Mn(II) trifluoromethanesulfonate (Mn(OTf)₂, 99.5%) was obtained from Dodo Chemical Reagent Co., Ltd. Bismuth(III) chloride (BiCl₃, >98%), copper(II) chloride dihydrate (CuCl₂·2H₂O, 99.99%), thiourea (CH₄N₂S, 99%), ethylene glycol (C₂H₆O₂, 98%), diglyme (G2, 99.8%), and lithium chloride (LiCl, anhydrous, 99.9%) were purchased from Aladdin Biochemical Technology Co., Ltd. Manganese powder (Mn, 99.9%, 200 mesh) was supplied by Macklin Biochemical Co., Ltd. All chemicals were used as received unless otherwise specified.

AE: Mn(OTf)₂ was dissolved in deionized water to prepare an AE containing 0.5 M Mn(OTf)₂.

All G2-based electrolytes were prepared in an argon-filled glovebox (uNIVERSAL, 2440/750/900, Mikrouna, China). The electrolyte components were weighed using an analytical balance (Quintix 125D-1CN, Sartorius, China) and continuously stirred at room temperature at 200 rpm for 24 h using a magnetic stirrer (TMHB-180CL, Taisite Instrument Co., Ltd., China).

G2-Mn(OTf)₂: Mn(OTf)₂ was dissolved in G2 solvent to obtain a 0.5 M Mn(OTf)₂/G2 electrolyte. The solution was sealed and stirred for 24 h to ensure complete dissolution and homogeneous mixing.

G2-Mn(OTf)₂-BiCl₃: BiCl₃ was added to the 0.5 M Mn(OTf)₂/G2 electrolyte, and the mixture was stirred for 24 h. By adjusting the amount of BiCl₃, electrolytes with different BiCl₃ concentrations (5, 10, and 25 mM) were prepared and denoted as G2-5mM BiCl₃, G2-BiCl₃, and G2-25mM BiCl₃, respectively.

G2-Mn(OTf)₂-LiCl: Based on the 0.5 M Mn(OTf)₂/G2 electrolyte, LiCl was added to achieve a final concentration of 30 mM. After stirring for 24 h, the G2-Mn(OTf)₂-LiCl electrolyte (denoted as G2-LiCl for simplicity) was obtained. The LiCl concentration was selected to provide the same Cl^- concentration as that introduced by 10 mM BiCl₃.

Mn anode: The preparation procedure for the Mn anode in this study was based on previously reported methods^[24]. Mn powder, Super P, and poly(vinylidene fluoride) (PVDF) were mixed in a weight ratio of 8:1:2, followed by adding N-methyl-2-pyrrolidone (NMP) to form a homogeneous slurry (TMHB-180CL, Taisite Instrument Co., Ltd., China; room temperature; 200 rpm for 2 h). Using a 500 μm doctor blade, the well-stirred slurry was uniformly coated onto a glass plate and dried in a vacuum oven (MSK-H200A, Hefei KJ Magnetic Electronics Co., Ltd., China) at 80 °C for 12 h. The dried electrode film was peeled off the glass plate and cut into circular electrodes with a diameter of 12 mm. The mass of each Mn anode was approximately 12-13 mg. As shown in Supplementary Figures 1 and 2, the Mn anode prepared in this study exhibits high chemical purity and good mechanical integrity, showing no cracking or powder shedding after operations such as tweezer handling and battery cell encapsulation.

