Porous Multicomponent Mn-Sn-Co Oxide Microspheres as Anodes for High-Performance Lithium-Ion Batteries.

Porous multicomponent Mn-Sn-Co oxide microspheres (MnSnO3-MC400 and MnSnO3-MC500) have been fabricated using CoSn(OH)6 nanocubes as templates via controlling pyrolysis of a CoSn(OH)6/Mn0.5Co0.5CO3 precursor at different temperatures in N2. During the pyrolysis process of CoSn(OH)6/Mn0.5Co0.5CO3 from 400 to 500 °C, the part of (Co,Mn)(Co,Mn)2O4 converts into MnCo2O4 accompanied with structural transformation. The MnSnO3-MC400 and MnSnO3-MC500 microspheres as secondary nanomaterials consist of MnSnO3, MnCo2O4, and (Co,Mn)(Co,Mn)2O4. Benefiting from the advantages of multicomponent synergy and porous secondary nanomaterials, the MnSnO3-MC400 and MnSnO3-MC500 microspheres as anodes exhibit the specific capacities of 1030 and 750 mA h g-1 until 1000 cycles at 1 A g-1 without an obvious capacity decay, respectively.

Introduction

Recently, with the increasing attention on clean energy and more demand for high energy density of various portable devices and electric vehicles, green energy storage and conversion devices, such as lithium-ion batteries (LIBs),[1−5] fuel cells,[6−10] and supercapacitors,[11−14] have been widely studied and applied. Among them, LIBs have been attracting researchers’ interests because of their high capacity and long cycle life.[15−17] In current commercial LIBs, graphite has been used as a classic anode material because of its excellent cycle performance and low charging/discharging potential.[18,19] However, the disadvantage of its low capacity density limits its wide application in high-performance LIBs.[20,21] Therefore, the search for a new generation of electrode materials with higher energy density and excellent cycle stability has become a top priority.[22,23] Among the numerous electrode materials for anodes, metal oxides are expected to be a promising anode material for replacing graphite because of their high theoretical capacity and good electrochemical properties.[24−26] In particular, binary metal oxides including cobalt or manganese have been constantly studied as anodes for next-generation LIBs because of their high theoretical capacity, low cost, low discharge plateau (0.3–0.6 V), and synergistic effects.[27−29] Yang’s group reported CoMn2O4 nanofibers via an electrospinning method combined with heat treatment, showing a reversible capacity of 526 mA h g–1 at 400 mA g–1 after 50 cycles.[30] Mesoporous NiCo2O4 microspheres synthesized by a facile solvothermal method with pyrolysis could deliver 1198 mA h g–1 after 30 cycles at 200 mA g–1.[31] Uniform hierarchical porous MnCo2O4 microspheres were also prepared via a solvothermal process with a post-annealing treatment, maintaining a specific capacity of 740 mA h g–1 after 1000 cycles.[32] Those porous binary metal oxides as anodes exhibited enhanced electrochemical properties because of the higher surface area, short lithium-ion transport pathway, and synergistic effect between cobalt and manganese oxides. However, further improvement of their electrochemical performances is still required for extensive practical application.

On the other hand, tin-based materials such as SnO2,[33,34] Zn2SnO4,[35] Co2SnO4,[36] and ASnO3 (A = Mn, Zn, Ca, and Co)[37,38] are also other promising anodes in LIBs, owing to their high theoretical capacities and rich electrochemical activities. Like most of the metal oxide electrode materials, the large volume variations during alloying or conversion reactions could cause the pulverization of electrodes, eventually leading to capacity deterioration and poor cycle performance. In order to overcome these thorny problems, it is a more effective strategy to construct a unique pore micro/nanostructure material or to introduce buffer matrices. One of the more mature methods in research is to design micro/nanostructured materials by self-assembly adapting to the volume expansion of lithium-ion insertion/extraction, such as porous microstructures,[39] nanotubes,[23] and nanospheres.[13] In addition, the hybridization of different composition nanostructured tin oxide, cobalt oxide, solid-solution CoSnO3, MnSnO3, and Co2SnO4 has also been studied to achieve the expected electrochemical performances for lithium-ion storage.[40−42] A hollow CoO-in-CoSnO3 nanostructure constructed by atomic layer deposition could maintain a specific capacity of 695.7 mA h g–1 after 100 cycles as an anode material of LIBs.[43] However, there is no report on the hybridization of Mn–Co metal oxides and tin-based oxides to improve their electrochemical performances through the multicomponent synergy. Therefore, it is very interesting to design and synthesize a multicomponent metal oxide including cobalt, manganese, and tin and to further study their electrochemical properties.

