Electrochemical Conversion of CO₂ via Molten Salt Electrolysis to Produce Hard Carbon Materials for Sodium-Ion Battery

Abstract

The conversion of CO₂ into valuable carbon materials is essential for establishing a circular economy. This study syntheses hard carbon materials from CO₂ using high-temperature molten salt electrolysis and explores their use as anodes in sodium-ion batteries. CO₂ gases are absorbed as carbonate ions in molten salts and then electrochemically decomposed into solid carbon. When electrolysis was conducted at 600 °C in molten LiCl-KCl containing Li₂CO₃, the generated carbon was amorphous and nongraphitizable, as determined by XRD analysis and Raman spectroscopy. The obtained carbons before and after heat treatments at 900–1800 °C were employed as the anode materials in two-electrode cells with metallic sodium counter electrodes and 1 M NaPF₆/EC-DMC electrolyte. for sodium ion battery. The carbon treated at 900 °C showed the highest capacity, exceeding that of commercial hard carbon materials. These results demonstrate the potential of CO₂ reduction and technologies for converting CO₂ into valuable carbon materials for energy storage devices.

1. Introduction

CO₂ emissions are a major cause of global warming. Therefore, there is considerable interest in technologies for its capture and utilization. One promising option is molten salt electrolysis, which can decompose CO₂ into solid carbon and O₂ gas. − During this process, CO₂ reacts with O^2– in molten salts to form CO₃ ^2– (eq ), which is then electrochemically reduced at the cathode to form solid carbon and O^2– according to eq . Meanwhile, the generated O^2– is converted to O2 gas at the anode (eq ). Several anode materials, including Pt, Au, B-doped diamond, and semiconducting ceramics ,, have been explored for this reaction.

$${\mathrm{CO}}_2(\mathrm{g})+{\mathrm{O}}^{2-}\rightleftharpoons {\mathrm{CO}}_3^{2-}$$ 1

$${\mathrm{CO}}_3^{2-}+4{\mathrm{e}}^{-}\rightleftharpoons \mathrm{C}(\mathrm{s})+3{\mathrm{O}}^{2-}$$ 2

$${\mathrm{O}}^{2-}\rightleftharpoons {\mathrm{O}}_2(\mathrm{g})+4{\mathrm{e}}^{-}$$ 3

Molten salt electrolysis is a particularly important CO₂ decomposition technology because it has the potential to produce valuable carbon materials, including spherical carbon, , carbon nanotube/nanofiber, − porous/hollow carbon, , graphene, , and highly graphitic carbon. , These carbon materials are chemically stable and have high electrical conductivity, making them effective electrode materials for energy storage devices such as Li-ion batteries, which are essential for achieving sustainable society. , Several researchers have reported on the performance of molten salt electrolysis-derived carbon materials as anode materials for Li-ion batteries. ,,, However, the growing demand for Li-based energy storage systems has raised concerns about the depletion of Li resources, which are limited and unevenly distributed. Na has emerged as a promising alternative owing to its cost-effectiveness and abundance (its Clarke number is 500 times higher than that of Li). , Therefore, the use of molten salt electrolysis for producing carbon-based electrode materials for Na-ion batteries is of high importance.

Conventional hard carbon anodes for Na-ion batteries are commonly derived from oil or coal. − However, from the perspective of CO₂ emission reduction and increasing regulatory pressure, the development of hard carbons from sustainable resources has become increasingly important. Accordingly, a wide variety of biomass-derived hard carbons have been investigated in recent years. In parallel, the reutilization of CO₂ as a carbon source has attracted as a complementary strategy for sustainable carbon production. In particular, the synthesis of hard carbon via molten-salt electrolysis enables the direct conversion of CO₂ into functional carbon materials, offering a fundamentally different pathway from biomass-based approaches. Although CO₂-derived carbon materials have been increasingly studied for Li-ion batteries, their application in Na-ion batteries remains very limited. Licht et al. produced tangled carbon nanotubes via molten salt electrolysis at 750 °C for use as anode materials in Na-ion batteries; however the capacity was low, requiring significant improvement. Thapaliya et al. reported the conversion of CO₂ to porous carbon materials via electrochemical reduction in molten carbonates at temperatures of 450 to 550 °C. The porous carbon exhibited capacitive behavior and excellent rate capability; however, its high specific surface area resulted in a large irreversible capacity and low initial Coulombic efficiency. Consequently, although the carbon materials generated from molten salt electrolysis has the potential for use in Na-ion batteries, further research is required to obtain electrochemically useful materials.

In this study, we examined the Na-ion storage performance of hard carbon materials created through molten salt electrolysis and heat treatment at 900–1800 °C. The effect of heat treatment on the generated carbon was evaluated by morphological and structural analyses. The electrochemical performance was then assessed in coin-type cells with a two-electrode configuration. This study demonstrates that molten salt electrolysis with postheat treatment can be used to convert CO₂ into valuable carbon materials for Na-ion batteries.

