Alkali Metal Ion Substituted Carboxymethyl Cellulose as Anode Polymeric Binders for Rapidly Chargeable Lithium‐Ion Batteries

The increasing demand for short charging time on electric vehicles has motivated realization of fast chargeable lithium‐ion batteries (LIBs). However, shortening the charging time of LIBs is limited by Li+ intercalation process consisting of liquid‐phase diffusion, de‐solvation, SEI crossing, and solid‐phase diffusion. Herein, we propose a new strategy to accelerate the de‐solvation step through a control of interaction between polymeric binder and solvent‐Li+ complexes. For this purpose, three alkali metal ions (Li+, Na+, and K+) substituted carboxymethyl cellulose (Li‐, Na‐, and K‐CMC) are prepared to examine the effects of metal ions on their performances. The lowest activation energy of de‐solvation and the highest chemical diffusion coefficient were observed for Li‐CMC. Specifically, Li‐CMC cell with a capacity of 3 mAh cm−2 could be charged to >95% in 10 min, while a value above >85% was observed after 150 cycles. Thus, the presented approach holds great promise for the realization of fast charging.


Introduction
The demand for electric vehicles with a higher mileage per charge and a shorter charging time necessitates the development of lithium-ion batteries (LIBs) with increased energy density and charging rate, with the respective target values set as >275 Wh kg À1 and >80% capacity within 15 min by the US Department of Energy. [1]However, given that the time required to fuel internal-combustion-engine vehicles does not exceed 5 min, even a 15 min charging time may be overly long.[26] Despite the numerous works on the control of these two factors, the limits imposed by conventional active materials and electrolytes have not been lifted yet.The rate with which Li + are intercalated and diffuse through the active material depends not only on intrinsic kinetic transition barriers but also on the local Li + concentration at the outer SEI layer.The liquid phase mass transport and de-solvation process and its accumulation at the interface are thus quite influential steps, albeit much less studied in the past. [2,3]n this work, we show that these steps are indeed not only a function of the electrolyte but also the cation identity in the binder material.
The increasing demand for short charging time on electric vehicles has motivated realization of fast chargeable lithium-ion batteries (LIBs).However, shortening the charging time of LIBs is limited by Li + intercalation process consisting of liquid-phase diffusion, de-solvation, SEI crossing, and solid-phase diffusion.Herein, we propose a new strategy to accelerate the de-solvation step through a control of interaction between polymeric binder and solvent-Li + complexes.For this purpose, three alkali metal ions (Li + , Na + , and K + ) substituted carboxymethyl cellulose (Li-, Na-, and K-CMC) are prepared to examine the effects of metal ions on their performances.The lowest activation energy of de-solvation and the highest chemical diffusion coefficient were observed for Li-CMC.Specifically, Li-CMC cell with a capacity of 3 mAh cm À2 could be charged to >95% in 10 min, while a value above >85% was observed after 150 cycles.Thus, the presented approach holds great promise for the realization of fast charging.
From experiments and multi-scale simulations, we show the change of the binder cation from K + to Li + to lead to increased concentration of Li + -ethylene carbonate (EC) coordinating complexes at the interface enhancing the driving force for intercalation, the diffusion coefficient, and the charging rate.29][30][31][32] Although these studies also utilized the Li-CMC binder that was providing additional Li + transport channels through the binder, this is not the same as a concept for lowering de-solvation energy barrier by Li + substituents in the binder proposed in this study.Moreover, we proved the interactions between electrolytes and structural characteristics of the binder facilitate the step of the Li + de-solvation from the solvated EC-Li + complexes by experimental and computational modeling/ simulation methods and this will help to step forward to achieve the fast-charging lithium ion batteries.

Significance of Binder for Li + Intercalation
For the electrode design employed herein, ~30% of the active material surface area is covered by inactive polymeric binders even when their loading is as low as 3 wt%, that is, when the graphite: conductor: binder weight ratio equals 94:3:3 (Figure 2).Due to this situation, the intercalation of a large fraction of Li + might be substantially influenced by the presence of the polymeric binder material.In addition, the consideration of the substantial amount of liquid electrolyte that must be impregnated into the amorphous binder regions, [33,34] increases the significance of the Li + accumulation and de-solvation processes.Indeed, several studies have emphasized that the de-solvation energy barrier has a larger effect on electrode kinetics than cell design. [2,6]In conclusion, the functionalization of the binder material might provide a promising way to alter the intercalation process.

