that, the precipitate was washed, filtered, and dried for 12 h at 110°C. To transform the dried powders into oxide powder, they were calcined at 1000°C for 2 h. After calcination, the dried powders were compressed into pellets of ϕ 10 mm using a uniaxial hydraulic press. The green pellets were sintering through two-step sintering at 1350°C for 1 h and 1250°C for 20 h [39].

Evaluation of tribological properties
A pin on disc wear equipment (Ducom, India) was used to evaluate the samples' wear properties at room temperature as per the ASTM-G99 standard [40]. The counter body was made of an alumina ceramic disc with a density of 99.7%, and the sample was used as a holder. The test loads were chosen following prior research, such as loads experienced by the femoral head in the hip socket when doing routine activities like walking, jogging, etc [41]. The test was carried out at a speed of 200 rpm over 1000 m with loads of 20, 30, and 50 N. The frictional force and wear depth data were obtained from the frictional force sensor (load cell) coupled with a linear variable differential transducer (LVDT). As per the ASTM G−99-17 standard methods, the tribological properties of CSZTA and CSZTA −4TiO 2 samples were evaluated. According to earlier studies, tribology testing of biomaterials was done following ASTM standards [42,43]. CSZTA−4TiO 2 samples were sintered using an optimized two-step sintering schedule (TSS) of 1350°C for 1 h and 1250°C for 20 h, whereas CSZTA samples were sintered using a single-step sintering schedule of 1600°C for 3 h [39].

Hydrothermal aging study
According to previous research, aging samples hydrothermally for 1 h at 134°C and 2 bar pressure is similar to 3-4 years in vivo [10,12]. The samples were heated in an autoclave using deionized water for the experiment. In order to ensure that the samples were free from contamination, polishing, and ultrasonic cleaning were done before the test. After that, the experiment was performed for a total duration of 100 h. The samples were thoroughly cleaned with distilled water and dried, and then XRD patterns originated after every 10 h of hydrothermal treatment.

Evaluation of bioactivity
In vitro, apatite formation in a simulated body fluid (SBF) with ion concentrations almost similar to human blood plasma can be used to determine the bioactivity of bioceramics [43][44][45]. Before being immersed in the SBF, the samples were incubated at 90°C in a 5 M H3PO4 solution for 4 days. After that, they were removed from the solution, cleaned with distilled water, and dried at room temperature. Every other day, the SBF was replaced, and the samples underwent characterization after 7 days. The samples were taken out and carefully cleaned with deionized water before being dried in a desiccator. SEM was used to examine the apatite layer formation.

Mechanical characterization of CSZTA and CSZTA−4TiO 2 composite
In a previous research [39], the authors evaluated the mechanical properties of CSZTA, and CSZTA−4TiO 2 composites are shown in Table 1. It is found that adding TiO 2 and a two-step sintering process improves the mechanical properties of the CSZTA system. These samples were chosen for further tribology, aging, and bioactivity investigation. Figure 1 depicts the variation of wear depth and coefficient of friction with the time or distance traveled of CSZTA and CSZTA−4TiO 2 samples. It can be observed from the plots Figure 1(a) that as the load increased, the wear depth of sintered samples increased, but the friction coefficient decreased as shown in Figure 1(b). The CSZTA sample shows the average coefficient of friction was 0.42, 0.40, and 0.39 at 20N, 30N, and 50N load conditions, respectively. It was also found that adding TiO 2 reduces friction's cocoefficiency by 0.35, 0.33, and 0.32 for different load conditions of 20N, 30N, and 50N, respectively. It is believed that the two-step sintering (TSS) process improved the relative density (97%), decreased grain size (0.96 µm) and mechanical properties such as hardness and fracture toughness. The TSS approach successfully modifies the interaction between grain boundary migration and diffusion, increasing densification without grain growth [39]. The improved mechanical properties caused to improvement in the wear resistance of CSZTA samples.

