Application of Ni-Oxide@TiO2 Core-Shell Structures to Photocatalytic Mixed Dye Degradation, CO Oxidation, and Supercapacitors

Performing diverse application tests on synthesized metal oxides is critical for identifying suitable application areas based on the material performances. In the present study, Ni-oxide@TiO2 core-shell materials were synthesized and applied to photocatalytic mixed dye (methyl orange + rhodamine + methylene blue) degradation under ultraviolet (UV) and visible lights, CO oxidation, and supercapacitors. Their physicochemical properties were examined by field-emission scanning electron microscopy, X-ray diffraction analysis, Fourier-transform infrared spectroscopy, and UV-visible absorption spectroscopy. It was shown that their performances were highly dependent on the morphology, thermal treatment procedure, and TiO2 overlayer coating.


Results and Discussion
Scanning electron microscopy (SEM) analysis was conducted to examine the sample morphology ( Figure 1). The morphologies of the as-synthesized A (NiO x ) and B (NiO y ) samples resembled hexagonal plates and short rods, respectively, while their color appeared bluish-green, which is characteristic of Ni(II) complexes [32,33]. Hexagonal plate morphology was also reported for a Ni-rich Ni-Co complex with green color [33]. When the A and B samples were annealed at 500 • C-denoted as A1 (NiO x -500 • C) and B1 (NiO y -500 • C)-the color changed to grey, indicating a modification in crystal structure. In addition, the morphologies underwent some changes, exhibiting particles of various sizes and worm-like structures of 1-2 µm, respectively. In the case of the A2 (NiO x @TiO 2 -500 • C) and B2 (NiO y @TiO 2 -500 • C) samples, obtained by coating with TiO 2 and subsequently annealing (500 • C) the A and B samples, the colors changed to dark yellow. The particle sizes (100-300 nm) of sample A2 were bigger than those of sample A1. Sample B2 showed an aggregated sphere-like morphology, with sizes of 100-400 nm. The samples obtained upon TiO 2 coating and annealing of the A1 and B1 samples are denoted as A3 (NiO x 500 • C@TiO 2 -500 • C) and B3 (NiO y 500 • C@TiO 2 -500 • C). The A3 and B3 samples had a similar yellowish-grey color as well as a sphere-like morphology with some small particles. Figure 2 shows the X-ray diffraction (XRD) patterns of the materials in Figure 1. The XRD pattern of sample A matched that of hexagonal Ni(OH) 2 (reference code: 98-016-1897) [34][35][36][37]. The peaks at 2θ = 19.2 • , 33.0 • , and 38.5 • can be assigned to the (001), (010), and (011) crystal planes of the hexagonal crystal phase, respectively. The XRD pattern of sample B was close to that of monoclinic Ni oxalate dehydrate (reference code: 98-015-0590) with a nearly overlapped strong peak at 2θ = 18.8 • , assignable to the (200)/(202) crystal planes. The A1 (NiO x -500 • C) and B1 (NiO y -500 • C) samples had identical XRD patterns matching that of cubic NiO (reference code: 98-064-6098). The three major peaks at 2θ = 37.3 • , 43.0 • , and 62.9 • can be assigned to the (111), (002), and (022) planes of the cubic crystal phase, respectively. The different Ni complexes of Ni hydroxide and oxalate assumed the same NiO structure upon thermal annealing at 500 • C. Cui et al. synthesized Ni(OH) 2 nanosheets and flower-like microspheres by hydrothermal method, using ammonia and polyvinylpyrrolidone at 150 • C for 15 h, and obtained NiO nanosheets and NiO microspheres by thermally annealing the corresponding samples at 500 • C [34]. For the other four TiO 2 -coated samples (A2, A3, B2, and B3), XRD patterns corresponding to that of tetragonal anatase TiO 2 (reference code: 98-020-2243) were observed. The strongest peak at 2θ = 25.3 • for TiO 2 corresponds to the (011) plane of the tetragonal phase. For samples A2 (NiO x @TiO 2 -500 • C) and B2 (NiO y @TiO 2 -500 • C), additional XRD peaks were