(Digital Presentation) The Role of the Electrolyte in the Charge Storage Mechanism of TiS2 in Sodium-Ion Batteries

Abstract

The development of high-performance energy storage devices has recently attracted much attention for applications such as electrical vehicles, grid storage and portable devices. In this sense, sodium-ion batteries (SIBs) has appeared as an alternative to traditional lithium-ion batteries as a result of the availability, lower environmental impact and reduced cost of the battery due to the typical SIB chemistry.1 Even though graphite is a well-established and highly commercialized anode material, its application in SIBs suffers from several drawbacks, such as the practical impossibility of using traditional carbonates-based electrolytes due to unfavorable thermodynamic behavior – resulting in low specific capacities. Despite this, certain solvents (such as ethers) allow the use of graphite in SIBs through the formation of ternary graphite intercalation compounds by a co-intercalation reaction, showing excellent cyclability and kinetics. However, its capacity at 100-150 mAh g-1 is still low compare to other anode materials, in addition to showing large volume expansion. 2, 3 Transition metal dichalcogenides (TMDs) have emerged as promising anode materials owing to their low cost, high electric conductivity, good thermal stability and environmental friendliness. Among them, titanium sulphide (TiS2), with a two-dimensional framework, exhibits several advantages such as a high conductivity (compare to other metal oxides), a larger interlayer distance (0.569 nm) compared to graphite (0.335 nm) and high stability. 4 Several papers have reported on the use of TiS2 as an anode material for SIBs, and observed that the voltage profiles and electrochemical behaviour of the system is highly dependent on the choice of electrolyte, yet this has seldomly been explicitly stated, nor the cause of the electrolyte dependence investigated.Here, we demonstrate that depending on the electrolyte employed, an electrochemical co-intercalation reaction, similar to those reported with graphite as active material can also occur. Galvanostatic charge-discharge experiments were first performed to assess differences between different electrolytes (linear ethers, cyclic ethers and carbonates). Here, differences were found on the voltage profiles among the linear ethers and the rest of the solvents employed. By using in-situ and ex-situ techniques, a huge expansion on the TiS2 layers was found when diglyme was the solvent used. It was concluded here that co-intercalation of the solvent with sodium was the mechanism that occurs. On the contrary, when a cyclic ether and/or a mixture of carbonates were used, a lower expansion of the layered structure occurred and remained constant during the cycling. Dilatometry experiments were used to confirmed these mechanisms. Thus, with this technique, it is possible to monitor the variation on the thickness of the electrode during cycling, i.e. operando. In this sense, a greater expansion was observed when diglyme was used as the solvent electrolyte.Finally, DFT was used to study the solvents investigated, and their interaction with the sodium cation. It was found that among the solvents, diglyme produced by far the smallest and most stable solvation shells, making them prime candidates for co-intercalation studies. Comparing the size and stability of the solvation shell, with the interlayer binding energy of TiS2, we found it was energetically favorable to expand the lattice and allow a solvation shell into the host structure, compared to forming a bare ion. Figure 1. Illustration on the variation of the layer distance from pristine to sodiated diglyme. D. Kundu, E. Talaie, V. Duffort and L. F. Nazar, Angew. Chem. Int. Ed., 2015, 54, 3431-3448.B. Jache, J. O. Binder, T. Abe and P. Adelhelm, Phys. Chem. Chem. Phys., 2016, 18, 14299-14316.M. Goktas, C. Bolli, J. Buchheim, E. J. Berg, P. Novák, F. Bonilla, T. f. Rojo, S. Komaba, K. Kubota and P. Adelhelm, ACS applied materials & interfaces, 2019, 11, 32844-32855. Q. Yun, L. Li, Z. Hu, Q. Lu, B. Chen and H. Zhang, Adv. Mater., 2020, 32, 1903826. Figure 1