Evaluation of tribological properties
The wear resistance of ceramics depends on the following parameters(i) grain refinement, (ii) increased hardness of composite, (iii) improved grain boundary toughness (iv) decreased internal residual stresses [46]. It observed that the added TiO 2 improved the tribological properties due to increased hardness, toughness and reduced grain size. In other words, the addition of TiO 2 in CSZTA sample caused the form of softer secondary-phase ZrTiO 4 (ρ = 4.86 g·cm − 3, E = 130-150 GPa) in comparison to base phases alumina (ρ = 4.2 g·cm − 3, E = 360-400 GPa) and zirconia (ρ = 6.2 g·cm − 3, E = 200-220 GPa) [47]. This softer phase readily degraded during sliding motion at the worn track and created wear debris with the dislodgment of nearby alumina particles, which acts as a roller between mating surfaces and lowers the co-efficient of friction [42,48]. The addition of TiO 2 decreased the CSZTA composite's coefficient of friction to 17% [49].
The tribological properties such as wear volume and specific wear rate were obtained after the wear test, as shown in Figure 2. The CSZTA sample showed higher wear volume and specific wear rates were 0.124(mm 3 ) and 3.89 × 10 −6 (mm 3 /N-m), respectively, under a 50N load condition. The TiO 2 added to the CSZTA system reduced the wear volume and specific wear rate of 0.08 mm 3 and 2.86 × 10 −6 mm 3 /N-m, respectively, at the 50 N load condition. The fractured grains and grain pull-out led to higher wear in the CSZTA sample than in the CSZTA−4TiO 2 sample. Jansen et al. [50] obtained a similar observation in the case of ceriadoped Y-TZP.
After the wear test, the worn surfaces of the samples were examined using SEM, as seen in Figure 3. The SEM images of CSZTA revealed the presence of microcracks, grain pull-out, and    fractured grains, as depicted in Figure 3(a,b) at 30N and 50N load conditions. It shows several grain pullouts and significant damage to the CSZTA sample at 50N. Due to grain pull-outs, debris forms between the specimen and the counter body, causing abrasive wear. On the other hand, the wear mechanism for CSZTA−4TiO 2 samples appears different, as shown in Figure 3(c,d) at 30N and 50N load conditions. Few grain pull-outs and fractured grains indicated that only minor wear occurred in TiO 2 added to the sample at 50N. Due to the increased tetragonal-phase stability of small particle sizes, no fractured grains exist [51]. In other words, fish-scale morphology is seen as micro-cracks within the perpendicular direction of the sliding direction, as shown square dotted mark in Figure 3(d). Luo et al. found that the fish-scale morphology occurs due to brittle fracture in the TiO 2 -added sample due to smaller grain size. Previous literature has described a similar morphology of worn surfaces with TiO 2 added in zirconia [52,53] Figure 4 displays the XRD results of the CSZTA samples following a 100-h hydrothermal treatment. The CSZTA samples showed that the rate of the phase change from tetragonal (T) to monoclinic (M) increased with hydrothermal aging time. Monoclinic transition in zirconia-alumina ceramics could be caused by internal stresses caused by differences in thermal expansion coefficients between the alumina and zirconia phases [54]. In addition, the increased grain size of the CSZTA samples caused the phase change from the T-M phase to occur [55,56]. Figure 5(a,b) depict the Rietveld refinement parameters and phase quantities of CSZTA and CSZTA−4TiO 2 samples after 100 h of hydrothermal treatment. After 50 h, the phase transformation (T-M) was seen in the CSZTA sample.

