Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2022 Dec 13;7(51):48301–48309. doi: 10.1021/acsomega.2c06480

Synthesis of Rhenium-Doped Molybdenum Sulfide by Atmospheric Pressure Chemical Vapor Deposition (CVD) for a High-Performance Photodetector

Xinke Liu †, Jiangchuan Wang †, Yuheng Lin †, Jie Zhou †, Qiang Liu ‡, Wenjie Yu ‡, Yongqing Cai §, Xiaohua Li †, V Divakar Botcha †, Tingke Rao †, Shuangwu Huang †,*
PMCID: PMC9798488  PMID: 36591144

Abstract

graphic file with name ao2c06480_0007.jpg

Two-dimensional layered materials have attracted tremendous attention as photodetectors due to their fascinating features, including comprehensive coverage of band gaps, high potential in new-generation electronic devices, mechanical flexibility, and sensitive light–mass interaction. Currently, graphene and transition-metal dichalcogenides (TMDCs) are the most attractive active materials for constructing photodetectors. A growing number of emerging TMDCs applied in photodetectors bring up opportunities in the direct band gap independence with thickness. This study demonstrated for the first time a photodetector based on a few-layer RexMo1–xS2, which was grown by chemical vapor deposition (CVD) under atmospheric pressure. The detailed material characterizations were performed using Raman spectroscopy, photoluminescence, and X-ray photoelectron spectroscopy (XPS) on an as-grown few-layer RexMo1–xS2. The results show that both MoS2 and ReS2 peaks appear in the RexMo1–xS2 Raman diagram. RexMo1–xS2 is observed to emit light at a wavelength of 716.8 nm. The electronic band structure of the few layers of RexMo1–xS2 calculated using the first-principles theory suggests that the band gap of RexMo1–xS2 is larger than that of ReS2 and smaller than that of MoS2, which is consistent with the photoluminescence results. The thermal stability of the few layers of RexMo1–xS2 was evaluated using Raman temperature measurements. It is found that the thermal stability of RexMo1–xS2 is close to those of pure ReS2 and MoS2. The fabricated RexMo1–xS2 photodetector shows a high response rate of 7.46 A W–1 under 365 nm illumination, offering a competitive performance to the devices based on TMDCs and graphenes. This study unambiguously distinguishes RexMo1–xS2 as a future candidate in electronics and optoelectronics.

Introduction

Transition-metal dichalcogenides (TMDCs) are a semiconductor material with the general formula of MX2, where M is the transition metal atom of group VI or VII (e. g., Mo, Re, W) and X is the chalcogen atom (e.g., S, Se, Te). Due to their excellent optical and electrical properties, two-dimensional TMDCs are often applied to make a variety of logical devices.1−4 Molybdenum disulfide (MoS2) is the most studied material in TMDCs. MoS2 has a wide range of polymorphs, such as 1T, 2H, and 3R. The 2H structure of MoS2 is thermodynamically stable, while the 3R phase is the metastable phase with the lowest energy. The 1T phase is the most unstable phase of the three types. Both 1T and 3R phases can be transformed into 2H phases at high temperatures.5 As a result, 2H-MoS2 is a stable semiconductor. In the atomically thin monolayer limit, 2H-MoS2 is a direct gap semiconductor with a band gap of about 1.9 eV. The band gap decreases from 1.9 to 0.7 eV for bulk with the increase in the number of MoS2 layers, and MoS2 is converted to an indirect semiconductor. To use MoS2 in the construction of electronic devices, the performance of MoS2 is often improved by adjusting the band gap. In particular, MoS2 is doped/alloyed with other elements, mechanically strained, or stacked in the form of heterostructures to achieve the tunable band gap.6 Among all these approaches, doping/alloying provides an easily scalable route to engineer the band structure.7

The two-dimensional ternary/quaternary materials synthesized by doping or alloying methods can alter the band gap compared with the intrinsic MoS2, manipulating traditional binary materials’ optical and electrical properties.8 In addition to band gap modification, alloying can result in structural and electronic phase transformation. At the same time, alloying can also improve the thermodynamic stability of the material due to the increase in entropy. Current research indicates that Re- and W-doped MoS2 are standard, such as Mo1–xWxS2, Mo1–xRexS2.9,10 Due to the similar structure of WS2 and MoS2, it is easy to form better doping and high alloying. In contrast, ReS2 (1T′) and MoS2 (2H) not only have different properties but also have very different phase structures and atomic structures. The atomic arrangement of the ReS2 layer can be thought of as a distorted 1T arrangement and low in-plane crystal symmetry compared with MoS2. Therefore, ReS2 and MoS2 cannot be alloyed uniformly, resulting in no phase separation, which is not conducive to obtaining intrinsic properties and may degrade performance. Hence, obtaining high-performance RexMo1–xS2 ternary materials is challenging, which are currently less studied.11

In this work, few-layer RexMo1–xS2 films were synthesized by the chemical vapor deposition (CVD) method using a certain proportion of sulfur, MoO3, and ReO3 powders at a high temperature. Then, Raman spectroscopy and X-ray photoelectron spectroscopy (XPS) analysis of the resulting alloy materials were performed. To further study the optoelectronic properties of the synthesized RexMo1–xS2, the structure of the ternary material was simulated, and the change of the band gap was calculated. A photodetector based on RexMo1–xS2 was finally fabricated, which achieved high performance of photoresponse.12 The synthesized material is still in the 2H phase, which has better stability than the 1T phase (MoS2)n cluster model.13 The present study offers a competitive performance to the devices based on TMDCs and graphenes, suggesting RexMo1–xS2 as a future candidate in electronics and optoelectronics.

