Fabrication of MoS₂ Nanoflakes Supported on Carbon Nanotubes for High Performance Anode in Lithium-Ion Batteries (LIBs)

Abstract

Molybdenum disulfide (MoS₂), an inorganic-layered material similar to structure of graphite, was randomly dispersed onto the surface of functionalized multiwalled carbon nanotubes to synthesized nanocomposite MoS₂/CNT. The as-obtained product was characterized via SEM, TEM, TGA, X-ray diffraction, and Raman spectroscopies. It was confirmed from XRD that MoS₂ layers with interlayer spacing of 0.614 nm were successfully produced. TEM images and Raman spectra indicated a random distribution of 20 nm sized nanoflake MoS₂ on the surface of MWNTs. The electrochemical performance of materials are expected to pave the way for the utilized anode material for lithium-ion batteries.

1. Introduction

For many years, energy storage is a major concern for all businesses and individuals. Many methods are developed to capture renewable energy sources such as solar cells, wind turbines, and tidal power. The conversion of these energy sources into direct electricity does not meet the demand for production and consumption. Storing energy obtained in the form of electricity, as rechargeable batteries, is an important solution. The emergence of rechargeable batteries in general and lithium-ion batteries (LIBs) in particular has become an effective alternative to battery research.

In landscape of LIBs, transition metal dichalcogenide (TMD), especially molybdenum disulfide (MoS₂), has currently attracted considerable attention due to its important role in many applications. In literature of knowledge, MoS₂ present its potential in areas, such as catalysts for the hydrogen evolution reaction (HER) [1], photoelectrode for energy conversion [2], lithium or sodium batteries [3], chemical sensing [3, 4], fillers in polymer composites [5, 6], photocatalysts [7–9], and solid lubricants for metallic and ceramic surfaces [10]. In recent years, MoS₂ has emerged as a promising anode material in lithium-ion batteries (LIBs) due to its typical-layered transition metal sulfide composing of three stacked atom layers (S-Mo-S) (Figure 1) [10]. Such MoS₂ nanostructures have a large specific surface area as well as abundant voids and defects that may provide more and shorter Li-ion diffusion during lithiation/delithiation processes [11]. In addition, the weak van der Waals interaction between MoS₂ layers allows the diffusion of Li⁺ ions without significant volumetric change during charge/discharge processes.

Figure 1 

Structure of molybdenum disulfide.

However, the poor electrical conductivity, low cycling stability, and high agglomerate risk remain the major drawbacks of bulk MoS₂. To overcome these problems, numerous studies of growth of MoS₂ on different conductive substrate or supports, such carbon paper [12–15], graphite [12, 13], 1D carbon nanotubes (CNTs) [10, 11, 14, 16–19], 3D nanocarbon [20], carbon nanofiber [21], or graphene [10, 14, 22, 23], were conducted to improve their electrochemical performance and stability. Well reversible specific capacity as high as 1,290 mAh·g^-1 were reported for nanostructured MoS₂-graphene anodes [10, 14]. Nanocomposite of MoS₂ and carbon nanotubes also remains attractive in terms of capacity. For example, MoS₂ nanosheet decorated on multiwalled carbon nanotubes (MWNTs) and single-walled carbon nanotubes (SWNTs) reported as larger than 1,000 and 1,400 mAh·g^-1, respectively. Constructing a composite material with reduced MoS₂ agglomerate on an interconnected conducting network of CNTs is crucial for high-rate capability and long-term cyclability of lithium-ion battery.

To date, the synthesis of MoS₂ nanomaterials has mainly been based on reactions in liquid media, i.e., hydrothermal [24], solvothermal [22], or even sonochemical synthesis [25]. Significant number of publications demonstrated that mass loadings of nanocomposite of MoS₂ and carbon-based materials can be created by hydrothermal synthesis. Also, solvothermal-synthesized nanoflake MoS₂ was reported with 700–800 mAh·g^-1 specific capacity. Although nanosized MoS₂ with different morphologies has successfully been prepared, the synthesized conditions are generally carried out under harsh conditions with high temperatures and pressures. Herein, a facile method was reported using gentle synthesis procedure to prepare MoS₂ nanoflakes with the support of MWNTs. Using our experimental procedure, MoS₂ nanoflakes with thickness of ~ 20 nm were readily obtained and randomly dispersed onto the surface of carbon nanotubes. Moreover, the MoS₂ interlayer spacing was around 0.62 nm, which makes them attractive for applications in lithium batteries.

2. Materials and Methods

The raw MWNTs were produced by the thermal chemical vapor deposition (T-CVD) method and purified by the wet chemical method. The details for MWNT synthesis and purification were given elsewhere [26–28]. For a better wetting of the surface, MWNTs were treated by refluxing in a mixture of C₆H₇NSO₃ and NaNO₂ [29].

All the chemicals used in this work were of reagent grade. Ammonium heptamolybdate tetrahydrate (NH₄)₆Mo₇O₂₄·4H₂O, sodium sulfide nonahydrate Na₂S·9H₂O, and ethylene glycol were purchased from Merck.

