Stabilizing electrode–electrolyte interphases using soluble metal–organic polyhedra for high-performance lithium metal batteries

Abstract

Lithium metal batteries (LMBs) are promising next-generation technology for achieving energy densities above 400 Wh kg⁻¹. Pairing lithium metal anodes with layered LiCoO₂ (LCO) offers high volumetric energy density for “3C” devices. However, LCO becomes unstable above 4.3 V, causing cobalt dissolution, lattice oxygen loss, and rapid capacity decay. Meanwhile, lithium metal suffers from dendrite growth and electrolyte depletion, compromising safety and cycle life. Herein, a novel class of metal–organic polyhedra (MOP)-based electrolytes is designed and synthesized to tackle the above issues. On the cathode side, the NAMI-designed MOP (NAMI-MOP) decomposes on the LCO surface, enabling in situ Cu doping on the LCO particles to protect the LCO lattice. On the anode side, it modulates anion reactivity towards lithium metal and forms a stable, LiF and lithium (poly)sulfide-rich solid-electrolyte interphase (SEI), suppressing dendrites and enhancing anode stability. Unlike conventional MOP with poor solubility, the NAMI-MOP dissolves well in both carbonate and ether-based electrolytes, ensuring uniform performance across various cell formats. As a result, the cycle life of cells improved from 900 cycles to 2000 cycles with 80% capacity retention at 2C/4C. Pouch cells with NAMI-MOP achieved high energy densities, 362 Wh kg⁻¹ (6.4 Ah Li‖LCO) and 412 Wh kg⁻¹ (5.7 Ah Li‖NMC811), under lean electrolyte conditions (<1.8 g Ah⁻¹), highlighting the promise of MOP as effective, soluble electrolyte additives for practical LMBs.

---

1 Introduction

The fast evolution of consumer electronics and the burgeoning low-altitude economy, encompassing applications such as drones and electric-powered aerial vehicles, have driven an unprecedented demand for advanced energy storage solutions that feature both compact and lightweight while maintaining high battery performance. Lithium metal batteries (LMBs) have emerged as a leading candidate to fulfill these requirements, offering substantially higher energy densities (>400 Wh kg⁻¹) compared to conventional lithium-ion batteries (150–250 Wh kg⁻¹). Their potential to revolutionize portable electronics and power next-generation aerial devices stems from the intrinsic properties of lithium metal, including its low density (0.59 g cm⁻³) and exceptional theoretical capacity (3860 mA h g⁻¹).^1–5 When coupled with high-voltage lithium cobalt oxide (LiCoO₂, LCO) cathodes, LMBs become even more attractive for portable electronics due to LCO's high volumetric energy density (2300–3000 Wh L⁻¹, 4.2–4.45 V vs. Li/Li⁺),^6–10 which is a crucial parameter for space-constrained applications, particularly in computers, communication devices, and consumer electronics, collectively termed “3C” devices.

However, the commercialization of LCO-based LMBs faces significant challenges, including the high reactivity of lithium metal, severe dendrite formation and unstable electrode–electrolyte interfaces. These issues lead to continuous electrolyte depletion, poor cycling stability and safety concerns.¹¹ Additionally, when LCO cathodes are operated at voltages beyond 4.3 V in order to deliver high energy densities, they suffer from accelerated structural degradation mechanisms, including irreversible phase transitions, cobalt dissolution and lattice oxygen loss.¹² The formation of an unstable cathode–electrolyte interphase (CEI) at high voltages further exacerbates side reactions.⁸ These multifaceted challenges have spurred extensive research into innovative materials and engineering strategies to stabilize both lithium metal anodes and high-voltage LCO cathodes.

The electrolyte additive strategy offers a simple and efficient approach to improve the cycling stability of Li metal anodes and prolong the lifespan of LMBs. This strategy is highly compatible with industrial processes of battery fabrication, making it a practical solution for commercialization.¹³ Conventional electrolyte additives such as lithium nitrate (LiNO₃),¹⁴ fluoroethylene carbonate (FEC),^15–17 lithium difluorophosphate (LiDFP),¹⁸ and 1,3-propanesultone (PS),¹⁹ have been reported to improve the interfacial stability of LMBs. These additives are either fully organic or fully inorganic, and thus lack the multifunctionality required to simultaneously address the key challenges in LMBs such as high-temperature cycling stability and compromised electrolyte viscosity.²⁰ Metal–organic frameworks (MOFs) have become promising candidates as electrolyte additives for LMBs to solve the above issues owing to their organic–inorganic hybrid porous structures with versatile functionalities, highly ordered ion channels and abundant anionophilic metal sites. These features enable multiple advantages, including homogenization of lithium ion flux, enhanced ion transport, and scavenging of electrolyte impurities.^21–25 However, their microcrystalline nature, insolubility and limited dispersion in liquid electrolyte systems bring new concerns such as phase separation during long-term storage and doubtful electrolyte uniformity.

In this context, metal–organic polyhedra (MOP) become particularly well-suited for serving as soluble electrolyte additives. Being a family member of metal–organic porous materials, MOP materials possess similar characteristics to MOFs and feature additional advantages such as solution processability and discrete nano-cage structures which can promote homogeneous mixing with liquid electrolytes. By powerful organic functional design and metal cluster chemistry, novel MOP materials can be made and employed as electrolyte additives or monomers for facilitating practical utilization of LCO-based LMBs. To date, only limited studies have investigated the integration of MOP into lithium metal battery systems.^26,27 For example, Liu et al. introduced the design and synthesis of a novel MOP additive for compositing with polyethylene oxide (PEO).²⁶ The resulting polymer electrolyte facilitates lithium ion conduction while simultaneously hindering anion mobility, leading to suppression of interfacial side reactions. Lu and Zhang et al. reported the design and fabrication of a cationic hyper-crosslinking MOP polymer which serves as a quasi-solid electrolyte (QSE) with enhanced mechanical strength and lithium salt dissociation.²⁷ These studies mainly focus on the formulation of polymer electrolytes with MOP nanofillers to immobilize anions of Li salts and improve the ionic conductivity of composite solid electrolytes. The solubility of MOP materials in common electrolyte systems and the effects of MOP additives on electrode–electrolyte interphase formation have not yet been investigated.

