Rationally Designed 2D CZIS/2D Ti₃CNT_x Heterojunctions for Photocatalytic Hydrogen Evolution Reaction

Abstract

Highly efficient photocatalysts for solar energy conversion require effective charge carrier separation and rapid interfacial transport kinetics to maximize electron availability. Two-dimensional Ti₃CNT_x, a novel conductive material in the MXene family with exceptional electrical conductivity, has emerged as an ideal electron transfer mediator due to its large specific surface area and abundant active terminal groups. In this work, we strategically integrated the 2D multi-metal sulfide Cu-Zn-In-S (CZIS) with 2D Ti₃CNT_x nanosheets through physical mixture, constructing a heterostructured 2D/2D CZIS/Ti₃CNT_x composite photocatalyst for the hydrogen evolution reaction. The unique architecture significantly accelerates electron migration from CZIS to Ti₃CNT_x, while synergistically promoting the spatial separation and directional transfer of photogenerated electron–hole pairs (e⁻/h⁺). When the hydrogen evolution reaction is carried out under identical conditions, the hydrogen yield rate is 4.3 mmol g⁻¹ h⁻¹ with pristine CZIS but is improved dramatically to 14.3 mmol g⁻¹ h⁻¹ when the composite containing an adequate amount of 2D Ti₃CNT_x is used. This study offers new insight into the rational design and controllable synthesis of Ti₃CNT_x-based composite photocatalytic systems for efficient photocatalytic hydrogen production.

1. Introduction

Advanced catalysts for solar-driven water splitting provide a transformative approach to harnessing clean energy and addressing global sustainability challenges [1,2,3,4,5,6,7,8]. Over recent decades, significant progress has been made in the development of nanostructured photocatalytic materials that are pivotal to advancing solar energy technologies. Among them, polymetallic sulfides have gained prominence as efficient semiconductor photocatalysts due to their exceptional spectral response, narrow bandgap, and high photochemical sensitivity [9]. Compared to binary metal sulfides, multinary metal sulfides (e.g., ZnIn₂S₄, Zn_xCd_1−xS) are distinct as the synergistic interactions between their metal ions help to deliver stable and superior performance during extended photocatalytic processes such as the hydrogen evolution reaction (HER) [10]. Notably, quaternary metal sulfides with single-crystalline architecture exhibit outstanding photocatalytic activity and have become intriguing materials in modern research [11].

The three pillars of photocatalytic activity are photon absorption efficiency, charge carrier migration dynamics, and surface reaction kinetics [12]. Among sulfide semiconductors, Cu-Zn-In-S (CZIS) has demonstrated exceptional photocatalytic efficiency for HER under visible light, thanks to its optimized band structure [13,14,15,16]. However, it has inherent limitations, such as low charge separation efficiency, limited stability under prolonged operation, and suboptimal band structure that restricts the broader utilization of the solar spectrum. Various strategies have been developed to address these drawbacks [17,18], including elemental doping, heterojunction engineering, surface modifications, etc.

Earth-abundant transition metal can replace noble metals to provide efficient cocatalysts for solar-driven hydrogen production [19,20,21,22,23]. MXenes, a burgeoning family of two-dimensional (2D) transition metal carbides, nitrides, and carbonitrides, have emerged as exceptional cocatalysts in photocatalysis [24]. Their superior performance stems from their high electrical conductivity, large surface area, and tunable composition [25]. They suppress charge recombination and promote charge separation to boost the photocatalytic efficiency, and they interact effectively with reactants through their abundant surface functional groups (e.g., –OH, –O, –F) [26]. Thanks to these advantages, MXenes have been proven to be a highly versatile material in HER as well as applications like CO₂ reduction, and pollutant degradation.

The integration of Ti₃CNT_x, a member of the MXene family, with CZIS creates an intimate heterojunction interface allowing effective spatial separation of photogenerated carriers. This configuration optimizes carrier mobility and makes use of trapping channels to effectively mitigate the inherent limitations of pristine CZIS, including low electrical conductivity, rapid recombination of photogenerated holes and electrons, and insufficient hydrophilicity [18,27,28,29,30,31]. The 2D/2D architecture is known to render superior interfacial characteristics, particularly in terms of shorter carrier diffusion length, larger contact interface, and greater charge carrier mobility [32,33]. A planar heterojunction configuration facilitates interfacial charge transfer kinetics via built-in electric fields, and ultrathin Ti₃CNT_x MXene nanosheets enable broad-spectrum light absorption spanning UV to NIR regions.

