Photocatalytic Activity of TiNbC-Modified TiO₂ during Hydrogen Evolution and CO₂ Reduction

Abstract

Photocatalytic CO₂ reduction and the production of hydrogen are urgent tasks of green energy. One of the most studied semiconductor photocatalysts for this purpose is titanium dioxide. However, it has a number of fundamental limitations that do not allow its application for such tasks on an industrial scale. Another class of promising materials, which is being investigated very actively, are two-dimensional materials based on MXenes. In this work, we present the first results on photocatalytic hydrogen evolution and CO₂ reduction using TiNbC/TiO₂ heterostructures with TiNbC contents of 1, 5, and 10%. The approach to the creation of heterostructures proposed in this work may become a significant breakthrough in the search for new highly efficient systems for carbon dioxide reduction and hydrogen production.

1. Introduction

The environment and energy are critical to human existence and survival. Over the past few decades, the rapid growth of fossil fuel energy consumption has not only maximized the impact on the environment but has also caused a devastating crisis in the energy production cycle, which is an extreme concern for the flourishing human race. The search for alternative, sustainable, and clean sources of energy that can replace dwindling supplies in non-renewable energy sources is an important and high priority for humanity for the foreseeable future [1]. The global market for renewable energy is expected to continue to grow in the coming years. According to Group Next Move Strategy Consulting, the market will exceed two trillion dollars by 2030. Environmental concerns related to fossil fuels, rapid urbanization, and economic growth in developing regions are major contributors to the projected market growth. The use of clean, abundant, and inexhaustible solar energy using photocatalytic reactions [2] caused by sunlight, the splitting and photofixation of nitrogen (N₂) [3], reductions in carbon dioxide (CO₂) emissions [4], and the decomposition of pollutants [5] is considered a feasible strategy to meet the ever increasing energy needs of people and remedy the environmental crisis. Absorption of sunlight, production of photogenerated charge carriers, separation and transfer of photogenerated charges, and interfacial charge transfers/reactions are a sequence of key events that occur in any photocatalytic system. Improving the efficiency of any of these events would ultimately increase the overall performance of the photocatalytic reaction or lead to efficient utilization of solar energy. Therefore, a great deal of research has been devoted to preparing photocatalytic materials with a broader spectrum of sunlight absorption; multiple active centers on the surface of the photocatalytic material; and effective charge separation capability in the volume and on the surface of the photocatalytic material, using various strategies and designs to maximize the use of solar energy.

Traditional semiconductor photocatalysts, including TiO₂, g-C₃N₄, and CdS, usually have drawbacks such as the inability to absorb visible and near-infrared light, solar illumination, and rapid recombination of photogenerated charge carriers [6]. Heteroatomic doping using co-catalysts and the designs of heterostructures are accepted as possible strategies to overcome the drawbacks associated with traditional single-photocatalyst materials. Co-catalyst materials, which have many traps for photogenerated charge carriers and have excellent electrical conductivities, lead to efficient charge separations in combination with photocatalytic materials. In addition, the presences of co-catalysts on the surfaces of semiconductor photocatalysts reduce activation energy, which contributes to the acceleration of reactions.

There are a number of requirements for confident photocatalytic reactions, such as the presence of abundant surface functional groups to form sufficient chemical bonds with the photocatalyst—a necessary condition for the effective extraction of charge carriers from a photocatalyst to thereby slow down the recombination kinetics in the photocatalyst; good electrical conductivity of a photocatalyst favors easy charge migration in its volume, which promotes efficient photocatalytic reactions; the presence of hydrophilic surface functional groups contribute to their interactions with water, which lead to good photocatalytic efficiencies; and good chemical stabilities and/or the need for non-aqueous media. In addition, the size and shape of the co-catalyst intend to demonstrate its influence on the photocatalytic reaction spectra. A new class of two-dimensional materials, known as Mxenes, meets all these requirements. MXenes typically include atomic layers of transition metals as their upper and lower layers; these metal sites can give a strong redox potential.

