## Introduction

With increasing demands in modern society, the energy crisis and environmental problems are currently in the spotlight. Hydrogen is the main sustainable source of renewable energy and is highly required for advanced energy conversion systems. Recently, photoelectrocatalytic and photoelectrochemical water-splitting methods are an efficient approach for the scalable generation of hydrogen. The concept of light-induced energy production through semiconductor catalysts attracted many research efforts aiming at the development of low-cost, efficient bifunctional materials for hydrogen production and environment-sensing response. The performance of these materials can be tuned while illuminating the light of various wavelengths. The occurred photo-induced electron transfer in the targeted nanosheets-based sensitive material will generate photoelectrochemical water splitting, e.g., faster hydrogen evolution reaction (HER) or broadband light-sensitive detector. These will pave new ways for emerging materials as a cheaper and more productive alternative to most common fossil fuels-based methods of hydrogen production as well as traditional photodetectors.

Two-dimensional (2D) transition-metal dichalcogenides (TMDs) have been demonstrated as promising catalysts for water splitting, hydrogen reduction, and water oxidation1,2. The unique chemical and physical properties of 2D TMDs are demonstrated by their small size enriched by the number of edge active sites, band-edge position, quantum confinement effect and thus, exhibit photo-induced catalytic efficiency3. TMDs are considered the most promising cost-effective catalysts, and their properties are determined by the TMDs’ polymorph type, namely hexagonal 2H or trigonal 1T. It has been reported that stable 2H-MoS2 shows semiconducting properties; meanwhile, metastable 1T-MoS24 or 1T-WS25 forms demonstrate metallic properties. The metallic phase of a material is capable of a more efficient performance than the semiconducting 2H counterpart due to its enhanced electrical conductivity. The metallic character is associated with undercoordinated chalcogen atoms at the 1010 edge of 2D TMD, providing more efficient adsorption of hydrogen, which is a crucial step for electrochemical water splitting6. The MoS2 and WS2 exhibit superior catalytic activity and are used as alternatives to Pt-based catalysts7,8,9. The WSe2 is demonstrated as an excellent output stability photodetector due to its high absorption coefficient in the visible and near-infrared regions10,11. Other known 2D materials, such as TiSe212,13,14, NiSe215,16, TaSe217, and TaS218,19,20, are interesting due to their superconductivity, charge-density wave, and metal-insulator transitions.

Among other 2D TMD materials, tantalum disulfide (TaS2) has been subject to numerous studies due to the variety of the material’s unique structural and electronic phases21. Its initial metallicity and, thus, electrical conductivity can lead to potential applications such as a light-responsive active electrode in the HER and photosensing. Albeit the study of TaS2 is mainly focused on the superconductivity22,23,24, field emission25,26, photo-induced features of the material for HER and photosensing have not been explored yet. Aiming to enhance the photoresponse performance of the TaS2, the quality and dimensions of the material must be considered. For instance, bulk TaS2 lacks exposed active sites that significantly affect the conversion efficiency of the material and narrows its application. Therefore, a thorough cleavage of the van der Waals structure by selecting the most appropriate method is essential to increase the conversion efficiency of TaS2.

Several methods have been demonstrated to produce layered TaS2, such as chemical vapor deposition (CVD)27, mechanical cleavages21, intercalation28, ultrasonication29,30,31, or liquid-phase exfoliation22. However, the CVD method remains to be improved before it can be used to produce large-domain homogeneous TaS2 films. The mechanical cleavage is not scalable and undergoes a lack of control over product thickness and size. Alternatively, electrochemical exfoliation (one of the solution-based techniques) is the most convenient, controlled, and straightforward method that can be employed in ambient temperature for large-scale production of 2D TaS2. Recently the electrochemical approach has been successfully applied to exfoliate few-layer phosphorene32 and arsenene33 and has been pioneered for single-step production of platinoid-decorated phosphorene34.