observed and attributed to hexagonal Ni(II) titanate NiTiO 3 (reference code: 98-006-6198). Kim et al. also reported the formation of the NiTiO 3 phase for thermally annealed NiO-TiO 2 NTAs at 600 • C [8]. This indicates that the NiTiO 3 phase is commonly formed by heating NiO and TiO 2 via NiO + TiO 2 + heat → NiTiO 3 . Based on the above results, the A2 and B2 samples were NiO@TiO 2 core-shell structures with interfacial NiTiO 3 , whereas the A3 and B3 samples were NiO@TiO 2 core-shell structures with no XRD detectable interface species. A and B after annealing at 500 °C (A1 and B1), TiO2-coated A and B followed by annealing at 500 °C (A2 and B2), and TiO2-coated A1 and B1 followed by annealing at 500 °C (A3 and B3). The starting materials A and B show different morphologies. Figure 2 shows the X-ray diffraction (XRD) patterns of the materials in Figure 1. The XRD pattern of sample A matched that of hexagonal Ni(OH)2 (reference code: 98-016-1897) [34][35][36][37]. The peaks at 2 = 19.2°, 33.0°, and 38.5 can be assigned to the (001), (010), and (011) crystal planes of the hexagonal crystal phase, respectively. The XRD pattern of sample B was close to that of monoclinic Ni oxalate dehydrate (reference code: 98-015-0590) with a nearly overlapped strong peak at 2 = 18.8°, assignable to the (200)/(202) crystal planes. The A1 (NiOx-500 °C) and B1 (NiOy-500 °C) samples had identical XRD patterns matching that of cubic NiO (reference code: 98-064-6098). The three major peaks at 2 = 37.3°, 43.0°, and 62.9 can be assigned to the (111), (002), and (022) planes of the cubic crystal phase, respectively. The different Ni complexes of Ni hydroxide and oxalate assumed the same NiO structure upon thermal annealing at 500 °C. Cui et al. synthesized Ni(OH)2 nanosheets and flower-like microspheres by hydrothermal method, using ammonia and polyvinylpyrrolidone at 150 °C for 15 h, and obtained NiO nanosheets and NiO microspheres by  (A and B), A and B after annealing at 500 • C (A1 and B1), TiO 2 -coated A and B followed by annealing at 500 • C (A2 and B2), and TiO 2 -coated A1 and B1 followed by annealing at 500 • C (A3 and B3). The starting materials A and B show different morphologies. Figure 3 shows the Fourier-transform infrared (FT-IR) spectra for the eight samples in Figure 1. The samples with identical XRD patterns showed similar FT-IR spectra-namely, the A1, A2, and A3 spectra-were similar to the B1, B2, and B3 spectra, respectively. The formation of the Ni oxalate complex of the B sample was more advanced than that of the A sample. Ni-OH stretching vibrations were observed at 3630 and 3380 cm −1 for the A and B samples, respectively [33,38]. For samples A1 and B1, the Ni-O vibration peak was observed at~560 cm −1 [33,38]. For the TiO 2 -coated samples, broad peaks were commonly observed at~650 cm −1 , and attributed to the Ti-O-Ti vibration of the TiO 2 lattice [39]. whereas the A3 and B3 samples were NiO@TiO2 core-shell structures with no XRD detectable interface species.  Figure 3 shows the Fourier-transform infrared (FT-IR) spectra for the eight samples in Figure 1. The samples with identical XRD patterns showed similar FT-IR spectra-namely, the A1, A2, and A3 spectra-were similar to the B1, B2, and B3 spectra, respectively. The formation of the Ni oxalate complex of the B sample was more advanced than that of the A sample. Ni-OH stretching vibrations were observed at 3630 and 3380 cm −1 for the A and B samples, respectively [33,38]. For samples A1 and B1, the Ni-O vibration peak was observed at ~560 cm −1 [33,38]. For the TiO2-coated samples, broad peaks were commonly observed at ~650 cm −1 , and attributed to the Ti-O-Ti vibration of the TiO2 lattice [39].  (A and B), A and B after annealing at 500 °C (A1 and B1), TiO2-coated A and B followed by annealing at 500 °C (A2 and B2), and TiO2-coated A1 and B1 followed by annealing at 500 °C (A3 and B3).
