Skip to main content
NCBI home page
ACS Materials Au logoLink to ACS Materials Au
. 2024 Dec 9;5(2):299–307. doi: 10.1021/acsmaterialsau.4c00084

CVD Grown Sub 10 nm Size g-C3N4 Particle-Decorated TiO2 Nanotube Array Composites for Enhanced Photocatalytic H2 Production

Kosei Ito 1,*, Sho Yoneyama 1, Shusuke Yoneyama 1, Paul Fons 1, Kei Noda 1,*
PMCID: PMC11907282  PMID: 40093843

Abstract

graphic file with name mg4c00084_0008.jpg

TiO2 nanotube arrays (NTA) have attracted much attention among photocatalysts because of their large specific surface area and easy surface transfer of excited electrons, and in recent years, attempts have been made to further improve their properties by forming Z-schemes when they are composited with other photocatalysts. However, as the spacing within and between nanotubes is only a few nanometers, the formation of heterojunctions is extremely difficult when TiO2–NTA is composited with other photocatalytic materials with larger grain sizes. Creating nanoparticle photocatalysts with dimensions smaller than those of the nanotube system is thus required to effectively form heterojunctions. We have constructed an original vacuum chemical vapor deposition (CVD) system with fine temperature control, an attribute that we believe is necessary for the preparation of small nanoparticles. Using this system, it is possible to greatly reduce the polymerization rate of melamine, the precursor of the carbon nitride (g-C3N4) photocatalyst, which offers the benefits of increased reduction power and a metal-free composition. As a result, g-C3N4 small nanoparticles with particle sizes of about 10 nm were successfully prepared, and heterojunctions could be formed even inside TiO2–NTA. The fabricated TiO2–NTA/g-C3N4 composite structure exhibited significantly improved redox power and photocatalytic hydrogen production compared to TiO2–NTA and g-C3N4 alone. In addition, while the hydrogen production rates for TiO2–NTA and g-C3N4 were almost constant, TiO2–NTA/g-C3N4 showed a rapid increase in the hydrogen production rate after a certain period of light irradiation, which was presumably caused by oxygen desorption from g-C3N4. The results of this study provide a method for supporting small nanoparticle materials on nanotube substrates and their importance in improving photocatalytic properties, and will also make a significant contribution not only to the field of photocatalysis but also to other fields requiring small nanoparticle materials.

Keywords: TiO2 nanotube arrays, carbon nitride, chemical vapor deposition, small nanoparticles, photocatalyst

1. Introduction

Titanium dioxide (TiO2) photocatalysis has been studied by many researchers since it was reported by Honda et al. in 1972.1 Among TiO2 photocatalysts, TiO2 nanotube arrays (NTA) have attracted much attention because of their large specific surface area and easy surface transfer of excited electrons (Figure 1a).2 In addition, NTA is a substrate material and can be used for a wide range of practical applications. Recently, attempts to improve the performance of TiO2–NTA have been carried out by combining TiO2–NTA with other materials to enhance both the reduction power and spatial separation of electron–hole pairs, which have been weak points of TiO2–NTA.3,4 This form of composite photocatalysts with enhanced oxidation and reduction power is widely known as Z-scheme photocatalysts, and the key to an efficient Z-scheme system is the formation of appropriate heterojunctions. Photocatalysts with strong reducing power, such as Cu2O,5 CdS,6 and g-C3N4,7 have been investigated in combination with TiO2–NTA. In this study, we attempted to complex g-C3N4 (which can be synthesized semipermanently, inexpensively, and easily) with TiO2–NTA.

Figure 1.

Figure 1

Open in a new tab

(a) SEM image of the as-grown TiO2–NTA. (b) The SEM image of TiO2–NTA/g-C3N4 synthesized by heating melamine (2 g) and TiO2–NTA in the same crucible. (c) The SEM image of TiO2–NTA/g-C3N4 synthesized by heating melamine (0.1 g) and TiO2–NTA in the same crucible.

Initially, TiO2–NTA was heated with melamine (a precursor of g-C3N4) in the same crucible. After growth, a g-C3N4 film was found to have covered the TiO2–NTA, as shown in Figure 1b. The experiment was then carried out again with a large reduction in the amount of melamine in the crucible, but a heterogeneous composite was synthesized with a thin film and large g-C3N4 particles on top of the TiO2–NTA (Figure 1c). This is not favorable for the formation of heterojunctions, which are key to Z-scheme photocatalysis and have been found to be a common problem in other previous studies.8 One way to overcome these problems is to thermally dissociate the g-C3N4 powder into small nanoparticles and support the small nanoparticles on TiO2–NTA.9 However, the thermal dissociation process leads to defects and amorphization of the original g-C3N4 material. In addition, physical vapor deposition is undesirable because it weakens the adhesion of the film to the substrate.

