Nanoporous TiN/TiO₂/Alumina Membrane for Photoelectrochemical Hydrogen Production from Sewage Water

Abstract

An aluminum oxide, Al₂O₃, template is prepared using a novel Ni imprinting method with high hexagonal pore accuracy and order. The pore diameter after the widening process is about 320 nm. TiO₂ layer is deposited inside the template using atomic layer deposition (ALD) followed by the deposition of 6 nm TiN thin film over the TiO₂ using a direct current (DC) sputtering unit. The prepared nanotubular TiN/TiO₂/Al₂O₃ was fully characterized using different analytical tools such as X-ray diffraction (XRD), Energy-dispersive X-ray (EDX) spectroscopy, scanning electron microscopy (SEM), and optical UV-Vis spectroscopy. Exploring the current-voltage relationships under different light intensities, wavelengths, and temperatures was used to investigate the electrode’s application before and after Au coating for H₂ production from sewage water splitting without the use of any sacrificing agents. All thermodynamic parameters were determined, as well as quantum efficiency (QE) and incident photon to current conversion efficiency (IPCE). The QE was 0.25% and 0.34% at 400 mW·cm⁻² for the photoelectrode before and after Au coating, respectively. Also, the activation energy was 27.22 and 18.84 kJ·mol⁻¹, the enthalpy was 24.26 and 15.77 J·mol⁻¹, and the entropy was 238.1 and 211.5 kJ⁻¹·mol⁻¹ before and after Au coating, respectively. Because of its high stability and low cost, the prepared photoelectrode may be suitable for industrial applications.

1. Introduction

Renewable energy sources are very important for our life, whereas most of these sources are based on photocatalytic materials that are used in the preparation of solar cells, photodetectors, and photocatalytic electrodes [1,2,3]. The production of H₂ gas due to water-splitting reaction is a very important energy source for airplanes and factories that used H₂ fuel [4]. The prepared photoelectrode must have high efficiency for water splitting and properties that qualify it for industrial applications, such as low cost, easy preparation, and easy operation [5,6]. The working of the electrode can be carried out under direct sunlight or using another artificial light. Many semiconductor materials were used for the synthesis of the electrode, including oxides, sulfides, and some polymers [7,8]. The properties of the photocatalytic materials improve with increasing the active sites inside the materials, in which the nanowires or nanotube morphology increase the active sites through increasing the surface area of the materials [9,10]. The other way for increasing the efficiency of the electrode is using plasmonic materials for light capture, in which a surface plasmonic resonance is produced from the oscillation of electrons in the conductive bandgap [11].

The most used plasmonic materials are noble metals, such as Au, Ru, Rh, and Pt, but these metals are very expensive [12]. Rabia et al. used Au to enhance the water-splitting reaction for H₂ production under light [13]. Shi et al. used Pt to enhance the activity of CdS materials for water splitting, in which the photoactivity is enhanced very much by using Pt metal [14]. At the same time, many researchers used active metals such as Co, Fe, Ni, and Cu as plasmonic materials. This appears in Cu/ZnO materials, in which the Cu nanomaterials enhance the light capture and efficiency for the light detection in the prepared photodetector [15]. Although these elements are cheap, they are active and easy to form oxides.

With its high noble features and coating protection, metal nitride ceramic can serve as a photo-plasmonic resonance material for light capture, comparable to noble plasmonic metal. These behaviors originate from the chemical stability and high melting point (>2000 K) [16]. Moreover, these metal nitrides are used in designing devices with spectral windows and operating conditions more efficiently in hazardous environmental conditions than noble metals [17]. Some studies were carried out for comparing the effect of TiN and ZrN plasmonic materials on the sensing properties of a plasmonic refractometry sensor with noble metals, finding that the nitride metals were more efficient than the noble metals [16,18,19,20].

Due to the good properties of the TiN material, such as high hardness, nontoxicity, high thermal conductivity, high corrosion resistance, chemical stability, and high UV-Vis light absorbance, Mohamed et al. used a 9 nm TiN layer to enhance the properties of TiO₂ photoactive material for solar cells [18]. With an increase in TiN thickness to 9 nm, the optical properties, absorption, and bandgap improved, resulting in increased photocurrent. Awad et al. used tri-layered TiO₂/TiN/TiO₂ self-cleaning systems of different TiN thicknesses to optimize their optical properties and facilitate the photodegradation of the organic materials [21]. Also, many researchers have concentrated on the properties of TiN and TiN-based nanostructures for application in many devices, such as fuel cells, supercapacitors, superconductors, and solar cells [22,23,24].