Herein, we will report porous heterogeneous multicomponent Mn–Sn–Co oxide microspheres (MnSnO3–MC400 and MnSnO3–MC500) fabricated by a solvothermal process followed with controlling pyrolysis of CoSn(OH)6/Mn0.5Co0.5CO3 precursor at different temperatures in nitrogen. As expected, the as-prepared microspheres as anode materials for LIBs exhibit enhanced electrochemical performances benefiting from the multicomponent synergy and porous secondary nanomaterials.

Results and Discussion

Characterizations of MnSnO3–MC400 and MnSnO3–MC500 Microspheres

The schematic illustration of the synthesis process of MnSnO3–MC400 and MnSnO3–MC500 is depicted in Scheme . The CoSn(OH)6 nanocubes were prepared in a stoichiometric coprecipitation method in an alkaline aqueous solution containing Sn4+ and Co2+. Then, the CoSn(OH)6 nanocubes were coated with MnCo-carbonate through a solvothermal treatment to obtain the CoSn(OH)6/Mn0.5Co0.5CO3 precursor. During the solvothermal process, the uniformly mixed microspheres are finally formed by diffusion because of the difference in the bonding energy between OH– and CO32– to metal ions. Subsequently, the MnSnO3–MC400 and MnSnO3–MC500 microspheres were obtained by controlling pyrolysis of CoSn(OH)6/Mn0.5Co0.5CO3 in nitrogen at 400 and 500 °C, respectively.

Scheme 1

Schematic Synthesis of MnSnO3–MC400 and MnSnO3–MC500 Microspheres

The X-ray diffraction (XRD) patterns of the CoSn(OH)6/Mn0.5Co0.5CO3 precursor, MnSnO3–MC400, and MnSnO3–MC500 are all displayed to identify their crystallographic structures, as indicated in Figure . The pattern of the CoSn(OH)6/Mn0.5Co0.5CO3 precursor can be assigned to Mn0.5Co0.5CO3 (no. 167) and CoSn(OH)6 (JCPDF no. 74-0365).[44] For MnSnO3–MC400, an obvious (112) peak at 33.1° is ascribed to MnSnO3 (JCPDS no. 47-0464); the characteristic peaks at 43.76° and 63.62° show the (400) and (440) crystal planes of MnCo2O4 (JCPDS no. 23-1237), and other peaks could be indexed as (Co,Mn)(Co,Mn)2O4 (JCPDS no. 18-0410). However, for MnSnO3–MC500, the (112) peak of MnSnO3 is still present. The obvious differences are that the (311) and (400) peaks of (Co,Mn)(Co,Mn)2O4 are converted into MnCo2O4, the (220) and (202) peaks of (Co,Mn)(Co,Mn)2O4 merge into the peak (220) of MnCo2O4, and the (511) peak of MnCo2O4 is formed, indicating that more MnCo2O4 is generated. Thermogravimetric (TG) analysis (Figure S1) was carried out to explore the calcination conditions of the CoSn(OH)6/Mn0.5Co0.5CO3 precursor. The mass loss below 150 °C (a stage) is attributed to the moisture volatilization, which has been physically and chemically adsorbed in air and the residual organic solvent volatilization. The (b) stage is attributed to the decomposition loss of CoSn(OH)6 converted to CoSnO3 and the thermal decomposition of Co0.5Mn0.5CO3.[45] In this process, with the decomposition of Co0.5Mn0.5CO3, the manganese mass gradually increases, while the electronegativity of Mn (1.5) is lesser than Co (1.8), which causes CoSnO3 to gradually transform into MnSnO3. It is very interesting to note that there is a weight loss at 400–500 °C (c stage), which is ascribed to the part conversion of (Co,Mn)(Co,Mn)2O4 into MnCo2O4 accompanied by a small amount of gas release. Thus, the content of (Co,Mn)(Co,Mn)2O4 in MnSnO3–MC400 is higher than that of MnSnO3–MC500, which could influence the electrochemical properties (Table S1). Furthermore, the elemental compositions of MnSnO3–MC400 and MnSnO3–MC500 were further confirmed by energy-dispersive X-ray spectroscopy (EDXS; Figure S2a,b), and the Co/(Mn + Sn) molar ratio is close to 1:1, being consistent with the initial reactants.