86.5 mg of (NH₄)₆Mo₇O₂₄·4H₂O was immersed into 50 mL of deionized water (DI) and sonicated until being completely dissolved to achieve a homogeneous solution. Then, 960 mg of Na₂S·9H₂O and 50 mg of MWNTs were added into the suspension and dispersed via sonication for 5 min to obtain a precursor solution. 50 mL of ethylene glycol was mixed with the precursor solution, and the reaction was carried out at 80°C for 30 min under stirring with gradual dropping of 2 mL of 1 M HCl solution. After that, the precipitates were collected, washed with DI water, and dried at 80°C in an oven.

The morphology and structure of the prepared samples were analyzed by TEM (JEOL-JEM-1400), SEM (JEOL-JSM-7401F), XRD (D8 ADVANCE), Raman (Labram 300), TGA (Netzsch TG 209 F3). The electrochemical measurements were performed by PARSTAT 2273 instrument using a Swagelok cell, where counter electrode was a lithium foil, reference electrode was Li/Li⁺, and working electrode was prepared by the following steps: (i)15 mg of composite material binder free was compressed on a 10 mm in diameter discs of copper foil(ii)3 layers of Nafion membrane were soaked with few drops of 1 M LiPF₆ in EC : DMC (1 : 1 ) solution, then sandwiched between the composite working electrode and counter electrode(iii)the entire assembly was secured in a cylinder of Swagelok cell

All the electrode preparing process and cyclic voltammetry were accomplished in argon full filled glovebox. The cyclic voltammograms were obtained from 0 to 3.5 V (versus Li/Li⁺) of voltage range at 1 mV/s of scanning rate.

3. Results and Discussion

Figure 2 shows XRD patterns of MoS₂ prepared with and without CNTs. The diffraction peaks of MWNTs appeared at 2θ of about 26° and 53° corresponding to (002) and (004) planes (according to JCPDS 25-0284), respectively (Figures 2(a) and 2(c)). Four other distinct peaks at 2θ of 14.41°, 32.76°, 39.60°, and 58.31° corresponding to (002), (100), (103) and (110) planes of MoS₂ (according to JCPDS 37-1492), respectively, were observed (Figures 2(b) and 2(c)). The d-spacing for the (002) plane of MoS₂ layers was calculated to be 0.617 nm according to its diffraction peak at using Bragg’s equation, which is consistent with that of hexagonal MoS₂ (2H-MoS₂) [27]. It should be noted that the relative diffraction peak intensities of the MoS₂/CNT sample are enhanced in comparison with those of MoS₂ obtained in the absence of CNTs. The enhanced diffraction peaks indicate that the MoS₂ layers growing on CNTs have higher crystallinity than that obtained in the absence of CNTs. This also demonstrates that CNTs can support the formation of MoS₂ and improve its crystallinity. In the case of MoS₂/CNTs, the (002) plane diffraction peak has a small shift to a lower angle of 14.34° from the standard angle of 14.41°, indicating that the interlayer distance of the MoS₂ nanoflakes is slightly increased which will potentially provide a sufficiently larger space for lithium-ion intercalation. It has been reported that with 1 mol of lithium intercalation, the c parameter of MoS₂ undergoes an increment of 0.03 Å. The expanded interlayer d-spacing would relieve the strain caused by electrochemical lithiation/delithiation during cycling and provide more space for Li-ion intercalation with reduced diffusion barriers [11, 30, 31].

Figure 2 

XRD patterns of (a) MoS₂/CNTs and (b) MoS₂ prepared in the absence of CNTs.

To verify the morphology of the as-prepared product, scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were conducted. The surface morphology of MoS₂ and carbon nanotubes could be clearly observed in Figures 3(a) and 3(b). Figure 3(a) illustrates that MoS₂ obtained in the absence of CNTs composes of nanoflake coalescence into large particles. The average crystallite size of the nanoflakes is around 20 nm in the thickness direction. In the presence of CNTs, a similar morphology of MoS₂ is observed under TEM. However, the MoS₂ nanoflakes appeared to be randomly distributed on the surface of CNTs in contrast to the agglomeration of MoS₂ nanoflakes synthesized without CNTs (Figures 3(c) and 3(d)).

(a)

(b)

(c)

(d)

Figure 3 

(a) SEM image of MoS₂ prepared in the absence of CNTs; (b) SEM image of prepared MoS₂/CNTs; (c, d) TEM images of MoS₂/CNTs. Inset in (d) shows the FFT patterns taken from the marked area.

