1 Introduction

In recent decades, rechargeable Li-ion batteries (LIBs) have been extensively employed for modern mobile electronics due to their virtues of high voltage, long lifespan, and environmentally benign nature [1, 2]. Nonetheless, commercialized graphite anode offers a limited theoretical specific capacity (372 mAh g−1), hampering its practical application for advanced LIBs as well as the future electric vehicles (EVs) [3, 4]. Recently, transition metal oxides (TMOs) have emerged as candidates to replace the traditional graphite owing to their much higher theoretical capacities [5,6,7]. In contrast to single-component oxides, spinel mixed transition metal oxides, such as ZnMn2O4, ZnFe2O4, ZnCo2O4, MnCo2O4, etc., have been exploited as anode materials, owing to the complementary and synergetic effects caused by mixed metal cations, which can bring better electrical conductivity and electrochemical performances [8,9,10,11].

Among different transition metal oxides, tetragonal ZnMn2O4 holds promise as a competitive anode for LIBs. On one hand, both Zn and Mn resources are low-priced, eco-friendly, and abundant in nature with respect to many other metals [12, 13]. On the other hand, as an anode for LIBs, ZnMn2O4 provides a larger energy density on account of its large theoretical lithiation capacity (784 mAh g−1) and low oxidation potential (~ 1.2 V vs. Li+/Li) [14, 15]. Nevertheless, like other transition metal oxides, ZnMn2O4 suffers from drastic volume variation throughout the repeated lithiation/delithiation processes, which gradually results in the destruction of electrode structure and an inevitable capacity fade [16, 17]. To overcome this problem, porous structured ZnMn2O4 has been designed and fabricated so as to alleviate the stress and strain induced by volume variation and thereby improve the structural stability [18]. For example, Dang et al. [19] constructed uniform ZnMn2O4 porous microspheres via a solvothermal method, which provided a discharge capacity of 602 mAh g− 1 after 100 cycles at 100 mA g−1. Rong et al. [20] adopted a two-step synthesis route to obtain porous nanoparticle-composed ZnMn2O4 microspheres, delivering a specific capacity of 724 mAh g−1 at 350th cycle at 400 mA g−1. Wang et al. [21] synthesized porous ZnMn2O4 microspheres composed of interconnected nanoparticles, which preserved a steady capacity of 800 mAh g−1 after 300 cycles. Except for various types of microspheres, other porous structured ZnMn2O4 is rare and thus has great research potential.

We report herein novel peanut-shaped porous ZnMn2O4 microparticles, which are prepared by annealing treatment of Zn1/3Mn2/3CO3 precursors synthesized by using a triethanolamine-assisted solvothermal reaction. Triethanolamine (TEA) acts as both the reaction solvent and surfactant during the solvothermal reaction process. The particle size and shape of Zn1/3Mn2/3CO3 vary with the volume ratio of water to TEA. As a result, a uniform peanut-shaped microstructure can be formed in the water-TEA (1:1, v/v) mixture. After thermal decomposition in air, the nanoparticle-composed peanut-shaped ZnMn2O4 microparticles are well formed and possess plenty of void space. As anode materials for LIBs, the ZnMn2O4 micro-/nanostructures illustrate outstanding rate capability and long-term cycling stability.

2 Experimental section

2.1 Preparation of ZnMn2O4 microparticles

Firstly, 1 mmol zinc acetate dihydrate, 2 mmol manganese acetate tetrahydrate and 0.06 mol urea were dissolved in 30 mL of deionized water and TEA (1:1, v/v) under continuous stirring for 0.5 h. Secondly, the resultant homogeneous solution was placed in a 50 mL Teflon-lined autoclave and heated to 160 °C for 12 h, resulting in the formation of the Zn1/3Mn2/3CO3 precursors. Then the white precursors were collected by centrifugal separation and washed alternately with deionized water and alcohol. After drying, the Zn1/3Mn2/3CO3 precursors were calcined at 600 °C in air for 5 h to obtain the final ZnMn2O4 products. To understand the influence of solvent medium on the formation of the precursors, the solvothermal reactions using various volume ratios of deionized water to TEA (30:0, 25:5, 20:10, 10:20 and 5:25) were performed for comparison under the same experimental conditions.

2.2 Materials characterizations

The crystal phase was measured by X-ray diffraction (XRD) (PANalytical X’pert Pro) in the 2θ range of 10°–80°. The morphologies and microstructures were detected through scanning electron microscope (SEM) (Phenom ProX) and field-emission transmission electron microscope (TEM) (FEI, Tecnai G2 F20). Brunauere-Emmette-Teller (BET) characterization and pore size analysis were characterized via Nitrogen adsorption-desorption isotherms at 77 K using Micromeritics ASAP 2460. X-ray photoelectron spectroscopy (XPS) was carried out on an AXIS Ultra DLD spectrometer with Al-Ka radiation.