Figure 1

XRD patterns of CoSn(OH)6/Mn0.5Co0.5CO3 and the calcination products at different temperatures.

X-ray photoelectron spectroscopy (XPS) measurement was used to probe the elemental composition and valence states of the metal ions in MnSnO3–MC400 and MnSnO3–MC500. The overall XPS spectra (Figure S3) confirm the presence of Mn, Co, Sn, and O elements. As displayed in Figure a, two peaks observed at 780.4 and 796.1 eV match with the spin orbit peak of Co 2p3/2 and Co 2p1/2 for MnSnO3–MC400, accompanied by two prominent shake-up satellite peaks at 786.6 and 803.6 eV, respectively, where Co 2p3/2 at 780.4 eV is close to CoO.[41] The binding energy difference between the two spin orbits and the adjacent satellite peaks is 6.2 and 7.5 eV, respectively, indicating that the Co element is +3, +4 valence.[46] The results confirmed that the Co elements in MnSnO3–MC400 and MnSnO3–MC500 coexist with +2, +3, and +4. Figure b shows the Mn 3s spectrum showing multiple splitting peaks. It can be seen that the splitting peak magnitude of the Mn 3s spectrum is 5.43 eV for MnSnO3–MC400 and 5.49 eV for MnSnO3–MC500, showing that the average valence of Mn in MnSnO3–MC400 with +2.7 is higher than that of MnSnO3–MC500 with +2.58.[47,48] As shown in Figure c,d, Mn 2p3/2 at 641.9 eV and Mn 2p1/2 at 653.6 eV are ascribed to the Mn 2p spectrum. After refined fitting, the Mn 2p spectrum could be divided into six peaks. The peaks at 641.5 and 653.4 eV belong to the Mn 2p3/2 and Mn 2p1/2 of Mn2+, and 642.6 and 654.6 eV are attributed to the Mn 2p3/2 and Mn 2p1/2 of Mn3+, respectively.[10,19] Furthermore, the other two peaks (642.8 and 654.4 eV) could be ascribed to Mn4+.[45,48]Figure e,f shows the high-resolution spectra of Sn 3d for MnSnO3–MC400 and MnSnO3–MC500. The characteristic signals of Sn 3d5/2 and Sn 3d3/2 are found to be located at 486.6 and 495 eV for MnSnO3–MC400 and 486.4 and 494.8 eV for MnSnO3–MC500, indicating the existence of Sn4+.[37,49] Besides, compared with MnSnO3–MC500, the binding energy of Sn4+ in MnSnO3–MC400 is relatively high.[43]

Figure 2

XPS survey high-resolution spectra of MnSnO3–MC400 and MnSnO3–MC500: (a,b) for Co 2p and Mn 3s of MnSnO3–MC400 and MnSnO3–MC500, (c,e) for Mn 2p and Sn 3d of MnSnO3–MC400, and (d,f) Mn 2p and Sn 3d of MnSnO3–MC500.

The morphologies of the as-synthesized products were first characterized by the field emission scanning electron microscopy (FESEM). As exhibited in Figure a, the CoSn(OH)6 nanocubes with the edge around 200 nm could be observed, while the CoSn(OH)6/Mn0.5Co0.5CO3 precursor shows microsphere morphology with a diameter of 700 nm or so after solvothermal treatment (Figure b). Figure c,d shows the highly uniform and rough surface microspheres of MnSnO3–MC400 and MnSnO3–MC500 obtained at 400 and 500 °C with an average diameter of about ∼650 and 600 nm, respectively.