Raman spectroscopy was also employed to further characterize the MoS₂/CNT hybrid structure. Raman experiments were carried out with 632.8 nm excitation. Figures 4(a) and 4(b) compare the Raman spectra taken from the as-obtained MoS₂/CNTs and bare CNTs. The Raman peaks at around 1573 cm^-1 and 1325 cm^-1 belong to MWNTs, while the G band of MWNTs is located at 2,650 cm^-1, consistent with previous reports [28]. The appearance of two other peaks at 377 and 405 cm^-1 corresponds to the out-of-plane (A_1g) and in-plane (E_2g¹) vibrational modes of hexagonal MoS₂, respectively. A more detailed view of the vibrational modes is presented in Figure 4(c). Two MoS₂ layers and corresponding vibrations are displayed. The ratio of these two peaks is usually employed to evaluate the thickness and orientation of MoS₂ [32–36]. The MoS₂ thin sheets with the basal plane exposed to laser beam (composed of less than 10 layers) are often reported to have this value being about 1, while this value of the bulk ones is about 1.5. In contrary, a value above 3 was reported for the nanosheet (NS) with the edge side exposed to laser beam [32–37]. The peak intensity ratio was observed to be higher than 3 (approximately 4) in this study, suggesting that the MoS₂ NS growth direction is perpendicular to the CNT surface.

(a)

(b)

(c)

Figure 4 

(a) Raman spectra of (A) MoS₂/CNTs and (B) bare CNTs; (b) comparison of the magnified Raman signals of MoS₂ in prepared MoS₂/CNTs and bare CNTs; (c) symmetric displacement of Mo and S atoms in A_1g and E_2g¹ vibrational modes.

New peaks at around 450 cm^-1 and 630 cm^-1 (Figure 4(b)) might be caused by the MoS₂ being partially oxidized if the laser power was too large and focused on the sample in the measurement. It was also found that the signals related to the CNTs mostly overlap with those of the MoS₂ sheets, and their intensity becomes significantly weaker, which suggests a good grafting of MoS₂ around the surface of CNTs.

The thermogravimetric analysis curves (Figure 5) were used to calculate the loading percentage of MoS₂ in the nanocomposite. Similar TGA curves for bare MoS₂ and MoS₂/CNTs showed the first mass loss from 100°C to about 200°C related to evaporation of absorbed moisture on the surface of materials. Larger specific area of MoS₂/CNTs allowed it containing more moisture that make evaporation of the sample occur at over 300°C. Afterward, the consecutive reaction is oxidation of MoS₂ to MoO₃ in air from 300°C. At over 400°C to 850°C, oxidation of CNTs occurred parallel to oxidation of MoS₂ in nanocomposite sample that makes its weight lose faster than bare MoS₂. The calculated results from TGA curves, MoS₂, and MWNTs in the nanocomposite have approximate loading percentage of 76 wt% and 24 wt%, respectively.

Figure 5 

TGA curves for bare MWNTs, MoS₂, and prepared MoS₂/CNTs (heating in air with 10 K/min of rate, from room temperature to 980°C).

Figure 6 showed voltammograms for prepared MoS₂/CNT anode materials. In knowledge of literature, lithiation of MoS₂ occurred within the voltage window of 3–0 V (vs. Li/Li⁺). The first intercalation step of Li⁺ into MoS₂ is shown as reaction (1) [17]. Reaction (1) is attributed by the peak at about 3.0 to 1.1 V. In the first scan, a reduction peak at approximately 1.1 to 1.2 V described the transformation of trigonal prismatic 2H-MoS₂ to octahedral 1T-Li_xMoS₂ [17, 18]. The second peak at 0.6 to 0.7 V attributed to lithiation into MWNTs and the formation of Li₂S to release Mo which is shown as reaction (2) [17]. Lithiation peak of MoS₂/CNTs presented larger peak area than that of MoS₂ that mainly resulted higher capacity of MoS₂/CNTs.

Figure 6 

The cyclic voltammograms for MoS₂/CNTs as anode materials.

In charging scan, Figure 6 showed a shallow peak at 1.7–1.8 V and a large peak at 2.6–2.7 V, approximately. The first shallow peak attributed delithiation of enduring Li_xMoS₂ which was not conversed by reaction (2). The officious delithiation step is the conversion of Li₂S to S₈^2-, which is attributed by the large peak at 2.6–2.7 V. The reversible conversion of MoS₂/CNTs demonstrates the absence or the exiguous formation of rigid Li₂S, which is constantly attributed by disappeared peak at approximately 0.4 V.

4. Conclusions

In summary, nanoflake MoS₂ were grafted directly on the surface of carbon nanotubes. These MoS₂/CNT nanocomposites were conveniently synthesized by the wet-chemical process, in which grafted MoS₂ nanoflakes are high crystalline and likely to be grown parallel to the MWNT surface. The MoS₂ nanoflakes had about 20 nm size with the interlayer spacing of 0.614 nm, which makes their electrochemical performance suitable for applications in lithium batteries. The nanocomposite contained 24 wt% of CNTs to rise the rate of lithiation.

Data Availability

The SEM, TEM, XRD, and Raman spectroscopy used to support the findings of this study have not been made available because they were provided only as image data type by analysis centres.

Conflicts of Interest

The authors declare that there is no conflict of interest regarding the publication of this paper.

Acknowledgments

This research is funded by the Viet Nam National University Ho Chi Minh City (VNU-HCM) under grant number B2017-20-07/HĐ-KHCN.

Supplementary Materials

S1: materials. Synthesis and purification of multiwalled carbon nanotubes (MWNTs). Surface treatment of MWNTs. S2: cyclic voltammograms of MoS₂/CNTs. (Supplementary Materials)