Based on the photocatalytic reaction, H₂ production and CO₂ reduction technology is carried out, which are promising for fuel and energy production because they rely exclusively on natural solar energy to trigger the reaction. In addition, the abundance of available CO₂ and H₂O acts as a cheap feedstock for photocatalytic reactions. Consequently, CO₂ reduction through solar energy and water splitting is a promising approach to obtaining renewable fuels because of its cheapness and environmental friendliness.

Water splitting with release of H₂ and reduction of CO₂ is based on photocatalysis [7,8,9,10,11,12,13,14,15,16,17,18,19,20,21] and electrocatalysis [22,23,24,25,26,27,28,29,30]. But, the greenest technology for hydrogen production and carbon dioxide reduction is the photocatalytic reaction. Most works on photocatalytic H₂ production and CO₂ reduction are devoted to model MXene Ti₃C₂T_x in combination with various semiconductors: TiO₂, ZnS, CdS, MoS₂, g-C₃N₄, etc. [7,8,9,10,11,12,13,14,15,16]. Several works are devoted to Nb₂C [17,18]. It was found that a Nb₂C sample obtained by chemical etching and 4 h ultrasound treatment (Nb₂C-4h), with a high specific surface area in an aqueous suspension of 161 m², measured by methylene blue, showed the highest rate of hydrogen formation of 290 μmol/h*g, which was 2–3 times higher than that of Nb₂C obtained by chemical etching without post-synthetic ultrasound treatment, which has a 55 m² specific surface in suspension. An important property of 2D Zr₂CO₂ and Hf₂CO₂ is that they have unexpectedly high and directionally anisotropic carrier mobilities, which can effectively promote the migrations and separations of photogenerated electron–hole pairs [19]. In [20], it was shown that the rich chemical composition of transition metal dichalcogenides (TMD), combined with numerous strategies to tune their electronic properties, make these materials very attractive for understanding the fundamental principles of electro- and photocatalysis and for developing highly efficient, renewable, and affordable catalysts for large-scale H₂ production [20].

Recently, scientists around the world have been intensively searching for new photocatalysts operating in the visible range of the spectrum because such photocatalysts activate photocatalytic processes for splitting water with the release of hydrogen, as well as the reduction of carbon dioxide. These processes are related to green energy and are promising research areas for mankind due to the new vector in the energy industry—environmental preservation and the increasing use of renewable energy sources. Dozens of well-established photocatalyst semiconductors are already known.

Solar energy can be directly turned to thermal energy by the photothermal effect or converted to fuel and stored for later exploitation. Also, sun irradiation can be used directly for the purpose of water depollution. Historically, titanium dioxide (TiO₂) and zinc oxide (ZnO) are the materials that really sparked worldwide interest in oxide semiconductors for the above applications. Their main advantages, compared to other types of semiconducting materials, is that they are plentiful in nature and considered environmentally benign. Furthermore, TiO₂ has many useful properties, such as ease of process ability, chemical stability, high surface area, low cost, non-toxicity, and excellent charge migration qualities. Years of research progress have actually allowed the theoretical thermodynamics limits to be approached in aforementioned applications. However, use of solar energy is still not widespread.

Semiconductors have wide bandgaps, and they are active only in the UV region. Sun radiation spans from X-rays to radio waves, with the most intense solar radiation occurring in the visible light range, such that 43% of the solar energy reaching the Earth’s surface is at the wavelengths from 400 to 700 nm. For example, if only UV radiation would be utilized by the photocatalyst-in-water-splitting reaction, solar to hydrogen conversion efficiency would be only 2%, even if a quantum efficiency of 100% would be assumed, and entropic losses would be neglected. It is generally recognized that two common bottlenecks often exist in applications of semiconducting materials in solar utilization schemes—limited light absorption and rapid charge recombination. Although many different strategies have been used (such as defect engineering or doping) in order to increase the efficiencies of semiconductors for solar energy harvesting, the efficiency-to-cost ratio has yet to meet the values needed to match traditional fossil fuel sources and achieve widespread commercial acceptance.