Herein, we report a synergetic strategy of stabilizing the CEI and regulating solid-electrolyte interphase (SEI) compositions via rational design and synthesis of a novel Cu(II)-based MOP (denoted as NAMI-MOP). As seen in Fig. 1a, the NAMI-MOP is functionalized by short-chain PEO and terminal allyl functions to enhance its solubility in non-aqueous electrolytes and introduce film-forming properties. Moreover, the coordinatively unsaturated Cu₂(COO)₄ clusters in NAMI-MOP act as Lewis acidic sites to interact with Lewis basic electrolyte components such as anions of lithium salts and regulate the anion reactivity towards Li metal anodes. As a result, the NAMI-MOP features excellent solubility and a favorable HOMO–LUMO structure to regulate the compositions and stabilities of both the CEI and SEI. The battery cycling results from coin cells and single-layer pouches to Ah-level pouches indicate the remarkable improvement of cycling stability by simple addition of NAMI-MOP to ether-based electrolytes.

Fig. 1 (a) A schematic diagram of the self-assembled structure of NAMI-MOP and MOF-17. For clarity, the side chains of the MOP structure are hidden. (b) Small-angle X-ray scattering (SAXS) profile of NAMI-MOP. Inset: packing structure of NAMI-MOP in the bulk solid state. The (c) high-resolution transmission electron microscopy (HRTEM) image of NAMI-MOP. The yellow circles highlight the discrete cage of NAMI-MOP. Scale bar 10 nm. The average cage size of NAMI-MOP was estimated to be 2.2 nm. Photographs of the solubility tests of MOP-17 (d) and NAMI-MOF (e) in 1M LiPF₆ in EC/DMC/DEC (1 : 1 : 1 by volume). Photographs of the solubility tests of MOP-17 (f) and NAMI-MOF (g) in 3 M LiFSI in DME. The mass loading of MOP materials (in d–g) is 2.0%.

2 Experimental

2.1 Materials and methods

Chemicals and solvents were purchased from commercial sources and used as received without further purification. FT-IR spectra were collected using a Nicolet Avatar 360 spectrophotometer. Nuclei magnetic resonance (NMR) spectra (¹H NMR) were recorded using Bruker DPX 400 MHz spectrometers. Elemental analysis was conducted using an ICP-OES (HORIBA JY2000-2) for cycled Li metal anodes. Thermogravimetric analysis (TGA) was performed on a PerkinElmer TGA 8000 analyzer. Ultraviolet and visible (UV-vis) spectroscopy was performed on a UV-2450 spectrophotometer. HRTEM was performed using a JEM-100F microscope. Raman spectra were obtained with an inVia™ confocal Raman microscope (RENISHAW) using a 532 nm laser as an excitation source at room temperature. Electrochemical impedance spectroscopy (EIS) was performed using a scanning electrochemical workstation (AMETEK Scientific Instruments) within a frequency range between 1 × 10⁶ and 0.05 Hz at a sinusoidal perturbation of 10 mV. X-ray photoelectron spectrometry (XPS) was performed on a Thermo Scientific Escalab 250Xi spectrometer to investigate the chemical states of the selected elements. The XPS data were processed using CasaXPS, and the binding energies were calibrated to be adventitious C 1s at 284.8 eV for MOP and 284.6 eV for CEI and SEI analyses. All XPS spectra were normalized for comparison.

2.2 Battery fabrication

To examine the electrochemical performance of the NAMI-MOP as a novel electrolyte additive for Li‖LCO coin-type full cells, CR2032 coin cells were assembled in an argon-filled glove box (H₂O < 0.1 ppm, O₂ < 0.1 ppm) with 60 µL electrolyte, which contains 3 M LiFSI in DME as the baseline electrolyte. The LCO cathode mass loading was 3.0 mg cm⁻². For the Li‖LCO pouch-type full cells, the dimensions of the Li anode and cathode were 2.5 cm × 5.0 cm and 2.0 cm × 4.5 cm, respectively, using 270 µL electrolyte, which contains 3 M LiFSI in 1,3-dimethoxypropane (1,3-DMP) as the baseline, and LCO cathode active mass loading was 12.2 mg cm⁻². For high-temperature (45 °C) cycling, the dimensions of the Li anode and cathode were 2.5 cm × 5.0 cm and 2.0 cm × 4.5 cm, respectively, using 180 µL baseline electrolyte (3 M LiFSI in 1,3-DMP), and LCO cathode active mass loading was 12.2 mg cm⁻². Li metal anodes with a lithium thickness of 50 µm and Celgard 2325 membrane separators were used in the above cell assembly. For Ah-level pouch cell fabrication, the dimensions of the Li anode and cathode were 5.4 cm × 6.4 cm and 5.0 cm × 6.0 cm, respectively, using the baseline electrolyte (3 M LiFSI in 1,3-DMP) at an E/C ratio of 4.76 Ah g⁻¹, and LCO cathode active mass loading was 12.2 mg cm⁻². The thickness of Li metal was 20 µm. The thickness of the PE membrane separator was 20 µm. To fabricate 6.4 Ah Li‖LCO, the same dimensions of electrodes except for high-loading LCO (22 mg cm⁻²) and PE separators (8 µm in PE thickness, one-side coated with 3 µm ceramic coating facing cathodes) were used. To fabricate 5.7 Ah Li‖NMC811 pouch cells, the same dimensions of electrodes except for high-loading NMC811 (20 mg cm⁻²) and PE membrane separators (8 µm in thickness) were used.