Effect of aging on the developed material's phase and microstructure
In contrast, even after 100 h of hydrothermal treatment, adding TiO 2 to the CSZTA system revealed no phase change. The CSZTA−4TiO 2 composite has  a substantially lower rate of aging than the 20%Al 2 O 3 -Y-ZrO 2 composite [57,58]. In other words, no phase transformation was observed even after 100 h in CSZTA−4TiO 2 due to the smaller grain size (0.9 µm). The tetragonal phase stability was enhanced by the presence of TiO 2 and its ability to act as a stabilizing agent [59].
The SEM images of CSZTA and CSZTA−4TiO 2 samples after different intervals of aging studies as shown in Figure 6. There are no grain pull-outs formed up to 25 h in the CSZTA sample during the aging study. More grain pull-outs were formed after 50 hours of hydrothermal studies, as shown in Figure 6(b,c). Due to phase transformation, volume increased (~4), causing surface uplift and increased friction between two surfaces, resulting in wear [12,60]. This wear leads to grain pull-outs in the material. No significant grain pull-outs were observed even after 100 h of hydrothermal study of the CSZTA-TiO 2 system, as shown in Figure 6(d,e). In the CSZTA sample, the addition of TiO2 caused a reduction in grain size, causing LTD to be resistant [39]. Figures 7 and 8 show the SEM images of sintered specimen surfaces after immersion in SBF. It was found that the samples treated with 5 M H 3 PO 4 seemed to have a thicker apatite layer. The elemental compositions of the formed mineral layer were identified using EDX analysis. It was observed that the CSZTA sample showed lower Ca/P = 1.5 for the 7 days, and its value increased to 1.61 after 14 days, as shown in Figure 7(a,c). As illustrated in Figure 7(b,d), the Ca/P ratio in the CSZTA−4TiO 2 system increased from 1.61 to 1.64 after 7 and 14 days. SBF-treated samples increased the mineral layer thickness for the TiO 2added sample compared to the CSZTA sample and exhibited cauliflower-like minerals for the 21-and 27day samples. The TiO 2 -added sample obtained a Ca/P ratio of 1.65 and 1.67 after 21 and 27 days. Still, the CSZTA sample showed a lower Ca/P ratio of 1.63 and 1.65 after 21 and 27 days, as shown in Figure 8(a,c). The larger alumina concentration in CSZTA resulted in a lower apatite layer (Ca/P). It was found that adding TiO 2 to the sample increased the Ca/P ratio, illustrated in Figure 8(b,d). According to P. Li et al., Al 2 O 3 is positively charged at physiological pH, while TiO 2 is slightly negatively charged [61,62]. In this regard, the Al 2 O 3 interacts with the negatively charged HPO 4 2-,

Bioactivity study of developed materials
whereas TiO 2 interacts with only positively charged Ca 2+ ions, resulting in a higher Ca/P ratio obtained in TiO 2 added sample than the CSZTA sample [32]. In other words, TiO 2 has a dielectric constant value (ɛ = 48), and Al 2 O 3 contains a dielectric constant value (ɛ = 10-11) [62]. TiO 2 has a dielectric constant nearly comparable to water (ɛ = 78), whereas Al 2 O 3 has a dielectric constant much lower than water and TiO 2 . Al 2 O 3 ceramics are less hydrophilic than other bio ceramics [63], with a lower wettability angle than TiO 2 [64]. As a result, solvated cations are less likely to approach Al 2 O 3 substrates, resulting in reduced Ca/P. Additionally, it is envisaged that functional groups like Zr-OH, Ta-OH, Ti-OH, and Si-OH will serve as apatite nucleating sites in bioactive ceramics [31,33,34]. The chemical treated with a 5 MH 3 PO 4 solution causes many zirconium hydroxides (Zr -OH) bonds to form on the surface. The calcium ions in the SBF react with these zirconium hydroxides (Zr-OH) groups to create calcium zirconate. Then, it reacts with phosphate ions to produce an appetite layer highly concentrated in calcium and phosphate ions [65,66].

Conclusions
The CSZTA−4TiO 2 sample obtained improved wear resistance due to finer grain size and tetragonal stability. Even after 100 h of study, the CSZTA−4TiO 2 sample showed no phase shift, indicating its excellent resistance to low-temperature deterioration. The bioactivity of CSZTA and CSZTA−4TiO 2 was increased after chemical treatment with a 5 M H 3 PO 4 solution. CSZTA −4TiO 2 is a possible substitute in biomedical applications because of its better tribological characteristics and effective resistance to LTD and biocompatibility.