Ultraviolet-visible (UV-Vis) absorption spectra ( Figure 4) were obtained to further characterize the samples showing different sample colors. For the as-synthesized samples A (NiOx) and B (NiOy), two strong absorption regions were observed at ~400 and ~700 nm, attributed to the d-orbital bonding transitions commonly observed for Ni(II) complexes [40]. For the other samples with NiO crystal phase (A1, A2, A3, B1, B2, and B3), broad absorption peaks were observed at ~700-800 nm, which could be attributed to the various transitions (e.g., from the ground state of 3 A2g to the excited states of 3 T2g and 3 T1g) of Ni(II) with octahedral coordination [34]. For the NiO@TiO2 core-shell structures (A2, A3, B2, and B3), band gap absorption edges were observed near 500 nm. whereas the A3 and B3 samples were NiO@TiO2 core-shell structures with no XRD detectable interface species.  (A and B), A and B after annealing at 500 °C (A1 and B1), TiO2-coated A and B followed by annealing at 500 °C (A2 and B2), and TiO2-coated A1 and B1 followed by annealing at 500 °C (A3 and B3) samples. Reference patterns are also shown at the bottom. Figure 3 shows the Fourier-transform infrared (FT-IR) spectra for the eight samples in Figure 1. The samples with identical XRD patterns showed similar FT-IR spectra-namely, the A1, A2, and A3 spectra-were similar to the B1, B2, and B3 spectra, respectively. The formation of the Ni oxalate complex of the B sample was more advanced than that of the A sample. Ni-OH stretching vibrations were observed at 3630 and 3380 cm −1 for the A and B samples, respectively [33,38]. For samples A1 and B1, the Ni-O vibration peak was observed at ~560 cm −1 [33,38]. For the TiO2-coated samples, broad peaks were commonly observed at ~650 cm −1 , and attributed to the Ti-O-Ti vibration of the TiO2 lattice [39]. Ultraviolet-visible (UV-Vis) absorption spectra ( Figure 4) were obtained to further characterize the samples showing different sample colors. For the as-synthesized samples A (NiOx) and B (NiOy), two strong absorption regions were observed at ~400 and ~700 nm, attributed to the d-orbital bonding transitions commonly observed for Ni(II) complexes [40]. For the other samples with NiO crystal phase (A1, A2, A3, B1, B2, and B3), broad absorption peaks were observed at ~700-800 nm, which could be attributed to the various transitions (e.g., from the ground state of 3 A2g to the excited states of 3 T2g and 3 T1g) of Ni(II) with octahedral coordination [34]. For the NiO@TiO2 core-shell structures (A2, A3, B2, and B3), band gap absorption edges were observed near 500 nm. Ultraviolet-visible (UV-Vis) absorption spectra ( Figure 4) were obtained to further characterize the samples showing different sample colors. For the as-synthesized samples A (NiO x ) and B (NiO y ), two strong absorption regions were observed at~400 and~700 nm, attributed to the d-orbital bonding transitions commonly observed for Ni(II) complexes [40]. For the other samples with NiO crystal phase (A1, A2, A3, B1, B2, and B3), broad absorption peaks were observed at~700-800 nm, which could be attributed to the various transitions (e.g., from the ground state of 3 A 2g to the excited states of 3 T 2g and 3 T 1g ) of Ni(II) with octahedral coordination [34]. For the NiO@TiO 2 core-shell structures (A2, A3, B2, and B3), band gap absorption edges were observed near 500 nm.
As TiO 2 is widely used in photocatalysis, the NiO@TiO 2 core-shells (A2, A3, B2, and B3) were tested for photocatalytic dye degradation [41][42][43][44][45][46][47][48][49][50][51]. To increase the novelty of this work, as pure dyes have been extensively investigated, we selected rarely studied mixed dyes, which are also more practical [41][42][43][44][45]. Figure 5 shows the UV-Vis absorption spectra of mixed dye solutions containing dispersions of A2 and B2 samples, and A3 and B3 samples, at increasing visible-light (>400 nm) exposure times. The mixed dye solution was a mixture of MO (5 ppm = 5 mg/L), RhB (5 ppm), and MB (5 ppm); its UV-Vis spectrum showed three absorption peaks at 450, 550, and 650 nm, corresponding to the absorption centers of MO, RhB, and MB, respectively [41][42][43][44][45]. The three peak positions were selected to analyze the degradation rate of each dye. Figure 6 displays the analyzed data for the corresponding UV-Vis absorption spectra in Figure 5. The UV-absorption spectra ( Figure 5) were taken upon achieving adsorption-desorption equilibrium for 1 h under dark conditions. MO was negligibly adsorbed on the A2, A3, and B3 samples, while 15% of MO was adsorbed on the B2 sample ( Figure 6). MB showed