Enhancement of Y5−xPrxSb3−yMy (M = Sn, Pb) Electrodes for Lithium- and Sodium-Ion Batteries by Structure Disordering and CNTs Additives

The maximally disordered (MD) phases with the general formula Y5−xPrxSb3−yMy (M = Sn, Pb) are formed with partial substitution of Y by Pr and Sb by Sn or Pb in the binary Y5Sb3 compound. During the electrochemical lithiation and sodiation, the formation of Y5-xPrxSb3-yMyLiz and Y5−xPrxSb3−yMyNaz maximally disordered–high entropy intermetallic phases (MD-HEIP), as the result of insertion of Li/Na into octahedral voids, were observed. Carbon nanotubes (CNT) are an effective additive to improve the cycle stability of the Y5−xPrxSb3−yMy (M = Sn, Pb) anodes for lithium-ion (LIBs) and sodium-ion batteries (SIBs). Modification of Y5−xPrxSb3−ySny alloys by carbon nanotubes allowed us to significantly increase the discharge capacity of both types of batteries, which reaches 280 mAh · g−1 (for LIBs) and 160 mAh · g−1 (for SIBs), respectively. For Y5−xPrxSb3−yPby alloys in which antimony is replaced by lead, these capacities are slightly smaller and are 270 mAh · g−1 (for LIBs) and 155 mAh · g−1 (for SIBs), respectively. Results show that structure disordering and CNT additives could increase the electrode capacities up to 30% for LIBs and up to 25% for SIBs.


Introduction
Intermetallics with sufficiently large interatomic voids are favorable for the insertion of lithium or sodium ions. The intermetallic compound LaSn 3 (AuCu 3 -type) as a possible negative electrode for the lithium-ion batteries was described by Vaughey et al. [1]. The insertion of lithium atoms into octahedral voids of hexagonal Zr 5 Sn 3 and RE 5 M 3 (RE = Y, La and Gd; M = Ge and Sn) binary phases were described earlier by us [2][3][4]. The crystal structure of RE 5 M 3 intermetallic compounds belongs to Mn 5 Si 3 -type. The RE atoms are located at 4d and 6g sites, respectively. The M components (Ge or Sn or Pb atoms) are located at the additional 6g site. For this binary structure, the presence of an octahedral void in the 2b site is typical. The incorporation of the third element (alkaline or transitions metals, marked as T) into the 2b site leads to the realization of the RE 5 TM 3 ternary phase with a Hf 5 CuSn 3 structure type [5]. We detected the new ternary phase of Tb 5 LiSn 3 with fully filled 2b site (Hf 5 CuSn 3 structure type) during the systematic study of the alloys of Tb-Li-Sn system [6]. The La 5 Li x Ge 3 phase, which formed by partial incorporation (up to x = 0.4) of lithium atoms into octahedral voids of the La 5 Ge 3 binary, was observed in La-Li-Ge system [3].
The rationale for the use of high entropy alloys as electrode materials is based on the assumption defined by Jien-Wei Yeh and colleagues [7] that the presence of several (five or more) elements in the alloy would increase the configurational entropy of mixing by an amount sufficient to overcome the enthalpies of compound formation, thereby, inhibiting the formation of potentially undesirable additional intermetallics. So, at the increasing of the number of elements in an alloy, the entropic contribution to the total free energy would overcome the enthalpic contribution and, thereby, stabilize solid solution phases [8]. Alloys of systems based on rare earth metals are prone to the formation of solid solutions, which can be both limited and unlimited. In some metallic systems, depending on the composition of the alloys, ordered intermetallic compounds, limited solid solutions, and unlimited solid solutions can be formed simultaneously [9][10][11]. A positive effect of component substitution on electrochemical properties was also observed for Li (Al1-zZnz) solid solution [12].
The concept of stabilizing a crystal structure with high entropy allows us to build electrodes that can be cycled for a long time without significantly reducing the capacity. Previously, this was shown for high-entropy oxides [13] such as cathode materials and MD-HEIP intermetallics as anode materials for Li-and Na-ion batteries [14].
It is shown that increasing entropy by introducing one or more elements leads to a completely different electrochemical behavior and stability of cycles. In addition, there is a unique opportunity to fine-tune the electrochemical characteristics of high-entropy materials, using the different effects of each element on the electrochemical processes. Based on these studies, a possible reaction mechanism is proposed, which takes into account the stabilization of entropy and the maintenance of the invariance of the structure of the initial matrix throughout the cycling process.
In addition to the structural modification of the main phase of the electrode due to the substitution of components [1][2][3][4][5][6][7][8][9][10][11][12][13][14], it is also significant addition of catalysts and CNT conductive additives that improve the cyclic stability of the electrodes. According to known literature data [15][16][17][18][19], the carbon nanotubes (CNTs) are a good candidate material for use in Li-ion batteries due to their perfect electrochemical properties. The insertion of CNTs as a conductive additive at a lower weight loading than conventional carbon black and other graphite representatives to the electrode matrix gives the opportunity to create more quality electrodes.
Chawla et al. [20] described the inclusion of Pd catalyst in CNT cathode materials in Li-O 2 batteries and received electrochemical data with significant improvement in discharge capacities. The batteries with CNTs and Pd catalyst cycled have 35% more cycles. The activity of the germanium catalyst was investigated to improve the electrochemical performance of SnO 2 /graphene nanocomposite anode material for LIBs [21].
In this study, we implemented two goals: first, we determined the effect of disordering the structure of Y 5−x Pr x Sb 3−y M y (M = Sn, Pb) electrodes due to component substitution, and second, we determined the effect of a modifying additive (CNT) on the electrochemical activity of electrodes for LIBs and SIBs. A CNT-modified electrode shows better kinetics of Li + /Na + intercalation into electrodes and higher corrosion resistance in electrolyte solutions and, as a result, increases the battery life. The positive effect of these two actions leads to improved cyclic stability and increased electrode capacity up to 30% for LIBs and up to 25% for SIBs.