Experiment Details

Growth of RexMo1–xS2 and Preparation of Photodetectors

The CVD method (SK-G05123K-3-655, Tianjin Zhonghuan Electric Furnace Co., Ltd., China) was used to prepare a few RexMo1–xS2 films on sapphire substrates. Before growth, the sapphire substrate was cleaned with ultrasonication, which included 10 min with acetone (CH3COCH3, 99.5%), 5 min with isopropanol (C3H8O, 99.5%), and deionized water. Sulfur powder (0.7 g) was placed in the alumina combustion boat upstream of the quartz tubular furnace. Molybdenum trioxide powder (MoO3, 99.95%) and rhenium trioxide powder (ReO3, 99.95%) were evenly dispersed in the alumina crucible and placed in the center of the furnace. Then, the cleaned sapphire substrates were placed directly on top of the MoO3 and ReO3 powder alumina crucible boat. After the position was fixed, the gas was pumped by a vacuum pump, and argon gas was introduced. The repeated pumping cycles were carried out to remove the impurity gas in the furnace tube, and the reaction was further continued under optimized conditions in the Ar ambiance. Before the boat was heated, the pressure inside the chamber reached atmospheric pressure by flowing argon gas at a constant flow rate of 120 sccm. The growth temperature of RexMo1–xS2 was then set to 850 °C and the holding time to 10 min. After the RexMo1–xS2 film was grown, UV lithography (Suss MA6) and thermal evaporation coating (ASB-EPI-C6) were used to obtain 10 nm Cr and 50 nm Au on these samples.

Material and Electrical Properties

In this work, X-ray photoelectron spectroscopy (XPS) measurement was conducted using a VG ESCALAB 220i-XL system, which has a monochromatic Al Kα (1486.6 eV) X-ray source and a 20 eV constant pass energy. The C 1s peak value of the core-level binding energy is 284.8 eV, which is used to eliminate the differential charging effect on the sample surface. Raman spectroscopy measurements were carried out in a confocal microscopy setup with a 514 nm solid-state green laser for excitation and a power of 0.15 mW. Transmission electron microscopy (TEM) was performed with America FEI-Titan Cubed Themis G2300, and the TEM information resolution (nonlinearity) was 0.06 nm. A Keithley 4200-SCS semiconductor analyzer was used to test the electrical characteristics of the corresponding photodetectors.

Theoretical Simulation

The first-principles calculations in density functional theory (DFT) were carried out by Cambridge Sequential Total Energy Package (CASTEP) using Generalized Gradient Approximation (GGA) in the Perdew–Burke–Ernzerhof (PBE) function for the structural optimization process.23,30 In the geometry optimization process, the final total energy change was less than 10–6 eV atom–1, the maximum force on each atom in the crystal was less than 0.03 eV nm–1, and the maximum stress of 0.03 eV nm–1 was the convergence criterion. The maximum Hellmann–Feynman force was 0.05 eV Å–1, which was sufficient to obtain a relaxed structure. Brillouin zone sampling adopts a 5 × 5 × 1 Monkhorste–Pack grid. To ensure convergence, the electronic wave function was extended to a plane wave base with a cutoff value of 650 eV. Also, in the energy properties calculation process, the plane wave energy cutoff of 420 eV was used to calculate the energy band structure. For the sampling of the Brillouin zone of the body system, the structure optimization process and the energy band structure were calculated using 4 × 4 × 1 point grids. At the same time, all calculations of structural relaxation were configured with a vacuum layer with a thickness of 20 Å on the c-axis.5,8,14

Results and Discussion

MoS2 is a direct band gap in a single layer and an indirect band gap in a multilayer, while ReS2 is a direct band gap in a single layer or a multilayer and will not change with the number of layers.13 MoS2 is relatively stable in the 2H phase, while ReS2 is relatively stable in the 1T′ phase. In addition to the 2H phase, MoS2 also has a metastable 1T′ phase, it has a zero band gap and superior Weyl semimetal properties but is difficult to prepare. ReS2 is extremely unstable in the 2H phase and has not been prepared yet.4−6,14 Few-layer RexMo1–xS2 nanosheets were synthesized on sapphire substrates in a chemical vapor deposition (CVD) furnace using ReO3 and MoO3 as the source materials (see the Experiment Details section). The schematic diagram of the CVD system is shown in Figure 1a. Weighed sulfur powder (0.7 g) was placed in a ceramic boat and placed in the first temperature zone of the tube furnace, which was upstream of the argon gas flow, and then MoO3 (0.015 g) and ReO3 (0.005 g) powders were put into the ceramic and placed in the second temperature zone of the tube furnace. The sapphire substrate was put on the top of the ceramic boat with the polished surface that needed to be used to deposit the thin film facing downward.15Figure 1b shows the optical image of the large-area continuous RexMo1–xS2 film. The cross-sectional TEM image of the film on the sapphire substrate is shown in Figure 1c, which indicates that the layers were a few layers of a thickness of about ∼2 nm. The thickness of the grown RexMo1–xS2 was measured using an atomic force microscope (AFM), and it is found to be about 1.9 nm. A single layer of two-dimensional material film was about 0.6 nm. The AFM result reveals that the grown RexMo1–xS2 thin film had three layers, which proves that the large-area continuous thin film is multilayered, as described in ref (16). By adjusting the ratio of the precursors and the front and back positions of the precursors, a RexMo1–xS2 film with x from 0 to 1 can be synthesized. As the value of x increases, the phase structure will change from the 2H phase to the 1T′ phase. During the synthesis of RexMo1–xS2, the bottleneck was the structural compatibility of these two monomers, since MoS2 is the 2H phase while ReS2 is the 1T′ phase. Generally speaking, two materials with different phase structures will increase the entropy when forming an alloy, so the synthesis of RexMo1–xS2 requires more energy than the original monomer.17 The low temperature cannot provide sufficient energy to meet the considerable formation energy of alloying required during growth. Fortunately, this problem can be effectively solved by increasing the growth temperature of the RexMo1–xS2 alloy.18 Increasing the growth temperature can provide enough energy to overcome the large energy barrier and form a uniform phase (1T′ or 2H) structure19 and the composition of the RexMo1–xS2 alloy. To further observe the thickness and number of layers of RexMo1–xS2, the TEM image of the RexMo1–xS2 film was taken. As shown in Figure 1c, it can be observed that the thickness of the RexMo1–xS2 film is about 1.9 nm, with three layers in total,20 which is consistent with the results measured by AFM (Supporting Information S1). To further verify Re, Mo, and S in the RexMo1–xS2 film, energy-dispersive X-ray (EDX) mapping was performed. The results are shown in Figure 1d–f. The three elements of Re, Mo, and S shown in Figure 1e–g are purple, red, and gray, respectively. These three elements are not uniform in our synthesized film, and the spatial segregation of various elements can be seen in the figure.21

Figure 1.