Photocatalytic properties of TiO₂ materials, especially in structures with a large surface area, such as nanowires and nanorods, are of particular interest [25]. TiO₂ nanotubes tolerate more active sites on their internal and external surfaces that increase their activity in light absorption, in addition to their large surface area-to-volume ratios. TiO₂ also has several desirable characteristics, including nontoxicity, low cost, ease of preparation, and high stability [22]. These properties qualify TiO₂ materials for application in sensors, supercapacitor, H₂ production, solar cell, and light absorbance [26]. Since TiO₂ has a wide bandgap > 3 eV, it can only be used in photocatalytic and light detection devices that operate in the UV range [27].

Many studies on H₂ generation by water splitting have recently been released, but almost all of them use a sacrificing agent, such as NaOH, HCl, Na₂S₂O₃, or Na₂SO₃ for water splitting [28,29,30,31]. Also, the majority of these studies do not demonstrate long-term H₂ production stability. Furthermore, the morphology of the microsized materials is uncontrolled [30,31,32]. In this study, a TiN/TiO₂/Al₂O₃ template photoelectrode was created and used to generate H₂ from sewage water splitting without the use of any sacrificing agents. The effect of depositing a thin Au film on the TiN/TiO₂/Al₂O₃ electrode was also investigated. An Al₂O₃ template was prepared with high hexagonal ordering using the Ni imprinting technique. The TiO₂ was deposited using the ALD technique, whereas the thin TiN layer was deposited using the DC sputtering technique. The electrode’s application for H₂ production was studied by exploration the current–voltage relationship under various light intensities, wavelengths, and temperatures. The quantum efficiency (QE), incident photon to current conversion efficiency (IPCE), and all thermodynamic parameters were calculated. The prepared photoelectrode can be applicable in the industrial field with high stability and low cost.

2. Experimental Methods

2.1. Al₂O₃ Template Synthesis

The synthesis of the Al₂O₃ template was carried out with a novel method of high controllability in the shape and size of the pores. The aim is to prepare pores with a hexagonal shape with a pore diameter of about 320 nm. For the preparation of the Al₂O₃ template, Al metal (99.99%) was used, in which an electropolishing process was carried out for removing any impurities or corrosions. To determine the position of the pores, the Ni imprinting method was applied at a pressure of ~10 kN/cm² for 180 s using an oil pump [24]. Then, a two-step anodization process is carried out using a mixture of H₃PO₄, ethylene glycol, and H₂O with a volume ratio of 1:100:200. The first step was carried out for 15 min, whereas the second step was performed for 120 min at 2 °C. The first step was etched using a mixture of H₂CrO₄ and H₃PO₄ with concentrations of 1.5 wt% and 6 wt%, respectively. As a final step, the nanopores were widened at 60 °C for 20 min.

2.2. Synthesis of TiN/TiO₂ and Au/TiN/TiO₂

TiN/TiO₂ nanotubes were prepared inside the Al₂O₃ template through two different steps. The pore of the template was about 320 nm with a length of about 1 µm. First, TiO₂ was synthesized by the atomic layer deposition device (ALD, Picosun SUNALE R150 ALD, Poscosun, Espoo, Finland) at 300 °C for 300 running cycles with a rate of 0.3 nm/s. TiO₂ is prepared from TiCl₄ as Ti source and H₂O as O source. After that, TiN thin film (8 nm) was deposited over TiO₂ using DC sputtering device under initial and working pressures of 6.7 × 10⁻⁷ and 1.3 × 10⁻² mbar, respectively. The sputtering process was carried out from a mixture of N₂ and Ar gases with volumes of 75 and 25 sccm, respectively. The Ti source was a Ti-metal (99.99%). Finally, for synthesis Au/TiN/TiO₂, Au nanoparticles are sputter-coated over TiN/TiO₂/Al₂O₃ template under vacuum conditions for 3 min. Au deposition was carried out utilizing a fairly basic sputter-coater at pressure 2 torr (low vacuum), current 13 mA, and a distance of 8 cm in front of the Au target. The rate of growth of the Au layer was 5 nm/min.