Figure 3

SEM images of (a) CoSn(OH)6, (b) Mn0.5Co0.5CO3/CoSn(OH)6, (c) MnSnO3–MC400, and (d) MnSnO3–MC500.

To further probe the microstructure of the as-obtained products, transmission electron microscopy (TEM) images were also studied. As can be seen in Figure a, the CoSn(OH)6/Mn0.5Co0.5CO3 microspheres are denser in the middle and sparse in the edges. To confirm the elemental distribution in the microspheres during the formation process, the elemental line profiles were determined (Figure b). It can be clearly seen that the tin, cobalt, and manganese elements exist in the whole microspheres from the core to the outer shell. This is because Co2+ and Sn4+ in CoSn(OH)6 are out-diffused, and Mn2+ and Co2+ in Mn0.5Co0.5CO3 attract inward because of the stronger binding energy of OH– to metal ions than CO32– during the formation process of CoSn(OH)6/Mn0.5Co0.5CO3 by solvothermal method. Figure c,d exhibits the TEM images of MnSnO3–MC400 and MnSnO3–MC500. It can be seen that hollow MnSnO3–MC400 has a smaller pore micro/nanostructure than MnSnO3–MC500, which is very different from their precursor CoSn(OH)6/Mn0.5Co0.5CO3 microspheres. More microscopic details can be observed from high-resolution TEM (HRTEM) images. It can be found that the fast Fourier transform (FFT) mode (inset of Figure S4) of the Mn0.5Co0.5CO3/CoSn(OH)6 precursor has CoSn(OH)6 and Mn0.5Co0.5CO3 diffraction rings. Figure e,f shows the selected-area electron diffraction (SAED) patterns of MnSnO3–MC400 and MnSnO3–MC500. Among them, the (Co,Mn)(Co,Mn)2O4 and MnCo2O4 structures could be indexed, respectively. As shown in Figure g, two distinct lattice spacings of 0.384 and 0.270 nm correspond to the (002) and (112) planes of MnSnO3. In addition, the (111) lattice spacing (0.482 nm) of (Co,Mn)(Co,Mn)2O4 can also be observed. The FFT mode (inset of Figure g) converted from the green square area further confirms the presence of MnSnO3, MnCo2O4, and (Co,Mn)(Co,Mn)2O4 diffraction rings. In Figure h, the HRTEM image, the (202) lattice spacing (0.301 nm) of (Co,Mn)(Co,Mn)2O4, the (112) lattice spacing (0.270 nm) of MnSnO3, and the (111) lattice spacing (0.478 nm) of MnCo2O4 can be clearly observed. It can be concluded that MnSnO3–MC400 and MnSnO3–MC500 all contain (Co,Mn)(Co,Mn)2O4, MnCo2O4, and MnSnO3. It should be noted that the difference between MnSnO3–MC400 and MnSnO3–MC500 is the content of (Co,Mn)(Co,Mn)2O4, resulting in the difference of electrochemical properties. The elemental mapping of Figure i,j shows that Mn, Co, Sn, and O are uniformly distributed in the microspheres.

Figure 4

TEM images of (a) Mn0.5Co0.5CO3/CoSn(OH)6; (b) elemental line profiles of Mn0.5Co0.5CO3/CoSn(OH)6, (c) MnSnO3–MC400, and (d) MnSnO3–MC500; SAED patterns of (e) MnSnO3–MC400 and (f) MnSnO3–MC500; HRTEM images of (g) MnSnO3–MC400 and (h) MnSnO3–MC500; and elemental mappings of (i) MnSnO3–MC400 and (j) MnSnO3–MC500.

N2 adsorption–desorption isotherm curves were also applied to study the surface areas and pore size properties of the MnSnO3–MC400 and MnSnO3–MC500 microspheres. As can be seen in Figure S5a,b, the isotherm curves of MnSnO3–MC400 and MnSnO3–MC500 could be classified as type IV with a type H1 hysteresis loop, indicating that their structures are mesoporous.[50] According to the corresponding Barrett–Joyner–Halenda plots (inset of Figure S5a,b), the average pore sizes of MnSnO3–MC400 and MnSnO3–MC500 are about 12.12 and 21.26 nm, respectively, confirming that the two samples contain mesoscale pores. It may be noted that the different pore size could cause a difference in the electrochemical properties for MnSnO3–MC400 and MnSnO3–MC500. The Brunauer–Emmett–Teller (BET) surface areas and pore volumes of the MnSnO3–MC400 and MnSnO3–MC500 were 53.213 m2 g–1 and 0.205 cm3 g–1, and 55.869 m2 g–1 and 0.283 cm3 g–1, respectively.

