Strategies for Alleviating Electrode Expansion of Graphite Electrodes in Sodium‐Ion Batteries Followed by In Situ Electrochemical Dilatometry

The electrochemical intercalation/deintercalation of solvated sodium ions into graphite is a highly reversible process, but leads to large, undesired electrode expansion/shrinkage (“breathing”). Herein, two strategies to mitigate the electrode expansion are studied. Starting with the standard configuration (−) sodium | diglyme (2G) electrolyte | graphite (poly(vinylidene difluoride) (PVDF) binder) (+), the PVDF binder is first replaced with a binder made of the sodium salt of carboxymethyl cellulose (CMC). Second, ethylenediamine (EN) is added to the electrolyte solution as a co‐solvent. The electrode breathing is followed in situ (operando) through electrochemical dilatometry (ECD). It is found that replacing PVDF with CMC is only effective in reducing the electrode expansion during initial sodiation. During cycling, the electrode breathing for both binders is comparable. Much more effective is the addition of EN. The addition of 10 v/v EN to the diglyme electrolyte strongly reduces the electrode expansion during the initial sodiation (+100% with EN versus +175% without EN) as well as the breathing during cycling. A more detailed analysis of the ECD signals reveals that solvent co‐intercalation temporarily leads to pillaring of the graphite lattice and that the addition of EN to 2G leads to a change in the sodium storage mechanism.

. General advantages and disadvantages of PVDF and CMC.

PVDF CMC
Advantages: + High chemical stability + High electrochemical stability + Mechanically stable [1] Advantages: + Aqueous processing possible (noncritical, low cost, fast electrode drying) + Low-cost + Thermal stability + Better adhesion between electrode film and current collector than if PVDF is used (strong hydrogen and chemical bonding) [2] [3] [3] [1] Disadvantages: -NMP for processing (hazardous Disadvantages: -More hydrophilic (more moisture adsorption) -Brittle [4b] [8] a) PVDF is poly(vinylidene difluoride), CMC is sodium salt of carboxymethyl cellulose, NMP is N-methylpyrrolidone.   The graphite particles can be observed on the top, which are held together by the binder. In the case of PVDF, the binder can be clearly seen on the surface of the electrode (see Figure S1a). This is also confirmed by the elemental mapping displayed in Figure S2a and also shows that nearly no PVDF binder can be found in the bulk part of the electrode. In the case of CMC, the binder is much more homogeneously distributed within the electrode.  The previous results are confirmed by performing similar experiments in lithium-ion half-cells.
The results of these experiments are shown in the Figure S5. The voltage profiles is slightly more uneven compare to SIBs, a behavior known from the literature. [9] Regarding the binder influence, a similar trend was found as in the case of sodium-ion half-cells. Thus, the expansion in the first cycle is 19% smaller for the CMC electrodes compared to those made with PVDF (142% for CMC vs 176% for PVDF). In addition, the breathing of the electrodes in the following cycles is similar for CMC (38-42%) and PVDF (36-40%). It is worth to mention that for conventional LIBs, the total expansion of the battery should be limited to around 5%. [10] This, however, includes the whole cell system, i.e. expansion/shrinkage of both anode and cathode needs to be considered which can also compensate each other. Electrochemical experiments (a. and b.) conducted in a two-electrode set-up with sodium as counter electrode and a current of 11 mA g -1 (corresponds to a C-rate of 0.1 C for [Na(2G)x]C20).
Labeling of the signals according to the results from Zhang et al. [11] , where m indicates the monolayer t-GIC and b the bilayer t-GIC. Conducted with 11 mA g -1 (corresponds to a C-rate of 0.1 C for [Na(2G)x]C20).

Comments on experimental challenges due to the use of ethylenediamine in electrochemical cells
Ethylenediamine (EN) is a highly reactive solvent and therefore electrochemical measurements in half cells were challenging. Pure EN cannot be used as solvent as it readily forms electrides and H2 with sodium, which is used as counter and, or reference electrode. Due to that the surface of the sodium appears blue and H2 gas evolves which could lead to a cell blow off. [11] The blue color can be seen in Figure S9. 2G + 10 v/v EN Figure S9. The picture shows a piece of sodium after 24 h of exposure to pure ethylenediamine.