To solve the above problems, it is essential to create pure small nanoparticles of g-C3N4, which can enter into nanotube apertures of several tens of nanometers in diameter; it is desirable to combine them using a simple deposition method. Thus, we focused on chemical vapor deposition (CVD) apparatus. The main role of the CVD apparatus is to deposit films on devices. In fact, in previous research using a CVD apparatus to combine g-C3N4 with TiO2–NTA, a g-C3N4 film was coated onto TiO2–NTA.10 In this work, to combine g-C3N4 in the form of small nanoparticles with TiO2–NTA, we constructed an original CVD apparatus (Figure 2) that incorporated the following two new features. The first is that the temperature can be adjusted by heaters in regions 1 and 2, allowing the precursor (melamine) to be heated rather than the target material being reheated. This makes it possible to intentionally manipulate the thermal polymerization process from melamine to g-C3N4, thus allowing for chemical complexation with TiO2–NTA during the thermal polymerization process, preventing the formation of nitrogen defects or amorphization, controlling particle size during transport to the substrate, and overall simplification of the synthesis process. The second feature is that deposition is usually carried out at atmospheric pressure, but our system is evacuated to a pressure of 40–133 Pa. The resulting increase in mean-free path enables ultrasmall state particles to penetrate into the nanotubes because it suppresses polymerization caused by collisions of gas molecules during vapor transport. Furthermore, if the number of particles adsorbed on TiO2–NTA can be controlled, excessive thermal polymerization after adsorption can be prevented and the g-C3N4 can be deposited as particles rather than films. Also, the low-pressure conditions aid in the formation of highly pure particles and facilitate the formation of heterojunctions with TiO2–NTA.

Figure 2.

Figure 2

Open in a new tab

Schematic diagram of the original CVD system.

In this article, we first present the conditions for the synthesis of g-C3N4 small nanoparticles from melamine using the modified CVD system. Next, observations of the interface between TiO2–NTA and g-C3N4 small nanoparticles are discussed. Finally, the function of the combined TiO2–NTA/g-C3N4 as a Z-scheme photocatalyst is evaluated by the redox power and photocatalytic H2 production. Although the current study demonstrates how to composite small nanoparticle photocatalysts into nanotubular substrate photocatalysts and their importance in improving photocatalytic properties, the results of this study are expected to contribute more broadly to not only the field of photocatalysis but also to other fields that require small nanoparticles.

2. Experimental Methods

2.1. Materials and Reagents

In this study, the following chemicals and reagents were purchased and used without any further purification: ethylene glycol (C2H6O2, Nacalai Tesque; >99.5%), ammonium fluoride (NH4F, Nacalai Tesque; >98%), titanium foil (Ti, Japan Metal Service; >99.5%), 2-propanol (C3H8O, Nacalai Tesque; >99.7%), melamine (C3H6N6, Tokyo Chemical Industry; >98%), terephthalic acid (TA, C8H6O4, Nacalai Tesque; >98%), sodium hydroxide (NaOH, Nacalai Tesque; 5 M), 5,5-dimethyl-1-pyrroline-N-oxide (DMPO, C6H11NO, Tokyo Chemical Industry; >97%), and methanol (CH3OH, Nacalai Tesque; >99.5%).

2.2. Preparation of TiO2–NTA by the Anodic Oxidation Method11

The electrolyte was prepared by stirring ethylene glycol (100 mL), ammonium fluoride (0.559 g), and H2O (11 mL) for 30 min. Next, the anode (titanium foil, 2.5 cm × 3.0 cm) and cathode (platinum foil) were immersed in the prepared electrolyte and a 40 V DC was applied for 1.5 h at room temperature. After application, the samples were immersed in 2-propanol for 30 min for washing. Finally, the washed sample was annealed at 400 °C for 1 h, followed by annealing at 500 °C for 3 h for crystallization.

2.3. Support of g-C3N4 Small Nanoparticles on TiO2–NTA Using CVD

First, melamine (2.0 g) was added to a quartz cell (12.5 mm × 12.5 mm × 45 mm) mounted on a heater in the CVD apparatus, and a TiO2–NTA substrate was placed in region 2. Next, the pressure in the CVD apparatus was lowered to about 40–133 Pa and a 1.5 CCM Ar gas flow was established. Finally, the melamine was heated to 300 °C. 300 °C is the temperature at which melamine begins to sublimate in the thermal polymerization process. The heating ramp time was set to 20 min and held for 10 min. The heating conditions for g-C3N4 deposition in the tube furnace were set to have a variable temperature in region 1 and a constant temperature of 500 °C in region 2. Region 2 was set to 500 °C because the thermal polymerization temperature of g-C3N4 is generally 500 °C and the stability limit of TiO2–NTA is 500 °C. The heating ramp times for regions 1 and 2 were set to be 30 min for heating followed by 30 min of constant temperature. The heating of the region 1 heater was initiated when the temperature in region 2 reached 500 °C.