2.3. Samples Characterization

The prepared template and nanotubes were characterized using different devices. The chemical structure was confirmed using an X-ray diffractometer (XRD, Bruker/Siemens D5000, Aubrey, TX, USA) by Kα radiation of Cu (λ = 1.5405 Å) at 30 mA, and 40 kV. Moreover, the elements inside the materials and nanomorphologies were determined using a scanning electron microscope (SEM, Auriga Zeiss FIB, Zeiss company, Oberkochen, Germany) provided with energy-dispersive X-ray (EDX, Zeiss company, Oberkochen, Germany) unit. The optical properties were characterized using a double beam spectrophotometer (Perkin Elmer Lambda 950, Perkin Elmener company, Waltham, MA 02451, USA).

2.4. Water-Splitting Test

The water-splitting test reaction was carried out under a solar simulator from a two-electrode cell, in which the TiN/TiO₂/Al₂O₃ or Au/TiN/TiO₂/Al₂O₃ represented the working electrode that worked as a cathode, whereas a graphite electrode with the same dimension was worked as an anode, Figure S1 (Supplementary Data). The chemical composition of the sewerage water is mentioned in Table S1. The pH value was 5 for this wastewater. The effects of light intensity (100 to 400 mW·cm⁻²), incident wavelength (405 to 636 nm), temperature reaction (30 to 70 °C), and time stability were studied for water splitting for H₂ and O₂ evolutions. The effect of monochromatic light on the water splitting is tested using optical filters of different wavelengths; 405, 470, 490, 508, and 636 nm.

3. Results and Discussion

3.1. SEM and XRD Studies

The morphological analysis of the Al₂O₃ template after pore widening at 60 °C is mentioned in Figure 1a. From this figure, the Al₂O₃ template is prepared with high accuracy and ordering with a hexagonal shape, in which the pore diameter is about 330 nm. Figure 2b displayed the bottom of the membrane after the deposition of TiO₂ by the ALD. The bottoms of the tubes are uniformly coated, whereas the outer diameter of the bottom is about 320 nm. Moreover, the TiN/TiO₂ back surface is shown in Figure 1c, after the deposition of TiN, the diameter is increased to ~330 nm. The cross-section of TiN/TiO₂ is shown in Figure 1d with a length of about 1.0 µm. The bottoms of the tubes become rough with extremely small nanofeatures after TiN deposition. This increases the effective surface area and surface-active sites, allowing TiN to perform its primary function as a plasmonic layer for light absorption. Figure S2 (Supplementary Data) also shows a top-view SEM image of Au/TiN/TiO₂/Al₂O₃ with agglomerated Au nanoparticles covering the top-surface and template pores with an average diameter of 285 nm.

The structures and chemical constructions of the prepared materials are confirmed using XRD analyses as shown in Figure 2. The sharp peaks in the XRD of the Al₂O₃ template, Figure 2a, refer to the formation of a crystalline structure with excellent precision and ordering [32]. The Al₂O₃ has three peaks at 44.5°, 65.0°, and 78.1°, which correspond to the growth directions (113), (214), and (119), respectively. In addition, Figure 2a shows the XRD analysis of the synthesized TiN/TiO₂, which shows eight peaks at 25.7°, 38.0°, 48.2°, 54.2°, 55.2°, 62.7°, 69.7°, and 75.4°. These peaks, coupled with the growth directions (101), (103), (200), (105), (211), (213), (118), and (215), indicated the synthesis of anatase TiO₂. The deconvolution of the broadband peak in the XRD pattern of TiN/TiO₂ exhibits the overlapping of two diffraction peaks: (101) peak of anatase-TiO₂ (JCPDS 21-1272) and the adjacent (111) peak of TiN (JCPDS 38-1420). Otherwise, the deposition of ultra-thin TiN film using DC sputtering on TiO₂ does not create any phase change in the XRD pattern [33]. From the XRD of TiN/TiO₂ and the observed peaks for the anatase TIO₂, it can be concluded that the TiO₂ layer was crystalline with preferred growth orientation alongside the (101) direction.

The EDX analyses of the Al₂O₃ template and TiN/TiO₂ are shown in Figure 2. For Al₂O₃, the O and Al elements of the template are detected. For TiN/TiO₂ nanotube; Ti, O, and N signals are detected. In addition, Figure S3 (Supplementary Data) shows the EDX spectrum for Au/TiN/TiO₂/Al₂O₃, in which all elements (Ti, O, N, Al, and Au) are well detected. The weight% of the Au is approximately 3%.

