Graphene-like MoS2 Nanosheets on Carbon Fabrics as High-Performance Binder-free Electrodes for Supercapacitors and Li-Ion Batteries.

Two-dimensional layer-structure materials are now of great interest in energy storage devices, owing to their graphene-like structure and high theoretical capacity. Herein, graphene-like molybdenum disulfide (MoS2) nanosheets were uniformly grown on carbon fabrics by using a hydrothermal method. They were evaluated as binder-free electrodes for Li-ion batteries (LIBs) and supercapacitors. As expected, long cycling life and high capacity/capacitance are achieved. When used as self-standing electrodes for LIBs, they deliver a high area capacity of ∼0.5 mAh/cm2 even after 400 cycles and remarkable rate capability in the charge/discharge potential range of 1-3 V. In addition, a three-dimensional integrated electrode of the MoS2 nanosheet exhibits a high capacitance of 103.5 mF/cm2 and long cycling stability up to at least 15 000 cycles at a current density of 3 mA/cm2 for supercapacitors. The great cycling stability of MoS2 in supercapacitors is promising in the enhancement of cycling stability through their integration with other materials as alternatives to graphene in some special fields.

Introduction

Nowadays, energy storage devices have attracted worldwide attention due to their vital roles as dominant mobile power sources for a range of applications, such as mobile devices, new-energy vehicles, and smart devices.[1−4] To develop high-performance electrode materials for lithium-ion batteries (LIBs) and supercapacitors (SCs), metal oxide nanostructures with high specific capacity/capacitance, typically 2–3 times higher than that of the carbonous materials, have attracted much attention.[5−8] However, their poor cycling stability leads to worse electrochemical performance which cannot satisfy the practical applications. Therefore, developing electrode materials with long cycle lives is still an urgent mission.

Since graphene was discovered in 2004, many other two-dimensional (2D) layer-structure materials have also regained research interest.[9−15] So far, among those two-dimensional (2D) layer-structure materials, molybdenum disulfide (MoS2) has played a significant role, which has been extensively studied for many applications such as photodetectors, water treatments, memorizers, and especially for SCs and LIBs.[16−19] However, when evaluating as an electrode material for Li-ion batteries, due to the large volume change during the charging/discharging cycles, the conductivity is low and the structure is destroyed, resulting in faster capacity decay. To resolve this problem, a feasible way is to composite with other materials which have excellent electronic conductivity, such as graphene, carbon nanotubes, etc.[20−23] Another approach is to increase the interlayer distance of MoS2, which will also introduce voids and defects into MoS2 to increase its lithium ion storage capacity. The third way is to design binder-free electrodes.

Compared to traditional binder-enriched materials, the self-standing electrode does not have any additional current collector or binder, which leads to high-speed electron transport and ion diffusion. Therefore, the binder-free electrode can effectively enhance the electrochemical performance of active materials. In recent years, specific mechanical and electronic properties of three-dimensional (3D) flexible binder-free energy storage devices have attracted great attention for the next generation, such as LIBs and SCs.[24−27] For instance, the synthesis of graded 3D ZnCo2O4 nanowire array/carbonless cloth binder LIB integrated electrode is reported with high reversible capacity and good cycling performance.[28] Cheng et al. also developed an LTO/graphene foam self-standing electrode with excellent electrochemical properties.[29] SCs and LIBs have many common features in improving the electrochemical properties of active materials despite their different charge storage mechanisms. The rationally designed binder-free electrodes are among them. Very recently, the electrodes based on NiCo2O4 nanowire arrays on carbon textiles synthesized by Zhang et al. exhibited high reversible capacity/capacitance and excellent cycling ability for LIBs and SCs.[30] Oxygen-deficient Fe2O3 nanorods on carbon fabrics were synthesized by Lu et al., and high capacitance and excellent cycling stability are achieved when used as anodes for SCs.[31]

In previous reports, many kinds of composites based on MoS2 were synthesized to enhance the cycling stability of MoS2. Recently, to get a more excellent anode material for LIBs, hierarchical MoS2 nanosheets/activated carbon fiber cloth and three-dimensional hierarchical MoS2 nano-array/carbon cloth were synthesized and evaluated. However, the galvanostatic charging and discharging tests were performed at a potential ranging from 5 mV to 3 V and 1 mV to 3 V, respectively, and the cycling performance is not so good. Moreover, when the charge/discharge voltage was below 1 V, the carbon cloth and active carbon fiber can contribute a great deal of the capacity to the whole electrode, which results in the calculated capacity of active materials higher than real.[32−34] At the same time, MoS2-based SCs are mainly binder-enriched. According to previous reports, MoS2 can be widely used in almost all commonly used electrolytes, because of the relative stability of MoS2 in acid, neutral, and alkaline aqueous solutions. When used in alkaline aqueous solution, some researchers choose Ni foam as the current collector. However, the Ni foam can easily form NiO on the surface and has a great influence on the measurement of capacitor properties, which may result in error and exaggerated capacitance, especially when a little amount of electrode-active materials are used for the measurements.[35−37] To avoid the above problem, carbon fabrics used as the current collector of the SC electrode may be a good choice, which can reflect the SC properties of active materials more accurately.