2.4. Equipment Used for Materials Characterization

The morphology of the samples was investigated by using a scanning electron microscope (SEM, JSM-7600F, JEOL). Fine structure observation for TiO2–NTA/g-C3N4 was carried out using a transmission electron microscope (TEM) (Tecnai Osiris, FEI). Chemical state and bonding information were obtained using X-ray photoelectron spectroscopy (XPS, JPS-9010TR, JEOL) with Mg Kα radiation and Fourier transform infrared spectroscopy (FT-IR) (ALPHA, Bruker). The crystal structure was investigated by using an X-ray diffractometer (XRD, D8 ADVANCE, Bruker) with Cu–Kα radiation. To explore the detailed energy band structure, diffuse-reflectance ultraviolet–visible (UV–vis) absorption spectra were monitored by using a UV–vis-near-infrared (UV–vis-NIR) spectrophotometer (UV-3600Plus, Shimadzu). In addition, the flat band (FB) potential was determined by Mott–Schottky analysis with an electrochemical impedance analyzer (VersaSTAT3, AMETEK). For electrochemical impedance (EI) measurements, a 0.5 M Na2SO4 aqueous solution was employed as a liquid electrolyte, and Ag/AgCl and platinum wires were used as reference and counter electrodes, respectively. The light source was a xenon lamp (MAX-303, Asahi Spectra, 300 ≤ λ ≤ 600 nm, 500 W/cm2).

2.5. Photocatalyst Experiment

The oxidation potential of the photocatalytic material was evaluated from a qualitative assessment based on the amount of OH radicals produced using photoluminescence measurements (PL). First, 100 mL of pure water, 0.083 g of TA, and 0.2 mL of NaOH were added to a 100 mL beaker and stirred for 1 h. Next, 10 mL of the prepared solution was added to a 50 mL beaker, and a substrate sample cut into 2 cm × 1.5 cm was immersed and irradiated with UV–vis light (300 nm ≤ λ ≤ 600 nm) for 15 min. Finally, the solution was placed in a 3 mL quartz cell, and the fluorescence spectrum was measured with a fluorescence spectrophotometer (RF-6000, Shimadzu). The excitation and emission wavelengths were 315 and 425nm for 2-hydroxyterephthalic acid (TAOH), respectively.12

The reducing power of the photocatalytic material was evaluated from a qualitative evaluation of the generated reactive oxygen species (ROS) by using electron spin resonance (ESR). First, 5 mL of CH3OH and 5 mg of DMPO were added to a small container and stirred. Next, the substrate sample was immersed and irradiated with UV–vis light (300 ≤ λ ≤ 600 nm) for 15 min. Finally, the spectra of the ROS were analyzed by using ESR spectroscopy (E-500, Bruker).

A small cell with a quartz window at the top was attached to the original closed system, and a gas chromatograph (GC, GC-8A, Shimadzu) connected to the closed system was used to detect photocatalytic H2 production. First, 6.4 mL of pure water and 1.6 mL of methanol were added to the small cell, and the mixture was stirred. Next, a photocatalyst sample was added to the cell; the lid was closed, and the entire closed system was filled with nitrogen gas. Finally, the system was irradiated by UV–vis light (300 nm ≤ λ ≤ 600 nm), and the H2 production was measured every hour for 5 h.

3. Results and Discussion

3.1. Evaluation of Synthesized TiO2 Nanotube Arrays

Figure S1a shows an X-ray diffraction pattern (XRD) that indicates that the Ti substrate was oxidized to anatase TiO2 by anodic oxidation.13 Furthermore, the XRD patterns shown in Figures 1a and S1b, along with microscopy observations, demonstrate that the synthesized TiO2 forms NTA with an orientation perpendicular to the substrate with a diameter of about 80 nm and a depth of about 1.5 μm. From the Mott–Schottky (Figure S2a) and Tauc plots (Figure S2b), the CB and VB of the TiO2–NTA were estimated to be 0.1 and 3.4 V (vs normal hydrogen electrode (NHE)), respectively.14−17

3.2. Synthesis and Evaluation of g-C3N4 Small Nanoparticles by CVD Method and Their Support on TiO2–NTA

XPS spectra of g-C3N4 deposited on a TiO2–NTA substrate when the temperature in region 1 was set between 650 and 800 °C are shown in Figure 3a. The peak at 288.6 eV is the peak of carbon nitride derived from the bonding of carbon and nitrogen and appears strongly at temperatures above 750 °C.16,18,19 The peak at 284.6 eV is reported to be due to incomplete polymerization or C=C bonding due to defects, and the peak appeared strongly at 650 and 700 °C.16,18,20,21 The C/N ratio at 650 °C is C/N = 3:5.83, clearly indicating a large ratio of nitrogen. On the other hand, at 750 °C, the C/N ratio was close to the theoretical value, C/N = 3:4.07, and the peak at 284.6 eV was clearly smaller than at 700 °C and below.22 It is known that during the thermal polymerization process of melamine to carbon nitride, the end group is removed as ammonia and the content nitrogen decreases.23 Therefore, when the temperature in region 1 was set below 700 °C, the ratio of nitrogen is thought to have been larger because melamine did not thermally polymerize completely to g-C3N4. The g-C3N4 yield was highest when the temperature of region 1 was above 750 °C.