In this work, we prepared 2D CZIS/2D Ti₃CNT_x MXene heterostructures via physical mixture. In the nanohybrid system, the CZIS component serves as a photoactive semiconductor for visible light harvesting, while the Ti₃CNT_x MXene enables directional charge transfer to allow efficient charge carrier separation. The synergy thus creates a stable 2D/2D photocatalytic system for HER that gives a remarkable hydrogen evolution rate of 14.3 mmol·h⁻¹·g⁻¹ under visible light illumination, surpassing most reported MXene-based photocatalysts.

2. Results

Figure 1 illustrates the synthesis of nanocomposite, denoted as 2D CZIS/2D Ti₃CNT_x. Two-dimensional nanosheets of titanium carbonitride (Ti₃CNT_x) were synthesized by firstly etching Ti₃AlCN with an HCl/LiF solution to remove the Al layer and then exfoliating and delaminating the layers into thin sheets by ultrasonication. Two-dimensional CZIS nanobelts were synthesized via a colloidal approach mediated by oleylamine (OLA) and 1-dodecanethiol (DDT). The assembly of Ti₃CNT_x and CZIS, at different mass ratios then gave the CZIS/Ti₃CNT_x nanocomposite (referred to as CN-y%, where y denotes the mass ratio of Ti₃CNT_x in the composite).

Figure 2 shows scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of Ti₃CNT_x, CZIS, and CN-8%. The lamellar morphology of Ti₃CNT_x is clearly seen (Figure 2a). For CZIS, a distinct nanobelt structure exists (Figure 2b,f) [13]. Figure 2c,d verify the formation of CZIS and Ti₃CNT_x to give CN-8% composite. In CN-8%, a coupling interface exists between the Ti₃CNT_x nanosheets and the CZIS nanobelts (Figure 2e), and the 0.27 nm and 0.34 nm lattice spacings can be attributed to Ti₃CNT_x and CZIS, respectively (Figure 2f) [13,34]. Elemental mapping (Figure 2g) confirms the coexistence and uniform distribution of S, Cu, In, Zn, Ti, and N within CN-8%.

The characteristic XRD peaks of CZIS at 26.7°, 28.3°, 30.3°, 39.3°, 47.2°, 51.2°, and 55.9° correspond to its (1010), (0002), (1011), (1012), (1120), (1013), and (1122) crystal planes, respectively (Figure 3a) [13]. The XRD peaks of Ti₃CNT_x at 7.6° and 15.1° can be attributed to the (002) and (004) crystal planes and are characteristic of the layered structure, and the successful etching of the Al layer in Ti₃AlCN is verified by the complete disappearance of diffraction peaks resulting from the Al-containing layers in Ti₃AlCN (Figure S1). The XRD pattern of CN-8% inherits the features of CZIS and contains the (002) peak of Ti₃CNT_x, suggesting that the composite is assembled successfully without phase alteration [35]. The intensity of the (002) peak in CN-y% increases with the mass fraction of Ti₃CNT_x (Figure S2).

The interfacial electronic interactions between CZIS and Ti₃CNT_x and the valence states of the elements in CN-8% were investigated using X-ray photoelectron spectroscopy (XPS). The survey spectra of CN-8% verify the successful introduction of Ti₃CNT_x into CZIS (Figure 3b). In the Cu 2p spectrum (Figure 3c), the characteristic peaks of Cu 2p_3/2 and Cu 2p_1/2 are at 931.65 and 951.54 eV for CZIS but shift notably to 932.34 and 952.23 eV in CN-8%. Figure 3d–f show similar shifts for Zn, In, and S. Therefore, electron transfer from CZIS to Ti₃CNT_x must have occurred in CN-8% [36,37,38]. This electronic interaction allows robust interfacial coupling between CZIS and Ti₃CNT_x. The Ti 2p spectrum of Ti₃CNT_x (Figure 3g) has four distinct components, i.e., Ti–C (455.33 eV), Ti–N (456.00 eV), Ti(II) (457.16 eV), and Ti–O (458.03 eV) [39,40]. In CN-8%, all four components of Ti 2p shift to a lower energy, and similar shifts also occur to the C and N peaks (Figure 3h,i). The results thus further verify the electron transfer from CZIS to Ti₃CNT_x. In CN-8%, the diminished intensity of C–Ti, C–N, C–O and C–F peaks (Figure 3g) likely originates from the relatively low mass fraction of Ti₃CNT_x, rather than chemical degradation [41].

The photocatalytic performance of Ti₃CNT_x, CZIS, and CN-y% in the hydrogen evolution reaction (HER) under visible light irradiation (λ ≥ 420 nm) was systematically investigated using Na₂S/Na₂SO₃ as a hole scavenger. Under the same conditions, no hydrogen is generated when Ti₃CNT_x alone is used as the catalyst, and CZIS gives a hydrogen yield rate of 4382 μmol·g⁻¹·h⁻¹ (Figure S3) [42]. The combination of Ti₃CNT_x and CZIS enhances H₂ generation remarkably (Figure 4a), as the hydrogen yield rate increases to 14,335 μmol·g⁻¹·h⁻¹ for CN-8%, although it decreases to 4869 μmol·g⁻¹·h⁻¹ for CN-16%. Presumably, incorporating a moderate amount of Ti₃CNT_x in CZIS facilitates charge transport/separation, but an excessive amount of Ti₃CNT_x obscures the active sites of CZIS and weakens the catalytic activity [43]. The stability of the photocatalysts was assessed through 12 h of continuous illumination tests. For CN-8%, the cumulative hydrogen yield increases linearly over the 12 h (Figure 4c), and the XRD of the used catalyst is identical compared to that of the fresh catalyst (Figure S4). The results collectively demonstrate the excellent structural integrity and stability of CN-8%. The apparent quantum yield (AQY) of CN-8% is 3.7%, 2.2%, and 0.7% at 420, 450, and 520 nm, respectively (Figure 4d). The close correlation between AQY and the optical absorption characteristics confirms the efficient utilization of visible light.

The charge separation and transport in CN-y% were systematically investigated through transient photocurrent response and electrochemical impedance spectroscopy (EIS) to find out why the addition of Ti₃CNT_x to CZIS can enhance the performance of the photocatalyst. Figure S5 shows that under visible light irradiation, CN-y% exhibits a notable decrease in electrochemical impedance, and the change in impedance matches the photocatalytic activity. The most significant amplification occurs to CN-8% (Figure 5a), as the current density becomes about three times that of CZIS, which indicates superior charge transfer kinetics. In the Nyquist plot (Figure 5b), compared to CZIS, CN-8% has a semicircle with a reduced diameter, which suggests diminished charge transfer resistance. These findings collectively demonstrate that doping CZIS with an appropriate amount of Ti₃CNT_x significantly enhances the carrier separation and transport efficiency, leading to a significant improvement in the photocatalytic performance [44].

Carrier separation was also investigated through steady-state photoluminescence (PL) and time-resolved photoluminescence (TRPL) spectroscopy. Figure 5c shows that the PL of CN-8% is quenched substantially. Hence, adding Ti₃CNT_x to CZIS facilitates the directional migration of photogenerated electrons and holes toward the active sites, and the suppressed carrier recombination enhances the photocatalytic activity [45]. Figure 5d shows that the average carrier lifetime is 2.25 ns for CZIS but 1.97 ns for CN-8%. A shorter fluorescence lifetime corresponds to a lower probability of charge recombination and a higher probability of surface migration, leading to stronger catalytic performance [46]. The findings all explain the stronger photocatalytic activity of CN-8%.

Some distinctions exist between CZIS and CN-8% in the UV-Vis diffuse reflectance spectroscopy spectra (Figure S6). The enhanced light-harvesting efficiency of CN-8% compared to CZIS proves that incorporating Ti₃CNT_x helps to optimize photo-responsive behavior. The flat-band potential of CZIS is determined through Mott–Schottky analysis (Figure 6a) to be −0.64 V vs. Ag/AgCl, which is −0.34 V vs. reversible hydrogen electrode (RHE). In Figure 6b, the DRS-derived Tauc plot was applied to deduce the bandgap width (about 1.71 eV) of CZIS according to the formula αhv = A(hν − E_g)^n/2, where n = 1 as CZIS is a direct semiconductor. Figure 6c sketches how Ti₃CNT_x improves the photocatalytic activity of CN-8% through band engineering. The favorable thermodynamic conditions, particularly in terms of conduction band positioning, make CN-8% suitable for the photocatalytic HER [47]. Photoexcitation promotes electron transitions from the valence band (VB) to the conduction band (CB), followed by rapid interfacial transfer across the heterojunction for participation in surface reactions.

When CN-8% is used as the photocatalyst, the reaction pathways are as follows. Because of the superior electronic conductivity of Ti₃CNT_x, the photogenerated electrons migrate efficiently to Ti₃CNT_x, where they participate in the surface-mediated reaction with the adsorbed H⁺ to generate hydrogen. Meanwhile, S²⁻ and SO₃²⁻ are oxidized by the photogenerated holes to form S_n²⁻ and SO₄²⁻, thus acting as sacrificial reagents for the photocatalytic hydrogen generation [48]. However, the oxidation of S²⁻ to S_n²⁻ has been shown to decrease in H₂ formation over time because S_n²⁻ has a high level of light absorption in the visible region [47]. Fortunately, SO₃²⁻ can reduce S_n²⁻ back to S²⁻ to render the solution colorless again, thus preventing S_n²⁻ (which gives the solution a yellow color) from interfering with effective light harvesting for water reduction [49]. That is, in the water/semiconductor suspension, the S²⁻/SO₃²⁻ mixture contains electron donors that enhance photocatalytic activity and catalyst stability in the context of the HER [48].

The enhanced photocatalytic efficacy of CN-8% primarily stems from the 2D/2D heterostructure formed by integrating CZIS nanobelts with Ti₃CNT_x nanosheets, which significantly exposes abundant photoactive sites. Engineering heterojunctions is a key strategy for improving the photocatalytic performance of quaternary sulfides. The highly conductive Ti₃CNT_x nanosheets substantially enhance the charge carrier separation in CZIS and suppress electron–hole recombination. In addition, the hydrophilic functional groups (–OH, –O) on the Ti₃CNT_x surfaces promote intimate photocatalyst–water interactions (Figure S7), facilitating proton transfer dynamics and elevating hydrogen evolution rates [50]. Table S1 shows that CN-8% has superior photocatalytic performance in the HER compared to most reported MXene-based photocatalysts.

The physical mixture of 2D CZIS nanobelts and 2D Ti₃CNT_x nanosheets gave a composite as a superior photocatalyst in the hydrogen evolution reaction. The performance of the photocatalyst, referred to as CN-y%, depends on the mass fraction of Ti₃CNT_x in the composite (denoted as y) and is the best when y = 8. The excellent photocatalytic performance of CN-8% is attributed to the intimate interfacial contact between CZIS and Ti₃CNT_x. Photogenerated electrons migrate efficiently from CZIS to Ti₃CNT_x, and the reduction in protons takes place on the surfaces of Ti₃CNT_x [51]. The CZIS/Ti₃CNT_x composite facilitates charge separation and suppresses electron–hole recombination, thereby markedly improving both the activity and the stability of the catalyst [52,53]. This work uncovers critical design principles for optimizing 2D/2D heterojunction photocatalysts, offering targeted strategies for advancing solar energy conversion technologies.

3. Materials and Methods

Materials: All chemicals were ACS grade or analytical reagent grade and used as received without further purification. Titanium aluminum carbonitride (Ti₃AlCN, 500 mesh, >98%), sodium diethyldithiocarbamate trihydrate (C₅H₁₀NNaS₂·3H₂O, 99%), 1-dodecanethiol (98%), oleylamine (80%–90%), 1-octadecene (>90.0%), ethanol (≥99%), anhydrous sodium sulfite (Na₂SO₃, ≥98%), sodium sulfide nonahydrate (Na₂S·9H₂O, ≥98%), and n-hexane (97%) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Hydrochloric acid (36%–38%) was obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Deionized (DI) water was purified using a laboratory water purification system (RODI-220A1, Qingdao Labtech Scientific, Qingdao, China).

Preparation of Cu(S₂CNEt₂)₂ (Cu(dedc)₂): The solution of Cu(NO₃)₂·3H₂O (1.208 g, 5 mmol) in water (100 mL) was added dropwise into the solution of N-ethyldithiocarbamate sodium (NaS₂CNEt₂, 2.2531 g, 10 mmol) in water (100 mL) under vigorous magnetic stirring (800 rpm). The mixture was stirred at room temperature for 30 min before centrifugation (6000 rpm, 10 min). The precipitates were washed with 1:1 v/v ethanol and deionized water three times and finally dried at 60 °C and 0.1 MPa for 12 h. Zn(dedc)₂ and In(dedc)₃ were synthesized analogously.

Preparation of CZIS nanobelts: To a three-necked flask containing OLA (16 mL), DDT (16 mL), and ODE (8 mL), the following were added sequentially: Cu(dedc)₂ (0.144 g, 0.4 mmol), In(dedc)₃ (0.224 g, 0.4 mmol), and Zn(dedc)₂ (0.58 g, 1.6 mmol). The mixture was stirred vigorously until complete dissolution, and the homogeneous solution was heated to 100 °C under continuous magnetic agitation and degassed for 20 min. The temperature was then increased to 250 °C at 3 °C/min under N₂ protection and maintained at 250 °C for 1 h before cooling to room temperature. The crude precipitates were washed alternately with n-hexane and ethanol for three cycles, with centrifugation at 8000 rpm for 10 min following each wash. After the final cycle, the precipitates were lyophilized for 24 h, yielding CZIS as a fine powder.

Preparation of Ti₃CNT_x nanosheets: To a three-necked round-bottom flask containing concentrated HCl (12 M, 20 mL) was added LiF (1.5× g), and the mixture was stirred magnetically at 45 °C for 30 min to give a homogeneous solution before the addition of Ti₃AlCN powder (0.5× g). The suspension was stirred at 45 °C for 24 h before cooling to ambient temperature. The solid material was iteratively separated by centrifugation (8000 rpm, 5 min) and washed with DI water until neutral pH, thus giving multilayer Ti₃CNT_x (m-Ti₃CNT_x). The multilayer Ti₃CNT_x was redispersed in DI water (20 mL) and sonicated for 5 h. The supernatant was collected and centrifuged at 3500 rpm for 15 min to give delaminated Ti₃CNT_x nanosheets, which were lyophilized and stored in amber vials under argon atmosphere.

Preparation of 2D CZIS/2D Ti₃CNT_x nanocomposites: Solution A was prepared by ultrasonically dispersing CZIS (500 mg) in ethanol (100 mL). Solution B was prepared by ultrasonically dispersing Ti₃CNTₓ (x mg, x = 10, 20, 40, 60, 80) in ethanol (100 mL). Under nitrogen (N₂) protection and at room temperature, Solution A was added dropwise into Solution B with continuous magnetic stirring at 150 rpm. The resulting mixture was stirred in the dark at room temperature for 12 h. The solid materials were collected after centrifugation and lyophilization to give the 2D CZIS/2D Ti₃CNTₓ nanocomposites, designated as CN-y (y = 2, 4, 8, 12, 16).

4. Conclusions

This work presents the rational design of a 2D/2D heterojunction photocatalyst through the electrostatic self-assembly of Cu-Zn-In-S (CZIS) nanobelts and Ti₃CNTₓ MXene nanosheets. The unique architecture facilitates efficient spatial separation of photogenerated charge carriers, wherein electrons rapidly transfer from CZIS to highly conductive Ti₃CNTₓ with its large specific surface area. This yields a hydrogen evolution rate of 14.3 mmol·g⁻¹·h⁻¹ for the CN-8% composite—3.3-fold higher than pristine CZIS (4.3 mmol·g⁻¹·h⁻¹)—along with a 3.7% apparent quantum yield at 420 nm, confirming effective visible light utilization. Comprehensive characterizations (SEM, TEM, XRD, XPS) verify heterojunction formation and interfacial electron transfer, while electrochemical analyses demonstrate reduced charge transfer resistance and suppressed carrier recombination. The enhanced performance is attributed to shortened carrier diffusion paths, increased active site exposure, and proton transfer facilitation via hydrophilic groups on Ti₃CNTₓ. This study uncovers a critical design principle for 2D/2D heterojunctions: synergistic integration of conductive MXenes with semiconductors to optimize charge dynamics, offering a viable strategy for advancing solar hydrogen production and sustainable energy technologies.