3.1. Catalyst characterization
X-ray diffraction (XRD) analysis was conducted to investigate the crystalline structure of the photocatalysts, including NWT, Pt-NWT, Cu-NWT, and Ag/Cu-NWT (Fig. 2a and S1). The XRD pattern of NWT revealed a highly crystalline anatase structure with dominant peaks at 2θ = 25.3◦, 37.9◦, 48◦, and 54.9◦ corresponding to the (101), (004), (200), and (211) planes, respectively, as indicated by the JCPDS Card No. 21-1272 (Zhu et al. 2021). Figure 2a shows the XRD patterns of NWT, 0.5Pt-NWT, 1.0Cu-NWT and 0.1Ag/1.0Cu-NWT photocatalysts. For all these four photocatalysts the major peaks observed were of the anatase phase of TiO2. There was no peak of Pt, Cu or Ag found in the XRD pattern because of the very small quantity of these metals in the metal-decorated NWT photocatalysts. All the optimized photocatalysts also showed the characteristic peak at 2θ = 25.3◦, 37.9◦, 48◦ and 54.9◦ corresponding to (101), (004), (200) and (211) planes respectively which not only confirms the formation of high crystallinity but also pure compound. The intensity of the reflections related to the TiO2 anatase phase decreased following the deposition of metals, suggesting that the presence of an additional crystalline phase had modified the crystalline structure of TiO2 (Alshehri and Narasimharao 2020). High crystallinity and co-presence of Ag and Cu on NWT photocatalysts are responsible for enhanced light harvesting capacity and ultimately the photocatalytic hydrogen production. Fig. S1a shows the XRD pattern of NWT and Pt-NWT (0.2Pt-NWT, 0.4Pt-NWT, 0.5Pt-NWT, 0.6Pt-NWT, 0.8Pt-NWT and 1.0Pt-NWT) photocatalysts and Fig. S1b shows the XRD of Cu-NWT (0.5Cu-NWT, 0.75Cu-NWT, 1.0Cu-NWT, 1.5Cu-NWT and 2.0Cu-NWT) photocatalysts. The XRD pattern of all Pt-NWT and Cu-NWT photocatalysts retained the peak of bare NWT and no other peak or impurity was observed.
Raman measurements for all four samples i.e., NWT, 0.5Pt-NWT, 1.0Cu-NWT and 0.1Ag/1.0Cu-NWT were carried out. Figure 2b exhibits the Raman spectroscopy analysis, elucidating the characterization of the Ti-O-Ti network structure of NWT materials within the wavenumber range of 100–900 cm− 1. Notably, the well-established anatase peaks at 142, 393, 511, and 635 cm− 1 are distinctly observed. In the case of NWT, these four pronounced peaks at 142, 393, 511, and 635 cm− 1 correspond to the vibration modes of Eg, B1g, A1g, and Eg, respectively, aligning with the TiO2 anatase phase. The Eg peak arises predominantly from the symmetric stretching vibration of the O–Ti–O linkages in TiO2, whereas the B1g peak corresponds to the symmetric bending vibration of O–Ti–O and the A1g peak is attributed to the antisymmetric bending vibration of the O–Ti–O linkage (Balachandran and Eror 1982; Melvin et al. 2015). In the case of Pt-NWT, Cu-NWT, and Ag/Cu-NWT, no distinct separate peak was observed. However, a slight shift towards lower wavenumbers and broadening of the peak were noted compared to bare NWT. This phenomenon can be attributed to the strong electronic interaction between the metals and the surface of NWT (Fan et al. 2014). Thus, Ag and Cu decoration improves the crystal structure of anatase TiO2. The phase of NWT was retained even after the metal decoration. These findings are consistent with the XRD data mentioned earlier. Fig. S2 shows the FTIR spectra of NWT, 0.5Pt-NWT, 1.0Cu-NWT and 0.1Ag/1.0Cu-NWT photocatalysts. The FTIR spectra manifest prominent and comprehensive peaks, signifying the existence of diverse functional groups within the materials. Notably, a slight dip around 3360 cm− 1 and 1643 cm− 1 can be attributed to the stretching and bending vibrations of adsorbed molecules and surface hydroxyls on TiO2 particles, respectively (Li et al. 2016; Chougala et al. 2017). The intense peak at 485 cm− 1 is assigned to the Ti-O stretching band which is the characteristic peak of TiO2 (Zhang et al. 2014). There is no significant peak-shift or extra peak observed in optimized metal decorated NWT photocatalysts because of a very small percentage of metal present in it and/or can be mainly attributed to the metal loading effect.