Figure 1

Open in a new tab

(a) Diagram of the CVD experimental device used in this work. (b) Optical image of a large-area continuous RexMo1–xS2 thin film. (c) Cross-sectional projection electron microscopy image with a few layers of RexMo1–xS2/sapphire after 1.9 nm (TEM). (d) Thickness of the RexMo1–xS2 film measured by AFM is 1.9 nm. (e–g) Energy-dispersive X-ray (EDX) mapping of the three elements of Re, Mo, and S.

In addition, Raman spectroscopy was performed to characterize the composition-dependent lattice vibration mode of the RexMo1–xS2 film. Figure 2a shows its four different vibration modes. For pure MoS2, the two characteristic peaks of the E2g and A1g modes are located at 382 and 403 cm–1, respectively, while the two characteristic peaks E2g and A1g of pure ReS2 are located at 151 and 213 cm–1, respectively. When RexMo1–xS2 is in the 1T′ phase, it will mainly appear as the peak of ReS2, and when RexMo1–xS2 is in the 2H phase, it will mostly show MoS2 peaks, which are attributed to the E2g and A1g modes of MoS2. In addition, other peaks at 152 and 214 cm–1 indicate the 2H phase of RexMo1–xS2. These two Raman peaks should be due to Re doping activating a new vibration mode caused by the phonon momentum with a large wave vector. The increase of the doping concentration makes the intensity of the Raman peaks of the 2H and 1T′ phase alloys weaker, and the widening of the full width at half-maximum (FWHM) is due to the softening or reforming of phonons caused by the enhancement of the electron–phonon coupling effect. Compared with pure MoS2, the Eg and Ag vibration modes at RexMo1–xS2 move to lower wavenumbers. Eg is shifted to the left by 4 cm–1, and Ag is shifted to the left by 1 cm–1, causing the mode peak separation from the original of 21 cm–1 to increase to 24 cm–1. Compared with pure ReS2, the Eg-like and Ag-like vibration modes at 151 and 213 cm–1 change little, respectively, but the FWHM widens. This is because the main body or most atoms of RexMo1–xS2 are still Re. The broadening is due to alloy scattering or doping of Mo atoms.22 The hardening of the Ag mode of RexMo1–xS2 may be due to changes in the bonding environment of S atoms. Based on the results obtained from the above Raman spectroscopy, the dependence of the phase structure of the prepared alloy material on the composition change is confirmed, and the preliminary analysis of the RexMo1–xS2 gold composition using the Raman test is possible.23 The Raman data of ReS2, RexMo1–xS2, and MoS2 are distinguished, proving that Re and Mo atoms diffuse each other and form bonds and synthesize the RexMo1–xS2 ternary alloy. The photoluminescence of RexMo1–xS2 in Figure 2b indicates adjustable band gaps of 1.77–1.72 eV in these three materials. The peak positions of the pure few-layer ReS2, few-layer RexMo1–xS2, and pure few-layer MoS2 are about 721, 717, and 700 nm, respectively. The calculated band gaps are 1.72, 1.73, and 1.77 eV, respectively. The measured results are consistent with the narrowing of the band gap caused by Re doping achieved by the first-principles calculations.24 The exact composition of the alloy samples was confirmed and their preliminarily electronic structures were preliminarily analyzed based on the XPS data.25 In pure MoS2, Mo 3d core-level binding energy peaks (Figure 2d) are observed at 232.9 and 229.8 eV, corresponding to Mo 3d3/2 and Mo 3d5/2 peaks. After doping Re into 2H-MoS2, Mo 3d3/2 transitions from 232.9 to 233.4 eV and Mo 3d5/2 transitions from 229.8 to 230.2 eV, indicating the upward movement of EF in the conduction band and the n-type doping in the MoS2 crystal. In pure ReS2, Re 4f core-level binding energy peaks were observed at 44.8 and 42.3 eV (Figure 2c), corresponding to the Re 4f5/2 and Re 4f7/2 peaks. It is worth noting that after doping Mo into 1T′-ReS2, Re 4f5/2 changed from 44.8 to 43.6 eV and Re 4f7/2 changed from 42.3 to 41.1 eV, indicating that EF moved down to the valence band and the p-type-doped crystals in ReS2.26 This doping-induced binding energy shift is also reflected in the S 2p core-level binding energy peak (Figure 2e), where obvious mutations and phase transitions are observed. As shown in Figure 1a, the downward shift of EF of the 1T′ alloy and the upward shift of the 2H alloy are because the d-orbital of the Re element has one more valence electron than the Mo element.27 In addition, the movement of EF indicates that the effective modulation and doping effect of the electronic structure of these alloys are atomic substitution rather than physical adsorption.28

Figure 2.

Figure 2

Open in a new tab

(a) Comparison of Raman signals of few layers of MoS2, Re0.05Mo0.95S2, and ReS2. (b) Comparison of photoluminescence (PL) signals of a few layers of MoS2, Re0.05Mo0.95S2, and ReS2. (c–e) Core-level XPS spectrum of Re 4f, Mo 3d, and S 2p of the multilayer Re0.05Mo0.95S2/sapphire sample.

Figure 3a shows the structure diagram of the three materials. It can be seen that RexMo1–xS2 is different from the original atomic arrangement. Here, the first principles in density functional theory (DFT) were used to study the three-layer MoS2, the three-layer RexMo1–xS2, and the three-layer ReS2, so the total energy program is calculated from scratch using Cambridge Sequential Total Energy Package (CASTEP).20,29 The direct gap between the three-layer MoS2, the three-layer RexMo1–xS2, and ReS2 has been pointed out. The calculated result was the same as the trend shown in Figure 2b, which shows that due to the doping of elements, the band gap adjustment of the semiconductor film can be realized, proving the present calculation’s accuracy.5

Figure 3.