3 Results and discussion

A novel linker molecule H₂L and the final NAMI-MOP product equipped with pre-designed functional groups were successfully prepared (see SI for details of the synthesis and chemical analyses). The three-dimensional (3D) microstructures of the NAMI-MOP solid were examined using the small-angle X-ray scattering (SAXS) method. As seen in Fig. 1b, the SAXS profile shows a broad Bragg reflection peak confirming a short-range structural order in 3D microstructures. The measured d-spacing value corresponding to the distance between adjacent NAMI-MOP is 2.84 nm. This packing distance is consistent with the reported value (2.9 nm) for another isostructural Cu(II)-based MOP with similar PEO chain lengths.²⁸ To further estimate the cage size and packing of the NAMI-MOP solid, a high-resolution transmission electron microscopy (HRTEM) image was taken. As seen in Fig. 1c, spherical NAMI-MOP particles with an average particle size of 2.2 nm can be observed. The estimated particle size indicates the discrete cage structure of NAMI-MOP and is comparable to the expected values for a typical Cu(II)-isophthalate MOP core (2–3 nm).²⁹

The investigation of NAMI-MOP as a novel electrolyte additive starts with solubility tests in liquid electrolytes. An isostructural MOP material, MOP-17,³⁰ is selected as the benchmark for comparison due to the same cage structure. MOP-17 is functionalized with allyl groups (Fig. 1a). As seen in Fig. 1d and f, the benchmark MOP solid, MOP-17, shows sedimentation issues in a carbonate-based electrolyte and poor solubility in an ether-based electrolyte, leading to poor uniformity of electrolyte and hindering the further utilization of MOP-based additives in LMBs. In contrast, clear blue solutions were obtained by dissolving NAMI-MOP solids in the same electrolytes (Fig. 1e and g), indicating the remarkably improved solubility. The blue color of the solutions originates from the Cu(II) complex motifs in NAMI-MOP. In short, the poor solubility and dispersibility of MOP in liquid electrolytes are successfully addressed by rational functional design.

The capability of NAMI-MOP to serve as a film-forming additive can be predicted by computational analysis of frontier molecular orbital energies. Fig. S9 shows the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energies of NAMI-MOP, DME, LiFSI and the solvation complex (LiFSI-3DME). The HOMO level of NAMI-MOP is calculated to be −4.70 eV, which is higher than that of DME (−9.05 eV), LiFSI (−10.78 eV) and the solvation complex (−8.70 eV), suggesting that NAMI-MOP is easier to be oxidized on the cathode surface and participates in CEI formation. In addition, NAMI-MOP possesses the lowest LUMO energy (−3.88 eV) compared to DME (0.01 eV), LiFSI (−2.03 eV) and the solvation complex (−0.60 eV), indicating that NAMI-MOP is more prone to gain electrons and capable of forming a SEI film on the anode surface. Moreover, leveraging a well-known chemical effect called neighboring group participation (NGP), the lithiophilic ether oxygen near the terminal allyl group can first coordinate with a lithium ion and helps bring the lithium ion close to the double bond (C=C) of the allyl group, allowing both to bind together and form a bidentate coordination structure (Fig. S10a). In addition, the presence of the lithium ion lowers the electron density of the bond in the allyl group, making it more susceptible to reduction (i.e., accept electrons), which is confirmed by the lower LUMO energy level shown in Fig. S10b. The frontier orbital energy calculation results reveal that NAMI-MOP features film-forming properties, which can contribute to the formation of MOP-derived passivation layers with improved interfacial stability on both electrode surfaces.

Raman spectroscopy was applied to probe the solvation behavior of FSI⁻ anions. As shown in Fig. S11, the Raman signals between 680 cm⁻¹ and 780 cm⁻¹ arise from the S–N–S vibrations in FSI⁻, which have been commonly used to analyze the binding behavior of FSI⁻ in electrolytes.³¹ The S–N–S vibration band of FSI⁻ was deconvolved into three peaks: free FSI⁻ (solvent-separated ion pair, SSIP), contact ion pairs (CIP) and ion aggregates (AGG, formed when FSI⁻ coordinates with two or more Li⁺).^32,33 Compared to the SSIP content (35%) of the baseline electrolyte (Fig. S11b), the decreased SSIP content (30%) of the electrolyte with the NAMI-MOP additive (Fig. S11d) is attributed to the Lewis acidic nature of the Cu-based clusters in NAMI-MOP, which likely interact with the Lewis basic FSI⁻ anions and disrupt the solvent-separated state. To further verify this effect across different solvent systems, the Raman spectra of 1,3-DMP-based electrolytes (3 M LiFSI in 1,3-DMP) with and without the NAMI-MOP additive were also measured. As shown in Fig. S12, a consistent trend was observed, with the SSIP content decreasing from 62% to 47% upon addition of NAMI-MOP. These results confirm that NAMI-MOP exhibits strong anionophilic behavior, interacting with FSI⁻ anions and thereby reducing the proportion of SSIP species. This interaction likely promotes the preferential reduction of FSI⁻ at the lithium metal surface, contributing to the formation of a more stable SEI.

