ZHANG Xiu, DENG Ya-kai, WANG Yan-li， ZHAN Liang, YANG Shu-bin, SONG Yan

(1. State Key Laboratory of Chemical Engineering，Key Laboratory for Specially Functional Polymers and Related Technology of Ministry of Education, Shanghai Key Laboratory of Multiphase Materials Chemical Engineering, East China University of Science and Technology，Shanghai200237，China; 2. Key Laboratory of Aerospace Advanced Materials and Performance of Ministry of Education, School of Materials Science and Engineering, Beihang University, Beijing100191, China;3. CAS Key Laboratory of Carbon Materials, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan030001, China)

Abstract: One-dimensional MoS2/carbon nanofibers (MoS2/CNFs) were synthesized by electrospinning using exfoliated MoS2 nanosheets and polyacrylonitrile as the precursors. The exfoliated MoS2 nanosheets about 150 nm across were encapsulated in carbon, and the free-standing MoS2/CNF film was easily cut into a flexible tablet that could be directly used as a binder-free anode for lithium storage. The MoS2/CNFs showed a high reversible capacity of 700 mAh g-1 at 100 mA g-1 after 50 cycles, a high rate capacity of 450 mAh g-1 at 1000 mA g-1 after 200 cycles and good cycling stability.

Key words: MoS2; Carbon nanofibers; Electrospinning; Anode material

1 Introduction

Lithium ion batteries (LIBs) have been regarded as one of the most important rechargeable energy storage devices with broad applications in hybrid and electric vehicles owing to their high potentials and environmental friendliness[1-3]. The reversible capacities of commercial graphite-based anode materials cannot satisfy the increasing requirements for high-performance LIBs due to the low theoretical capacity of graphite (372 mAh g-1)[4]. Therefore, many endeavours have been concentrated on exploring novel anode materials, such as transition metal oxides[5], molybdenum disulfide (MoS2)[6], stannum (Sn)[7]and silicon (Si)[8]. Among these novel anode candidates, MoS2has been recognized as one of the most promising and attractive one for LIBs, owing to its high theoretical capacity (670 mAh g-1), relatively low discharge potential, low-cost, safety and environment-friendly[9-11]. However, bulk MoS2suffers from a low electrical conductivity and a high volume expansion during cycling, leading to a poor electrochemical performance[10]. To overcome these shortcomings, researchers focus on fabricating single-layer or few layered MoS2by chemical vapor deposition[11,12], chemical exfoliation[13]and mechanical exfoliation[14,15]. Beside the extremely low yield, the single-layer MoS2or MoS2nanosheets are easy to restack, leading to the structural instability during cycling, as a result, an obvious volume expansion and rapid capacity fading will occur[16]. Thus, researchers further focus their attentions on combining nanostructured MoS2with carbonaceous materials (such as carbon nanotubes[17], graphene[18,19]and conductive polymers[20]) to resolve the low electrical conductivity, huge volume change and restacking problem. However, how to develop a simple and efficient approach to fabricate MoS2/carbon nanostructures is still a big challenge.

Herein, we develop an efficient approach to fabricate one-dimensional MoS2/carbon nanofibers (denoted as 1D MoS2/CNFs) by electrospinning. Exfoliated MoS2nanosheets with small lateral sizes were encapsulated in carbon to form nanofibers. The distinctive structure of 1D MoS2/CNFs can improve the electrical conductivity of pure MoS2nanosheets, but also can prevent the restacking of nanosheets. Importantly, the free-standing MoS2/CNF film is flexible and easily cut into tablets that are directly used as anode of LIBs without need of a binder. The unique structures offer the resultant 1D MoS2/CNFs excellent electrochemical performance. For instance, the resultant 1D MoS2/CNFs have a high capacity of 700 mAh g-1at 100 mA g-1after 50 cycles and good high-rate performance (450 mAh g-1at 1 000 mA g-1after 200 cycles). We believe that this efficient method can be further extended to fabricate other active nanomaterials encapsulated in carbon nanofibers or carbon nanotubes for broad applications in batteries, supercapacitors and catalysts.