The optical and electronic properties of materials have been characterized by using Diffuse Reflectance Spectroscopy (DRS). To gain insights into the optical absorbance of the optimized photocatalysts, namely NWT, 0.5Pt-NWT, 1.0Cu-NWT, and 0.1Ag/1.0Cu-NWT, Fig. 2c presents the UV-visible diffuse reflectance spectroscopy (DRS) data. The strong absorption bands around 380 nm, attributed to the intrinsic absorption of TiO2, are thought to result from the excitation of O2p electrons to the Ti3d level (Xu et al. 2010; Zhang et al. 2015b). Absorption edges of NWT, 0.5Pt-NWT, 1.0Cu-NWT and 0.1Ag/1.0Cu-NWT are around 390 nm, 408 nm, 420 nm and 435 nm respectively. The DR spectra of the modified NWT is noteworthy for exhibiting a significant shift in the absorption transition towards longer wavelengths (400 nm < λ < 435 nm). The extended visible light absorption of 0.1Ag/1.0Cu-NWT among all photocatalysts is due to the plasmonic resonance and light harvesting capacity of Ag and Cu respectively which shows absorption above 400 nm (Wang et al. 2022). The absorption of 0.5Pt-NWT and 1.0Cu-NWT was found to be significantly lower compared to 0.1Ag/1.0Cu-NWT, which could be attributed to the lower sensitivity of Pt and Cu to visible light absorption relative to Ag. Ultimately co-decoration of both Ag and Cu on NWT is found to be responsible for improved photocatalytic hydrogen production. Optical band gap studies have been carried out using Tauc Plot (Fig. S9), observed that the absorption shifts are consistent with a reduction in the optical band gap energies, mirroring the trends observed in the absorption spectra. Notably, Ag/Cu-NWT (3.06 eV) exhibits a significantly lower band gap energy compared to bare NWT (3.27 eV), Pt-NWT (3.20 eV), and Cu-NWT (3.15 eV). The reduction in band gap energy in Ag/Cu-NWT is likely due to the synergistic effect of silver and copper decoration, which enhances charge carrier separation and facilitates better light absorption in the visible region, thereby narrowing the band gap (Scarisoreanu et al. 2020). While Pt-NWT and Cu-NWT also exhibit some degree of band gap narrowing, they are not as effective as the Ag/Cu combination. Platinum, although excellent for enhancing photocatalytic activity, is costly and does not significantly affect light absorption in the visible range. Copper alone provides some enhancement in visible light absorption, but it is the addition of silver that amplifies the effect due to the plasmonic resonance (Zhu et al. 2019; Scarisoreanu et al. 2020; Liu et al. 2022).
To gain insight into the growth process of the NWT structure, we monitored the growth steps of the samples using SEM at various NaOH concentrations. Fig. S3 shows the SEM images of synthesised TiO2 at different NaOH concentrations at the same magnification i.e., at 1µm. Irregular morphology of TiO2 (TiO2@5M) was observed with 5M NaOH solution in the synthesis method. Well-organized 3D spheres made up of nanowire TiO2 (NWT) morphology was obtained with 10M NaOH solution. The rod like (TiO2@15M) and scale like (TiO2@20M) morphologies of TiO2 were obtained when 15M and 20M NaOH solution was used in the synthesis method. It is obvious from the above data that the NaOH concentration in the reaction mixture plays crucial role in morphology elucidation. It can also be inferred that the morphology of the product can be controlled by the choice of solvent (Lu et al. 2021). Solvent provides sufficient space to grow the titanium oxide into nano-wire ball like morphology. The observed variations in morphologies can be attributed to the polarity and coordinating ability of the co-solvent, which significantly affects the solubility, reactivity, and diffusion behaviour of the reactants, thereby influencing the structural and morphological characteristics of the final products (Das et al. 2008; Lu et al. 2021). The optimized NWT morphology not only enhanced light harvesting ability due to its unique nanowire ball like structure but also improved its surface area due to high porosity for enhanced photocatalytic hydrogen production.