3.2. Optical Properties of TiN/TiO₂/Al₂O₃

The optical properties of the prepared template and TiN/TiO₂/Al₂O₃ are illustrated in Figure 3. The reflectance spectra of the Al₂O₃ template and TiN/TiO₂/Al₂O₃ are displayed in Figure 3a. The reflectance of TiN/TiO₂/Al₂O₃ shows stronger interference fringes than that of the Al₂O₃ template alone. In the case of TiN/TiO₂/Al₂O₃, constructive and destructive interferences between reflected waves from TiO₂/Al₂O₃, TiN/TiO₂, and air/TiN interfaces are responsible for these strong interference fringes. These interferences are low in the UV and Vis regions and increase in the near IR region. The strength of the interference ripples grows with increasing wavelength; nevertheless, the interference ripple below 600 nm is much stronger. This could be due to the Al₂O₃ blue emission band, which is attributed to mixed emission from F and F⁺ centers [34]. After Au nanoparticles deposition, the reflectance values decrease, this indicates the increasing absorbance values. This confirms the main plasmonic role of Au nanoparticles. The reduction of the reflectance in the UV/Vis region indicating that the TiN/TiO₂ materials have strong absorbance in the UV/Vis regions. To confirm that for the TiN/TiO₂, we have measured the absorbance spectra for TiO₂ and TiN/TiO₂ materials, as declared in Figure 3b. Because the anatase bandgap forms between the Ti3d and O2p states, TiO₂ exhibits a good absorbance [34,35]. After TiN deposition, the absorbance is enhanced in which the right edge of the absorption peak is extended in the visible region till 600 nm. This improvement in the absorption behavior came from the main role of the plasmonic TiN material. The bandgap (Eg) values for TiO₂ and TiN/TiO₂ materials, which are 3.1 and 2.2 eV, respectively, demonstrate this improvement in optical properties (Figure 3c). The Eg value was calculated using the Tauc equation, Equation (1), the optical absorption (A), absorption coefficient ($\mathsf{\alpha }$), material thickness (d), light frequency (v), Planck constant (h), and Boltzmann constant ($\mathsf{\beta }$) [36]:

$$\mathsf{\alpha }\mathrm{h}\mathsf{\nu }=\mathsf{\beta }{\left(\mathrm{h}\mathsf{\nu }-\mathrm{Eg}\right)}^{\frac{1}{2}}$$

(1)

$$\mathsf{\alpha }=\left(\frac{2.303}{\mathrm{d}}\right)\mathrm{A}$$

(2)

3.3. The Photoelectrochemical Performance

The electrochemical performances of the prepared electrodes, TiN/TiO₂/Al₂O₃ and Au/TiN/TiO₂/Al₂O₃, for the H₂ generation are measured under lighting from a metal-halide Lamp (Newport, 66926-500HX-R07). The measurements were carried out using the Keithley measurement source unit (model:2400), whereas the J_ph value represents the H₂ generation rate from the waste H₂O (sewage water). The relation between current density and voltage for the TiN/TiO₂/Al₂O₃ electrode with and without Au coating is revealed in Figure 4A,B shows how the TiN/TiO₂/Al₂O₃ electrode functioned as a photocathode for the generation of H₂. Under light irradiation, the resultant current density is −0.0924 mA·cm⁻² @ −1 V, as shown in Figure 4A. Figure 4B shows that this electrode’s performance for O₂ production is poor, with a current density of just 0.0295 mA·cm⁻² @ 1 V and an onset value of −0.0202 mA·cm⁻² @ 0 V. While the Au-coated electrode in Figure 4C,D is utilized to split water and produce O₂, the opposite electrode produces H₂ (graphite-electrode). With an onset value of 0.0143 mA·cm⁻² @ 0 V, the current density was −0.127 mA·cm⁻² @ −1 V and 0.140 mA·cm⁻² @ 1 V. As a result, the Au/TiN/TiO₂/Al₂O₃ electrode performed better as a photoanode for O₂ generation, as shown in Figure 4C,D). The generated current density has increased significantly, Figure 4D, reaching 0.140 mA·cm⁻². This enhancement in the produced current density is due to the plasmonic resonance of Au nanoparticles and their role in the enhancement of near IR absorbance and hot-electron allocation [36,37,38]. In addition to that, the existence of Au nanoparticles on the electrode surface causes light multiple reflections and mean photon path elongation alongside the catalytic electrode. Additionally, because the Fermi level (EF) of Au NPs is lower than that of TiO₂, light excites electrons at the Fermi level of plasmonic Au NPs and raises them to the localized surface plasmon energy level. This permits surface plasmon-generated hot electrons to be transported to the conduction band (CB) of TiO₂, resulting in enhanced charge carriers and photocurrent. Also, Au nanoparticles facilitate the excitation of the electron to overcome the Schottky barrier and reach the conducting band of TiO₂/TiN. Also, plasmonic Au NPs were able to widen the light absorption band to the visible region. As a result, the incorporation of Au NPs reduces electron-hole recombination and improves light absorption, resulting in improved photocatalytic activity.