Cu_1.8S cathode: The preparation procedure for the Cu_1.8S cathode material in this study follows the reported literature^[35]. Briefly, 2.0 g CuCl₂·2H₂O (11.7 mmol) and 3.6 g thiourea (47.3 mmol) were dissolved in 80 mL ethylene glycol at 120 °C individually. Then, the two hot, transparent solutions were mixed and stirred at 300 rpm for another 5 min before transferring the yellowish solution into a 200 mL autoclave (LC-KH-200, Lichen Scientific Instruments (Hunan) Co., Ltd., China). The autoclave was sealed and heated at 140 °C for 6 h to produce the black suspension liquid, which was vacuum filtered, washed with de-ionized water and anhydrous ethanol 3 times each, and vacuum dried at 80 °C to obtain the dark gray precursor. This precursor was then sintered at 350 °C at a ramping rate of 5 °C min^-1 in an argon atmosphere for 2 h using a tube furnace (OTF-1200X, Hefei KJ Magnetic Electronics Co., Ltd., China) and cooled to room temperature naturally to obtain the final product. To prepare the cathode slurry, 210 mg Cu_1.8S powder, 60 mg Ketjen Black, and 30 mg PVDF were ground in a weight ratio of 7:2:1 and uniformly stirred with 3 mL NMP at 200 rpm to obtain a homogeneous slurry (TMHB-180CL, Taisite Instrument Co., Ltd., China; room temperature; 200 rpm for 2 h). The slurry was directly coated onto Cu foil, dried in an air-blowing oven (DHG-9070A, Shanghai Yiheng Scientific Instrument Co., Ltd., China) for 30 min, followed by vacuum drying for 2 h to obtain the Cu_1.8S cathode. The cathode was cut into circular electrodes with a diameter of 12 mm, with an active material loading of 1-2 mg cm^-2 per electrode.

X-ray diffraction (XRD, 2θ range: 10°-80°, Shimadzu XRD-6100) and X-ray photoelectron spectroscopy (XPS, ESCALAB250Xi, Thermo Fisher Scientific) were used to characterize the chemical composition of the Mn anode surface. Scanning electron microscopy (SEM, accelerating voltage: 20 kV, HITACHI SU8220, Japan; ZEISS 300 Carl Zeiss, Germany) was used to observe the surface morphology of the Mn electrode. High-resolution mass spectrometry (HRMS, Bruker ESI-Q-TOF MS/MS) was used to characterize the electrolyte. Ultraviolet photoelectron spectroscopy (UPS, Thermo Fisher Scientific ESCALAB Xi) was used to determine the work function and valence band maximum of the electrodes.

All batteries were assembled in an argon-filled glovebox with moisture and oxygen levels below 0.01 ppm. The electrochemical performance was evaluated using CR2032-type coin cells in several configurations: Mn||Mn symmetric cells, Mn||Ti, Mn||Cu, and Mn||SS asymmetric cells, as well as Mn||Cu_1.8S full cells. For the asymmetric cells, Ti foil (20 μm thick, 12 mm diameter) and Cu foil (100 μm thick, 12 mm diameter) were utilized as current collectors. Whatman GF/D glass fiber (17 mm diameter) served as the separator. Following assembly, the cells were sealed under a constant pressure of 85 kg cm^-2 and allowed to equilibrate for at least 8 h prior to testing. Galvanostatic charge discharge (GCD) profiles were recorded using a LANHE battery testing system (CT3004A, Wuhan, China). Additionally, Cyclic voltammetry (CV) and Tafel polarization curves were obtained via an electrochemical workstation (CHI660E, Shanghai, China).

Tafel polarization curves: Three-electrode electrochemical measurements were performed using a Swagelok-type cell to evaluate the interfacial redox kinetics of the Mn anode. A pristine Mn anode or a BiCl₃-modified Mn anode was used as the working electrode, while an untreated Mn anode was used as both the counter and reference electrodes. The BiCl₃-modified Mn anode was obtained after cycling in the G2-BiCl₃ electrolyte, followed by rinsing with G2 and drying. To eliminate the direct contribution of BiCl₃ or dissolved Bi-containing species in the electrolyte, all three-electrode measurements were conducted in the G2-Mn(OTf)₂ electrolyte without BiCl₃. Prior to the measurements, the cells were allowed to equilibrate at the open-circuit potential for 30 min. Tafel measurements were subsequently performed at a scan rate of 0.001 V s^-1. All electrochemical measurements were conducted at a controlled temperature of 28 °C.