In order to overcome this limitation, we will utilize double-transition metal MXene (D-MXenes). MXenes are a class of plasmonic 2D materials that hold a great promise for the conversion of sun energy, owing to diverse elemental compositions, unique 2D structures, large surface areas, abundant surface terminations, and excellent photoelectronic properties. The recent introduction of double-transition metal MXene (D-MXenes) further increases the number of possible MXene structures and possibilities of tuning their properties to suit the specific application. Depending on the chemical composition and fabrication method, these materials exhibit large absorptions over a broad bandwidth (∼1.55 μm), covering a significant visible-to-near-IR spectral window, tunable plasmon resonances, and adjustable work functions in a large range from below 2.14 eV (the lowest work function in all metals, of cesium) to over 5.65 eV (the highest work function in all metals, of platinum).

All works on photocatalytic and electrocatalytic H₂ productions were performed with or based on MXenes derived from three-component MAX-phases. Recently, scientists have paid attention to MAX-phases synthesized from four-component compositions [31,32,33,34], which also exhibit various bright properties due to the combination of different metals and the features of embedding in the crystal lattices. The aim of this work is to study the parameters of photocatalytic reactions of hydrogen evolution as well as CO₂ reduction using the TiNbC/TiO₂ compound as a photocatalyst.

2. Materials and Methods

2.1. Obtaining of the TiNbC/TiO₂ Heterostructures

TiNbC MXenes were purchased from the Advanced 2D Materials Co. Ltd. store, Wuzhong District Industrial Park, Suzhou City, Jiangsu Province, China, with claimed material purities of 99% and particle sizes 1–5 μm. Photocatalysts were prepared from TiO₂ Evonik P25 (anatase 80%, rutile 20%) and MXene with weight contents of 1, 5, and 10% for MXene. Acetone (50 mL) was added to a mixture of TiO₂ and MXene, then the suspension was kept on a water bath at 80 °C until the acetone completely evaporated (approximately 60 min), the precipitate was dried at 60 °C for 12 h, and then calcined at 400 °C and 500 °C for 1 h at a heating rate of 3–4 °C/min.

The samples obtained according to the described procedure will be hereinafter referred to as 1% TiNbC/TiO₂, 5% TiNbC/TiO₂, and 10% TiNbC/TiO₂.

2.2. Characterization

XRD investigations were performed with a D8 Advance powder diffractometer (Bruker, Karlsruhe, Germany) using CuKα radiation. XRD-patterns were recorded in the 2θ range from 20° to 80°. Morphologies and compositions of the materials were characterized by a scanning electron microscopy (SEM) system (MAIA 3, Tescan, Brno-Kohoutovice, Czech Republic) coupled with an energy dispersive spectrometry (EDS) detector (X-act, Oxford Instruments, High Wycombe, UK). Electron images and EDS spectra were obtained at a 25 keV accelerating voltage. Samples for SEM were prepared by dropping 1 μL of the material solution onto a cleaned crystalline silicon substrate with subsequent drying at ambient conditions. Images of TiNbC/TiO₂ particles were obtained using high-resolution Transmission Electron Microscopy (TEM, Jeol JEM-2100 microscope, Tokyo, Japan) equipped with 200 kV field emission gun and point resolution of 0.19 nm. Transmission Electron Microscopy equipped with an analytical attachment for energy dispersive analysis Aztech X-Max 100. The optical extinction spectra of colloidal NPs were measured using an ultraviolet-visible spectrophotometer (Cary 5000; Agilent Technologies, Santa Clara, CA, USA) in a 200 to 1000 nm (1.24 to 6.2 eV) spectral interval with a spectral resolution of 1 nm using 2 mm optical path length cuvettes. Samples were prepared by mixing TiNbC/TiO₂ micropowder with deionized water (MilliQ, 18.2 MΩ*cm) followed by sonicating in an ultrasonic bath.