Figure 3

Open in a new tab

(a) Illustration of the three-dimensional (3D) drawing (purple symbol: W atoms; cyan symbol: Mo atoms; yellow symbol: S atoms) of ReS2, RexMo1–xS2, and MoS2 layers, in which Mo and Re elements are spatially segregated. Electronic band structure of (b) three-layer ReS2, (c) three-layer Re0.05Mo0.95S2, and (d) three-layer MoS2.

Figure 4a–c shows the effect of temperature changes on the Eg and Ag Raman modes of the few layers of ReS2, RexMo1–xS2, and MoS2. These three materials were measured in the range of 300–500 K with a 325 nm excitation laser based on the Grüneisen model to fit the relationship between the peak position of the Eg and Ag modes and the temperature

graphic file with name ao2c06480_m001.jpg 1

where ω0 is the temperature at the peak position of the Raman mode at 0 K and x is the first-order temperature of the same mode coefficient.6 The slope of the fitted line gives the first-order temperature coefficient of the specific Raman mode. As shown in Figure 4d, the first-order temperature coefficient x of the A1g and E2g1 modes of pure ReS2 are −0.01278 and −0.00784 cm–1 K–1, respectively. As shown in Figure 4e, the first-order temperature coefficient x of ReS2-like Eg and ReS2-like Ag modes of RexMo1–xS2 are −0.01176 and −0.01001 cm–1 K–1, respectively. The first-order temperature coefficient x of MoS2-like Eg and MoS2-like Ag modes of RexMo1–xS2 are −0.01496 and −0.02116 cm–1 K–1, respectively. As shown in Figure 4f, the first-order temperature coefficient x of pure MoS2 in A1g and E2g modes is −0.01139 and −0.01198 cm–1 K–1, respectively. In contrast, nondoped compounds have higher thermal stability due to the greater free and internal energy of the ternary RexMo1–xS2 alloy. The opaque photodetector array with Cr/Au as the electrode was fabricated on Re0.05Mo0.95S2/sapphire (1 × 1 cm2) by semiconductor processing technology, and the MoS2/sapphire photodetector used as a control was fabricated using the same process.

Figure 4.

Figure 4

Open in a new tab

Temperature-dependent Raman spectra. (a) Sample ReS2, (b) Re0.05Mo0.95S2, and (c) MoS2. The Raman shift of each mode as a function for temperature (d–f), including the value of χ obtained from linear fitting.

The 3D schematic diagram of the Re0.05Mo0.95S2/sapphire photodetector is shown in Figure 5a, where the thickness of the electrode (Cr/Au) on the device is 10/50 nm, and the effective area of each device array is 47 376 μm2. Meanwhile, Figure 5a shows the device structure of the Re0.05Mo0.95S2/sapphire Schottky junction-based UV photodetector. Using the above method, a two-dimensional (2D) Re0.05Mo0.95S2 layer was in situ van der Waals epitaxy grown on a sapphire substrate to form a Schottky junction, which is beneficial to constructing high-quality Schottky junction and effectively reducing the interface defects introduced by the wet transfer process. Figure 5b shows the optical picture of a single device. Figure 5c plots the current–voltage (I–V) curves of the Schottky junction in darkness and 365 nm light. The device exhibits prominent photovoltaic characteristics, enabling the photodetector to operate in the self-driven mode. The significant enhancement of reverse current under UV irradiation can be attributed to two aspects: (i) the reduction of the Schottky barrier under UV irradiation will result in a large photocurrent under reverse bias and (ii) the multiplier gain may be caused by carrier scattering in the Re0.05Mo0.95S2 layer. Figure 5c shows the dark current (Idark) and photocurrent (Ilight) of the two devices (Re0.05Mo0.95S2 and MoS2) as a function of the bias voltage under 365 nm laser irradiation. Idark is the current generated by the device under the bias voltage when there is no incident laser and Ilight is the current generated by the device under the bias voltage and incident laser irradiation.31 The dark current of the Re0.05Mo0.95S2 device is 128 nA at 10 V, while the dark current of the MoS2 device is only 29.7 nA, which may be attributed to the smaller band gap and the lower potential barrier of the Re0.05Mo0.95S2 device due to the doping of the Re element. At the same time, under 10 V and 365 nm (5.647 μW) illumination, the photocurrent of the device increased from 2734 nA (MoS2) to 42 288 nA (Re0.05Mo0.95S2), and the Ilight/Idark ratio of the Re0.05Mo0.95S2 device increased by 3.62 times. This can be explained by the fact that the light scattering caused by the Re0.05Mo0.95S2 structure is stronger than that of MoS2, and the band gap is reduced due to local strain, which leads to an increase in the mobility of photogenerated carriers. Further observation of the improvement of the device through the performance of the Re0.05Mo0.95S2 structure, the response time of two photodetectors (MoS2 and Re0.05Mo0.95S2) was also studied. Figure 5d shows that the photocurrent of the two devices periodically changes as the 365 nm laser was turned on/off at a voltage of 10 V, whereas the 365 nm (5.647 μW) laser was turned on/off every 20 s. Both devices show stable switching characteristics, and the photocurrent observed in the Re0.05Mo0.95S2/sapphire photodetector is more significant than that in the MoS2/sapphire photodetector. Figure 5d defines the rise time (trise) and the fall time (tfall) of the two devices under 365 nm laser and 4 V bias irradiation. Here, the rise time of the device can be obtained by calculating the time required for the photocurrent to increase from 10% of the maximum photocurrent to 90% under illumination. The fall time is defined as the photocurrent falling from 10% of the maximum photocurrent to 90%.

Figure 5.

Figure 5

Open in a new tab

(a) 3D schematic diagram of the RexMo1–xS2/sapphire photodetector, where the electrodes are Cr/Au. (b) Optical micrograph of the RexMo1–xS2/sapphire photodetector. (c) Optical response characteristics of the RexMo1–xS2/sapphire and the MoS2/sapphire photodetector as a function of voltage under 365 nm laser irradiation. Response time of RexMo1–xS2/sapphire and MoS2/sapphire photodetectors under 10 V illumination and 365 nm laser irradiation, photocurrent (Iph), responsivity (R), and noise equivalent power (NEP) at these three wavelengths. The noise-specific detection rates (D*) below correspond to (d)–(f).

