Fluorinated imine modulating efficient sulfur redox kinetics and a stable solid electrolyte interphase in lithium–sulfur batteries

Abstract

The shuttle effect of lithium polysulfides (LiPSs) and the instability of the solid electrolyte interphase (SEI) lead to lithium dendrite growth and severe corrosion of lithium anodes (Li-anodes) for lithium–sulfur (Li–S) batteries. Herein, we introduce phenyl-1-(4-(trifluoromethyl)phenyl)ethan-1-imine (PTPEI), a fluorinated Schiff base molecule, as a novel dual-function electrolyte additive to enhance the sulfur redox kinetics at the cathode side. Simultaneously, it facilitates the formation of a dense, robust, and stable SEI enriched with lithium fluoride (LiF) on the anode side. With theoretical calculations, we reveal that molecular structure regulation strengthening van der Waals forces between PTPEI and LiPSs facilitates charge transfer by affecting the highest occupied molecular orbital (HOMO) level and improves the role of the PTPEI molecule catalyst in accelerating the sulfur redox kinetics. Furthermore, we demonstrate that the Schiff base molecular configuration facilitates the decomposition of PTPEI and expedites the formation of a stable LiF-enriched SEI, effectively protecting the Li-anode during cycling. As a result, the Li–S cell with PTPEI delivers an initial discharge capacity of 1190.9 mA h g⁻¹ with a capacity decay rate of 0.90% per cycle at 0.1C for 50 cycles at a high sulfur loading of 5.5 mg cm⁻² and low E/S ratio of 8 μL mg⁻¹.

---

Tong Wu

Dr Tong Wu, a master's supervisor, has extensive experience in the application research of lithium secondary batteries. In recent years, his research has increasingly focused on the intersection of nanoscience and energy chemistry. Specifically, he has explored the structure–activity relationships between material catalysis, energy storage properties, and nanoscale microstructures from the perspectives of intermolecular interactions, atomic valence states, and charge transfer mechanisms. Dr Wu has published nine peer-reviewed academic papers in prestigious international journals such as Journal of Power Sources, ACS Applied Materials & Interfaces, and Electrochimica Acta. Additionally, he has successfully obtained funding for two scientific research projects from the Inner Mongolia Autonomous Region Science Foundation.

1. Introduction

Lithium–sulfur (Li–S) batteries, with a remarkable theoretical energy density (2600 W h kg⁻¹) achieved through precipitation–dissolution conversion reactions occurring at the cathode interface, exhibit significant potential as next-generation rechargeable batteries.^1–5 One major issue facing Li–S batteries is the low sulfur utilization, which reduces the overall capacity of the battery.^6–10 The efforts for designing sulfur host materials have effectively addressed the issue of low sulfur utilization, enabling the achievement of a high specific capacity even under high sulfur loading.^11–15 However, the shuttle effect caused by long-chain lithium polysulfide (LiPS) dissolution during the cycling of Li–S batteries results in the loss of active sulfur.^16–18 Even worse, Li-anodes in Li–S batteries experience dendrite growth and the corrosion induced by soluble LiPSs, exacerbating uneven Li deposition, and ultimately leading to the short cycle life.^19–21 Various strategies, including the design of polysulfide anchoring materials,²² utilization of electrolyte additives,^23,24 and optimization of the electrolyte,^25–27 have been extensively explored to address the above challenges.

Among these strategies, the incorporation of functional additives into the electrolyte emerges as a cost-effective and convenient approach to effectively suppress the shuttle effect, to protect Li-anodes and to improve LiPS conversion. These additives can promote halide formation (e.g., lithium fluoride (LiF),²⁸ lithium chloride (LiCl),²⁹ and lithium bromide (LiBr)³⁰) on the surface of Li-anodes, facilitating the establishment of a stable solid electrolyte interphase (SEI). Notably, fluorinated molecules regulate the Li-anodes to form a SEI enriched with LiF.^31,32 Besides, the fluorinated molecule can enhance LiPS conversion by modifying reaction pathways or accelerating redox kinetics.^33,34 The aforementioned capability of the fluorinated molecule additives can be further enhanced through modification of their molecular structure. Jin et al. demonstrated that incorporating active end groups effectively enhances the release of fluoride ions in fluorinated molecules, thereby facilitating the formation of a fluorinated SEI.³⁵ Jing et al. indicated that among the three isomers of benzenedithiols, 1,4-benzenedithiol exhibits a more pronounced catalytic effect on LiPSs.³⁶ With these advancements, a new fluorinated molecule additive is urgently needed to promote the formation of a stable fluorinated SEI on the Li-anode and to catalyze the conversion of LiPSs to alleviate the LiPS shuttling for sulfur cathodes.

Herein, we present the employment of the fluorinated Schiff base organic molecule phenyl-1-(4-(trifluoromethyl)phenyl)ethan-1-imine (PTPEI) as a bifunctional electrolyte additive in a Li–S cell to modulate the sulfur reduction kinetics and construct a fluorinated SEI. The Li–S cell with PTPEI in electrolytes was evaluated to present the effective modulation of a molecular catalyst on sulfur redox kinetics and Li⁺ diffusion, suppressing the shuttling effects and improving the cycling stability. Based on the distinctive structure of the Schiff base molecule, the lone pair of electrons on the nitrogen atom possesses a remarkable capacity for electron donation, which can serve as an electron donor during the oxidation reaction of LiPSs and thereby exert a catalytic effect on the conversion of LiPSs. The energy levels and low-gradient isosurfaces (RDG) of PTPEI–Li₂S_x (x = 1, 2, 4, 6, and 8) compounds were then calculated to reveal the impact of the molecular configuration on the van der Waals interactions and charge transfer capability between PTPEI and Li₂S_x. The surface of the Li-anode after 100 cycles in a Li–S cell was examined by SEM, EDS mapping and XPS etching to reveal the protective effects of PTPEI on the Li-anode, and the composition and formation of a stable and robust SEI. The Li‖Li symmetric cells were further assembled to investigate the role of PTPEI in facilitating the stable overpotential and SEI formation at high current density and under long-life cycling conditions. And, the Li–S cell with PTPEI demonstrates stable cycling performance with a high sulfur loading and lean electrolyte.