2.3. Catalytic Activity Measurements

The catalytic activities of the investigated heterostructures were evaluated in two ways: by the rate of hydrogen production in an aqueous suspension of ethyl alcohol and by the rate of carbon dioxide reduction in the gas phase.

2.3.1. Photocatalytic Hydrogen Evolution

The photocatalytic activities of the obtained heterostructures were measured in a static reactor in the photocatalytic hydrogen evolution from an aqueous ethanol solution (Figure 1a). A suspension consisting of 10 mL of C₂H₅OH, 90 mL of H₂O, and 50 mg of photocatalyst was placed in the reactor. Before the start of the photocatalytic experiment, the reactor was purged with Ar for 30 min to remove O₂. The activity was measured for 1.5 h under UV irradiation with a wavelength of 380 nm (Figure 1b), sampling every 15 min. The evolved hydrogen was analyzed on a Khromos GH-1000 gas chromatograph (Russia) using Ar as a carrier gas.

The experimental setup (Figure 1a) was a glass reactor with samplers, inside which the reaction suspension was stirred using a magnetic stirrer. The reactor lid contained a quartz window, through which the photocatalyst was illuminated by an LED with a wavelength of 380 nm (Figure 1b).

2.3.2. Photocatalytic Reduction of Carbon Dioxide

The activities of photocatalysts were tested in a CO₂ reduction reaction in a batch reactor (170 mL). A photocatalyst (30 mg) was deposited on a glass support and placed in the reactor. After that, the reactor containing 1 mL of water was purged with CO₂ for 1 h, and the light source was turned on. The details were described elsewhere [35]. The gas phase composition was determined as in the previous case and using a gas chromatograph “GH-1000” (Chromos, Russia).

On the basis of the data obtained, the rates of formations of products were calculated with the total reduction rates of CO₂ (W(CO₂RR)). The rates of the formations of products (CH₄, CO, and H₂) were determined by a linear approximation of the kinetic curves. The total reduction rate of CO₂ was calculated using the following formula:

$$\mathrm{W}\left({\mathrm{C}\mathrm{O}}_2\mathrm{R}\mathrm{R}\right)=8\times \mathrm{W}\left({\mathrm{C}\mathrm{H}}_4\right)+2\times \mathrm{W}\left(\mathrm{C}\mathrm{O}\right),$$

where 2 and 8 are coefficients to account for the electron balance, and W(CH₄) and W(CO) are the rates of formation of methane and CO, respectively.

The selectivity of the photocatalytic CO₂ reduction process was calculated using the following formula:

$$\mathrm{S}\left({\mathrm{C}\mathrm{O}}_2\mathrm{R}\mathrm{R}\right)=\frac{\mathrm{W}\left({\mathrm{C}\mathrm{O}}_2\mathrm{R}\mathrm{R}\right)}{\mathrm{W}\left({\mathrm{C}\mathrm{O}}_2\mathrm{R}\mathrm{R}\right)+2\times \mathrm{W}\left({\mathrm{H}}_2\right)}\times 100\mathrm{\%}.$$

3. Results and Discussion

3.1. X-ray Diffraction

X-ray diffraction patterns of the original MXenes and the titanium dioxide used to produce the heterostructures, as well as the XRDs of the synthesized heterostructures, are shown in Figure 2. The X-ray diffraction patterns of titanium dioxide contained characteristic peaks of rutile and anatase. From the ratio of the intensities of the characteristic reflexes of their widths, it can be concluded that the used titanium dioxide consists of 86% of anatase, with a crystallite size of 18 nm, and 16% of rutile, with a crystallite size of 24 nm. X-ray diffraction patterns of the heterostructures show characteristic peaks of rutile and anatase, as well as the main peak of MXene. At the same time, there are no peaks of compounds such as titanium carbonate or niobium oxide or carbonate, from which we can conclude that the chosen strategy of obtaining the heterostructures does not lead to a change in the structure of MXene itself.

3.2. SEM and TEM Characterizations

Figure 3 shows a SEM image of the 5% TiNbC/TiO₂ heterostructure. It can be seen that most of the particles correspond to the manufacturer’s declared sizes of 1–5 μm (Figure 3). EDS analysis confirms the elemental composition of the heterostructure for all heterostructure concentrations (Figure S1). The element distribution map on a typical 5% TiNbC/TiO₂ particle shows a uniform distribution of the elements Ti, Nb, O, and C (Figure S2). The ratio of the elements was measured by the EDX method. As can be seen from

Table 1. TiNbC/TiO₂ heterostructures compound ratios.

Sample	Ti, %	Nb, %	C, %	O, %
1% TiNbC/TiO2	36.1 ± 0.4	1.0 ± 0.2	1.2 ± 0.2	62.9 ± 0.9
5% TiNbC/TiO2	39.4 ± 0.5	3.1 ± 0.3	4.1 ± 0.3	57.6 ± 0.9
10% TiNbC/TiO2	39.0 ± 0.3	4.5 ± 0.1	8.2 ± 0.4	57.5 ± 0.4

, the contents of 1, 5, and 10% MXene satisfied the reported compositions of heterostructures. The TiNbC/TiO₂ heterostructures compound ratios are shown in Table 1. All MXenes were completely covered by TiO₂ nanoparticles (Figure S3).

Figure 4 shows the structure of the obtained titanium dioxide and the MXene 5% TiNbC heterostructures obtained by transmission electron microscopy.

The image of a typical particle (Figure 4, left) clearly shows hexagonal nanoparticles covering a large particle with no clear facets. Comparison of the EDX maps of the distribution of individual elements (Figure 4, right) allows us to conclude that the nanoparticles with sharp edges are titanium dioxide in the presented image, and the larger particles without sharp edges are MXene TiNbC. This fully corresponds to the granulometric characteristics of the components declared by the MXene manufacturer (see Section 2.1) and obtained by XRD. The electron diffraction on the particle in question forms a system of concentric rings and individual point reflexes. The analysis of their radii (inset in Figure 4) shows that the diffraction is caused mainly by anatase, which also correlates with the powder X-ray diffraction data. It is also possible to specify the parameters of the unit cell [36].

Figure 5 shows the extinction spectrum of the heterostructures with different concentrations of TiNbC.

The values of the band gap width for TiNbC/TiO₂ heterostructures were determined from the absorption spectra (Figure 6). The obtained band gap values were E_g = 3.43 eV for the 1% of TiNbC and 3.33 eV for the 5% and 10% of TiNbC. The obtained values were quite close to the bandgap width of pure titanium dioxide, which was due to the small content of MXene in the heterostructure.

3.3. Study of the Photocatalytic Activity of TiNbC/TiO₂ in the Process of CO₂ Reduction and Hydrogen Evolution

The activities of the synthesized composite samples were tested in CO₂ reductions and hydrogen formations from aqueous solutions of ethanol under LED irradiation. The results of the photocatalytic reductions of CO₂ and hydrogen productions are summarized in

Table 2. Photocatalytic activities and selectivity of TiNbC/TiO₂.

Photocatalyst	Yield, μmol h−1g−1			W (CO2RR), μmol h−1g−1	S (CO2RR), %
	CH4	CO	H2
TiO2 Evonik P25	5.0	2.1	1.2	44.1	94.8
1% TiNbC/TiO2	5.0	0.9	3.0	41.8	87.5
5% TiNbC/TiO2	5.0	1.5	1.8	43.0	92.1
10% TiNbC/TiO2	2.0	1.0	0	18.0	100

and

Table 3. Photocatalytic activities of TiNbC/TiO₂ in the process of H₂ production.