Comparing the two devices, it can be observed that the rise time is reduced from 6 s (MoS2) to 3.7 s (Re0.05Mo0.95S2), and the fall time is decreased from 6.8 s (MoS2) to 5.7 s (Re0.05Mo0.95S2). The shortened rise and fall times may be due to a decrease in effective electron quality and an increase in electron mobility. In addition, the photocurrent (Iph) is one of the parameters reflecting the light response of the photodetector, which can be calculated using the below formula

graphic file with name ao2c06480_m002.jpg 2

where Ilight and Idark have been defined in Figure 5c.29 The responsivity (R) is a significant indicator to measure the photoelectric conversion characteristics and photoelectric conversion spectrum of a photodetector and can be calculated by the below formula

graphic file with name ao2c06480_m003.jpg 3

where Iph is the photocurrent and S and Pi are the active area of the film and the power density of the incident laser, respectively.32,33Figure 5e shows the relationship between the Iph and R of the photodetector based on MoS2 and Re0.05Mo0.95S2 at the incident wavelength of 365 nm light and the different incident power at 10 V voltage. It can be found that under the irradiation of three wavelengths, the Re0.05Mo0.95S2 device shows a higher photocurrent and responsivity than the MoS2 device. The maximum responsivity of the device under a 365 nm incident laser was increased from 0.85 A W–1 (MoS2) to 7.46 A W–1 (Re0.05Mo0.95S2). In particular, the relationship between responsivity and different incident powers is studied. The responsivity reaches a peak first and then decreases as the incident power increases.

This may be due to the saturation of trap states at the interface, as well as the low recombination frequency and long carrier lifetime of low-power lasers. To further observe the improvement of the device performance by the Re0.05Mo0.95S2 material, the noise equivalent power (NEP) and specific detection rate (D*) at different incident powers were calculated. NEP can quantitatively calculate the signal-to-noise ratio, defined as the excitation power required to generate a signal equal to the noise level in a 1 Hz bandwidth. It can be calculated using the below formula

graphic file with name ao2c06480_m004.jpg 4

where Idark and R are the dark current and responsivity, respectively, and Δf is the bandwidth (Δf is defined as 1 in this article).20 The dark current excitation noise is the main contribution to the total noise of the photodetector. D* can be calculated using the below formula

graphic file with name ao2c06480_m005.jpg 5

where A is the active area of the MoS2 film under the incident laser irradiation. Figure 5f contains the relationship between the NEP and D* of the photodetector based on MoS2 and Re0.05Mo0.95S2 and the different incident power at a wavelength of 365 nm and a voltage of 10 V.28 As it is well-known, the larger the D* value, the smaller the NEP value can be, which means that the noise of the device is lower and the detection capability is stronger. In contrast, the decrease in NEP decrease and increase in D* of the Re0.05Mo0.95S2 photodetector under the irradiation of three different incident lasers indicate that the Re0.05Mo0.95S2 material can significantly improve the photodetection compared with the weak growth in the dark current. Table 1 summarizes some of the reported performance parameters of photodetectors, showing the photocurrent of the detector, which has a stronger detection ability.

Table 1. Comparison of the Critical Parameters of Our Device to the Reported 2D Material-Based Photodetectors.

photodetector R[A/W] D* [Jones] response time [s] refs
RexMo1–xS2/sapphire 7.46 8.0 × 109 3.70 this work
MoS2/sapphire 0.85 1.9 × 109 6.0 this work
MoS2/glass 0.032 1.0 × 109   (34)
MoS2/SiO2 0.048 1.3 × 107   (35)
MoS2/SiC 0.25 5.6 × 108 2.19 (36)
ReS2(1–x)Se2x/SiO2 0.25   0.015 (37)
MoS2/PSS 0.032 1.2 × 1010 2.24 (38)
MoS2/PGS 0.25 5.6 × 108 2.19 (39)
MoS2/sapphire 0.071   0.50 (40)
MoS2(1–x)Se2x/SiO2 2.5 × 10–4   0.038 (41)
Open in a new tab

Conclusions

In conclusion, this study has demonstrated that the few-layer RexMo1–xS2 films were synthesized by the CVD method at 1 atm pressure and that the high-performance photodetector has been successfully constructed on the as-prepared RexMo1–xS2 on a sapphire substrate.42 The Raman, PL, and XPS measurements have been performed on the as-grown RexMo1–xS2 few layers to understand film quality. The results show that both MoS2 and ReS2 peaks appear in the RexMo1–xS2 Raman diagram. RexMo1–xS2 is observed to emit light at a wavelength of 716.8 nm. The electronic band structure of the few-layer RexMo1–xS2 is calculated using the first-principles theory, demonstrating that the band gap of RexMo1–xS2 is larger than that of ReS2 and smaller than that of MoS2, which is consistent with the photoluminescence results. Temperature-dependent Raman spectroscopy on ReS2, Re0.05Mo0.95S2, and MoS2 layers have been carried out for the purpose of comparative study, and further, the thermal stability of the few layers of RexMo1–xS2 has been evaluated. It was found that the thermal stability of RexMo1–xS2 is close to that of pure ReS2 and MoS2. The manufactured RexMo1–xS2 photodetector shows superiority to the pure MoS2 detector, with a higher responsivity of 7.46 A W–1. Our results provide a positive prospect for further improving the performance of the few-layer Re0.05Mo0.95S2 photodetector.

Acknowledgments

This work was supported by the National Key Research and Development Program of China (2017YFB0403000), the National Natural Science Foundation of China (61974144), the Guangdong Province Key Research and Development Plan (2020B010174003), the Science and Technology Foundation of Shenzhen (JSGG20191129114216474), the Science and Technology Project of Shenzhen City (JSGG20210802154213040 and JSGG20201102152403008), and the Project of Innovation and Strong School (PT2020C002). The authors wish to acknowledge the assistance on HRTEM observation received from the Electron Microscope Center of Shenzhen University and the device fabrication process in the Photonics Center of Shenzhen University. The sulfur powder was provided by Xintai Wenhe Chemical Pte Ltd. (China, Shandong). The CVD system (CVD-12IIH-3Z/G) was provided by Tianjin Zhonghuan Furnace Corp. The Open Project of State Key Laboratory of Functional Materials for Informatic.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c06480.