In this work, graphene-like MoS2 nanosheets on carbon fabrics are successfully synthesized in a typical and easy hydrothermal method and directly used as self-standing electrodes for LIBs and SCs. When evaluated as anodes of LIBs, this kind of integrated electrode showed high regional capacity, long cycling stability, and excellent rate performance in a galvanostatic charging and discharging potential window ranging from 1 to 3 V. When evaluated as electrodes for SCs, the as-synthesized integrated electrodes exhibited long cycling stability until 15 000 cycles with a capacitance fading of only 19.4% in a two-electrode coin cell configuration, when the current density is 3 mA/cm2.

Results and Discussion

Morphologies and Structural Analysis

The graphene-like MoS2 nanosheets grown in carbon fabrics were synthesized via an easy hydrothermal method. Figure a shows the X-ray diffraction (XRD) pattern of the MoS2 nanosheets. The pattern can be well indexed to hexagonal MoS2 (JCPDS card no. 73-1508). The peak located at 2θ = 26° is a typical diffraction caused by carbon fabrics. The other peaks located at 2θ = 14.39, 33.62, 39.65, 44.14, 49.87, and 58.56° can be assigned to the (002), (101), (103), (006), (105), and (110) planes of the hexagonal MoS2 phase, respectively. Their morphologies were characterized through the use of field-emission scanning electron microscopy (FESEM) and FE-transmission electron microscopy (FE-TEM). Figure b shows the SEM image of the as-synthesized products, which were evenly coated high-density samples. The higher magnification of the SEM image shown in Figure d reveals that the samples on the carbon fabrics are graphene-like ultrathin nanosheet structures. To clearly observe the thickness of the sample on the carbon fabrics, the sectional view of the as-synthesized product is shown in Figure c. We can clearly estimate that the thickness of MoS2 nanosheets coated on carbon fabrics was ca. 200–300 nm. Figure e shows the TEM image of the MoS2 nanosheets which were scratched from the carbon fabrics, and we can clearly see a two-dimensional layer-structure, corresponding to the SEM images very well (Figure d). The selected-area electron diffraction (SAED) and high-resolution TEM (HR-TEM) analyses effectively indicate the polycrystalline texture of MoS2 nanosheets (Figure e,g). The lattice fringes are visible, and the d-spacing of 0.2635 nm was computed to correspond well to the (101) lattice plane of the hexagonal MoS2 nanosheet. The higher magnification image of MoS2 nanosheets is shown in Figure f. We can observe a well-defined layered structure of MoS2 nanosheets. The thickness of the MoS2 nanosheet is about 2.5 nm, which was characterized by an atomic force microscope (AFM) in Figure h.

Figure 1

(a) XRD pattern, (b–d) SEM images of graphene-like MoS2 nanosheets on carbon fabrics. (e, f) TEM images of the MoS2 nanosheets and the inset (e) shows the corresponding SAED pattern. (g) HR-TEM image of MoS2 nanosheets. (h) Three-dimensional AFM image of MoS2 nanosheets.

The detailed elemental composition and the valence states on the surface of MoS2 nanosheets are further characterized by X-ray photoelectron spectroscopy (XPS). The investigated spectrum of the MoS2/carbon fabrics (Figure a) shows the existence of Mo, S, C, and O elements. Figure b,c show the high-resolution XPS spectra of Mo 3d and S 2p obtained from the MoS2/carbon fabrics. The high-resolution Mo 3d peak (Figure b) shows two sub-peaks and a small peak at the binding energies of 229.6, 232.8, and 226.8 eV that correspond to the Mo 3d5/2, Mo 3d3/2, and S 2s states, respectively, indicating that molybdenum is in its Mo (IV) state. In Figure c, the peaks at the binding energies of 163.6 and 162.5 eV correspond to the S 2p1/2 and S 2p3/2 states, respectively. All of the binding energy values are well consistent with those reported for MoS2.[38]

Figure 2

XPS spectra of graphene-like MoS2 nanosheets on carbon fabrics. (a) Survey spectra and high-resolution spectra of (b) Mo 3d and S 2s, and (c) S 2p.