The chopped current density versus the applied voltage is illustrated in Figure 4A,B. From Figure 4A, there is a change in the current density value with on and off light. This indicates the high activity of the TiN/TiO₂/Al₂O₃ electrode for water splitting and the PEC H₂ generation process. Also, Au/TiN/TiO₂/Al₂O₃ electrode chopped current density is appeared clearly with on and off light.

The effect of light intensity on the prepared TiN/TiO₂/Al₂O₃ electrode is referred to in Figure 5a. The produced J_ph increases with increasing the light intensity from 100 to 400 mW. This is clearly appearing in the inset figure, whereas the J_ph is increased from 0.076 to 0.096 mA·cm⁻², respectively. As the light intensity increased, the J_ph increases due to the increasing electron-hole pair generation [39,40]. With increasing the light intensity, many photons per second reach the active sites of the photocatalytic material to generate free electrons at the active sites of the TiO₂/TiN catalyst. The produced current is directly proportional to the absorbed light intensity [41]. The J_ph represents the current density produced in the cell due to the water splitting, in which J_ph can represent the rate of H₂ or O₂ evolution [41,42].

On the other hand, after coating the electrode with Au nanoparticles, the produced J_ph value increases, Figure 5b, in which the J_ph value changed from 0.10 to 0.139 mA·cm⁻² with increasing the light intensity from 100 to 400 mW·cm⁻², respectively.

The quantum efficiency (QE) is the relation between the incident light photons (intensity) and produced electrons (current intensity), in which QE can be calculated using Equation (4) [43], whereas the photon flux (number of photons per second), N, can be calculated using Equation (3).

$$\mathrm{N}=\mathrm{P}·\mathsf{\lambda }/\mathrm{hc}$$

(3)

$$\mathrm{QE}={\mathrm{J}}_{\mathrm{ph}}\mathrm{C}/\mathrm{N}={\mathrm{J}}_{\mathrm{ph}}·\mathrm{h}·\mathrm{C}·\mathrm{c}/\mathrm{P}·\mathsf{\lambda }$$

(4)

Here, h is the Plank’s constant (6.626 × 10⁻³⁴ J/s), c is the light speed in space (3 × 10⁸ m/s), P is the light intensity (W·m⁻²), and $\mathsf{\lambda }$ is the light wavelength (405 × 10⁻⁹ m).

The value of N is directly proportional with the light intensity as shown in Figure S4, in which N is changed from 2 × 10²¹ to 8 × 10²¹ photon/sec as the light intensity changed from 100 to 400 mW·cm⁻². On the other hand, the QE and light intensity are almost inversely proportional to each other. The QE for TiN/TiO₂/Al₂O₃ is changed from 0.25% to 0.08% with changing the light intensity from 100 to 400 mW·cm⁻², Figure 6a. Also, the QE for Au/TiN/TiO₂/Al₂O₃ is changed from 0.34% to 0.1% with changing the light intensity from 100 to 400 mW·cm⁻², Figure 6b. From Figure 6a,b, it can be seen that there is more enhancement in the QE after Au coating. This is related to the role of Au in enhancing the catalytic properties of the catalyst and improving light absorption. The contact of the electrode with the sewage water causes increasing of the local charges around Au nanoparticles. These local charges enhance the photocatalytic activity of the electrode and increase the water-splitting reaction rate. Furthermore, due to the strong corrosion resistance of Au under most environmental conditions, the layer of the Au nanoparticles on the surface protects the underlying layers and increases the photoelectrode stability [44].

The effect of monochromatic light on the produced J_ph for TiN/TiO₂/Al₂O₃ and Au/TiN/TiO₂/Al₂O₃ electrodes is shown in Figure 7a,b, respectively. From Figure 7a, it can be seen that the J_ph value is changed under the monochromatic light effect, in which 405 nm light has the optimum J_ph value (−0.092 mA·cm⁻²). This behavior is matched well with the absorption light spectrophotometry for this electrode.

