Introduction

Nanotube arrays (NTAs) prepared by electrochemical anodization have captured tremendous interest in energy1,2,3, environment3,4,5, biomedicine3,6,7,8,9, as well as many other fields. The growth of NTAs during anodization is mainly ascribed to the presence of the appropriate amount of F− ion and the underlying mechanism has been summarized in recent works3,10. Based on the same principle, NTAs have been fabricated on a variety of valve metals3,10, including titanium (Ti), zirconium (Zr), hafnium (Hf), tantalum (Ta), niobium (Nb), vanadium (V) and so on. The most widely investigated one is TiO2 NTAs because of the excellent semiconducting property and biocompatibility of TiO23.

In order to improve the performance of TiO2 NTAs and widen their applications, elemental doping is a feasible approach. In this respect, a straightforward and efficient way is an anodization treatment of Ti-X alloys where X is the expected dopant. Doped TiO2 NTAs have hitherto been produced by anodization of various alloys such as TiAu, TiPt, TiZr, TiW, TiMo, TiNb, TiAl, TiCu, TiAg, TiFe, TiMn, TiRu, TiHf, TiV and TiNi to attain specific functions3,11,12,13,14,15,16,17,18,19. For example, Au doped TiO2 NTAs have been proven to be a highly efficient photocatalyst to produce H2 from splitting water11. Ag doped TiO2 NTAs show long-term antibacterial ability and simultaneously good biocompatibility thus being potential biomedical implant coatings13. The anodization behavior of Ti-X alloys depends closely on the alloying element X. For instance, TiRu alloy with small concentrations of Ru (a typical non-valve element that cannot form a compact electronic insulating passive layer when anodized) of less than 5 at% allows the formation of well-defined NTAs16, whereas regular NTAs can be formed on the TiZr alloy in a wider concentration range of Zr (a typical valve element) of up to 50 wt%20. NTAs appear to be relatively easy to form on Ti alloys with valve alloying elements in a wide content range of the alloying elements, but much harder on Ti alloys with non-valve alloying elements confined to only a small concentration of the alloying element. Fabrication of well-defined NTAs with a large content of non-valve metals is thus challenging.

It is of interest to study the anodization behavior of nearly equiatomic NiTi alloy (50.8at% Ni) in terms of the following aspects. Firstly, NiO and NiO/TiO2 composites have many unique properties that can be exploited in electrochemical energy storage, photocatalysis, biosensors, resistance switching and electrochromic fields19,21,22,23,24. In most of these applications, usually a large surface area per unit mass is needed to achieve a high efficiency. Ni doped NTAs are expected to cater to this requirement. Secondly, NiTi alloys have been widely used in biomedical implants due to its notable shape memory and super-elastic properties, but further improvement in terms of bioactivity is needed to achieve more satisfactory tissue integration. NTAs formed by anodization may constitute a nano-functional coating on NiTi based implants to improve the biological performance25,26. Thirdly, the work will provide insights into the anodization of Ti alloys with a large content of non-valve elements. Since the report on anodization of NiTi alloy to form Ni-Ti-O NTAs by Frank and co-authors19, there have been very limited publications regarding this topic and in particular, precise structural control of Ni-Ti-O NTAs is quite difficult due to the large Ni content26,27,28. Herein, the influence of various anodization parameters such as anodization duration and voltage as well as the composition and temperature of electrolyte on the formation and structure of the Ni-Ti-O NTAs is investigated systematically. The feasibility of producing well-controlled Ni-Ti-O NTAs on NiTi alloy by modulating these parameters is demonstrated and their potential applications to the energy storage as well as environmental and biomedical fields are explored.