Insights from Multi-Scale Simulations
To follow this idea, we first attempted to change the cation identity within the binder.Inspired by the fact that the interaction of CMC with organic solvents becomes stronger with decreasing substituted cation size, [35] we herein chose CMC as a simple but efficient binder and prepared Li + , Na + , and K + substituted binders (Li-, Na-, and K-CMC, respectively) using a well-established titration method (Figure 3a and Figures S1-S3, Supporting Information). [36,37]We first performed molecular dynamics (MD) simulations to gain insights into the distribution of electrolyte components close to the interface.As shown in Figure 3b, we found interfacial Li + in the Li-CMC binder case to be coordinated by about three times as many EC molecules compared to the bulk electrolyte.In contrast, we observed significantly reduced interfacial Li + coordination numbers in the case of Na-CMC and K-CMC.Similar trends have been found before by Pham et al. [38] Moreover, we found the Li + concentration at the graphite-binder interface to be higher for Li-CMC than for Na-and K-CMC, likely due to the stronger interaction of Li-CMC and EC (Figure 3c).The simulation results thus suggest that Li-CMC is substantially more wetted than the other binder materials, significantly increasing the local Li + concentration.
To evaluate the effect of the accumulated Li + concentration, we built digital-twin  graphite electrode models (Figure S7, Supporting Information) differing only in the binder type (Li-, Na-, or K-CMC) and examined their behavior during charging at 4 C (Figure 3d).Despite the small binder content of 3 wt%, the charge capacity of the Li-CMC electrode is 63.9%, exceeding those of Na-CMC (62.5%) and K-CMC (59.6%).To visualize this difference more clearly, we constructed three color maps of Li + concentration at the same voltage (4.2 V in the full cell of LiNi 0.6 Mn 0.2 Co 0.2 O 2 /graphite; 3 mAh cm À2 , N/P ratio = 1.15), as shown in Figure 3e-g.As the 3D electrode structure was held constant, we could effectively compare both the overall Li + concentration and the interfacial Li + concentration near the binder.As similar to the results of Figure 3c and Figure 3d, the Li-CMC electrode showed an increased Li + concentration at the interface the binder as well as in the bulk.

Activation Energies from Impedance Spectroscopy
To experimentally confirm the significance of the binder-solvent interaction for Li + accumulation and de-solvation, we fabricated a Li/CMC binder film/Li cell (Figure 4a) and examined the motion of Li + at the interface between Li metal and the binder film using electrochemical impedance spectroscopy (EIS).As reported elsewhere, the temperature dependence of impedance can be used to obtain the activation energies of SEI crossing and de-solvation steps (Figure 4b). [11,34,39,40]Herein, raw impedance spectra measured at different temperatures were fitted using the equivalent circuit depicted in Fig- ure 4b, and the temperature dependence of two resistance factors, R SEI and R de-sol (Figure 4c), was examined to determine the activation energies of SEI crossing and de-solvation.As shown in Fig- ure 4d, the activation energies of SEI crossing were almost identical (Li-CMC: 353.75 kJ mol À1 , Na-CMC: 353.74 kJ mol À1 , and K-CMC: 353.73 kJ mol À1 ), whereas those of de-solvation exhibited marked differences (Li-CMC: 242.05 kJ mol À1 , Na-CMC: 319.46 kJ mol À1 , and K-CMC: 322.41 kJ mol À1 ).In particular, the much lower activation energy of desolvation observed for Li-CMC could greatly support higher Li + concentrations and stronger binder-solvent interactions near the active material surface.