Figure 3.

Figure 3

Open in a new tab

(a) XPS spectra of g-C3N4 deposited on a TiO2–NTA substrate for different region 1 temperatures of the CVD apparatus. (b) XPS spectra of g-C3N4 deposited on a TiO2–NTA substrate at different holding times when the region 1 temperature of the CVD apparatus was set to 750 °C. (c) Model of the g-C3N4 structure optimized by quantum chemical calculations and simulated defect formation free energy changes for each carbon or nitrogen atom. (The calculation conditions are described in the Supporting Information). (d) XPS spectra of g-C3N4 deposited on TiO2–NTA substrates for different temperatures in region 2 of the CVD apparatus.

As can be seen in Figure 3b, when the temperature in region 1 was fixed at 750 °C and the holding time was changed from 30 min to 1 h, the peak intensity at 288.6 eV and N 1s clearly decreased, while the peak intensity at 284.6 eV increased. This is thought to be due to the longer holding time breaking unstable bonds between carbon and nitrogen atoms, leading nitrogen to desorb and the formation of a stable C=C bond.24,25 We have examined the relative defect formation energies in g-C3N4 for a variety of sites (with density functional theory (DFT)-based quantum chemical calculation) as shown in Figure 3c. These calculations showed that a defect at the N3 site had the lowest formation energy, supporting the above XPS discussion. Details of the calculation process are reported in the Supporting Information.

The XPS spectra for when region 1 was fixed at 750 °C with a holding time of 30 min while region 2 was held at 400 and 500 °C are shown in Figure 3d. When region 2 was 400 °C, a large peak indicating incomplete polymerization appeared at 284.6 eV. This result indicates that thermal polymerization progresses not only during transport but also after deposition on the substrate.

Based on these results, the optimal conditions for synthesizing g-C3N4 by CVD are with the temperatures in regions 1 and 2 set for 750 and 500 °C, respectively, with a holding time of 30 min.

From the XRD patterns shown in Figures 4a and S1a, it was confirmed that the crystalline phase of TiO2–NTA did not change after CVD. The reason why no peaks corresponding to loaded g-C3N4 were observed is thought to be due to the small amount of g-C3N4 loaded.

Figure 4.

Figure 4

Open in a new tab

(a) XRD spectrum of TiO2–NTA/g-C3N4. (b) XPS spectra of TiO2–NTA/g-C3N4 prepared under optimal conditions and g-C3N4 prepared by the simple heating of powder. (c) FT-IR spectra of melamine, g-C3N4 powder, TiO2–NTA, and TiO2–NTA/g-C3N4, and a magnified view of the 1200–1350 nm–1 wavenumber region.

XPS spectra comparing g-C3N4 synthesized by CVD and simple g-C3N4 powder synthesized by the thermal polymerization of melamine are shown in Figure 4b. It can be seen that g-C3N4 synthesized by CVD and g-C3N4 formed by the heating of powder lead to similar shapes.

Figure 4c shows the FT-IR spectra of melamine, g-C3N4 powder, TiO2–NTA, and TiO2–NTA/g-C3N4. Compared to melamine, g-C3N4 powder shows a characteristic peak at 1200–1400 cm–1, which originates from aromatic C–N groups.26 The enlarged FT-IR spectrum shows that TiO2–NTA has no peaks between 1200 and 1400 cm–1, while TiO2–NTA/g-C3N4 has a small peak therein. This result confirms that g-C3N4 is slightly loaded in TiO2–NTA.

