Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2025 Apr 5;17(15):22822–22830. doi: 10.1021/acsami.5c02273

Oxygen Functional Groups Regulating Sulfur Distribution in Carbon Micropores to Enhance Solid-Phase Conversion Reactions for Lithium–Sulfur Batteries

Luna Yoshida †, Takashi Hakari ‡, Yukiko Matsui ‡, Yuki Orikasa §, Masashi Ishikawa †,*
PMCID: PMC12013621  PMID: 40186549

Abstract

graphic file with name am5c02273_0006.jpg

The performance of lithium–sulfur (Li–S) batteries is determined by the cathode, which is to a large extent affected by the low electrical conductivity of S and the dissolution of lithium polysulfides (Li2Sx). The confinement of S within microporous C is a promising method to address these challenges. The introduction of O-containing functional groups inside the C micropores improves the capacity for solid-phase conversion in Li–S batteries. However, the mechanism behind this enhanced performance, particularly the role of the O-containing functional groups on S inside the pores, remains unclear. In this study, we investigate the effect of these functional groups on S and/or the Li2Sx inside the C micropores, focusing on their impact on the electrochemical efficiency and the suppression of polysulfide migration. Electrochemical impedance spectroscopy measurements show that these O-containing functional groups accelerate charge transfer reactions and Li+ ion diffusion. Cross-sectional scanning transmission electron microscopy-electron energy loss spectroscopy of the S–C composites reveals that, without O-containing functional groups, S and/or Li2Sx migrate and localize to the inner edge of the carbon host during cycling. In contrast, the presence of O-containing functional groups inside the pores of the microporous C host maintains a uniform distribution of S and/or Li2Sx within the C micropores, explaining the improved solid-phase conversion performance in Li–S batteries. In conclusion, this paper proposes a new design for the cathode of high-performance Li–S batteries. For the first time, experimental evidence is provided to confirm the mechanism whereby the introduction of O-containing functional groups into microporous C enhances the performance of Li–S batteries by lowering the resistance and preventing Li2Sx migration. These modifications improve the electrochemical efficiency and offer insights for developing more effective cathodes to advance the commercialization of Li–S batteries.

Keywords: lithium−sulfur batteries, microporous carbon, oxygen-containing functional group, charge transfer, Li+ diffusion, sulfur distribution

Introduction

The development of advanced rechargeable batteries with high energy densities is essential for efficiently utilizing renewable energy and combating global warming. Among post-Li-ion battery technologies, Li–S batteries (LSBs) have attracted considerable attention owing to their high theoretical energy density (2600 Wh kg–1 and 2800 Wh L–1) and the abundance of S as a cathode active material.1 However, the LSB performance is affected by the cathode properties, such as the low electrical conductivity of S and its discharge product, Li2S (5.0 × 10–30 and 9.2 × 10–9 S cm–1 at 25 °C, respectively),2 significant volume changes during lithiation/delithiation,3 and the dissolution of intermediate products (Li2Sx, x = 6–8). These issues lead to active material loss, contamination of the Li-metal anode, and the shuttle effect.4,5 Consequently, they induce low S utilization, reduce the Coulombic efficiency, and lead to capacity decay, all of which significantly degrade the LSB performance and delay their large-scale commercialization.

The confinement of S within mesoporous C (pore diameter 2–50 nm) is regarded as a promising method to address the above-mentioned challenges faced by LSBs.6 This approach enhances the electronic conductivity of the cathode, mitigates volume expansion of the active cathode material, and prevents the dissolution of Li2Sx (x = 6–8). Nevertheless, the weak van der Waals adsorption interactions between nonpolar C materials and polar Li2Sx are insufficient to fully prevent Li2Sx (x = 6–8) dissolution. Attempts to more effectively block the dissolution of Li2Sx (x = 6–8) have encouraged the development of various modified mesoporous C materials that incorporate polar materials such as heteroatoms (N, S, and B),7,8 functional groups (amine, amide, and lithium sulfonate),9,10 polymers (polyethylene glycol and poly(3,4-ethylene-dioxythiophene)-poly(styrenesulfonate) (PEDOT–PSS)),6,11 and metal compounds (MnO2 and MoS2).12,13 These modified mesoporous C–S composite cathodes exhibit stable cycling performance and high capacity due to their strong Li2Sx chemisorption.

In our previous study, we identified microporous C (MC, pore diameter <2 nm) as a promising host material for S.14−16 The S that impregnated the MC pores underwent direct transformation into solid Li2Sx (x = 1, 2) without forming Li2Sx (x = 4–8) to effectively suppress Li2Sx (x = 6–8) dissolution. In addition, polar O-containing functional groups formed on MC through oxidation treatment lower the internal resistance and enhance the capacity of LSBs based on solid-phase conversion reactions (Figure S1).17,18 Oxidized MC–S composites with abundant O-containing functional groups exhibit lower electrolyte wettability compared with those of untreated MC–S composites, alleviating excessive electrolyte decomposition and minimizing the formation of a thick cathode electrolyte interphase (CEI).18 This reduces the interface resistance (Rint) of the oxidized MC–S cathode, thereby improving the LSB performance. However, although the role of O-containing functional groups that formed on the surface of MC has been studied extensively, the direct effect of O-containing functional groups formed inside the micropores on the S that impregnated the MC remains unclear.