Electrochemical Properties

The electrochemical performances of MnSnO3–MC400 and MnSnO3–MC500 as anodes for LIBs were investigated. Figure a,b exhibits the first three cyclic voltammetry (CV) curves of MnSnO3–MC400 and MnSnO3–MC500 at a rate of 0.2 mV s–1 in the voltage range of 0.01–3.0 V. As exhibited in Figure a, in the first cycle, the smaller reduction peak around 1.85 V is ascribed to the reduction of Mn4+ and Co4+ to Mn3+ and Co3+. Three sharp peaks were observed, one peak at 1.2 V could be ascribed to the reduction of Mn3+ to Mn2+ and Co3+ to Co2+, one sharp peak at 0.69 V could be attributed to the reduction of Co2+ to metallic Co, and another sharp peak at 0.51 V is assigned to the reduction of Mn2+ and Sn4+ to metallic Mn and Sn.[44] The peak below 0.5 V is ascribed to the lithiation of Sn to Li4.4Sn alloys. A small peak at 1.03 V which disappears in subsequent cycles generally could be assigned to the irreversible decomposition of the solvent in the electrolyte and the formation of the solid electrolyte interface (SEI).[37] In the anodic scan, two broad oxidation peaks are observed at 0.52 and 0.91 V, corresponding to the delithiation of Li4.4Sn alloy.[51] At the same time, two weak peaks around 1.34 and 1.81V can be ascribed to oxidation of Sn to Sn4+ and Mn to Mn2+, respectively.[52] One broad oxidation peak at 2.08 V is observed, which corresponds to the oxidation of Mn2+ to Mn3+ and Co to Co2+. In subsequent cycles, the aforementioned peaks at 0.51 and 0.69 V shift to 0.53 and 0.72 V, respectively, which is due to the fact that the formation of Li2O and metal stimulates the microstructure of the MnSnO3–MC400 electrode after the first lithiation process, thereby accelerating the kinetics.[53] A reduction peak at 1.12 V can be found, indicating that part of Mn3+/Mn2+ is reversible. Similar electrochemical properties of MnSnO3–MC500 are exhibited in Figure b; two obvious peaks at 0.48 and 0.64 V are found in the first reduction scan, one could be ascribed to the reduction of Mn2+ to Mn and Sn4+ to Sn and the other could be assigned to the reduction of Co2+ to Co and the formation of the SEI. In addition, the peak at 1.9 V is ascribed to the reduction of Mn4+ and Co4+ to Mn3+ and Co3+, and the peak at 1.28 V can be ascribed to the reduction of Mn3+ to Mn2+ and Co3+ to Co2+. In the anodic scan, two broad oxidation peaks at 0.54 and 1.51 V correspond to the delithiation of Li4.4Sn alloy oxidation and Mn to Mn2+, respectively.[54] A peak at 1.28 V that cannot be ignored is attributed to Sn-oxidation to Sn4+.[37] A strong broad peak at 2.02 V is attributed to the oxidation of Co to Co2+ and Mn2+ to Mn3+. It is interesting that the reduction peak at 0.64 V moves to 0.74 V and the peak at 0.48 V moves to 0.51 V in the second and third circles, respectively. These oxidation/reduction processes can be described as follows

Figure 5

CV curves of MnSnO3–MC400 (a) and MnSnO3–MC500 (b) between 0.01 and 3.0 vs (Li/Li+)/V at a rate of 0.1 mV s–1, and the charge/discharge curves of (c) MnSnO3–MC400 and (d) MnSnO3–MC500 for the first three cycles at a current density of 1 A g–1 within the voltage range of 0.01–3.0 vs (Li/Li+)/V.

