1. Introduction
Using titanium (Ti) as a dental implant material has advantages owing to its excellent mechanical properties, biocompatibility, and tight bonding to the bone tissue, which is known as osseointegration. However, aesthetic, and hypersensitive reactions including allergies to Ti implants have been reported [
1,
2,
3]. An yttria-stabilized tetragonal zirconia polycrystal (Y-TZP) was used as another option as a dental implant material [
4]. However, the osseointegration of zirconia (ZrO
2) implants is controversial. Kohal et al. [
5]. evaluated 65 one-piece ZrO
2 implants incorporated in humans for three years post-operation and reported that the survival rates of ZrO
2 implants were inferior to those of Ti implants. In contrast, Roehling et al. [
6]. studied 2758 publications and reported that ZrO
2 and Ti implants showed similar soft and hard tissue integration capacities.
It is well known that surface structure and surface roughness is key to the osseointegration of dental implants [
7]. The surface treatment of Ti implants by combining the large-grit and acid-etched procedures, which is generally named SLA, produced positive effects on the activation of blood platelets and cell migration, and clinical success rate [
8,
9,
10,
11]. Surface modifications of Y-TZP implants have been reported [
12,
13,
14], and large-grit and acid etching procedures were also applied to Y-TZP implants [
12,
13,
14,
15]. However, Y-TZP is known to exhibit high mechanical strength and chemical stability [
16], and achieving a similar surface structure and roughness on Y-TZP as that on Ti is difficult. Surface roughness is a popularly used parameter for evaluating the material’s surface. Bormann et al. [
17]. compared the bone tissue response to those of the micro-structured ZrO
2 and SLA-Ti implants. ZrO
2 implants were chemically treated with a hot solution of hydrofluoric acid. They reported that the mean roughness of the SLA-Ti implants (Sa = 1.19 μm) was approximately twice that of similar ZrO
2 implants (Sa = 0.63 μm).
Apart from surface structure and roughness, the surface chemical composition also affects the osseointegration of dental implants. It is not clear which factor, surface structure, roughness, or chemical composition will contribute more to osseointegration for ZrO2 implants. It is presumed that this is the reason for the controversy regarding the osseointegration of ZrO2. To elucidate the osseointegration of ZrO2 implants, it is important to evaluate the bone response of the ZrO2 surface for a similar structure and roughness as that of Ti. If we can prepare a ZrO2 surface with a similar structure and roughness as that of Ti, we can say something about the contribution of the surface chemical composition to the osseointegration for ZrO2 implant.
The molecular precursor method is a surface modifying method, i.e., an arbitrary metal oxide film coating on the ceramic and metal surfaces achieved using a metal complex liquid [
18,
19,
20]. A thin apatite film can be deposited onto the Ti surface using the Ca–EDTA complex precursor solution [
21,
22]. The thickness of the thin apatite film was in the range of 0.6–0.8 μm. The molecular precursor method is a wet process [
23] and can only change the chemical composition of substrates without changing the surface structure and roughness because of the very thin film coating. A thin ZrO
2 film can be deposited on the Ti surface using the Zr-complex precursor solution. Therefore, we can achieve a similar surface structure and roughness of the ZrO
2 surface as roughened Ti implant surface.
In this study, we tried to distinguish the effects of the surface chemical composition, Ti or ZrO2, from those of the surface structure and roughness between Ti and ZrO2 coated Ti implants in the osseointegration process. The ZrO2 coating fabricated by the molecular precursor method changed the surface chemical composition from Ti to ZrO2 without changing the surface structure and roughness of roughened SLA-Ti. The main purpose of this study is to reveal the effects of the difference in surface chemical composition on the bone response with respect to ZrO2 and Ti with a similar surface structure and roughness. We can make clear the difference in bone responses between Ti and ZrO2 surfaces. The effectiveness of the ZrO2 surface on the bone response will be clarified in this study. To the best of our knowledge, this is the first trial to elucidate the effectiveness of the surface chemical composition of implant materials on a bone formation without considering surface structure and roughness.