2. Experimental Section

2.1. Production of Carbon Materials via Molten Salt Electrolysis

The electrolyte for molten salt electrolysis was a mixture (75:25 mol %) of LiCl (>99%, Fujifilm Wako Pure Chemical Co.) and KCl (>99.9%, Fujifilm Wako Pure Chemical Co.) and it was used in a high-purity MgO crucible after vacuum drying overnight at 200 °C. Cyclic voltammetry was conducted at 600 °C in the melt containing 4 mol % of Li₂O (>95%, Fujifilm Wako Pure Chemical Co.) to confirm the incorporation of CO₂ before and after 50% CO₂/Ar gas mixture was bubbled through the melt for 2.5 h at a flow rate of 20 cm³ s^–1. Next, potentiostatic electrolysis was performed at 600 °C to generate carbon materials for the measurement of Na-ion storage performance. To ensure repeatability, a standardized melt containing 3 mol % of Li₂O and Li₂CO₃ (>99%, Fujifilm Wako Pure Chemical Co.) was employed. Subsequently, the generated carbon materials were ultrasonicated several times in pure water and 10% HCl to remove impurities derived from the molten salts, followed by drying at 150 °C. The experimental apparatus used for the electrochemical measurements is shown in Figure . The working electrode for cyclic voltammetry was a Mo wire, whereas the for potentiostatic electrolysis was a Mo plate. A glassy carbon counter electrode and Ag⁺/Ag reference electrode were used for both procedures. The Ag⁺/Ag reference electrode was prepared by immersing Ag wire in a melt with 0.5 mol % AgCl inside a mullite tube sealed at one end. Electrochemical measurements were performed using an electrochemical measurement system SP-150, under Ar atmosphere, and the measured potentials were adjusted to Li⁺/Li redox potentials.

1.

Experimental apparatus of the electrochemical measurement for molten salt electrolysis (W.E.: working electrode, C.E.: counter electrode, R.E.: Reference electrode).

2.2. Heat Treatment and Characterizations of Carbon Materials

The solid carbon obtained from electrolysis was heat treated for 1 h under an Ar atmosphere at 900 °C in a horizontal tube furnace. Heat treatments at 1200, 1500, or 1800 °C were conducted using an induction heating furnace. The morphologies of the carbon materials were evaluated using scanning electron microscope (SEM; S-4700, Hitachi, Japan), and elemental analysis was performed using SEM-energy-dispersive X-ray spectroscopy (EDS). The carbon structures were analyzed using X-ray diffraction (XRD; D8 Advance, Bruker, Billerica, MA) with Cu Kα radiation. Raman spectra were collected using a Raman microscope (inVia, Renishaw, Wotton-under-Edge, U.K.) with Ar-ion laser excitation at 514.5 nm. The Brunauer–Emmett–Teller (BET) specific surface areas were estimated from N₂ adsorption isotherms obtained at 77 K using a surface area and porosity analyzer (Autosorb-iQ, Quantachrome Instruments Co.).

2.3. Electrochemical Evaluation for Na-Ion Storage Performance

After mixing the carbon powders with 5 wt % carboxymethyl cellulose (D2200, Daicel FineChem Ltd.) and 5 wt % styrene–butadiene rubber (BM-400B, Zeon Co.) in distilled water, the resulting slurry was coated on an Al current collector (20 μm thick) using the doctor blade method. The carbon electrodes had diameters of 10 mm and thicknesses of approximately 0.05 mm. The electrodes were dried overnight under vacuum at 150 °C, along with glass fiber filters (GA-55, Toyo Roshi Kaisha, Ltd.), which were used as separators. Subsequently, the electrodes and separators were immersed for 30 min in ethylene carbonate and dimethyl carbonate containing 1 M sodium hexafluorophosphate (1 M NaPF₆/EC-DMC, Kishida Chemical Co., Ltd.). Finally, coin-type cells in a two-electrode configuration with carbon electrodes and Na metal. Electrochemical measurements were conducted using a BioLogic VMP3 multichannel galvanostat–potentiostat. Charge/discharge tests were performed at room temperature within a voltage range of 2.00 to 0.00 V. The discharge (oxidative) capacity was estimated based on the weight of the active materials.