3.3. Surface Morphology and Band Structure of TiO2–NTA/g-C3N4

By analysis of SEM images, we compared the grain size of g-C3N4 deposited on TiO2–NTA/g-C3N4 (2.0 g) and TiO2–NTA/g-C3N4 (0.2 g). As shown in Figure 5a, TiO2–NTA/g-C3N4 (2.0 g) had 30–100 nm of g-C3N4 deposited on TiO2–NTA. On the other hand, g-C3N4 was not observed on TiO2–NTA in TiO2–NTA/g-C3N4 (0.2 g), as shown in Figure 5b. This result contradicts the results of Figures 3 and 4 and implies that g-C3N4 of TiO2–NTA/g-C3N4 (0.2 g) is supported on TiO2–NTA as small nanoparticles so that it cannot be observed in the SEM image. Therefore, a more detailed image was obtained by using TEM (Figure 5c). The results show that TiO2–NTA/g-C3N4 (0.2 g) has evenly spaced g-C3N4 deposits with a uniform size of about 10 nm on TiO2–NTA. It was also confirmed that the lattice spacing of this g-C3N4 was 0.34 nm, indicating a 002 peak, which is an interphase stacking of g-C3N4 (Figure S3).27 A large amount of melamine was placed in a crucible for the growth of the TiO2–NTA/g-C3N4 (2.0 g) sample. g-C3N4 growth of 30 to 100 nm was observed to be nonuniform due to the high molecular density, which facilitated thermal polymerization and aggregation. On the other hand, the TiO2–NTA/g-C3N4 (0.2 g) sample, which used a smaller amount of melamine, has a lower molecular density, making polymerization and aggregation less likely to occur and leads to deposition on the TiO2–NTA in the form of uniform small nanoparticles of 10 nm in size. A schematic diagram of the TiO2–NTA/g-C3N4 (0.2 g) growth process is shown in Figure 5d.

Figure 5.

Figure 5

Open in a new tab

(a) SEM image of TiO2–NTA/g-C3N4 (2.0 g). (b) The SEM image of TiO2–NTA/g-C3N4 (0.2 g). (c) The TEM image of TiO2–NTA/g-C3N4 (0.2 g). (d) The schematic diagram of TiO2–NTA/g-C3N4 (0.2 g) grown by our CVD apparatus. (e) Relative band alignments of TiO2–NTA and g-C3N4.

For electrochemical measurements, g-C3N4 was synthesized on indium tin oxide coated substrates (ITO) by placing them in region 2. From Mott–Schottky (Figure S4a) and Tauc plots (Figure S4b) of g-C3N4 loaded on ITO, the CB and VB of g-C3N4 were estimated to be −0.8 and 1.9 V (vs NHE), respectively.14−17 The band diagrams of TiO2–NTA and g-C3N4 that are estimated from Figures S2 and S4 are shown in Figure 5e.

3.4. Evaluation of the Photocatalytic Redox Power of TiO2–NTA/g-C3N4 by PL and ESR Measurements

Figure 6a shows the results of a series of PL measurements, where the amount of g-C3N4 deposited in TiO2–NTA was varied. It can be seen that TiO2–NTA/g-C3N4(0.2 g) produced the largest number of OH radicals. Figure 6b shows ESR results for TiO2–NTA and TiO2–NTA/g-C3N4 (0.2 g).28 No ROS peak was detected for TiO2–NTA, whereas a ROS peak was detected for TiO2–NTA/g-C3N4 (0.2 g). The band diagram in Figure 5e shows that TiO2–NTA satisfies the formation potential of OH radicals but not that of reactive oxygen species, while g-C3N4 satisfies the formation potential of reactive oxygen species but not that of OH radicals.29,30 Therefore, taking into account the results of the PL and ESR measurements, it is concluded that TiO2–NTA/g-C3N4 (0.2 g) forms a Z-scheme reaction, meaning that efficient charge separation occurred.

Figure 6.

Figure 6

Open in a new tab

(a) PL measurements of TiO2–NTA/g-C3N4 for varying amounts of melamine placed in the crucible. (b) ESR measurements of TiO2–NTA/g-C3N4 (0.2 g).

3.5. Photocatalytic Hydrogen Production by TiO2–NTA/g-C3N4 (0.2 g)

A comparison of the photocatalytic hydrogen production from methanol aqueous solutions (20%) of TiO2–NTA, g-C3N4 (0.2 g), and TiO2–NTA/g-C3N4 (0.2 g) is shown in Figure 7a. TiO2–NTA/g-C3N4 (0.2 g) generated more H2 than did TiO2–NTA and g-C3N4 (0.2 g). The band structure of TiO2–NTA does not satisfy the redox potential for generating hydrogen from water, but the reason why it generates a small amount of hydrogen is believed to be due to the band-filling effect.31 In addition, the results of a cycling experiment in Figure 7b show that TiO2–NTA/g-C3N4 (0.2 g) produced more H2 in the second cycle than in the first, while the amount of H2 produced in the second and third cycles remained unchanged. On the other hand, there was no significant change in the amount of hydrogen produced by TiO2–NTA in the three cycle experiments (Figure S5). Considering that g-C3N4 is known as a material that easily adsorbs oxygen and the XPS results in Figure 7c, the first reduction reaction is thought to be a simultaneous proton reduction and adsorbed oxygen reduction.32,33 After a certain period of time elapses, it is speculated that the adsorbed oxygen desorbs and the excited electrons of g-C3N4 are active only for proton reduction, suggesting that the photocatalytic hydrogen production in the second reaction was higher than that in the first. The graph in Figure 7b, which shows a change from a nonlinear to a linear increase, confirms the validity of this premise. Previous research has also shown that the initial stage of photocatalytic hydrogen production by g-C3N4 tends to have a nonlinear increase in the amount of hydrogen produced.16 The fact that the results of photocatalytic hydrogen production did not change between the second and third cycles indicates that the adsorbed oxygen was lost and that catalytic degradation did not occur. The reason why the hydrogen production of TiO2–NTA alone is linear is thought to be because of the absence of impurities on the TiO2 surface. The oxygen peak in the XPS O 1s spectra of TiO2–NTA measured before and after the cycling experiment (Figure S6) also did not change. Since the O 1s signals in Figure S6 are due to oxygen derived from TiO2, this cycling experiment suggests that TiO2–NTA does not deteriorate even after long-term use. Finally, we confirmed that this reaction was proceeded by photocatalysis (Figure S7).