2. Experimental

2.1. Preparation of CMK-3/S cathodes and electrolytes

The CMK-3/S composite was synthesized by blending sulfur powder (Aladdin Reagent, 99.95%) and ordered mesoporous carbon (CMK-3, XFNANO, INC) in a mass ratio of 7 : 3. The mixture was sealed in a 50 mL autoclave and gradually heated at a rate of 5 °C min⁻¹ until reaching 155 °C for 20 h. The sulfur content of the composite was determined through thermogravimetric analysis (TGA), where data were collected using a TG analyzer (NETZSCH STA449) under an N₂ atmosphere up to 600 °C, with a temperature step of 10 °C min⁻¹, revealing a sulfur content of 70.5% as shown in Fig. S1 in the ESI.†

The CMK-3/S cathodes were prepared using a slurry casting method. A slurry was created by mixing the CMK-3/S composite, Super P (Temigao Graphite Co., Ltd), and polyvinylidene fluoride (PVDF, Solef@5130, Solvay) with N-methyl-2-pyrrolidone (NMP, Aladdin Reagent, 99.5%) in a weight ratio of 8 : 1 : 1. The resulting slurry was then coated onto carbon-coated aluminum foil (Shanghai Zhaoyuan Industrial Co., Ltd) and dried at 60 °C overnight in a vacuum oven. Subsequently, circular shapes with a diameter of 14 mm were cut from the CMK-3/S cathode. Each cathode had an active sulfur loading of approximately 1.4 mg cm⁻² (5.5 mg cm⁻² for high sulfur loading).

The blank electrolyte used in this study is a commercial Li–S battery electrolyte (LS-002, DoDoChem) consisting of 1 M lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) and 0.01 M LiNO₃ dissolved in a solvent mixture of DOL/DME (1 : 1 by volume). PTPEI was synthesized via the acetylenic amination reaction between trifluoromethylaniline (TFMA) (Macklin Reagent, purity ≥ 98%) and phenylacetylene. Thin-layer chromatography analysis was subsequently conducted to confirm the absence of any residual phenylacetylene and TFMA, as previously reported.³⁷ The molecular structure and synthesis of PTPEI are illustrated in Fig. S2.† The hydrogen spectrum of the sample was analyzed by nuclear magnetic resonance (¹H NMR, Bruker Ascend 500 MHz NMR spectrometer). The four peaks a, b, c, and d within the range of 6.8–8 ppm correspond to the four types of hydrogen atoms (i.e., aromatic hydrogen) on the two benzene ring structures in PTPEI. The signal peak e at 2.23 ppm is attributed to the methyl (–CH₃) hydrogen atoms in PTPEI, confirming the expected product synthesized, as shown in Fig. S3a.† Fourier transform infrared spectroscopy was also used for auxiliary verification. The wavenumber range of 680–780 cm⁻¹ typically corresponds to the absorption peak of the C–F bond in the trifluoromethyl (CF₃) group, and the wavenumber range of 1520–1690 cm⁻¹ typically corresponds to the absorption peak of the C=N bond. Meanwhile, no obvious N–H absorption peak was observed in the wavenumber range of 3400–3600 cm⁻¹ (Fig. S3b†). The spectra in Fig. S3† provide evidence for the absence of phenylacetylene and TFMA residues. To optimize the electrolytes, different molarities (0.02, 0.01, and 0.005 mol L⁻¹) of PTPEI were introduced into the blank electrolyte while TFMA was added for comparison. All electrolyte optimizations were conducted within an argon-filled glove box (VIGOR, LG1200/750TS, H₂O < 1.0 ppm, O₂ < 1.0 ppm).

2.2. Assembly of the test cell and electrochemical measurements

The 2025-type coin cells were tested within the glove box. Li‖Li symmetric cells were assembled using Li foil (China Energy Lithium Co., Ltd) as both the cathode and anode, along with a single-layer separator (Celgard 2400) soaked in the aforementioned electrolytes. For the assembly of Li–S cells, CMK-3/S was used as the cathode instead. The electrolyte-to-sulfur (E/S) ratio was set at 20 μL mg⁻¹ (8 μL mg⁻¹ for high sulfur loading). The electrochemical properties of the Li–S cell using electrolyte with PTPEI (referred to as “cell with PTPEI”) are compared to those of the Li–S cell using blank electrolyte (referred to as “cell with blank”) and electrolyte with TFMA (referred to as “cell with TFMA”). Electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV) measurements were performed using an electrochemical workstation (VMP-3, Biologic), applying a voltage amplitude of 10 mV in the frequency range from 100 kHz to 100 mHz for EIS and a scanning rate of 0.1 mV s⁻¹ in the voltage range from 1.6 V to 2.8 V for CV. The Li₂S deposition experiment was conducted in a coin cell using carbon paper as the cathode and Li foil as the anode. During the assembly process, 25 μL of 0.25 M Li₂S₈ solution was added to the cathode side, while another 25 μL of electrolyte, with or without additives (TFMA or PTPEI). According to Faraday's law, the conversion capacity of Li₂S is calculated by using the amount of charge. The galvanostatic discharge/charge profile, rate capability, cyclability, and galvanostatic intermittent titration technique (GITT) measurements were performed using a LAND system (CTA2001A, Wuhan Land Electronic Co., Ltd) at various cycling rates with cut-off voltages ranging from 1.6 V to 2.8 V (0.1C, 0.2C, 0.5C, 1C, and 2C; where 1C corresponds to a sulfur capacity of 1675 mA h g⁻¹).

2.3. Electrode characterization

UV-visible absorption spectroscopy (UV-2310II, Shanghai Tianwei) was employed to examine the changes in sulfur in the solution. The morphologies of the Li-anode before and after cycling were examined using a scanning electron microscope (SEM, Hitachi, S-4800, 5 kV, 5 μA). Following the completion of 100 cycles, we disassembled Li–S cells and washed the Li-anodes with anhydrous dimethyl carbonate (DMC) inside a glove box. We followed a similar procedure for extracting and treating the Li-metal from symmetrical Li‖Li cycles after a duration of 20 h. To investigate their surface chemical bonds, we utilized X-ray photoelectron spectroscopy (XPS) with Kratos AXIS SUPRA⁺ equipment that was equipped with Al Kα radiation (hv = 4000 eV), employing an emission angle of 90°. The SEI was collected from the surface of the Li-anode, followed by immersion in deuterated chloroform for extraction of organic constituents, which were subsequently analyzed using ¹H NMR spectroscopy.