Methods

The NiTi alloy (50.8 at % Ni) was cut into small sheets (7.5 mm × 7.5 mm × 2 mm) and polished to a mirror finish. A copper wire of 100 mm long was attached to the non-working face of the NiTi sheet with the copper conductive tapes, which was then sealed with the silicone rubber. The distance between the working electrode and the counterpart of Pt foil (10 mm × 10 mm × 0.2 mm) was set at 20 mm. Each sample was anodized in the electrolyte of 100 ml with a DC power supply. The electrochemical cell was covered with a PVC film during anodization to prevent the dissolution of H2O from atmosphere to the electrolyte. The temperature of the electrolyte was controlled by a thermostatic bath. After anodization, all samples were washed with distilled water for 30 s immediately. To investigate the effects of the anodization voltage, the samples were anodized at 30°C and 5 V, 10 V, 15 V, 20 V, 25 V, 30 V, 40 V, 60 V and 90 V in ethylene glycol (Sinopharm Chemical Reagent, China) containing 0.2 wt% NH4F (Sinopharm Chemical Reagent, China) and 1.0 vol% H2O. To investigate the influence of the anodization temperature, the samples were anodized at 25 V and 10°C, 20°C, 30°C, 40°C and 50°C in ethylene glycol supplemented with 0.2 wt% NH4F and 0.5 vol% H2O for 1 h and to investigate the effects of the anodization time, the samples were anodized at 30°C and 25 V in ethylene glycol containing 0.2 wt% NH4F and 1.0 vol% H2O for 5 min, 15 min, 30 min, 60 min and 120 min. To study the influence of the H2O concentration, the samples were anodized at 30°C and 25 V in ethylene glycol with 0.2 wt% NH4F and 0.0 vol%, 0.25 vol%, 0.5 vol%, 1.0 vol%, 1.5 vol% and 2.0 vol% H2O for 1 h and to investigate the influence of the F− content, the samples were anodized at 30°C and 25 V in ethylene glycol supplemented with 1.0 vol% H2O and 0.025 wt%, 0.05 wt%, 0.1 wt%, 0.2 wt%, 0.4 wt%, 0.8 wt% and 1.0 wt% NH4F for 1 h.

Field-emission scanning electron microscopy (FE-SEM, JSM-6700F, JEOL) and high resolution transmission electron microscopy (HR-TEM, JEM-2100F, JEOL) were conducted to examine the morphology and microstructure of the anodized samples. The elemental composition and chemical states were determined by X-ray photoelectron spectroscopy (XPS, K-Alpha, Thermo) employing monochromatic Al Kα radiation (6 mA, 12 kV and 1486.68 eV). The three-dimensional morphology of the corrosion micropits was assessed by laser scanning confocal microscopy (LSCM, OLS4100, OLYMPUS) and an electrochemical workstation (CS350, CorrTest) was employed to acquire the current-time curves during anodization.

The electrochemical capacitance performance of the NTAs is assessed. The electrode composed of NTAs on the NiTi alloy with an exposed area of 0.5 cm2 was used in the electrochemical tests conducted in a 1 M KOH solution at room temperature on the electrochemical workstation (CS350, CorrTest). A three-electrode system with SCE as the reference electrode and Pt foil as the counter electrode was used. The electrochemical behavior was characterized by CV and GCD tests. The CV tests were performed in the potential range between 0 and 0.5 V (vs. SCE) at scanning rates of 5, 20, 50, 100, 200, 500 and 1000 mV/s and the GCD curves were obtained between 0 and 0.5 V (vs. SCE) at a current density of 0.4 mA/cm2. The electrochemical impedance spectroscopy (EIS) measurement was conducted at open circuit potential (OCP) with a sinusoidal perturbation potential of 10 mV within a frequency range of 100 kHz to 1 mHz.

The glucose sensing ability of the Ni-Ti-O NTAs was measured by cyclic voltammetry on the electrochemical workstation at room temperature. NaOH solutions (0.1 M) of 50 ml containing different concentrations of glucose were used as the electrolyte and the Pt foil (10 mm × 10 mm × 0.2 mm) was used as the counterpart. All potentials were referenced to the Ag/AgCl reference electrode in the measurement. The mirror polished NiTi alloy was also used as a comparison. The electrodes were activated in the NaOH solution (0.1 M) at a scanning rate of 50 mV/s for 100 cycles before the electrochemical tests.

