Biocompatibility of Nanoscale Hydroxyapatite Coating on TiO2 Nanotubes

In this study, a highly-ordered TiO2 nanotube array was successfully fabricated on the surface of a pure titanium foil using the anodization method, and a hydroxyapatite (HA) layer was electrochemically deposited on the vertically aligned titania (TiO2) nanotube array. The TiO2 nanotubes exhibited an inner diameter ranging from 44.5 to 136.8 nm, a wall thickness of 9.8 to 20 nm and a length of 1.25 to 3.94 µm, depending on the applied potential, and the anodization time and temperature. The TiO2 nanotubes provided a high number of nucleation sites for the HA precipitation during electrochemical deposition, resulting in the formation of a nanoscale HA layer with a particle size of about 50 nm. The bond strength between the HA coating and the nanotubular layer with an inner diameter of 136.8 nm was over 28.7 MPa, and the interlocking between the nanoscale HA and the TiO2 nanotubes may have been responsible for the high bond strength. The biocompatibility assessment was conducted on Ti foil with a composite coat of nanoscale HA and the TiO2 nanotube array by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) array with mesenchymal stem cells (MSCs). The mesenchymal stem cells adhered to and spread onto the nanoscale HA layer with plenty of extending filopodia, indicating excellent biocompatibility of the HA coat, the composite coat of nanoscale HA and the TiO2 nanotube array. The findings suggest that the nanoscale HA coating on the TiO2 nanotube array might be a promising way to improve the bond strength and the compatibility of the HA layer.


Introduction
Biomedical metal materials, such as precious metals, stainless steels, cobalt alloys, and titanium alloy materials, have been widely used in the field of hard tissue replacement due to their high strength and toughness, excellent fatigue resistance, and good processing performance [1]. Titanium and titanium alloys have better mechanical compatibility than other three alloys and are currently the most promising metal materials for clinical applications. It was found that titanium and titanium alloy implants are biologically inert materials that cannot be distinguished by the immune system in living organisms [2]. A layer of cystic fiber membrane is formed on the surface of the titanium implant to shield the stimulation, so titanium alloy implants are difficult to organically combine with the bone tissue [3]. If only short-term, metal implants are acceptable. In many cases, however, long-term implantation of metal implants is required. For example, fractures require long-term placement of the

Preparation of TiO 2 Nanotube Array
Titanium foils with 99.9% purity were used as starting materials. Prior to anodic oxidation, the samples were ultrasonically cleaned in ethanol and acetone for 20 min each and then dried in a nitrogen stream. After degreasing, the samples were exposed to a solution of HF-HNO 3 -H 2 O at a volume ratio of 1:1:2 to etch the surface for 30 s. Finally, the samples were cleaned in deionized water and dried in air. The nanotubes on the Ti foil samples were anodized using an electrolyte with glycerol and water (volume ratio 1:1) containing 0.3 mol/L ammonium fluoride (NH 4 F) at an applied potential ranging from 10 to 40 V in a water bath. After the anodization, the samples were immediately rinsed in deionized water and dried in air. The nanotubes were annealed at 550 • C for 3 h in air at a heating rate of 10 min −1 to obtain the crystalline phase of TiO 2 .

Hydroxyapatite Coating
Electrochemical deposition of the HA coating on the TiO 2 nanotubes was conducted at a constant current density of 3 mA/cm 2 at 50 • C in a water bath. The electrolyte contained 0.084 mol/L Ca(NO 3 ) 2 , Materials 2019, 12,1979 3 of 12 0.05 mol/L (NH 4 ) 2 HPO 4 , 0.1 mol/L NaNO 3 and 0.0024 mol/L Na 3 C 6 H 5 O 7 . The Ca:P ratio of the electrolyte was maintained at 1.67 and the pH value was adjusted to 4.5-5.0 using HNO 3 and NH 3 ·H 2 O. After deposition, the samples were washed in a 0.1 mol/L NaOH solution for 2 h at 80 • C using an external heating system and then rinsed with deionized water and dried in air.