To examine the applicability of NAMI-MOP as an electrolyte additive for the lithium metal anode, Li metal coin cells were assembled using a thin lithium anode (thickness of 50 µm) and low-loading LCO cathode (3.0 mg cm⁻²) and cycled at 2C charge (0.8 mA cm⁻²) and 4C discharge (1.6 mA cm⁻²). As shown in Fig. 2a, the Li‖LCO coin cells with the baseline electrolyte maintained 80% capacity retention after 900 cycles. Impressively, NAMI-MOP exhibited a much slower capacity drop and achieved a double increase in cycle life when compared to the baseline electrolyte (3 M LiFSI in DME) at the same capacity retention. Under high rate cycling conditions (2C charge and 4C discharge) and high upper cut-off voltage (>4.2 V), the fast (de)intercalation of lithium ions in LCO lattices tends to accelerate unfavorable cobalt dissolution and lattice oxygen loss, which result in the structural degradation of LCO and capacity decay.^34,35 To investigate the effects of NAMI-MOP on the LCO stability, inductively coupled plasma atomic emission spectrometry (ICP-AES) was performed on the cycled Li metal anodes after 100 cycles with different mass loadings of NAMI-MOP. As shown in Fig. 2b, the concentration of deposited Co on the Li metal anode cycled with the baseline electrolyte was determined to be 58.61 ppm. In contrast, the concentrations of deposited Co were significantly decreased when the cells were cycled with the NAMI-MOP additive. For example, at 0.5% loading of NAMI-MOP, the Co on the cycled anode was reduced to 22.89 ppm. Moreover, increasing the loading of NAMI-MOP further lowered the concentrations of Co deposition on the Li metal anodes and 1.5–2.0% loadings were found to be the optimal amount for suppressing the Co dissolution and subsequent deposition onto the Li metal anode. In addition, X-ray photoelectron spectroscopy (XPS) indicates strong Co 2p_3/2 and Co 2p_1/2 peaks of cobalt species deposited on the Li metal anode after 100 cycles with the baseline electrolyte, as shown in Fig. 2c. Conversely, the Li metal anode after being cycled with NAMI-MOP showed no obvious XPS signals of Co species (Fig. 2d). The XPS results are consistent with the ICP-AES results, suggesting that the Co dissolution from the cathode and subsequent deposition onto the Li anode surface were suppressed by the NAMI-MOP additive. This can be attributed to the formation of a MOP-derived CEI layer which can contribute to the stabilization of the LCO lattices and inhibition of the Co dissolution.

Fig. 2 (a) Cycling performance of low-loading LCO (3.0 mg cm⁻²) lithium metal coin cells in an ether-based electrolyte (3 M LiFSI in DME) with and without NAMI-MOP at 2C charge and 4C discharge at 3.0–4.4 V. (b) ICP-AES results of cobalt concentration on cycled Li metal anodes after 100 cycles at 3.0–4.4 V, 2C charge (0.8 mA cm⁻²) and 4C discharge (1.6 mA cm⁻²). (c) Normalized XPS depth profiles of Co 2p spectra of a recovered Li metal anode after 100 cycles with the baseline electrolyte (3 M LiFSI in DME). (d) Normalized XPS depth profiles of Co 2p spectra of a recovered Li metal anode after 100 cycles with the NAMI-MOP added electrolyte. (e) Cycling performance of high-loading LCO (12.2 mg cm⁻²) lithium metal (thickness of 50 µm) single-layer pouch cells in an ether-based electrolyte (3 M LiFSI in 1,3-DMP) with and without NAMI-MOP at 1C and 3.0–4.4 V. (f) and (g) the charge/discharge curves of (e).

Based on the preliminary results of coin cell studies, the cycling tests of Li‖LCO (12.2 mg cm⁻²) single-layer pouch cells have also been carried out at a rate of 1C (2.0 mA cm⁻²) charge and discharge to further evaluate the effectiveness of NAMI-MOP as an electrolyte additive. As shown in Fig. 2e–g, the pouch cell with the baseline electrolyte (3 M LiFSI in 1,3-DMP) delivered an initial capacity of 162 mA h g⁻¹ and an initial coulombic efficiency (CE) of 93.84%. However, it exhibited a rapid capacity decay after 100 cycles and retained 80% capacity after 150 cycles. In comparison, the pouch cell fabricated with the NAMI-MOP additive manifested remarkably improved cycling stability and retained 96% capacity after 150 cycles, achieving an extended cycle life from 150 cycles to 500 cycles. These results further illustrate the advantages of employing NAMI-MOP as electrolyte additives to stabilize the cycling of LCO-based lithium metal cells.