2.4. Computational details

The underlying mechanism governing the regulation of electrochemical reactions by PTPEI and TFMA was elucidated through density functional theory (DFT) calculations, which were performed using the Gaussian 09 program package. Full ground state geometry optimizations of Li₂S_x (x = 1, 2, 4, 6, 8) compounds, both before and after binding to PTPEI and TFMA, were conducted at the wb97xd/6-311+g(d,p) level.

3. Results and discussion

To investigate the efficacy of PTPEI in interacting with LiPSs, we monitored the color change of PTPEI in a Li₂S₆ solution and subsequently extracted the supernatant for UV-visible absorption spectroscopy analysis. After the addition of PTPEI, the color of Li₂S₆ solution changed from yellow to light yellow, and the S²⁻ absorption peak intensity at 264 nm has been significantly reduced, as shown in Fig. S4.† The experimental results demonstrate a decrease in the concentration of Li₂S₆ in the solution, indicating that PTPEI is an effective additive capable of reacting with unstable LiPSs. Subsequently, the effect of PTPEI on the electrochemical performance of the Li–S cell was assessed. The redox peaks in the cell with PTPEI exhibit a noticeable shift compared to those of the cell with blank and the cell with TFMA (Fig. 1a). Specifically, the oxidation peak (A) in the cell with PTPEI occurs at 2.38 V, with reduction peaks (C₁ and C₂) observed at 2.31 V and 2.03 V respectively, contrasting with those of the control cells (blank: A at 2.48 V, C₁ at 2.26 V and C₂ at 1.97 V; TFMA: A at 2.40 V, C₁ at 2.26 V and C₂ at 1.98 V). Additionally, the cell with PTPEI demonstrates a higher peak current, indicating its active electrochemical kinetics and rapid charge transfer capability.³⁸ Moreover, the cell with PTPEI provides a significant degree of overlap, suggesting superior reversibility of electrochemical reactions when comparing CV profiles of varying cycles for both controls (Fig. S5†). The Tafel slope values derived from the oxidation–reduction peaks indicate that the battery containing PTPEI has the lowest slope value, with Q1, P1, and P2 peaks having slope values of 42.20, 56.04, and 42.79 mV dec⁻¹, respectively. These slope values are lower than those of the cell with TFMA (68.73, 52.97, and 54.45 mV dec⁻¹) and the cell with blank (75.41, 64.41, and 77.01 mV dec⁻¹). This indicates that PTPEI is more effective in catalyzing the conversion of LiPSs and the charge transfer of oxidation products (Fig. S6†). The influence of the amounts of PTPEI on the initial charge–discharge performance of the Li–S cell was investigated at 0.1C (Fig. S7†). The results demonstrate that including the 0.01 mol L⁻¹ additive in the cell enhances charge–discharge specific capacity and mitigates polarization, establishing this concentration as the preferred choice for all subsequent Li–S cells.

Fig. 1 Modulation of sulfur redox kinetics. (a) CV curves at a scanning rate of 0.1 mV s⁻¹. (b) Initial galvanostatic charge–discharge test curves at 0.1C. (c) Typical GITT response curves (charge and discharge). (d) Rate performance at various cycling rates from 0.1 to 2C. (e) Cycling performance at 1C for 200 cycles. (f) CV curves at different scanning rates from 0.1 to 0.9 mV s⁻¹. (g) Plot of Li⁺ ion diffusion kinetics for a cell with PTPEI. (h) Cycling performance of a cell with PTPEI at an E/S = 8 μL mg⁻¹, 5.5 mg cm⁻², and 0.1C.

The impact of PTPEI on the capacity and cycle performance of Li–S cells was further investigated. The initial galvanostatic charge–discharge curves of the cells are depicted in Fig. 1b. The cell with PTPEI exhibits not only a higher initial discharge specific capacity of 1214 mA h g⁻¹ at 0.1C but also a smaller polarization of 140 mA, while the cells with TFMA and blank exhibit initial discharge specific capacities of 1089 and 1027 mA h g⁻¹, respectively. After 100 cycles, the capacity of the cell with PTPEI remains at 80.81% with a corresponding capacity of 981 mA h g⁻¹, significantly surpassing that of the control cells (Fig. S8†). Additionally, the GITT test has been employed to investigate the polarization potential of the electrochemical reaction. The cell with PTPEI, as shown in Fig. 1c, exhibits smaller polarization and higher capacity throughout the cycling process when compared to the control cells. Additionally, the cell with PTPEI demonstrates significantly improved rate performance (Fig. 1d). Specifically, at current densities of 0.1, 0.2, 0.5, 1, and 2C respectively, the cell with PTPEI delivers discharge specific capacities of 1207, 957, 890, 786, and 669 mA h g⁻¹. Then, when back to a current density of 0.1C, the capacity can be restored to 973 mA h g⁻¹. The cell with PTPEI demonstrates a discharge capacity of 580 mA h g⁻¹ (49.2% of the initial capacity) even after 200 cycles at a rate of 1C, as shown in Fig. 1e. In contrast, both the cell with blank and the cell with TFMA exhibit inferior rate performance characterized by significant capacity fluctuation with increasing current densities and a rapid attenuation trend during cycling at a rate of 1C. The enhancement of electrochemical performance signifies that the introduction of PTPEI can effectively alleviate the sluggish sulfur redox kinetics.

The EIS measurement has been employed to explore the impedance change in the Li–S cell upon the introduction of PTPEI. The EIS spectra (Fig. S9a†) show that the cell with PTPEI exhibits much lower resistance than the cell with TFMA and the cell with blank. After 100 cycles at 0.1C (Fig. S9b†), the impedance of all cells exhibited a decreasing trend with varying degrees. According to the data obtained from equivalent circuit fitting, the impedance decreases in cells with PTPEI and TFMA can be attributed to the additives accelerating electron/ion transfer kinetics (Table S1†).^34,36 This enhancement facilitates the formation of a highly conductive SEI on the surface of a Li-anode while mitigating the accumulation of dead lithium, thereby reducing charge transfer resistance, and significantly improving overall electrochemical performance.

Fig. 1f and S10† depict the CV measurements over a scan rate range from 0.1 to 0.9 mV s⁻¹ for the cells with PTPEI, TFMA, and blank, respectively. The cell with PTPEI demonstrates the current of the redox peaks at different scan rates, exhibiting a clearer and stronger response even at a high scan rate of 0.9 mV s⁻¹. The CV results at various scan rates demonstrate the remarkable electrochemical stability of the cell with PTPEI, which significantly contributes to its exceptional rate performance. As shown in Fig. 1g, the redox peak currents in the cell with PTPEI are linearly proportional to the square root of scan rates v^1/2. The slopes obtained from peaks A, C₁, and C₂ of the cell with PTPEI are significantly higher compared to those observed in the cell with TFMA (Fig. S11a†) and blank (Fig. S11b†), indicating an enhanced Li⁺ diffusion rate and improved redox kinetics through the introduction of PTPEI.