Biocompatibility Assessment
MTT assay was carried out on bare Ti, TiO 2 nanotubes and the nanoscale HA coating on TiO 2 nanotubes using mesenchymal stem cells (MSCs) (Cell Resource Center of Shanghai Institute of Life Science). Bare titanium specimens were used as a control. Industrial pure titanium sheets (≥99.5% purity) were cut into pieces by a wire electric discharge machine, ground with SiC sandpaper of 600#, 800#, and 1000#, and then mechanically polished. Under ultrasonic conditions, the titanium sheets were washed in deionized water, absolute alcohol, and acetone for 20 min to completely dissolve the residual grease on the surface. After the degreasing treatment, the titanium sheet was rinsed with deionized water and then air-dried. The surface of the titanium was pretreated for 30 s in a mixed solution of HF, HNO 3 and water at a volume ratio of 1:1:2, and then deionized. The surface was rinsed with water and dried for use.
The cells were cultured in low glucose Dulbecco's modified Eagle medium (DMEM) with 30% neonate bovine serum and 3.7 g/L sodium bicarbonate to adjust the acid/base balance (Sigma, AR, St. Louis, MO, USA). The medium was changed every 2 day and the cell culture was carried out in an incubator at 35 • C with 5% CO 2 and saturated humidity. The mesenchymal stem cells were seeded on each disc at a density of 1 × 10 4 and cultured for 3 and 7 days.
After 7 days of culture on the sample, the cells were fixed with 3% glutaraldehyde for 2 h, then washed three times with phosphate buffered saline (PBS), fixed with 1% acid for one hour, and washed three times with PBS for 10 minutes each. Then, all samples were dehydrated with a gradient of absolute ethanol solution at 30%, 50%, 70%, 90%, and 100% for 10 min, and finally placed in a freeze dryer to dry the water in the mesenchymal stem cells. The dehydrated and fixed cells were observed by scanning electron microscopy (SEM) after gold spraying.
A Cell Counting KIT-8 (CCK-8, Dojindo Co., Kumamoto, Japan) was used to measure the cell viability. The cell viability of the MC3T3-E1 cells of the Ti/TiO 2 /nHA coating was calculated according to the following formula [21]: where OD t is the absorbance value of the experimental group, OD nc is the absorbance value of the negative control group, and OD blank is the absorbance value of the α-MEM medium.

Characterization of Surface Properties
The phase structures present in the surface microstructure were identified by X-ray diffraction (XRD, Rigaku, Tokyo, Japan) using a Rigaku D/Max 2500PC diffractometer operated at 50 kV and 100 mA with Cu K radiation (=1.5406 nm) at room temperature. The morphology and size of the TiO 2 nanotube array and HA coating were characterized using a scanning electron microscope (JSM-6480LV, JEOL, Tokyo, Japan) equipped with energy dispersive X-ray spectrometry (EDS) at 20 kV (XL30 S-FEG). The transmission electron microscopy (TEM, JEM-2100, JEOL, Tokyo, Japan) observations were conducted using a JEM-2100 operating at 160 kV. The samples for TEM observation were scraped off of the titanium layer with a knife. They were then dispersed in an ethanol solution, dropped on a TEM copper wire with a carbon film, put through vacuum-assisted drying, and placed in an electric sample stage.

Characterization of Bond Strength of the Coating
The bond strength of the coating was tested using an electronic universal testing machine (RG4100, Pioneer Times Technology Co., Ltd., Wuhan, China). The bond strength of the HA coating was measured via the modified ASTM C-633 method [22], and the test schematic is shown in Figure 1. The test procedure began with the selection of a thin steel sheet of 1 cm × 5 cm, which was polished. To reduce the surface chemical energy of the steel sheet, it was necessary to remove the oil by ultrasonic cleaning. Then, the thick coating of the steel plate was bonded with epoxy resin glue. The bonding area was 1 cm × 1 cm and the tensile test was performed after the epoxy resin had dried for 24 h. When the electronic universal testing machine was stretched, the loading speed was set to 1 mm/min, and the highest point of the curve, that is, the maximum pulling force, was recorded during the entire stretching process.