Li + Diffusion Coefficients
Based on the understanding of binder-electrolyte interactions, we determined the chemical Li + diffusion coefficients by cyclic voltammetry (CV) (Figure 5a), EIS (Figure 5b), and galvanostatic intermittent titration technique (GITT) (Figure 5c) using a symmetric cell with a graphite anode (graphite: carbon conductor: CMC binder: SBR binder = 94:3:1.5:1.5, w/w/w/w) and 1 M LiPF 6 in EC:ethyl methyl carbonate (EMC) (3:7, v/v) as the electrolyte.Unlike intrinsic diffusion coefficients, which depend on the active solid material, chemical diffusion coefficients depend not only on the solid-state diffusion coefficient of the active material but also on factors such as SEI crossing and de-solvation at the electrode interface. [41]However, as we used the same active material and confirmed that the difference in activation energy for SEI crossing was negligibly affected by CMC binder type, the chemical diffusion coefficients were mainly governed by the CMC type-dependent de-solvation step.Despite the significant difference between the chemical diffusion coefficients obtained by Energy Environ.Mater.2024, 7, e12509 different methods, they were always highest for the Li-CMC electrode.Thus, the simple exchange of cations in CMC from Na + to Li + was concluded to strongly facilitate rapid charging.In contrast, the graphite anode using commercial PVdF binder shows the lowest chemical diffusion coefficient as shown in Figure S9, Supporting Information.Because both electrons and Li + movement distance become longer as reported previously due to the higher wettability and swelling behavior of PVdF binder.

Experimental Charging Behavior
To experimentally confirm the effect of alkali metal ions in CMC, we compared the fast charging behaviors of coin-type full cells (LiNi 0.6 Mn 0.2 Co 0.2 O 2 /graphite, designed areal capacity = 3 mAh cm À2 ) based on the two charging protocols described in the insets of Figure 6a-1,b-1 and corresponding to 15 min (4 C) and 10 min (6 C) charging, respectively.According to Figure 6a-2, the Li-CMC cell required the longest time to reach a voltage of 4.3 V, at which charging started in the CV mode, which indicated that Li + easily intercalated into the graphite anode with the help of Li + in the binder.Specifically, the Li-CMC cell could be charged to 95.7% and 90.1% of the designed capacity even at 4 and 6 C, respectively, while lower charging efficiencies were observed for Na-CMC (88.8% and 87.5%, respectively) and K-CMC (85.2% and 81.3%, respectively), as expected from the results of theoretical calculations, activation energies, and chemical diffusion coefficients.To support the above results, we further examined lithiation/de-lithiation resistances and cell DC-IRs, revealing that the lowest values were observed for Li-CMC in both cases (Figures S4 and S5, Supporting Information).The effects of substituted cation type within CMC were further examined by characterization of two highly loaded graphite anodes (>6 mAh cm À2 ).As shown in Figure S6, Supporting Information, Li-CMC enabled the intercalation of more Li ions into the highly loaded graphite anode than Na-CMC (by 4%), which is a remarkable achievement given the small binder loading of 3 wt% and the fact that simple cation replacement was used.Furthermore, as shown in Figure S10, Supporting Information, Li-CMC shows significant development in capacity retention compared with commercial PVdF binder based one in 4C rate cycle.This ability of Li-CMC to promote fast charging was also held responsible for superior cycling performance in both 4-and 6-C charging tests (Figure 6a-3,b-3, respectively).In the case of charging at 4 C (12 mA cm À2 ), the Li-CMC cell maintained its initial capacity after 200 cycles with high and stable Coulombic efficiencies, while the Na-CMC cell featured a lower capacity retention of 94.5%, and the K-CMC cell featured a significantly decreased capacity retention of 62.3% with slightly unstable and low Coulombic efficiencies.When the charging rate increased to 6 C (18 mA cm À2 ), the Li-CMC cell maintained >80% of the initial capacity after 150 cycles, while Na-CMC and K-CMC cells retained <50% and 30% of their initial capacities, respectively, and exhibited strong Coulombic efficiency fluctuations, which indicated the occurrence of irreversible reactions.Even though the capacity of the Li-CMC cell was well maintained at ~80% compared to the initial capacity, the electrode surface was significantly covered by dead Li as shown in Figure S11, Supporting Information.According to the case study of the fast-charging experiment reported by Dasgupta et al., [42] they have used the condition of 4 C charge/ 1 C discharge with the graphite/hard carbon blend anode, resulting in ~85% of initial capacity after 100 cycles.Although this test showed high capacity retention over 85% after 100 cycles as similar to our study, a significant amount of dead Li formed was observed on the electrode surface and this implies that the dead Li formation seems to be likely happened during fast-charging experiments.For this time, we analyze the surface properties of Li-CMC electrodes only, but we expect the electrode employing Na-and K- CMC has bigger and larger amounts of dead Li formed than that observed from Li-CMC electrodes.
[45][46] As shown in Figure S12, Supporting Information, each of graphite electrode showed a similar adhesion (dry electrode: ~40 g f /cm & wet electrode: ~32 g f /cm) and cohesion strength (dry electrode: ~22 g f /cm & wet electrode: ~18 g f /cm), implying that the cation exchanging in the CMC binder was not helping to improve the mechanical strength of battery electrode.Thus, the improvement of battery performance observed on Li-CMC cell was likely derived from the chemical interaction between binder and electrolyte.
Above-mentioned Li-CMC effect was also investigated by testing a single unit pouch cells (~85 mAh).As shown in Figure S13, Supporting Information, the Li-CMC effects on the fast charging increase as the total area of the graphite anode increases.Thus, Li-CMC was concluded to play a pivotal role in facilitating Li + intercalation at the graphite-binder interface by lowering the activation energy of Li + de-solvation or accelerating the chemical diffusion of Li + .