To realize the underlying mechanism of the exceptionally improved cycling performance of Li‖LCO batteries enabled by the NAMI-MOP additive, the surface microstructure of LCO particles was investigated by HRTEM. The cycled LCO cathodes were obtained from single-layer pouch cells after formation cycles (same charge and discharge rates, 0.1C × 2, 0.2C × 4, 0.5C × 4) between 3.0 and 4.4 V. As shown in Fig. 3a, a thick and uneven (3.9–10 nm) CEI layer is observed on the LCO surface after being cycled with the baseline electrolyte, indicating extensive accumulation of decomposition products due to serious side reactions between electrode and electrolyte components. Comparatively, the surface of LCO particles is covered by a much thinner CEI layer (1.2–1.5 nm, see Fig. 3e) after being cycled with NAMI-MOP, suggesting fewer decomposition products on the LCO surface. Meanwhile, the chemical composition of these CEI layers on LCO particles was determined by XPS tests. As shown in Fig. 3b and f, the O 1s XPS peak of the cathode lattice oxygen can be detected in all samples, revealing that the CEI layer thickness is smaller than the limited probing depth of XPS (<10 nm), and thus the change in relative intensity of lattice oxygen could be regarded as a describer of CEI formation.³⁶ For example, the surface O 1s spectrum of LCO after being cycled with NAMI-MOP exhibited much higher intensity of lattice oxygen, suggesting the formation of a thinner CEI layer, which is consistent with the observation of HRTEM. Furthermore, the intensities of the C=O peak owing to the oxidation of ether solvent molecules and the intensities of SO₄²⁻ and SO₃²⁻ peaks due to the oxidation of FSI⁻ anions were significantly decreased after the LCO cathode was cycled with the NAMI-MOP additive (Fig. 3g), revealing the mitigated oxidative decomposition of electrolyte components on the cathode surface. Importantly, the rise of the C–O peak in the surface O 1s XPS spectrum (Fig. 3f) and Cu(0) species in the surface Cu 2p XPS spectrum (Fig. 3h) can be ascribed to the preferred decomposition of NAMI-MOP, indicating the participation of the NAMI-MOP additive in CEI formation. In contrast, the weaker intensity of lattice oxygen on the surface of LCO after being cycled with the baseline electrolyte (Fig. 3b) reveals a much thicker CEI layer and this can be attributed to the uncontrolled electrolyte decomposition which results in the overgrowth of the CEI layer on the cathode surface. Based on the results of HRTEM and surface XPS analyses, the NAMI-MOP exhibited film-forming behavior and plays a key role in constructing a thinner CEI with enhanced stability and suppressing electrolyte decomposition. Furthermore, XPS depth analysis of the LCO electrode exhibited a stronger peak intensity of LCO lattice oxygen after being cycled with NAMI-MOP, as seen in the normalized O 1s spectra of Fig. 3f, suggesting the effective retention of LCO crystal structures. Moreover, the rise of Cu(0) signals in the LCO lattice (Fig. 3h) indicates the formation of a copper-doped solid solution by NAMI-MOP decomposition. It is worth noting that copper doping has been reported as an effective strategy to stabilize LCO cathodes³⁷ and other cathode systems.^38,39 Therefore, these XPS results are consistent with the aforementioned ICP-AES results and suggest that the Cu doping effect induced by NAMI-MOP decomposition contributes to the stabilization of LCO lattices and inhibition of Co dissolution and lattice oxygen loss.

Fig. 3 (a) HRTEM image of a LCO cathode after formation cycles with the baseline electrolyte (same charge and discharge rates, 0.1C × 2, 0.2C × 4, 0.5C × 4). (b–d) Normalized O 1s, Cu 2p and S 2p XPS depth profiles for LCO cathodes after being cycled with the baseline electrolyte. (e) HRTEM image of a LCO cathode after formation cycles with a NAMI-MOP added electrolyte (same charge and discharge rates, 0.1C × 2, 0.2C × 4, 0.5C × 4). (f–h) Normalized O 1s, Cu 2p and S 2p XPS depth profiles for LCO cathodes after being cycled with a NAMI-MOP added electrolyte.

To further explore the effect of NAMI-MOP on improving the cycling performance of Li‖LCO batteries, the SEI composition of cycled metallic Li anodes was characterized by XPS. As shown in Fig. 4a and b, both SEIs featured an organic-rich outer layer due to the high ratio of organic species (indicated by C–C/C–H and C=C signals in C 1s spectra) on the anode surface and an inorganic-rich inner layer due to the lowered concentration of organic components. In comparison, the C 1s and Cu 2p XPS depth profiles of the Li anode after being cycled with NAMI-MOP exhibited higher organic content (Fig. 4b) as well as the presence of Cu(0) and Cu(I) species (Fig. 4d). The increased content of organic components and the formation of copper species with low oxidation states should be attributed to the reduction of NAMI-MOP on the Li metal anode. Moreover, the Li anode cycled with the NAMI-MOP-added electrolyte showed a stronger F 1s signal (685 eV, corresponding to LiF) in the SEI at various depths (Fig. 4f). Notably, LiF formation arises from the desirable FSI⁻ decomposition, and LiF is widely considered as one of the most preferable interphase components for Li metal anodes due to its low electronic conductivity and high physical/chemical stability.^40–43 The LiF enrichment in the SEI suggests the promoted anion decomposition on the Li anode surface. Such enhancement is closely related to the Lewis acidic open Cu(II) sites of NAMI-MOP which can interact with Lewis basic FSI⁻ anions and facilitate preferential anion decomposition and LiF formation at the SEI.⁴⁴ Additionally, the Lewis acid–base interactions between the NAMI-MOP and FSI⁻ anions tend to lower the LUMO level of the FSI⁻ anions and make it more prone to reduction.⁴⁵ As seen in Fig. 4g and S13, the total proportion of sulfides (assigned as Li₂S) and polysulfides (such as Li₂S₂ and Li₂S₄) in the surface S 2p XPS spectrum significantly increases from 6% (baseline electrolyte) to 58% (NAMI-MOP added electrolyte). Specifically, 25% and 23% of S-containing species are lithium sulfide and polysulfides, respectively, on the Li anode surface after being cycled with the NAMI-MOP added electrolyte. Lithium sulfide and polysulfides have been identified as beneficial SEI components with high ionic conductivity (10⁻⁵ S cm⁻¹) that can improve the ionic conductivity of the SEI and stabilize the Li metal anode.^46–52 Therefore, the XPS data clearly show the participation of the NAMI-MOP additive in the SEI formation as predicted by our frontier orbitals analysis. Meanwhile, the NAMI-MOP additive can regulate anion reactivity towards Li metal anodes and promote the formation of favorable inorganic SEI components such as LiF and lithium (poly)sulfides, contributing to the stabilization of Li metal anodes.

Fig. 4 Normalized C 1s, Cu 2p, F 1s XPS depth profiles for Li metal anodes after formation cycles (same charge and discharge rates, 0.1C × 2, 0.2C × 4, 0.5C × 4) with the baseline electrolyte (a, c and e) and NAMI-MOP added electrolyte (b, d and f). (g) Analysis of sulfur-containing SEI species based on the sulfur 2p XPS spectra of Li metal anodes after formation cycles (same charge and discharge rates, 0.1C × 2, 0.2C × 4, 0.5C × 4) with and without the NAMI-MOP additive.

Building on the above cycling results and post-mortem analysis, high temperature (45 °C) cycling tests were conducted to further examine the advantages of the NAMI-MOP additive. As shown in Fig. 5a, a Li‖LCO (12.2 mg cm⁻²) pouch cell with the baseline electrolyte (3 M LiFSI in 1,3-DMP) delivered an initial capacity of 162 mA h g⁻¹ comparable to the capacity delivered at room temperature (see Fig. 2e) at the same current density but suffered fast capacity fading and CE fluctuation after 70 cycles because of the accelerated electrolyte consumption at elevated temperature. Interestingly, the pouch cell cycled with the NAMI-MOP additive exhibited a two-fold increase in cycle life although it delivered a smaller initial capacity (158.9 mA h g⁻¹). These cycling results indicate that the rationally designed NAMI-MOP additive enhances the cycling stability of Li‖LCO under high temperature conditions. In addition, the scalability of the NAMI-MOP additive was demonstrated in Ah-level multi-stacked pouch cells. Specifically, Ah-level cells were fabricated using LCO cathodes (12.2 mg cm⁻²), ultrathin Li metal anodes (20 µm in thickness) and PE separators (20 µm in thickness). The long-term cycling stability is illustrated in Fig. 5b, where the pouch cell with the NAMI-MOP additive maintained a capacity retention of 80% at 1000 mA after 260 cycles. In comparison, the control pouch cell exhibited a comparable initial capacity at the same current but suffered accelerated capacity fading after 100 cycles. The cycling results of Ah-level cells give strong evidence to verify the effectiveness of the NAMI-MOP additive in large capacity LMBs.

Fig. 5 (a) High-temperature cycling performance of LCO (12.2 mg cm⁻²) lithium metal (thickness of 50 µm) single-layer pouch cells with and without NAMI-MOP at 1C charge and discharge, 3.0–4.4 V and 45 °C. (b) Cycling performance of multi-stacked Ah-level LCO (12.2 mg cm⁻²) lithium metal (thickness of 20 µm) pouch cells with and without NAMI-MOP at 1000 mA charge and discharge currents, 3.0–4.3 V.

Finally, the practicality of applying the NAMI-MOP additive in lithium metal batteries with high gravimetric energy density (GED) was demonstrated using pouch cells with much larger capacities. Specifically, a LCO-based lithium metal pouch cell was fabricated using high-loading LCO cathodes (22 mg cm⁻²), ultrathin Li metal anodes (20 µm in thickness) and thin PE separators (8 µm in PE thickness, one-side coated with 3 µm ceramic coating facing cathodes). Moreover, the E/C ratio of the cell was strictly limited to 1.71 g Ah⁻¹. The specific energy density of the fabricated pouch cell is determined based on the discharge energy at 0.1C and the total weight of the cell including anode, cathode, separator, electrolyte, tabs and battery package. As seen in Fig. 6a, the Li‖LCO pouch cell delivered an energy density of 362 Wh kg⁻¹ and exhibited stable cycling at 0.2C charge (1200 mA) and 0.5C discharge (3000 mA) for at least 32 cycles. Additionally, another pouch cell with higher GED was fabricated with high-loading NCM811 cathodes (20 mg cm⁻²), ultrathin Li metal anodes (20 µm in thickness) and thin PE separators (8 µm in thickness), yielding a N/P ratio of 1.03. A low E/C ratio of 1.75 g Ah⁻¹ was applied in the Li‖NMC pouch cell. As seen in Fig. 6c and d, the resulting pouch cell delivered a high energy density of 412 Wh kg⁻¹ and exhibited stable cycling at 0.5C (2500 mA) for at least 30 cycles. These two practical and workable LMBs demonstrate the promising potential of NAMI-MOP for large capacity and high GED LMBs.

Fig. 6 (a) Cycling performance of a 6.4 Ah LCO (22 mg cm⁻²) lithium metal (thickness of 20 µm) pouch cell in an ether-based electrolyte with NAMI-MOP at 0.2C charge and 0.5C discharge, 3.0–4.4 V. (b) The corresponding charge/discharge curves of the Li‖LCO pouch cell at different rates and cycles. (c) Cycling performance of a 5.7 Ah NMC811 (20 mg cm⁻²) lithium metal (thickness of 20 µm) pouch cell in an ether-based electrolyte with NAMI-MOP at 0.5C charge and discharge, 3.0–4.3 V. (d) The corresponding charge/discharge curves of the Li‖NMC811 pouch cell at different rates and cycles.

4 Conclusions

To sum up, a novel MOP material (NAMI-MOP) that serves as a multifunctional electrolyte additive for enhancing the cycling performance of LMBs was designed and synthesized. The NAMI-MOP possesses excellent solubility in commercial carbonate-based and ether-based electrolytes. Consistent cycling results from coin-type and single-layer pouch cells to multi-stacked Ah-level pouch cells indicate the effectiveness of the NAMI-MOP additive for stabilizing and lengthening the cycle life of LMBs. Post-mortem analysis revealed the film-forming nature of NAMI-MOP which forms a MOP-derived CEI to stabilize LCO and suppress electrolyte degradation. Moreover, the anionophilic open metal sites of NAMI-MOP can regulate the anion chemistry and promote the formation of a LiF and lithium (poly)sulfide-rich SEI to stabilize Li metal anodes. When coupled with high-loading LCO (22 mg cm⁻²) or NMC811 (20 mg cm⁻²) cathodes, the 6.4 Ah Li‖LCO and 5.7 Ah Li‖NMC811 pouch cells with the NAMI-MOP additive offered high energy densities of 362 Wh kg⁻¹ and 412 Wh kg⁻¹, respectively, illustrating the practical utilization of the NAMI-MOP additive in LMBs.