In this study, we focused on the relationship between the O-containing functional groups within the MC pores and the S that impregnated these pores to elucidate the mechanism behind the improved LSB performance. First, X-ray absorption near edge structure (XANES) measurements were carried out to confirm that some O-containing functional groups formed on/within MC remained intact even after charge–discharge cycling. Next, electrochemical impedance spectroscopy (EIS) measurements revealed that the O-containing functional groups promote charge transfer reactions and Li+ diffusion within the cathodes. Finally, thinned MC–S composite particles extracted from the cathodes before and after cycling were examined with scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS). These results led us to conclude that the O-containing functional groups and/or oxygen compounds derived from these O-containing functional groups suppressed the migration and localization of S and Li2Sx within the MC pores during cycling, thereby enlarging the reaction area to ultimately improve the LSB performance. This is the first time the effect of O-containing functional groups residing inside the carbon micropores on the sulfur active materials has been clarified by analyzing the cathode materials.

Results and Discussion

Characterization of O-Containing Functional Groups Using X-ray Absorption Near-Edge Structure Measurement

XANES measurements were performed to analyze the state of O-containing functional groups on/within MC oxidized with 70 wt % nitric acid (70 HNO3 MC) before and after cycling. The O K-edge XANES spectra of the MC–S and 70 HNO3 MC–S powders were measured using the total electron yield (TEY) mode for surface-sensitive detection and partial fluorescence yield (PFY) mode for bulk-sensitive detection, as shown in Figure 1a,b, respectively. All spectra exhibited sharp pre-edge features at 530.1 and 531.5 eV and broad whiteline in the region 539.1–543.3 eV, which were attributed to carbonyl and carboxyl functional groups.20 The pre-edge peak intensity for the 70 HNO3 MC–S powder was higher than that of the MC–S powder, indicating that oxidation led to the formation of O-containing functional groups on/within the 70 HNO3 MC. Figure 1c,d shows the O K-edge XANES spectra of pristine cathodes composed of MC–S and 70 HNO3 MC–S. The PFY-mode spectra exhibited a higher pre-edge peak at 530.1 and 531.5 eV for the pristine 70 HNO3 MC–S cathode compared with that of the pristine MC–S, indicating that O-containing functional groups produced during the oxidation treatment remained intact throughout the electrode-preparation process. Conversely, the TEY-mode spectra of pristine MC–S and 70 HNO3 MC–S cathodes exhibited similar shapes. This observation was attributed to the coverage of the electrode surface by carboxymethyl cellulose and styrene–butadiene rubber. To obtain the XANES spectra of cycled MC–S or 70 HNO3 MC–S cathodes, the LSBs containing these cathodes were subjected to charge–discharge cycles, and the cathodes were removed from the cell at a discharge voltage of 1.5 V or charge voltage of 2.5 V during the 10th cycle (Figure 1e–h). In the TEY mode, the spectra of these cathodes were nearly identical, regardless of whether the MC was oxidized, due to the deposition of electrolyte decomposition products on the cathodes (Figure 1e,g). Conversely, in the PFY mode, the peak corresponding to O-containing functional groups (530.1 and 531.5 eV) was observed for the charge–discharged cathodes, with the intensity being higher for the 70 HNO3 MC–S cathode than for the MC–S cathode (Figure 1f, h). These results confirmed the presence of O-containing functional groups formed on/within the micropores of 70 HNO3 MC, even after cycling. On the other hand, the spectra of the charge–discharged cathodes showed a new pre-edge peak around 532.1 eV. The peak intensity of the 70 HNO3 MC–S cathodes was higher than that of the MC–S cathodes, which implied that some O-containing functional groups changed into oxygen compounds. Moreover, the cathodes subjected to charge–discharge cycling for 50 cycles were investigated using the same procedure, yielding results consistent with the observations above (Figure S2).

Figure 1.

Figure 1

Open in a new tab

O K-edge X-ray absorption near edge structure spectra of MC–S powders and MC–S electrodes before and after cycling. Spectra for MC–S and 70 HNO3 MC–S powders obtained in the (a) total electron yield (TEY) and (b) partial fluorescence yield (PFY) modes. Pristine MC–S and 70 HNO3 MC–S cathodes, not yet in contact with the electrolyte, were measured in the (c) TEY and (d) PFY modes. Charge–discharged MC–S and 70 HNO3 MC–S cathodes, extracted from the cells at the 10th discharge cycle at 1.5 V, analyzed in the (e) TEY and (f) PFY modes. Similarly, charge–discharged MC–S and 70 HNO3 MC–S cathodes, extracted from the cells at the 10th charge cycle at 2.5 V, analyzed in the (g) TEY and (h) PFY modes.