The combined strength calculation formula is as follows: where F is the maximum tensile force, S is the bonding area, and σ b is the bonding strength of the coating. The number of samples used for each test was three.
Materials 2019, 12, x FOR PEER REVIEW 4 of 12 by ultrasonic cleaning. Then, the thick coating of the steel plate was bonded with epoxy resin glue. The bonding area was 1 cm × 1 cm and the tensile test was performed after the epoxy resin had dried for 24 h. When the electronic universal testing machine was stretched, the loading speed was set to 1 mm/min, and the highest point of the curve, that is, the maximum pulling force, was recorded during the entire stretching process. The combined strength calculation formula is as follows: where F is the maximum tensile force, S is the bonding area, and σb is the bonding strength of the coating. The number of samples used for each test was three.

Results and Discussion
Figure 2a-c shows the SEM images of the top view and cross-section of TiO2 nanotubes anodized in 0.3 mol/L NH4F + glycerol-water (volume ratio 1:1) at 20 V Ti foil after 9 h anodization. The average nanotube diameter measured was 87.2 nm and the average thickness (or length of the nanotubes) was 3.4 µm. Close observation of the microstructures of the nanotubes were conducted using a TEM, and the bright field TEM image of the nanotubes is shown in Figure 2d. The nanotubes exhibited a hollow bamboo structure. This sort of hollow structure is beneficial for inducing the deposition of Ca 2+ and PO4 3− ions, which will be discussed in a later section. In addition, the numerous nanotubes were expected to act as nucleation sites, thereby accelerating HA deposition and refining the HA grain crystals during electrochemical deposition [22].
Moreover, to obtain TiO2 nanotubes with different dimensions (i.e., wall thickness and inner diameter), a different anodization time was used to prepare the nanotubes under the anodization voltage of 20 V. Figure 3 illustrates the top views of the morphologies of the nanotubes obtained at anodization times ranging from 3 to 12 h under the anodization voltage of 20 V. It was clear that prolonging the anodization time significantly increased the inner diameter but decreased the wall thickness. The nanotubes obtained by anodization at 20 V for 3 h had inner diameters of 44.5 nm and wall thicknesses of 20.0 nm (see Figure 3a), while the TiO2 nanotubes obtained by anodization at 20 V for 6 h had inner diameters of 71.2 nm and wall thicknesses of 14.4 nm (see Figure 3b). When the anodization time was prolonged to 12 h, the inner diameter of the TiO2 nanotubes further increased but the wall thickness further decreased, reaching 136.8 nm and 10.0 nm, respectively (see Figure 3c). The increasing tube diameter may have promoted the adsorption ability of the calcium and

Results and Discussion
Figure 2a-c shows the SEM images of the top view and cross-section of TiO 2 nanotubes anodized in 0.3 mol/L NH 4 F + glycerol-water (volume ratio 1:1) at 20 V Ti foil after 9 h anodization. The average nanotube diameter measured was 87.2 nm and the average thickness (or length of the nanotubes) was 3.4 µm. Close observation of the microstructures of the nanotubes were conducted using a TEM, and the bright field TEM image of the nanotubes is shown in Figure 2d. The nanotubes exhibited a hollow bamboo structure. This sort of hollow structure is beneficial for inducing the deposition of Ca 2+ and PO 4 3− ions, which will be discussed in a later section. In addition, the numerous nanotubes were expected to act as nucleation sites, thereby accelerating HA deposition and refining the HA grain crystals during electrochemical deposition [22]. Moreover, to obtain TiO 2 nanotubes with different dimensions (i.e., wall thickness and inner diameter), a different anodization time was used to prepare the nanotubes under the anodization voltage of 20 V. Figure 3 illustrates the top views of the morphologies of the nanotubes obtained at anodization times ranging from 3 to 12 h under the anodization voltage of 20 V. It was clear that prolonging the anodization time significantly increased the inner diameter but decreased the wall thickness. The nanotubes obtained by anodization at 20 V for 3 h had inner diameters of 44.5 nm and wall thicknesses of 20.0 nm (see Figure 3a), while the TiO 2 nanotubes obtained by anodization at 20 V for 6 h had inner diameters of 71.2 nm and wall thicknesses of 14.4 nm (see Figure 3b). When the anodization time was prolonged to 12 h, the inner diameter of the TiO 2 nanotubes further increased but the wall thickness further decreased, reaching 136.8 nm and 10.0 nm, respectively (see Figure 3c). The increasing tube diameter may have promoted the adsorption ability of the calcium and phosphorus ions of the highly ordered TiO 2 arrays [23].      