Li-Ion Battery Performance

Coin-type cell configuration was used to assess the energy storage properties of MoS2/carbon fabrics for LIBs. The mass loading of MoS2 nanosheets on carbon fabrics used for LIB tests is about 2.5–3 mg. MoS2 nanosheet/carbon fabric as a cathode was compared with Li metal under the constant current circulation conditions under the potential window of 0.01–3 and 1–3 V at the current rate of 0.2 mA/cm2 at room temperature. When the discharge potential was below 1 V, the carbon fabrics will react with Li metal, and this process is nearly irreversible. As a result, the Li metal will be wasted and lead to low Coulombic efficiency. Furthermore, the reduction peak of MoS2 does not exist below 1 V. Therefore, we test the electrochemical performance of MoS2/carbon fabrics at a potential ranging from 1 to 3 V. Figure S1 shows the charge/discharge profiles of MoS2/carbon fabrics and pure carbon fabrics for the 1st, 5th, and 60th, respectively. We can see that the pure carbon fabrics exhibit high area capacity (Figure S1b). The first-discharge reaction of MoS2/carbon fabrics comprises mainly two regions. The first potential plateau at ∼1.1 V associated with the reaction of MoS2 and Li+ as in eq . The second plateau at ∼0.6 V is associated with the reduction of Mo4+ into the metallic Mo embedded into a cubic Li2S matrix through the conversion process (eq ) and the formation of a gel-like solid electrolyte interphase layer at the interface of the electrolyte and the electrode. In the charge profiles, the two plateaus at ∼1.7 and ∼2.3 V are associated with the oxidation of cubic Mo to hexagonal MoS2 and dissociation oxidation of Li2S to sulfur, respectively. In the second and subsequent discharge profiles, two potential plateaus at ∼1.8 and ∼1.15 V are associated with reduction of S to Li2S and the formation of lithium intercalate of LiMoS2, respectively.Considering that the main lithiation/delithiation reaction of MoS2 is above 1 V after the first cathodic process, and the potential plateau is above 2 V, the capacity contribution below 1 V cannot be taken into consideration when assembled into full battery, then, we think that the galvanostatic charge/discharge tests of MoS2/carbon fabrics performed at a potential ranging from 1 to 3 V are more suitable.

Figure a shows the charge/discharge profiles of MoS2/carbon fabrics for the 1st, 20th, 100th, 300th, and 400th at the potential between 1 and 3 V, respectively. Because the reduction process (eq ) may not occur in the first discharge process, the corresponding oxidation process may not exist either. The reaction mechanism can be changed when the potential window changed. According to Figure a, during the first discharge, a large plateau at about 1.25 V can be attributed to the reaction of MoS2 and Li+ as in eq . The reaction of eq does not exist. In the first charge process, a large inconspicuous plateau between 1.5 and 2.5 V may be associated with the extraction of Li+ from the LiMoS2 lattice. In the subsequent cycles, only a simple insertion/extraction process occurs. It should be noted that the Li+ intercalate potential of the first discharge process is lower than the subsequent cycles, which can be attributed to the different discharge depth resulting in the different Li+ insertion potential. Figure b shows the cycling performance of MoS2/carbon fabrics and pure carbon fabrics at a current density of 0.2 mA/cm2 in the potential window between 1 and 3 V. Because the pure carbon fabrics contribute such little capacity to the whole electrode, the reaction of Li with carbon fabrics can be neglected. After 400 cycles, the MoS2/carbon fabrics still hold a high reversible area capacity of 0.4 mAh/cm2, about 80% capacity retention after the 1st cycle, compared to the cycling performance of MoS2 carbon fabrics tested at a potential window from 0.01 to 3 V (Figure S2), which shows a capacity retention of only 63% in the 60th cycle after the 1st cycle. The enhanced cycling performance of MoS2/carbon fabrics when tested at a potential ranging from 1 to 3 V can be attributed to two reasons. First, the depth of discharge can result in the easier destruction of the electrode. Second, the current collector taking part in the electrochemical reaction may result in the active materials peeling off from the current collectors and the capacity fading rapidly. To provide more direct evidence, the morphologies of MoS2/carbon fabrics tested at a potential ranging from 0.01 to 3 V (a, b, c) and 1 to 3 V (d, e, f) after the cycles were compared. Figure c shows the rate performance of MoS2/carbon fabrics at various current densities from 0.2 to 4 mA/cm2 in a potential window of 1–3 V. When the current density increases to 4 mA/cm2, the MoS2/carbon fabric electrode still exhibits a high area capacity of 0.29 mAh/cm2, and when returning back to 0.2 mA/cm2, the capacity of MoS2/carbon fabrics is recovered with a capacity retention of about 86.7%. The first two charge/discharge curves of pure carbon fabrics above 1 V are shown in Figure S3. As shown in Figure S3, pure carbon fabrics contribute a low area capacity of less than 0.015 mAh/cm2 between 1 and 3 V. Figure a–c show the SEM images of MoS2/carbon fabrics tested at a potential ranging from 0.01 to 3 V after 60 cycles, and it was observed that the structures of the carbon fabrics and graphene-like nanosheets were completely destroyed. The materials are partly peeled off from the carbon fabrics. Figure d–f show the SEM images of MoS2/carbon fabrics tested at a potential ranging from 1 to 3 V after 400 cycles. Compared to the electrode that is tested at 0.01–3 V, the morphologies of MoS2 were not destroyed completely, and thus we can see the nanosheet structure clearly.