The charging/discharging curves of MnSnO3–MC400 and MnSnO3–MC500 for the first three cycles are exhibited in Figure c,d at a current density of 1 A g–1. The initial discharge/charge capacities of MnSnO3–MC400 and MnSnO3–MC500 are 1194.2/777.9 mA h g–1 and 1393.1/866.2 mA h g–1 with the initial Coulombic efficiency of 65.14 and 62.18%, respectively. The capacity loss is due to the side reaction of the electrolyte and the formation of the SEI film, resulting in a lower Coulombic efficiency in the first cycle. It can be clearly seen that the discharge platform is basically the same as the CV.

Figure a shows the cycle stabilities of MnSnO3–MC400 and MnSnO3–MC500 at the high rate of 1 A g–1. The capacity fading of MnSnO3–MC400 is seen obviously from 895.9 mA h g–1 (at the 2nd cycle) to 226.4 mA h g–1 (at the 100th cycle), while MnSnO3–MC500 delivers the capacity from 952.6 mA h g–1 (at the 2nd cycle) to 145.4 mA h g–1 (at the 160th cycle). Amazingly, after the lowest value, the specific capacity sequentially increases during the subsequent cycles without further decay. The specific capacity of MnSnO3–MC400 could deliver 1030 mA h g–1 at the 560th cycle and remain stable at the 1000th cycle, while MnSnO3–MC500 could show 730 mA h g–1 capacity at the 750th cycle. This phenomenon is also normal in some porous metal oxides reported.[20,53,55] First, it could be attributed to the reactivation effect induced by high-rate lithiation of porous metal oxide microspheres during the lithiation/delithiation process. Before activation, the destruction of the SEI caused by the volume expansion and the blockage of the electrolyte led to the significant capacity attenuation.[55] After reactivation, the microspheres were reconstructed as well as the optimizing stable SEI, resulting in more Co and Mn active sites, are exposed.[53] At the same time, the optimizing SEI is also beneficial to the reversible formation and decomposition of an organic polymer/gel layer in the electrolyte. Another reason is about the reversible formation and decomposition of an organic polymeric/gel-like layer from the electrolyte.

Figure 6

(a) Continued cycle performance of MnSnO3–MC400 and MnSnO3–MC500 at 1 A g–1 within the voltage range of 0.01–3.0 vs (Li/Li+)/V. (b,c) SEM images of MnSnO3–MC400 and MnSnO3–MC500 after 1000th cycles at 1 A g–1, and (d) cycle performance of the MnSnO3–MC400 with a current density from 2 to 1 A g–1 at 0.01–3.0 V and 0.01–2.1 V. The differential charge capacity vs potential curves of (e) MnSnO3–MC400 and (f) MnSnO3–MC500 at different cycles.

Compared with MnSnO3–MC500, the specific capacity of MnSnO3–MC400 is higher, which could be ascribed to the more content ratio of MnCo2O4/(Co,Mn)(Co,Mn)2O4 and the smaller pore size. SEM tests were also carried out on the recycled electrode to explore its integrity. As shown in Figure b,c, MnSnO3–MC400 and MnSnO3–MC500 could still maintain the microsphere structure after a long cycle even at a large current rate. Compared with the MnSnO3–MC500 electrode, the MnSnO3–MC400 electrode surface has obvious porous pore. This result indicated that the pores of MnSnO3–MC400 are not blocked, demonstrating the better cycle stability of MnSnO3–MC400 microspheres as anode electrodes. The organic polymeric/gel-like layer could form at low voltage and decompose at high voltage.[56] For more detailed electrochemical characteristics, the MnSnO3–MC400 electrodes were tested under the different voltage range at high-rate current density. Figure d shows the cycling performance of the MnSnO3–MC400 anode from 2 to 1 A g–1 at 0.01–3.0 V and 0.01–2.1 V. The specific capacity increases during the 200th to 450th cycles in the voltage range of 0.01–3.0 V at 1 A g–1 but does not increase when the cutoff voltage is from 0.01 to 2.1 V. This result may be due to the fact that the organic polymeric/gel-like layer could not decompose below 2.1 V. As expected, the rising trend could recover after the cutoff voltage back to 3.0 V. The above results show that the reversible formation and decomposition of the organic polymeric/gel-like layer could result in the increase of capacity. Thereafter, the charge differential capacity curves of MnSnO3–MC400 and MnSnO3–MC500 at different cycle numbers were further verified. As shown in Figure e,f, it can be seen that the MnSnO3–MC400 electrode exhibits several sharp peaks at 2.15–2.85 V after 1000 cycles, indicating the further oxidation of Co2+ or Mn2+. In addition, a blunt peak appeared at 2.85–3.0 V further indicates the decomposition of the electrolyte. It should be noted that there is no peak in the range of 2.85–3.0 V at 300th and 500th cycle of MnSnO3–MC400. Comparing the 10th and 100th cycles in the 2.15–2.85V and 2.85–3.0 V stages, it was found that the oxidation of Co and Mn gradually increased before the activation of MnSnO3–MC400 electrode, and the reversible degradation of electrolyte was continuously suppressed. A sharp peak was observed in the 2.4–3.0 V range of the MnSnO3–MC500 electrode after 10 cycles, but only a weak blunt peak was observed after 1000 cycles, indicating that MnSnO3–MC500 was favorable for lithium storage reaction at the initial cycle stage, and the pore blockage after continuous cycles was unfavorable for lithium implantation and the reversible formation and decomposition of organic polymer/gel layer. This is consistent with the SEM structure of the electrode after circulation.