Author contributions

Yan-Lung Wong: conceptualization, funding acquisition, investigation, formal analysis, data curation, validation, writing – original draft, writing – review & editing. Wei Huang: investigation, data curation, writing – review & editing. Cuili Zhang: formal analysis. Lang Wang: resources. Shengbo Lu: conceptualization, resources, project administration, supervision. Chenmin Liu: conceptualization, writing – review & editing, supervision.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: including the synthesis route, structural characterization, theoretical calculations, experimental details, and spectral analysis. See DOI: https://doi.org/10.1039/d5ta07061a.

Acknowledgements

The authors are grateful for financial support from the Innovation and Technology Support Programme (ITP-026-22NP) offered by the Innovation and Technology Commission, the Government of the Hong Kong SAR.

References

1. W. Xu, J. Wang, F. Ding, X. Chen, E. Nasybulin, Y. Zhang and J.-G. Zhang, Energy Environ. Sci., 2014, 7, 513–537.
2. X.-B. Cheng, R. Zhang, C.-Z. Zhao and Q. Zhang, Chem. Rev., 2017, 117, 10403–10473.
3. D. Lin, Y. Liu and Y. Cui, Nat. Nanotechnol., 2017, 12, 194–206.
4. J. Liu, Z. Bao, Y. Cui, E. J. Dufek, J. B. Goodenough, P. Khalifah, Q. Li, B. Y. Liaw, P. Liu, A. Manthiram, Y. S. Meng, V. R. Subramanian, M. F. Toney, V. V. Viswanathan, M. S. Whittingham, J. Xiao, W. Xu, J. Yang, X.-Q. Yang and J.-G. Zhang, Nat. Energy, 2019, 4, 180–186.
5. C. Jin, T. Liu, O. Sheng, M. Li, T. Liu, Y. Yuan, J. Nai, Z. Ju, W. Zhang, Y. Liu, Y. Wang, Z. Lin, J. Lu and X. Tao, Nat. Energy, 2021, 6, 378–387.
6. J. B. Goodenough and K.-S. Park, J. Am. Chem. Soc., 2013, 135, 1167–1176.
7. Q. Liu, X. Su, D. Lei, Y. Qin, J. Wen, F. Guo, Y. A. Wu, Y. Rong, R. Kou, X. Xiao, F. Aguesse, J. Bareño, Y. Ren, W. Lu and Y. Li, Nat. Energy, 2018, 3, 936–943.
8. L. Wang, B. Chen, J. Ma, G. Cui and L. Chen, Chem. Soc. Rev., 2018, 47, 6505–6602.
9. Y. Lyu, X. Wu, K. Wang, Z. Feng, T. Cheng, Y. Liu, M. Wang, R. Chen, L. Xu, J. Zhou, Y. Lu and B. Guo, Adv. Energy Mater., 2021, 11, 2000982.
10. G. Wang, Z. Bi, A. Zhang, P. Das, H. Lin and Z.-S. Wu, Engineering, 2024, 37, 105–127.
11. M. Qi, L. Xie, Q. Han, L. Zhu, L. Chen and X. Cao, J. Energy Storage, 2022, 47, 103641.
12. Q. Wu, B. Zhang and Y. Lu, J. Energy Chem., 2022, 74, 283–308.
13. K. Xu, Chem. Rev., 2014, 114, 11503–11618.
14. W. Zhang, Y. Lu, L. Wan, P. Zhou, Y. Xia, S. Yan, X. Chen, H. Zhou, H. Dong and K. Liu, Nat. Commun., 2022, 13, 2029.
15. Z. Li, Y. Liao, H. Ji, X. Lin, Y. Wei, S. Hao, X. Hu, L. Yuan, Z. Huang and Y. Huang, Adv. Energy Mater., 2025, 15, 2404120.
16. J. Heine, P. Hilbig, X. Qi, P. Niehoff, M. Winter and P. Bieker, J. Electrochem. Soc., 2015, 162, A1094.
17. Y. Zhang, Y. Wu, H. Li, J. Chen, D. Lei and C. Wang, Nat. Commun., 2022, 13, 1297.
18. P. Shi, L. Zhang, H. Xiang, X. Liang, Y. Sun and W. Xu, ACS Appl. Mater. Interfaces, 2018, 10, 22201–22209.
19. Y. Li, B. Wen, N. Li, Y. Zhao, Y. Chen, X. Yin, X. Da, Y. Ouyang, X. Li, P. Kong, S. Ding, K. Xi and G. Gao, Angew. Chem., Int. Ed., 2025, 64, e202414636.
20. W. Zhang, Y. Lu, Q. Feng, H. Wang, G. Cheng, H. Liu, Q. Cao, Z. luo, P. Zhou, Y. Xia, W. Hou, K. Zhao, C. Du and K. Liu, Nat. Commun., 2025, 16, 3344.
21. F. Chu, J. Hu, C. Wu, Z. Yao, J. Tian, Z. Li and C. Li, ACS Appl. Mater. Interfaces, 2019, 11, 3869–3879.
22. X. Li, P. Xu, Y. Tian, A. Fortini, S. H. Choi, J. Xu, X. Tan, X. Liu, G. Chen, C. Zhang, X. Lu, L. Jin, Q. Wang, L. Shen and Y. Lu, Adv. Mater., 2022, 34, 2107787.