Figure 3

(a) Charge/discharge profiles, (b) cycling performance at 0.2 mA/cm2 and (c) rate performance of MoS2 carbon fabrics in the potential range of 1–3 V.

Figure 4

SEM images of the MoS2/carbon fabric electrode after 60 cycles (a–c) and after 400 cycles (e, d, f). The charge/discharge potential range was 0.01–3 V (a–c) and 1–3 V (e, d, f), respectively.

Supercapacitor Performance

The MoS2/carbon fabrics were also applied as self-standing electrodes of supercapacitors. The mass loading of MoS2 on carbon fabrics used for SC tests is about 1.8–2.2 mg/cm2. The capacitive properties of the MoS2/carbon fabrics were first tested with cyclic voltammetry (CV) and galvanostatic charge–discharge measurements in three-electrode configurations. Figure a displays the CV curves of the MoS2/carbon fabrics at scan speeds varying from 10 to 50 mV/s. The pseudocapacitance behavior can be confirmed obviously from the fact that the CV curve shape is different from the electrical double-layer capacitance, which is a nearly rectangular CV curve. Notably, redox peaks between −0.8 and 0 V can be seen in all CV curves due to the Faraday effect. As the scanning rate increases from 10 to 50 mV/s, the peak current rises up, but the CV curve shape shows no significant change. This result suggests that rapid redox reactions towards electrochemical energy storage can be performed with MoS2/carbon fabrics.

Figure 5

(a) CV curves, (b) constant-current charge–discharge voltage profiles, (c) specific capacitance as a function of current density, and (d) cycling performance at a current density of 2 mA/cm2.

Constant current charge/discharge profiles were conducted at current densities ranging from 1 to 8 mA/cm2, which are displayed in Figure b. The representative capacitance of the electrode material is calculated as followswhere I represents the constant-discharge current, S means the geometrical area of the electrode, Δt is the discharged time after IR drop, and ΔV is the voltage drop upon discharge. The specific capacitance of the MoS2/carbon fabric electrode calculated from each discharge curve is about 191.9, 152.3, 136.7, 127.4, 119.3, and 92.6 mF/cm2 at 1, 2, 3, 4, 5, and 8 mA/cm2, respectively, as shown in Figure c. Figure d shows the cycling performance of MoS2/carbon fabrics at a current density of 2 mA/cm2. After 2000 charge/discharge cycles, specific capacitance attenuation is negligible, indicating that the MoS2/carbon fabric electrode has superior cycling stability performance.