Figure 4 presents the results of EDS microanalysis and X-ray diffraction measurements for the nanotubes after annealing at 550 °C for 3 h in air. The nanotubes were anodized with the use of NH4F + glycerol-water electrolyte at the applied potential of 20 V for 12 h. Only Ti and O peaks were detected in the nanotube array; and the O/Ti molar ratio calculated by EDS (Figure 4a) was 2.01, which indicates a stoichiometric TiO2. The XRD results in Figure 4b show that the as-formed  Figure 4 presents the results of EDS microanalysis and X-ray diffraction measurements for the nanotubes after annealing at 550 • C for 3 h in air. The nanotubes were anodized with the use of NH 4 F + glycerol-water electrolyte at the applied potential of 20 V for 12 h. Only Ti and O peaks were detected in the nanotube array; and the O/Ti molar ratio calculated by EDS (Figure 4a) was 2.01, which indicates a stoichiometric TiO 2 . The XRD results in Figure 4b show that the as-formed nanotubular layer was an amorphous structure, which turned into an anatase structure after annealing in air at 550 • C for 3 h.   Figure 5 shows the SEM image of the HA coating electrochemically deposited on the TiO2 nanotubes at 3 mA/cm 2 with the use of a calcium-phosphorus electrolyte and calcinated at 700 °C. The top view of the HA coating shows the HA particles with an average diameter of about 50 nm. The inserts refer to the EDS microanalysis of the HA coating, the value of Ca/P of the HA was calculated as 1.57, which was very close to the 1.67 standard stoichiometric HA. HA with a Ca/P ratio ~1.57 is reported to be more beneficial for bone formation [24].   The inserts refer to the EDS microanalysis of the HA coating, the value of Ca/P of the HA was calculated as 1.57, which was very close to the 1.67 standard stoichiometric HA. HA with a Ca/P ratio~1.57 is reported to be more beneficial for bone formation [24]. Figure 6 presents the XRD patterns for the HA coatings on TiO 2 nanotubes at different deposition times of 1 min, 5 min, 10 min, and 30 min. The peaks on the XRD patterns are well fitted to the standard spectrum of HA (ICDD No. 74-566). We noted that with the increase of deposition time, the content of the HA phase gradually increased, which was evidenced by the varying intensities of the peaks from the HA phases in the XRD profiles. However, we could not determine whether the growth of HA was in monoaxial mode according to the XRD pattern [25], so SEM characterization is carried out below. Figure 5 shows the SEM image of the HA coating electrochemically deposited on the TiO2 nanotubes at 3 mA/cm 2 with the use of a calcium-phosphorus electrolyte and calcinated at 700 °C. The top view of the HA coating shows the HA particles with an average diameter of about 50 nm. The inserts refer to the EDS microanalysis of the HA coating, the value of Ca/P of the HA was calculated as 1.57, which was very close to the 1.67 standard stoichiometric HA. HA with a Ca/P ratio ~1.57 is reported to be more beneficial for bone formation [24].   to the standard spectrum of HA (ICDD No. 74-566). We noted that with the increase of deposition time, the content of the HA phase gradually increased, which was evidenced by the varying intensities of the peaks from the HA phases in the XRD profiles. However, we could not determine whether the growth of HA was in monoaxial mode according to the XRD pattern [25], so SEM characterization is carried out below. micro-rough and bioactive titanium dental implants in vivo using histometry and pull-out tests and found that the pull-out strength was 3.96 MPa [26]. This result confirmed that the TiO2 nanotubes can act as an anchor and enhance the mechanical interlocking between the HA coating and Ti substrates. We noted that the HA coating on the TiO2 nanotubes with an inner diameter of 157.7 nm exhibited a lower bond strength of 15.6 MPa. This was probably due to the oversized nanotubes having a long length and thinner tube wall thickness and, therefore, breaking easily. and bioactive titanium dental implants in vivo using histometry and pull-out tests and found that the pull-out strength was 3.96 MPa [26]. This result confirmed that the TiO 2 nanotubes can act as an anchor and enhance the mechanical interlocking between the HA coating and Ti substrates. We noted that the HA coating on the TiO 2 nanotubes with an inner diameter of 157.7 nm exhibited a lower bond strength of 15.6 MPa. This was probably due to the oversized nanotubes having a long length and thinner tube wall thickness and, therefore, breaking easily. Figure 8 shows the HA coating on TiO 2 nanotubes with different nanotube dimensions. Figure 8a shows the cross-section of the HA coating on TiO 2 nanotubes anodized at 20 V for 3 h. The nanotubes exhibited inner diameters of 44.5 nm, wall thicknesses of 20.0 nm and lengths of 1.25 µm. During electrochemical deposition, the particle size of the HA crystals (50 nm) was larger than the inner diameter of the TiO 2 nanotubes (44.5 nm), and the HA crystals attached to the ends of the TiO 2 nanotubes, rather than growing into the nanotubes. The bond strength between the HA coating and the nanotubes was weak in this deposition model. The cross-section of the HA coating on the TiO 2 nanotubes, with inner diameters of 71.2 nm, wall thicknesses of 14.4 nm and lengths of 2.25 µm, anodized at 20 V for 6 h is illustrated in Figure 8b. The HA crystals grew into the tubes because of the larger inner diameter of the nanotubes (71.2 nm) compared to the HA particles (50 nm). The TiO 2 nanotubes acted as an anchor [27], thus the HA nanoparticles grew into the nanotubes and formed strong mechanical interlocking between the nanotubes and the HA coating. The cross-section of the HA coating on the TiO 2 with inner diameters of 136.8 nm, wall thicknesses of 9.8 nm and lengths of 3.94 µm anodized at 20 V for 12 h is shown in Figure 8c, the nanotube diameter is more than double the size of the HA particles, hence, the nanoparticles not only grew into the nanotubes but also grew into the gaps between the nanotubes. The results of the bond strength test are shown in Figure 7. The bond strength of a layer of HA coating alone is only about 2.41 MPa. However, the bond strength of a layer of HA coating on TiO2 nanotubes was significantly higher. The bond strength of HA coating on TiO2 nanotubes with an inner diameter of 23.4 nm was 8.87 MPa; on TiO2 nanotubes with an inner diameter of 44.5 nm it was 12.5 MPa. The higher bond strength of 19.6 MPs for the HA on the TiO2 nanotubes was achieved when the inner diameter of the nanotubes reached 71.2 nm. The highest bond strength was 28.7 MPa for the HA on the nanotubes with an inner diameter of 136.8 nm. Aparicio et al. investigated micro-rough and bioactive titanium dental implants in vivo using histometry and pull-out tests and found that the pull-out strength was 3.96 MPa [26]. This result confirmed that the TiO2 nanotubes can act as an anchor and enhance the mechanical interlocking between the HA coating and Ti substrates. We noted that the HA coating on the TiO2 nanotubes with an inner diameter of 157.7 nm exhibited a lower bond strength of 15.6 MPa. This was probably due to the oversized nanotubes having a long length and thinner tube wall thickness and, therefore, breaking easily.   Figure  8a shows the cross-section of the HA coating on TiO2 nanotubes anodized at 20 V for 3 h. The During electrochemical deposition, the particle size of the HA crystals (50 nm) was larger than the inner diameter of the TiO2 nanotubes (44.5 nm), and the HA crystals attached to the ends of the TiO2 nanotubes, rather than growing into the nanotubes. The bond strength between the HA coating and the nanotubes was weak in this deposition model. The cross-section of the HA coating on the TiO2 nanotubes, with inner diameters of 71.2 nm, wall thicknesses of 14.4 nm and lengths of 2.25 µm, anodized at 20 V for 6 h is illustrated in Figure 8b. The HA crystals grew into the tubes because of the larger inner diameter of the nanotubes (71.2 nm) compared to the HA particles (50 nm). The TiO2 nanotubes acted as an anchor [27], thus the HA nanoparticles grew into the nanotubes and formed strong mechanical interlocking between the nanotubes and the HA coating. The cross-section of the HA coating on the TiO2 with inner diameters of 136.8 nm, wall thicknesses of 9.8 nm and lengths of 3.94 µm anodized at 20 V for 12 h is shown in Figure 8c, the nanotube diameter is more than double the size of the HA particles, hence, the nanoparticles not only grew into the nanotubes but also grew into the gaps between the nanotubes. According to the literature [11,28], the growth mechanism of the anodized TiO2 nanotubes can be explained as follows: According to the literature [11,28], the growth mechanism of the anodized TiO 2 nanotubes can be explained as follows: Titanium was firstly oxidized into TiO 2 under the anodic voltage, as shown in Reaction (1). The TiO 2 that formed on the surface of the Ti substrate reacted with F − ions and dissolved when the electrolyte contained F − ions, as shown in Reaction (2). Then pits formed on the Ti substrate. The formation of TiO 2 nanotubes was the result of a dynamic circle between the electrochemical oxide and the chemical dissolution. Eventually, the pits developed into pores and grew into nanotubes.
After anodization, the HA coating was electrochemically deposited on the TiO 2 nanotubes. Firstly, H 2 O and NH 4+ were reduced and hydrogen gas was released. The formation of hydroxide ions led to a local increase in pH in the vicinity of the cathode (reactions are given below).
The conditions for the deposition of brushite (CaHPO 4 ·2H 2 O) were fulfilled at a pH value of 9 through the formation of OH − ; then HPO 4 2− reacted with Ca 2+ to produce the deposition of CaHPO 4 ·2H 2 O on the surface of the TiO 2 nanotubes, as follows: When the coating was treated with an alkaline solution (1 M NaOH for 2 h), a hydrolysis reaction converted the brushite (CaHPO 4 ·2H 2 O) into HA (Ca 10 (PO 4 ) 6 (OH) 2 ): The SEM images of mesenchymal stem cells cultured for 1 d on pure Ti, TiO 2 nanotubes, and HA coating on TiO 2 nanotubes are shown in Figure 9. We saw that the HA coating on the TiO 2 nanotubes exhibited the highest cell numbers (Figure 9a) compared to the TiO 2 nanotubular surface ( Figure 9b) and the bare surface of pure Ti (Figure 9c), which exhibited the lowest cell numbers and no cell differentiation. Furthermore, the mesenchymal stem cells on the HA coating of the TiO 2 nanotubes and the TiO 2 nanotubular surface were healthier compared to those on the bare Ti surface. These cells exhibited a fusiform structure and adhered to the surface with a high density of filopodia extending out and anchored on the nanoscale characteristics of HA and TiO 2 nanotubes. Cai et al. [29] reported that the typical morphology of healthy mesenchymal stem cells shows adhering and spreading onto coatings with some filopodial and lamellipodial extensions, indicating a high viability of the cells. Based on the morphological observations, we concluded that the nanoscale HA coating on TiO 2 nanotubes exhibited superior biocompatibility and was more favorable for cellular activities compared to bare Ti and TiO 2 nanotubes alone.
The proliferation rate of mesenchymal stem cells after culture for 3 days and 7 days are shown in Figure 10. The initial cell density seeded on bare Ti, TiO 2 nanotubes, and the HA coating on TiO 2 nanotubes was the same (1 × 10 4 ). On Day 3, the cell number on bare Ti was significantly lower than that of the HA coated surface and the TiO 2 nanotubes. In addition, the cell number on the HA coating on TiO 2 nanotubes was higher than that of the TiO 2 nanotubes. On Day 7, the HA coating exhibited greater cell growth than that of the TiO 2 nanotubes and bare Ti. The cell viability displayed a sequence: HA on TiO 2 nanotubes > TiO 2 nanotubes > bare Ti. We concluded that the nanoscale HA coating on TiO 2 nanotubes possessed excellent biocompatibility.