The Li⁺ diffusion coefficients (D_Li⁺) can be determined by analyzing the linear relationship between the peak current and the square root of the voltage sweeping rate obtained from CV tests conducted in one cell at various scan rates, using the Randles–Sevcik equation:

Ip = (2.69 × 105)n3/2ADLi+1/2ν1/2CLi+ (1)

where I_p represents the peak current, n denotes the number of electrons transferred during the reaction, A represents the electrode area, v indicates the scan rate, and C_Li⁺ refers to the lithium-ion molar concentration.^39–42 The D_Li⁺ values for A, C₁, and C₂ of the cell with PTPEI are 3.01 × 10⁻⁷, 2.45 × 10⁻⁷, and 2.75 × 10⁻⁷ cm² s⁻¹, respectively (Table S2†). These values demonstrate improvements of 1.16, 1.63, and 1.35 times when compared to those of the cell with TFMA, as well as enhancements of 1.63, 2.31, and 2.00 times when compared to those of the cell with blank. The D_Li⁺ serves as a crucial parameter for assessing kinetic behaviors, and the aforementioned numerical variations caused by PTPEI provide additional evidence to elucidate the enhanced electrochemical properties of Li–S cells resulting from the introduction of PTPEI in electrolyte. In addition, the nucleation kinetics experiment of Li₂S was also carried out to verify the catalytic ability of TFMA and PTPEI, as shown in Fig. S12.† The potentiostatic discharge curve of Li₂S deposition from LiPSs at 2.05 V was collected. The cell with PTPEI has higher dissolution current, lower dissolution time and deposition capacity of Li₂S (231.87 mA h g⁻¹), indicating that PTPEI has better conversion kinetics for Li₂S. The deposition capacity without any additive was measured to be 159.39 mA h g⁻¹, whereas with the addition of TFMA, it significantly increases to 220.63 mA h g⁻¹. The performance of the cell with PTPEI has been further assessed under the conditions including high sulfur loading (5.5 mg cm⁻²) and a lean electrolyte (E/S ratio is 8 μL mg⁻¹). Benefiting from enhanced D_Li⁺ and electron transfer, the cell with PTPEI delivers a remarkable discharge specific capacity of 1190.9 mA h g⁻¹ and sustains a capacity retention of 645.9 mA h g⁻¹ after 50 cycles at 0.1C, demonstrating a capacity decay rate of merely 0.9% per cycle (Fig. 1h). In contrast, the cell with blank exhibits a limited cycling capability of only 7 cycles and experiences rapid decay in its specific capacity (from 1022.4 to 799 mA h g⁻¹). The cell with TFMA gradually diminishes in specific capacity before eventually shorting out after 30 cycles (Fig. S13†). Compared with the reported studies on other electrolyte additives, this work has advantages in both capacity and stability at high sulfur loading, as shown in Table S3,† which indicates that PTPEI has potential in the development of high-performance Li–S cells. PTPEI exhibited enhanced electrochemical stability, sulfur redox kinetics, D_Li⁺ and electron transfer during the process of electrochemical experimental analysis.

The DFT calculations have been conducted to explore the mechanism of PTPEI as a molecule catalyst in regulating the sulfur redox kinetics during the cycling process of Li–S batteries.⁴³Fig. 2a depicts the energy levels of the HOMO and lowest unoccupied molecular orbital (LUMO) of TFMA–Li₂S_x and PTPEI–Li₂S_x (x = 1, 2, 4, 6, and 8) complexes. The values for the energy difference (E_gap) between the HOMO and LUMO for all compounds fall within the range of 5.63–7.21 eV (Table S4†). Compared with the energy levels of Li₂S_x obtained in our previous study,⁴⁴ the E_gap values of PTPEI–Li₂S_x are consistently lower than those of pristine Li₂S_x. Notably, the E_gap values for TFMA–Li₂S₂ and TFMA–Li₂S₆ are marginally higher than those of Li₂S₂ and Li₂S₆, respectively. Consequently, in the previous CV tests, TFMA did not exhibit a significant contribution to the reduction of LiPSs. The introduction of PTPEI modifies these values by enhancing the HOMO energy levels in PTPEI–Li₂S_x compounds. The observed changes in the HOMO energy levels for PTPEI–Li₂S_x suggest an increased propensity for these compounds to act as electron acceptors, thereby indicating their potential as a molecular catalyst. Simultaneously, it is evident that the HOMO energy levels predominantly reside on the Li₂S_x units in PTPEI–Li₂S_x, implying that oxidation primarily occurs at the Li₂S_x units. The introduction of additives can expedite the sulfur redox kinetics of Li–S batteries.

Fig. 2 Theoretical evidence of van der Waals interaction for PTPEI–Li₂S_x and TFMA–Li₂S_x (x = 1, 2, 4, 6, and 8). (a) Schematic energy levels of the HOMO and LUMO. (b) Low-gradient isosurfaces (RDG) (s = 0.5 a.u.).

In comparison with TFMA–Li₂S_x, the values of PTPEI–Li₂S_x exhibit a more pronounced decrease, indicating a stronger enhancement of the Li₂S_x reaction within PTPEI. The low-gradient isosurfaces (RDG) (s = 0.5 a.u.) for these compounds are depicted in Fig. 2b to further investigate the underlying factors contributing to this phenomenon. The noncovalent interaction regions of all compounds, depicted in shades of green or light brown in Fig. 2b, unequivocally validate the predominant presence of van der Waals interactions among the molecules within these compounds. This observed interaction relationship substantiates the ability of TFMA and PTPEI to enhance electron transfer in Li₂S_x. Meanwhile, incorporating phenylacetylene into TFMA molecules alters their molecular structure. The V-shaped molecular structure enhances van der Waals interactions between PTPEI and Li₂S_x. The nitrogen atoms in Schiff bases exhibit a strong electron-donating ability owing to their lone pairs of electrons, thereby increasing the electron transfer capability of PTPEI to Li₂S_x.⁴⁵ According to theoretical calculations, PTPEI can enhance its impact on the sulfur reaction kinetics by modulating the molecular structure.