Electrochemical Impedance Spectroscopy

The LSBs with oxidized MC–S had a higher discharge voltage compared with those with untreated MC–S,18 suggesting that the O-containing functional groups formed on/within the oxidized MC lowered the internal resistance of the LSB during cycling. To investigate the effect of these O-containing functional groups on the resistive components of the LSB, EIS measurements were performed on MC–S/Li and 70 HNO3 MC–S/Li cells for various numbers of cycles and at different voltage levels (Figures 2a and S3). Figure 2b, c shows the Nyquist plots recorded at 2.5 V during the charge process in the 10th and 50th cycles. The plots of both MC–S and 70 HNO3 MC–S intercept the horizontal axis in the high-frequency region, and exhibit two semicircles around 3.1 kHz and 0.15 Hz and an inclined linear section. The high-frequency semicircle, low-frequency semicircle, and linear section correspond to the Rint, charge transfer resistance (Rct), and Warburg impedance (Zw), respectively. The Rint values for the MC–S/Li and 70 HNO3 MC–S/Li cells at various voltage levels during the 10th and 50th discharge and charge processes are summarized in Figure 2d,e. At all measuring points, Rint of 70 HNO3 MC–S was lower than that of MC–S. This can be understood as follows. The O-containing functional groups formed on the oxidized MC–S decreased the affinity of the cathode for the electrolyte.18 This is because the polar O-containing functional groups with their negative charges are strongly repelled by the TFSI– anion and nonpolar HFE, both of which are components of the electrolyte. The lowered affinity of the cathode for the electrolyte suppressed electrolyte decomposition and formation of the CEI. Therefore, Rcei of 70 HNO3 MC–S was lower than that of MC–S. Figure 2f shows the Arrhenius plots of 1/Rint derived from the temperature dependence of the impedance spectra of the cells at 2.5 V during the 10th and 50th charge processes. The activation energy of 1/Rint (Eaint) measured at the 10th cycle was 0.52 eV for MC–S and 0.23 eV for 70 HNO3 MC–S. The Eaint of MC–S differs from that of 70 HNO3 MC–S because of the different Li+ conduction mechanism in their CEIs, attributable to the different CEI morphologies or CEI compositions. After the 50th cycle, MC–S and 70 HNO3 MC–S exhibited Eaint values nearly identical to those of 70 HNO3 MC–S after the 10th cycle (0.30 and 0.28 eV, respectively). The significant difference in the Eaint values for MC–S between the 10th and 50th cycles indicated that continuous electrolyte decomposition, or changes in the CEI, occurred on the MC–S cathode during cycling, with a stable CEI gradually forming over time. In contrast, the nearly constant Eaint for 70 HNO3 MC–S that extended from the early to the later cycles suggested that the O-containing functional groups on 70 HNO3 MC contributed to the formation of a stable CEI. Our previous study revealed that electrolyte decomposition was effectively suppressed on the 70 HNO3 MC–S cathode with the large amount of O-containing functional groups, which caused a thin and stable CEI to start forming during the early cycles.18 Considering that the formation of the CEI would be sensitive to the forming conditions, especially those of the initial cycle, the cycling performance and impedance of the cells may depend on the prevailing charge/discharge conditions during the initial cycle, and we aim to investigate this in the future.

Figure 2.

Figure 2

Open in a new tab

Electrochemical performances, impedance spectroscopy, and activation energy measurements of the MC–S and 70 HNO3 MC–S cathode/Li anode cells at the 10th and 50th cycles. (a) Discharge–charge curves for the 10th and 50th cycles. The blue solid circles indicate the measurement points for electrochemical impedance spectroscopy. Data reconstructed from the original results in ref (18). Nyquist plots at 2.5 V during the (b) 10th and (c) 50th charge process. Interfacial resistance (Rint) at various voltages during the (d) discharge and (e) charge processes. (f) Activation energy of the 1/Rint (Eaint) at 2.5 V during the charge process. Charge transfer resistance (Rct) at various voltages during the (g) discharge and (h) charge processes. (i) Activation energy of 1/Rct (Eact) at 2.5 V during the charge process. Warburg coefficient (σ) at various voltages during the (j) discharge and (k) charge processes. The open circles and solid circles in (d–k) indicate the 10th and 50th discharge and charge processes, respectively.

The influence of these O-containing functional groups on/within 70 HNO3 MC on the Rct was explored, and the results are presented in Figure 2g, h. The Rct value in the 10th and 50th cycles, determined at various voltage levels, was lowered by the oxidation treatment of MC, indicating that the O-containing functional groups formed on/within the 70 HNO3 MC facilitated the charge transfer reaction. To further analyze the effect of the oxidation treatment on the charge-transfer process, Arrhenius plots were generated (Figure 2i). The activation energy of 1/Rct (Eact) measured in the 10th cycle increased from 0.50 to 0.67 eV after the oxidation treatment. The Eact values for the MC–S/Li (0.65 eV) and 70 HNO3 MC–S/Li cells (0.72 eV) after the 50th cycle also showed trends consistent with these observations. Theoretically, the Eact should have remained the same for MC–S and 70 HNO3 MC–S. This is because S, as the active material, has impregnated the MC pores in both electrodes; thus, the reaction mechanism would be expected to be the same with or without the oxidation treatment. However, the Eact of MC–S clearly differs from that of 70 HNO3 MC–S. This difference is elaborated upon in the discussion of the STEM-EELS analysis provided below.

To evaluate the Li+ diffusion within the cathode, the Warburg coefficient (σ) of the MC–S/Li and 70 HNO3 MC–S/Li cells was calculated from the EIS spectra (Figure 2j,k). During the 10th and 50th cycle, regardless of the voltage level, 70 HNO3 MC–S exhibited a lower σ compared with that of MC–S. The oxidation treatment produced O-containing functional groups not only on the surface of 70 HNO3 MC but also within its micropores. These O-containing functional groups facilitated Li+ diffusion both on the surface and inside the 70 HNO3 MC–S particles. In addition, the rate capabilities of the MC–S and 70 HNO3 MC–S/Li cells were measured (Figure S4). The higher discharge capacity and Coulombic efficiency of 70 HNO3 MC–S than those of MC–S for rates from 0.1 to 0.5 C indicate that the O-containing functional groups remain effective irrespective of the current. Based on these results, we concluded that the internal resistance, including the Rint, Rct, and σ, of the LSBs was lowered by the presence of the O-containing functional groups formed on/within 70 HNO3 MC, thereby improving the LSB performance.

