Introduction

With the rapid development of portable electronic devices and electric vehicle industries, LIBs [1,2,3,4], lithium-sulfur batteries [5,6,7,8], and fuel cells are regarded as energy storage devices with great potential for development. Among them, LIBs have been widely used in various industries because of their long life span, high energy density, light weight, and environmental protection [9]. Graphite, employed as the anode material of current LIBs, own good stability, but its inherent low specific capacity also seriously restricts the development of high-energy–density LIBs. In order to meet the growing application demands, many researchers have invested a lot of time in developing high-performance anode materials, hoping to replace traditional graphite and further improve the electrochemical performance of LIBs [10, 11].

TMCs, as the promising candidate, such as FeS2, CoS2, MoS2, ReS2, NiS2, and NiS, make full use of all oxidation states of transition metals and result in a rich reversible redox reaction, which makes them own high theoretical capacity and has attracted extensive attention [12,13,14,15,16,17,18,19]. Nonetheless, the inherently low electrical conductivity, the structural pulverization, and the large volume change of TMCs lead to rapid capacity decay during charge and discharge, which significantly restricts the cycling stability of TMCs in LIBs [20,21,22]. To solve the above problems, lots of strategies to enhance the performance of lithium storage have been proposed. One effective approach is to design nanostructures with a high specific surface area, thereby shortening the transfer length of Li+ intercalation [23,24,25,26,27,28,29]. For example, Hao et al. reported that a material composed of MnS nanocrystals only provided a specific capacity of 287 mA h g−1 at 100 mA g−1 [30]. The problem of capacity fading caused by its pulverization, inherent low electronic conductivity, and aggregation of metal particles during cycling cannot be solved by mere nanostructure engineering. So another approach of integrating nanostructured TMCs with carbon-based materials has been proposed to solve the aforementioned issues [31,32,33,34,35,36]. In this method, in addition to the advantages of TMCs, the introduced carbon component provides a stable conductive framework, which can enhance the conductivity of the material and well alleviate the volume change during charging and discharging, thereby improving the electrochemical performance.

Herein, the yolk-shell CoS2@FeS2@NC hollow microspheres are successfully synthesized by a multistep process containing the fabrication of solid Co-glycerate (Co-G) precursor, the growth of Fe-glycerate (Fe-G) nanosheets, polydopamine (PDA) coating, and subsequent sulfurization process. The CoS2@FeS2@NC hybrid materials have the following advantages: (1) the yolk-shell structure can effectively buffer the volume change, provide more active centers, and greatly shorten the transfer paths of lithium-ions since the large specific surface area plays an important role in facilitating ion transportation. (2) Compared to single metal sulfides, the integration of CoS2 with FeS2 into nanostructure not only preserves advantages of individual components, but also effectively exploits the synergistic effect between different metal sulfides, which can achieve excellent electrochemical performance [37,38,39]. (3) The coating of the NC layer not merely effectively enhances the electrical conductivity and structural stability of the material, but also inhibits the exfoliation and aggregation of FeS2 nanoparticles during the lithium insertion/extraction process. Therefore, after 100 cycles, the specific capacity of the CoS2@FeS2@NC composites can be maintained at 614 mA h g−1 at 1 A g−1, showing high specific capacity and outstanding cycling stability.

Experimental section

Preparation of Co-G microspheres

The synthesis process of Co-G microspheres is referred to the previous literature [40]. In a typical synthesis, Co(NO3)2·6H2O (0.546 g) was added into a mixed isopropanol and glycerol solution (240 mL) with an isopropanol/glycerol volume ration of 5:1. After rapid stirring for a few minutes, the solution turned clear pink and was transferred into a 300 mL Teflon-lined autoclave, followed by heating at 180 ℃ for 6 h. After that, the pink products were centrifugally separated and rinsed with ethanol and dried at 70 ℃ for 10 h, denoted as Co-G.