In the biological experiments, the cell viability was evaluated by live/dead staining. After autoclave sterilization, each sample was placed on one of the wells on a 24-well cell culture plate, seeded with MC3T3-E1 subclone 14 pre-osteoblastic cells at a density of 2.0 × 104 cells/cm2 and cultured in α-MEM (Gibco) supplemented with 10% calf bovine serum (CBS), 100 units/ml penicillin and 100 μg/ml streptomycin in a humidified atmosphere of 5% CO2 at 37°C. After culturing for 1, 3 and 7 days, the samples were rinsed with PBS, stained with the LIVE/DEAD® viability/cytocoxicity kit for mammalian cells (Invitrogen) for 1 h at 37°C, rinsed with PBS again and observed by fluorescence microscopy (BM-20AYC, BM).

The photocatalytic activity of the samples was evaluated using a 300 W Xenon lamp. The preparation parameters of the Ni-Ti-O NTAs are the same as those in the electrochemical capacitance experiments. The TiO2 NTAs with the same morphology as Ni-Ti-O NTAs were used as the control. Each sample with an apparent area of 2 cm2 was incubated in 2 ml of a solution of methylene blue (1.2 × 10−5 M) in a glass cuvette for 1 h before irradiation to establish the adsorption-desorption equilibrium on the NTAs. The decomposition rate of the dye was evaluated by measuring the absorbance of the solutions periodically every 10 min on a UV/Vis spectrophotometer (UV-1100, Mapada).

Results and Disscusion

The effect of the anodization voltage on the structure of Ni-Ti-O NTAs is shown in Figure 1. The samples are fabricated by anodization in an ethylene glycol solution containing 0.2 wt% NH4F and 1.0 vol% H2O at 30°C. At a low voltage of 5 V, nanotubes (NTs) with a small diameter of about 15 nm are observed. Increasing the voltage from 10 to 25 V results in a linear increase in the NT diameter to 70 nm at 25 V. The NT diameter remains the same in the voltage range of 25-40 V and further voltage increase leads to a slight decrease in the NT diameter to 65 and 60 nm at 60 and 90 V, respectively. The variation in the NT length versus anodization voltage shows a similar trend as that of the diameter, with the largest length of about 1100 nm generated at 25 V. When the voltage exceeds 25 V, evenly distributed micropits occur (Figure S1f-i). The entire surface of micropits is covered by NTAs, but the diameter and length are smaller than those outside the micropits (Figure S2). The elemental composition inside and outside the micropits shows no obvious difference as evidenced by Energy dispersive spectroscopy (EDS) elemental mapping (Figure S3). The variation of current during anodization is similar to that of pure Ti, but its steady-state value is higher. When the voltage exceeds 25 V, current fluctuation in the steady-state phase during the anodization process can be observed (Figure S4a). It is generally accepted that a larger voltage leads to increased NT diameter and length in anodization of pure Ti29, but in case of NiTi alloy, this rule no longer applies when the voltage exceeds 25 V. The deviation may be explained by runaway anodization at an elevated voltage, namely current self-amplification resulting from the resistance heating effect. The resistance heating effect elevates the temperature, which on the one hand accelerates oxide dissolution reducing the NT diameter and length, but on the other hand promotes the formation of electrochemical active sites on the sample surface causing local thinning, breakdown of the oxide film and formation of micropits30. The micropits act as short-circuit channels to increase the current further heating the sample. After initiation, the micropits will proliferate in both depth and width (Figure S5). At the same time, hydrolysis of metal cations such as Ti4+ and Ni2+ enriched in the micropits reduces the local pH thus accelerating oxide dissolution and decreasing the NT length and diameter in the micropits.

Figure 1
figure 1

Surface FE-SEM images of the Ni-Ti-O NTAs formed in ethylene glycol containing 0.2 wt% NH4F and 1.0 vol% H2O at 30°C at different anodization voltages.

