Facile Synthesis of MPS3/C (M = Ni and Sn) Hybrid Materials and Their Application in Lithium-Ion Batteries.

Herein, we successfully synthesized two novel metal thiophosphites (MTPs) hybridized with carbon, that is, NiPS3/C and SnPS3/C composites, via an environment-friendly and cost-effective approach without harsh reaction conditions. Subsequently, the electrochemical performances of NiPS3/C and SnPS3/C composites have been investigated in coin-cells, and it is revealed that MTPs/C have a significantly higher Li-storage capacity and better stability compared to the MTPs without carbon. Moreover, the SnPS3/C electrode shows a lower internal resistance and a better rate performance compared to NiPS3/C. We employed extensive ex situ experiments to characterize the materials and interpreted the remarkably improved performance of MTPs/C.

Introduction

Lithium-ion batteries (LIBs) are considered to be novel substitutes of finite fossil fuels and have been developed very fast in the last few decades. Nowadays, LIBs have been applied widely in energy storage devices, portable electronics, and electric vehicles.[1−5] However, as the present anode material of commercial LIBs, graphite has a low lithiation capacity with a theoretical limitation of 372 mA h g–1 and a poor rate performance.[6] Therefore, the research of new anode materials to replace graphite is one of the most attractive topics in LIBs and considerable achievements have been made.[7−11] As reported in previous studies, double-anion MXY (M: metal and X, Y: anion ions) materials have attracted considerable attention and especially have shown some intriguing applications in electrocatalysis.[9−11] Very recently, we synthesized tin phosphides and tin chalcogenides and applied them as the anode materials for LIBs, showing considerably good performances.[12,13] Therefore, for a further step, we attempted to synthesize double-anion MPS3 (M: Ni and Sn) materials and apply them for LIBs. Previously, metal thiophosphites (MTPs) were also studied as the anode of LIBs.[14−19] For instance, Dangol et al. reported an ultrathin 2D NiPS3 nanosheet electrode that gave a high reversible capacity of 796 mA h g–1 at current density of 0.1 A g–1.[16] Du et al. reported a novel graphene-supported Ni0.5Co0.5PS3 nanohybrid that showed a capacity of 456 mA h g–1 at 0.5 A g–1.[17] In 2018, Edison et al. for the first time reported novel SnPS3 as the anode of LIBs that delivered 532 mA h g–1 at 0.1 A g–1.[18] In contrast, our MPS3 materials are fabricated in a much simpler way with a large size of ∼10 μm, but showing fairly good lithiation and delithiation properties as anodes of LIBs compared to those nanosized materials reported in the literature, for example, a capacity of 683 mA h g–1 over 200 cycles for our SnPS3/C electrode.

As we know, the introduction of carbon could be a useful route to improve the Li-storage properties in LIBs.[20−25] In this work, we provide the first report about successful preparation of two novel NiPS3/C and SnPS3/C composites via a facile two-step process, and applied them as the anodes of LIBs to investigate the lithium storage properties in a coin-cell configuration. For comparison, we also fabricated NiPS3 and SnPS3 without carbon and measured their electrochemical properties. Interestingly, we find that MTPs/C has a significantly higher lithiation capacity and better stability compared to the MTPs without carbon. In order to understand that, we applied X-ray diffraction (XRD), Raman spectroscopy, X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), and high-resolution transmission electron microscopy (HRTEM) to study the morphologies and structures of MTPs/C. We find that the existence of carbon improves the structural stability of MTP anodes, and also enhances the conductivity and capability of the lithium storage. That is because carbon can serve as a buffer to cushion the stress induced on the anode materials and mitigate the aggregation of the materials during the cycling, as well as increase the structural stability and enhance the kinetics of charge transfer during the electrochemical process.

In our studies, the SnPS3/C anode shows a lower internal resistance and a better rate performance compared to NiPS3/C, due to a more uniform carbon distribution. In addition, many synthetic pathways such as the chemical vapor transport method, hydrothermal method, chemical exfoliation method, high pressure method, and so forth were utilized to synthesize the abovementioned double-anion TMPC energy nanomaterials, while these synthesis routes have shown some shortcomings, for example, harsh reaction conditions, lack of phase purity, or difficulty in tuning the atomic ratio of the final sample.[16,17,26] Herein, MTPs/C composites were successfully grown in an evacuated quartz tube via a simple high-temperature reaction (see Scheme ). This kind of confined reaction system utilized here plays a crucial role in the formation of a MTPs/C sample due to a strong inhibition effect on the loss of volatile P and S elements. Compared with conventional synthesis methods for double-anion materials, the facile two-step process employed here for synthesis of MTPs/C shows significant advantages, that is, no complex equipment and harsh reaction conditions.

Scheme 1

Schematic Illustration of Synthetic Routes for MPS3/C (M = Ni and Sn) Composites

Results and Discussion

The crystal structures of NiPS3 and SnPS3 are depicted as insets to Figures a and 2a. As we have seen, the metal and phosphorus atoms are sandwiched between the layers of sulfur atoms. Note that these materials possess typical van der Waals’ layered structures, suitable for application as energy storage materials.[27−29] The XRD patterns for the sample evaluation on phase purity and crystallinity are shown in Figures a and 2a: monoclinic NiPS3/C (JCPDS card no. 33-0952, C2/m, a = 5.81 Å, b = 10.07 Å, c = 6.63 Å, β = 106.98°); monoclinic SnPS3/C (JCPDS card no. 65-0647, P21/c, a = 6.55 Å, b = 7.49 Å, c = 11.31 Å, β = 124.19°). Note that quality of XRD patterns for NiPS3/C is better than that for SnPS3/C, in terms of less noise, due to the existence of more carbon in the SnPS3/C composite confirmed by EA analysis and SEM images. Furthermore, the Raman spectrum of NiPS3/C and SnPS3/C samples (Figures b and 2b) shows two characteristic peaks located around 1350 and 1600 cm–1, attributed to the G- and D-bands of carbon, respectively. The G-band corresponds to an E2g vibration mode of carbon and associated with the vibration of carbon sp2 atoms in a 2D hexagonal lattice, while the D-band is attributed to the vibration of carbon atoms with dandling bonds in plane terminations of disordered carbon. The G- and D-band features confirm that the coated carbon exists as the form of carbon in our samples.[30−32] Elemental analysis reveals that the accurate carbon content of NiPS3/C and SnPS3/C are 9.74 and 15.68 wt %, respectively.

Figure 1

(a) Powder XRD pattern of the NiPS3/C composite. The inset illustrates the crystal structure of NiPS3. (b) Raman spectrum of the NiPS3/C composite. (c) SEM, (d) TEM, and (e) HRTEM images of the NiPS3/C composite. X-ray photoelectron spectral regions for (f) Ni 2p, (g) P 2p, and (h) S 2p levels.

Figure 2

(a) Powder XRD pattern of the SnPS3/C composite. The inset illustrates the crystal structure of SnPS3. (b) Raman spectrum of the SnPS3/C composite. (c) Low-magnified and (d) high-magnified SEM images of the SnPS3/C composite. (e) TEM image of the SnPS3/C composite. X-ray photoelectron spectral regions for (f) Sn 3d, (g) P 2p, and (h) S 2p levels.

Furthermore, the morphology of the as-synthesized composites was observed using a scanning electron microscope. The microstructure of NiPS3/C shows a feature of typical 2D stacked NiPS3 plates with carbon wrapped around, as shown in Figure c. The TEM image (Figure d) of the product indicates that the dark NiPS3 is wrapped with carbon, consistent with the SEM image very well. The HRTEM image of the NiPS3/C composite in Figure e shows the clear crystal lattice with a d-spacing of 0.287 nm, corresponding to the (130) facet of NiPS3 crystals. XPS was carried out to further analyze the bonding characteristics as well as the composition of as-synthesized samples. The high-resolution XPS spectra shown in Figure f reveal a positively charged state of Ni species corresponding to the main peak at ∼854.0 eV, and the other two satellite peaks located at 859.0 and 864.1 eV are of Ni 2p.[33] Specifically, the main positively charged state peak can be resolved into two peaks at 855.4 and 854 eV, attributed to the species of Ni3+ and Ni2+, respectively.[34] The P 2p core level XPS spectrum (Figure g) has its spin–orbit doublet in the 2p3/2 and 2p1/2 peaks positioned at 131.9 and 132.7 eV, respectively, while the peaks situated at 134.3 and 135.1 eV can be ascribed to 2p3/2 and 2p1/2 of the P species as a positively charged state.[35] In addition, Figure h shows the S 2p spectrum, where S 2p3/2 and S 2p1/2 are located at 162.4 and 163.6 eV, respectively.[36,37] The C 1s spectrum is shown in Figure S1 (Supporting Information), where the main peak at 284.8 eV is attributed to the graphitic carbon coating of NiPS3/C and the small peak at ∼286.4 eV is resulted from a slight surface oxidation.[37,38]

The morphology of the as-synthesized SnPS3/C composite was also characterized, and Figure c–e shows the representative SEM and TEM images of the SnPS3/C composite with different magnifications. It is found that the microstructure of the SnPS3/C composite shows an irregular aggregation between the matrix and carbon as compared to the NiPS3/C composite, due to the existence of more carbon in the SnPS3/C composite. Figure S2 depicts the corresponding EDX mapping characterization of the as-synthesized SnPS3/C composite, and the images confirm the uniform distributions of tin, phosphorus, and sulfur elements, which are evenly encapsulated in the carbon layers. Figure f shows the XPS spectrum of the Sn 3d level, where the species at 486.4 and 495.1 eV are assigned to Sn2+ 3d5/2 and 3d3/2, while the other two species at 487.5 and 496.5 eV are assigned to the more positive state of the Tin element, such as Sn3+.[39]Figure g shows the P 2p spectrum. The outline of our P 2p spectrum is very similar to that of the SnPS3 material synthesized by Edison et al.[18] However, the fitting of P 2p in Edison et al.’s work is obviously wrong as the intensity of the P 2p1/2 peak is much higher than that of the P 2p3/2 peak, but actually the intensity of P 2p1/2 should be a half of that of P 2p3/2. We fit the P spectrum correctly in our work as four species, assigned to two groups of 2p3/2 and 2p1/2 doublets shown in Figure g, where the intensity of 2p1/2 is exactly a half of that of 2p3/2.[35] Edison et al. made the similar mistake for S 2p fitting as well, where the S 2p1/2 is much bigger than S 2p3/2.[18] In Figure h, our S 2p spectrum shows a correct fitting, where the S 2p3/2 and 2p1/2 peaks are located at 162.3 and 163.5 eV, respectively, and the lower binding energies indicate a negatively charged S species in our SnPS3/C sample. Thus, it is indicated that the charges transfer from the Sn and P elements to S in the SnPS3/C composite.[36,37] Figure S3 in Supporting Information shows C 1s of SnPS3/C, very similar to that of NiPS3/C.

The suitability of the NiPS3/C composite to host Li ions was studied in half-cell configurations in a coin-cell assembly. The electrochemical properties of the as-synthesized product were further tested by cyclic voltammetry (CV) between 0.01 and 3.00 V at a scan rate of 0.1 mV s–1, and the initial five consecutive cycles are shown in Figure a. The cyclic voltammogram reveals the initial lithiation and conversion reaction of the NiPS3/C electrode occurring at a voltage of 1.7 and 1.1 V versus Li/Li+ in the first cathodic cycle. In the anodic scanning curve, three oxidation peaks located at 1.3, 1.9, and 2.3 V correspond to the lithium-ion delithiation reaction.[40−42] From the charge–discharge voltage profiles of the NiPS3/C composite electrode at 0.5 A g–1 in the potential range of 0.01–3.00 V (vs Li/Li+) for the 1st, 2nd, 3rd, and 100th discharge–charge cycles (Figure b), an obvious platform at ∼1.2 V can be observed in the first discharge process, consistent with the above CV data. During the first charge process, a step located at 2.2 V is observed, followed by plateaus moving to higher potential gradually in the subsequent cycles. Figures c and S4 display the representative cycling performance of the NiPS3/C composite electrode and NiPS3 electrode at a specific current of 0.5 A g–1, respectively. The capacities and currents were calculated based on the mass of NiPS3/C and NiPS3 anode materials in the electrode. The NiPS3/C composite electrode exhibits a capacity of about 441 mA h g–1 over 200 cycles (Figure c), while the NiPS3 electrode shows sustained capacity fading during the electrochemistry cycling process and decreased to 172 mA h g–1 after 50 cycles (Figure S4). In comparison to the NiPS3 electrode, the NiPS3/C electrode evidently shows a better electrochemical performance. As for the initial capacity fading trend of the NiPS3/C electrode (Figure c), it is likely due to the capacity losses incurred by the degradation of active materials in the electrolyte during the cycling process. The corresponding electrochemical mechanism for the NiPS3 electrode has been reported in previous literature: NiPS3 + xLi+ + xe– ↔ LiNiPS3, LiNiPS3 + yLi+ + ye– → 3Li2S + Ni + P and P + yLi+ + ye– → LiP.[14,15] Figure S5 in Supporting Information depicts the excellent rate capability of the NiPS3/C composite electrode, although the cycling stability could be further improved. Recently, Dangol et al. reported ultrathin 2D NiPS3 nanosheets that were obtained by liquid-phase exfoliation, which showed a high reversible capacity of 796 mA h g–1 at a current density of 0.1 A g–1 after 150 cycles.[16] Moreover, Du et al. reported a novel graphene-supported Ni0.5Co0.5PS3 nanohybrid electrode that shows a capacity of 456 mA h g–1 after 500 cycles at 0.5 A g–1,[17] which is comparable to that of the NiPS3/C composite electrode in this work.

Figure 3

(a) First five voltammograms of the NiPS3/C composite electrode in the voltage ranging from 0.01 to 3 V at a scan speed of 0.1 mV s–1. (b) Galvanostatic charge–discharge curves of the NiPS3/C composite electrode at a current density of 0.5 A g–1. (c) Cycling performance of the NiPS3/C composite electrode at a current density of 0.5 A g–1.

Similarly, the electrochemical performance of the SnPS3/C composite was also investigated in half-cell configurations in a coin-cell assembly. The first five consecutive CVs (Figure a) of the SnPS3/C composite electrode were collected. For the first cycle, two reduction peaks can be found at 1.6 and 1.2 V, attributed to the initial lithiation and conversion reaction of Li with SnPS3.[18,40−42] The irreversible reduction peak at 0.75 V can be ascribed to the reaction between the active materials and the electrolyte for the formation of a solid electrolyte interface membrane.[33,41,42] During the subsequent positive sweep, the oxidation peak located at 2 V corresponds to lithium-ion delithiation reaction. From the second cycle, the CV curves mostly overlap, indicating a good reversibility for the SnPS3/C electrode during the electrochemical reactions. The discharge and charge curves for the 1st, 2nd, 3rd, 10th, and 100th cycles at a current density of 0.2 A g–1 are shown in Figure b. Consistent with the above CV results, a platform at 1.25 V is observed in the discharge process. In the charge process, a weakening plateau situated at 1.7 V can be observed. Notably, the reaction mechanism of SnPS3 toward lithium also has been confirmed by earlier report: SnPS3 + xLi+ + xe– → LiSnPS3, LiSnPS3 + yLi+ + ye– → 3Li2S + Sn + P, Sn + xLi+ + xe– ↔ LiSn and P + yLi+ + ye– ↔ LiP.[18]Figures c and S6 display the cyclic performance of the SnPS3/C composite and SnPS3 electrodes at a current density of 0.2 A g–1, respectively. The capacities and currents were calculated based on the mass of the corresponding anode materials in the electrode. The SnPS3/C composite electrode delivers a capacity of about 683 mA h g–1 over 200 cycles, indicating that the SnPS3/C electrode shows a significantly enhanced cycling performance than the SnPS3 electrode. As seen in Figure S6, the SnPS3 electrode initially experiences a sharp capacity fading and only maintains a capacity of 108 mA h g–1 over 70 cycles. It is worth noting that an observable downward trend related to the delivered capacity of the SnPS3/C composite electrode could be mainly due to volume expansion accompanying the Li-alloying electrochemical process occurring in the electrode. The rate performance of the SnPS3/C electrode is distinctly better than that of the NiPS3/C electrode, as shown in Figure S7 in Supporting Information. Note that Edison et al. recently reported a SnPS3 anode that delivered a capacity of 532 mA h g–1 after 100 cycles at a current density of 0.1 A g–1.[18] It is significant that the SnPS3/C composite electrode in this work displays better specific capacitance as well as cycling stability during the charge–discharge process compared to Edison et al.’s work. Based on the above analysis, we can find that both NiPS3/C and SnPS3/C composite electrodes display a significantly higher lithiation capacity and better cycling stability compared to the corresponding MTPs without carbon in our experiments. It is because carbon serves as a buffer to cushion the stress induced on the anode and mitigates the aggregation of the materials during the cycling, thus enhancing the stability of the anode. Simultaneously, the electrons involved in electrochemical reactions can easily penetrate the carbon during the lithiation and delithiation, leading to excellent conductivity.