3. Results and Discussion

3.1. Production and Postelectrolysis Heat Treatment of Carbon Materials

Figure shows the cyclic voltammograms of the Mo electrode before and after the introduction of CO₂ gas in molten LiCl-KCl-Li₂O at 600 °C. Before adding CO₂ gas (Figure a), the cathodic (A) and the corresponding anodic (A′) peaks were observed at approximately 0 V vs Li⁺/Li. These peaks were attributed to the deposition and dissolution, respectively, of metallic Li on the Mo electrode. After CO₂ was introduced, current densities (D) and (D′) were observed due to Li deposition and dissolution, respectively. In addition, new cathodic peaks (B and C) appeared at higher potentials. This indicates that CO₂ was absorbed as CO₃ ^2–in the molten salt. Current density (B) is assumed to carbon deposition, as expressed by eq . Specifically, the literature , links current densities (B) and (C) to the reactions in eqs and , respectively. Another potential explanation for peak (B) is the formation of an intercalation compound between carbon and Li.

$${\mathrm{CO}}_3^{2-}+2{\mathrm{e}}^{-}={\mathrm{CO}}_2^{2-}+{\mathrm{O}}^{2-}$$ 4

$${\mathrm{CO}}_2^{2-}+2{\mathrm{e}}^{-}=\mathrm{C}(\mathrm{s})+2{\mathrm{O}}^{2-}$$ 5

2.

Cyclic voltammograms of the Mo electrode in molten LiC-KCl (a) before and (b) after CO₂ gas bubbling.

Next, we conducted potentiostatic electrolysis for several hours at 0.7 V vs Li⁺/Li in molten LiCl-KCl-Li₂O–Li₂O₃. The current efficiency was nearly 100% based on the weight of the deposited carbon particles. Figure illustrates the morphology of the carbon material deposited during electrolysis at 600 °C, as well as those obtained after heat treatment at 900, 1200, 1500, and 1800 °C. Spherical particles were predominantly observed, with no significant difference in their size or shape before and after heat treatment. EDS elemental analysis (Table ) revealed that the as-electrolyzed carbon contained some oxygen functional groups and impurities derived from molten salts such as K and Cl. Nevertheless, the purity was significantly enhanced through the subsequent heat treatment.

3.

SEM images of the carbons: (a) as-electrolyzed, (b–e) obtained after heat treatment at 900, 1200, 1500, and 1800 °C.

1. EDX Analysis of the Carbons before and after Heated at 900, 1200, 1500, and 1800 °C.

heating temp./°C	C/at%	O/at%	others/at%	O/C ratio
600 °C (As-electrolyzed)	91.7	7.0	1.3	0.077
900	94.8	4.5	0.7	0.047
1200	96.3	3.3	0.1	0.034
1500	96.0	4.0	0	0.047
1800	96.4	3.6	0	0.037

Figure shows the XRD patterns of the (002) diffraction peak, corresponding to the stacking of hexagonal carbon layers. The crystallinity improved with increasing heat treatment temperature. At 1800 °C, two distinct sharp peaks emerged, which were attributed to turbostratic and graphitic carbon. This phenomenon is typical of hard carbon materials and is known as multiphase graphitization. The Raman spectra indicate a similar trend in crystallinity (Figure ). The peaks at approximately 1350 and 1580 cm^–1 correspond to the D and G bands, respectively. The full width at half-maximum of the G band decreased with increasing temperature, indicating the development of a graphitic structure. The intensity of the D band increased at 1200 and 1500 °C, which aligns with previous reports on the heat treatment of carbon in this temperature range. −

4.

XRD pattern of the carbons: (a) as-electrolyzed, (b)­(c)­(d)­(e) obtained after heat treatment at 900, 1200, 1500, and 1800 °C.

5.

Raman spectra of the carbons: (a) as-electrolyzed, (b–e) obtained after heat treatment at 900, 1200, 1500, and 1800 °C.

3.2. Electrochemical Storage Behavior of Carbon Materials in Na-Ion Batteries

The generated carbon materials, before and after heat treatment, were investigated as anode materials for Na-ion batteries. Figure shows the cyclic voltammograms of the carbon electrodes in 1 M NaPF₆/EC-DMC electrolyte. Reduction and oxidation peaks were observed at approximately 0.1 V, which were attributed to the insertion/deinsertion, respectively, of Na ions in the carbon layers. These peaks increased in size after heat treatment at 900 °C; however, treatment at higher temperatures reduced the peak size. Similar results were observed in the charge/discharge curves, as shown in Figure . The potential profiles contained a slope above 0.1 V and a plateau region from 0.1 to 0 V, which corresponds to the cyclic voltammetry results. Hard carbon materials can accommodate Na ions in the interlayer space between carbon layers and within closed nanovoids. − The former takes place within a voltage range of 1.0–0.1 V, while the latter occurs at approximately 0.1–0 V. It is speculated from these results that heat treatment at 900 °C increases the number of Na-ion insertion sites in both the interlayer spaces and nanovoids surrounded by carbon layers with locally short-range order. The discharge capacity of the carbon materials treated at at 900 °C reached a maximum of 240 mAh g^–1, which exceeded that of a commercially available hard carbon material (Figures S1 and S2). After heat treatment at higher temperatures, the plateau region at 0.1 to 0 V was extended but the slope region above 0.1 V was narrowed, resulting in the decreased total discharge capacity. This was attributed to the formation of more graphitic regions in the carbon structure, which provide limited storage for Na ions. ,

6.