(a) 5 V, (b) 10 V, (c) 15 V, (d) 20 V, (e) 25 V, (f) 30 V, (g) 40 V, (h) 60 V and (i) 90 V. The anodization durations at different voltages are 12, 6, 6, 4, 1, 0.5, 0.5, 0.25 and 0.25 h, respectively, to reach their balanced (maximum) nanotube length. Variation in the diameter and length of the NTs as a function of anodization voltage (j).

Full size image

The influence of the electrolyte temperature on the structure of NTAs is displayed in Figure 2. The samples are anodized at 25 V in ethylene glycol with 0.2 wt% NH4F and 0.5 vol% H2O for 1 h. As the electrolyte temperature goes up, the NT diameter increases initially and then decreases when the temperature is over 30°C, but the variation is not too large (38–50 nm). The NT length decreases lineally from 1,320 to 386 nm when the electrolyte temperature is raised from 10 to 50°C. A higher temperature reduces the electrolyte viscosity that can induce two opposite effects during electrochemical anodization. Firstly, it increases the mobility of H2O and F− ions in the electrolyte to accelerate NT growth, as evidenced by the increasing current with electrolyte temperature illustrated in Figure S4b and secondly, it accelerates the chemical etching rate to shorten the NTs. During anodization of pure Ti, a linear increase in the tube length is observed as the anodization temperature is raised29, providing evidence of the domination of the first effect. However, anodization of NiTi alloy here exhibits a completely opposite tendency, implying that chemical dissolution of nickel oxide may be more sensitive to the temperature rise and the second effect is dominant in the NiTi alloy anodization process. It can be concluded that in order to fabricate long Ni-Ti-O NTAs, a low electrolyte temperature is required. Another phenomenon is that the Ni-Ti-O NTAs formed at a low electrolyte temperature form bundles separated by cracks (Figure 2 and Figure S6). With increasing NT length, their mechanical strength may not withstand the capillary force thus resulting in clustering3. In addition, irregular macroscopic corrosion grooves are observed after anodization at a relatively high temperature (Figure S7) and this is consistent with the current fluctuation in the steady stage on the current-time curves (Figure S4b), which may be related to runaway anodization.

Figure 2
figure 2

Surface and cross-sectional FE-SEM images of Ni-Ti-O NTAs formed in ethylene glycol with 0.2 wt% NH4F and 0.5 vol% H2O at 25 V at different anodization temperature.

(a) 10°C, (b) 20°C, (c) 30°C, (d) 40°C and (e) 50°C for 1 h. Variation in the diameter and length of the NTs as a function of anodization temperature (f).

Full size image

The effects of the anodization time on the NTA structure are investigated. Anodization is conducted at 30°C and 25 V in an ethylene glycol solution containing 0.2 wt% NH4F and 1.0 vol% H2O. As shown in Figure 3, the NT length increases rapidly initially and slows down gradually afterwards. After anodization for 60 min, the NT length is nearly steady at about 1,100 nm. It is well known that the final NT length is determined by the anodization duration as well as growth and dissolution rates of the oxide. Our results indicate that after anodization for 60 min, a balance is established between the growth rate on the bottom of NTs and dissolution rate on the top. The anodization time influences the NT diameter in a way similarly to the NT length. With increasing anodization duration and NT length, the transport of F− to the electrolyte/oxide interface becomes more difficult. Hence, etching of the NT by F− slows and consequently, the bottom oxide thickness and NT diameter increase3. After anodization of 60 min, the tube length does not increase further because the availability of F− at the electrolyte/oxide interface and etching rate of F− ion on the bottom of the NTs become constant thereby leading to a steady state with respect to the NT diameter. Another phenomenon that should be noted is the irregular surface layer. Generally, in the initial stage of anodization, an irregular surface layer forms and it is gradually removed as anodization proceeds. Our results indicate that an anodization duration of 60 min is required to remove this layer in order to expose the underlying nanotubular structure.