2. Materials and Methods
2.1. Specimen Preparation
Two shapes of Ti samples (JIS2 type, 99.9% mass, Furuuchi Chemical, Tokyo, Japan), disks, and rectangular plates were used in this study as shown in
Figure 1. Ti disks with 12.0 mm in diameter and 1.0 mm in thickness were used for surface observation, crystal structure and atom elements analyses, phosphate-buffered saline (PBS), and simulated body fluid (SBF) immersion experiments. Ti rectangular plates with dimensions of 1.0 mm × 10.0 mm × 20.0 mm were used for zeta potential measurements and those with dimensions of 2.0 mm × 3.0 mm × 1.0 mm were for animal implantation experiments. The number of disks and rectangular plates was three for each experiment in addition to AFM measurements (
n = 4).
A rectangular plate is better to be set into the cell of the apparatus for zeta potential measurements. Regarding the animal experiment, the bone defect is box-shaped, and a rectangular plate fits this shape of the bone defect. It is difficult to fit the disks to the bone defect. In our previous studies for surface observation or SBF immersion experiments for coating materials, we standardized the shape of the sample on a disk. So, disks were used for other experiments.
Before ZrO
2 coating, the surfaces of the specimens were polished with a #1200 waterproof paper under running water. Then, specimens were subjected to SLA treatment. Namely, sandblasting was performed perpendicular to the titanium surface from a distance of 20 mm with 180 μm alumina particles at a 0.6 N/mm
2 air pressure. Then, acid etching was performed on the blasted surface with a mixture of 36% hydrochloric acid (HCl) and a 96% sulfuric acid (H
2SO
4) solution for 3 min at 70 °C [
24]. SLA-treated specimens were then ultrasonically cleaned in ethanol and distilled water for 20 min (ultrasonic cleaner; VS-100III; AS ONE, Osaka, Japan).
Ti disks and rectangular plates were coated with ZrO2 (ZrO2/Ti) using the molecular precursor method. A molecular precursor solution (25 mL) for the ZrO2 film (Zr complex ethanol solution; Zr ion concentration = 0.32 mmol/g; TFTECH, Tokyo, Japan) was poured on the Ti surface. It was then spin-coated using a spin coater (1H-D7; MIKASA, Tokyo, Japan) in the double step mode; first at 500 rpm for 5 s, and then at 2000 rpm for 30 s. Ti disks and rectangular specimens covered with precursor films were then heated at 550 °C in air for 30 min using a tubular furnace (EPKPO12-K; ISUZU, Niigata, Japan). A thin film coating of ZrO2 can be achieved on Ti disks and rectangular plates using this method.
2.2. Surface Characterization of the ZrO2 Coating
The surface structure of ZrO2/Ti was observed using scanning electron microscopy (SEM; SU1510; Hitachi High-Technologies, Tokyo, Japan). Au supper-coated specimens were observed at a 15 kV accelerating voltage.
The three-dimensional surface structure of ZrO2/Ti was observed using atomic force microscopy (AFM; Nanosurf Easyscan 2, Nanosurf AG; Liestal, Switzerland). The tapping mode was employed using a TapAL-G cantilever (Budget sensors, Bulgaria; resonance frequency = 190 kHz, spring constant = 48 N/m). The average arithmetic roughness of the three-dimensional surface (Sa) was calculated within the 1.0 μm × 1.0 μm, 2.5 μm × 2.5 μm, 5.0 μm × 5.0 μm, and 10.0 μm × 10.0 μm areas of the AFM images. The surface of roughened Ti was also observed using SEM and AFM by the same method.
X-ray diffraction was used to characterize the crystal structure of the coated ZrO2 thin film on Ti disks with a thin layer technique (XRD; attachment incidence angle θ = 0.3°, SmartLab; Rigaku, Tokyo, Japan). An X-ray source is Cu Kα which powers 45 kV × 200 mA. The presence of Ti or Zr atoms on Ti and ZrO2/Ti disks on each surface was confirmed using an electron probe micro analyzer (EPMA; JXA-8900R; JEOL, Tokyo, Japan) at an accelerating voltage of 15 kV. Cross-sectional observation was also performed. The disks were then embedded in an epoxy resin. After curing the resin, the specimens were vertically cut through the middle using a cutting machine.
The atom elements in the surface of Ti and ZrO2/Ti were analyzed by using X-ray photoelectron spectroscopy (XPS; AXIS-ULTRA; Kratos Analytical, Stretford, UK), with an X-ray source of Al Kα and a power of 15 kV × 10 mA.