Therefore, in this study, the photoresponse of electrochemically exfoliated TaS2 nanosheets which have preliminarily been characterized by micro- and spectroscopic techniques (Supplementary Notes 1) was investigated in an effort to develop a low-cost, efficient photosensitive electrode that can replace noble metal-based catalysts35 (as a photoelectrocatalyst—PEC) and CMOS-based photodetectors36 (as a photodetector—PD). Pioneered low-potential electrochemical exfoliation of TaS2 crystals was carried out in a nonaqueous electrolyte medium of tetrabutylammonium hexafluorophosphate (TBAPF6) in N, N-dimethylformamide (DMF). The as-exfoliated 2D TaS2 nanostructures consist of few-layer nanosheets and nanoparticles. Detailed characterization of bulk and exfoliated TaS2 was performed through various microscopic and spectroscopic techniques. The photo-induced high HER electrocatalytic activity of the as-exfoliated TaS2 nanostructures with low overpotential (>575 mV and 320 mV) was reported. A room-temperature TaS2-based broadband photodetector in the acidic medium of 0.5 M H2SO4 has been presented. The higher electrical conductivity of 2D TaS2 compared to its bulk form can improve the efficiency of charge distribution in the catalytic/sensing systems and, thus, broaden the material’s application area.

## Results

### Synthesis and characterization of the materials

Electrochemical exfoliation of bulk TaS2 crystal [see scanning electron microscopy (SEM) images in Fig. 1a–c] was performed in anhydrous electrolyte of 0.01 M TBAPF6 in DMF. Following three main steps of the procedure, namely wetting of the crystal, accumulation (or activation) of the tetrabutylammonium cations (TBA+), and their intercalation between TaS2 layers were carried out at −1, −2.5, and −3.8 V (vs. 0.1 M Ag/AgNO3), respectively. The first two steps were completed in 2 min each, followed by 4 h of TBA+-assisted intercalation-delamination process at a potential of −3.8 V. It has to be noted that the duration of the process is defined by the initial size of the TaS2 crystal and can be terminated when a sufficient amount of the exfoliated material is produced.

After the exfoliation of the TaS2 crystal, flakes of various sizes and thicknesses were observed in the resulted dispersion based on STEM (Fig. 1d–f) and AFM analysis (Fig. 2). These include TaS2 nanosheets of lateral size up to 6 μm with a thickness of 30 nm and 2.3 μm nanosheets with 7 nm thickness. (Fig. 2a, b, respectively). In addition, exfoliated material consists of TaS2 nanoparticles of 100–200 nm lateral size and thickness in the range between 0.5 and 3 nm. (Fig. 2c). The successful exfoliation of the TaS2 crystal into nanosheets and nanoparticles, as well as their size distribution, were confirmed by the statistical analysis (Supplementary Fig. 1) obtained from STEM images. The results are also in good agreement with high-resolution TEM images (Supplementary Fig. 2) of two types of exfoliated nanostructures with a lateral size of ten-hundreds of nanometers and several micrometers.

TEM and SAED microscopies demonstrate good quality of exfoliated TaS2 nanosheets (Fig. 3a–d) of hexagonal structure (schematic image, Fig. 3e, f). High-resolution TEM images in Supplementary Fig. 3 also show the formation of round-shaped TaS2 nanoparticles localized on the surface of the larger flakes. This is consistent with the statistical analysis of particle size distribution (Supplementary Fig. 1) that confirms the production of two types of TaS2 nanostructures. Based on the fast Fourier transform (FFT) analysis of TEM images (Supplementary Fig. 4), the crystal symmetry of TaS2 is demonstrated as the fringes oriented in a three-fold rotation axis. The d-spacing of the symmetry of the TaS2 crystal of about 2.9 Å and 2.93 Å defines its (100) planes’ orientations. The electron diffraction patterns of a flat area of the TaS2 nanosheet show this growing direction. Different magnification TEM images of the TaS2 exhibit in-plane atoms are ordering with the lattice spacing of 0.305 nm. The morphology of the exfoliated material and thus sharply defined layered appearance are caused by the initial structure of the TaS2 crystal.

The SAED patterns of the large area TaS2 flakes (Supplementary Fig. 5) verified several configurations of 2H-type structure with the different twist angles of 7°, 8°, 9°, and 10° (Fig. 4). The value of twist angle of as-exfoliated TaS2 nanosheets is relatively regular for all analyzed samples (7‒10 degrees) and increases between rows: 1 layer < 2 layers < 3 layers. The trigonal arrangement of Ta atoms in Fig. 3e, f corresponds to the standard stacking of S-Ta-S in the unit cell. This direct stacking of each layer via Ta atoms was theoretically proven in a few publications37,38; however, experimentally was confirmed only for 1T-TaS2 bilayer structure39. Based on the density of state calculations, the electronic properties of the 1T-TaS2 bilayer depend on the twist angles. The increase of both formation energy and bandgap has been theoretically explained due to the interlayer interaction of twisted TaS2 flakes and the redistribution of charge density induced by the in-plane distortion38. This feature can also lead to the increase of the electronic specific-heat coefficient of the randomly twisted TaS2 nanosheets22, establishing the material as a promising candidate for photo-induced applications40.

The analysis of the crystal structure and optical characteristics of the bulk and exfoliated TaS2 is demonstrated in Fig. 5. X-ray diffraction (XRD) was used to verify the successful synthesis of hexagonal 2H-TaS2 (Supplementary Notes 2, 3). The XRD pattern in Fig. 5a confirmed the P-3m1 hexagonal structure of TaS2 (PDF: 04-003-4190). No other phases were observed, indicating the purity of the starting material. The sharp peaks in the pattern ascertaining the high crystallinity of the sample. Similar peaks were observed in the exfoliated material’s XRD pattern, albeit broadened and with lower intensity caused by the disruption of initial crystallinity due to the exfoliation procedure. The broadening of the XRD peaks indicates a decrease in crystal size and lattice strain increase. Figure 5b shows the Raman spectra of bulk and as-exfoliated TaS2 samples. Three main Raman features depicted at 239, 301, and 373 cm−1 from the bulk crystal were assigned to the E2g, E1g, and A1g modes, respectively28,41. More prominent peaks of the same modes and their 4 cm−1 redshifts were observed for the exfoliated nanosheets. The shift is attributed to the decrease of the interlayer van der Waals forces and the material’s thickness reduction. UV-VIS absorption spectrum of the as-exfoliated TaS2 nanostructures recorded at room temperature is demonstrated in Fig. 5c. The TaS2 represents low in-plain reflectivity at high energies (from 400–720 nm) and vice versa at low energies accordingly42.

To confirm chemical composition and binding states, we have measured the survey (Supplementary Fig. 6) and high-resolution X-ray photoelectron spectra (HR-XPS) of Ta 4 f and S 2p (Fig. 6). Survey spectra confirmed composition, which was close to TaS2 for bulk and exfoliated samples as well. Since exfoliated samples were measured on a gold substrate, Au peaks can also be seen in its survey spectrum. HR-XPS Ta 4 f spectra were deconvoluted with two binding states TaIV (TaS2) and TaV (Ta2O5), with binding energies 22.9 and 26.1 eV, respectively43. Although O 2 s at 24.0 eV overlaps with the Ta 4 f, it was not used for deconvolution because its amount was less than 1% due to a low relative sensitivity factor of 0.14 compared to 8.62 for Ta 4 f. Each binding state in Ta 4 f has spin-orbit components separated by 1.92 eV. Binding state ratios are summarized in Supplementary Table 1. Bulk crystal was covered by 20% of Ta2O5, and after exfoliation, this amount increased to 30%. It has to be noted that TaS2 is susceptible to oxidation, and this ability is even higher after the exfoliation43. Thus, operating the material in an oxygen-free environment and storing it in the dry solvents will increase the stability of as-exfoliated TaS2.

### Photoelectrocatalytic performance of as-exfoliated TaS2 for HER

Light-driven electrocatalysis using semiconducting materials (e.g., TaS2) or their heterostructures is one of the most promising sustainable approaches for hydrogen generation by water splitting44,45,46. Even though a thermal treatment or catalyst support texturization of the catalytic films is used to increase the number of active centers of the TaS2 flakes, these methods are time-consuming and require a high-temperature and inert gas atmosphere47. The light-induced electrochemistry (photoelectrochemistry) can be a simple approach to enhance the HER activity of the TaS2 nanosheets. To the best of our knowledge, the photoelectrocatalytic activity of TaS2 has not been explored to date.

Herein, the photoelectrocatalytic performance of electrochemically exfoliated TaS2 for HER was explored (see Supplementary Notes 4). The three-electrode linear sweep voltammetry (LSV) was performed in the 0.5 M H2SO4 employing a few-layer TaS2 deposited on a glassy carbon as a working electrode. The measurements have been repeated in the light mode for four different wavelengths (420, 532, 630, and 720 nm). In addition, the results were compared with the dark mode for as-exfoliated TaS2 nanosheets of mixed composition. The mixed composition consists of nanosheets of various sizes and nanoparticles, as was confirmed by the AFM, STEM, and statistical analysis. Few-layer TaS2 samples of a smaller lateral size (~30–40 nm, Supplementary Fig. 7) were obtained as follows: first, the TaS2 dispersion was centrifuged at 3500 rpm for 10 min; second, the collected supernatant was centrifuged at 10,000 rpm for 10 min, and the resulting precipitate was used for the test.

As shown in Fig. 7a, b, the HER for initial as-exfoliated TaS2 starts at a potential of 575 mV (vs. RHE) under dark mode at a current density of 10 mA cm−2. Under the illumination light of different wavelengths, the overpotential shifts towards lower potentials, specifically: 537, 549, 531, and 454 mV when blue, green, red, and far-red light-emitting diodes (LEDs) are implemented, respectively. This behavior is attributed to the carrier increment generated in the TaS2 nanosheets with light irradiation enhanced by the intrinsic defect (wrinkles) and mutual twisting of the individual layers. The current density of the TaS2 catalyst under dark decreases after 100 cycles (Supplementary Fig. 8), demonstrating a slow degradation of the as-investigated catalyst. It has to be noted that the onset overpotential for the TaS2 catalyst based on flakes of smaller size was significantly reduced to ~320 mV (Fig. 7c). Considering the influence of the light on the as-exfoliated samples, the ~100 mV decrement of the overpotential is expected that will make a few-layer TaS2 a superior HER photoelectrocatalyst. This study on bare TaS2 catalysts can compete with recently reported results48, where the performance of TaS2 was enhanced by the incorporation of conductive films49 or thermal treatment in the H2-environment47. Furthermore, thoroughly prepared catalysts under light illumination will outperform the HER activity of commercial Pt/C electrode (~40 mV), providing a cheaper alternative.

### TaS2-integrated photodetector

Solution-based PDs offer unique advantages over standard photodetectors in the form of a simple and easy device manufacturing process. Recently 2D nanomaterials have demonstrated their feasibility for PDs ranging from visible to THz; however, this is not inherent to one material42,50,51. Therefore, a broadband photodetection capability of materials is highly demanded to its wide application possibility. The intrinsic parameters of the material’s light sensitivity and response can be tuned by changing the applied voltage, analyte concentration, and power of the illuminated light. Herein, the photoelectrochemical response of TaS2 in a 0.5 M H2SO4 acidic medium was measured in a three-electrode system at ambient temperature. The TaS2-integrated PD was tested under four illumination wavelengths (420, 520, 630, and 720 nm) with a scanning speed of 10 mV·s−1. The procedure of the electrode preparation and photoelectrochemical measurements are described in Supplementary Notes 5.

The results illustrating a steady response toward different illumination wavelengths are depicted in Fig. 8. As shown in Fig. 8a, the power-dependent photocurrent measurement was achieved by using chronoamperometry in 0.5 M H2SO4 at 0.5 V against saturated calomel electrode (SCE) under the illumination of a 420 nm blue LED light source. The response of the TaS2-based electrode toward blue illumination was the highest in comparison with other LEDs. The response toward green illumination (Fig. 8b) was decreased and slightly increased after applying red (Fig. 8c) and far-red (Fig. 8d) light sources. The most prominent response of the electrochemically exfoliated few-layer TaS2 is demonstrated under blue illumination and well consistent with the light absorbance of the material (Fig. 5d). The near-linear increment in current density with the increase of the illumination power from 100 to 1200 mW was illustrated (Supplementary Fig. 9a) for all lights. Additionally, I–V measurements were performed using LSV with a scanning speed of 20 mV s−1 and under continuous blue light illumination with the power of 600 mW (Supplementary Fig. 9b). The results display an increment in photocurrent with the increase of applied voltage, demonstrating the maximum photoelectrochemical performance of the TaS2 nanosheets in 0.5 M H2SO4 at around −0.6 V vs. SCE. As the intensity increases, the current response jumps to a higher value corresponding to a higher conductance, indicating that the separation of electron-hall pairs is triggered.

The photodetection potential of the material has been assessed in more detail by calculating a few essential parameters. At first, the photoresponsivity (Rph) of the TaS2-integrated PD has been calculated using the following equation: Rph = ΔI/P, where ΔI represents the difference in photocurrent density between the dark and photocurrent, and P is the irradiation power intensity per unit area. The calculated photoresponse demonstrated the highest responsivity of about 0.62 mA·W−1 under a 420 nm blue LED source illumination at 0.5 V applied potential (vs. SCE). Figure 9a highlighted the photoresponsivity of the devices concerning the power for different illumination wavelengths. Another important parameter for the evolution of photodetection capability is specific detectivity; it has been determined accordingly52

$$D^ \ast = \frac{{\left( {AB} \right)^{1/2}}}{{{\mathrm{NEP}}}}$$
(1)

In the above equation, A, B, and NEP stand for the active area, measured bandwidth, and noise equivalent power, respectively. NEP is defined as the input signal power resulting in a signal-to-noise ratio (S/R) of 1 in a 1 Hz output bandwidth. Furthermore, in the equation form, NEP can be presented as52,53

$${\mathrm{NEP}} = \frac{{I_N}}{{R_{{\mathrm{ph}}}}}.$$
(2)

Here, Rph and IN represent the photoresponsivity of the PD and the noise current, respectively. In addition, the noise, which is associated with the dark current (ID) of the PD current, can be ascribed as IN2 = 2eIDB, where e represents the electronic charge42. The calculated specific detectivity of our PD devices of about 1.28 × 109 cm·Hz1/2 W−1 (Fig. 9b).

Prepared TaS2-based PD exhibited a broadband photodetection capability with the highest photoresponsivity (0.62 mA W−1) and specific detectivity (1.28 × 109 cm Hz1/2 W−1) toward 420 nm light illumination, which is nearly double compared with 532, 630, and 720 nm illumination wavelengths. This photoresponsivity value is higher than the response of solution-based photodetectors integrated with other TMDs (Supplementary Table 2). Our results demonstrate the versatility of 2H-TaS2-based PD, which exhibits metal-like characteristics coupled with broadband light absorption and thus emission due to energy-separated bands. The original metallicity of the TaS2 that is assisted by undercoordinated atoms at edge sites is enhanced by the random twisting between the TaS2 monolayers, which form a rich homogeneous interface with enhanced conductivity22. This consequently increases the electron transfer in the “TaS2 electrode/0.5 M H2SO4 electrolyte” system and thus improves the catalytic performance of the material, establishing it favorable for electrocatalysis1,2,43.

## Discussion

The light-induced efficiency of electrochemically exfoliated TaS2 for photoelectrocatalysis and photodetector performance has been demonstrated. A well-controlled exfoliation process that is based on the TBA+-cation intercalation between the TaS2 interlayers was performed in the nonaqueous DMF medium applying a potential of −3.8 V. Few-layer TaS2 nanosheets of different lateral sizes, namely, 10–100 nm and 1–30 μm, were identified and characterized by numerous characterization and microscopy techniques. It was shown that exfoliated TaS2 nanosheets possess a trigonal prismatic structure (2H type, hexagonal) and, thus, metallic character. The metallicity of the 2H-TaS2 nanosheets was enhanced by the twisting of the individual flakes and their mutual interaction, causing charge-density redistribution. This ability was further increased by the external illumination of the light of the following wavelengths: 420 nm (blue), 532 nm (green), 630 nm (red), and 720 nm (far-red). Based on the above, TaS2 nanosheets were tested as PEC for HER and PD of visible light. The as-exfoliated TaS2-based catalyst, under light illumination, demonstrated high HER catalytic activity with the overpotential for –10 mA·cm−2 below 575 mV, which is ~100 mV lower than during dark mode. Additionally, it was proven that the accurate preparation of the TaS2-based catalyst via sample centrifugation can significantly enhance the catalytic performance of the material. Solution-processed TaS2-enabled PD exhibited broadband light sensitivity in the visible range. The highest photoresponsivity (0.62 mA W−1) and specific detectivity (1.28 × 109 cm Hz1/2 W−1) were observed under blue light illumination, nearly double compared to green, red, and far-red lights. These findings can potentially ignite multiple insights of 2H-TaS2 in photo-induced electrochemistry and (nano)optoelectronics.

## Methods

### Synthesis and crystal growth

TaS2 crystals were grown by the vapor transport method in quartz glass ampoule using iodine as a transport medium. For the synthesis were used tantalum (powder—200 mesh; 99.9%) and sulfur (99.9999%, granules) corresponding to 10 g of TaS2 were placed together with 250 mg of iodine (granules, 99.999%) in quartz glass ampoule (35 × 180 mm) and melt sealed under high vacuum (1 × 10−3 Pa). The ampoule was first heated at 600 °C for 50 h, following 50 h of treatment at 800 °C and 1000 °C. The vapor transport was performed at 1000 °C for the hot zone with the TaS2 charge and 950 °C for the cold crystal growth zone for seven days. Formed hexagonal crystals with sizes up to 20 mm were separated from the ampoule in the glovebox.

### Electrochemical exfoliation

The electrochemical exfoliation of TaS2 crystal was carried out in the organic-based electrolyte composed of 0.01 M TBAPF6 in DMF. The procedure was conducted in the four-port electrochemical cell equipped with a reference (0.1 M Ag/AgNO3 in acetonitrile), counter (platinum), and working (TaS2 crystal) electrodes as well as argon purging. Three different potentials were applied to the working electrode to initiate its wetting at −1 V for 120 s, activate the tetrabutylammonium cations (TBA+) at −2.5 V for 120 s, and finally stimulate cations’ intercalation between TaS2 interlayers at −3.8 V for several hours. As-exfoliated TaS2 precipitate was collected, ultrasonicated for 30 min, and then washed in several polar organic solvents (acetonitrile, methanol, and DMF) using vacuum filtration. The exfoliated TaS2 was filtered through the cellulose membrane, redispersed, and stored in the dried DMF solution.

### Electrocatalytic, photoelectrocatalytic measurements, and characterization

More details on the exploration of electrocatalytic and photoelectrocatalytic measurements and material characterization methods are given in the Supplementary Information file. B.W. was supported from the grant of Specific university research – grant No A2_FCHT_2021_055. P.K.R. was supported by the European Structural and Investment Funds, and CHEMFELLS III' (No.CZ.02.2.69/0.0/0.0/19_074/0014006).