The symmetric two-electrode system adopted to evaluate the capacitive performance in this study due to the three-electrode configuration may easily overestimate the energy storage capability of an electrode material for practical supercapacitor use. Moreover, the cycling performance in a three-electrode system can be affected by some other aspects, such as the active materials peeling off from the current collectors. Figure a presents the CV curves of the two-electrode device at scan rates from 10 to 300 mV/s with a potential window between −0.8 and 0.8 V. Not only because of the surface electrosorption of Li+ cations but also because of the fast, reversible successive surface redox reactions of MoS2, the measured CV curves show quasirectangular shapes. It can be seen from the absence of redox peaks that the supercapacitor is charged and discharged at a pseudoconstant rate throughout the voltammetric cycle. From the galvanostatic charge–discharge measurements at current densities between 0.5 and 8 mA/cm2, we obtained the accurate electrochemical values of the symmetric two-electrode system (Figure b). The capacitance of the electrode can be calculated according to the following equationwhere I is the constant discharge current, S is the geometrical area of the electrode, Δt is the discharged time after IR drop, and ΔV is the voltage drop upon discharge, respectively.[39−41] The area capacitance values of the electrode were about 159.38, 134.5, 114.5, 103.5, 93.5, 85.63, and 71 mF/cm2 at 0.5, 1, 2, 3, 4, 5, and 8 mA/cm2 that are calculated from the charge/discharge curves, respectively, as shown in Figure c. Compared with the three-electrode configurations, the capacitance of the electrode is lower in a two-electrode system, which may be a normal phenomenon in the supercapacitor test. Figure d shows the cycling performance of the two-electrode system at 3 mA/cm2. After 15 000 cycles, the electrode still retains a high specific capacitance of about 83.7 mF/cm2 (about 80.6% capacitance retention).

Figure 6

(a) CV curves, (b) constant-current charge–discharge voltage profiles, (c) specific capacitance as a function of current density, and (d) cycling performance at a current density of 3 mA/cm2.

Conclusions

In summary, graphene-like MoS2 nanosheets have been successfully grown on carbon fiber fabrics with strong adhesion by a simple hydrothermal method. The MoS2/carbon fabrics are self-standing electrodes for energy storage devices. When used as an anode for Li-ion batteries, graphene-like MoS2/carbon fabrics give a high reversible area capacity of 0.5 mAh/cm2, long cycling stability (80% capacity retention after 400 cycles), and excellent rate performance. A high specific capacitance of about 95 mF/cm2 at 3 mA/cm2 and extraordinary cycling stability (near 81% capacitance retention after 15 000 cycles) can be achieved as an electrochemical capacitor electrode. Because of the excellent cycling stability of MoS2, we can composite with other electrode materials that have large specific capacitance but poor cycling stability to improve the electrochemical properties of electrode materials just like graphene and partly replace graphene in some special aspects.

Methods

Materials Synthesis

Graphene-like MoS2 nanosheets were obtained by a typical and simple hydrothermal method. Since all reagents used are analytical without further purification. To directly grow MoS2 nanosheets on carbon fabrics, commercial carbon fabrics which are used as substrates were first cleaned with acetone, ethanol, and deionized water, and then the carbon fabrics were cut into the needed sizes (2 cm × 4 cm) and immersed in 0.5 M NaOH solution for about 1 h. After the above pretreatment, a piece of the above carbon fabrics was placed in a Teflon-lined stainless-steel autoclave (50 mL) containing a mixture of 0.3 g Na2MoO4·2H2O, 0.4 g CH4N2S, and 35 mL deionized water. The sealed autoclave was heated in an oven at 240 °C for 24 h. Immediately after the completion of the reaction, the reacted carbon cloth was washed with deionized water and ethanol, and then dried in a vacuum oven at 80 °C for 12 h.

Characterization

The X-ray diffraction (XRD) pattern was obtained by using an X-ray diffractometer (X’Pert PRO, PANalytical B.V., the Netherlands) with Cu Kα radiation (λ = 0.154 nm). X-ray photoelectron spectrometry was performed on a VG Multi-lab 2000 system with a monochromatic Al Kα X-ray source. The morphologies and structures of the as-synthesized products were characterized by a field-emission scanning electron microscope (Nova NanoSEM 450, FEI, the Netherlands), a field-emission transmission electron microscope (Tecnai G2 F30, FEI, the Netherlands), and an atomic force microscope (AFM, SPM9700, Shimadzu).

Electrochemical Measurements

The carbon fabric-supported MoS2 nanosheets directly acted as the working electrode. Lithium metal was used as the counter electrode, and a polypropylene film (Celgard-2300) was used as the separator. The electrolyte used was 1.0 M LiPF6 solution with a mixture of ethylene carbonate and dimethyl carbonate (EC/DMC, 1:1 in volume). Galvanostatic charging and discharging tests were conducted using a battery testing system (Land, China) at a potential ranging from 1 to 3 V. The SC tests were performed on a CHI 660C electrochemical workstation in an aqueous LiOH electrolyte (1 M). The three-electrode cell tests were conducted with Pt foil as the counter electrode and a saturated calomel electrode as the reference electrode. The two-electrode SC tests were conducted in a two-electrode coin cell (CR2032) configuration assembled in air, where MoS2/carbon fabrics serve as both the cathode and anode, glassy fibers serve as the separator. Coin cells were assembled in an argon-filled glovebox for a lithium-ion battery.