2 Experimental

2.1 Materials and preparation

MoS2nanosheets were initially fabricated by a modified sheer exfoliation method as literature reported[21]. 90 mg exfoliated MoS2nanosheets were dispersed in 10 mL N,N-Dimethylformamide (DMF) by ultrasonication, then 60 mg polyacrylonitrile (PAN) were dissolved in the solution by stirring at 40 ℃ for 12 h. Subsequently, above mixture solution was poured into a syringe for electrospinning. Detailly, the diameter of the needle, the distance between needle and collector, the working voltage and the flow rate of the solution are 0.34 mm, 15 cm, 10 kV and 0.5 mL/h, respectively. Finally, the resultant 1D MoS2/PAN nanofibers were oxidized at 280 ℃ in air for 2 h and then carbonized at 850 ℃ for 3 h under Ar atmosphere to obtain the 1D MoS2/C nanofibers.

2.2 Characterization

The morphologies of the samples were characterized by scanning electron microscopy (SEM, Zeiss MERLIN Compact) and transmission electron microscopy (TEM, JEOL 2100F). The structure and composition were characterized by X-ray diffraction (XRD, Rigaku D/MAX2500), X-ray photoelectron spectroscopy (XPS, Thermo Scientific Escalab 250Xi) and TGA (STA449 Jupiter, NETZSCH) measurements. The nitrogen adsorption test was performed on a Quantachrome QDS-MP-30 analyzer (USA) at 77 K.

2.3 Electrochemical measurement

Electrochemical experiments were performed using standard CR2031 type coin cells assembled in the glovebox. After the MoS2/CNFs were cut into flexible tablet that was directly used as the anode electrode. The exfoliated MoS2and bulk MoS2electrodes were fabricated by mixing the active material with acetylene black and poly(vinyl difluoride) (PVDF) at a weight ratio of 8∶1∶1 for comparison. In the process of fabrication of LIBs, pure lithium foil were used as the counter electrode, propene polymer (PP) membrane as the separator and 1 mol/L LiPF6in ethylene carbonate as the electrolyte. The galvanostatic discharge/charge behavior of coin cells was tested on a battery testing system (Land CT2001A) at the voltage of 0.01-3 V. Both cyclic voltammetry (CV) and electrochemical impedance spectrometry (EIS) tests were performed on an Autolab equipment (PGSTAT302N).

3 Results and discussion

Fig. 1 illustrates schematically the synthetic procedure of the MoS2/CNFs. Bulk MoS2were added to the deionized water and isopropylamine (IPA) solvent with a volume ratio of 1∶1, and exfoliated by an emulsification machine at 10 000 r/min. Subsequently, the exfoliated MoS2nanosheets were treated through freeze-drying, and then dispersed in the PAN/DMF solution. After the free-standing 1 D MoS2/PAN based nanofibers were achieved by electrospinning, they were oxidized at 280 ℃ for 2 h and then carbonized at 850 ℃ for 3 h under Ar atmosphere, resulting in the 1D MoS2/CNFs.

Fig. 1 Schematic illustration of the fabrication of 1D MoS2/CNFs.

The MoS2nanosheets are achieved by solvent exfoliation with a lateral size of about 150 nm (Fig. 2a), and the exfoliated MoS2nanosheets are dispersed well in the DMF solvent even after ten days. The exfoliated MoS2exhibits an interplanar space of 0.63 nm corresponding to (002) plane (Fig. 2b). Subsequently, 1D MoS2/CNFs were also successfully fabricated by the electrospinning method. As shown in Fig. 3a, a large amount of interlaced 1D MoS2/CNFs are observed, and the nanofibers are uniform dispersed and have a diameter of about 180 nm. The surface of resultant MoS2/CNFs is quite smooth, and no bulk MoS2or restacked MoS2nanosheets are detected among the nanofibers (Fig. 3b), illustrating that all the exfoliated MoS2nanosheets are encapsulated in the carbon matrix due to their small sizes and good affinity with PAN molecules. The TEM image in Fig. 3c also shows that the nanofibers have uniform diameter of about 180 nm, agreeing well with the results observed in the SEM images (Fig. 3a and 3b). Additionally, two-dimensional MoS2nanosheets can be detected and are well encapsulates in the carbon matrix (Fig. 3d). After carbonization, the PAN molecules have transformed into amorphous carbon (Fig. 3e), and the typical lattice fringes with an interplanar spacing of 0.63 nm are observed among the amorphous carbon (Fig. 3f), indexed to be the (002) facets of MoS2[20], which has the same characteristics as the exfoliated MoS2nanosheets (Fig. 2b). Fig. 2g-j show a typical scanning transmission electron microscopy (STEM) bright field image and elemental mapping analysis of the 1D MoS2/CNFs, where carbon, molybdenum and sulfur species are all distributed in this area. Importantly, the free-standing 1D MoS2/CNFs can be easily cut into flexible tablet and directly used as the binder-free anode for LIBs (inserted in Fig. 3a).