Zeta potential is known to indicate the surface properties, especially charging behavior of the material surface. Next, we tried to determine the apparent zeta potentials using an electrokinetic analyzer (SurPASS 3; Anton Paar GmbH, Graz, Austria). For determining the zeta potential, an aqueous electrolyte solution (1.0 mM of KCl) was used at a pH level of 5.6. KCl solution passed through a thin slit channel formed by two Ti or ZrO
2/Ti plates with identical surfaces of rectangular plate implant. The pressure-driven flow of the KCl solution gave the streaming current and potential. The zeta potential was then calculated according to the Helmholtz–Smoluchowski equation [
25,
26]. Furthermore, the pH-dependent curves of the zeta potential were plotted by measuring the zeta potential at any pH level adjusted by 0.05 M HCl and 0.1 M KOH. The isoelectric points of Ti and ZrO
2/Ti were obtained from these curves. Measurements of apparent zeta potential at a pH level of 5.6 were performed three times. In these experiments, coefficient of variation was less than 3%, and it was concluded the measurements possessed high reliability. Thus, measurements of zeta potential at different pH were performed only once.
For determining the durability of ZrO2 coating, the ZrO2/Ti disks were immersed in PBS solution (pH = 7.4). The volume of PBS solution was 20 mL. The specimens were stored in 20 mL of PBS in a polypropylene bottle for 90 days. After immersion, the specimens were dried in a desiccator. The morphological changes in ZrO2 films were observed using SEM at an accelerating voltage of 15 kV after sputter-coating with Au.
2.3. SBF Immersion Experiments
As SBF, Hanks’ balanced salt solution (HBSS) was prepared without organic species and adjusted pH = 7.4 [
27]. Ti and ZrO
2/Ti disks were immersed in 20 mL of HBSS at 37 °C in a polypropylene bottle. After immersing for 6 and 24 h, the excess HBSS on the disks was removed using soft paper, and the disks were immediately dried in a desiccator. The surfaces of the Ti and ZrO
2/Ti disks were then observed using SEM at an accelerating voltage of 15 kV after sputter-coating with Au. The crystallographic structures of the precipitates on the Ti and ZrO
2/Ti disks were analyzed using XRD (RINT-RAPID2 MM007; Rigaku, Tokyo, Japan), with an X-ray source of Cu-Kα at a power of 40 kV × 30 mA, and FT-IR (FT/IR-430; Jasco, Tokyo, Japan) using the KBr method. The precipitated products were detached from the Ti and ZrO
2/Ti disks for FT-IR measurements.
2.4. Animal Implantation Experiment
The approval of the animal implantation experiment was obtained from the Animal Experimental Ethical Guidelines of the Tsurumi University, School of Dental Medicine (certificate nos. 19A042, 20A001, 21A009, and 22A005). Overall, 12 male Wistar rats were used for animal implantation experiments. Their weight was approximately 180 g and ages are 6 weeks. The rats were housed in pairs of two per cage at 20–25 °C in a 12 h circadian light rhythm environment. They feed on water and food ad libitum during the experiment.
Ti or ZrO
2/Ti rectangular plate implant was placed in a femur bone defect [
28,
29], and each rat received one implant. Overall, 3 ZrO
2/Ti and 3 Ti rectangular plate implants were inserted for 2- and 4-week implantation periods, respectively.
For general inhalation, anesthesia using a 3% isoflurane and oxygen mixture was used. The concentration of isoflurane was reduced to 2% during surgical manipulation. The hind limb was shaved and then operating field was disinfected. Next, operation field was locally anesthetized by the injection of xylocaine. The distal surface of the hind limb was longitudinally incised, and femur was exposed. The cortical bone defect was prepared through the cortex and the medulla. A very gentle surgical technique was employed and internal cooling with a physiological saline solution was continuously introduced during the defect preparation. The size of the defect was 1 mm × 2 mm. A rectangular plate implant was press fitted into the bone defect as shown in
Figure 2. The muscle tissue and skin were separately sutured using non-absorbable. After surgery, antibiotics (benzyl penicillin G procaine, 3,000,000 U/kg) were subcutaneously injected.