The comparisons of the electrochemical properties for previous Co––Mn binary metal oxides, ASnO3 (A = Mn, Co), and MnSnO3–MC400 and MnSnO3–MC500 of this work are displayed in Table S2. It can be found that MnSnO3–MC400 and MnSnO3–MC500 as advanced anodes for LIBs exhibit high reversible capacities and excellent cycle stabilities even at the high rate of 1 A g–1, which may be due to the following reasons: first, the Co and Mn nanoparticles in Sn/Li2O and LixSn/Li2O matrices after full lithiation could work as anchors to prohibit the diffusion and coarsening of Sn nanocrystals during the dealloying process, which could improve the conversion reaction reversibility between Sn and Li2O during lithiation/delithiation process.[52,57] Second, the appropriate pore size of porous second nanomaterials could reduce the diffusion paths of Li+ and electrons, which is beneficial to enhance electrochemical kinetics, and third, the synergistic effect among the multicomponent Mn–Sn–Co oxides.

To probe the rate capabilities of the electrode, the charge/discharge tests of MnSnO3–MC400 and MnSnO3–MC500 were also evaluated at different rates from 0.1 to 2 A g–1 (Figure S6). The average discharge capacities of the MnSnO3–MC400 electrode were 1584, 1220, 834, 590, and 403 mA h g–1 at the rates of 0.1, 0.2, 0.5, 1, and 2 A g–1, respectively. Excitingly, when the rate went back to 0.1 A g–1, the discharge capacity of MnSnO3–MC400 could still return to 1123 mA h g–1, while MnSnO3–MC500 only provided a specific capacity of 286 mA h g–1 at 1 A g–1 and a discharge capacity of 608 mA h g–1 when the rate went back to 0.1 A g–1. It can be easily seen that the rate performance of MnSnO3–MC400 is superior to MnSnO3–MC500. On the basis of the above electrochemical capacity and rate capabilities, it can be found that the appropriate content ratio of MnCo2O4/(Co,Mn)(Co,Mn)2O4 is conducive to the capacity retention and rate performance in the case of a certain content of MnSnO3.

To further investigate the electrochemical performances, the Nyquist plots of MnSnO3–MC400 and MnSnO3–MC500 electrodes at 1000th cycle are exhibited in Figure S7. As shown in Figure S7, the Nyquist diagram presents two semicircles and one oblique line corresponding to the high–medium-frequency and the low-frequency region, respectively. The equivalent circuit model (inset of Figure S7) consisted of electrolyte resistance (Re), SEI resistance (Rs), charge-transfer resistance (Rct), Warburg impedance (W1), and two constant phase element.[53,58] It can be obviously seen in the fitted impedance parameters of Table S3 that the Rct of MnSnO3–MC400 (210 Ω) at 1000th cycle is much smaller than that of MnSnO3–MC500 (347 Ω), showing the faster charge-transfer reaction of MnSnO3–MC400 for Li+ insertion and extraction. Therefore, the electrode reaction kinetics for MnSnO3–MC400 during charging and discharging is enhanced, resulting in the improved cycle performance of the cells.