The efficient 0.1Ag/1.0Cu-NWT photocatalyst's microscopic nature was first characterized by SEM (Fig. 3a-c), and then by low-resolution transmission electron microscopy (LRTEM) and high-resolution transmission electron microscopy (HRTEM) (Fig. 3d-h). Uniformly dispersed ball like structure made up of nanowire is shown in Fig. 3a having diameter of nanowire as 10–20 nm and the whole NWT structure is around 3–4 µm as confirmed by SEM images (Fig. S4). Figure 3d-h shows LRTEM and HRTEM images of optimized 0.1Ag/1.0Cu-NWT photocatalyst. The lattice fringes observed in the HRTEM image (Fig. 3g-h) of 0.1Ag/1.0Cu-NWT correspond to the (111) plane of Ag (d = 0.24 nm) (Li et al. 2017), (111) plane of Cu (d = 0.211 nm) (Kainthla et al. 2018) and (101) plane of anatase TiO2 (d-space value of 0.326 nm) (Reddy et al. 2017). These results provide clear evidence of homogeneously dispersed Ag and Cu in their metallic forms on the NWT surface, indicating a close association between them. The crystalline quality of the 0.1Ag/1.0Cu-NWT photocatalyst was further demonstrated by the Selected Area Diffraction (SAED) pattern shown in Fig. 3i, which reveals a well-defined diffraction pattern with a lattice spacing of 0.326 nm, matching the (101) plane of anatase TiO2. SAED pattern of the TiO2 nanowires reveal their single crystallinity and conclude the growth of the single-crystal TiO2 nanowire/nanorod along the [0 0 1] direction (Sun et al. 2013). Corresponding LR and HRTEM along with SAED pattern for 0.5Pt-NWT and 1.0Cu-NWT photocatalysts are shown in Fig. S5a-d and S5e-h respectively.
Figure 4 presents the elemental mapping and TEM-EDAX analysis to visualize the distribution of elements and determine the atomic percentage of each element across the 0.1Ag/1.0Cu-NWT photocatalyst. Figure 4a-f shows elemental mapping of 0.1Ag/1.0Cu-NWT photocatalyst depicting the homogeneous dispersion of elements throughout the sample. Figure 4g shows TEM-EDAX spectra of 0.1Ag/1.0Cu-NWT photocatalyst which confirms the presence of 0.13% Ag, 6.72%Cu, 53.56% O and 39.58% Ti atomic percentages. Cu atomic percentage has been found to be higher than the actual percentage of synthesized 1%Cu-NWT which might be due to the copper gride used for the sample analysis. Overall observation confirmed the presence of Ag and Cu in the 0.1Ag/1.0Cu-NWT photocatalyst.
The impressive organization of nanowires in TiO2 serves as a driving force for our in-depth examination for its Brunauer–Emmett–Teller (BET) surface area and pore volume. Consequently, we conducted an investigation onto the N2 sorption isotherm and corresponding BET surface area plot of both P25 TiO2 and NWT, aiming to validate the superior photocatalytic performance of NWT compared to P25 TiO2 (as shown in Fig. S6). TiO2 being the main photocatalyst which would contribute to the photocatalytic efficiency majorly due to its surface area and other properties, therefore BET surface area, pore volume, and pore size diameter of the P25 and NWT has been reported in Table 1. Metal decoration or co-decoration did not contribute much to the surface area. As expected, the BET surface area of NWT (123 m2 g− 1) was found to be significantly higher than that of P25 TiO2 (52.90 m2 g− 1). The well-organized porous morphology of NWT exhibits a 2.5 times greater surface area than P25 TiO2, leading to improved light harvesting ability through repeated reflection and refraction. Additionally, the increased surface area provides more active reaction sites and facilitates mass transfer, allowing for efficient transport of reactant and product molecules.
Table 1
BET surface area, pore volume and pore size diameter values of the P25 TiO2, and NWT samples.