Observation of Oxidized Microporous C–S Composites using Scanning Transmission Electron Microscopy-Electron Energy Loss Spectroscopy

To investigate the effect of the O-containing functional groups on S inside the MC–S particles, STEM-EELS observations were performed on the cross-section of the MC–S and 70 HNO3 MC–S particles extracted from the cathodes. Figure 3a,b shows the STEM images and corresponding EELS mappings of the MC–S and 70 HNO3 MC–S particles from the pristine cathodes. In the high-angle annular dark field (HAADF) images, cross sections of the particles of both MC–S and 70 HNO3 MC–S exhibited no contrast, indicating a uniform elemental distribution within the particles. The C and S mappings revealed homogeneous distributions in both samples, with S being uniformly confined to the pores of the MC and 70 HNO3 MC. Moreover, the high-intensity O signal inside the 70 HNO3 MC–S particles confirmed that O-containing functional groups had formed within the micropores as a result of the oxidation treatment.

Figure 3.

Figure 3

Open in a new tab

Scanning transmission electron microscopy images and corresponding electron energy loss spectroscopy mappings after focused ion beam milling. Cross sections of (a) pristine MC–S, (b) pristine 70 HNO3 MC–S, (c) 50th cycle MC–S, and (d) 50th cycle 70 HNO3 MC–S particles extracted from each cathode. The white scale bars represent 0.2 μm or 100 nm.

The elemental distribution in the MC–S particles extracted from the 50th charge–discharge cathode stood in marked contrast to that of the 70 HNO3 MC–S particles (Figure 3c). The near-surface area of the MC–S particles, which appeared white in the HAADF images, contained a higher concentration of Li and S compared with that of the bulk area, which appeared black. This proved that S and/or Li2Sx migrated and were localized in the near-surface area during charge–discharge cycling. The migration and localization of S and/or Li2Sx, induced by cycling, could be explained as follows: During charge–discharge, the volume changes of S and/or Li2Sx partially destroyed and expanded the MC pore walls. This weakened the capillary force that held S and/or Li2Sx,21 causing the S and/or Li2Sx in the MC pores to migrate to and localize in the near-surface area. In contrast, the HAADF images of the cycled 70 HNO3 MC–S particles presented no contrast and revealed a uniform elemental distribution in the EELS mappings (Figures 3d and 4b). Moreover, the 50th cycle MC–S particles were smaller than the 50th cycle 70 HNO3 MC–S particles (Figure 4). This could be understood as follows: As shown in Figure 4a, cracks appeared in the 10th cycle MC–S particles along the areas where the active materials were localized, which appeared white in the HAADF images. This was due to the repeated expansion and shrinkage of localized S and/or Li2Sx during cycling. This localized expansion and shrinkage fragmented the MC–S particles and decreased the particle size. In the case of 70 HNO3 MC–S, the particle shape and size remained unchanged after charge–discharge cycling. The uniform distribution of S within the 70 HNO3 MC–S particles may have dispersed the stress caused by the volume changes of S to prevent the particles from cracking. Based on the above-mentioned results, the migration and localization of S and/or Li2Sx to the region near the particle surface, as well as the observed disintegration of particles in MC–S during charge–discharge cycling, were suppressed by the O-containing functional groups within the MC pores. This dramatic difference in the distribution of S within the MC particles induced by the oxidation treatment affected the reaction mechanism and resulted in the difference in Eact. Furthermore, the formation of the CEI on the MC–S surface during charge–discharge cycling was suppressed in 70 HNO3 MC–S. This phenomenon is consistent with findings from previous studies.18

Figure 4.

Figure 4

Open in a new tab

Scanning transmission electron microscopy observation after focused ion beam milling. High-angle annular dark and bright field images of (a) MC–S and (b) 70 HNO3 MC–S particles extracted from the pristine cathodes and after the 1st, 10th, and 50th cycles. White scale bars represent 2 μm.

Mechanism behind the Improved Li–S Battery Performance due to O-Containing Functional Groups