All spin-polarized DFT calculations were performed by employing the Vienna ab initio simulation package. The exchange-correlation interaction was described using the Perdew-Burke-Ernzerhof functional within the generalized gradient approximation. Grimme’s DFT-D3 method was used to treat van der Waals interactions. The Brillouin zone was sampled using a 3 × 3 × 1 k-point grid, and the plane-wave energy cutoff was set to 400 eV. The convergence criteria were set to 10-5 eV for energy and 0.05 eV/Å for force. A 15 Å vacuum layer was added in the z-direction to avoid the interaction between periodic images.

The binding energy (E_b) between two isolated systems can be calculated by the formula:

(1) $$ \begin{equation} \begin{aligned} E_b = E_{AB} - E_A - E_B \end{aligned} \end{equation} $$

where E_AB represents the total energy of the AB system. E_A and E_B are the energies of isolated A and B systems, respectively.

The adsorption energy (E_ads) was defined as:

(2) $$ \begin{equation} \begin{aligned} E_{ads} = E_{total} - E_{sub} - E_{adsorbate} \end{aligned} \end{equation} $$

where E_total, E_sub, and E_adsorbate are the total energies of the substrate with adsorbate, the substrate, and the free adsorbate, respectively.

Firstly, the chemical stability of the Mn anode in various electrolytes was evaluated. It is evident from Supplementary Figure 3 that the Mn anode demonstrated vigorous gas evolution upon immersion in the AE, whereas it remained stable in the nonaqueous counterparts (G2-Mn(OTf)₂ and G2-BiCl₃). This underscores the chemical incompatibility between the Mn anode and aqueous media, which precludes stable Mn deposition.

Electrochemical assessments in Mn||Mn symmetric cells further validated these results [Supplementary Figure 4]. At a current density of 0.1 mA cm^-2, the Mn||AE||Mn symmetric cell displayed a considerable polarization voltage of about 4 V and a limited cycle life of less than 40 h, primarily due to the parasitic HER. However, symmetric cells using G2-Mn(OTf)₂ and G2-BiCl₃ electrolytes exhibited excellent cycling stability of over 1,000 h [Figure 1A]. The G2-BiCl₃-based symmetric cells maintained stable voltage profiles without obvious short-circuiting or abrupt polarization growth during long-term cycling [Figure 1B, Supplementary Figure 5A]. To further evaluate the robustness of Mn plating/stripping under more demanding conditions, additional symmetric cell tests were conducted at higher current densities and areal capacities [Supplementary Figure 6]. The G2-BiCl₃ electrolyte maintained stable cycling performance for over 1,000 h at 0.5 mA cm^-2 with an areal capacity of 0.25 mAh cm^-2. Even under a higher current density of 1.0 mA cm^-2 and an areal capacity of 0.5 mAh cm^-2, the G2-BiCl₃ electrolyte still enabled stable Mn plating/stripping for more than 1,000 h, corresponding to a cumulative plated capacity of approximately 500 mAh cm^-2. In contrast, the G2-Mn(OTf)₂ electrolyte exhibited pronounced polarization fluctuations and unstable voltage profiles under the same high-loading condition, indicating unstable Mn deposition/stripping behavior.

The overpotential of symmetric cells decreased significantly with BiCl₃ addition. Concentration-dependent studies [Supplementary Figure 7] indicated that 10 mM BiCl₃ was optimal, as a lower concentration (5 mM) was insufficient to reduce polarization, whereas the higher concentration (25 mM) showed an abrupt increase in overpotential at the end of cycling. The rate-capability measurements [Figure 1C] also showed the kinetic benefits of the BiCl₃ additive. In G2-BiCl₃, no significant polarization was observed with an increase in current density, whereas in G2-Mn(OTf)₂, polarization was observed to increase with the rise in current density. Furthermore, the overpotential was restored to the initial level when the current density was reset back to 0.1 mA cm^-2, indicating good electrochemical reversibility and improved reaction kinetics.

