Ang
Li‡
a,
Changfeng
Li‡
a,
Peixun
Xiong‡
a,
Jinfeng
Zhang
a,
Dongling
Geng
b and
Yunhua
Xu
*a
aSchool of Materials Science and Engineering, Key Laboratory of Advanced Ceramics and Machining Technology (Ministry of Education), Tianjin Key Laboratory of Composite and Functional Materials, Tianjin University, Tianjin 300072, China. E-mail: yunhua.xu@tju.edu.cn
bSchool of Materials Science and Engineering, Nanjing University of Science and Technology, 200 Xiaolingwei, Nanjing 210094, P. R. China
First published on 6th June 2022
Layered transition metal oxides (LTMOs) are a kind of promising cathode materials for potassium-ion batteries because of their abundant raw materials and high theoretical capacities. However, their synthesis always involves long time calcination at a high temperature, leading to low synthesis efficiency and high energy consumption. Herein, an ultra-fast synthesis strategy of Mn-based LTMOs in minutes is developed directly from alkali-transition metal based-metal–organic frameworks (MOFs). The phase transformation from the MOF to LTMO is systematically investigated by thermogravimetric analysis, variable temperature optical microscopy and X-ray diffraction, and the results reveal that the uniform distribution of K and Mn ions in MOFs promotes fast phase transformation. As a cathode in potassium-ion batteries, the fast-synthesized Mn-based LTMO demonstrates an excellent electrochemical performance with 119 mA h g−1 and good cycling stability, highlighting the high production efficiency of LTMOs for future large-scale manufacturing and application of potassium-ion batteries.
Recently, layered transition metal oxides (LTMOs) have been extensively investigated as cathode materials for KIBs.18–21 They have ionic diffusion paths with a large interlayer distance, benefiting the migration of K+. Among them, layered potassium manganese oxide (KxMnO2) has attracted great attention owing to their high theoretical capacities and abundance of raw materials.22–25 Generally, layered KxMnO2 is synthesized through a long-time high-temperature calcination (≥800 °C), which typically takes more than 10 hours (Table S1†). Apparently, this procedure takes time and consumes energy, leading to low production efficiency and high cost.
It is well known that the atomic distance seriously impacts the diffusion process during pyrolysis treatment, and further influences the formation of a new phase. Aiming at rapid synthesis of LTMO, a short distance between alkali metal ions and transition metal ions as well as a uniform distribution is necessary. Metal–organic frameworks (MOFs) are a kind of porous crystalline material formed by metal ions and organic ligands through coordination bonds, which have a homogenous distribution of different elements at the atomic level.26 With the merits of large surface area, ultra-high porosity, and structural diversity, they have attracted extensive attention in energy fields.27,28 The investigation was mainly focused on MOF-derived anode materials but rarely on cathodes.29–33 A lot of efforts have been devoted to developing transition metal (Fe, Co, Ni, Mn, Cu, and Zn)-based MOF precursors for anode materials, while less on alkali-transition metal based-MOFs. Despite differences in the properties between alkali and transition metal ions, alkali-transition metal based-MOFs can be constructed and are suitable for rapid synthesis of LTMO cathodes due to their atomic level distribution of different elements.
Herein, an ultra-fast synthesis method is developed for preparing a layered cathode material of KxMnO2 directly from the K[Mn(HCOO)3] MOF (KM-MOF). The formate anion (HCOO−) is the simplest and smallest carboxylate ligand and the atomic-level homogenous distributions of elements promises a short migration distance of metal ions, enabling a fast phase formation of layered KxMnO2 in minutes (Fig. 1). The phase transformation process from MOF to layered metal oxide was systematically investigated by thermogravimetric analysis (TGA), variable temperature optical microscopy and X-ray diffraction (XRD). The fast-synthesized layered KxMnO2 achieved excellent electrochemical performance with 119 mA h g−1 and good cycling stability, that surpasses most previously reported LTMO cathodes (Table S2†), highlighting the high production efficiency of LTMOs for future large-scale manufacturing and application of KIBs.