The rats were euthanized at 2 and 4 weeks after implantation using carbon dioxide gas to harvest the rectangular plate implants and the surrounding femoral bone.
After removing the excess tissue, the specimens were fixed in a 10% neutral buffered formalin solution (pH 7.4). Then, tissue blocks including implants were dehydrated through a gradient series of ethanol. Subsequently, they were embedded in methyl methacrylate. A cutting–grinding technique was applied to make non-decalcified thin sections. Cutting of the tissue blocks was performed in a direction perpendicular to the axis of the implants. After cutting and grinding by EXAKT-Cutting Grinding System (BS-300CP band system and 400CS microgrinding system; EXAKT, Norderstedt, Germany), thin sections in which thickness was approximately 50–70 μm were obtained.
Thin sections were double stained with methylene blue and basic fuchsin. Histological and histomorphometrical evaluation was performed using a light microscope (Eclipse N
i, Nikon, Tokyo, Japan, magnifications of ×40 and ×100). Bone response towards rectangular plate implant was histologically evaluated. As a histomorphometric analysis, the bone-to-implant contact (BIC) ratio and the bone mass (BM) around the implant were measured within the regions of interest (ROI). The ROIs were determined in accordance with a previous study [
30], as illustrated in
Figure 3. The BIC ratio was defined as the percentage of the length of direct bone–implant contact along with the implant surface. BM was calculated as the percentage of newly formed bone around the implant. The BIC ratio and BM values were obtained by the image analysis (WinROOF, Visual System Division Mitani, Tokyo, Japan).
2.5. Statistical Analysis
All data were calculated with the help of SPSS for Windows (SPSS Statics 17.0; SPSS, Chicago, IL, USA). The results of the surface roughness from AFM measurements were compared using the student t-test. A p-value < 0.05 was considered to be significant. The results of the BIC ratio and BM values were statistically analyzed by one-way analysis of variance (ANOVA) and Tukey’s test for multiple comparisons (p < 0.05).
4. Discussion
In this study, we prepared a ZrO2-coated Ti surface with a similar structure and roughness as that of a Ti surface using the molecular precursor method and evaluated the bone response toward ZrO2 thin films. It was revealed that the ZrO2 surface with the same structure and roughness as those of Ti promotes osteogenesis equivalent to or better than that of Ti in the early stages of bone formation.
The surface structure and roughness analysis of ZrO2/Ti using SEM and AFM confirmed that the surface appearances and roughness of Ti and ZrO2/Ti were almost similar. The molecular precursor method can change the surface chemical composition without changing the surface structure and roughness. EPMA showed that ZrO2 films were uniformly observed on all parts of the substrate.
Some studies have been reported on the osseointegration of ZrO
2 dental implants in comparison with that of Ti implants [
31,
32]. Gahlert et al. [
33]. investigated the BIC ratio and peri-implant bone density of ZrO
2 implants with a rough acid-etched surface structure and roughness and compared it with Ti-SLA implants in the maxilla of pigs. They reported that there was no difference in the bone responses of ZrO
2 implants and Ti-SLA controls. However, an unacceptable difference was observed in the surface structure and roughness of the ZrO
2 and Ti implants using confocal three-dimensional white light microscopy. The Sa values with roughened ZrO
2 implants were higher than those of Ti-SLA implants. Thus, a majority of the studies on bone responses have not focused on the same surface structure and roughness between the ZrO
2 and Ti specimens. This present study is the first study comparing the bone responses of the Ti and ZrO
2 implants with almost similar surface structure and roughness.
The XRD patterns of the ZrO2 films on the Ti disk revealed that the deposited coating comprised monoclinic and tetragonal zirconia. The broad peaks in XRD and FR-IR are due to the low crystallinity of precipitated apatite. It is well-known that the crystal structure of ZrO2 changes with temperature. Therefore, a stabilized agent, such as yttria, is often used to stabilize tetragonal zirconia at room temperature. However, the molecular precursor solution did not contain the stabilized agents. The presence of tetragonal zirconia in this study was due to the extreme thinness of the ZrO2 films. The atoms in ZrO2 films can easily move at low energies, which might reduce the phase transition temperature. This provided different results from the bulk of ZrO2.