relatively good adsorption. Adsorption performance has commonly been explained by electrostatic interactions between the dyes and the catalysts surface [49,50]. Because MB and RhB molecules are more positively charged than MO, the negatively surface-charged catalyst will adsorb MB and RhB more strongly than MO. For the increased adsorption of MO for B2 sample, the combination of NiO/NiTiO 3 /TiO 2 could govern the surface interactions between MO and the B2 sample surface. The order of dye adsorption was found to be MO < RhB < MB for the A2, A3, and B3 samples, and RhB < MO < MB for the B2 sample. The B2 sample showed the best adsorption performance. More interesting results were observed by increasing the visible-light exposure time. Among the three dyes, MO was the fastest to degrade, showing a similar behavior on all the samples. The degradation rate typically showed the following order: RhB < MB << MO. In addition, MB exhibited similar degradation in all the samples; conversely, RhB showed drastic differences between the different samples. The photocatalytic activity for RhB showed an order of B3 << A3 < A2 < B2. The B3 sample showed negligible photocatalytic activity for RhB, while the B2 sample showed a good catalytic activity, comparable to that of MB. The A2 sample showed better activity than the A3 sample for photocatalytic RhB degradation. It appeared that the interfacial NiTiO 3 phase formation (for A2 and B2) facilitated the RhB degradation. For all three dyes, the B2 sample showed the best catalytic performance for mixed dye degradation. Some studies on catalysts for the photocatalytic degradation of mixed dyes are summarized in Table 1 [41][42][43][44]52].   A and B), A and B after annealing at 500 °C (A1 and B1), TiO2-coated A and B followed by annealing at 500 °C (A2 and B2), and TiO2-coated A1 and B1 followed by annealing at 500 °C (A3 and B3).
As the dyes and samples absorb visible light, photogenerated electron hole pairs can be formed by direct visible-light absorption by mixed dye and NiO@TiO 2 . The electron (e − ) and hole (h + ) pairs are separated by interfacial charge transfer at the interface of the p-n junction [5,6,18,46]   . Adsorption (C/Co) and photodegradation (C/Cad) performance profiles of photocatalytic mixed dye solutions containing as-synthesized NiO followed by TiO2 coating and annealing at 500 °C (A2 and B2), and pre-annealed (500 °C) NiO followed by TiO2 coating and annealing at 500 °C (A3 and B3). The corresponding ultraviolet-visible (UV-Vis) absorption spectra are shown in Figure 5. C0 and Cad indicate the UV-Vis absorption intensities of the mixed solution before and after adsorption, respectively.  Photocatalytic mixed dye degradation under UV light (365 nm) was also tested (Figure 7). Unlike under visible-light conditions, the photocatalytic activity showed different behavior under UV-light conditions. MO did not show the fastest degradation rate. The A2 sample showed better photocatalytic activity for RhB than the other samples. The B3 sample showed better photocatalytic activity for MO and MB than the other samples. The B2 sample showed the poorest photocatalytic activity for the mixed dye solution under UV light ( Figure 7); however, it showed the best photocatalytic activity under visible light (Figures 5 and 6). Based on these results, the photocatalyst selection depends on the light wavelength. Figure 5. Ultraviolet-visible absorption spectra for photocatalytic mixed dye degradation under visible-light irradiation obtained using as-synthesized NiO followed by TiO2 coating and annealing at 500 °C (A2 and B2), and pre-annealed (500 °C) NiO followed by TiO2 coating and annealing at 500 °C (A3 and B3). Figure 6. Adsorption (C/Co) and photodegradation (C/Cad) performance profiles of photocatalytic mixed dye solutions containing as-synthesized NiO followed by TiO2 coating and annealing at 500 °C (A2 and B2), and pre-annealed (500 °C) NiO followed by TiO2 coating and annealing at 500 °C (A3 and B3). The corresponding ultraviolet-visible (UV-Vis) absorption spectra are shown in Figure 5. C0 and Cad indicate the UV-Vis absorption intensities of the mixed solution before and after adsorption, respectively. . Adsorption (C/C o ) and photodegradation (C/C ad ) performance profiles of photocatalytic mixed dye solutions containing as-synthesized NiO followed by TiO 2 coating and annealing at 500 • C (A2 and B2), and pre-annealed (500 • C) NiO followed by TiO 2 coating and annealing at 500 • C (A3 and B3). The corresponding ultraviolet-visible (UV-Vis) absorption spectra are shown in Figure 5. C 0 and C ad indicate the UV-Vis absorption intensities of the mixed solution before and after adsorption, respectively.   (Figure 7). Unlike under visible-light conditions, the photocatalytic activity showed different behavior under UV-light conditions. MO did not show the fastest degradation rate. The A2 sample showed better photocatalytic activity for RhB than the other samples. The B3 sample showed better photocatalytic activity for MO and MB than the other samples. The B2 sample showed the poorest photocatalytic activity for the mixed dye solution under UV light ( Figure 7); however, it showed the best photocatalytic activity under visible light (Figures 5 and 6). Based on these results, the photocatalyst selection depends on the light wavelength. The materials were tested as CO oxidation catalysts. Figure 8 displays the first and second CO oxidation reaction run profiles. CO2 (mass = 44 amu) production by CO oxidation was monitored with the increasing reaction temperature by using mass spectrometry [33,[53][54][55][56]. Qualitative analysis was performed using the mass profile data. For the first runs of the as-synthesized samples-i.e., A (NiOx) and B (NiOy)-the CO2 production onsets were observed at 265 and 300 °C, respectively; while, in the second runs, they were observed at 300 and 250 °C, respectively. In the  (A2 and B2), and pre-annealed (500 • C) NiO followed by TiO 2 coating and annealing at 500 • C (A3 and B3).
The materials were tested as CO oxidation catalysts. Figure 8 displays the first and second CO oxidation reaction run profiles. CO 2 (mass = 44 amu) production by CO oxidation was monitored with the increasing reaction temperature by using mass spectrometry [33,[53][54][55][56]. Qualitative analysis was performed using the mass profile data. For the first runs of the as-synthesized samples-i.e., A (NiO x ) and B (NiO y )-the CO 2 production onsets were observed at 265 and 300 • C, respectively; while, in the second runs, they were observed at 300 and 250 • C, respectively. In the second run, sample A showed degraded activity (by +35 • C), whereas sample B showed enhanced activity (by −50 • C). The Ni hydroxide (sample A) and Ni oxalate (sample B) complexes were found to change to the same NiO crystal phase after the CO oxidation reactions, based on the XRD results. For sample A1, the onsets were observed above 470 • C in the first and second CO oxidation runs. The catalytic activity dramatically degraded for sample A1 with NiO crystal phase. For sample B1, the onsets were observed at 300 and 350 • C in the first and second runs, respectively. Although the B1 sample showed the same NiO crystal phase, its catalytic activity was considerably superior to that of sample A1. Similarly, samples A2 and B2 showed an onset above 450 • C and at 300 • C, respectively. Samples A3 and B3 showed similar onsets near 400 • C for the first and second runs. As a result, the sample preparation method is crucial to determine the catalytic activity. In the present study, the TiO 2 coating on the B2 sample did not to degrade the CO oxidation activity. However, when Ni(OH) 2 plates were used as starting material, the CO oxidation activity significantly decreased. second run, sample A showed degraded activity (by +35 °C), whereas sample B showed enhanced activity (by −50 °C). The Ni hydroxide (sample A) and Ni oxalate (sample B) complexes were found to change to the same NiO crystal phase after the CO oxidation reactions, based on the XRD results. For sample A1, the onsets were observed above 470 °C in the first and second CO oxidation runs. The catalytic activity dramatically degraded for sample A1 with NiO crystal phase. For sample B1, the onsets were observed at 300 and 350 °C in the first and second runs, respectively. Although the B1 sample showed the same NiO crystal phase, its catalytic activity was considerably superior to that of sample A1. Similarly, samples A2 and B2 showed an onset above 450 °C and at 300 °C, respectively. Samples A3 and B3 showed similar onsets near 400 °C for the first and second runs. As a result, the sample preparation method is crucial to determine the catalytic activity. In the present study, the TiO2 coating on the B2 sample did not to degrade the CO oxidation activity. However, when Ni(OH)2 plates were used as starting material, the CO oxidation activity significantly decreased. Figure 8. CO2 production (by CO oxidation) mass profiles for as-prepared NiO (A and B), annealed (500 °C) A and B (A1 and B1), TiO2-coated A and B followed by annealing at 500 °C (A2 and B2), and TiO2-coated A1 and B1 followed by annealing at 500 °C (A3 and B3).