On the other hand, the effect of monochromatic light on Au/TiN/TiO₂/Al₂O₃ electrode is shown in Figure 7b; the J_ph value is changed with the wavelength of the incident light. The irradiance power versus the optical wavelength (irradiance spectrum) of the 66,142–500W Hg (Xe) lamp is presented in Figure S5. The wavelength 470 nm has the optimum J_ph value (0.137 mA·cm⁻²), then wavelength 405 nm has a value of 0.131 mA·cm⁻². The incident photon to current conversion efficiency (IPCE) describes the photocurrent produced due to photon flux. It can be calculated from the wavelength values [45]. The IPCE is calculated from Equation (5).

$$\mathrm{IPCE}=\frac{{\mathrm{J}}_{\mathrm{ph}}\left(\mathrm{mA}·{\mathrm{cm}}^{-2}\right)·1240\left(\mathrm{V}·\mathrm{nm}\right)}{\mathrm{P}\left(\mathrm{mW}·{\mathrm{cm}}^{-2}\right)·\mathsf{\lambda }\left(\mathrm{nm}\right)}$$

(5)

The IPCE is calculated for Al/Al₂O₃/TiO₂/TiN as shown in Figure 7c at a light intensity of 100 mW·cm⁻². From Figure 7c, it can be seen that the electrode has the optimum IPCE of 0.25% at 405 nm. This value is decreased by increasing the wavelength to 636 nm. Moreover, after coating the electrode with Au nanoparticles, the IPCE increases to 0.39% at 405 nm. Then, the value decreases with increasing the wavelength from 405 to 636 nm. The IPCE value results from the effect of the photocatalytic property of the electrode for sewage water splitting without adding an external electrolyte. Therefore, the electrode works for converting the sewage water to H₂ and O₂ with higher efficiency than that reported by other previous studies [28,46,47,48,49,50].

3.4. Effect of Temperature and Thermodynamic Parameters

The temperature effect on the TiN/TiO₂/Al₂O₃ electrode for H₂ production due to water splitting is presented in Figure 8a, in which the J_ph increases from −0.1 to −0.28 mA·cm⁻² with increasing the temperature from 30 to 60 °C. On the other hand, the temperature has the same effect on the electrode after Au coating, Au/TiN/TiO₂/Al₂O₃, as shown in Figure 8b. in which the J_ph value increases from 0.13 to 0.33 mA·cm⁻² with increasing the temperature from 30 to 70 °C, respectively. The general increases in J_ph values represent the reaction rate, which reflects the H₂ and O₂ production rate [51,52].

From the Arrhenius equation, Equation (6), the activation energy (Ea) can be calculated depending on the particles collision and the rate of water splitting, in which k, R, and A are the rate, universal gas, and Arrhenius constants, respectively, and T is the absolute temperature [53]. From the Ea value, the degree of reaction occurrence is determined [54,55,56].

$$\mathrm{k}={\mathrm{Ae}}^{-\mathrm{Ea}/\mathrm{RT}}$$

(6)

From Figure 8c,d, it can be seen that the slope values of the relation ln J_ph and 1/T give Ea values. The Ea values are 27.22 and 18.84 kJ·mol⁻¹ for electrodes Al/Al₂O₃/TiO₂/TiN and Al/Al₂O₃/TiO₂/TiN/Au, respectively. The Ea values for both electrodes are low when compared to earlier published data for other photocatalysts [54,55,56,57,58]. So, the prepared electrodes are efficient for H₂ and O₂ evolution due to the water-splitting reaction [57]. Therefore, the Au coating over the Al/Al₂O₃/TiO₂/TiN electrode has a significant role in decreasing the Ea energy and increasing the probability of water-splitting occurrence. This is related to the plasmonic effect of Au nanoparticles and light capture phenomena [58]. In the same manner, the enthalpy (ΔH*) and entropy (ΔS*) can be estimated from the Eyring equation, Equation (7), using the Boltzmann constant (kB) and the Planck constant (h) [59,60].

$$\mathrm{k}=\mathrm{T}·\frac{\mathrm{kB}}{\mathrm{h}}·{\mathrm{e}}^{\mathsf{\Delta }\mathrm{S}/\mathrm{R}}·{\mathrm{e}}^{-\mathsf{\Delta }\mathrm{H}/\mathrm{RT}}$$

(7)

For calculating the ΔH* and ΔS* values, the slope, and intercept from Figure 8e,f are used. The values of ΔH* for TiN/TiO₂/Al₂O₃ and Au/TiN/TiO₂/Al₂O₃ electrodes are 24.26 and 15.77 J·mol⁻¹, respectively, while ΔS* values are 238.1 and 211.5 kJ⁻¹·mol⁻¹.