Based on the results of this study, the oxidation of MC as an S host improved the LSB performance through the following mechanism (Scheme 1): For MC–S that had not been oxidized, the uniformly confined S within the MC pores migrated to and localized in the vicinity of the particle surface during charge–discharge cycling. This localization of S and/or Li2Sx reduced the contact area with the pore walls, that is, the electrical contact area, which led to a lower utilization rate of S and/or Li2Sx. However, excess electrolyte decomposition and the formation of a thick CEI on the surface of the untreated MC–S cathode18 suppressed Li2Sx dissolution, which enabled the LSBs with untreated MC–S cathodes to continue charging and discharging.22 In contrast, for oxidized MC–S, which had a large number of O-containing functional groups on/within the pores, the migration of S and/or Li2Sx was suppressed for the following reasons: (i) dipole–dipole interactions or the formation of covalent Li–O bonds between the O-containing functional groups introduced into the oxidized MC particles and polar Li2Sx,23,24 and (ii) the anchoring effect due to the bulky structure of the O-containing functional groups. In addition, after cycling, the oxidized MC–S cathodes had abundant oxygen compounds, as confirmed by XANES measurements; the products may also prevent Li2Sx (or S) migration and localization. The suppression of S and/or Li2Sx migration and subsequent localization enlarged the reaction area within the pores, which in turn lowered the Rct and facilitated Li+ diffusion to improve the LSB performance. To date, the use of functional groups, heteroatoms, and metal compounds etc. to modify carbon materials that serve as S hosts has been reported to improve the LSB performance as a result of the following effects: (i) chemisorption toward soluble Li2Sx suppresses the loss of cathode active materials, pollution of the Li anode, and undesired shuttle effects, and (ii) the catalytic effect accelerates the Li2Sx conversion reaction.25−28 The effects discovered in our study, that is, the suppression of Li2Sx (or S) migration and localization inside the MC pores to explain the performance improvement of LSBs, are therefore new mechanisms that differ from the established theory.

Scheme 1. Mechanism Underlying the Improved Performance of Li–S Batteries Attributed to O-Containing Functional Groups on/within Microporous C.

Scheme 1

Open in a new tab

In this study, MC–S composites were used as a model for S cathodes capable of solid-phase conversion reactions. These reactions are also known to occur in the case of electrolytes and additives that form an effective CEI on mesoporous C-containing cathodes.29 The improved LSB performance owing to the O-containing functional groups was also observed for solid-phase conversion-based LSBs due to the electrolytes and additives (Figure S5).30,31 This suggests that the effects of the O-containing functional groups identified in this study are likely to be generally effective for solid-phase conversion-based LSBs. Our findings, therefore, have significant implications for the design of future LSB cathodes.

Conclusions

We elucidated the mechanism behind the improved LSB performance owing to the oxidation of MC as an S host. O-containing functional groups were formed on and within the MC during oxidation treatment and remained intact even after cycling, as proven by XANES measurements. The presence of these O-containing functional groups lowered not only the Rint but also the Rct and σ. To investigate the direct effect of O-containing functional groups on the cathode active materials within the micropores, the MC–S composite cathodes were analyzed before and after cycling using STEM-EELS. The results revealed that the O-containing functional groups suppressed the migration and localization of Li2Sx (or S) inside the MC during charge–discharge cycling. This provided a large reaction area within the micropores, thereby enhancing the LSB performance. The findings for the first time highlight the importance of modifications within the C micropores, which are anticipated to significantly influence the future design of high-performance LSB cathodes.

Methods

Preparation of Oxidized Microporous C–S Composites

Commercial alkali-activated MC (Kansai Coke and Chemicals Co. Ltd.; ash content <1.0%) was used as a starting material. MC (2 g) was mixed with 70 wt % nitric acid (HNO3, 100 mL, FUJIFILM Wako Pure Chemical Co.) and refluxed at 120 °C for 2 h. The excess HNO3 was removed by washing the MC with deionized water, and the product was dried under vacuum at 80 °C to yield 70 HNO3 MC. The 70 HNO3 MC–S composite was prepared using the melt–diffusion method.6 A mixture of 70 HNO3 MC and S (S, FUJIFILM Wako Pure Chemical Co. > 99%) was prepared in a weight ratio of 52:48 (70 HNO3 MC:S). The mixture was sealed in a stainless-steel container and heated at 155 °C for 5 h to form 70 HNO3 MC–S. The same procedure was used to confine S in untreated MC, resulting in MC–S. All experiments were performed in air.

Assembly of Oxidized Microporous C–S Composite Cathode/Li Anode Cells

The 70 HNO3 MC–S cathode was prepared by mixing 70 HNO3 MC–S, acetylene black (AB, HS-100, Denka Co. Ltd.), carboxymethyl cellulose (F-AG, DKS Co. Ltd.), and styrene–butadiene rubber (TRD 2001, JSR Co.) in a weight ratio of 89:5:3:3. The resulting viscous slurry was cast onto an Al current collector. The electrode sheet was pressed by applying a constant load and punched into circular disks (diameter: 12 mm). The cathodes were dried under vacuum at 50 °C before use. The MC–S cathode was prepared following the same procedure. The areal S loading for both the 70 HNO3 MC–S and MC–S cathodes was 1.0 ± 0.4 mg cm–2. All cathodes were prepared in air.

The Li-metal anode was prepared by punching Li foil (200 μm, Honjo Metal Co. Ltd.) into circular disks (diameter: 13 mm). The electrolyte was prepared by mixing lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, Solvay Japan, Ltd.), tetraglyme (G4, Nippon Nyukazai Co.), and 1,1,2,2-tetrafluoroethyl 2,2,3,3-tetrafluoropropylether (HFE Daikin Industries Co.) in a molar ratio of 10:8:40.19 Two-electrode cells were assembled in a high-purity Ar-filled glovebox using a 70 HNO3 MC–S or MC–S cathode, Li-metal anode, electrolyte, and two microporous polypropylene separators. The electrolyte-to-S ratio (E/S) was fixed at 18.7 μL mg–1.