Furthermore, supercapacitor performance tests were briefly performed. The related results of cyclic voltammograms (CVs), galvanostatic charge/discharge curves (CD), and impedance plots are shown in Figure 9. For the selected samples (A1 and B3), cyclic voltammetry curves were obtained at various scan rates. Broad anodic and cathodic peaks were typically observed, associated with the redox reactions of NiO + OH − → NiOOH + e − + H2O [33,57]. The other samples showed similar qualitative behaviors. The symmetrical peaks indicated reversible redox reactions. The peak current increased with the increasing scan rate, and the gap between the anodic and cathodic peak positions became wider, revealing a diffusion-controlled pseudocapacitive behavior of the material [4,22,57]. The galvanostatic CD curves were obtained at the current density of 0.83 A/g. The measured specific capacitance (F/g) was found to be in the following order: A2  A3 < B2 < B1 < B3 < A1. For sample A1 with NiO crystal phase, the specific capacitance was measured to be 320 F/g, while, the specific capacitances of the other two TiO2-coated samples (A2 and A3) were found to be identical, 50 F/g. For sample B1 with NiO crystal phase, the specific capacitance was measured to be 189 F/g, lower than that of the A1 sample (320 F/g). However, the other two TiO2 coated samples, B2 Figure 8. CO 2 production (by CO oxidation) mass profiles for as-prepared NiO (A and B), annealed (500 • C) A and B (A1 and B1), TiO 2 -coated A and B followed by annealing at 500 • C (A2 and B2), and TiO 2 -coated A1 and B1 followed by annealing at 500 • C (A3 and B3).
Furthermore, supercapacitor performance tests were briefly performed. The related results of cyclic voltammograms (CVs), galvanostatic charge/discharge curves (CD), and impedance plots are shown in Figure 9. For the selected samples (A1 and B3), cyclic voltammetry curves were obtained at various scan rates. Broad anodic and cathodic peaks were typically observed, associated with the redox reactions of NiO + OH − → NiOOH + e − + H 2 O [33,57]. The other samples showed similar qualitative behaviors. The symmetrical peaks indicated reversible redox reactions. The peak current increased with the increasing scan rate, and the gap between the anodic and cathodic peak positions became wider, revealing a diffusion-controlled pseudocapacitive behavior of the material [4,22,57]. The galvanostatic CD curves were obtained at the current density of 0.83 A/g. The measured specific Materials 2016, 9, 1024 9 of 15 capacitance (F/g) was found to be in the following order: A2 ≈ A3 < B2 < B1 < B3 < A1. For sample A1 with NiO crystal phase, the specific capacitance was measured to be 320 F/g, while, the specific capacitances of the other two TiO 2 -coated samples (A2 and A3) were found to be identical, 50 F/g. For sample B1 with NiO crystal phase, the specific capacitance was measured to be 189 F/g, lower than that of the A1 sample (320 F/g). However, the other two TiO 2 coated samples, B2 and B3, exhibited values of 211 F/g and 134 F/g, respectively, which were higher than those of the A1 and A2 samples (50 F/g). Kim et al. reported specific capacitances in the range of 40-100 F/g for NiO-TiO 2 NTAs and 120-300 F/g for NiO-TiO 2 nanotube films [8], very close to the values observed in the present study. Some literature values for TiO 2 -NiO hybrid materials are summarized in Table 2, where TiO 2 was mainly core [4,7,8,58,59]. To the best of our knowledge, the specific capacitance for NiO@TiO 2 core-shell structures has not been reported. The high-frequency region of the impedance Nyquist plots with real (Z ) and imaginary (Z ) parts are shown in Figure 9. Ascending straight lines were commonly observed (not shown here) for all the samples in the low-frequency region, corresponding to Warburg diffusion resistance [33,53]. Sample A3 clearly showed a semicircle before and after CD measurements, corresponding to charge transfer resistance. The other samples showed no clear semicircles. The interfacial resistance of the A1 sample was smaller than those of the other two samples (A2 and A3). The resistance of the B2 sample was larger than those of the B1 and B3 samples. Thus, the resistance in the impedance plots was found to be consistent with the order of the measured specific capacitance values. and B3, exhibited values of 211 F/g and 134 F/g, respectively, which were higher than those of the A1 and A2 samples (50 F/g). Kim et al. reported specific capacitances in the range of 40-100 F/g for NiO-TiO2 NTAs and 120-300 F/g for NiO-TiO2 nanotube films [8], very close to the values observed in the present study. Some literature values for TiO2-NiO hybrid materials are summarized in Table  2, where TiO2 was mainly core [4,7,8,58,59]. To the best of our knowledge, the specific capacitance for NiO@TiO2 core-shell structures has not been reported. The high-frequency region of the impedance Nyquist plots with real (Z′) and imaginary (Z′′) parts are shown in Figure 9. Ascending straight lines were commonly observed (not shown here) for all the samples in the low-frequency region, corresponding to Warburg diffusion resistance [33,53]. Sample A3 clearly showed a semicircle before and after CD measurements, corresponding to charge transfer resistance. The other samples showed no clear semicircles. The interfacial resistance of the A1 sample was smaller than those of the other two samples (A2 and A3). The resistance of the B2 sample was larger than those of the B1 and B3 samples. Thus, the resistance in the impedance plots was found to be consistent with the order of the measured specific capacitance values.