  • AFM image of RexMo1–xS2 (PDF)

The authors declare no competing financial interest.

Supplementary Material

ao2c06480_si_001.pdf (245.7KB, pdf)

References

  1. Du H.; Zhang Q. P.; Zhao B.; Marken F.; Gao Q. C.; Lu H.-X.; Guan L.; Wang H.-L.; Shao G.; Xu H.-L.; Zhang R.; Fan B.-B. Novel hierarchical structure of MoS2/TiO2/Ti3C2Tx composites for dramatically enhanced electromagnetic absorbing properties. J. Adv. Ceram. 2021, 10, 1042–1051. 10.1007/s40145-021-0487-9. [DOI] [Google Scholar]
  2. Wu M.; He Y.; Wang L.-B.; Xia Q.-X.; Zhou A.-G. Synthesis and electrochemical properties of V2C MXene by etching in opened/closed environments. J. Adv. Ceram. 2020, 9, 749–758. 10.1007/s40145-020-0411-8. [DOI] [Google Scholar]
  3. Xu Q.; Zhou Y.-C.; Zhang H.-M.; Jiang A.-N.; Tao Q.-Z.; Lu J.; Rosen J.; Niu Y.-H.; Grasso S.; Hu C.-F. Theoretical prediction, synthesis, and crystal structure determination of new MAX phase compound V2SnC. J. Adv. Ceram. 2020, 9, 481–492. 10.1007/s40145-020-0391-8. [DOI] [Google Scholar]
  4. Gao J.; Kim Y.-D.; Liang L.-B.; Idrobo J.-C.; Chow P.; Tan J.-W.; Li B.-C.; Li L.; Sumpter B.-G.; Lu T.-M.; Meunier V.; Hone J.; Koratkar N. Transition-Metal Substitution Doping in Synthetic Atomically Thin Semiconductors. Adv. Mater. 2016, 28, 9735–9743. 10.1002/adma.201601104. [DOI] [PubMed] [Google Scholar]
  5. Deng Q.-X.; Li X.-B.; Si H.-Y.; Hong J.-H.; Wang S.-Y.; Feng Q.-L.; Hu C.-X.; Wang S.-S.; Zhang H.-L.; Suenaga K.; Xu H. Strong Band Bowing Effects and Distinctive Optoelectronic Properties of 2H and 1T ’ Phase-Tunable Mo(x)Re(1-)(x)S(2)Alloys. Adv. Funct. Mater. 2020, 30, 2003264 10.1002/adfm.202003264. [DOI] [Google Scholar]
  6. Ding Y.-H.; Cheng Z.; Zhu X.-L.; Yvind K.; Dong J.-J.; Galili M.; Hu H.; Mortensen N.-A.; Xiao S.-S.; Oxenlowe L.-K. Ultra-compact integrated graphene plasmonic photodetector with bandwidth above 110 GHz. Nanophotonics 2020, 9, 317–325. 10.1515/nanoph-2019-0167. [DOI] [Google Scholar]
  7. Hafeez M.; Gan L.; Li H.-Q.; Ma Y.; Zhai T.-Y. Large-Area Bilayer ReS2 Film/Multilayer ReS2 Flakes Synthesized by Chemical Vapor Deposition for High Performance Photodetectors. Adv. Funct. Mater. 2017, 26, 4551–4560. 10.1002/adfm.201601019. [DOI] [Google Scholar]
  8. Liang Q.-J.; Wang Q.-X.; Zhang Q.; Wei J.-X.; Lim S.; Zhu R.; Hu J.-X.; Wei W.; Lee C.; Sow C.; Zhang W.-J.; Wee A. High-Performance, Room Temperature, Ultra-Broadband Photodetectors Based on Air-Stable PdSe2. Adv. Mater. 2019, 31, 1807609 10.1002/adma.201807609. [DOI] [PubMed] [Google Scholar]
  9. Chow P. C. Y.; Someya T. Organic Photodetectors for Next-Generation Wearable Electronics. Adv. Mater. 2020, 32, 1902045 10.1002/adma.201902045. [DOI] [PubMed] [Google Scholar]
  10. Hu P.-A.; Wen Z.-Z.; Wang L.-F.; Tan P.-H.; Xiao K. Synthesis of Few-Layer GaSe Nanosheets for High Performance Photodetectors. ACS Nano 2012, 6, 5988–5994. 10.1021/nn300889c. [DOI] [PubMed] [Google Scholar]
  11. Li X.-B.; Cui F.-F.; Feng Q.-L.; Wang G.; Xu X.-S.; Wu J.-X.; Mao N.-N.; Liang X.; Zhang Z.-Y.; Zhang J.; Xu H. Controlled growth of large-area anisotropic ReS2 atomic layer and its photodetector application. Nanoscale 2016, 8, 18956–18962. 10.1039/C6NR07233J. [DOI] [PubMed] [Google Scholar]
  12. Liu X.-K.; Wu J.; Yu W.-J.; Chen L.; Huang Z.-H.; Jiang H.; He J.-Z.; Liu Q.; Lu Y.-M.; Zhu D.-L.; Liu W.-J.; Cao P.-J.; Han S.; Xiong X.-B.; Xu W.-Y.; Ao J.-P.; Ang K.-W.; He Z.-B. Monolayer WxMo1-xS2 Grown by Atmospheric Pressure Chemical Vapor Deposition: Bandgap Engineering and Field Effect Transistors. Adv. Funct. Mater. 2017, 27, 1606469 10.1002/adfm.201606469. [DOI] [Google Scholar]
  13. Wang Y.-Y.; Deng J.-J.; Wang X.; Che J.-T.; Ding X.-L. Small stoichiometric (MoS2)n clusters with the 1T phase. Phys. Chem. Chem. Phys. 2018, 20, 6365–6373. 10.1039/C7CP07914A. [DOI] [PubMed] [Google Scholar]