Photocatalyst	The Rate of H2 Production, μmol/min	Activity in H2, μmol h−1g−1
TiO2 Evonik P25	0.003	4.0
1% TiNbC/TiO2	0.038	45.6
5% TiNbC/TiO2	0.047	56.4
10% TiNbC/TiO2	0.019	22.8

, respectively. Note that only the TiNbC/TiO₂ heterostructures showed photocatalytic activities in both processes.

Figure 7 shows the results of the photocatalytic tests of the investigated titanium dioxide and MXene heterostructures in the photocatalytic CO₂ reduction. One can see that the main reduction products were CH₄ and CO; hydrogen was evolved as a by-product in the water-splitting reaction.

In terms of the overall rate of CO₂ reduction, the deposition of TiNbC on commercial titanium dioxide did not lead to an increase in activity. At the same time, the presence of a rather large amount of hydrogen in the case of a 1% TiNbC/TiO₂ photocatalyst should be noted.

Also, the activity was tested in hydrogen evolution from aqueous solutions of ethanol (Table 3). In the case of this test reaction, an increase in activity was observed when MXene was deposited on the surface of commercial titanium dioxide Evonik P25, and the dependence of activity on the mass fraction of the max-phase passed through a maximum at a content of 5 wt.%. In this case, the deposited maxene could play a role similar to that of a metal deposited on the surface of a semiconductor, increasing the separation of photogenerated charge carriers.

The resulting catalytic activities of the heterostructures were not considered commercially applicable (more than 1000 μmol h⁻¹g⁻¹ is required), but there are ways to improve the catalytic activities of TiNbC/TiO₂. One such way could be to translate the heterostructures into the nanoparticle range by laser ablation or fragmentation [37,38]. Laser fabrication of nanoparticles in a liquid from a mixture of two types of semiconductors as an initial powder yields spherical nanoparticles with a heterostructure. In each case, it is necessary to select the laser modes, as well as the type of solvent.

4. Conclusions

Note that we have carried out photocatalytic tests on photocatalytic hydrogen evolutions and CO₂ reductions using the heterostructures based on TiO₂ and MXenes Mo₂TiC₂, Mo₂Ti₂C₃, VCrC, and TiNbC with MXene contents of 1, 5, and 10% as photocatalysts. Only the TiNbC/TiO₂ heterostructures showed photocatalytic activities.

The study of the photocatalytic activities of TiNbC/TiO₂ in the processes of hydrogen evolution and reduction of CO₂ demonstrated for the first time that the creation of heterostructures of MXene and titanium dioxide can be considered a promising way to create new materials for the needs of green energy. It was also shown that there was an area of an optimal ratio between the components of a heterostructure in the creation of such heterostructures. In the case considered in this article, the optimal ratio was the addition of 5% MXene to titanium dioxide. The question remains open about the promising use of this heterostructure in the form of nanoparticles. But, for this purpose, it is necessary to work out the technology of the synthesis of such nanoparticles. From the general physical concepts, heterostructures in the forms of nanoparticles should be more active in photocatalytic processes.

In a recent work, the photocatalytic activities of TiO₂/Ti₃C₂ nanocomposites with MXene contents of Ti₃C₂ < 10%, 38.7%, and 96% were investigated on fabrication methods for MXene-based photocatalysts and photocatalytic performances for contaminant degradation, CO₂ reduction, H₂ evolution, and N₂ fixation with various MXene-based nanocomposites [39]. The use of compounds such as TiO₂, BiVO₄, Ag₃PO₄, In₂S₃, CdS, BiFeO₃, BiOBr, Fe₂O₃, RuO₂, C₃N₄, AgInS, and MIL-100(Fe) as co-catalysts remains a promising direction.