Fig. 2 (a) SEM and (b) HRTEM images of exfoliated MoS2 nanosheets. Exfoliated MoS2 nanosheets dispersed in DMF solvent after ten days is inserted in (a).

To further explore the chemical composition of the 1D MoS2/CNFs, XRD measurement was performed. As shown in Fig. 4a, there are several strong peaks at 14.2°, 32.6° and 39.6°, corresponding to the (002), (100), (103) planes of MoS2, respectively[22]. It should be noted that the peak of the MoS2/CNFs at 14.2° is much weaker than that of bulk MoS2, suggesting that the bulk MoS2has been successfully exfoliated into two-dimensional nanosheets. There is a broad and weak peak at about 25°,which refers to the typical diffraction peak of amorphous carbon[13]. There are no impurities detected by XRD analysis, demonstrating the high crystallinity and phase purity of the resultant 1D MoS2/CNFs. The type-IV hysteresis loop of the isotherms with pronounced adsorptions was obtained at relative pressuresp/p0from 0 to 1 (Fig. 4b). The specific surface area of the 1D MoS2/CNFs is 23.94 m2g-1. Interestingly, the pore size distribution indicates that there exists micropores and mesopores in the MoS2/CNFs (Fig. 4c), which should be related to the layered and restacked MoS2nanosheets as well as the interlaced structure of carbon. Thermogravimetric analysis was also performed to confirm the content of MoS2in the nanofibers (Fig. 4d). The weight loss occurs before 100 ℃, attributing to the evaporation of physically adsorbed water. The main weight loss occurs in the range of 300-500 ℃ because of the consumption of carbon and the oxidation of MoS2to MoO3in air. According to the final yield of MoO3, the content of MoS2encapsulated in the CNFs is more than 27.6%.

Fig. 3 (a, b) SEM, (c, d) TEM and (e, f) HRTEM images of synthesized 1D MoS2/CNFs. (g) STEM image of the 1D MoS2/CNFs and its corresponding (h) C, (i) Mo and (j) S elemental mapping. The MoS2/CNF film shows a good flexibility as inserted in (a).

The elemental contents of 1D MoS2/CNFs were elucidated by XPS measurement. Based on the XPS survey, it is distinct that carbon, molybdenum, sulfide and oxygen species are assumed in the MoS2/CNFs (Fig. 5a). The fitted C 1s peaks are demonstrated in Fig. 5b, the peaks located at 284.5, 285.5 and 287 eV are related to C—C, C—O and C—C bonds of amorphous carbon, respectively[23]. The high-resolution Mo 3d spectrum can be fitted to three types at the binding energies of 236.7, 233.4 and 230 eV, corresponding to the Mo6+, Mo 3d3/2and Mo 3d5/2peaks (Fig. 5c), respectively. The Mo 3d3/2and Mo 3d5/2peaks represent to the Mo4+in MoS2. The Mo6+is attributed to the existence of MoO3which is caused by the unavoidable surface oxidation of MoS2in carbonization. The high-resolution S 2p spectrum can be fitted to two types at 164 and 162.7 eV, corresponding to the S 2p1/2and S 2p3/2peaks, respectively, representive of the S2-in MoS2[18]. The O 1s peaks observed in the spectrum are mainly caused by the oxygen in the testing environment.

Fig. 4 (a) XRD patterns of the resultant MoS2/CNFs and bulk MoS2 samples. (b)Nitrogen adsorption/desorption isotherms of the MoS2/CNFs. (c) Pore size distribution of the MoS2/CNFs. (d) TG curve of the MoS2/CNFs.

Fig. 5 (a) XPS spectrum of the 1D MoS2/CNFs. High-resolution XPS spectra of (b) C 1s, (c) Mo 3d and (d ) S 2p.