2.3 Electrochemical measurements

The anode performances of the ZnMn2O4 microparticles were tested with CR2016-type coin cells under room temperature using lithium metal as the counter/reference electrode and Celgard 2400 film as the separator. The working electrode consisted of 70 wt% ZnMn2O4, 20 wt% Super P, and 10 wt% polyvinylidene fluoride. The mass loading of active materials is ca. 1 mg cm−2 in the electrode. The electrolyte solution was 1 M LiPF6 in a mixture of 50 wt% ethylene carbonate and 50 wt% dimethyl carbonate. The cells assembled in an argon-filled glove box were set quietly for 4 h, and then cycled between 0.01 and 3.0 V on a battery testing system (LANHE CT2001). Cyclic voltammetry (CV) was tested by an electrochemical workstation (CHI750E) with a scan rate of 0.1 mV s−1.

3 Results and discussion

Figure 1a exhibits the XRD pattern of the Zn1/3Mn2/3CO3 precursor, which can be indexed as a solid mixture of ZnCO3 and MnCO3 because of the similar crystal structure and close lattice constants [21, 22]. After annealing treatment of the precursor, all diffraction peaks of the product (Fig. 1b) can be assigned to the tetragonal spinel-phase ZnMn2O4 (JCPDS card no. 24-1133). The high peak density and no detectable peaks from impurities suggest that the resultant ZnMn2O4 has good crystallinity and high purity.

Fig. 1
figure 1

XRD patterns of the Zn1/3Mn2/3CO3 precursor (a) and ZnMn2O4 product (b)

Full size image

The elemental composition and metallic oxidation states of the ZnMn2O4 product were determined by XPS characterization, as shown in Fig. 2. The survey spectrum (Fig. 2a) confirms the existence of Zn, Mn and O as well as C elements, of which the C 1 s peak acts as the reference for calibration. In Fig. 2b, two sharp peaks centered at binding energies of 1044.5 eV for Zn 2p1/2 and 1021.4 eV for Zn 2p3/2 with the binding energy difference of 23.1 eV indicate the presence of Zn2+ ion in ZnMn2O4 [4, 7]. In Fig. 2c, two strong peaks loaded at binding energies of 654.0 eV for Mn 2p1/2 and 642.2 eV for Mn 2p3/2 with the binding energy difference of 23.1 eV illustrate the existence of Mn3+ ion [12, 16]. As depicted in Fig. 2d, the O 1 s spectrum could be divided into two peaks at 530.1 eV and 531.8 eV, corresponding to the lattice oxygen and physically adsorbed oxygen, respectively [12, 20]. The above results are well in accordance with previously reported ZnMn2O4 materials.

Fig. 2
figure 2

XPS spectra of survey spectrum, a Zn 2p (b), Mn 2p (c), and O 1 s (d) of the ZnMn2O4 product

Full size image

The SEM image (Fig. 3a) of the ZnMn2O4 product reveals that a large quantity of uniform peanut-shaped microparticles are obtained with the width of ca. 0.5 µm and the length of ca. 1 µm. As shown in a high-magnification SEM image (Fig. 3b), the monodisperse microparticles are composed of tremendous interconnected nanoparticles with the size ranging from 50 to 90 nm. Moreover, the microparticles possess specific porous structure, resulting from the carbonate pyrolysis with release of CO2. The contrast of bright and shade in the TEM image (Fig. 3c) further confirms the homogeneously distributed pores inside the ZnMn2O4 microparticles. The HRTEM image displays the distinct lattice fringes with an interplanar spacing of 0.25 nm, corresponding to the ZnMn2O4 (211) plane (Fig. 3d) [7].