X-ray Absorption Near-Edge Structure Measurement

Charge–discharge cathodes were prepared at various potentials for XANES measurements by conducting galvanostatic charge–discharge measurements in the voltage range of 1–3 V and at a current density of 0.1 C (1 C = 1672 mA g–1). After cycling (10th and 50th cycles), the cathodes were extracted from the cell in a glovebox and washed three times with HFE to remove residual electrolyte. XANES measurements were conducted at the O K-edge at the SR center (BL-11, Ritsumeikan University) to analyze the O-containing compounds on and within the cathodes before and after cycling. The XANES spectra were acquired simultaneously in the TEY, for surface-sensitive detection, and PFY modes, for bulk-sensitive detection. All procedures were performed under vacuum or in an Ar atmosphere.

Electrochemical Impedance Spectroscopy

To reach the target cycle number and voltage for EIS measurements, galvanostatic charge–discharge was performed under the same conditions as described above. After maintaining the cell at the open-circuit voltage for 2 h at various potentials during the 10th and 50th cycles, EIS was performed using a multipotentiostat (VersaSTAT 4, AMETEK Inc.) to evaluate the voltage and temperature dependence of the resistive components in LSBs. Measurements were conducted in the frequency range of 1.0 × 10–2–1.0 × 106 Hz with an alternating current voltage amplitude of 10 mV. Activation energy measurements were performed within a temperature range of 0–25 °C. The resulting Nyquist plots were fitted to an equivalent electrical circuit comprising an ohmic resistor in series with two R//CPE elements and a Warburg element (Figure S6). σ, which is inversely proportional to the Li+ diffusion coefficient in the cathode active material, is expressed as

graphic file with name am5c02273_m001.jpg

where Z′, f, and B denote the real part of the Zw, frequency, and real part of the impedance excluding Zw, respectively.

Scanning Transmission Electron Microscopy-Electron Energy Loss Spectroscopy

Charged–discharged cathodes were prepared for STEM-EELS analysis by conducting galvanostatic charge–discharge measurements under the same conditions as described above. After completing 50 cycles, the cathode was removed from the cell in a glovebox and washed three times with HFE to eliminate residual electrolyte. A portion of the active materials from the MC–S and 70 HNO3 MC–S cathodes, both pre- and post-cycling, was extracted and thinned to approximately 70 nm using a focused ion beam/scanning electron microscope (NB5000, Hitachi High-Tech Co.). Elemental concentration maps of the cathode active materials before and after cycling were obtained using STEM-EELS (HD-2700, Hitachi High-Tech Co.) with an accelerating voltage of 200 kV. All procedures were performed under vacuum or in an Ar atmosphere.

Acknowledgments

This study was supported by JST ALCA-SPRING (grant number JPMJAL1301, Japan), JST SPRING (grant number JPMJSP2150, Japan), and JST GteX (grant number JPMJGX23S0, Japan). We extend our gratitude to Kansai Coke and Chemicals Co. Ltd. for providing the MC.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.5c02273.

  • Electrochemical performances of MC–S or 70 HNO3 MC–S cathode/Li anode cells; O K-edge XANES spectra of MC–S or 70 HNO3 MC–S electrodes during the 50th cycle; measurement points for EIS; Nyquist plots of MC–S or 70 HNO3 MC–S cathode/Li anode cells; summary of EIS for MC–S/Li and 70 HNO3 MC–S/Li cells at the 50th cycle; electrochemical performances of micromesoporous C–S cathode/Li anode cells; and equivalent electrical circuit used for EIS (PDF)

Author Present Address

∥ Department of Energy and Environment, National Institute of Advanced Industrial Science and Technology, 1-8-31 Midorigaoka, Ikeda 563-0026, Japan

Author Contributions

L.Y.: Writing – original draft, Data curation, Investigation, Visualization, Conceptualization. T.H.: Writing – review and editing. Y.M.: Writing – review and editing, Conceptualization. Y.O.: Writing – review and editing, Resources, Supervision. M.I.: Writing – review and editing, Supervision, Project administration, Funding acquisition.

The authors declare no competing financial interest.

Supplementary Material

am5c02273_si_001.pdf (864.7KB, pdf)