Figure 3
figure 3

Surface and cross-sectional FE-SEM images of the Ni-Ti-O NTAs formed in ethylene glycol with 0.2 wt% NH4F and 1.0 vol% H2O at 25 V and 30°C for different anodization time.

(a) 5 min, (b) 15 min, (c) 30 min, (d) 60 min and (e) 120 min. Variation in the diameter and length of the NTs as a function of anodization time (f).

Full size image

The composition of the anodization electrolyte is another important parameter determining the NTA structure. The effects of the H2O content in the electrolyte are shown in Figure 4. The samples are anodized at 30°C and 25 V in ethylene glycol with 0.2 wt% NH4F but different H2O concentrations for 1 h. NTAs are observed for H2O contents between 0.0 and 1.0 vol% in the electrolyte. Excess H2O leads to an irregular porous structure (Figures 4e and f) because of fast oxide dissolution. When the H2O content is 0 vol%, a thick irregular surface layer can be observed on the NTAs. As the H2O content is increased, this irregular surface layer dissolves gradually and finally disappears at a cocentration of 0.5 vol%. In the range of 0.0–1.0 vol%, the NT diameter and length increase linearly with the H2O content. Since the F− ion concentration in the electrolyte is constant, the etching rate of the NT bottom is the same. As a result of the electric field, H2O in the electrolyte is deprotonated to generate H+, O2−, and/or OH− and anions migrate to the oxide/metal interface forming the oxide. As the H2O content goes up, the availability of anions at the oxide/metal interface increases and consequently the oxide growth rate as well. The NT bottom thickness is proportional to the NT diameter3 and so a linear relationship between the electrolyte H2O content and NT diameter is observed. By the same token, the constant chemical etching rate of F− of the top of NTs and fast growth of NTs at a larger H2O concentration increase the NT length. Micropit formation and current fluctuations are observed when the H2O content is less than 0.5 vol% (Figure S8 and S4c), but the morphology of the micropits is different from that shown in Figures S1 and S7 indicative of different formation mechanisms.

Figure 4
figure 4

Surface and cross-sectional FE-SEM images of the Ni-Ti-O NTAs formed at 25 V and 30°C for 1 h in ethylene glycol with 0.2 wt% NH4F and different H2O contents.

(a) 0.0 vol%, (b) 0.25 vol%, (c) 0.5 vol%, (d) 1.0 vol%, (e) 1.5 vol% and (f) 2.0 vol%. Variation in the diameter and length of the NTs as a function of H2O content (g).

Full size image

The presence of F− in the electrolyte is essential to the formation of NTAs3. The major effect is to etch the bottom of the NTs at a proper rate to maintain a relatively constant bottom oxide thickness. The F− concentration has a significant influence on the NT diameter and length as illustrated in Figure 5. The samples are anodized at 30°C and 25 V in ethylene glycol with 1.0 vol% H2O and different NH4F contents for 1 h. As the NH4F concentration is increased from 0.025 wt% to 1.0 wt%, the NT diameter increases from 33 nm to a maximum of 70 nm for 0.2 wt% NH4F and then diminishes gradually to 39 nm for 0.8 wt% NH4F. The variation in the NT length exhibits the same tendency as the NT diameter. Excess F− leads to rapid dissolution of the oxide thus hampering the ability to generate the regular nanotubular structure (Figure 5g). The variations in the Ni-Ti-O NT diameter and length is the same as those of the TiO2 NTs29. One possible explanation is that a small F− concentration reduces the growth rate of the NTs and prolongs the attack by H+ produced by H2O deprotonation of the NT bottom. In comparison, a high F− concentration expedites the anodization process thus reducing the impact of H+ etching and increasing the NT diameter. However, too much F− in the electrolyte accelerates electrochemical etching of the NT bottom and chemical dissolution of the NT top thus decreasing the NT diameter and length. It is noted that corrosion micropits can occasionally be observed from the samples when the NH4F concentration is small (Figure S9).

Figure 5
figure 5

Surface and cross-sectional FE-SEM images of the Ni-Ti-O NTAs formed at 25 V and 30°C for 1 h in ethylene glycol solution with 1.0 vol% H2O and different NH4F concentrations.

(a) 0.025 wt%, (b) 0.05 wt%, (c) 0.1 wt%, (d) 0.2 wt%, (e) 0.4 wt%, (f) 0.8 wt% and (g) 1.0 wt%. Variation in the diameter and length of the NTs as a function of NH4F concentration (h).

Full size image

By modulating the various anodization parameters discussed above, precise structural control of the Ni-Ti-O TNAs can be accomplished. In order to better understand the microstructure of the NTAs, HR-TEM inspection is conducted (Figure S10). The low magnification TEM image (Figure S10(a)) shows the length and diameter of the NTs, being consistent with that observed by FE-SEM. The HR-TEM image and SAED pattern (Figure S10(b)) display the amorphous structure of the NTs. The EDS line scanning of Ni, Ti and O along the longitudinal direction of the NTs (Figure S10(c)) displays that the ratio Ni to Ti is lower at the top area of the NTAs compared to that at the bottom area, indicating that the Ni oxide dissolves preferentially compared to the Ti oxide during anodization.

The potential applications of NTAs to energy, biomedical and environment fields are subsequently explored. One potential application of the Ni-Ti-O NTAs is the electrode in electrochemical energy storage and the preparation parameters and morphology of the NTAs are presented in Figure S11. Figure 6 displays the electrochemical characterization results. A pair of distinct anodic and cathodic peaks with a symmetrical structure is observed from the cyclic voltammetry (CV) curves (Figure 6a) and the intensity increases with the scanning rate, implying good reversibility in the faradic oxidation and reduction reactions at the interface between the NTAs and electrolyte. The XPS results of the Ni-Ti-O NTAs in Figure S12 have shown that the Ni2p1/2 and Ni2p3/2 is centered at 856.0 and 873.6 eV, respectively, with a spin-energy separation of 17.6 eV, which well matches the characteristic of Ni(OH)2 as previously reported31. Based on our XPS results and previous report32, the two peaks can be attributed to the conversion between the different oxidation states of Ni according to the follow reaction:

Figure 6
figure 6

Electrochemical characterization of the Ni-Ti-O NTAs.

(a) CV curves acquired at various scanning rates. (b) Capacitance obtained at various scanning rates. (c) GCD voltage profiles obtained at a current density of 0.4 mA/cm2. (d) Cycling performance at a current density of 0.4 mA/cm2. (e) Nyquist plot of the Ni-Ti-O NTAs.

Full size image

The variation in the capacitance versus scanning rate derived from the CV curves in Figure 6a is displayed in Figure 6b which shows relatively rapid charging/discharging kinetics. As shown in the galvanostatic charge-discharge (GCD) voltage profiles in Figure 6c, a Columbic efficiency of nearly 100% can be observed from each cycle at a current density of 0.4 mA/cm2. The good cycling stability of the Ni-Ti-O NTAs is demonstrated in Figure 6d. The capacitance increases by about 9% during the first 100 cycles and it may be attributed to the activation process of the electrodes. Remarkably, the capacitance only decreases by only 3.2% after the 1000-cycle test. Nearly twice the capacitance is observed from the Ni-Ti-O NTAs at a current density of 0.4 mA/cm2 compared to previous work19. It is well known that high electrical conductivity of the electrode will contribute to the capacitance properties. The conducting behavior of the Ni-Ti-O NTAs is investigated by EIS and the Nyquist plot is shown in Figure 6e where the high frequency region of the plot is shown as an inset. Conventional electrode materials usually show a semicircular arc, corresponding to the charge transfer between the electrode and the electrolyte. However, no semicirlce is observed for the Ni-Ti-O NTAs, which indicates the absence of electrical resistance and the potential availability as the supercapacitor electrode. Salari and co-authors ascribe this phenomenon to the greatly improved electron transfer pathways in the highly ordered NTAs compared to disordered structure33. The aforementioned results suggest that the Ni-Ti-O NTAs are potential electrode materials in electrochemical energy storage due to fast charging/discharging kinetics and good cycling stability.