Synthesis of Ni Oxide and TiO 2 Coating
Two different Ni oxide complexes-i.e., NiO x (A) and NiO y (B)-were synthesized by hydrothermal method as described below, and then used as starting materials. For material A, 0.475 g of NiCl 2 ·6H 2 O (GR 98%, Duksan Pure Chemical Co., Ansan, Korea) was dissolved in 40 mL of deionized water, and 2 mL of 1.0 M NaOH solution was added to it. The solution was then transferred to a Teflon-lined stainless autoclave and placed in an oven setting at 120 • C for 12 h. For material B (NiO y ), 0.951 g of NiCl 2 ·6H 2 O was dissolved in a mixed solvent (20 mL of H 2 O + 20 mL of ethylene glycol); then, 1.0 mL of 0.1 M NaOH solution and 0.238 g of oxalic acid (≥99%, Sigma-Aldrich, St. Louis, MO, USA) were added to it. Upon complete dissolution, the entire solution was transferred to a Teflon-lined stainless autoclave and placed in an oven setting at 120 • C for 12 h. After the hydrothermal reaction, the obtained precipitates were fully washed with deionized water and ethanol, and then centrifuged, repeatedly. The washed sample powders were dried in an oven at 70 • C for two days. For materials A1 (NiO x -500 • C) and B1 (NiO y -500 • C), the dried sample powders were thermally treated at 500 • C in an electric furnace for 4 h. For materials A2 (NiO x @TiO 2 -500 • C) and B2 (NiO y @TiO 2 -500 • C), the dried A and B sample powders (0.1 g) were dispersed in ethanol solvent, and 0.5 mL of titanium (IV) isopropoxide (TTIP; 97%, Sigma-Aldrich, St. Louis, MO, USA) was added to it. While stirring the solution, water vapor was slowly introduced using a humidifier for 2 h. After the reaction, the samples were fully washed with ethanol solvent to remove uncoated TTIP. Then, the collected sample was dried and thermally annealed at 500 • C for 4 h. For materials A3 (NiO x 500 • C@TiO 2 -500 • C) and B3 (NiO y 500 • C@TiO 2 -500 • C), the A1 and B1 sample powders were used for TiO 2 coating. The other procedures were the same as those used for the A2 and B2 samples.

Characterization of the Materials
The morphology of the prepared samples was examined using field-emission SEM (FE-SEM, Hitachi SE-4800, Tokyo, Japan). The samples were placed on an HF-etched Si substrate. The crystal phases were identified using powder XRD; the patterns were obtained by using a PANalytical X'Pert Pro MPD diffractometer (PANalytical Inc., Westborough, MA, USA) equipped with a Cu Kα radiation source. FT-IR spectra were measured using a Nicolet iS 10 FT-IR spectrometer (Thermo Scientific, West Palm Beach, FL, USA) with an attenuated total reflection mode between 500 and 4000 cm −1 . UV-Vis absorption spectra of the samples were recorded using a Neosys-2000 double beam UV-Vis spectrometer (Scinco, Seoul, Korea).