References

  1. Manthiram A.; Chung S.-H.; Zu C. Lithium–Sulfur Batteries: Progress and Prospects. Adv. Mater. 2015, 27, 1980–2006. 10.1002/adma.201405115. [DOI] [PubMed] [Google Scholar]
  2. Choi S.; Yoon I.; Nichols W. T.; Shin D. Carbon-Coated Li2S Cathode for Improving the Electrochemical Properties of an All-Solid-State Lithium-Sulfur Battery Using Li2S-P2S5 Solid Electrolyte. Ceram. Int. 2018, 44, 7450–7453. 10.1016/j.ceramint.2018.01.104. [DOI] [Google Scholar]
  3. Miao Z.; Li Y.; Xiao X.; Sun Q.; He B.; Chen X.; Liao Y.; Zhang Y.; Yuan L.; Yan Z.; Li Z.; Huang Y. Direct Optical Fiber Monitor on Stress Evolution of the Sulfur-Based Cathodes for Lithium–Sulfur Batteries. Energy Environ. Sci. 2022, 15, 2029–2038. 10.1039/D2EE00007E. [DOI] [Google Scholar]
  4. Ren W.; Ma W.; Zhang S.; Tang B. Recent Advances in Shuttle Effect Inhibition for Lithium Sulfur Batteries. Energy Storage Mater. 2019, 23, 707–732. 10.1016/j.ensm.2019.02.022. [DOI] [Google Scholar]
  5. Hofmann A. F.; Fronczek D. N.; Bessler W. G. Mechanistic Modeling of Polysulfide Shuttle and Capacity Loss in Lithium–Sulfur Batteries. J. Power Sources 2014, 259, 300–310. 10.1016/j.jpowsour.2014.02.082. [DOI] [Google Scholar]
  6. Ji X.; Lee K. T.; Nazar L. F. A Highly Ordered Nanostructured Carbon–Sulphur Cathode for Lithium–Sulphur Batteries. Nat. Mater. 2009, 8, 500–506. 10.1038/nmat2460. [DOI] [PubMed] [Google Scholar]
  7. Pang Q.; Tang J.; Huang H.; Liang X.; Hart C.; Tam K. C.; Nazar L. F. A Nitrogen and Sulfur Dual-Doped Carbon Derived from Polyrhodanine@Cellulose for Advanced Lithium–Sulfur Batteries. Adv. Mater. 2015, 27, 6021–6028. 10.1002/adma.201502467. [DOI] [PubMed] [Google Scholar]
  8. Yang C.-P.; Yin Y.-X.; Ye H.; Jiang K.-C.; Zhang J.; Guo Y.-G. Insight into the Effect of Boron Doping on Sulfur/Carbon Cathode in Lithium–Sulfur Batteries. ACS Appl. Mater. Interfaces. 2014, 6, 8789–8795. 10.1021/am501627f. [DOI] [PubMed] [Google Scholar]
  9. Fretz S. J.; Agostini M.; Jankowski P.; Johansson P.; Matic A.; Palmqvist A. E. C. Amine- and Amide-Functionalized Mesoporous Carbons: A Strategy for Improving Sulfur/Host Interactions in Li–S Batteries. Batteries & Supercaps 2020, 3, 757–765. 10.1002/batt.202000027. [DOI] [Google Scholar]
  10. Fretz S. J.; Pal U.; Girard G. M. A.; Howlett P. C.; Palmqvist A. E. C. Lithium Sulfonate Functionalization of Carbon Cathodes as a Substitute for Lithium Nitrate in the Electrolyte of Lithium–Sulfur Batteries. Adv. Funct. Mater. 2020, 30, 2002485 10.1002/adfm.202002485. [DOI] [Google Scholar]
  11. Xiao P.; Bu F.; Yang G.; Zhang Y.; Xu Y. Integration of Graphene, Nano Sulfur, and Conducting Polymer into Compact, Flexible Lithium–Sulfur Battery Cathodes with Ultrahigh Volumetric Capacity and Superior Cycling Stability for Foldable Devices. Adv. Mater. 2017, 29, 1703324 10.1002/adma.201703324. [DOI] [PubMed] [Google Scholar]
  12. Li Z.; Zhang J.; Lou X. W. D. Hollow Carbon Nanofibers Filled with MnO2 Nanosheets as Efficient Sulfur Hosts for Lithium–Sulfur Batteries. Angew. Chem., Int. Ed. 2015, 54, 2886–12890. 10.1002/anie.201506972. [DOI] [PubMed] [Google Scholar]
  13. You Y.; Ye Y.; Wei M.; Sun W.; Tang Q.; Zhang J.; Chen X.; Li H.; Xu J. Three-Dimensional MoS2/rGO Foams as Efficient Sulfur Hosts for High-Performance Lithium-Sulfur Batteries. Chem. Eng. J. 2019, 355, 671–678. 10.1016/j.cej.2018.08.176. [DOI] [Google Scholar]
  14. Takahashi T.; Yamagata M.; Ishikawa M. A Sulfur–Microporous Carbon Composite Positive Electrode for Lithium/Sulfur and Silicon/Sulfur Rechargeble Batteries. Prog. Nat. Sci. 2015, 25, 612–621. 10.1016/j.pnsc.2015.11.011. [DOI] [Google Scholar]
  15. Yoshida L.; Matsui Y.; Deguchi M.; Hakari T.; Watanabe M.; Ishikawa M. Improvement of Lithium-Sulfur Battery Performance by Porous Carbon Selection and LiFSI/DME Electrolyte Optimization. ACS Appl. Mater. Interfaces 2023, 15, 37467–37476. 10.1021/acsami.3c06624. [DOI] [PubMed] [Google Scholar]
  16. Yoshida L.; Hakari T.; Matsui Y.; Ishikawa M. Polyglycerol-Functionalized Microporous Carbon/Sulfur Cathode for Li-S Battery. Electrochim. Acta 2022, 429, 141000 10.1016/j.electacta.2022.141000. [DOI] [Google Scholar]
  17. Okabe S.; Uchida S.; Matsui Y.; Yamagata M.; Ishikawa M. Performance Enhancement of Rechargeable Sulfur Cathode Utilizing Microporous Activated Carbon Composite. Electrochemistry 2017, 85, 671–674. 10.5796/electrochemistry.85.671. [DOI] [Google Scholar]