Kokubo et al. [
34] reviewed the history of SBF, the correlation of the ability of apatite to form on various materials in SBF with their in vivo bone bioactivities, and some examples of the development of novel bioactive materials based on apatite formation in SBF and concluded that examination of apatite formation on material in SBF is useful for predicting the in vivo bone bioactivity of a material. Thus, we first employed the SBF immersion experiments and then performed the animal experiments. As a result, SBF immersion experiments suggested that the surface property of ZrO
2 promotes osteogenesis equivalent to or better than that of Ti in the early stages of bone formation. This result corresponded with that observed in the animal experiments.
The animal study suggested that with the same surface structure and roughness as that of Ti rectangular plate implants, the ZrO2 surface exhibited the promotion of new bone formation surrounding the implants. In this study, the apparent zeta potential measurement showed that the value of ZrO2/Ti was higher than that of Ti at any pH level. It is inferred that the electrostatic repulsion force of the ZrO2 against bone proteins, such as osteocalcin and osteopontin, which are negatively charged, may be smaller than that of Ti. Thus, these reactions may promote bone formation on ZrO2.
In the animal experiments, rectangular plate implants were inserted into the femurs of rats, and 2- and 4-week implantation periods were evaluated because we intended to observe the early stage of bone formation. A significantly higher value in BIC and BM for ZrO
2/Ti was observed. Albrektsson et al. [
35] argued that osseointegration corresponded to approximately 60% bone contact for Ti implants. In this study, the BIC ratio of the Ti implant was 50–70%, whereas that of the ZrO
2/Ti implant was approximately 75%. In other words, the BIC ratio of the ZrO
2/Ti implant was expected to be above the limit prescribed by Albrektsson et al.
This is a new approach to distinguishing the effects of surface chemistry from those of surface structure and roughness between ZrO
2 and Ti in the osseointegration process. This study is basic research to evaluate the bone response of the ZrO
2 surface with the same structure and roughness as that of Ti for elucidating the osseointegration of ZrO
2 implants. In contrast, it was suggested that ZrO
2-coated Ti implants also have a sufficient clinical value from the viewpoint of BIC ratio and BM value. Non-metallic implants have been reported to exhibit the risks of fracture [
36]. Therefore, ZrO
2 is often used as implant abutments only clinically. However, ZrO
2-coated Ti implants can resolve this fracture problem because of the better mechanical properties of Ti. Moreover, it is expected that a ZrO
2 film preventing corrosion on the Ti substrate will lead to a decrease in allergies and hypersensitivity problems.
Two-step rotations for spin coating were determined according to a previous report by Sato et al. [
18]. The first step of spin coating, 500 rpm (low speed), was the setting for spreading the precursor solution over the entire substrate. The second step, 2000 rpm (high speed), was aimed to blow away the excess coating precursor solution and achieve a film thickness of fewer than 1 μm.
In these experiments, we inserted the rectangular plates into the cortical limb bone of rats. This is the basic model for evaluating the bone response to surface-modified materials as previously reported [
28,
29,
30]. In dental clinics, cylindrical or screw-type implants were normally used. We already performed the thin film coating on tooth-shaped or cylindrical implants in the molecular precursor method using the spin-coating technique [
37,
38]. Moreover, bone responses after the implantation into the jaw bone should be examined. We also developed a mandibular bone defect model in rats [
15]. In the next series of our experiments, ZrO
2 coating was deposited on the cylindrical or screw-type implants and assessed the bone formation after the implantation of ZrO
2 coating implants into the bone defects in the maxilla or mandible.
According to the results of this study, it is suggested that the ZrO
2 surface with the same surface structure and roughness as that of Ti implants can promote bone formation. In other words, ZrO
2 implant surfaces are required to have the same roughness as that Ti implants despite the difficulty in processing. Moreover, sandblasting Y-TZP surfaces caused a problem of phase transformation [
39]. Hirota et al. prepared a unique roughened surface of Y-TZP by nano-pulsed laser irradiation without phase transformation, thereby enhancing the bone response [
28]. Thus, it is important to have efficient roughened ZrO
2 surfaces, such as those in Ti implants, without phase transformation to achieve ZrO
2 osseointegration.