surface. These cells exhibited a fusiform structure and adhered to the surface with a high density of filopodia extending out and anchored on the nanoscale characteristics of HA and TiO2 nanotubes. Cai et al. [29] reported that the typical morphology of healthy mesenchymal stem cells shows adhering and spreading onto coatings with some filopodial and lamellipodial extensions, indicating a high viability of the cells. Based on the morphological observations, we concluded that the nanoscale HA coating on TiO2 nanotubes exhibited superior biocompatibility and was more favorable for cellular activities compared to bare Ti and TiO2 nanotubes alone.  The proliferation rate of mesenchymal stem cells after culture for 3 days and 7 days are shown in Figure 10. The initial cell density seeded on bare Ti, TiO2 nanotubes, and the HA coating on TiO2 nanotubes was the same (1 × 10 4 ). On Day 3, the cell number on bare Ti was significantly lower than that of the HA coated surface and the TiO2 nanotubes. In addition, the cell number on the HA coating on TiO2 nanotubes was higher than that of the TiO2 nanotubes. On Day 7, the HA coating exhibited greater cell growth than that of the TiO2 nanotubes and bare Ti. The cell viability displayed a sequence: HA on TiO2 nanotubes > TiO2 nanotubes > bare Ti. We concluded that the nanoscale HA coating on TiO2 nanotubes possessed excellent biocompatibility.

Conclusions.
For this paper, we fabricated a highly-ordered TiO2 nanotube array on the surface of a pure titanium foil using the anodization method. Then, a hydroxyapatite (HA) layer was electrochemically deposited on the vertically aligned titania (TiO2) nanotube array. The effect of the inner diameters on the bond strength between the nanoscale HA and the TiO2 nanotubes was investigated.
(1) Highly-ordered TiO2 nanotubes with different dimensions were successfully fabricated on titanium using the anodization method. The nanotubes were had inner diameters ranging from 44.5 to 136.8 nm, wall thicknesses ranging from 9.8 to 20.0 nm, and tube lengths ranging from 1.25 to 3.94 µm, depending on the applied potential, and the anodization time and temperature.
(2) A layer of hydroxyapatite (HA) with the average particle size of 50 nm was electrochemically deposited on to the anodized TiO2 nanotubes. The TiO2 nanotubes promoted the formation of the nanoscale HA coat by providing a high number of nucleation sites during electrochemical deposition.

Conclusions
For this paper, we fabricated a highly-ordered TiO 2 nanotube array on the surface of a pure titanium foil using the anodization method. Then, a hydroxyapatite (HA) layer was electrochemically deposited on the vertically aligned titania (TiO 2 ) nanotube array. The effect of the inner diameters on the bond strength between the nanoscale HA and the TiO 2 nanotubes was investigated.
(1) Highly-ordered TiO 2 nanotubes with different dimensions were successfully fabricated on titanium using the anodization method. The nanotubes were had inner diameters ranging from 44.5 to 136.8 nm, wall thicknesses ranging from 9.8 to 20.0 nm, and tube lengths ranging from 1.25 to 3.94 µm, depending on the applied potential, and the anodization time and temperature.
(2) A layer of hydroxyapatite (HA) with the average particle size of 50 nm was electrochemically deposited on to the anodized TiO 2 nanotubes. The TiO 2 nanotubes promoted the formation of the nanoscale HA coat by providing a high number of nucleation sites during electrochemical deposition. (3) The dimensions of the TiO 2 nanotubes had a significant impact on the bond strength of the HA coat on the nanotubes. The HA coat, deposited on the nanotube, with an inner diameter of 136.7 nm exhibited the highest bond strength, reaching 28.7 MPa. The TiO 2 nanotubes with proper dimensions were favorable to the growth of HA into the nanotubes and the gaps between the nanotubes, leading to a strong mechanical interlocking effect between the HA and the nanotubes. The bond of the long oversized nanotubes with thinner tube walls could be broken easily, deteriorating the bond strength of the HA coat.