Figure 7.

Figure 7

Open in a new tab

(a) Photocatalytic hydrogen production of TiO2–NTA/g-C3N4 (0.2 g), TiO2–NTA, and g-C3N4 (0.2 g). (b) Photocatalytic hydrogen production cycling experiment with TiO2–NTA/g-C3N4 (0.2 g). (c) Change in the XPS O 1s spectra of TiO2–NTA/g-C3N4 (0.2 g) during a photocatalytic hydrogen production cycling experiment.

4. Conclusions

This paper demonstrates that it is possible to uniformly support g-C3N4 small nanoparticles on nanotube TiO2 by adjusting the temperature and heating time in a CVD apparatus operating at reduced pressure. Furthermore, it was found that TiO2–NTA loaded with g-C3N4 was more efficient at various photocatalytic properties than TiO2–NTA and g-C3N4 alone. This indicates that g-C3N4 small nanoparticles prepared from melamine under CVD conditions function as a semiconductor photocatalyst similar to g-C3N4 powder prepared by powder heating.

These findings are expected to lead to the development and use of similar CVD apparatuses not only in the field of photocatalysis but also in other fields requiring the support of small nanoparticles.

Acknowledgments

This study was supported by a Grant-in-Aid for Scientific Research (KAKENHI No. 19H02174) of the Japan Society for the Promotion of Science (JSPS). K.I. is very grateful for the support from JST SPRING Grant Number JPMJSP2123.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmaterialsau.4c00084.

  • XRD; SEM; Mott–Schottky plot; Tauc plot; TEM; photocatalytic hydrogen production cycling experiment; XPS; and control experiments for photocatalytic H2 generation (PDF)

Author Contributions

K.I.: conceptualization, methodology, investigation, writing—original draft, data curation, and writing—review and editing. S.Y.: investigation and data curation. S.Y.: investigation and data curation. P.F.: methodology and writing—review and editing. K.N.: conceptualization, methodology, writing—review and editing, supervision, and project administration.

The authors declare no competing financial interest.

Supplementary Material

mg4c00084_si_001.pdf (1.2MB, pdf)