Sample | BET surface area (m2/g) | Pore volume (cm3/g) | Pore size diameter (nm) |
P25 | 52.90 | 0.354 | 25.94 |
NWT | 123.00 | 0.636 | 16.20 |
Since 0.1Ag/1.0Cu-NWT display high crystallinity and well optical response ability, it is very necessary to investigate its surface compositions and electronic status, X-ray photoelectron spectroscopy (XPS) analysis was also performed. Figure 5a illustrates the survey spectrum of the 0.1Ag/1.0Cu-NWT photocatalyst, revealing the presence of elements Ti, O, Cu, and Ag. Figure 5b shows the peaks corresponding to Ti 2p at 458.80 eV and 464.55 eV, and 530.23 eV and observed data of O1s peak (Fig. 5c ) includes one major peak at 530.23 eV and small shoulder also appeared at 532.22 eV corresponding to the binding energies of O1s, which was attributed to O 1s electron binding energy for TiO2 and the H–O from the absorbed H2O on their surface respectively (Erdem et al. 2001; Kruse and Chenakin 2011; Zhu et al. 2017). The Ti 2p region displays an asymmetric peak at 458.80 eV and 464.55 eV, corresponding to the binding energies of Ti 2p3/2 and Ti 2p1/2, respectively (Erdem et al. 2001; Su et al. 2013). The binding energy values at 367.85 and 373.82 eV refers to Ag 3d5/2 and Ag 3d3/2, respectively (Fan et al. 2014). This observation confirms the presence of metallic silver (Ag⁰) in the sample. The peaks at 932.75 and 952.56 eV are identified as Cu 2p3/2 and Cu 2p1/2, respectively, indicating the presence of Cu in reduced form (Cu⁰) (Zhang et al. 2015a; Reddy et al. 2017). High resolution spectra of Cu 2p and Ag 3d serving as proof of Cu and Ag in their metallic states. The XPS spectra of 0.5Pt-NWT photocatalyst is depicted in Fig. S8, and the presence of Platinum species was confirmed through deconvolution of Pt 4f peak into two components, Pt 4f7/2 and Pt 4f5/2, at binding energies 68.97 and 72.52 eV, respectively (Fig. S8d) (Shinde et al. 2018b; Sravani et al. 2020).
3.2. Photocatalytic hydrogen production
A series of methodical experiments on photocatalytic hydrogen production were conducted, employing simulated solar light irradiation, and utilizing methanol as a sacrificial agent across all the prepared photocatalysts. A comparison was made with commercially available P25 TiO2. P25 TiO2, being the widely recognized benchmark photocatalyst in commercial use, is the preferred starting point for further development and modifications aimed at overcoming its limitations. The objective is to augment the efficiency while maintaining the cost-effectiveness of both the photocatalyst and the experimental configuration for solar-driven photocatalytic hydrogen generation. Thus, motivated by this idea, focused attempts were made to obtain and optimize the morphology of synthesized TiO2 which would deliver much more superior results as compared to commercially available P25 TiO2 in terms of efficiency.
Thus initially, the effect of morphology of TiO2 on the hydrogen production capacity was studied. Moreover, the optimized morphology was further improved by decorating it with noble metal Pt, non-noble metal Cu, and a combination of non-noble Cu and a minimal amount of noble metal Ag to further enhance hydrogen production. The trend observed for photocatalytic hydrogen production, as depicted in Fig. 6a, is as follows: P25 TiO2 < NWT < Cu-NWT < Pt-NWT ≤ Ag/Cu-NWT. Optimized low-cost 0.1Ag/1.0Cu-NWT (10,184 µmol/g) photocatalyst showed the highest hydrogen production as compared to all P25 TiO2, NWT, Cu-NWT photocatalysts and at par hydrogen production with 0.5Pt-NWT (10,050 µmol/g) photocatalyst.
Morphology obtained by using 10 M NaOH concentration i.e., NWT showed a dramatic increment in hydrogen production than other morphologies obtained at 5 M, 15 M and 20 M respectively (Fig. 6b). In comparison to the commercially available P25 TiO2 (99 µmol/g), NWT exhibited a remarkable nearly 7-fold increase in hydrogen production capacity (678 µmol/g). This significant improvement in photocatalytic efficiency can be attributed to the favourable morphology achieved through the specific arrangement of atoms. This morphology enhances the absorption of light energy (as shown in Fig. 2d and S7) and facilitates mass transfer through the increased specific surface area (see BET explanation).