Figure 4

(a) First five voltammograms of the SnPS3/C composite electrode in the voltage ranging from 0.01 to 3 V at a scan speed of 0.1 mV s–1. (b) Galvanostatic charge–discharge curves of the SnPS3/C composite electrode at a current density of 0.2 A g–1. (c) Cycling performance of the SnPS3/C composite electrode at a current density of 0.2 A g–1.

In order to further understand the electrochemical performances of the NiPS3/C composite and SnPS3/C composite electrodes, electrochemical impedance spectroscopy (EIS) was applied to investigate the conductivity change of the as-prepared composite electrode before charge–discharge cycling and after 200 cycles. Figure a displays the equivalent circuit model, where Rs is the internal resistance between the electrolyte and the electrode; CPE and Rct correspond to the constant phase element and charge transfer resistance, respectively. The Nyquist plots of the as-prepared composites are shown in Figure c,e. Specifically, the semicircle located in the medium frequency region in the Nyquist plots is associated with internal resistances of the electrode.[43] Therefore, clearly the NiPS3/C and SnPS3/C composite electrodes show an observable impedance decrease after 200 cycles compared to that before cycling, suggesting that the as-prepared composites maintain good electron-transport during the cycling process.[33] This may be attributed to the existence of carbon in the composites.[44] Meanwhile, the ex situ SEM is carried out to study the structural change of the composite electrodes after 200 cycles. From the ex situ SEM images (Figure b,d), we can see that these two composites show an irregular morphology after cycling, inducing improved electron-transport properties due to the good contact between NiPS3 (or SnPS3) and carbon. Note that electrochemical experiments show that SnPS3/C (Figure S7) is better than the NiPS3/C (Figure S5) composite in terms of the delivered Li-storage capacity. From the SEM images, it is found that the integration of the carbon and SnPS3 is higher than that of NiPS3 and C, thus a better conductivity for SnPS3/C during the charge–discharge process of Li-ion batteries. This also is evidenced by EIS investigations as shown in Figure c,e, clearly suggesting a lower internal resistance for the SnPS3/C electrode compared to the NiPS3/C electrode, thus more excellent electrochemical properties for the SnPS3/C electrode.

Figure 5

(a) Equivalent circuit model to fit the Nyquist plots, Rs: contact resistance between the electrode and the electrolyte; Rct: charge transfer resistance; and CPE: constant phase element. (b) SEM image of the NiPS3/C composite electrode after 200 cycles at 0.5 A g–1 and (c) electrochemical impedance spectra of the NiPS3/C composite electrode before cycling and after 200 cycles at 0.5 A g–1. (d) SEM image of the SnPS3/C composite electrode after 200 cycles at 0.2 A g–1 and (e) electrochemical impedance spectra of the SnPS3/C composite electrode before cycling and after 200 cycles at 0.2 A g–1.