3.5. Stability of the Photoelectrode

The stability of the prepared TiN/TiO₂/Al₂O₃ and Au/TiN/TiO₂/Al₂O₃ photoelectrodes was studied as illustrated in Figure 9a,b. The stability represents the relation between time and the produced J_ph value by applying a potential of −0.9 and 0.9 V on the two electrodes before and after Au coating, respectively. From Figure 9a, the electrode without Au is stable with time (1000 s) at about −0.06 mA·cm⁻². After Au coating, the J_ph value becomes stable at about 0.07 mA·cm⁻², Figure 9b. These values indicate the TiN and Au coating materials have high stability and anticorrosion properties that maintain the produced J_ph at the same values with time [13,61]. This means that every small change in the produced J_ph value refers to a small corrosion reaction due to the acidity of the sewage water.

As illustrated in Figure S6, the number of H₂ and O₂ moles was determined using Equation (8) [62,63], where F is the Faraday constant and dt is the time change.

$${\mathrm{H}}_2\mathrm{mole}={{\displaystyle \int }}_0^{\mathrm{t}}{\mathrm{J}}_{\mathrm{ph}}·\mathrm{dt}/\mathrm{F}$$

(8)

According to the graph, both gases increase with time, corresponding to 7369.6 and 7915.0 mol h⁻¹ cm⁻² for H₂ and O₂, respectively.

The reproducibility of the TiN/TiO₂/Al₂O₃ and Au/TiN/TiO₂/Al₂O₃ photoelectrodes is demonstrated in Figure 9c,d, respectively, by repeating the relation between current density and voltage three times at 30 °C under the simulated sunlight. The standard deviation (SD) values for the two photoelectrodes are 0.3% and 1.29%, respectively.

For comparison, the performance indicators, J_ph and IPCE, of the prepared electrodes are displayed in

Table 1. Comparison of the IPCE and J_ph values of the prepared photoelectrodes with previous work.

Photoelectrode	Electrolyte	Jph	IPCE % (405 nm)	Ref.
BiFeO3	H2SO4	0.1 mA·cm−2	0.21	[28]
PrFeO	H2SO4	−130 µA·cm−2	-	[46]
Au/Pb(Zr, Ti)O3	H2SO4	0.06 mA·cm−2	0.2	[47]
Poly(3-aminobenzoic acid) frame	H2SO4	0.08 mA·cm−2	-	[50]
TiN-TiO2	H2SO4	3.0 nA·cm−2	0.03	[64]
TiN/TiO2/Al2O3 Au/TiN/TiO2/Al2O3	Sewage water Sewage water	0.0.9 mA·cm−2 0.14 mA·cm−2	0.25 0.39	Present work

relative to some previously reported electrodes for the water-splitting reaction to evaluate O₂ and H₂ [28,46,47,50,63,64]. In addition to the low cost and high controllability of the prepared nanomaterials, our prepared photoelectrode has relatively high IPCE and J_ph values comparative to the previously reported photoelectrodes in Table 1.

3.6. Mechanism of Electron Transition

There are two types of photocatalytic mechanisms for TiN/TiO₂ and Au/TiN/TiO₂: (1) photo-induced hot-electron transfer to an adjacent reactant, namely interfacial charge transfer, and (2) localized surface plasmon resonance (LSPR)-based electron-hole separation, namely radiative energy transfer, which requires the plasmonic band to be connected with the TiO2 band gap. Optical and electrochemical experiments were used to investigate the two types (Figure 3a, Figure 4 and Figure 10). Furthermore, the number of H₂ and O₂ moles has been computed (Figure S6).

To boost the photocatalytic optical trapping capability and photoelectric conversion rate, this approach uses LSPR to produce enhanced local electromagnetic fields around the TiO₂ photocatalyst. The aim is to create finely regulated TiN or Au nanostructures with suitable LSPR coupling to the incident light [65,66]. Plasmonic improves light absorption and broadens TiO₂ absorption in this case [67]. The increased electromagnetic field formed by the coupling between incident light and plasmonic materials transmits radiative energy from the TiN or/and Au plasmonic materials to the semiconducting TiO₂. The UV-visible spectra of the TiN/TiO₂ reveal absorption bands between 350 and 600 nm representing the surface plasmon resonance energy, as shown in Figure 3b. The intense absorption is associated with the considerable enhancement in the electromagnetic field in the vicinity of the TiN nanomaterials. As a result, it is realistic to expect TiN nanoparticles to enable a long-lived LSPR process with high-optical cross-sections and tunable throughout a broad energy range, including deep into the UV and Vis [68].