Photocatalytic Dye Degradation, CO Oxidation, and Supercapacitor Performance Tests
For photocatalytic mixed dye degradation, equal amounts of three different dye (i.e., MO, RhB, and MB) solutions with concentrations of 5 mg/L (=ppm) were mixed to prepare a mixed dye test solution. Then, 20 mg of sample powder was dispersed in 50 mL of the mixed dye solution to achieve adsorption-desorption equilibrium under dark conditions for 1 h. Upon achieving the equilibrium, 2 mL of the solution was taken and centrifuged to remove the residual powder, and a UV-Vis absorption spectrum was recorded by using a V-530 UV-Vis spectrometer (Jasco, Tokyo, Japan). As the equilibrium was reached, the mixed dye solution was examined under UV (or visible) irradiation. For visible light irradiation, a 500 W Halogen lamp (>400 nm) was used and the distance from the lamp and the dye solution in 100 mL beaker (a diameter of 5.5 cm) was fixed at about 30 cm. For UV light irradiation, four 6W UV (365 nm) lamps were used. Every 2 h, 2 mL of the solution was collected to record the UV-Vis absorption spectrum. For the CO oxidation reaction, 10 mg of sample powder was loaded in a flow-type quartz U-tube with an inner diameter of 4 mm. CO oxidation experiments were performed by increasing the temperature up to 500 • C at the rate of 10 • C·min −1 under flow of CO (1%) and O 2 (2.5%) mixed gas in N 2 balance. The reaction gas products, such as CO 2 (mass = 44 amu), were examined using an SRS RGA200 quadrupole mass spectrometer (Stanford Research System, Sunnyvale, CA, USA). After the first CO oxidation run, the sample was naturally cooled to room temperature; then, the second CO oxidation run was performed. The electrochemical measurements were conducted using a CHI 660D electrochemical work station (CH Instruments, Austin, TX, USA) with a conventional three-electrode configuration (Ag/AgCl reference electrode, Pt wire counter electrode, and sample mounted on a Ni-foam working electrode) in 6.0 M KOH electrolyte solution. For the working electrode, the sample powder (60 wt %) was fully mixed with acetylene black (20 wt %) and poly(vinylidene fluoride) (20 wt %) in 2 mL of N-methyl-2-pyrrolidone solvent. The dispersion was slowly dropped onto Ni foam (1 × 1 cm 2 ), dried, and pressed to fabricate a thin Ni foam sheet. The CVs were obtained between −0.2 and 6.0 V, galvanostatic CD experiments were conducted with potentials ranging from 0.1 to 0.4 V at the charge density of 0.83 A·g −1 , and electrochemical impedance measurements were performed over a frequency range from 0.1 MHz to 0.01 Hz.

Conclusions
In this work, we investigated the preparation method, characteristics, and various application performances of NiO@TiO 2 core-shell nanostructures. Generally, TiO 2 coating on Ni(OH) 2 plates and on annealed NiO showed poor photocatalytic, thermocatalytic, and supercapacitor performances. Therefore, it is advisable to use rod-shaped Ni oxalate complex as a starting material for TiO 2 coating. The main findings are as follows: 1.
Nanoplates and nanorods were synthesized by hydrothermal method, and showed XRD patterns of Ni(OH) 2 and Ni oxalate complexes, respectively. NiO crystal phase was commonly obtained by thermal annealing at 500 • C.

2.
NiO@TiO 2 core-shells with interfacial NiTiO 3 could be prepared by TiO 2 coating on as-synthesized Ni samples, followed by thermal annealing at 500 • C. TiO 2 coating on annealed (500 • C) NiO, followed by thermal annealing at 500 • C, showed no XRD detectable NiTiO 3 at the interface.

3.
For photocatalytic dye degradation, TiO 2 coating on Ni oxalate complex followed by thermal annealing (sample B2) showed the best photocatalytic activity for mixed dye degradation under visible light.

4.
For CO oxidation, the B2 sample (NiO y @TiO 2 -500 • C) was also more efficient for CO oxidation than other samples.

5.
For the case of specific capacitance, the B3 sample (NiO x 500 • C@TiO 2 -500 • C) was more efficient for specific capacitance than other samples. Specific capacitances were obtained to be in the range of 50-320 F/g. When Ni(OH) 2 plates were used as starting material, the TiO 2 coating showed a specific capacitance of only 50 F/g. However, when rod-shaped Ni oxalate complex was used, the TiO 2 coating showed specific capacitance values of 211 and 134 F/g.
Briefly, the present study showed that the sample preparation procedures and the annealing steps vary greatly in the application fields to achieve a higher performance. The obtained valuable information on the preparation methods of the investigated core-shell structures could be extended to other transition metal oxide core-shell structures, promoting their development and diverse applications.