  14. Zhang K.-H.; Bersch B.-M.; Joshi J.; Addou R.; Cormier C.-R.; Zhang C.-X.; Xu K.; Briggs N.-C.; Wang K.; Subramanian S.; Cho K.; Fullerton-Shirey S.; Wallace R. M.; Vora P. M.; Robinson J. A. Tuning the Electronic and Photonic Properties of Monolayer MoS2 via In Situ Rhenium Substitutional Doping. Adv. Funct. Mater. 2018, 28, 1706950 10.1002/adfm.201706950. [DOI] [Google Scholar]
  15. Butanovs E.; Kuzmin A.; Piskunov S.; Smits K.; Kalinko A.; Polyakov B. Synthesis and characterization of GaN/ReS2, ZnS/ReS2 and ZnO/ReS2 core/shell nanowire heterostructures. Appl. Surf. Sci. 2021, 536, 147841 10.1016/j.apsusc.2020.147841. [DOI] [Google Scholar]
  16. Cui F.-F.; Li X.-B.; Feng Q.-L.; Yin J.-B.; Zhou L.; Liu D.-Y.; Liu K.-Q.; He X.-X.; Liang X.; Liu S.-Z.; Lei Z.-B.; Liu Z.-H.; Peng H.-L.; Zhang J.; Kong J.; Xu H. Epitaxial growth of large-area and highly crystalline anisotropic ReSe2 atomic layer. Nano Res. 2017, 10, 2732–274. 10.1007/s12274-017-1477-7. [DOI] [Google Scholar]
  17. Dolui K.; Rungger I.; Das Pemmaraju C.; Sanvito S. Possible doping strategies for MoS2 monolayers: An ab initio study. Phys. Rev. B 2013, 88, 075420 10.1103/PhysRevB.88.075420. [DOI] [Google Scholar]
  18. Kochat V.; Apte A.; Hachtel J.-A.; Kumazoe H.; Krishnamoorthy A.; Susarla S.; Idrobo J.-C.; Shimojo F.; Vashishta P.; Kalia R.; Nakano A.; Tiwary C.-S.; Ajayan P.-M. Re Doping in 2D Transition Metal Dichalcogenides as a New Route to Tailor Structural Phases and Induced Magnetism. Adv. Mater. 2017, 29, 1703754 10.1002/adma.201703754. [DOI] [PubMed] [Google Scholar]
  19. Lin J.-H.; Zhou J.-D.; Zuluaga S.; Yu P.; Gu M.; Liu Z.; Pantelides S.-T.; Suenaga K. Anisotropic Ordering in 1T ’ Molybdenum and Tungsten Ditelluride Layers Alloyed with Sulfur and Selenium. ACS Nano 2018, 12, 894–901. 10.1021/acsnano.7b08782. [DOI] [PubMed] [Google Scholar]
  20. Lin Y.-C.; Dumcenco D.-O.; Komsa H.-P.; Niimi Y.; Krasheninnikov A.-V.; Huang Y.-S.; Suenaga K. Properties of Individual Dopant Atoms in Single-Layer MoS2: Atomic Structure, Migration, and Enhanced Reactivity. Adv. Mater. 2014, 26, 2857–2861. 10.1002/adma.201304985. [DOI] [PubMed] [Google Scholar]
  21. Loh L.-Y.; Chen Y.-F.; Wang J.-Y.; Yin X.-M.; Tang C.-S.; Zhang Q.; Watanabe K.; Taniguchi T.; Wee A.; Bosman M.; Quek S.-Y.; Eda G. Impurity-Induced Emission in Re-Doped WS2 Monolayers. Nano Lett. 2021, 21, 5293–5300. 10.1021/acs.nanolett.1c01439. [DOI] [PubMed] [Google Scholar]
  22. Wang Z.-X.; Zhao X.-X.; Yang Y.-K.; Qiao L.; Lv L.; Chen Z.; Di Z.-F.; Ren W.; Pennycook S.-J.; Zhou J.-D.; Gao Y.-F. Phase-Controlled Synthesis of Monolayer W1-xRexS2 Alloy with Improved Photoresponse Performance. Small 2020, 16, 2000852 10.1002/smll.202000852. [DOI] [PubMed] [Google Scholar]
  23. Miao J.-L.; Zhang F.-J. Recent Progress on Photomultiplication Type Organic Photodetectors. Laser Photonics Rev. 2019, 13, 1800204 10.1002/lpor.201800204. [DOI] [Google Scholar]
  24. Gu H.; Lu Y.-M.; Zhu D.-L.; Li K.-L.; Zheng S.-N.; Wang J.-F.; Ang K.-W.; Xu K.; Liu X.-K. High temperature study on the thermal properties of few-layer Mo0.5W0.5S2 and effects of capping layers. Results Phys. 2017, 7, 4394–4397. 10.1016/j.rinp.2017.11.017. [DOI] [Google Scholar]
  25. Qiao H.; Yuan J.; Xu Z.-Q.; Chen C.-Y.; Lin S.-H.; Wang Y.-S.; Song J.-C.; Liu Y.; Khan Q.; Hoh H.-Y.; Pan C.-X.; Li S.-J.; Bao Q.-L. Broadband Photodetectors Based on Graphene-Bi2Te3 Heterostructure. ACS Nano 2015, 9, 1886–1894. 10.1021/nn506920z. [DOI] [PubMed] [Google Scholar]
  26. Cai S.; Xu X.-J.; Yang W.; Chen J.-X.; Fang X.-S. Materials and Designs for Wearable Photodetectors. Adv. Mater. 2019, 31, 1808138 10.1002/adma.201808138. [DOI] [PubMed] [Google Scholar]
  27. Dalmatova S.-A.; Fedorenko A.-D.; Mazalov L.-N.; Asanov I.-P.; Ledneva A.-Y.; Tarasenko M. S.; Enyashin A. N.; Zaikovskii V. I.; Fedorov V. E. XPS experimental and DFT investigations on solid solutions of Mo1-xRexS2 (0 <x <0.20). Nanoscale 2018, 10, 10232–10240. 10.1039/C8NR01661E. [DOI] [PubMed] [Google Scholar]