Preparation of yolk-shell Co-G@Fe-G hollow microspheres

Typically, Fe(NO3)3·9H2O (50 mg) was homogeneously dispersed in the solution containing glycerol (10 mL) and isopropanol (70 mL) by ultrasonication. Then, as-prepared Co-G (35 mg) was added into the solution and sonicated for 10 min. Finally, the resulting mixture was poured into 100 mL Teflon-lined autoclave and reacted at 190 ℃ for 10 h. The dark brown sediments were obtained through centrifugation, washed three times with ethanol and deionized, and dried at 70 ℃ overnight, denoted as Co-G@Fe-G.

Preparation of yolk-shell CoS 2 @FeS 2 @NC hollow microspheres

To coat the NC layer, the prepared Co-G@Fe-G microspheres (50 mg) were uniformly dispersed into 100 mL Tris-buffer solution (10 M) and sonicated for 30 min. Then, dopamine hydrochloride (25 mg) was added into the suspension, which was stirred at 25 ℃ for 6 h. The suspension was suction filtered, washed with deionized water, and dried at 70 ℃ overnight to obtain black products. Finally, as-prepared black products (0.1 g) and sublimed sulfur (0.3 g) were thoroughly ground for 5 min. In nitrogen atmosphere, the mixtures were placed in a small quartz boat and heated at 350 ℃ for 4 h with a ramp rate of 2 ℃·min−1. After cooling, the final products were obtained, denoted as CoS2@FeS2@NC. For comparison, yolk-shell CoS2@FeS2 microspheres were synthesized with the same procedure except for the absence of the PDA coating process.

Results and discussion

The fabrication procedure of yolk-shell CoS2@FeS2@NC hollow microspheres is vividly depicted in Fig. 1. Firstly, Co-G microspheres precursor samples are synthesized and Fe-G nanosheets are sequentially grown on the surface of Co-G microspheres via the solvothermal process. The phenomenon of uniform growth of Fe-G nanosheets on the surface of the Co-G microspheres can be explained by the previously reported heterogeneous nucleation and self-templated evolution processes. [40,41,42]. In the coexistence of isopropanol (IPA) and glycerol, iron ions are firstly coordinated with IPA molecules and then uniformly grown on the prepared Co-G microspheres. Under the current synthesis conditions, with the progress of the reaction, the Fe-IPA (forms of coordination binding of iron ions with isopropanol molecules) layers gradually transform into Fe-G nanosheets. While the Fe-G nanosheets are growing, Co-G begins to decompose under the influence of Fe3+. Due to the different decomposition speeds of the inner and outer surfaces, the yolk-shell structure is finally formed. Afterward, a layer of PDA is coated on the yolk-shell Co-G@Fe-G hollow microspheres to obtain Co-G@Fe-G@PDA hollow microspheres. Finally, the as-prepared Co-G@Fe-G@PDA composites are transformed into CoS2@FeS2@NC by in situ carbonization and sulfuration.

Fig. 1
figure 1

Fabrication process of yolk-shell CoS2@FeS2@NC hollow microspheres

Full size image

Typical scanning electron microscopy (SEM) and transmission electron microscopy (TEM) well demonstrate the morphologies and detailed structures of CoS2@FeS2@NC hollow microspheres. The as-obtained Co-G microspheres demonstrate a smooth surface and spherical morphologies with a rather uniform size of about 600 nm (Fig. 2a). The uniform Fe-G nanosheets are synthesized by the solvothermal process on the surface of Co-G nanospheres, which can be well proved by the SEM image (Fig. 2b and Fig. S1). As can be seen from Fig. 2c, the thickness of the nanosheets on the surface of the microspheres increases significantly, indicating that Co-G@Fe-G microspheres are successfully coated with PDA. Additionally, we can clearly observe the yolk-shell structure from one of the broken microspheres (Fig. S1 and Fig. 2c). The TEM images are adopted to investigate the interior structure of CoS2@FeS2@NC microspheres, which further confirms the yolk-shell and hollow structure (Fig. 3a, b). As shown in Fig. 2d, after high-temperature sulfuration, the diameter of the microspheres reduces from 850 to 750 nm before calcination. The reason for the shrinkage is that the nanosheets are partially cracked into nanoparticles at this temperature, and the external polydopamine is transformed into nitrogen-doped carbon. It is clearly seen from the TEM image that the thickness of the carbon layer is approximately 35 nm (Fig. 3c). Furthermore, it is evident from the high-resolution TEM (HRTEM) image that the lattice spacings of 0.19 nm and 0.27 nm correspond to the (220) plane of the CoS2 and the (200) plane of the FeS2, respectively, which agrees with XRD spectra (Fig. 3d). The high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image and X-ray mapping images show that Fe, Co, S, C, and N elements are homogeneously distributed throughout the microsphere (Fig. 3e). Table S1 further demonstrates the presence and content of above elements. For comparison, the morphology of CoS2@FeS2 is also illustrated by the SEM image in Fig. S2.

Fig. 2
figure 2

a SEM images of Co-G, b Co-G@Fe-G, c Co-G@Fe-G@PDA, d CoS2@FeS2@NC

Full size image
Fig. 3
figure 3

a–c TEM images of CoS2@FeS2@NC with different magnifications, d HRTEM image of CoS2@FeS2@NC, and e corresponding EDX elemental mapping images of CoS2@FeS2@NC

Full size image

Figure 4a exhibits the X-ray diffraction (XRD) pattern of CoS2@FeS2@NC composites. The characteristic peaks at 28.7°, 32.5°, 36.7°, 46.8°, 50.9°, and 64.4° can be indexed to the (111), (200), (210), (220), (221), and (321) planes of FeS2 (JCPDS#42–1340), respectively [43, 44]. The peaks located at 32.5°, 40.3°, 46.8°, 55.4°, and 60.6° perfectly correspond to the (200), (211), (220), (311), and (230) planes of CoS2 (JCPDS#41–1471), respectively [44, 45]. Simultaneously, there is no obvious C peak in the XRD results, indicating the carbon in the material mainly exists in the form of amorphous carbon. From the XRD pattern, the CoS2@FeS2@NC composites exhibit excellent crystallinity.

Fig. 4
figure 4

a XRD patterns and b Raman spectra of CoS2@FeS2@NC

Full size image

Figure 4b depicts the Raman spectrum of CoS2@FeS2@NC, in which two prominent peaks located around 1398 cm−1 and 1550 cm−1 correspond to the D band and G band, respectively. The D band is mainly related to Raman scattering introduced by internal structure defects or disorder, while the G band is chiefly due to the stretching vibration of sp2 hybrid carbon atoms [46,47,48]. The calculated value of ID/IG is relatively high 0.95. Combined with XRD analysis, it indicates that there are abundant topological defects and disorder in NC, which is favorable for lithium storage [49]. The BET analysis showed that the specific surface area of the yolk-shell CoS2@FeS2@NC hollow microspheres is 14.94 m2 g−1 and the pore size is mainly distributed around 2.77 nm and 15–85 nm (Fig. S3), which greatly shortens the diffusion length of Li+ and enhances the electrochemical performance of LIBs.

The surface chemical valence states of the CoS2@FeS2@NC nanospheres are further verified by X-ray photoelectron spectroscopy (XPS) analyses. Fig. 5a indicates that the samples are composed of Fe, Co, S, C and N five elements. Furthermore, the spectra of Fe 2p (Fig. 5b) and Co 2p (Fig. 5c) are fitted with two spin–orbit doublets and shakeup satellites (denoted as “Sat.”). From the Fe 2p spectrum (Fig. 5b), it can be seen that two distinct peaks located at 713.2 and 726.6 eV correspond to the Fe 2p3/2 and Fe 2p1/2 binding energies, respectively [44, 50]. Similarly, two typical peaks at 781.8 and 801.0 eV can be indexed to Co2+ 2p3/2 and Co2+ 2p1/2, which indicates the existence of Co2+ [44, 51]. There are three fitted peaks in the S 2p spectrum (Fig. 5d), of which the two peaks at 163.6 and 164.6 eV are assigned to 2p3/2 and 2p1/2 of the S element in the composites, demonstrating the existence of S–S bonds in the CoS2@FeS2@NC. Additionally, the peak located at 168.6 eV is attributed to the S–O bond generated by the surface oxidation of CoS2@FeS2@NC composites [15, 51]. As can be seen from Fig. 5e, the C 1 s spectrum is divided into three peaks at 284.6 eV, 285.9 eV, and 288.4 eV, indicating the presence of C–C, C-N and C-O, respectively [52]. The N 1 s XPS spectrum is curve-fitted with three characteristic peaks at 398.8 eV, 400.0 eV, and 400.9 eV, corresponding to pyridinic N, pyrrolic N, and graphitic N, respectively [43, 53].

Fig. 5
figure 5

a XPS survey scan, b Fe 2p, c Co 2p, d S 2p, e C 1 s, and (f) N 1 s high-resolution spectrum for CoS2@FeS2@NC

Full size image

The electrochemical properties of the CoS2@FeS2@NC composites is evaluated. Fig. 6a exhibits typical CV curves of CoS2@FeS2@NC for the first five cycles in the 0.01–3.0 V (vs. Li/Li+) voltage window. During the first cathodic scan, the peaks at 0.51 and 0.68 V can be attributed to the formation of a solid electrolyte interface (SEI) layer on the electrode surface [54,55,56,57]. The two reduction peaks near 1.18 and 1.29 V are caused by the insertion of Li+ into FeS2 (Eq. (1) and Eq. (2)) [58]. In addition, the reduction peaks appearing at 1.50 and 1.93 V are due to the insertion of Li+ into CoS2 (Eq. (3) and Eq. (4)) [54]. The first anodic curve revels four oxidation peaks at 1.91, 2.05, 2.30, and 2.48 V, respectively, of which the oxidation peaks located at 2.05 and 2.30 V are related to the Li+ extraction of CoS2 (Eq. (5) and Eq. (6)). The other two oxidation peaks at 1.91 and 2.48 V correspond to the formation of Li2FeS2 (Eq. (7)) and FeSy/S (Eq. (8) and Eq. (9)), respectively [55]. The two reduction peaks of CoS2 move from their previous positions to 1.26 and 1.82 V, respectively, during the subsequent cycling. Moreover, the reduction peak at 2.12 V is assigned to the formation of Li2FeS2 (Eq. (10)), while the peak at 1.37 V is ascribed to the formation of Fe/Li2S (Eq. (11)) [43, 58]. Except for the first circle, the remaining CV curves basically overlap, demonstrating the excellent cycling stability and reversibility of the CoS2@FeS2@NC electrode.

Fig. 6
figure 6

Electrochemical performance of yolk-shell CoS2@FeS2@NC hollow microspheres. a CV curves at a scan rate of 0.1 mV s−1. b The discharge–charge voltage profiles at 1 A g−1. c Rate performance and d the corresponding discharge–charge curves at different current densities. e Cycling performance and coulombic efficiency at 1 A g.−1