Conclusion
In reality, cases when a single ground-breaking discovery changes the game rules are rare, and most commonly, challenging but achievable goals such as high-energy-density LIBs capable of ultrafast charging (within <5 min) are realized by synergistically combining small but numerous improvements.Herein, we report how the exchange of cations in a CMC binder can improve the fast charging capability of highly loaded graphite anodes for LIBs.In particular, the theoretical multiscale modeling of the graphite-binder interface and the analysis of experimentally obtained desolvation activation energies and chemical diffusion coefficients support this superior enhancement.These results suggest a so far not considered importance of the choice of binder material for fast charging and gives first microscopic insights which will pave the way towards more controlled battery design.In combination with other improvements for fast charging such changes in composition, dopant, primary/secondary particle size, electron conductor, binder distribution, etc., the development of high-performance batteries is surely in closer reach.

Figure 1 .
Figure 1.Illustration of processes involved in the intercalation of Li + into a graphite anode and the corresponding energy landscape.

Figure 2 .
Figure 2. a) Three-dimensional (3D) representation of a digital-twin electrode comprising graphite, conductive carbon, and binder (e.g., CMC).b) Simulated contact areas of graphite with the binder and the liquid electrolyte.c) Expected effect of CMC modification on the surface concentration of Li + and their intercalation.

Figure 3 .
Figure 3. a) Molecular structures of Li-, Na-, and K-CMC.b) Relative EC coordination numbers of Li-, Na-, and K-CMC and c) corresponding interfacial and bulk Li + concentrations calculated using molecular dynamics simulations.d) Charge curves simulated for a rate of 4 C using 3D digital-twin modeling.3D color maps of lithium concentration in a charged (4.2 V) graphite electrode with e) Li-CMC, f) Na-CMC, and g) K-CMC binders.

Figure 4 .
Figure 4. a) Illustration of cell configuration and processes corresponding to different impedance spectrum regions.b) Temperature-dependent impedance spectra recorded for Li/(Li-, Na-, or K-CMC)/Li cells.c) Arrhenius behavior of the temperature dependence of R SEI and R de-sol .d) Activation energies of SEI crossing and de-solvation obtained by analysis of the plots in c).

Figure 5 .
Figure 5. a) Peak current versus square root of scan rate (m 1/2 ) plots and chemical Li + diffusion coefficients determined from the same.b) Nyquist plots of graphite/graphite symmetric cells and chemical Li + diffusion coefficients determined from the same.c) GITT profiles and chemical Li + diffusion coefficients determined from the same.

Figure 6 .
Figure 6.a) 15-and b) 10-min fast charging performances.a-1) 15-and b-1) 10-min charge protocols used to evaluate fast charging performance.Voltage profiles and charge capacities determined in CV mode under a-2) 15-and b-2) 10-min fast charging conditions.Effects of cycling on capacity retention and Coulombic efficiency under a-3) 15-and b-3) 10-min fast charging conditions and data obtained for new control cells showing capacity fading during 6-C charging.