Fig. 6 Electrochemical performance of 1D MoS2/CNFs for LIBs. Charge-discharge curves of (a) the 1D MoS2/CNFs and (b) bulk MoS2at a current density of 100 mA g-1. (c) Cycling performance of bulk MoS2 and the MoS2/CNFs at a current density of 100 mA g-1. (d) High-rate performances of bulk MoS2 and the MoS2/CNFs at various current densities of 50, 100, 200, 500 and 1 000 mA g-1. (e) Cycling performance of the MoS2/CNFs at a current density of 1 000 mA g-1. (f) Nyquist plots of bulk MoS2 and the MoS2/CNFs after rate-cycling with an amplitude of 5 mV.

The electrochemical performance of the 1D MoS2/CNFs was primary tested by galvanostatic charge-discharge measurement at a current density of 100 mA g-1. As shown in Fig. 6a, two obvious potential plateaus during the first discharge are visible for the MoS2/CNF electrode. The plateau at ～1.1 V is caused by the intercalation of Li+into MoS2to form LixMoS2which brings up the phase changing of MoS2from trigonal prismatic to octahedral. Another plateau at ～0.5 V is related to the conversion reaction of LixMoS2to Mo and Li2S[24]. Although the specific surface area of the 1D MoS2/CNFs is only 23.94 m2g-1, the initial discharge and charge capacities of the MoS2/CNF electrode are 1 155 and 799 mAh g-1at 100 mA g-1, respectively. Because the lithium storage in the 1D MoS2/CNFs obeys the insertion/extraction Li+storage mechanism. The Coulombic efficiency is 69.2%, and the irreversible capacity loss is mainly attributed to the formation of solid electrolyte interphase (SEI) film[25]. The reversible capacity of the MoS2/CNFs remains stable at 700 mAh g-1at 100 mA g-1after 50 cycles (Fig. 6a), which is much higher than that of bulk MoS2(360 mAh g-1, Fig. 6b). Fig. 6c indicates that the MoS2/CNFs also show better cycling performance than bulk MoS2. Importantly, the 1D MoS2/CNF sample exhibits distinguished high-rate capabilities at different current densities from 50 to 1 000 mA g-1(Fig. 6d). The reversible capabilities of the MoS2/CNF sample is up to 530 and 450 mAh g-1at 500 and 1 000 mA g-1, respectively, which are significantly higher than the bulk MoS2(164 mAh g-1at 1 000 mA g-1). And when the current rate is again reduced back to 50 mA g-1, the reversible capacity can be recovered and maintains at 769 mAh g-1. In addition, the MoS2/CNF sample keeps no attenuation at 1 000 mA g-1after 200 cycles (Fig. 6e), indicating the excellent cycle performance.

To explore the reasons of the excellent electrochemical performance of the MoS2/CNF sample, the EIS measurement was employed after rate cycles (Fig. 6f) and the equivalent circuit diagram of AC impedance is shown in Fig. 7.

Fig. 7 The equivalent circuit diagram of AC impedance for the MoS2/CNF and bulk MoS2 samples.

The two plots both contain two semicircles which locates at high and medium frequencies region and a straight line locates at low frequencies region. After data fitting, the solid electrolyte interface resistance (Rf, 4.07Ω) and charge transfer resistance (Rct,33.65Ω) of the MoS2/CNFs are obtained, which are much lower than those of bulk MoS2(Rf,16.24Ω;Rct, 228.3Ω). The fast diffusion of electron and high electrochemical activity for lithium storage in the MoS2/CNFs are clearly demonstrated by these results. The MoS2nanosheets homogenously encapsulated in the carbon matrix provide more active sites for lithium ions. The one-dimensional and small diameter microstructures of CNFs provide a short pathway for Li+transportation. And the intertwined carbon nanostructure forms a well conductive network, which also plays an important role as a buffering effect to effectively decrease the volume expansion during charge-discharge cycling[26].

4 Conclusions

We have developed an efficient approach to fabricate 1D MoS2/CNFs by the electrospinning method. The free-standing MoS2/CNF film can be easily cut into flexible tablet and directly used as binder-free anode for lithium storage. The unique structures offer the 1D MoS2/CNFs a high gravimetric capacity (700 mAh g-1at 100 mA g-1) and good high-rate performance (450 mAh g-1at 1 000 mA g-1) along with a stable cycle property. We believe that this simple and efficient method can be further extended to fabricate various active anode materials (such as transition metal oxides, Sn and Si) encapsulated in carbon for broad applications in batteries, supercapacitors and catalysts.