To demonstrate the effect of BiCl₃ on Mn deposition, asymmetric cell cycling was carried out. Stable cycling was also achieved for Mn||Cu cells at 0.1 mA cm^-2 and 0.025 mAh cm^-2 over 100 cycles [Figure 1D, Supplementary Figure 5B]. However, a kinetic difference was found in CV [Figure 1E]. The strong coordination of Mn²⁺ with solvent molecules in the G2-Mn(OTf)₂ electrolyte may contribute to sluggish interfacial Mn deposition kinetics, resulting in a more negative onset potential of -0.32 V (vs. Mn²⁺/Mn). This led to a deposition/dissolution overpotential of 1.29 V, corresponding to slow redox kinetics. The improved kinetics in the G2-BiCl₃ electrolyte are attributed to the BiCl₃ additive, which regulates the Mn anode interface and facilitates reversible Mn deposition and stripping.

The response current density and the integrated CV peak area of the symmetric cells [Supplementary Figure 8] indicated that BiCl₃ promoted the deposition/stripping activity. To quantitatively evaluate the interfacial redox kinetics of the Mn anode, exchange current density (j₀) measurements were further performed using a three-electrode configuration [Figure 1F]. The measurements were conducted in the G2-Mn(OTf)₂ electrolyte without BiCl₃ to exclude the direct contribution of BiCl₃ in the electrolyte. The BiCl₃-modified Mn anode exhibited a j₀ of 0.01449 mA cm^-2, approximately 5.75 times higher than that of the pristine Mn anode (0.00252 mA cm^-2), indicating substantially faster Mn²⁺/Mn interfacial redox kinetics. This result is consistent with the enhanced deposition/stripping activity observed in the symmetric-cell measurements. These findings demonstrate that the BiCl₃ additive effectively improves the interfacial electrochemical kinetics of Mn deposition/stripping.

To elucidate the interfacial effects of BiCl₃ incorporation on the Mn anode, XRD was used to characterize the Mn anode surface after immersion. By-product Mn₃O₄ (PDF#18-0803) was formed after the immersion of the Mn anode in AE [Figure 2A], while the Mn anode in the G2-Mn(OTf)₂ electrolyte was still pristine [Figure 2B]. In the Mn anode in the G2-BiCl₃ system, however, clear diffraction peaks of Bi metal were found [Figure 2C]. These results suggest that a spontaneous interfacial reaction between Mn and Bi³⁺ species occurs during immersion, forming Bi-related species on the Mn surface (3Mn + 2Bi³⁺ → 3Mn²⁺ + 2Bi). SEM and energy dispersive spectrometer (EDS) analyses further confirmed this surface change [Figure 2D-F, Supplementary Figure 9]. For the SEM/EDS immersion characterization, Mn powder was used to facilitate direct observation of the surface changes. The surface morphology of the Mn powder treated with G2-BiCl₃ electrolyte was evidently altered, and Bi was evenly distributed, suggesting the formation of Bi-containing surface species^[33]. Furthermore, XPS and transmission electron microscope (TEM) characterizations were performed on the immersed Mn anodes to further investigate the chemical characteristics of the interfacial layer. The XPS results reveal the coexistence of Mn, Bi, O, and Cl elements within the interphase [Figure 2G]. The TEM images indicate that the interphase is predominantly amorphous [Figure 2H]. Meanwhile, the TEM-EDS elemental mapping further demonstrates that the Bi-containing interphase is relatively uniformly distributed over the Mn anode surface [Figure 2I]. Considering the strong diffraction signals of metallic Bi and the XPS evidence, metallic Bi is considered to be the predominant functional component within the Bi-containing interphase, although the coexistence of other species cannot be excluded. However, the exact chemical composition of this interphase cannot be unambiguously determined at this stage, as the species detected may originate from manganese oxides, bismuth oxides, chlorine-containing species, and/or other mixed interfacial components. Accordingly, this interfacial layer is referred to as a “Bi-containing interphase” throughout this work.

Recent studies suggest that halogen atoms (X = Cl, Br, I), characterized by large ionic radii and low electronegativity, can reduce the coordination between Mn²⁺ and solvent molecules, reducing deposition overpotentials. Considering that BiCl₃ provides Cl atoms to the electrolyte, one hypothesis was that the kinetic advantages could be due to a halogen-mediated solvation effect. To investigate this, HRMS was performed on G2-Mn(OTf)₂ and G2-BiCl₃ electrolytes [Figure 3].