In general, the composition and structure of MOF precursors significantly affect the pyrolysis process and the corresponding derivative. Therefore, an insight into the KM-MOF to layered KxMnO2 transformation process is required to improve the understanding of the rapid synthesis. To probe the transformation process, component evolution was monitored during the calcination process by TGA. TGA measurements were performed from 50 to 900 °C with a heating rate of 10 °C min−1 in air. As shown in Fig. 2c, a remarkable weight loss of 42.0% occurred from 250 to 410 °C, which is assigned to the decomposition of the KM-MOF and the formation of metal oxide. Theoretically, the organic species were burned into CO2, leading to 57.6% weight loss according to the theoretical calculations, while the formation of metal oxide caused 14.0% weight gain upon the uptake of oxygen. The TGA measurement result is consistent with the theoretical values. A small weight loss of 3.0% was displayed from 700 to 800 °C, which should be attributed to the sublimation of potassium because of its low boiling point of 759 °C. The bond lengths of the KM-MOF obtained by using crystalline data are in the order of K–O > Mn–O > C–O (Table S3†), indicating that the bond strengths are in the order of K–O < Mn–O < C–O. Therefore, during the thermal heating process, the K–O and Mn–O bonds would break prior to the C–O bonds. The morphology evolution of the KM-MOF is also monitored by variable temperature optical microscopy, and the optical microscope photographs at different temperatures are shown in Fig. 2d–g. At low temperatures below 300 °C, there is no obvious change to the crystalline morphology (Fig. 2d and e). However, when the temperature increased to 350 °C, the crystals became opaque (Fig. 2f), and molten at 400 °C (Fig. 2g), suggesting damage of the crystalline structure. These results are well consistent with the TGA measurement results.
To gain further insights into the phase conversion process from the KM-MOF to layered KxMnO2, variable temperature XRD was conducted (Fig. 2h). The diffraction peak of the (200) crystal plane that is based on K–O bonds disappeared at 250 °C (Fig. S4†). When the temperature increased to 300∼500 °C, the KM-MOF almost lost its crystallinity. From 600 °C, new peaks appeared at 12.7 and 25.4°, which can be indexed to the planes of (001) and (002) of layered KxMnO2, respectively, indicative of the beginning of layered phase formation. These two peaks slightly shifted to lower diffraction angles of 12.6° for (001) and 25.2° for (002) planes from 800 to 900 °C, indicating that the interlayer spacing became larger. This can be explained by the enhanced coulombic repulsion between the MO6 slabs upon K+ sublimation. The interlayer spacing is calculated to be 0.69 nm based on the diffraction peak of the (001) plane. The variable temperature XRD measurement results show that layered KxMnO2 can be facilely synthesized from the KM-MOF through thermal sintering. Based on the aforementioned analysis, the transformation process from the KM-MOF to layered KxMnO2 is schematically illustrated in Fig. 2i. During the heating process, K–O and Mn–O bonds successively break along with the uptake of oxygen to form K and Mn oxides from 300 to 400 °C, and then the layered phase of KxMnO2 is formed above 600 °C.
An ultra-fast synthesis method was developed to produce layered KxMnO2, by sintering the KM-MOF at 1000 °C for only 8 minutes, and the resulting product is denoted as KMO-F. As a control, layered KxMnO2 was also prepared using a conventional long-time sintering method from the same MOF precursor, denoted as KMO-S. Their layered crystal structures were revealed by the XRD patterns, and they are in good agreement with the previously reported results in the literature.35 The pronounced peaks at 12.7/25.4° of KMO-S and 12.6/25.2° of KMO-F are assigned to the (001)/(002) planes (Fig. 3a and S5†), and the interlayer spacings are calculated to be 0.69 and 0.68 nm, respectively. Rietveld refinement suggests that both KMO-F and KMO-S belong to the hexagonal symmetry with a space group of P63/mmc. The lattice parameters of KMO-F phase are calculated to be a = b = 2.881 Å and c = 14.132 Å, while they are a = b = 2.880 Å and c = 14.062 Å for KMO-S. Therefore, they have the same crystalline structure, and the schematic crystal structure of the layered KxMnO2 is shown in Fig. 3b. Mn ions are octahedrally coordinated with oxygen anions while K ions are located between the layers of MnO6 octahedra. The overall K/Mn atomic ratios were measured by using inductively coupled plasma (ICP) to be 0.53 for KMO-F and 0.51 for KMO-S (Table S4†), corresponding to the molecular formulas of K0.53MnO2 and K0.51MnO2, respectively. The chemical properties of the layered compounds were investigated by X-ray photoelectron spectroscopy (XPS) (Fig. S6†). The high-resolution Mn 3s spectrum of KMO-F shows a peak energy separation (ΔE) of 5.06 eV (Fig. S7†), consistent with the previous reports.35 According to the linear relationship between the chemical valence of Mn and the ΔE value, the average oxidation state of Mn in KMO-F was calculated to be 3.47 (Table S5†). Correspondingly, the Mn 2p1/2 and Mn 2p3/2 peaks in the high-resolution Mn 2p spectrum can be deconvoluted into two peaks of Mn3+ and Mn4+ with equal integral areas (Fig. 3c), suggesting an average valence of 3.5, very close to the value of 3.47. Similar results were obtained for KMO-S.
The morphologies and microstructures of the layered KMOs were characterized by scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM). The SEM images of KMO-F and KMO-S present similar morphologies with crystal particles in random shapes and sizes (Fig. 3d and f). The HRTEM images reveal highly crystalline structures with interlayer spacings of 0.69 nm for KMO-F and 0.68 nm for KMO-S (Fig. 3e and g), corresponding to the (001) plane. Uniform elemental distributions of K, Mn and O were measured over the whole particles by scanning transmission electron microscopy (STEM) (Fig. 3h and S8†), suggesting a well-formed layered structure. Apparently, well-crystallized layered KxMnO2 can be produced by the fast synthesis method, similar to the conventional long-time synthesis methods from the same MOF precursor, demonstrating the effectiveness and superiority of the new method.
The electrochemical performances of KMO-F and KMO-S were investigated and compared under the same test conditions. Coin-type cells were fabricated using KMO-F or KMO-S cathodes and 2.5 M KFSI/TEP as the electrolyte. The cyclic voltammetry (CV) measurements were carried out at a scan rate of 0.1 mV s−1 in a voltage window of 2.0–4.2 V (vs. K+/K). Similar CV curves were displayed for the two cathodes with five redox couples (Fig. 4a and b), suggesting similar electrochemical behaviors. These current peaks can be assigned to the reversible potassium insertion/extraction into/from layered KxMnO2 with Mn3+/Mn4+ redox reaction. Obviously, the well-overlapped CV curves after the first cycle predict good cycling stability for the KMO cathodes. The galvanostatic charge/discharge profiles at a current density of 50 mA g−1 are shown in Fig. 4c and d. The first charge/discharge capacities of KMO-F and KMO-S are 119/133 and 84/111 mA h g−1, respectively. Although KMO-F delivered a slightly higher initial capacity, similar cycling performance was obtained for them (Fig. 4e). KMO-F and KMO-S retained 80% and 85% of their capacity after 100 cycles at 50 mA g−1, respectively, with high average coulombic efficiencies of about 99%. In addition, the KMO-F cathode delivers an initial discharge capacity of 81.14 mA h g−1 at 300 mA g−1 and retains 86% of its capacity after 120 cycles (Fig. S9†). Good rate performance was obtained for both KMO-F and KMO-S (Fig. 4f). When the current density increased from 50 to 1000 mA g−1, KMO-F maintained a specific capacity of 65 mA h g−1, compared with 73 mA h g−1 for KMO-S. The electrochemical performance once again manifests the feasibility of the fast synthesis method for high performance layered cathode materials.
To further investigate the K-storage behavior for the KMO cathodes, CV measurements at various sweep rates from 0.1–2.1 mV s−1 were carried out (Fig. 4g and h). The capacity contributions from surface pseudo-capacitance and ion intercalation can be quantified by separating the current response i at a fixed potential V into capacitive-controlled (k1v) and diffusion-controlled (k2v1/2) processes, according to the following equation:36i(V) = k1v1 + k2v1/2. In our work, based on the integration of CV curves, 91.2% of the total charge storage capacity of the KMO-F cathode is capacitive at a sweep rate of 1.5 mV s−1. Such high proportions of pseudocapacitive contribution result from fast kinetics, which should account for the good rate capability of the KMO-F cathode. The surface capacitive-controlled capacities and diffusion-controlled capacities for KMO-F and KMO-S at different scan rates are summarized in Fig. 4i. KMO-S shows slightly higher diffusion-controlled capacities at various sweep rates.
The K storage process of the layered cathodes was investigated by in situ XRD, and the data are shown in Fig. 5. By charging (extraction of K ions), the (001) and (002) peaks shifted to lower 2-theta degrees, indicating the increased interlayer spacing due to enhanced electrostatic repulsion between the adjacent oxygen layers upon K+ extraction. During the discharge process, a reversible peak shift was observed, demonstrating good structural reversibility and cycling stability.
Fig. 5 (a) Galvanostatic charge/discharge profiles in the first cycle of layered KxMnO2 at a current density of 50 mA g−1, (b) in situ XRD patterns, and (c) the image plots of the in situ XRD data. |
Footnotes |
† Electronic supplementary information (ESI) available: XRD, optical microscope photograph, XPS, elemental mapping, CV, and galvanostatic charge–discharge profiles. See https://doi.org/10.1039/d2sc02442j |
‡ Ang Li, Changfeng Li and Peixun Xiong contributed equally to this work. |
This journal is © The Royal Society of Chemistry 2022 |