Transition Metal Layer Substitution in Mo2CS2 MXene for Improving Li Ion Surface Kinetics

We study the adsorption and mobility of a Li ion on the surface of the Mo2CS2 MXene by means of Density Functional Theory. We find that by substituting the Mo atoms of the upper MXene layer with V the mobility of the Li ion can be improved up to 95% while the material retains its metallic character. This fact indicates that MoVCS2 is a promising candidate for anode electrode in Li-ion batteries, where the materials need to be conductive and the Li ion needs to have a small migration barrier.


■ INTRODUCTION
The tunable properties 1−5 via control of the surface termination atoms of two-dimensional (2D) materials has made the latter the new point of interest regarding their use as anode materials in Li-ion batteries. Since 2011, a new ceramic material belongs in the 2D class, the MXenes. MXenes have the general formula M n+1 X n T x , (n = 1, 2, 3), with T a surface termination atom 2 and M an early transition metal. X stands for either carbon (C) or nitrogen (N). 4,6−9 The most studied MXene is titanium carbide, Ti 3 C 2 , 10−17 which was also the first to be discovered 4,18 in 2011. Since then, zirconium carbide, Zr 2 C, has garnered attention 5,19−22 after it was experimentally synthesized 20 by etching Al 3 C 3 from layered Zr 3 Al 3 C 5 structures. However, the studies of Zr 2 C as an anode electrode in ion batteries have been limited to O, OH, F, and S surface termination atoms, while a more extensive study of its use as an anode electrode in Li and non-Li ion batteries was conducted by Papadopoulou et al. in 2022. 5 A year earlier than the synthesis of Zr 2 C, in 2015, another MXene, molybdenum carbide, Mo 2 C, was experimentally synthesized from the gallium-based atomic laminate Mo 2 Ga 2 C. 23 Band structure calculations have showed that Mo 2 C is metallic when it is terminated by O or OH atoms, while it becomes semiconducting when it is terminated by F or Cl atoms. 23 The need to find new electrode materials with large ion capacities and fast charging rates has led to the question of whether Mo 2 C can be used to enhance electrode performance.
In 2016, Sun et al. 24 theoretically verified that the bare and terminated Mo 2 C is dynamically stable, a crucial property for an electrode. Also in 2016, Çakır et al. 25 theoretically studied the adsorption and diffusion of Li, Na, K, and Ca atoms on the surface of the Mo 2 C MXene. They predicted that all the aforementioned atoms are strongly adsorbed on the surface of the monolayer MXene and found energy barriers for diffusions as low as 15 meV (K case).
In 2017, Mehta et al. 26 studied the adsorption of a Li ion on the Mo 2 C monolayer doped with N and Mn using firstprinciples calculations. In particular, they (a) substituted one Mo atom from the upper MXene layer with Mn and (b) substituted a C atom with N. They found that the presence of the dopant atoms increases the Li adsorption and, thus, the ion storage capacity. However, there was no investigation regarding the mobility of the Li ion.
Finally, Mehta et al. 27 in 2019 performed first-principles calculations in S-terminated Mo 2 C. They found that the Li ion has a diffusion barrier equal to 0.22 eV.
In the present work, we study a Li ion's adsorption and diffusion in the Mo 2 CS 2 structure where the upper Mo MXene layer has been entirely substituted by vanadium, V. We denote this structure as MoVCS 2 . We chose the S-terminated Mo 2 CS 2 because previous studies 5,28,29 have shown that S lowers the migration barrier of the Li ion. The significance of this study and the findings therein are 2-fold: First, we identify that the transition layer substitution results in a more stable material than the original configuration, a fact that opens the way for experimental work on MXenes with two different transition metals. Second, the very low Li-ion diffusion barriers calculated herein indicate the suitability of the aforementioned kind of MXenes for electrodes in Li-ion batteries.

■ COMPUTATIONAL METHODS
For the bulk structure of the bare Mo 2 C, we used a starting structure typical of that found in theoretical analysis. 30 We performed Density Functional Theory (DFT) calculations as implemented in the CASTEP package 31−35 to derive the Mo 2 CS 2 structure, where we randomly placed the S atom on top of the Mo atom, at a distance smaller than the bond length R 0 39 between the S and Mo which is equal to 2.33 Å. 40 DFT uses pseudopotentials 36 generated on the fly for the electron− ion interactions. 37 The exchange-correlation interactions between electrons were described using the corrected density functional of Perdew, Burke, and Ernzerhof (PBE) within the generalized gradient approximation (GGA).
A vacuum space of 30 Å was introduced to minimize the mirror interactions between adjacent MXene layers. A plane wave cutoff energy E cut = 650 eV and a k-point spacing of 7 × 7 × 1 was used to converge the overall energy per formula unit to 0.5 × 10 −5 eV. The structures were optimized using the BFGS geometry optimization method 35,38 and relaxed until the residual forces on the atoms were less than 0.01 eV × Å −1 . DFT produced a final structure where the interlayer distance between S and Mo increased by 28.3% from the one we had originally considered, while the interlayer distance between the two Mo layers decreased by 1.4%.
For the structures including the Li-ion, DFT was applied again after we inserted a single ion on the surface of Mo 2 CS 2 at a distance from the S atom smaller than the bond length where E MoVCSd 2 −Li is the total energy of the structure where the upper Mo layer has been substituted by V including the Li ion and E MoVCSd 2 is the total energy of MoVCS 2 without the added ion.
In order to calculate the energy barrier for diffusion, E bar , for the Li ion on each structure's surface, first we created a 2 × 1 × 1 supercell of each structure. The initial (reactant) position of the Li ion was the one predicted by the DFT calculations, while the final (product) position was the same position in the adjacent cell. We then applied an LST/QST transition state (TS) search algorithm in CASTEP. 41 The E bar was calculated as the barrier from the initial position.
Finally, in order to calculate the Density of States (DOS) and thus characterize each material regarding its conductivity, we used the OPTADOS code 42 as implemented in CASTEP. For convenience, the Fermi level was shifted at 0 eV.