It is evident from Figure 3A and B that in the G2-Mn(OTf)₂ electrolyte, Mn²⁺ coordinates with G2 and OTf^- to form various cationic and anionic complexes, including [Mn(G2)₂]²⁺, [Mn₂(OTf)₃(G2)₂]⁺, [Mn(OTf)₃]^-, and [Mn₂(OTf)₅]^-. It is important to note that the mass spectrometry measurements for the G2-BiCl₃ electrolyte [Figure 3C and D] show that the most abundant species are the same as those found in the G2-Mn(OTf)₂ electrolyte, namely various cationic and anionic complexes formed between Mn²⁺, G2, and OTf^-. Meanwhile, complexes directly formed between Bi³⁺ and Cl^- (e.g., [BiCl₂(G2)]⁺ and [Bi₂(OTf)Cl₄]⁺) are present only in trace amounts. These results show that the coordination environment of Mn²⁺ in the G2-BiCl₃ electrolyte is primarily governed by G2 and OTf^-, and that adding BiCl₃ has no significant effect on the dominant Mn²⁺ solvation structure. As additional support for refuting the halogen-mediated hypothesis, another control additive was selected, LiCl [Supplementary Figure 10]. In the cell with the electrolyte G2-LiCl (30 mM LiCl), no kinetic improvement was observed, although Cl^- was present in the electrolyte. Taken together, these results suggest that BiCl₃ primarily enhances the reaction kinetics by forming a Bi-containing interphase that regulates Mn deposition, rather than by significantly altering the dominant Mn²⁺ solvation structure, although a minor contribution from dissolved Cl-containing species to the local interfacial environment cannot be completely ruled out.

To explain the kinetic acceleration, DFT calculations and UPS characterization were carried out. The results obtained from the DFT calculations [Figure 4A and B] show that the adsorption energy of Mn atoms on the Bi(012) crystal plane is more negative than that on the Mn(411) plane. This stronger interaction between Mn species and the Bi surface suggests that the Bi-containing interphase may provide favorable adsorption sites, improve Mn adsorption and interfacial charge-transfer conditions, and consequently promote a more controlled Mn deposition process. Moreover, the work function (ϕ) of the Mn anode decreased from 4.45 eV in the G2-Mn(OTf)₂ electrolyte to 3.96 eV in the G2-BiCl₃ electrolyte, as determined by UPS analysis [Figure 4C-F]. The reduced work function suggests a modified surface electronic environment and a more favorable electronic structure for interfacial charge transfer, which may contribute to the enhanced interfacial redox kinetics.

To further understand the interfacial kinetic behavior, distribution of relaxation times analysis was performed based on the electrochemical impedance spectroscopy spectra. As shown in Supplementary Figure 11, the relaxation peaks can be assigned to different electrochemical processes, including solid electrolyte interphase (SEI)-related processes at high frequencies, charge-transfer processes at intermediate frequencies, and ion diffusion processes at low frequencies. Notably, a pronounced relaxation peak is observed in the 10^-4-10^-3 s region for the G2-Mn(OTf)₂ electrolyte, whereas no obvious relaxation peak is observed in this region for the electrolytes containing BiCl₃, including 5, 10, and 25 mM BiCl₃. Furthermore, among the BiCl₃-containing electrolytes, the 10 mM BiCl₃ electrolyte exhibits substantially lower charge-transfer resistance than those with 5 and 25 mM BiCl₃, indicating more favorable interfacial charge-transfer kinetics at the optimized BiCl₃ concentration. These results further support that the introduction of BiCl₃ facilitates more reversible Mn deposition/stripping and improves interfacial kinetics.