  28. Yang W.; Chen J.-X.; Zhang Y.; Zhang Y.-J.; He J.-H.; Fang X.-S. Silicon-Compatible Photodetectors: Trends to Monolithically Integrate Photosensors with Chip Technology. Adv. Funct. Mater. 2019, 29, 1808182 10.1002/adfm.201808182. [DOI] [Google Scholar]
  29. Yang S.-Z.; Gong Y.-J.; Manchanda P.; Zhang Y.-Y.; Ye G.-L.; Chen S.-M.; Song L.; Pantelides S.-T.; Ajayan P.-M.; Chisholm M.-F.; Zhou W. Rhenium-Doped and Stabilized MoS2 Atomic Layers with Basal-Plane Catalytic Activity. Adv. Mater. 2018, 30, 1803477 10.1002/adma.201803477. [DOI] [PubMed] [Google Scholar]
  30. Yang D.-Z.; Ma D.-G. Development of Organic Semiconductor Photodetectors: From Mechanism to Applications. Adv. Opt. Mater. 2019, 7, 1800522 10.1002/adom.201800522. [DOI] [Google Scholar]
  31. Wang Y.; Song L.; Chen Y.; Huang W. Emerging New-Generation Photodetectors Based on Low-Dimensional Halide Perovskites. ACS Photonics 2020, 7, 10–28. 10.1021/acsphotonics.9b01233. [DOI] [Google Scholar]
  32. Qiu Q.-X.; Huang Z.-M. Photodetectors of 2D Materials from Ultraviolet to Terahertz Waves. Adv. Mater. 2021, 33, 2008126 10.1002/adma.202008126. [DOI] [PubMed] [Google Scholar]
  33. Furchi M.; Urich A.; Pospischil A.; Lilley G.; Unterrainer K.; Detz H.; Klang P.; Andrews A.-M.; Schrenk W.; Strasser G.; Mueller T. Microcavity-Integrated Graphene Photodetector. Nano Lett. 2012, 12, 2773–2777. 10.1021/nl204512x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Gu H.; Chen L.; Lu Y.-M.; Tian F.-F.; Zhang Z.-Q.; Xu K.; Wu J.; Botcha V. D.; Li K.-L.; Liu X.-K. Low-temperature study of neutral and charged excitons in the large-area monolayer WS2. Jpn. J. Appl. Phys. 2018, 57, 060309 10.7567/JJAP.57.060309. [DOI] [Google Scholar]
  35. Xie Y.; Zhang B.; Wang S.-X.; Wang D.; Wang A.-Z.; Wang Z.-Y.; Yu H.-H.; Zhang H.-J.; Chen Y.-X.; Zhao M.-W.; Huang B.-B.; Mei L.-M.; Wang J.-Y. Ultrabroadband MoS2 Photodetector with Spectral Response from 445 to 2717 nm. Adv. Mater. 2017, 29, 1605972 10.1002/adma.201605972. [DOI] [PubMed] [Google Scholar]
  36. Huang Z.-Z.; Zhang T.-F.; Liu J.-K.; Zhang L.-H.; Jin Y.-H.; Wang J.-P.; Jiang K.-L.; Fan S.-S.; Li Q.-Q. Amorphous MoS2 Photodetector with Ultra-Broadband Response. ACS Appl. Electron. Mater. 2019, 1, 1314–1321. 10.1021/acsaelm.9b00247. [DOI] [Google Scholar]
  37. Xiao Y.-F.; Min L.; Liu X.-K.; Liu W.-J.; Younis U.; Peng T.-H.; Kang X.-W.; Wu X.-H.; Ding S.-J.; Zhang D.-W. Facile integration of MoS2/SiC photodetector by direct chemical vapor deposition. Nanophotonics 2020, 9, 3035–3044. 10.1515/nanoph-2019-0562. [DOI] [Google Scholar]
  38. Kang P.-P.; Nan H.-Y.; Zhang X.-M.; Mo H.-X.; Ni Z.-H.; Gu X.-F.; Ostrikov K.; Xiao S.-Q. Controllable Synthesis of Crystalline ReS2(1-x)Se2x Monolayers on Amorphous SiO2/Si Substrates with Fast Photoresponse. Adv. Opt. Mater. 2020, 8, 1901415 10.1002/adom.201901415. [DOI] [Google Scholar]
  39. Liu X.-K.; Hu S.-Q.; Luo J.-L.; Li X.-H.; Wu J.; Chi D.-Z.; Ang K.-W.; Yu W.-J.; Cai Y.-Q. Suspended MoS2 Photodetector Using Patterned Sapphire Substrate. Small 2021, 17, 2100246 10.1002/smll.202100246. [DOI] [PubMed] [Google Scholar]
  40. Liu X.-K.; Hu S.-Q.; Lin Z.-C.; Li X.-H.; Song L.-J.; Yu W.-J.; Wang Q.; He W. High-Performance MoS2 Photodetectors Prepared Using a Patterned Gallium Nitride Substrate. ACS Appl. Mater. Interface 2021, 13, 15820–15826. 10.1021/acsami.0c22799. [DOI] [PubMed] [Google Scholar]
  41. Cho B.; Kim A.-R.; Park Y.; Yoon J.; Lee Y.-J.; Lee S.; Yoo T.-J.; Kang C.-G.; Lee B.-H.; Ko H.-C.; Kim D.-H.; Hahm M.-G. Bifunctional Sensing Characteristics of Chemical Vapor Deposition Synthesized Atomic-Layered MoS2. ACS Appl. Mater. Interface 2015, 7, 2952–2959. 10.1021/am508535x. [DOI] [PubMed] [Google Scholar]
  42. Zhang J.-M.; Qian Y.-Z.; Nan H.-Y.; Gu X.-F.; Xiao S.-Q. Large-scale MoS2(1-x)Se2x monolayers synthesized by confined-space CVD. Nanotechnology 2021, 32, 355601. 10.1088/1361-6528/ac0026. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

ao2c06480_si_001.pdf (245.7KB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES