Kyubin
Shim‡
a,
Hyun Woo
Kim‡
a,
Sungwoo
Park
a,
Kyeong-Deok
Seo
b,
Chang-Yeon
Kim
a,
Jin Bae
Lee
a,
Jong Seong
Bae
c and
Hae Jin
Kim
*a
aResearch Center for Materials Analysis, Korea Basic Science Institute, Daejeon 34133, Republic of Korea. E-mail: hansol@kbsi.re.kr
bDepartment of Chemistry and Chemistry Institute for Functional Materials, Pusan National University, Busan 46241, Republic of Korea
cBusan Center, Korea Basic Science Institute, Busan 46742, Republic of Korea
First published on 23rd February 2024
To address the limitations of the anodes used in current Li-ion batteries (LIBs), a MnS/MnO-decorated S,N-doped carbon (MSNC) electrode was designed by introducing a Mn2+-coordinated polymer (Mn-DTOGA) bonded to a sulfur atom with acetate counter ions through an imine formation reaction for advanced LIBs. The MSNC electrode was prepared and characterized by X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), scanning electron microscopy, and transmission electron microscopy, which confirmed the ease of synthesis, superior controllability, and uniformity of the material. The electrochemical characteristics of the materials were assessed via cyclic voltammetry, electrochemical impedance spectroscopy, and galvanostatic charge–discharge tests to determine their viability as anode materials in LIBs. The results demonstrate excellent cycling stability, with the material exhibiting superior performance for 500 cycles at a rate of 0.2 C. Furthermore, in situ XRD and XPS measurements provided a fundamental understanding of the material, confirming the redox reaction between the MnS/MnO complex and metallic Mn/Li2S/Li2O.
In the last few decades, transition metal oxides (756 mA h g−1) and transition metal sulfides (616 mA h g−1) have been proposed as alternatives to carbon-based materials because of their high theoretical specific capacities resulting from the conversion processes.10–14 Mn-based materials, such as MnO and α-MnS, are regarded as desirable alternative anodes for LIBs, owing to their advantages, such as environmental friendliness, natural abundance, and low operating potentials.15,16 However, during repeated charge–discharge cycles, Mn-based compounds with high capacities experience several issues. In particular, a short cycle lifespan is caused by substantial structural changes during delithiation, and a poor rate capability is caused by low electrical conductivity and slow kinetics, which restrict their practical applications.17
Several methods have been suggested to address these issues and achieve a good Li storage performance. Embedding nanosized MnO or MnS into a carbon matrix can shorten the Li-ion diffusion pathway, reduce the volume change during repeated charging and discharging, and enhance electronic conductivity. In particular, the enhanced Li affinity and customized electrical structures of heteroatom-doped carbons make them ideal candidates for rapid Li+/electron transport.18,19
Recently, it has been shown that oxide/sulfide heterostructures, created using oxide and sulfide semiconductors with various bandgaps, can generate an internal electric field at the interfaces to accelerate the transport of ions and electrons, and are potential candidates for Li/Na-ion storage.20,21 Additionally, compared to sulfides, oxides provide a larger specific capacity for the same metal but poorer conductivity and more severe polarization during the charge–discharge process. Consequently, oxide/sulfide heterostructures can fully exploit the advantages of each material. Embedding oxide/sulfide heterostructures into a carbon matrix is a practical technique for achieving high-performance Li storage. For example, porous N,S-co-doped carbon-embedded MnO/MnS heterostructures developed by Zhang et al. demonstrated a high specific capacity of 591 mA h g−1 at 0.1 A g−1 and significant cycling performance.20 Chen et al. reported the synthesis of N and S co-doped carbon-confined MnO/MnS heterostructures derived from the one-step pyrolysis of Mn-methionine frameworks. When used as an anode material, these heterostructures exhibited capacities of 574/577 mA h g−1 at 0.5 A g−1.22 Nevertheless, the fabrication of composites with a wide range of properties typically requires labor-intensive and expensive syntheses, which are challenging to scale-up for practical applications. Furthermore, the production of well-defined oxide/sulfide heterostructures with carbon matrices remains challenging. Consequently, it is essential to develop a straightforward and efficient technique for reducing the complexity of the synthesis process.20–22 Given these aspects, we recently reported MnS/MnO-decorated N,S-doped carbon derived from an Mn(II)-coordinated polymer, providing a new synthesis method and high electrochemical performance.23
In this study, we synthesized a Mn2+-coordinated polymer bonded to a sulfur atom with acetate counter ions (Mn-DTOGA) via an imine formation reaction. Subsequently, the Mn ion-coordinated polymer was calcinated at 1000 °C to obtain homogeneous MnS/MnO-decorated S,N-doped carbon (herein, MSNC) nanoparticles for practical applications. The exceptional chemical and physical properties of the MSNC nanoparticles were thoroughly characterized using X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). To assess their suitability as anode materials in LIBs, the electrochemical properties of the MSNC materials were evaluated using electrochemical impedance spectroscopy (EIS), cyclic voltammetry (CV), and galvanostatic charge–discharge tests. In addition, in situ XRD and XPS analyses were performed. The results provide strong evidence that MSNC holds great promise for the development of high-energy-density LIBs, owing to their superior chemical, physical, and electrochemical properties.
High-resolution TEM and energy-dispersive X-ray spectroscopy (EDS) images were acquired using an ultra-corrected energy-filtered TEM microscope (Libra 200 HT MC Cs, Carl Zeiss, Germany) operating at an accelerating voltage of 200 kV. For the TEM analysis, a drop of a dilute solution of nanoparticles in ethanol was deposited onto ultrathin carbon-coated copper grids. The morphologies of the samples were characterized by field-emission SEM (S-4800, Hitachi, Japan) at the Korea Basic Science Institute, Daejeon. High-performance X-ray photoelectron spectroscopy (HP-XPS, VG Scientific ESCALAB 250) confirmed the chemical composition of the prepared samples (The Converging Materials Core Facility of Dong-Eui University, Busan and the Korea Basic Science Institute (KBSI), Busan.). Electrochemical measurements were performed using a potentiostat (VSP; Biological Science Instruments). The electrochemically active surface area (ECSA) was calculated based on the cyclic voltammograms recorded in a 1.0 M H2SO4 electrolyte saturated with N2 at various scan rates.24–26 The N2 adsorption–desorption analysis was conducted using a volumetric nitrogen sorption system by ASAP2020 (Micromeritics, USA) in Korea Basic Science Institute (KBSI), Daejoen. The adsorption–desorption measurements were performed at liquid nitrogen temperature (77 K). The specific surface area (SBET), the micropore volume (Vmic), and the external surface area were determined by means of a t-plot.27 The total pore volume (Vtot) was calculated at relative pressure of P/P0 = 0.9889. The pore sizes of the sample were analyzed using the Horvath–Kawazoe (H–K) method (microporosity)28 and the Barret–Joyner–Halenda model (BJH) method (mesoporosity).29
To prepare the anode, a slurry was formed by mixing the MSNC powder, super-P, and polyvinylidene fluoride (weight content of 8:1:1) in NMP. The slurry was gently cast on Cu foil using a doctor's blade. The resulting anode films were dried at 90 °C for 10 h under vacuum.31,32 Other materials, such as MnS, MnO, and hard carbon materials, were also carried out in the same manner as above.
CR-2032 cells (Li metal/MSNC, MnS, MnO, and hard carbon) were assembled in an argon-filled glove box with oxygen and moisture levels of less than 0.1 ppm. Commercial 1.0 M LiPF6 in EC:DEC (1:1 vol) (5.0 wt% FEC) electrolyte and Celgard® separator were used for discharge–charge cycling studies.
Fig. 1 (a) Schematic of the preparation of the MSNC material; (b) XRD measurement of MSNC; XPS spectra of MSNC: (c) C 1s, (d) N 1s, (e) O 1s, (f) S 2p, (g) Mn 2p, and (h) Mn 3s. |
XPS was used to determine the chemical composition of the MSNC nanoparticles. Fig. S1 (ESI†) shows the XPS spectrum of the MSNC, which reveals the presence of C, O, N, S, and Mn. The high-resolution C 1s XPS spectrum displays peaks corresponding to the C–C, C–O/C–N/C–S, and CO bonds (Fig. 1c). The N 1s spectrum (Fig. 1d) confirms the presence of pyrrolic-N, quaternary-N, and pyridinic-N bonds in the material. The high-resolution O 1s XPS spectrum revealed the presence of metal–O, C–O, and CO bonds (Fig. 1e). The S 2p spectrum (deconvoluted in Fig. 1f) exhibits peaks corresponding to C–S–C, MnS, and sulfate. The Mn 2p spectrum was deconvoluted into Mn2+ (MnO and MnS), Mn3+,4+, and Mn2+ satellite peaks (Fig. 1g). The Mn 3s region was used to examine the oxidation state of Mn. The interval between the Mn 3S1/2 and Mn 3S3/2 peaks varied depending on the oxidation state of the Mn species. In this case, the appearance of two peaks in the Mn 3s spectrum (Fig. 1h) indicates the coupling of electrons between Mn 3s and Mn 3d. The shift in the binding energy of the Mn 3s peaks indicated a change in the oxidation state of Mn. The presence of a 5.9 eV interval in the spectrum corresponds to the Mn2+ state, suggesting the existence of MnO bonds.33,34 All peak positions in the XPS spectra are listed in Table S1 (ESI†).
Fig. 2 (a) SEM (SE and BE modes), (b) high-resolution TEM, and (c) mapping images of MSNC (N, O, C, S, and Mn). |
To confirm the active area, the ECSA in a 1 M H2SO4 solution was measured, as shown in Fig. S6 (ESI†). The active surface area of the MSNC electrode is calculated using the double-layer capacitance, ECSA = Cdl/Cs, where Cdl represents the double-layer capacitance and Cs (0.029 mF cm−2) represents the specific capacitance of the sample. The surface area of the MSNC electrode was 0.242 cm2.
Fig. 3c–e illustrates the discharge/charge curves of the MSNC electrode over 500 cycles at a rate of 0.2 C, ranging from 0.05 to 2.0 V. The MSNC electrode initially exhibits discharge and charge capacities of 598 and 293 mA h g−1, respectively (Fig. S7, ESI†), which was confirmed similar to profiles of MnS and MnO cells. Further information was described in the ESI.† (Fig. S8, ESI†). The decrease in capacity was attributed to the irreversible formation of SEI layers and electrolyte decomposition in the low-potential regions. Furthermore, the presence of a plateau at approximately 0.63 V in the discharge curves is attributed to the formation of an Mn and Li sulfide/oxide matrix. During the initial charging process, the plateau at approximately 1.2 V was associated with Mn sulfidation/oxidation.22,35
Region 1 (2.0–0.01 V)
MnS/MnO + 2Li+ + 2e− → Mn0 + Li2S/Li2O | (1) |
Region 2 (0.01–2.0 V)
Mn0 + Li2S/Li2O → MnS/MnO + 2Li+ + 2e− | (2) |
Further discharge/charge profiles explain the reversible capacities at approximately 232 mA h g−1 with 100% coulombic efficiency for 500 cycles. Although the electrochemical performance of MSNC needs to be improved in terms of capacity and irreversibility, this result indicates a superior cycle performance compared to that of commercial HC (hard carbon) in Fig. S9 and 10 (ESI†) (HC-pac-2m, Aekyung Chemical Co., Ltd.). Fig. 3e shows the rate capability of the MSNC as an anode material, which was assessed to further determine its applicability. The capacity values provided by MSNC at various rates of 0.05, 0.1, 0.2, 0.5, 1.0, and 2.0 C are 359, 317, 273, 221, 183, and 145 mA h g−1, respectively. After 24 cycles, when the C-rate is reduced to 0.05C, the capacity returns to 332 mA h g−1. We compared electrochemical performances of recently reported MnS and/or MnO-based carbon composites anodes. As shown in Table S2 (ESI†), MSNC anodes (in this work) exhibit superior cyclic stability with capacity retention of 79% after the 500th cycle, and lower voltage operation as anode. However, the anodes presented relatively low initial charge capacity, which might be due to their low contents of Mn-based active materials and relatively narrow voltage cut-off range. Capacity values provided by MSNC at various rates of 0.05, 0.1, 0.2, 0.5, 1.0, and 2.0 C are 359, 317, 273, 221, 183, and 145 mA h g−1.
In situ XRD measurements were performed during the first cycle of the discharge/charge process to gain further insight into the Li storage mechanism of the MSNC material (Fig. S11, ESI†). The evolution of the XRD patterns (30 scans) and the corresponding discharge/charge profiles are shown in Fig. 4a. The intensities at 34.5° and 49.5° of MnS decreased during the first charging cycle. From this result, we could confirm that the MnS complex is reduced to metallic Mn during the first discharge, followed by the formation of the MnS/MnO complex, as mentioned above, during the subsequent charging process. The color-coded images (Fig. 4b) obtained from additional scans (144 scans over three cycles) provide evidence that the MnS complex is also reduced as the cycling progresses, which is considered to be attributed to SEI formation during cycling.
Fig. 4 (a) In situ XRD measurement of Li/MSNC cell during discharge/charge, (b) MSNC, and XPS spectra of MSNC electrode after 1 cycle (c)–(f), 20 cycles (g)–(j). |
XPS spectra were obtained to determine the chemical composition of the MSNC after one and 20 cycles. The spectra show the presence of C, N, O, S, P, Li, F, and Mn in the prepared electrodes (Fig. S12, ESI†). Fig. 4c–j shows the high-resolution spectra of S 2p, Mn 2p, O 1s, and Li 1s (1 and 20 cycles), and all the peak positions of the XPS are listed in Table S3 and S4 (ESI†). Fig. S13 (ESI†) shows the C 1s spectra after cycles one and 20 cycles, which reveal the presence of C–C, C–O/C–N/C–S, CO, and C–F bonds. In the C 1s spectra, C–F bonds were newly introduced compared to the powder sample, which was due to LiPF6 from the electrolyte. The high-resolution S 2p XPS spectrum displays peaks corresponding to MnS, C–S–C, and sulfate (Fig. 4c and g). The intensity of the sulfate bond peak decreased in the sample after 20 cycles compared with that after only one cycle. This is because O ions emerge from the broken S–O bonds during cycling. Furthermore, the intensity of the MnS peak increased in the sample after 20 cycles compared to that of the sample after only one cycle. In the Mn 2p region (Fig. 4d and h), Mn2+ (MnS and MnO), Mn3+,4+, and Mn2+ satellite peaks were observed. Interestingly, as shown in the Mn 2s spectra, the peak intensity of MnO decreased relative to that of MnS after 20 cycles, suggesting that MnO was primarily involved in the cycling reaction. The O 1s XPS spectra (after one and 20 cycles) were deconvoluted into peaks corresponding to Li2O, MnO, and CO groups (Fig. 4e, i). From the O 1s spectra, it can be observed that the peak intensity of Li2O is enhanced in comparison to that of MnO after 20 cycles. Based on this result, we propose that O ions are released during the decomposition of the predominant MnO bond with Li ions during cycling. Fig. 4f and j show the XPS profiles of the high-resolution Li 1s peaks corresponding to LiF and Li2O, respectively. Compared to that after one cycle, the Li2O peak intensity after 20 cycles increased, thereby the Li 1s peak results support the previous results for the O 1s peak.
Therefore, the in situ XRD and XPS data validate the roles of MnS and MnO during the charge/discharge process. Furthermore, MnO decomposed into Mn and O ions with formation of Mn–O and Li–O, as observed in the O 1s and Li 1s spectra. In particular, the amount of O1 decreased with depth after 1 and 20 cycles, the amount of S1 was slightly reduced as shown in XPS depth study (Fig. S14, ESI†). This indicates that, as mentioned previously, MnO undergoes reduction, contributing to the enhanced capacity during the cycle reaction. Likewise, MnS partly contributed to the redox reaction. These findings demonstrate that MnS/MnO plays a crucial role in the charging and discharging processes for a stable long cycles.
Footnotes |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ma00066h |
‡ The first two authors contributed to the work equally. |
This journal is © The Royal Society of Chemistry 2024 |