■ RESULTS AND DISCUSSION
The structure of the Mo 2 CS 2 MXene is shown in Figure 1. The S atom sits directly above the Mo atom of the upper MXene layer. Furthermore, in Figure 2 we can see the structure of the MoVCS 2 where the upper Mo layer of Mo 2 CS 2 has been substituted by V. This time, the termination atom S sits above a bottom layer Mo atom. The upper and bottom MXene layers are denoted as L1 and L2 in Figure 2.
The coordination number of the S atom is also different. In Mo 2 CS 2 S bonds with only one atom belonging in L1, and in MoVCS 2 S bonds with three L1 atoms. Finally, the MoVCS 2 structure has 3.25% lower final energy than the Mo 2 CS 2 one, a fact that indicates that it is also stable.  As a next step, we added the Li ion on the surface of the two structures. In Figure 3 we see the Mo 2 CS 2 case where the Li ion sits on top of the S termination, while in Figure 4 we see the MoVCS 2 case where the Li ion sits on top of a Mo atom of L2. In the latter case, the S termination atoms have slightly moved due to the bonds with Li and now sit above empty space. The MoVCS 2 −Li structure also has 3.1% lower final energy than the Mo 2 CS 2 −Li one, a fact that indicates its stability.
The adsorption energies for the Li ion on each structure's surface are shown in Table 1. On Mo 2 CS 2 , Li appears to be more strongly adsorbed despite its smaller coordination number as described above. As Papadopoulou et al. 9 have showed, E ads depends on the average distance between the termination atoms, the Li ions, and the L1 atoms. In fact, the smaller these distances are, the stronger the adsorption. Our results here are fully in support of this observation, as the distances between the aforementioned atoms in Mo 2 CS 2 are smaller, on average, than in MoVCS 2 as indicated in Table 2.
Also in Table 1, we can see the migration barriers E bar for the Li ion. One can see that the substitution of Mo with V lowers the energy barrier by 95% (0.59 eV in Mo 2 CS 2 vs 0.03 eV in MoVCS 2 ). This fact is significant since we seek to improve the mobility of the Li ion in order to achieve faster charge/ discharge rates in Li-ion batteries.
Mehta et al. 27 cite a migration barrier for Li in Mo 2 CS 2 equal to 0.22 eV. This difference with the results reported here has to do with the different structure reported for Mo 2 CS 2 −Li where Mehta et al. found that S sits on top of a C atom. This difference is a result of the different softwares used to perform DFT (VASP vs CASTEP), as well as the different pseudopotentials used. Still, if we were to accept that E bar = 0.22 eV, MoVCS 2 exhibits 86.4% lower Li diffusion barrier. This reduction is also of high significance.
This result of E bar = 0.03 eV we found here for MoVCS 2 is equal with the one found for the Ti 3 C 2 Cl 2 MXene in a previous study, 9 further reinforcing the use of MXenes as anode electrodes in Li-ion batteries.
Finally, in Figure 5, we see the calculated DOS for each structure. All materials are metallic (no bandgap at 0 eV, i.e., the Fermi level), a property that is essential for electrodes in Li-ion batteries.

■ CONCLUSIONS
Rechargeable ion batteries are energy storage devices whose operation is based predominantly on the intercalation of ions. 43 In general, an ion battery consists of a cathode (positive electrode) and an anode (negative electrode) in contact with an electrolyte which contains ions. The two electrodes are separated by a microporous polymer membrane (separator) which stops the electrons from passing between them alongside the ions. 44 During charging of an ion battery cell, ions leave the positive electrode and move through the electrolyte to the negative electrode. We have, therefore, a storage of energy to the anode. During discharging of the battery, this energy is released, powering the battery, and the ions move back to the cathode.
One of the most active research fields regarding Li-ion batteries is the increase of their rate performance in order to reduce charging time which is important to the electric vehicles' market. 45 The materials of the electrodes are crucial for the performance of Li-ion batteries, as they determine capacity, cell voltage, and cyclability. 44 Anode materials are still predominantly carbon-based; however, these anodes have almost reached their theoretical    Figure 6.
In the present study, we examined the use of the Mo 2 CS 2 MXene as an anode material for Li-ion batteries, where the upper layer of Mo atoms has been substituted by V. We found that this substitution renders the initial Mo 2 CS 2 more thermodynamically stable, although experimental work has not yet been applied in depth to the kind of materials like the MoVCS 2 MXene. Furthermore, the substitution of the entire top Mo layer with V was shown to improve the mobility of the Li ion by 95%, while the material retains its metallic character. This fact makes the MoVCS 2 material a promising candidate for the negative electrode in Li-ion batteries since it allows for faster charge/discharge rates than the Mo 2 CS 2 MXene.