References

  1. Fujishima A.; Honda K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972, 238 (5358), 37–38. 10.1038/238037a0. [DOI] [PubMed] [Google Scholar]
  2. Ikreedeegh R. R.; Hossen M. A.; Tahir M.; Aziz A. A. A comprehensive review on anodic TiO2 nanotube arrays (TNTAs) and their composite photocatalysts for environmental and energy applications: Fundamentals, recent advances and applications. Coord. Chem. Rev. 2024, 499, 215495 10.1016/j.ccr.2023.215495. [DOI] [Google Scholar]
  3. Pandey B.; Rani S.; Roy S. C. A scalable approach for functionalization of TiO2 nanotube arrays with g-C3N4 for enhanced photo-electrochemical performance. J. Alloys Compd. 2020, 846, 155881 10.1016/j.jallcom.2020.155881. [DOI] [Google Scholar]
  4. Pourhashem S.; Duana J.; Guana F.; Wanga N.; Gaoa Y.; Hou B. New effects of TiO2 nanotube/g-C3N4 hybrids on the corrosion protection performance of epoxy coatings. J. Mol. Liq. 2020, 317, 114214 10.1016/j.molliq.2020.114214. [DOI] [Google Scholar]
  5. Liu G.; Zheng F.; Li J.; Zeng G.; Ye Y.; Larson D. M.; Yano J.; Crumlin E. J.; Ager J. W.; Wang L. W.; Toma F. M. Investigation and mitigation of degradation mechanisms in Cu2O photoelectrodes for CO2 reduction to ethylene. Nat. Energy 2021, 6, 1124–1132. 10.1038/s41560-021-00927-1. [DOI] [Google Scholar]
  6. Wang F.; Hou T.; Zhao X.; Yao W.; Fang R.; Shen K.; Li Y. Ordered macroporous carbonous frameworks implanted with CdS quantum dots for efficient photocatalytic CO2 reduction. Adv. Mater. 2021, 33, 2102690 10.1002/adma.202102690. [DOI] [PubMed] [Google Scholar]
  7. Wang S.; Zhan J.; Chen K.; Ali A.; Zeng L.; Zhao H.; Hu W.; Zhu L.; Xu X. Potassium-doped g-C3N4 achieving efficient visible-light-driven CO2 reduction. ACS Sustainable Chem. Eng. 2020, 8, 8214–8222. 10.1021/acssuschemeng.0c01151. [DOI] [Google Scholar]
  8. Tan M.; Ma Y.; Yu C.; Luan Q.; Li J.; Liu C.; Dong W.; Su Y.; Qiao L.; Gao L.; Lu Q.; Bai Y. Boosting photocatalytic hydrogen production via interfacial engineering on 2D ultrathin Z-scheme ZnIn2S4 /g-C3N4 heterojunction. Adv. Funct. Mater. 2022, 32, 2111740 10.1002/adfm.202111740. [DOI] [Google Scholar]
  9. Li Y.; Ren Z.; He Z.; Ouyang P.; Duan Y.; Zhang W.; Lv K.; Dong F. Crystallinity-defect matching relationship of g-C3N4: Experimental and theoretical perspectives. Green Energy Environ. 2024, 9, 623–658. 10.1016/j.gee.2023.02.012. [DOI] [Google Scholar]
  10. Gao Z.-D.; Qu Y.-F.; Zhou X.; Wang L.; Song Y.-Y.; Schmuki P. Pt-Decorated g-C3N4/TiO2 Nanotube Arrays with Enhanced Visible-Light Photocatalytic Activity for H2 Evolution. ChemistryOpen 2016, 5, 197–200. 10.1002/open.201500219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Goto H.; Masegi H.; Sadale S. B.; Noda K. Intricate behaviors of gas phase CO2 photoreduction in high vacuum using Cu2O-loaded TiO2 nanotube arrays. J. CO2 Util. 2022, 59, 101964 10.1016/j.jcou.2022.101964. [DOI] [Google Scholar]
  12. Sotiles A. R.; Barbosab L. V.; Vedovatob Y. P.; Wypycha F.; Fariab E. H. Effect of Zn2Cr layered double hydroxide on photocatalytic hydroxylation of terephthalic acid and photodegradation of sodium diclofenac. Catal. Today 2023, 423, 114287 10.1016/j.cattod.2023.114287. [DOI] [Google Scholar]
  13. Sreekantan S.; Hazan R.; Lockman Z. Photoactivity of anatase–rutile TiO2 nanotubes formed by anodization method. Thin Solid Films 2009, 518, 16–21. 10.1016/j.tsf.2009.06.002. [DOI] [Google Scholar]
  14. Ito K.; Uchida R.; Noda K. Visible light-assisted hydrogen generation over platinum-loaded tungsten trioxide nanorods with the hexagonal and triclinic phases. J. Photochem. Photobiol. A 2023, 443, 114824 10.1016/j.jphotochem.2023.114824. [DOI] [Google Scholar]
  15. Zos’ko N. A.; Aleksandrovsky A. S.; Kenova T. A.; Gerasimova M. A.; Nikolay G.; Maksimov N. G.; Taran O. P. Cyclic voltammetry as an activation method of TiO2 nanotube arrays for improvement of photoelectrochemical water splitting performance. ChemPhotoChem 2023, 7, e202300100 10.1002/cptc.202300100. [DOI] [Google Scholar]
  16. Ito K.; Noda K. Highly efficient hydrogen production and selective CO2 reduction by the C3N5 photocatalyst using only visible light. Phys. Chem. Chem. Phys. 2024, 26, 153–160. 10.1039/D3CP04431A. [DOI] [PubMed] [Google Scholar]
  17. Montakhab E.; Rashchi F.; Sheibani S. Enhanced photocatalytic activity of TiO2 nanotubes decorated with Ag nanoparticles by simultaneous electrochemical deposition and reduction processes. Appl. Surf. Sci. 2023, 615, 156332 10.1016/j.apsusc.2023.156332. [DOI] [Google Scholar]