In the biomedical fields, an important application of the Ni-Ti-O NTAs is concerned with biofunctional coatings on NiTi-based biomedical implants. The NTAs have been shown to induce selective differentiation of mesenchymal stem cells (MSCs) into osteoblasts9, enhance osteoblast and endothelial functions25,34 and reduce adhesion of macrophages35. Hence, the biofunctionality of NiTi alloy is expected to be enhanced by forming the NTAs coating on the surface. The preparation parameters and morphology of the NTAs used in biological experiments are presented in Figure S13. As shown by the live/dead staining results in Figure 7, the NTAs support cell growth similar to the control. In addition, the tubular structure of the NTAs serves as a good carrier for delivery of drugs and species such as growth factors36, antibiotics37 as well as inorganic biofunctional elements8,13,38 to cater to different clinical requirements. Another potential application for the Ni-Ti-O NTAs in the biomedical fields is as the electrode of non-enzymatic glucose sensor. Ni has been recognized to be able to direct the electrooxidation of glucose to glucolactone through the redox reaction of equation (1) and the following equation39:

Figure 7
figure 7

Fluorescence images of Live/dead staining of MC3T3-E1 cells after culturing for 1, 3 and 7 days on the the Ni-Ti-O NTAs and NiTi alloy control.

Full size image

This electrochemical process is a kinetically controlled surface reaction, so the NTAs with large specific area can provide more active sites. As shown in Figure 8, the amperometric response of the Ni-Ti-O NTAs is ten times of that of the mirror polished NiTi alloy, indicating their significantly enhanced electrooxidation of glucose.

Figure 8
figure 8

CVs of the NiTi alloy (a) and Ni-Ti-O NTs (b) in 0.1 M NaOH with different glucose concentrations at a scanning rate of 50 mV/s.

All samples were activated at 50 mV/s for 100 cycles.

Full size image

TiO2 is an n-type semiconductor and NiO is a p-type semiconductor. It has been reported that the p-n junction can enhance the photocatalytic activity of TiO2 by retarding electron-hole recombination40. The Ni-Ti-O NTAs are promising materials in the energy field, for example, evolution of hydrogen21 and conversion of solar energy41. Nonetheless, our data show that the Ni-Ti-O NTAs have poor photocatalytic activity (Figure S14). One possible explanation is that the Ni content in the Ni-Ti-O NTAs is too high thus blocking the active site in TiO2. Systemic research concerning the photocatalytic activity of the NTAs is in progress and additional findings will be reported in due course.

Different applications may require Ni-Ti-O NTAs with specific diameter and length. For example, in order to promote adhesion and proliferation and inhibit differentiation of MSCs, NTs with a small diameter is required, while large NTs may induce selective differentiation of the MSCs into osteoblast cells42. NiO is inexpensive and desirable in gas sensors, resistance switches and electrochromic devices. In most of the applications, the specific surface area needs to be maximized to improve the efficiency. Our results demonstrate that by varying the processing parameters, the diameter, length and composition (Figure S12) of the Ni-Ti-O NTAs can be tailored to meet the various requirements to widen the use of the materials in the different fields.

Conclusions

In summary, the diameter and length of the Ni-Ti-O NTAs can be tailored in a controlled manner despite the large fraction of Ni in NiTi alloys. To form well-defined NTs, the anodization voltage, temperature and NH4F concentration in the electrolyte can be varied in a relatively wide range but the optimal concentration range of H2O in the electrolyte is narrower. By adjusting these conditions, the Ni-Ti-O NTAs with different diameters (15–70 nm) and lengths (45–1320 nm) can been fabricated. The results also provide insights into the anodization of other Ti alloys containing a large fraction of non-valve metals. The Ni-Ti-O NTAs show promising potential in the energy and biomedical fields.