The deterioration of the Li-anode presents another significant challenge to the stable cycling performance of Li–S batteries.⁴⁶ We disassembled the cell after 100 cycles at 0.1C and employed SEM to investigate the surface of the Li-anode, with the objective of assessing the protective efficacy of PTPEI on the Li-anode. The surface of the Li-anode in the cell with PTPEI (Fig. 3a) exhibits a remarkably smooth morphology without any discernible signs of corrosion, while the Li-anode in the cell with TFMA (Fig. 3b) maintains its intact morphology, indicating only minor corrosion. In contrast, SEM analysis reveals prominent “mossy” features on the surface of the Li-anode in the cell with blank (Fig. 3c), suggesting a relatively severe corrosion. To comprehensively observe changes in the Li-anode, SEM has been utilized to measure the cross-section of the Li-anode. A compact SEI forms on the surface of the anode in the cell with PTPEI (Fig. 3d), while no signs of corrosion are observed within the Li-anode. The cell with TFMA (Fig. 3e) shows corrosion within the Li-anode; however, it is primarily localized on the surface of the Li-anode. The Li-anode from the cell with blank exhibits irregular erosion and internal corrosion, as depicted in Fig. 3f. The introduction of PTPEI facilitates the formation of a stable SEI on the surface of the Li-anode, thereby effectively protecting the Li-anode and enhancing the electrochemical performance of Li–S batteries.

Fig. 3 Formation of a stable SEI for Li anodes. SEM images for PTPEI, TFMA and blank after 100 cycles at 0.1C. (a–c) Surface. (d–f) Cross-section. (g–l) EDS elemental mappings.

The EDS mapping of the corresponding cross-sections for the Li-anode is presented in Fig. 3h–m, aiming to investigate the elemental distribution on the surface of the Li-anode. Notably, in the cell with PTPEI, a weak signal of sulfur is detected and there is virtually no discernible presence of sulfur deep within the Li-anode. However, in the cell with TFMA, sulfur is found to be predominantly concentrated on the surface. The internal sulfur signal is relatively weak, and the detection areas for sulfur and fluorine are largely consistent. This suggests that TFMA can provide a certain degree of protection against LiPS induced erosion of Li-anodes, although its efficacy is not as pronounced as that of PTPEI. Conversely, the presence of sulfur within the Li-anode is observed to exhibit a stronger and more pronounced signal in the cell with blank. Meanwhile, in the cell with PTPEI, fluorine is uniformly dispersed on the Li-anode surface with a significantly stronger and denser signal compared to that observed in the cell with TFMA. In contrast, the fluorine signal in the cell with blank exhibits weak intensity and uneven dispersion. The results demonstrate that PTPEI and TMFA are capable of forming a fluorinated SEI on the surface of the Li-anode. The fluorinated SEI constructed by using PTPEI exhibits higher density, effectively retarding the corrosion of LiPSs on the Li-anode and significantly enhancing its stability.

The aforementioned Li-anodes have undergone XPS depth profiling to facilitate further analysis of the components on the surface of Li-anodes. The S 2p spectra are presented in Fig. S14a–c,† where peaks corresponding to short-chain sulfides such as Li₂S₂ and Li₂S can be observed before etching, along with the peaks of thiosulfate that inevitably arise during sample transfer. After etching, the characteristic peak area of thiosulfate is notably reduced, revealing a distinct peak corresponding to Li₂S₂ and Li₂S. As the etching time extends, these characteristic peaks gradually diminish in size. The peak area of the cell with PTPEI is comparatively smaller compared to others. The irreversible deposition of Li₂S₂ and Li₂S on the surface of the Li-anode is mitigated in the cell with PTPEI.

The Li 1s spectra shown in Fig. 4a–c exhibit peaks at 55.5, 54.8, and 53.9 eV respectively, which can be attributed to the presence of chemical bonds involving Li–F, Li–O, and Li–N.^47,48 After 60 s etching, the peak area of Li–O exhibits a significant decrease, but the ones of Li–F and Li–N demonstrate a noticeable increase, indicating the elimination of air-induced effects on the Li-anode. With prolonged etching time, although the peak intensity shows varying degrees of reduction, it is found that the proportion of peak area for Li–F from the cell with PTPEI (Fig. 4a) is significantly larger and substantially greater than that observed in the other two cells. This result provides evidence of LiF as the primary constituent of the fluorinated SEI in the cell with PTPEI.

Fig. 4 LiF-enriched SEI on a Li-anode. XPS spectra with different depth etching for Li 1s of (a) PTPEI, (b) TFMA, and (c) blank and F 1s of (d) PTPEI, (e) TFMA, and (f) blank.

For the F 1s spectra (Fig. 4d–f), the presence of two characteristic peaks corresponding to C–F (688 eV) and LiF (684.9 eV) were observed in all three samples. The peak of LiF exhibits significant intensity both before and after etching in the cell with PTPEI (Fig. 4d), once again demonstrating LiF as the primary component on the surface of the Li-anode. The presence of C–F can be attributed to both the primary decomposition products of LiTFSI and the contribution of the C–F bond in PTPEI and TFMA molecules. The intensity of C–F in the cell with PTPEI is comparable to that observed in the cell with blank (Fig. 4f). In contrast, there is an enhanced intensity of the C–F peak in the cell with TFMA (Fig. 4e). The peak of C–F for PTPEI and blank is almost eliminated after 60 s of etching, while it remains distinctly observable even after 180 s of etching in the cell with TFMA. The observed phenomenon can be attributed to the fact that PTPEI, under the influence of phenylacetylene, exhibits a higher propensity for C–F bond cleavage and transformation into the LiF solid in comparison to TFMA during the formation of the fluorinated SEI.

The composition of the SEI has been further analyzed by NMR spectroscopy. In the ¹H-NMR spectra for the SEI from the cell with TFMA (Fig. S15†), due to the minimal content of additives employed, there is significant interference from deuterated chloroform and DOL/DME. Nonetheless, when magnified, the spectroscopy reveals the presence of TFMA. In contrast, in the SEI from the cell with PTPEI (Fig. S16†), no detectable presence of PTPEI can be identified, but another compound can be identified as p-xylene resulting from PTPEI decomposition. Interestingly, the presence of PTPEI can still be detected in the electrolyte from the cell with PTPEI after cycling, as illustrated in Fig. S17.† The SEI of the cell with PTPEI was examined by HRTEM, revealing the existence of lattice fringes for LiF (Fig. S18†), which further support the deposition of LiF as a SEI on the Li-anode. These findings suggest the partial decomposition of PTPEI during cycling, leading to the formation of a LiF-enriched SEI.