Full size image
$${\mathrm{FeS}}_{2}+{2\mathrm{Li}}^{+}+{2\mathrm{e}}^{-}\to {\mathrm{Li}}_{2}{\mathrm{FeS}}_{2}$$
(1)
$${\mathrm{Li}}_{2}{\mathrm{FeS}}_{2}+{2\mathrm{Li}}^{+}+{2\mathrm{e}}^{-}\to {\mathrm{Fe}+2\mathrm{Li}}_{2}\mathrm{S}$$
(2)
$${\mathrm{CoS}}_{2}+{\mathrm{xLi}}^{+}+{\mathrm{xe}}^{-}\to {\mathrm{Li}}_{\mathrm{x}}{\mathrm{CoS}}_{2}$$
(3)
$${\mathrm{Li}}_{\mathrm{x}}{\mathrm{CoS}}_{2}+{(4-\mathrm{x})\mathrm{Li}}^{+}+{(4-\mathrm{x})\mathrm{e}}^{-}\to {\mathrm{Co}+2\mathrm{Li}}_{2}\mathrm{S}$$
(4)
$${\mathrm{Co}+2\mathrm{Li}}_{2}\mathrm{S}\to {\mathrm{Li}}_{\mathrm{x}}{\mathrm{CoS}}_{2}+{(4-\mathrm{x})\mathrm{Li}}^{+}+{(4-\mathrm{x})\mathrm{e}}^{-}$$
(5)
$${\mathrm{Li}}_{\mathrm{x}}{\mathrm{CoS}}_{2}\to {\mathrm{CoS}}_{2}+{\mathrm{xLi}}^{+}+{\mathrm{xe}}^{-}$$
(6)
$${\mathrm{Fe}+2\mathrm{Li}}_{2}\mathrm{S}\to {\mathrm{Li}}_{2}{\mathrm{FeS}}_{2}+{2\mathrm{Li}}^{+}+{2\mathrm{e}}^{-}$$
(7)
$${\mathrm{Li}}_{2}{\mathrm{FeS}}_{2}\to {\mathrm{Li}}_{2-\mathrm{x}}{\mathrm{FeS}}_{2}+{\mathrm{xLi}}^{+}+{\mathrm{xe}}^{-}$$
(8)
$${\mathrm{Li}}_{2-\mathrm{x}}{\mathrm{FeS}}_{2}\to {\mathrm{FeS}}_{\mathrm{y}}+{\left(2-\mathrm{y}\right)\mathrm{S}+(2-\mathrm{x})\mathrm{Li}}^{+}+{(2-\mathrm{x})\mathrm{e}}^{-}$$
(9)
$${\mathrm{FeS}}_{\mathrm{y}}+{\left(2-\mathrm{y}\right)\mathrm{S}+2\mathrm{Li}}^{+}+{2\mathrm{e}}^{-}\to {\mathrm{Li}}_{2}{\mathrm{FeS}}_{2}$$
(10)
$${\mathrm{Li}}_{2}{\mathrm{FeS}}_{2}+{2\mathrm{Li}}^{+}+{2\mathrm{e}}^{-}\to {\mathrm{Fe}+2\mathrm{Li}}_{2}\mathrm{S}$$
(11)

Figure 6b shows the typical discharge/charge voltage profiles of the CoS2@FeS2@NC electrode at a current density of 1 A g−1 in a voltage window of 0.01–3.0 V (vs. Li/Li+). The discharge and charge capacities in the first cycle are 1162.6 mAhg−1 and 946.6 mAh g−1, corresponding to an initial columbic efficiency (CE) of 81.4%. The low CE of the battery is primarily due to the reductive decomposition of the electrolyte and the formation of the SEI layer during the charge–discharge process [59,60,61]. What is more, the reaction between lithium ions and the remaining functional groups of the PDA-derived carbon also affects CE [51]. During the second cycle, the discharge capacity reduces to 877.1 mAhg−1 and the corresponding charge capacity is 854.7 mAhg−1, resulting in a high CE of 97.5%. In the subsequent cycles, the CE of 96.8% and 98.7% are observed at the 5th and 50th cycle, respectively, which manifests excellent reversibility of the charge/discharge process.