To achieve the highest possible activity, NWT was modified by decorating with the popular noble metal Pt since Pt has more work function as well as better reduction potential than other noble metals. Various % Pt-NWT photocatalysts were synthesized and checked for photocatalytic hydrogen production. Optimized 0.5Pt-NWT (10,050 µmol/g) showed a better increment in hydrogen production than other Pt percentages. 0.2 to 0.5 percentage of Pt with NWT i.e., 0.2Pt-NWT, 0.3Pt-NWT, 0.4Pt-NWT and 0.5Pt-NWT shows gradual increment in hydrogen production (Fig. 6c). The photocatalytic performance of nanowire-tethered (NWT) materials decreases when the Pt content exceeds 0.5%, a result of the light-blocking and scattering effects caused by the accumulation of Pt on the NWT surface (Sun et al. 2012; Lee and Chang 2019). However, the optimized 0.5Pt-NWT exhibited significantly enhanced hydrogen production, reaching nearly a 15-fold increase compared to NWT (678 µmol/g) and a remarkable 102-fold increase compared to P25 (99 µmol/g). The reason behind the betterment is the Pt decoration on NWT which reduces the electron hole recombination and increases the absorption of light energy (Fig. 2b-c and 7).
Further, as Pt-TiO2, is a very exhaustively studied system due to the very high work function of Pt to reduce H+ to H2 for enhanced efficiency, it is always a challenge to work out a system or photocatalyst which would be a cost-effective alternative to beat Pt-TiO2. Therefore, the study was further directed to modify the optimized NWT photocatalyst with bimetallic non-noble metal Cu and low-cost noble metal Ag to explore and exploit their very fascinating properties of SPR effect and extended solar light absorption for improved hydrogen production activity at par with Pt-TiO2 systems.
Firstly, TiO2 decoration with varying Cu concentration in the range of 0.5 to 2% has been studied. 1% Cu-NWT was found to show excellent improvement in hydrogen production sample among other concentrations; however these results were still lower than the Pt-NWT system.
In this attempt, initially, NWT was decorated with an inexpensive non-noble metal Cu with its percentage varying from 0.5 to 2%. (Fig. 6d). The increasing hydrogen production trend was observed from 0.5%to1.0%, and on further increase in Cu % the trend was found to decrease. The optimized 1.0Cu-NWT photocatalyst (3,907 µmol/g) demonstrated a substantial increase in hydrogen production, achieving nearly 6-fold improvement compared to NWT (678 µmol/g) and a remarkable 40-fold increase compared to P25 TiO2 (99 µmol/g). This enhancement in hydrogen production is due to the presence of Cu which absorbs plasmon-induced irradiation to generate hot electrons which are partly ejected to CB of TiO2 overcoming the Schottky barrier and separating the photogenerated charge carriers significantly (Zhang et al. 2015a; Kumaravel et al. 2019).
Further, optimized 1.0Cu-NWT was decorated with a minimal quantity of Ag to improve the efficiency of the photocatalyst and to get comparable results at par with Pt-NWT photocatalyst. A very small quantity of Ag i.e., 0.1% Ag was decorated on an optimized 1.0Cu-NWT photocatalyst which resulted in a dramatic increment in photocatalytic hydrogen production capacity as compared to Cu-NWT, NWT as well as P25 TiO2. 0.1Ag/1.0Cu-NWT (10,184 µmol/g) photocatalyst showed almost 2.6-fold increment in hydrogen production than 1.0Cu-NWT (3,907 µmol/g) photocatalyst.