  18. Yoshida L.; Hakari T.; Matsui Y.; Deguchi M.; Yamamoto H.; Inoue M.; Ishikawa M. Understanding the Improved Performances of Lithium–Sulfur Batteries Containing Oxidized Microporous Carbon with an Affinity-Controlled Interphase as a Sulfur Host. J. Power Sources 2024, 624, 235572 10.1016/j.jpowsour.2024.235572. [DOI] [Google Scholar]
  19. Dokko K.; Tachikawa N.; Yamauchi K.; Tsuchiya M.; Yamazaki A.; Takashima E.; Park J.-W.; Ueno K.; Seki S.; Serizawa N.; Watanabe M. Solvate Ionic Liquid Electrolyte for Li–S Batteries. J. Electrochem. Soc. 2013, 160, A1304–A1310. 10.1149/2.111308jes. [DOI] [Google Scholar]
  20. Kim K.; Zhu P.; Li N.; Ma X.; Chen Y. Characterization of Oxygen Containing Functional Groups on Carbon Materials with Oxygen K-Edge X-ray Absorption near Edge Structure Spectroscopy. Carbon 2011, 49, 1745–1751. 10.1016/j.carbon.2010.12.060. [DOI] [Google Scholar]
  21. Butt H.-J.; Kappl M. Normal Capillary Forces. Adv. Colloid Interface Sci. 2009, 146, 48–60. 10.1016/j.cis.2008.10.002. [DOI] [PubMed] [Google Scholar]
  22. Kensy C.; Leistenschneider D.; Wang S.; Tanaka H.; Dörfler S.; Kaneko K.; Kaskel S. The Role of Carbon Electrodes Pore Size Distribution on the Formation of the Cathode–Electrolyte Interphase in Lithium–Sulfur Batteries. Batteries & Supercaps. 2021, 4, 612–622. 10.1002/batt.202000195. [DOI] [Google Scholar]
  23. Wasalathilake K. C.; Roknuzzaman M.; Ostrikov K. K.; Ayoko G. A.; Yan C. Interaction between functionalized graphene and sulfur compounds in a lithium–sulfur battery–a density functional theory investigation. RSC Adv. 2018, 8, 2271. 10.1039/C7RA11628D. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hou T.-Z.; Chen X.; Peng H.-J.; Huang J.-Q.; Li B.-Q.; Zhang Q.; Li B. Design Principles for Heteroatom-Doped Nanocarbon to Achieve Strong Anchoring of Polysulfides for Lithium–Sulfur Batteries. Small 2016, 12, 3283–3291. 10.1002/smll.201600809. [DOI] [PubMed] [Google Scholar]
  25. Lee D.; Yang J.; Yun W. C.; Lee J. W. Nitrogen-doped hierarchical porous carbon derived from CO2 for the high-performance cathode of lithium-sulfur battery. Electrochim. Acta 2025, 520, 145867 10.1016/j.electacta.2025.145867. [DOI] [Google Scholar]
  26. Heo J.; Hong J.-W.; Gu H. W.; Sung J.; Kim D.-H.; Kim J. H.; Kang S.; Lee Y.-J.; Choi H. Y.; Kim D.; Baeg K.-J.; Han J. T.; Park J.-W. A Promising Approach to Ultra-Flexible 1 Ah Lithium–Sulfur Batteries Using Oxygen-Functionalized Single-Walled Carbon Nanotubes. Adv. Sci. 2025, 12, 2406536 10.1002/advs.202406536. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lian Z.; Ma L.; Wu H.; Xiao H.; Yang Y.; Zhang J.; Zi J.; Chen X.; Wang W.; Li H. Accelerating sulfur redox kinetics by rare earth single-atom electrocatalysts toward efficient lithium–sulfur batteries. Appl. Catal. B: Environ. Energy 2025, 361, 124661 10.1016/j.apcatb.2024.124661. [DOI] [Google Scholar]
  28. Wang Y.; Zhou X.; Liu X.; Ma X.; Li Y.; Duan D.; Yuan Q.; Wang J.; Liu S. Research on the performance and mechanism of sulfonated activated carbon as cathode carrier for L–S batteries. J. Energy Storage 2024, 92, 112130 10.1016/j.est.2024.112130. [DOI] [Google Scholar]
  29. Torii Y.; Matsui Y.; Yamamoto K.; Uchida S.; Yamazaki S.; Watanabe T.; Nakanishi K.; Uchiyama T.; Uchimoto Y.; Ishikawa M. Preventing Capacity Fading in Lithium–Sulfur Batteries Using Sulfur Confinement in Mesoporous Carbon and Fluorinated Solvent-Based Electrolytes. J. Phys. Chem. C 2023, 127, 15069–15077. 10.1021/acs.jpcc.3c02992. [DOI] [Google Scholar]
  30. Tonoya T.; Ando H.; Takeichi N.; Senoh H.; Kojima T.; Hinago H.; Matsui Y.; Ishikawa M. Capacity Decay Mechanism of Lithium–Sulfur Batteries Using a Microporous Activated Carbon–Sulfur Composite as the Cathode Material. J. Phys. Chem. C 2023, 127, 10038–10044. 10.1021/acs.jpcc.3c01961. [DOI] [Google Scholar]
  31. Tonoya T.; Matsui Y.; Hinago H.; Ishikawa M. Microporous Activated Carbon Derived from Azulmic Acid Precursor with High Sulfur Loading and Its Application to Lithium-Sulfur Battery Cathode. Electrochem. Commun. 2022, 140, 107333 10.1016/j.elecom.2022.107333. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

am5c02273_si_001.pdf (864.7KB, pdf)

Articles from ACS Applied Materials & Interfaces are provided here courtesy of American Chemical Society

RESOURCES