23. S. Bai, Y. Sun, J. Yi, Y. He, Y. Qiao and H. Zhou, Joule, 2018, 2, 2117–2132.
24. L. Shen, H. B. Wu, F. Liu, J. Shen, R. Mo, G. Chen, G. Tan, J. Chen, X. Kong, X. Lu, Y. Peng, J. Zhu, G. Wang and Y. Lu, Adv. Funct. Mater., 2020, 30, 2003055.
25. A. U. Mu, G. Cai and Z. Chen, Adv. Sci. (Weinh.), 2024, 11, 2305280.
26. C. Shen, J. Liu, D.-H. Zhang, Q. Li, H. Zhang, Z. Zhuo, W. Wang, Z.-A. Nan, L. Liu and Y.-G. Huang, J. Mol. Struct., 2025, 1321, 140063.
27. J. Liu, R. Zhang, X. Xie, J. Wang, F. Jin, Z. Wang, T. Wang, P. Cheng, J. Lu and Z. Zhang, Angew. Chem., Int. Ed., 2025, 64, e202414211.
28. S. S. Nagarkar, M. Tsujimoto, S. Kitagawa, N. Hosono and S. Horike, Chem. Mater., 2018, 30, 8555–8561.
29. X. Liu, X. Wang, A. V. Bavykina, L. Chu, M. Shan, A. Sabetghadam, H. Miro, F. Kapteijn and J. Gascon, ACS Appl. Mater. Interfaces, 2018, 10, 21381–21389.
30. H. Furukawa, J. Kim, N. W. Ockwig, M. O'Keeffe and O. M. Yaghi, J. Am. Chem. Soc., 2008, 130, 11650–11661.
31. J. Holoubek, H. Liu, Z. Wu, Y. Yin, X. Xing, G. Cai, S. Yu, H. Zhou, T. A. Pascal, Z. Chen and P. Liu, Nat. Energy, 2021, 6, 303–313.
32. Y. Yamada, M. Yaegashi, T. Abe and A. Yamada, Chem. Commun., 2013, 49, 11194–11196.
33. S. Chen, J. Fan, Z. Cui, L. Tan, D. Ruan, X. Zhao, J. Jiang, S. Jiao and X. Ren, Angew. Chem., Int. Ed., 2023, 62, e202219310.
34. G. G. Amatucci, J. M. Tarascon and L. C. Klein, Solid State Ion., 1996, 83, 167–173.
35. Y. Liu, Y. Zhu and Y. Cui, Nat. Energy, 2019, 4, 540–550.
36. L. Hu, Z. Qian, W. Gao, X. Wang and Y. Tian, J. Mater. Sci., 2020, 55, 2052–2067.
37. M. Zou, M. Yoshio, S. Gopukumar and J.-i. Yamaki, Chem. Mater., 2003, 15, 4699–4702.
38. L. Yang, F. Ren, Q. Feng, G. Xu, X. Li, Y. Li, E. Zhao, J. Ma and S. Fan, J. Electron. Mater., 2018, 47, 3996–4002.
39. X.-Z. Kong, D.-L. Li, K. Lahtinen, T. Kallio and X.-Q. Ren, J. Electrochem. Soc., 2020, 167, 140545.
40. X. Ren, L. Zou, X. Cao, M. H. Engelhard, W. Liu, S. D. Burton, H. Lee, C. Niu, B. E. Matthews, Z. Zhu, C. Wang, B. W. Arey, J. Xiao, J. Liu, J.-G. Zhang and W. Xu, Joule, 2019, 3, 1662–1676.
41. N. von Aspern, G.-V. Röschenthaler, M. Winter and I. Cekic-Laskovic, Angew. Chem., Int. Ed., 2019, 58, 15978–16000.
42. M. He, R. Guo, G. M. Hobold, H. Gao and B. M. Gallant, Proc. Nat. Acad. Sci., 2020, 117, 73–79.
43. Y. Chen, Z. Yu, P. Rudnicki, H. Gong, Z. Huang, S. C. Kim, J.-C. Lai, X. Kong, J. Qin, Y. Cui and Z. Bao, J. Am. Chem. Soc., 2021, 143, 18703–18713.
44. Z. Wang, Z. Xu, Y. Mu, B. Slater, J. Li, L. Zeng, B. Guo and K. Wang, ACS Appl. Mater. Interfaces, 2025, 17, 25223–25236.
45. T. Li, X.-Q. Zhang, N. Yao, Y.-X. Yao, L.-P. Hou, X. Chen, M.-Y. Zhou, J.-Q. Huang and Q. Zhang, Angew. Chem., Int. Ed., 2021, 60, 22683–22687.
46. W. Li, H. Yao, K. Yan, G. Zheng, Z. Liang, Y. M. Chiang and Y. Cui, Nat. Commun., 2015, 6, 7436.
47. C. Yan, X.-B. Cheng, C.-Z. Zhao, J.-Q. Huang, S.-T. Yang and Q. Zhang, J. Power Sources, 2016, 327, 212–220.
48. G. Li, Y. Gao, X. He, Q. Huang, S. Chen, S. H. Kim and D. Wang, Nat. Commun., 2017, 8, 850.
49. X.-B. Cheng, C. Yan, H.-J. Peng, J.-Q. Huang, S.-T. Yang and Q. Zhang, Energy Storage Mater., 2018, 10, 199–205.
50. H. Chen, A. Pei, D. Lin, J. Xie, A. Yang, J. Xu, K. Lin, J. Wang, H. Wang, F. Shi, D. Boyle and Y. Cui, Adv. Energy Mater., 2019, 9, 1900858.
51. Z. Wang, C. Chen, D. Wang, Y. Zhu and B. Zhang, Angew. Chem., Int. Ed., 2023, 62, e202303950.
52. P. Li, H. Zhang, J. Lu and G. Li, Angew. Chem., Int. Ed., 2023, 62, e202216312.

---

Footnote

† These authors contributed equally to this work.

---