Conclusions

In summary, the porous multicomponent heterogeneous microspheres (MnSnO3–MC400 and MnSnO3–MC500) have been successfully fabricated. As anodes for LIBs, MnSnO3–MC400 and MnSnO3–MC500 exhibit the specific capacities of 750 and 1030 mA h g–1 at 1 A g–1 without obvious capacity decay until 1000 cycles, respectively. The improved electrochemical properties could be attributed to the unique porous second micro/nanostructure, the lithiation-induced high-rate reactivation, and the synergistic effects among the multicomponent Mn–Sn–Co oxides.

Experimental Section

Synthesis of CoSn(OH)6 Nanocubes

CoSn(OH)6 nanocubes were synthesized according to a previous report.[37] SnCl4·5H2O (0.351 g, 1 mmol) was dissolved in 5 mL of ethanol to form solution A. CoCl2·6H2O (0.238 g, 1 mmol) and sodium citrate (0.294 g) were dissolved in 35 mL of ultrapure water to form solution B. Then, solution A and solution B were mixed and stirred to form a uniform solution, followed by dropwise addition of 5 mL of 2 M NaOH solution at room temperature. After 1 h, a pink precipitate is obtained by centrifuging and washing via ethanol and water each two times, which was then dried in a vacuum oven at 60 °C for 12 h.

Synthesis of MnSnO3–MC400 and MnSnO3–MC500 Microspheres

CoSn(OH)6 (0.1 g), 1.5 mmol MnCl2·4H2O, and 1.5 mmol CoCl2·6H2O were added into 40 mL of ethylene glycol under stirring to form solution A. Then, 2.0 g of NH4HCO3 was added into solution A under stirring combined with ultrasonication (15 min) to form suspension B. The as-obtained suspension B was transferred to 50 mL Teflon liner with stainless steel autoclaves and kept for 20 h at 200 °C in an oven. The CoSn(OH)6/Mn0.5Co0.5CO3 precursor was collected by centrifuging and washing with ethanol and water each two times. Finally, the dry precursor was placed in a furnace for 3 h at 400 °C with a 2 °C min–1 rate under the presence of N2 to obtain MnSnO3–MC400. For comparison, porous MnSnO3–MC500 could be obtained by heating CoSn(OH)6/Mn0.5Co0.5CO3 at 500 °C.

Material Characterization

The phase composition of the as-synthesized products was characterized by XRD (Shimadzu XRD-6000, Cu Kα radiation). The surface composition of the products was evaluated by XPS (PerkinElmer model PHI 5600). FESEM (JEOL, JSM-6700F) and TEM (FEI TF20 and JEM-2100F) were applied to analyze the surface and microstructure. TG analysis was carried on an instrument (Pyris Diamond TG-DTA) in air or N2. Fourier transform infrared and Raman spectra were recorded on an Agilent Cary 660 Fourier transmission infrared and a Renishaw Raman spectrometer, respectively. The pore volume and surface area of the as-obtained products were estimated on a Micromeritics ASAP 2020 by the BET method by nitrogen adsorption–desorption at 77 K.

Electrochemical Measurements

The electrochemical properties of the porous microspheres (MnSnO3–MC400 and MnSnO3–MC500) were measured via the half-cells (CR2032). The working electrodes (a diameter of 13 mm) were made of 80 wt % active materials, 10 wt % acetylene black, and 10 wt % poly(vinylidene difluoride). The loading mass of the as-prepared active materials on the electrode is about 0.93 mg cm–2. The reference electrode is lithium foil, and the electrolyte is 1 M LiPF6 in ethylene carbonate and diethylcarbonate (1:1 volume). The galvanostatic charging/discharging cycle tests were performed on a LAND CT-2001A system (Wuhan Kingnuo Electronics Co., Ltd., China). Electrochemical impedance spectroscopy and CV tests were carried out on an Autolab 302 N and a CHI 760E (Chenhua Ltd. Co., China) electrochemical workstation, respectively.