Based on these results, BiCl₃ acts as a multifunctional interfacial regulator, as shown in the schematic [Figure 5]. In the early stage of rest or in the initial stages of cycling, Bi³⁺ preferentially displaces surface Mn atoms to form a uniform Bi-containing interphase. This in-situ engineered interphase can work in several ways to optimize the Mn anode: (1) Improving interfacial kinetics: the Bi-containing interphase facilitates faster and more reversible Mn deposition/stripping processes; (2) accelerating the charge transfer: reduced surface work function can enhance the charge transfer rate at the electrochemical interphase; (3) protecting the interface: the Bi-containing interphase forms a good selective ion-conductor to block the formation of detrimental species such as Mn₃O₄, and ensures a high Mn²⁺ flux. To conclude, such a synergistic effect, both of strong atomic adsorption and favorable electronic structure, is an effective mechanism to overcome the slow redox reaction of the Mn anode and is an effective method for developing high-performance nonaqueous MIBs.

The practical effectiveness of the interface modified with BiCl₃ was investigated by synthesizing the cathode material Cu_1.8S by a solvothermal process^[35]. The successful preparation of Cu_1.8S was confirmed by XRD, XPS, and SEM elemental mapping analyses [Supplementary Figure 12].

The effect of the BiCl₃ additive on the full-cell performance was systematically studied. The Mn||G2-BiCl₃||Cu_1.8S cell delivered an average discharge capacity of 214.6 mAh g^-1 over 100 cycles, which was higher than that of the Mn||G2-Mn(OTf)₂||Cu_1.8S cell (158.0 mAh g^-1) at a current density of 20 mA g^-1 [Figure 6A]. Concentration-dependent studies [Supplementary Figure 13] showed that 10 mM BiCl₃ was optimal. When tested at a current density of 100 mA g^-1, the Mn||Cu_1.8S cell with 10 mM BiCl₃ delivered an average discharge capacity of 166.1 mAh g^-1 over 200 cycles, while maintaining almost 100% capacity retention after 200 cycles [Figure 6B]. These results are consistent with the symmetric cell findings. Moreover, the LiCl-based control cell [Supplementary Figure 14] showed no performance enhancement, reinforcing the conclusion that the kinetic gains are uniquely induced by the Bi-containing interphase rather than the presence of Cl atoms.

The GCD profile of the Cu_1.8S cathode in both electrolytes is similar to the dual storage mechanism described in the literature [Figure 6C]. However, the most significant finding is the comparison of the profiles, showing that the BiCl₃ additive significantly reduces voltage polarization, one of the most important parameters in electrochemical processes. This kinetic enhancement was confirmed by CV, which showed that the G2-BiCl₃ system exhibited narrower peak separations and more distinct redox features [Figure 6D], indicating improved reaction reversibility and accelerated redox kinetics^[36].

To understand how the BiCl₃ additive affects the full-cell kinetic characteristics, the rate capability of the full cells was measured [Figure 6E]. At current densities of 20, 50, 100, and 200 mA g^-1, the Mn||G2-BiCl₃||Cu_1.8S cell demonstrates outstanding discharge capacities of 245.4, 193.6, 152.0, and 112.1 mAh g^-1, respectively. By comparison, the full cell with the G2-Mn(OTf)₂ electrolyte exhibits much lower capacities of 113.7, 77.1, 42.9, and 29.7 mAh g^-1 at the same current densities. Further, by increasing the current density to 200 mA g^-1 for long-term cycling stability testing [Supplementary Figure 15], the Mn||G2-BiCl₃||Cu_1.8S full cell delivered an average specific capacity of 48.3 mAh g^-1 over 800 cycles. In contrast, the cell with the G2-Mn(OTf)₂ electrolyte exhibited almost no measurable capacity under these high-current-density conditions. Furthermore, the first-cycle charge/discharge profiles at different current densities were analyzed [Figure 6F], demonstrating the stable charge/discharge behavior and good reversibility of the Mn||G2-BiCl₃||Cu_1.8S full cell.