  18. Liu Y.; Gaob Y.; Chena L.; Lia L.; Dinga D.; Daia Z. MnOx-decorated oxygen-doped g-C3N4 with enhanced photocatalytic activity for efficient removal of uranium(VI). Sep. Purif. Technol. 2023, 307, 122794 10.1016/j.seppur.2022.122794. [DOI] [Google Scholar]
  19. Xu Z.; Chenb Y.; Wanga B.; Ranb Y.; Zhongb J.; Li M. Highly selective photocatalytic CO2 reduction and hydrogen evolution facilitated by oxidation induced nitrogen vacancies on g-C3N4. J. Colloid Interface Sci. 2023, 651, 645–658. 10.1016/j.jcis.2023.08.012. [DOI] [PubMed] [Google Scholar]
  20. Ding Y.; Tang Y.; Yang L.; Zeng Y.; Yuan J.; Liu T.; Zhang S.; Liu C.; Luo S. Porous nitrogen-rich carbon materials from carbon self-repairing g-C3N4 assembled with graphene for high-performance supercapacitor. J. Mater. Chem. A 2016, 4, 14307–14315. 10.1039/C6TA05267C. [DOI] [Google Scholar]
  21. Feng P.; Cuia K.; Haib Z.; Wang J.; Wangb L. Facile synthesis of activated carbon loaded g-C3N4 composite with enhanced photocatalytic performance under visible light. Diamond Relat. Mater. 2023, 136, 109921 10.1016/j.diamond.2023.109921. [DOI] [Google Scholar]
  22. Ma X.; Lv Y.; Xu J.; Liu Y.; Zhang R.; Zhu Y. A strategy of enhancing the photoactivity of g-C3N4 via doping of nonmetal elements: A first-principles study. J. Phys. Chem. C 2012, 116, 23485–23493. 10.1021/jp308334x. [DOI] [Google Scholar]
  23. Wang X.; Maeda K.; Thomas A.; Takanabe K.; Xin G.; Carlsson J. M.; Domen K.; Antonietti M. A metal-free polymeric photocatalyst for hydrogen production from water under visible light. Nat. Mater. 2009, 8, 70–80. 10.1038/nmat2317. [DOI] [PubMed] [Google Scholar]
  24. Liao J.; Cui W.; Li J.; Sheng J.; Wang H.; Dong X.; Chen P.; Jiang G.; Wang Z.; Dong F. Nitrogen defect structure and NO+ intermediate promoted photocatalytic NO removal on H2 treated g-C3N4. Chem. Eng. J. 2020, 379, 122282 10.1016/j.cej.2019.122282. [DOI] [Google Scholar]
  25. Dong G.; Wen Y.; Fan H.; Wang C.; Cheng Z.; Zhang M.; Ma J.; Zhang S. Graphitic carbon nitride with thermally-induced nitrogen defects: An efficient process to enhance photocatalytic H2 production performance. RSC Adv. 2020, 10, 18632–18638. 10.1039/D0RA01425G. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Pan Y.; Liu X.; Zhang W.; Shao B.; Liu Z.; Liang Q.; Wu T.; He Q.; Huang J.; Peng Z.; Liu Y.; Zhao C. Bifunctional template-mediated synthesis of porous ordered g-C3N4 decorated with potassium and cyano groups for effective photocatalytic H2O2 evolution from dual-electron O2 reduction. Chem. Eng. J. 2022, 427, 132032 10.1016/j.cej.2021.132032. [DOI] [Google Scholar]
  27. Ge L. Synthesis and photocatalytic performance of novel metal-free g-C3N4 photocatalysts. Mater. Lett. 2011, 65, 2652–2654. 10.1016/j.matlet.2011.05.069. [DOI] [Google Scholar]
  28. Yu Q.; Yang K.; Li H.; Li X. Z-scheme α-Fe2O3/g-C3N4 with the Fe–OC– bond toward enhanced photocatalytic degradation. Colloids Surf., A 2021, 616, 126269 10.1016/j.colsurfa.2021.126269. [DOI] [Google Scholar]
  29. Muñoz E.; Marotti R.; Navarrete E. Communication-hydroxyl radicals attack CdTe quantum dots. J. Electrochem. Soc. 2021, 168, 097503 10.1149/1945-7111/ac2282. [DOI] [Google Scholar]
  30. Zhan H.; Zhou Q.; Li M.; Zhou R.; Mao Y.; Wang P. Photocatalytic O2 activation and reactive oxygen species evolution by surface B-N bond for organic pollutants degradation. Appl. Catal., B 2022, 310, 121329 10.1016/j.apcatb.2022.121329. [DOI] [Google Scholar]
  31. Diwald O.; Thompson T. L.; Goralski E. G.; Walck S. D.; Yates J. T. Jr The effect of nitrogen ion implantation on the photoactivity of TiO2 rutile single crystals. J. Phys. Chem. B 2004, 108, 52–57. 10.1021/jp030529t. [DOI] [Google Scholar]
  32. Wang X.; Zhang X.; Wei Y.; Gao C.; Bao J.; Zhang N. In situ grown Co3O4 nanosheets in the interlayer space of g-C3N4 for efficient removal of Hg0 from flue gas. Fuel 2022, 324, 124660 10.1016/j.fuel.2022.124660. [DOI] [Google Scholar]
  33. Shang D.; Zhong Q.; Cai W. High performance of NO oxidation over Ce-Co-Ti catalyst: The interaction between Ce and Co. Appl. Surf. Sci. 2015, 325, 211–216. 10.1016/j.apsusc.2014.11.056. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

mg4c00084_si_001.pdf (1.2MB, pdf)

Articles from ACS Materials Au are provided here courtesy of American Chemical Society

RESOURCES