This improvement also showed comparable hydrogen production as of optimized 0.5Pt-NWT (10,050 µmol/g) photocatalyst (Fig. 6a). The main reason for choosing 0.1%Ag on 1.0 Cu-NWT system was to highlight that even the very small amount viz. 0.1% of Ag along with 1.0Cu-NWT will be sufficient to replace the 0.5Pt-NWT. It is indeed interesting to study the effect of variable concentration of Ag with 1.0 Cu-NWT system for further enhancement in hydrogen production. The detail study and optimization of variable metal concentrations and combinations of different metals will surely replace the Pt in future for cost-effective catalysts for cheap hydrogen production. The remarkable photocatalytic performance exhibited by the 0.1Ag/1.0Cu-NWT photocatalyst can be attributed to several key factors. Firstly, its unique morphology and bimetallic decoration play a crucial role. These features enable enhanced absorption of visible light, resulting in more efficient utilization of solar energy. Additionally, the modified structure of the photocatalyst leads to an increased surface area, providing more active sites for the photocatalytic reaction. Moreover, the bimetallic decoration facilitates accelerated charge transfer processes, enhancing the overall efficiency of the photocatalyst. Lastly, the presence of silver and copper on the nanowires effectively suppresses the recombination of photogenerated electron-hole pairs, leading to an extended lifetime of the charge carriers and further enhancing the photocatalytic performance. These attributes ultimately result in excellent photocatalytic hydrogen production, as explained in the mechanism section.
To evaluate the reusability of the optimized 0.1Ag/1.0Cu-NWT photocatalyst, cyclic experiments were performed under identical experimental conditions. As depicted in Fig. 6e, the amount of hydrogen production remained consistent throughout the five cycles. This observation indicates that the photocatalyst maintained its activity without any noticeable decrease, demonstrating its remarkable stability.
3.3. Mechanism of photoinduced charge transfer in Ag/Cu-NWT
The results obtained indicate that the 0.1Ag/1.0Cu-NWT photocatalyst possesses remarkable photocatalytic hydrogen production ability under simulated solar light irradiation. In order to elucidate the mechanism responsible for enhanced photocatalytic hydrogen production, a plausible charge transfer pathway is proposed and visually illustrated in Fig. 7.
The achieved nanowire ball-like morphology of the photocatalyst through the use of 10M NaOH resulted in a significant improvement in hydrogen production. This enhancement can be attributed to the unique structure of the photocatalyst, which plays a vital role in efficiently harnessing light energy through repeated reflection and refraction from the light source. This ball-like arrangement of nanowires also has been found to improve the porosity and surface area of the photocatalyst which ultimately facilitates more reaction sites and mass transfer. The 2.5 times improved BET surface area and UV-visible DRS (Fig. 2d) spectral data of NWT clearly demonstrates probable reasons for 7-fold improved photocatalytic hydrogen production as compared to P25 TiO2.
To achieve the highest possible activity at par or better than Pt-TiO2 with an added advantage of cost effectivity, the NWT was decorated with non-noble metal Cu and a minimal quantity of noble metal Ag. 0.1Ag/1.0Cu-NWT. The unique structure of this photocatalyst exhibited superior hydrogen production compared to all other prepared photocatalysts, and similar hydrogen production to that of 0.5Pt-NWT. This can be attributed to the proposed charge transfer mechanism, where upon light irradiation, electrons are excited from the valence band to the conduction band of NWT, and hot electrons are continuously supplied to the conduction band of TiO2 due to the collective oscillation of conduction band electrons of decorated Ag nanoparticles. The surface plasmon resonance (SPR) effect of the decorated Ag nanoparticles enhances the absorption of visible light, resulting in more photoelectrons being available in the conduction band of NWT. (Nyamukamba et al. 2017). Further, these photoexcited electrons are captured by Cu which acts as an electron sink for the effective H+ to H2 reduction reaction. Thus, Cu decoration not only acts as a co-catalyst but also plays an important role in charge separation (Reddy et al. 2017; Wang et al. 2022). The VB holes are trapped by methanol which acts as a hole scavenger or used for the oxidation process of H2O (Denisov et al. 2019). Hence, notwithstanding the significance of the SPR effect, the catalytic activity of co-metallic decorated Ag and Cu on NWT exerts a more influential effect by mitigating electron-hole recombination and promoting the charge transfer rate. In conclusion, the combination of the tailored morphology of TiO2 (NWT) and the co-metallic decoration (Ag/Cu-NWT) bestows exceptional characteristics upon the photocatalyst. These features encompass amplified light absorption in the visible spectrum, augmented surface area, expedited efficiency in charge transfer, and subdued recombination of photogenerated electron-hole pairs. Collectively, these factors contribute to the remarkable photocatalytic hydrogen production achieved by the catalyst.