When plasmonic materials’ LSPR modes are coupled with TiO₂, the resultant broadens, resulting in increased coupling efficiency with incident light. By enhancing light absorption, promoting electron-hole creation, and heating the surrounding area, this enhanced field can be used to improve photocatalytic performance [69]. When a TiO₂ photocatalytic layer is applied to the plasmonic structure, the increased field is confined to the interior of the TiO₂ layer, and the plasmon-induced electron-hole pairs diffuse to the photocatalytic surface, contributing to the photocatalytic process. The plasmonic response in the near UV spectrum rises with size t, it can be concluded.

The increased electromagnetic field formed by the interaction of incident light and plasmonic materials transfers radiative energy from the plasmonic materials to the semiconducting TiO₂ [70]. The radiative energy boosts the photocatalysis efficiency when the LSPR energy is coupled with the TiO₂ bandgap, otherwise, the energy is more likely to decay nonradiatively through electron–phonon interactions and no significant enhancement in the photocatalytic activity can be observed [71].

At the TiO₂/plasmonic interface, a significant redistribution of charge occurs depending on the plasmonic work function and the semiconductor electron affinity. In the n-type TiN/TiO₂ and Au nanomaterials, the charge redistribution forms the Schottky barrier which builds up an internal electric field inside the photocatalyst. The electric field facilitates the transfer of the photoexcited electrons and suppressing electron/ hole recombination and improving the quantum efficiency of photocatalysis.

The TiN nanostructures possess a localized plasmonic resonance in the visible and near IR region, similar to the Au nanostructures [72,73]. The difference between the behavior of TiN and Au nanostructure is that TiN produces Ohmic junction whereas Au forms Schottky junction with TiO₂ material [64]. Moreover, TiN is a low-cost material with corrosion resistance and mechanical strength properties. The enhancement in the photocatalytic properties with incorporation in the TiO₂ materials is related to the conversion of energy into electrons below the bandgap of TiO₂ but limits the potential barrier. TiN has a work function QM of about 4 eV in a vacuum, which matches with the TiO₂ electron affinity, and this forms an energetic alignment for the hot electrons [74].

From Figure 10a,c, it can be seen that the fabricated TiN/TiO₂ photoelectrode has an Ohmic junction, which facilitates the electron transition to TiO₂ without barriers. These electrons contribute to the water-splitting reaction and the negative potential is expected to facilitate the H₂ production. On the other hand, incorporating Au nanoparticles boosts photocurrent density while simultaneously forming a Schottky circuit, as seen in Figure 10b,d [74]. As a result, not all electrons absorbed by Au nanoparticles can travel to TiN and TiO₂ materials for water splitting (cold electrons), while hot electrons can only pass through the barrier for the water-splitting reaction. So, the addition of Au nanomaterial increases the potential to be a positive value.

4. Conclusions

This work is very interesting for H₂ production from sewage water using TiN/TiO₂/Al₂O₃ nanotube material. The Al₂O₃ template has been prepared with high accuracy and ordering with hexagonal pores using the Ni imprinting method. TiN/TiO₂ is deposited inside the pore using ALD and DC sputter-coating method, respectively. Characteristic analyses of the structural, morphological, and optical properties were carried out for the prepared template and nanotube materials. The application of the TiN/TiO₂/Al₂O₃ materials as photocathode carried out with and without Au nanoparticles coating, in which the Au materials facilitate the sewage water-splitting reaction. The effect of different parameters was studied for water-splitting reaction, such as light intensity, light wavelength, and temperature. The values of QE and IPCE were calculated with a value of 0.34 and 0.39% at 400 nm. Moreover, all thermodynamic parameters are calculated for the photocathode for H₂ production, wherein Ea, ΔH*, and ΔS* values were 18.84 kJ·mol⁻¹, 15.77 J·mol⁻¹, and 211.5 kJ⁻¹·mol⁻¹ for Au/TiN/TiO₂/Al₂O₃ photoelectrode.