The Li‖Li symmetric cells were assembled and subjected to electrochemical performance tests to further investigate the role of PTPEI in protecting the Li-anode. At a current density of 1 mA cm⁻² (Fig. 5a), both the cells with PTPEI and the cell with TFMA exhibit stable overpotential after cycling for 190 h and 100 h, respectively, maintaining this stability for more than 1000 h. In contrast, the cell with blank experienced a short circuit after cycling for 130 h. The uneven deposition of lithium in the cell with blank leads to severe growth of lithium dendrites, eventually puncturing the separator and causing a short circuit in the cell. Fig. S19† illustrates part 1 of Fig. 5a, where the symmetric cell with PTPEI exhibits a significantly lower overpotential (50 mV) compared to the other two cells (230 mV for the cell with blank and 130 mV for the cell with TFMA), indicating excellent kinetic properties for the cell with PTPEI. The voltage hysteresis of both the symmetric cells with TFMA and PTPEI exhibits a consistent increase with increasing current density, as shown in Fig. 5b. In contrast, the cell with blank displays an unstable voltage profile with fluctuating voltage hysteresis. The PTPEI facilitates the homogeneous deposition of lithium, thereby enhancing the stability of lithium even under conditions of high current density.

Fig. 5 Uniform lithium deposition on a Li-anode for a Li‖Li symmetric cell. (a) Voltage profiles of the different symmetric cells at 1 mA cm⁻². (b) Long-term stability of the different symmetric cells at different current densities. SEM images of the 20 h cycled Li-metal, (c) the cell with PTPEI, (d) the cell with TFMA, and (e) the cell with blank. Li 1s XPS spectra with different depth etching on the surface of the Li-anode for Li‖Li symmetric cells, (f) with PTPEI, (g) with TFMA, and (h) with blank.

Moreover, in order to investigate the impact of PTPEI on lithium deposition on the surface of Li-metal, we compared the SEM images (Fig. 5c–e) of Li-metal obtained from the symmetric cell after cycling for 20 h, including the cells with PTPEI, TFMA, and blank. In comparison to the phenomenon of uneven deposition on the surface of Li-metal in the cell with blank (Fig. 5e), the surface is flat and the deposits are uniform in the cell with TFMA (Fig. 5d) and the cell with PTPEI (Fig. 5c).^37,38 This phenomenon can be directly observed through the corresponding cross-section SEM images of Li-metal presented in Fig. S20a–c.† The EDS mapping images (Fig. S20d and e†) reveal a substantial presence of fluorine on the surface of Li-metal from the cell with PTPEI. XPS measurement enables a more accurate characterization of the elemental composition on the surface of Li-metal. Regardless of before-etching or after-etching, the relative abundance of Li–F in the Li 1s spectrum is significantly higher in the cell with PTPEI (Fig. 5f) compared to the other two cells. The incorporation of PTPEI into the electrolyte facilitates the formation of a stable SEI mainly consisting of LiF, which significantly promotes the uniform deposition of lithium.

The experimental results demonstrate that the incorporation of PTPEI into the electrolyte of a Li–S cell can effectively enhance the reaction kinetics of the Li–S cell to some extent and facilitate the formation of a stable SEI with abundant LiF. A mechanism of TFMA and PTPEI in the Li–S cell during cycling is proposed based on the presented scheme (Fig. 6a). PTPEI can be likened to a “multifunctional robot” equipped with advanced tools. Phenylacetylene modifies the molecular spatial structure of TFMA, thereby enhancing intermolecular interactions with Li₂S_x and further improving sulfur redox kinetics in Li–S cells while effectively mitigating shuttle effects. It is noteworthy that, due to changes in the molecular structure, a fraction of PTPEI undergoes decomposition on the surface of the Li-anode. This leads to the release of fluoride ions and consequent formation of a dense and stable fluorinated SEI during the cycling process. The enhancement of sulfur redox kinetics and the Li-anode protection provided by PTPEI are pivotal for augmenting the electrochemical performance in Li–S cells.

Fig. 6 Schematic illustration of the enhancement mechanism of PTPEI and TFMA on redox kinetics and SEI formation. (a) Cartoon diagram. (b) Possible reaction paths.

In contrast, TFMA, as an electrolyte additive, can be likened to a “multifunctional robot” that effectively enhances the reaction kinetics of the Li–S cell and mitigates the shuttle effect of polysulfide. Simultaneously, TFMA in the electrolyte actively participates in and promotes the formation of the fluorinated SEI on the surface of the Li-anode. However, the fluorinated SEI lacks sufficient density. As the cycling progresses, the SEI is susceptible to destruction by lithium dendrites, while the Li-anode is still affected by polysulfide. The protective capability of TFMA for the Li-anode is limited. The mechanism underlying the interaction of PTPEI with LiPSs and the construction of the SEI on the lithium anode is depicted in Fig. 6b, where the Schiff base molecule PTPEI is capable of facilitating the conversion of LiPSs by offering electrons via the nitrogen atom.

4. Conclusions

The present study introduces PTPEI, a fluorinated Schiff base molecule obtained by modifying TFMA through phenylacetylene, as a novel dual-function electrolyte additive for Li–S batteries to enhance the sulfur redox kinetics and establish a stable fluorinated SEI for protecting the Li-anode. PTPEI not only enhances sulfur reaction kinetics by strengthening the van der Waals force between PTPEI and polysulfide but also facilitates the release of fluorine ions from PTPEI to construct a dense fluorinated SEI during cycling. Consequently, the PTPEI molecule effectively mitigates the shuttle effect of LiPSs, improves the utilization rate of active sulfur and safeguards the integrity of the Li-anode. As a demonstration, the Li–S cell with PTPEI exhibits remarkable cycling stability for 50 cycles and maintains a high capacity of 645.9 mA h g⁻¹, even when operated with high sulfur loading (5.5 mg cm⁻²) and lean electrolyte (E/S ratio is 8 μL mg⁻¹). We anticipate that regulating the imine Schiff base molecular structure will provide novel insights for developing efficient multifunctional electrolyte additives for molecular catalysts and robust SEIs in Li–S batteries.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article [and/or its ESI†].