Conclusions

In this work, we utilized an environment-friendly and cost-effective route for the synthesis of two novel NiPS3/C and SnPS3/C composites. Taking advantage of the carbon layer, the MTPs/C composites show obviously enhanced Li-storage properties than the corresponding unmodified MTPs. Specifically, the NiPS3/C composite electrode delivers a specific capacity of 441 mA h g–1 after 200 cycles at a current density of 0.5 A g–1, while the SnPS3/C composite electrode shows 683 mA h g–1 over 200 cycles at 0.2 A g–1, respectively. A better rate performance for the SnPS3/C electrode compared to NiPS3/C is because the aggregation between SnPS3 and carbon is more homogeneous than that in the NiPS3/C composite, therefore a better conductivity for SnPS3/C. Moreover, ex situ EIS as another proof also showed lower internal resistances for the SnPS3/C electrode than the NiPS3/C electrode. These results highlight the introduction of carbon as the strategy to achieve high-performance anodes for LIBs and also help us to understand the mechanism behind.

Experimental Section

Materials

Nickel acetylacetonate [Ni(acac)2, 95% purity], tetraphenyltin [Sn(C6H5)4, 97% purity], red phosphorous [98.5% purity], and sublimate sulfur [99.95% purity] were purchased from Aladdin Company. The solvent, ethanol, was obtained from Shanghai Chemical Reagents Company, China. All reagents were used as received without further purification.

Sample Synthesis

MPS3/C (M = Ni, Sn) composites were grown via a facile two-step technical route. In a typical procedure, powder of nickel(II) acetylacetonate (95%) or tetraphenyltin (97%) was loaded into sealed quartz tubes, and heated at 873 K for 10 h. Then, the dark precipitates of Ni/C and Sn/C were collected without any purifying treatment. Stoichiometric amounts of Ni/C (or Sn/C), red phosphorus (99%), and sublimated sulfur (99%) were thoroughly mixed together in proportion, and encapsulated in evacuated quartz ampoules, then followed by a sintering procedure at 973 K for 2 days. Scheme schematically illustrates the synthesis route for MPS3/C (M = Ni, Sn) composites. After the furnace cooled down to room temperature naturally, the products were harvested carefully, washed with absolute alcohol several times, and dried in a vacuum furnace at 60 °C for subsequent characterization.

Structural Characterization

The phases and purity of the obtained products were characterized by X-ray powder diffraction on a Philips X’pert X-ray diffractometer equipped with Cu Kα radiation (λ = 1.5418 Å). The SEM images were obtained using a JEOL-JSM-6700F field-emission scanning electron microscope. The TEM images and HRTEM images were obtained on a JEOL-2010 with an acceleration voltage of 200 kV, and the samples for analysis were prepared by dipping carbon-coated copper grids into ethanol-dissolved samples, then dried under ambient conditions. XPS was performed on a Thermo ESCALAB 250. Raman spectroscopy was carried out on a JY LABRAM-HR confocal laser micro-Raman spectrometer using Ar+ laser excitation with a wavelength of 514.5 nm. The elemental ratios of Ni, Sn, P, and S were derived from XPS analysis. The weight percentage of carbon was analyzed by elemental analysis (EA, Elemental vario EL cube, Thermal Conductivity Detector) in a pure oxygen atmosphere.

Electrochemical Characterization

Coin-type 2016 cells were assembled in an argon-filled glovebox (O2, H2O < 1 ppm) with lithium foil as the anode, Celgard 2400 as the separator, and 1.0 M LiPF6 in ethylene carbonate (EC)/diethyl carbonate (DEC) (1:1 by volume) as the electrolyte. The working electrode was prepared by mixing active materials, carbon black, and polyvinylidene fluoride in a weight ratio of 6:2:2 in N-methyl-2-pyrrolidone. Then, the mixture was ball-milled for 10 h to mix uniformly, then the obtained slurry was spread on a Cu foil substrate and dried at 80 °C in a vacuum oven for 10 h. The loading mass of the NiPS3/C and SnPS3/C anode materials is around 1–1.5 mg. Galvanostatic measurements were performed using a LAND-CT2001A instrument in the potential range of 0.01–3 V (vs Li/Li+) at a designated current density at constant room temperature. CV was performed at a scan rate of 0.1 mV s–1 with an electrochemical workstation (CHI660E).