To quantify the charge-storage mechanism, a CV was carried out as a function of scan rate. The b-values (i = av^b) obtained for both electrolytes were found to be in the range 0.59-0.66, indicating that the electrochemical process is predominantly diffusion-controlled Figure 7). The calculated values of diffusion coefficients (D) for the G2-BiCl₃ system were significantly larger than those for the baseline, indicating that the Bi-containing interphase facilitates Mn²⁺ transport and charge transfer across the Mn/electrolyte interface. These results demonstrate that BiCl₃ effectively reduces interfacial kinetic limitations and enhances the electrochemical reaction kinetics of the Mn||Cu_1.8S full cell.

The schematic diagram of the Mn||Cu_1.8S cell structure and its operation mechanism is shown in Figure 8A. At the anode, the Mn atoms are oxidized and enter the electrolyte in the form of ions Mn²⁺. The electrons from the oxidation process at the anode travel to the cathode through the external circuit. At the cathode, Mn²⁺ and electrons react with Cu_1.8S to yield MnS and metallic Cu via a displacement reaction. The charge process proceeds through the corresponding reverse electrochemical pathways.

Ex-situ XRD measurements of the cathodes were conducted at different electrochemical states during the first electrochemical cycle to track the phase evolution during cycling and to elucidate the role of BiCl₃ in cathode reactions and the resulting improvement in full-cell kinetics. Figure 8B and Supplementary Figure 16 show GCD curves of both cells with the two electrolytes, with the three sampling points of pristine (P), fully discharged (D), and fully charged (C). The ex-situ XRD results of the pristine state (P) indicate that the diffraction peaks of the cathode in both electrolytes match the Cu_1.8S standard card [Figure 8C, Supplementary Figure 17A].

The phase evolution pathways of the two electrolytes are significantly different when the state of charge is fully discharged (D). The sluggish kinetics of the Mn anode cause severe polarization in the G2-Mn(OTf)₂ electrolyte, leading to the rapid onset of the cut-off condition and therefore premature end of the discharge process, as shown by the presence of the Cu_1.8S phases in the XRD pattern [Supplementary Figure 17B]. The latter G2-BiCl₃ electrolyte, in comparison, allows for the full conversion of the Cu_1.8S cathode, with all of the primary diffraction peaks from Cu_1.8S fully disappearing [Figure 8D]. At the same time, a comparison of the diffraction profiles in the fully discharged state (D) also shows a significant difference in the formation of the MnS reaction product. The characteristic peaks of the MnS phase are, in particular, more intense and well resolved for the G2-BiCl₃ electrolyte when compared to the G2-Mn(OTf)₂ electrolyte, which means that the transformation is more complete.

The difference in phase reversibility remains very large under fully charged conditions (C). The diffraction peaks corresponding to the discharge products MnS disappear entirely in the cell with the G2-BiCl₃ electrolyte, while the diffraction peaks of the Cu_1.8S phase are completely recovered and restored [Figure 8E]. In contrast, the fully reconstructed Cu_1.8S cathode in the baseline G2-Mn(OTf)₂ system was only partially reconstructed under standard conditions. The complete recovery of Cu_1.8S in the electrolyte G2-Mn(OTf)₂ can only be realized when the charge voltage is raised to 1.3 V [Supplementary Figure 17C and D]. This requires a higher overpotential due to sluggish Mn anode interfacial kinetics, which hinder efficient Mn deposition/stripping processes. To further verify the different conversion behaviors, ex situ XPS analyses were also performed on the cathodes at the pristine (P), fully discharged (D), and fully charged (C) states [Figure 8F-H, Supplementary Figure 17E-G]. Notably, in the fully discharged state, the Cu 2p spectrum of the G2-Mn(OTf)₂ control still exhibits a detectable Cu²⁺ signal, indicating incomplete conversion of Cu_1.8S. In contrast, the Cu 2p spectrum of the G2-BiCl₃ cathode shows no detectable Cu²⁺ signal, with the Cu species predominantly converted to metallic Cu⁰. These results provide additional chemical-state evidence for the more complete conversion of Cu_1.8S in the G2-BiCl₃ electrolyte. Collectively, the XRD and XPS results indicate that the BiCl₃ additive regulates the Mn anode interface, reduces the interfacial kinetic polarization, and thereby facilitates more complete and reversible Cu_1.8S conversion within the normal operating voltage range.