Author contributions

Tong Wu: methodology, writing – original draft, writing – review and editing. Mingxun Jia: investigation, data curation, formal analysis, methodology. Ye Lu: formal analysis. Jinting Ye: data curation. Daotong Yang: formal analysis. Yingying Zhang: formal analysis. Shuyuan Xie: data curation. Dawei Kang: data curation. Limei Duan: data curation. Haiming Xie: methodology, writing – review and editing. Jinghai Liu: supervision, conceptualization, methodology, writing – original draft, writing – review and editing. All authors have given approval to the final version of the manuscript.

Conflicts of interest

The authors declare no competing interests.

Acknowledgements

This work was financially supported by the Doctoral Scientific Research Foundation of Inner Mongolia University for Nationalities (Project No. BS614), the National Natural Science Foundation of China (No. 22150710516, 21961024, 21960125, 22361034, 22361035, 22365024, 22242003, and 22461030), the Performance Project of Introducing Outstanding Talents and Initiating Scientific Research in Public Institutions at the Autonomous Region Level (No. RCQD20002), Basic Research Funds for University at the Inner Mongolia (No. GXKY22087 and NJZY21433), the Inner Mongolia Grassland Talents-Young Leading Talent Program (No. KYCYYC23001 and KYCYYC24001), the Program for Innovative Research Team in Universities of Inner Mongolia Autonomous Region (No. NMGIRT2417), and fund for supporting the Reform and Development of Local Universities (Discipline Construction).