Fig. 3
figure 3

SEM images (a, b), TEM image (c) and HRTEM image (d) of the peanut-shaped porous ZnMn2O4 microparticles

Full size image

The influences of solvent composition on the shape and particle size of the carbonate precursor have been further investigated. Figure 4a–f illustrate the SEM images of the precursors synthesized employing different volume ratios of water to TEA. As seen in Fig. 4a, cubical and spherical particles beyond the size of 10 um could be achieved when using pure water as the solvent. With the increase of TEA content in the water-TEA mixed solvent, the cubes gradually disappeared and the spheres grew into the strips (Fig. 4b, c). Meanwhile, the particle size of the precursor reduced rapidly. When the TEA percentage reached 50%, the precursor particles became regular peanut-shape with lengths of ca. 1 um (Fig. 4d). Presumably TEA may work as the surfactant to regulate and control the crystal growth of Zn1/3Mn2/3CO3 because the hydroxyl and amine groups in TEA molecules could be selectively adsorbed onto the crystal surface [23,24,25]. However, with the increasing TEA content, the peanut-shaped particles grew into irregular and larger spheres again (Fig. 4e, f). Note that the certain amount of water in the urea-assisted solvothermal reaction is crucial to the shape and even purity of the precursors at the same time [26]. Consequently, TEA and water have a strong synergistic effect on the formation of a peanut-shaped micro/nanostructure with homogeneous shape and small particle size.

Fig. 4
figure 4

SEM images of Zn1/3Mn2/3CO3 precursors prepared with various volume ratios of water to TEA: a 30:0, b 25:5, c 20:10, d 15:15, e 10:20, and f 5:25

Full size image

The porous characteristic and BET surface area of the peanut-shaped ZnMn2O4 microparticles were investigated by the nitrogen adsorption/desorption isotherm and Barrett-Joyner-Halenda (BJH) method, as shown in Fig. 5. The isotherm with a hysteresis loop can be classified as a type-IV curve, suggesting the existence of mesoporous structure [8, 9]. The resultant surface area is calculated as large as 29.3 m2 g−1 with total pore volume of 0.19 cm3 g−1. The pore size distributes in the range of 2–90 nm with an average pore width of 26.7 nm. Such a benignant porous structure with a large surface area could not only provide the buffer space to alleviate the structural expansion during lithiation process, but also greatly facilitate the fast Li+ ion transport at the electrode/electrolyte interface, thus bringing about good electrochemical performance [27,28,29].

Fig. 5
figure 5

Nitrogen adsorption/desorption isotherm and BJH pore size distribution (inset) for the peanut-shaped porous ZnMn2O4

Full size image

Figure 6a gives the initial five successive CV profiles between 0.01 and 3.0 V. In the 1st cathodic scan, two peaks at 1.21 V and 0.70 V, disappearing during the following scans, are ascribed to the reduction of Mn3+ to Mn2+ and the generation of a solid electrolyte interphase (SEI) layer, respectively [3]. In addition, the strong peak centered at 0.26 V is caused by the reduction of Zn2+/Mn2+ to Zn/Mn and the subsequent Li–Zn alloying reaction [30]. The 1st anodic process exhibits a wide peak close to 1.28 V corresponding to the oxidation of Mn/Zn to Mn2+/Zn2+ [31]. The cathodic reduction peaks in the 2nd to 5th cycles positively shift to ca. 0.35 V, resulting from the irreversible electrode reactions in the 1st cycle. The CV curves after the initial cycle are gradually overlapped, revealing a high electrochemical reversibility of the ZnMn2O4 anode. The lithiation/delithiation process can be processed as follows [3, 5, 22]:

Fig. 6
figure 6

CV profiles (a), voltage profiles (b), cycling stability (c), and rate performance (d) of the ZnMn2O4 anode

Full size image
$$ {\text{ZnMn}}_{{{2}}} {\text{O}}_{{{4}}} \, + \,{9\text{Li}}^{ + } \, + \,{9\text{e}}^{ - } \to {\text{ ZnLi}}\, + \,{2\text{Mn}}\, + \,{4\text{Li}}_{{{2}}} {\text{O}} $$
(1)
$$ {\text{ZnLi}}\, + \,2{\text{Mn}}\, + \,3{\text{Li}}_{2} {\text{O }} \leftrightarrow {\text{ ZnO}}\, + \,2{\text{MnO}}\, + \,7{\text{Li}}^{ + } \, + \,7{\text{e}}^{ - } $$
(2)

Figure 6b presents the voltage profiles of the peanut-shaped porous ZnMn2O4 microparticle anode at a specific current of 0.1 A g−1 in the first five cycles. The porous ZnMn2O4 anode exhibits the initial lithiation and delithiation capacities of 1393 mAh g−1 and 878 mAh g−1 with a coulombic efficiency of 63%. The large irreversible capacity loss may basically be induced by the formation of SEI film on the electrode interface and irreversible consumption of Li+ in the defects of the ZnMn2O4 particles [21, 32]. Particularly, all the voltage plateaus and the finally overlapped voltage profiles conform well to the CV data. The coulombic efficiency rapidly approaches more than 96% in the 4th cycle, indicating a good lithium insertion/removal reversibility of the ZnMn2O4 anode.

Figure 6c depicts the rate performance of the peanut-shaped ZnMn2O4 anode. The average lithiation capacities of 879, 831, 783, 743, 688, and 579 mAh g−1 are obtained at specific currents of 0.1, 0.2, 0.5, 1, 2, and 4 A g−1, respectively. The reversible capacity at a high specific current of 4 A g−1 is about 1.5 times the theoretical capacity of the commercialized graphite, which demonstrates that the as-prepared ZnMn2O4 anode has great application potential for high-power LIBs. Particularly, a high lithiation capacity of 903 mAh g−1 can still be obtained when the specific current returns to 0.1 A g−1, indicting the good electrode reaction reversibility. It is believed that the good rate capability of the ZnMn2O4 anode results from its reasonable hierarchical micro-/nanostructure and appropriate particle size. The secondary microparticles can make a good contact with conductive carbon in the electrode due to the oblong shape and relatively small size (ca. 1 µm), facilitating the improvement of electronic conductivity of active materials [23]. The small particle size of the primary nanoparticles offers short diffusion length for Li+ and large electrode reaction interface [33]. The rich pores on the surface and inside the microparticles permit the easy permeation of electrolyte into the whole micro-/nanostructured active materials [34, 35]. All of the abovementioned factors are favorable for faster electrochemical kinetics, resulting in the high-rate lithium storage performance.

The ZnMn2O4 anode also exhibits good long-term cycling performance, as described in Fig. 6d. At a specific current of 0.5 A g−1, the lithiation capacity rapidly decreases from 840 mAh g−1 for the 2nd cycle to 690 mAh g−1 for the 6th cycle, and then gradually rises to a maximum value of 875 mAh g−1 for the 125th cycle. Afterwards, the capacity decreases very slightly in the following cycles and maintains at 797 mAh g−1 after 700 cycles with the coulombic efficiency retaining at 99% throughout the long-term cycling. A decrease followed by an increase in capacity has also been observed for other porous structured TMOs anodes [13]. The quick capacity decay is most likely related to the partial fracture and electrical isolation of the nanoparticles because of the large volume swelling during lithium insertion process [33]. The gradual capacity rise in subsequent cycles is attributable to the reactivation of the electrode, including the reconstruction of the interconnected three-dimensional porous structures and the reversible formation of a polymeric gel-like film on the anode surface [28, 29]. SEM images of the peanut-shaped porous ZnMn2O4 microparticles before cycling and after 100 cycles at 0.5 A g−1 are compared in Fig. 7a, b. After cycling, the particle size of ZnMn2O4 increases to ca. 1.4 times that of the fresh particles due to the irreversible volume expansion and microstructure reconstruction throughout the continuous lithiation/delithiation reactions. More importantly, the peanut-shaped microparticle morphology is basically preserved without obvious destruction, resulting from the strong porous structure that effectively buffers the volume change. The TEM image (inset in Fig. 7b) also illustrates the existence of void space inside the microparticle and interconnection of the nanoparticles after repeated cycles. Therefore, the outstanding structural stability of the porous micro-/nanostructure during cycling provides a long cycling life. As shown in Table 1, such remarkable electrochemical performances make the peanut-shaped porous ZnMn2O4 microparticle a more promising anode material compared with some previously reported porous ZnMn2O4 microspheres [5, 15, 17, 19,20,21,22, 36,37,38].

Fig. 7
figure 7

SEM image of the ZnMn2O4 microparticles before cycling (a) and after 100 cycles (The inset shows its TEM image) (b)

Full size image
Table 1 Electrochemical comparison of ZnMn2O4 microparticles with other ZnMn2O4 microspheres
Full size table

4 Conclusions

Herein, we report peanut-shaped porous ZnMn2O4 microparticles synthesized through a facile solvothermal reaction and post-calcination method. The shape and size of the carbonate precursor can be adjusted by changing the solvent composition, i.e. the volume ratio of water to triethanolamine. Owing to the uniform porosity inside the micro-/nanostructure and the relatively small size of nanoparticle-composed microparticle, the ZnMn2O4 microparticles not only accommodate the volume expansion over the repeated cycling, but also effectively facilitate the rapid Li+ transport at the same time. As expected, the ZnMn2O4 anode exhibits high lithiation capacities of 579 mAh g−1 at 4 A g−1 and 797 mAh g−1 up to 700 cycles at 0.5 A g−1. The outstanding electrochemical performance validates our novel peanut-shaped porous ZnMn2O4 microparticle a competitive anode material for high-energy LIBs.