Cyclic voltammograms (fifth cycle) in 1 M NaPF₆/EC-DMC electrolyte for the carbons (a) as-electrolyzed, (b–e) heated at 900, 1200, 1500, and 1800 °C, respectively. The scan rate is 0.1 mVs^–1.

7.

Charge/discharge curves (10th cycle) acquired at 25 mAg^–1 in 1 M NaPF₆/EC-DMC electrolyte for the carbons: (a) as-electrolyzed, (b–e) heated at 900, 1200, 1500, and 1800 °C, respectively.

The initial Coulombic efficiencies of the as-electrolyzed carbons and 900, 1200, 1500, and 1800 °C heat-treated carbons were 67, 75, 80, 84, and 83%, respectively. The as-electrolyzed carbon therefore had the lowest initial Coulombic efficiency. This aligns with its high specific surface area (230 m² g^–1; Table S1), which was significantly higher those of the heat-treated carbon materials. The irreversible capacity is believed to originate from the formation of a solid electrolyte interface layer via electrolyte decomposition. − Therefore, because heat treatment significantly reduced its specific surface area of the carbon material (Table S1), the irreversible capacity was reduced. Another possibility is that defects in the as-electrolyzed carbon contributed to irreversible Na-ion trapping. , Heat treatment at higher temperatures improved the crystallinity of the carbon material, thereby enhancing the initial Coulombic efficiency. Figure illustrates the rate capability of the carbon electrodes. The carbon treated at 900 °C showed the highest capacity at low current densities. However, the rate capability of the as-electrolyzed carbon surpassed that of the carbon treated at 900 °C at current densities at 1000 mA g^–1. This would be because Na ions storage in the interlayer space dominate the capacity of the as-electrolyzed carbons, and the migration of Na ion in the interlayer space between carbon layers was faster than within nanovoids. Another possible contribution to the good rate capability is the high electric double-layer capacitance of the as-electrolyzed carbon owing to its high specific surface area. , This interpretation is supported by the more rectangular shape observed in the CV curves for the carbons treated at lower temperatures. Notably, the rate performance of the carbon treated at 900 °C exceeded that of a commercially available hard carbon material (Figure S3). In addition, Figure illustrates the cycling performance of the carbon material treated at 900 °C, which retained approximately 70% of its initial capacity after 200 cycles. Thus, although electrolysis and heat-treatment conditions require further optimization, this study suggests that carbon derived from CO₂ electrolysis with postheat treatment is promising for application in the anodes of Na-ion batteries.

8.

Discharge (oxidative) capacity acquired at various current densities in a potential range of 2–0 V vs Na⁺/Na in 1 M NaPF₆/EC-DMC electrolyte for the carbons: (a) as-electrolyzed, (b–e) heated at 900, 1200, 1500, and 1800 °C, respectively.

9.

Cyclability test at 0.2 A g^–1 in a potential range of 2–0 V vs Na⁺/Na in 1 M NaPF₆/EC-DMC electrolyte for the carbons heated at 900 °C.

4. Conclusions

This study demonstrates the capture and decomposition of CO₂ via electrochemical reactions in molten salts, as well as the application of the obtained carbon in Na-ion batteries. The as-electrolyzed carbon was amorphous, and its crystallinity was improved by heat treatment. XRD analysis showed that multiphase graphitization, a typical trait of hard carbons, occurred upon heating at 1800 °C. The carbon treated at 900 °C had the highest discharge capacity as an anode material in Na-ion battery, exceeding that of a commercially available hard carbon material. The initial Coulombic efficiency was improved by heat treatment, likely because of the reduction in specific surface area and the improved crystallinity of carbon. This study demonstrates that molten salt electrolysis can be used to convert CO₂ into valuable carbon materials for applications like Na-ion battery anodes. Although further investigation with full-cell testing in combination with suitable cathode materials is required for practical applications, this work expands the possibilities of molten salt electrolysis for the capture and utilization of CO₂, which will aid the development of integrated and sustainable CO₂ mitigation strategies.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.6c00423.

• Additional tables and figures for complementary experiments described in the manuscript; initial coulombic efficiency and BET surface area of the carbons; Charge/discharge curves (10th cycle) acquired at 25 mAg–1 in 1M NaPF6/ECDMCelectrolyte for a commercially available hard carbon material (PDF)

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.