To highlight the electrochemical performance of CoS2@FeS2@NC, the rate performance of CoS2@FeS2 is also evaluated. Fig. 6c shows the rate performance of the electrodes made of yolk-shell CoS2@FeS2@NC hollow microspheres and yolk-shell CoS2@FeS2 hollow microspheres cycled under different current densities. The results show that CoS2@FeS2@NC battery own higher capacity and better stability than CoS2@FeS2 battery at various current densities. When the current density gradually increases from 0.1 to 0.2, 0.5, 1.0, 2.0, and 5 A g−1, the corresponding specific capacities decrease from 1097 to 1076, 922, 752, 549, and 209 mAhg−1, respectively. In addition, when the current density is restored to 0.1 A g−1, the specific capacity can quickly return to 1130 mAhg−1, demonstrating that the CoS2@FeS2@NC electrode has excellent reversibility for lithium storage. It can be seen that at the current density of 0.1 A g−1, the capacity of the last twenty cycles is higher than that of the first ten cycles. The main reason for this phenomenon is that the electrolyte penetrates into the inner cavity to activate the active material [14, 62,63,64]. The discharge–charge voltage curves at different current densities further confirm the excellent rate capability of the CoS2@FeS2@NC electrode (Fig. 6e).

Moreover, the CoS2@FeS2@NC electrode also shows excellent cycling stability. Fig. 6e exhibits the cycling performance over 100 cycles at a current density of 1 A g−1. The specific capacity of the CoS2@FeS2 electrode rapidly decays to 76 mAh g−1 along with a CE of about 97% after only 20 cycles. In contrast, the CoS2@FeS2@NC composite can still stabilize at a high specific capacity of 614 mAh g−1 without significant decrease after 100 cycles. The initial CE of 81% gradually increases and stabilizes at more than 98% after 16 cycles, further proving the outstanding cycling stability of CoS2@FeS2@NC. The performance improvement is due to the introduction of NC layer. On the one hand, the NC layer can prevent structural exfoliation of FeS2 nanoparticles and inhibit the dissolution of polysulfides. On the other hand, the NC layer coated on the surface of FeS2 nanoparticles can improve the conductivity and electrochemical activity.

Through electrochemical impedance spectra (EIS), we can further study the reason why the CoS2@FeS2@NC composites have better electrochemical performance than CoS2@FeS2 for LIBs. The Nyquist plots (Fig. S4) of CoS2@FeS2@NC and CoS2@FeS2 are composed of semicircles and oblique straight lines. The semicircle in the high-frequency region and the oblique straight lines in the low-frequency region correspond to the charge transfer resistance between electrolyte and electrode material (Rct) and Li+ diffusion impedance, respectively [31, 65, 66]. Obviously, in high-frequency range, the radius of the semicircle for CoS2@FeS2@NC is much smaller than that of the CoS2@FeS2 electrode, which indicates that the CoS2@FeS2@NC composite has better electrochemical reaction kinetics and faster charge transfer speed. Additionally, after 100 cycles at 1 A g−1, the charge transfer resistance becomes smaller compared with before cycling, indicating that the conductivity and electrochemical activity of CoS2@FeS2@NC and CoS2@FeS2 electrodes have been enhanced during cycling process. Compared with the previously reported TMCs-based anode materials for LIBs, the yolk-shell CoS2@FeS2@NC electrode exhibits the excellent lithium-ion property (Table S2 in the “Supplementary information”) [15, 35, 44, 51, 67,68,69,70,71,72]. The above results show that yolk-shell CoS2@FeS2@NC hollow microspheres own outstanding electrochemical performance and great application potential in lithium battery, which is consistent with our initial design.

Conclusion

In summary, yolk-shell CoS2@FeS2@NC hollow microspheres are fabricated and applied as advanced anode material for LIBs by a step-by-step strategy. This unique structure fully combines the advantages of NC coating and yolk-shell nanostructure, which markedly improve the specific capacity and cycling performance. Additionally, the FeS2 nanoparticles well constrained by external NC can effectively reduce their aggregation and exfoliation, further enhancing the cycling stability of LIBs. As advanced anode materials for LIBs, the CoS2@FeS2@NC hollow microspheres can maintain a high capacity of 614 mAh g−1 at 1 A g−1 after 100 cycles. The results show that CoS2@FeS2@NC hybrid electrode owns superior electrochemical properties and great application potential in high-performance LIBs.