References

1. P. K. Zou, Y. S. Lin, L. Li, J. X. Wang, Y. Chao, B. R. Li, H. Y. Ma, Z. Y. Liu, Y. Yu and C. K. Yang, J. Mater. Chem. A, 2024, 12, 26707–26717,  10.1039/D4TA06086E.
2. Z. W. Seh, Y. Sun, Q. Zhang and Y. Cui, Chem. Soc. Rev., 2016, 45, 5605–5634,  10.1039/C5CS00410A.
3. J. J. Zhao, Z. X. Chen, Q. Cheng, M. Zhao, X. Z. Ma, X. Q. Zhang, J. Q. Huang and B. Q. Li, J. Mater. Chem. A, 2024, 12, 21845–21852,  10.1039/D4TA01997K.
4. Y. Fei and G. Li, Adv. Funct. Mater., 2024, 2312550,  DOI:10.1002/adfm.202312550.
5. X. Han and X. F. Xu, J. Mater. Chem. A, 2023, 11, 18922–18932,  10.1039/d3ta03366j.
6. K. Yang, C. Li, H. Y. Qi, Y. F. Dai, Y. H. Cui and Y. B. He, J. Mater. Chem. A, 2023, 11, 10425–10434,  10.1039/D3TA01082A.
7. Z. Wang, H. Ge, S. Liu, G. Li and X. Gao, Energy Environ. Mater., 2023, 6(3), e12358,  DOI:10.1002/eem2.12358.
8. G. W. Gao, X. C. Yang, J. X. Bi, W. Q. Guan, Z. Z. Du and W. Ai, J. Mater. Chem. A, 2023, 11, 26318–26339,  10.1039/D3TA06057H.
9. R. X. Chen, Y. C. Zhou and X. D. Li, Energy Storage Mater., 2024, 68, 103353,  DOI:10.1016/j.ensm.2024.103353.
10. H. Yuan, J. Yang and Y. W. Zhang, J. Mater. Chem. A, 2023, 11, 17841–17847,  10.1039/D3TA03518B.
11. J. F. Wu, Y. Y. Feng, Y. Chen, T. Fan and Y. W. Li, J. Mater. Chem. A, 2023, 11, 12025–12033,  10.1039/D3TA02347H.
12. R. Y. Deng, H. Y. Yu, J. M. Liu, F. L. Chu, J. Lei, L. Z. Yang and F. X. Wu, J. Mater. Chem. A, 2023, 11, 15769–15777,  10.1039/D3TA02327C.
13. Z. Li, I. Sami, J. Yang, J. Li, R. V. Kumar and M. Chhowalla, Nat. Energy, 2023, 8, 84–93,  DOI:10.1038/s41560-022-01175-7.
14. X. Zhang, X. Y. Li, Y. Z. Zhang, X. Li, Q. H. Guan, W. Jian, Z. C. Zhuang, Q. Zhuang, X. M. Cheng, H. T. Liu, J. Zhang, C. Y. Shen, H. Z. Lin, Y. L. Wang, L. Zhan and L. C. Ling, Adv. Funct. Mater., 2023, 33, 2302624,  DOI:10.1002/adfm.202302624.
15. C. C. Wu, Y. C. Ho and S. H. Chung, J. Mater. Chem. A, 2023, 11, 24651–24660,  10.1039/D3TA05632E.
16. H. Li, H. Yang and X. Ai, Adv. Mater., 2023, 2305038,  DOI:10.1002/adma.202305038.
17. J. Qin, R. Wang, P. Xiao and D. Wang, Adv. Energy Mater., 2023, 13, 2300611,  DOI:10.1002/aenm.202300611.
18. T. Wang, J. He, Z. Zhu, X. B. Cheng, J. Zhu, B. Lu and Y. Wu, Adv. Mater., 2023, 2303520,  DOI:10.1002/adma.202303520.
19. J. Li, L. Gao, F. Pan, C. Gong, L. Sun, H. Gao, J. Zhang, Y. Zhao, G. Wang and H. Liu, Nano-Micro Lett., 2024, 16, 12,  DOI:10.1007/s40820-023-01223-1.
20. J. Wang, G. Li, X. Zhang, K. Zong, Y. Yang, X. Zhang, X. Wang and Z. Chen, Adv. Mater., 2024, 36, 2311019,  DOI:10.1002/adma.202311019.
21. C. Zhao, Y. Huang, B. Jiang, Z. Chen, X. Yu, X. Sun, H. Zhou, Y. Zhang and N. Zhang, Adv. Energy Mater., 2024, 14, 2302586,  DOI:10.1002/aenm.202302586.
22. D. Yang, T. Wu, H. Gao, M. Jia, L. Ji, J. Wang, Q. Zhuang, B. Yu, L. Lu, Y. Zhang and J. Liu, Mater. Today Sustainability, 2023, 24, 100573,  DOI:10.1016/j.mtsust.2023.100573.
23. W. Zhang, F. Ma, Q. Wu, Z. Cai, W. Zhong, Z. Zeng, S. Cheng and J. Xie, ACS Appl. Energy Mater., 2022, 5, 15719–15728,  DOI:10.1021/acsaem.2c03306.
24. Z. Ye, S. Zhai, R. Liu, M. Liu, Y. Xu, C. Li, X. Wang and T. Mei, Chem. Eng. J., 2024, 479, 147847,  DOI:10.1016/j.cej.2023.147847.
25. L. P. Hou, Y. Li, Z. Li, Q. K. Zhang, B. Q. Li, C. X. Bi, Z. X. Chen, L. L. Su, J. Q. Huang, R. Wen, X. Q. Zhang and Q. Zhang, Angew. Chem., Int. Ed., 2023, 62, e202305466,  DOI:10.1002/anie.202305466.
26. Z. P. Jiang, Z. Q. Zeng, W. Hu, Z. L. Han, S. Cheng and J. Xie, Energy Storage Mater., 2021, 36, 333–340,  DOI:10.1016/j.ensm.2021.01.008.
27. Y. W. Song, L. Shen, N. Yao, S. Feng, Q. Cheng, M. Jin, X. Chen, B. Q. Li and Q. Zhang, Angew. Chem., Int. Ed., 2024, 63, e202400343,  DOI:10.1002/anie.202406336.
28. S. Bag, C. Zhou, P. J. Kim, V. G. Pol and V. Thangadurai, Energy Storage Mater., 2020, 24, 198–207,  DOI:10.1016/j.ensm.2019.08.019.
29. K. Nie, Q. Fu, R. Gao, K. Wang, H. Wang, C. Teng, X. Wang, J. Ren and R. Wang, Energy Storage Mater., 2023, 63, 103011,  DOI:10.1016/j.ensm.2023.103011.
30. F. Wu, S. Thieme, A. Ramanujapuram, E. Zhao, C. Weller, H. Althues, S. Kaskel, O. Borodin and G. Yushin, Nano Energy, 2017, 40, 170–179,  DOI:10.1016/j.nanoen.2017.08.012.
31. M. He, R. Guo, G. M. Hobold, H. Gao and B. M. Gallant, Proc. Natl. Acad. Sci. U. S. A., 2020, 117, 73–79,  DOI:10.1073/pnas.1911017116.
32. Z. Han, S. Li, M. Sun, R. He, W. Zhong, C. Yu, S. J. Cheng and J. Xie, J. Energy Chem., 2022, 68, 752–761,  DOI:10.1016/j.jechem.2021.12.038.
33. W. Guo, W. Zhang, Y. Si, D. Wang, Y. Fu and A. Manthiram, Nat. Commun., 2021, 12, 3031,  DOI:10.1038/s41467-021-23155-3.
34. W. Zhang, F. Ma, Q. Wu, Z. Zeng, W. Zhong, S. Cheng, X. Chen and J. Xie, Energy Storage Mater., 2023, 6, e12369,  DOI:10.1002/eem2.12369.
35. Z. Li, Y. Li, C. X. Bi, Q. K. Zhang, L. P. Hou, X. Y. Li, J. Ma, X. Q. Zhang, B. Q. Li, R. Wen and Q. Zhang, Adv. Funct. Mater., 2024, 34, 2304541,  DOI:10.1002/adfm.202304541.
36. J. Lian, W. Guo and Y. Fu, J. Am. Chem. Soc., 2021, 143, 11063–11071,  DOI:10.1021/jacs.1c04222.
37. Y. Lu, Y. Wang, H. Li, P. Li, X. Feng, Y. Yamamoto, M. Bao and J. Liu, RSC Adv., 2023, 13, 3371–3376,  10.1039/D2RA07120G.
38. Y. Liu, K. Xiao, S. Yang, J. Sun, S. Li, X. Liu, D. Cai, Y. Zhang, H. Nie and Z. Yang, Small, 2024, 2309890,  DOI:10.1002/smll.202309890.
39. R. Li, Z. Bai, W. Hou, Z. Wu, P. Feng, Y. Bai, K. Sun and Z. Wang, Chin. Chem. Lett., 2023, 34, 108263,  DOI:10.1016/j.cclet.2023.108263.
40. H. Wang, X. Lai, C. Chen, P. Shi, H. Wan, H. Wang, X. Chen and D. Su, Chin. Chem. Lett., 2024, 35, 108473,  DOI:10.1016/j.cclet.2023.108473.
41. Y. Liu, F. Wu, Z. Hu, F. Zhang, K. Wang, L. Li and R. J. Chen, Angew. Chem., Int. Ed., 2024, e202402624,  DOI:10.1002/anie.202402624.
42. A. Huang, L. Kong, B. Zhang, X. Liu, L. Wang, L. Li and J. Xu, ACS Nano, 2024, 18, 12795–12807,  DOI:10.1021/acsnano.3c12279.
43. R. Meng, X. He, S. J. H Ong, C. Cui, S. Song, P. Paoprasert, Q. Pang, Z. J. Xu and X. Liang, Angew. Chem., Int. Ed., 2023, 135, e202309046,  DOI:10.1002/anie.202309046.
44. T. Wu, J. Ye, T. Li, Y. Liu, L. Jia, L. Sun, J. Liu and H. Xie, J. Power Sources, 2022, 536, 231482,  DOI:10.1016/j.jpowsour.2022.231482.
45. J. Zhou, T. Wu, Y. Pan, J. Zhu, X. Chen, C. Peng, C. Shu, L. Kong, W. Tang and S. L. Chou, Adv. Funct. Mater., 2022, 32, 2106966,  DOI:10.1002/adfm.202106966.
46. X. B. Cheng, C. Yan, J. Q. Huang, P. Li, L. Zhu, L. Zhao, Y. Zhang, W. Zhu, S. T. Yang and Q. Zhang, Energy Storage Mater., 2017, 6, 18–25,  DOI:10.1016/j.ensm.2016.09.003.
47. X. Ren, X. Zhang, Z. Shadike, L. Zou, H. Jia, X. Cao, M. H. Engelhard, B. E. Matthews, C. Wang, B. W. Arey, X. Q. Yang, L. Jun, J. G. Zhang and W. Xu, Adv. Mater., 2020, 32, 2004898,  DOI:10.1002/adma.202004898.
48. Z. Huang, J. Meng, M. Xie, Y. Shen and Y. Huang, J. Mater. Chem. A, 2020, 8, 14198–14204,  10.1039/D0TA05147K.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta08542f

‡ Tong Wu and Mingxun Jia contributed equally.

---