Synthesis and Phase Analysis
The maximally disordered Y 5−x Pr x Sb 3−y M y (M = Sn, Pb; 0 ≤ x,y ≤ 0.5) alloys were prepared from high purity metals: yttrium (ingot, 99.9 at.%, Sigma-Aldrich, Saint Louis, MO, USA), praseodymium (ingot, 99.9 at.%, Sigma-Aldrich, Saint Louis, MO, USA), antimony (rod, 99.99 at.%, Sigma-Aldrich, Saint Louis, MO, USA), tin (99.8%, shot, 3 mm, Sigma-Aldrich, Saint Louis, MO, USA) and lead (shot, <2 mm, 99.9%, Sigma-Aldrich, Saint Louis, MO, USA). These alloys were synthesized by arc-melting of pure metals under the argon atmosphere. The Y 5−x Pr x Sb 3−y M y (Li/Na) z alloys were synthesized by both thermal and electrochemical methods. For the thermal insertion of lithium and sodium into Y 5−x Pr x Sb 3−y M y , two separate pieces of this alloy were mixed with Li or Na, respectively, according to the aimed stoichiometry of the product and filled into tantalum crucibles under argon atmosphere that were sealed by arc-welding. For the electrochemical method, the lithiation and sodiation of Y 5−x Pr x Sb 3−y M y alloy were used. The samples synthesized by the thermal method were annealed at 670 K for 24 h. The reaction product was analyzed by X-ray powder diffraction (XRD) using Rigaku MiniFlex 600 powder diffractometer (Curadiation), Rigaku, Tokyo, Japan. The phase content of alloys was determined by means of a TESCAN VEGA3 electron microscope (Tescan, Brno, Czech Republic) equipped with Oxford Instruments energy dispersive X-ray analyzer, Aztec ONE system, High Wycombe, UK. In some cases, the flame photometry (Flapho-4 flame photometer, Carl Zeiss Jena, Jena, Germany) for detection of lithium was used.

Structure Refinement
Both the single crystal and powder X-ray diffraction methods were used for the investigation of the crystal structures of Y 5−x Pr x Sb 3−y M y and Y 5−x Pr x Sb 3−y M y (Li/Na) z phases. Rietveld refinements against X-ray powder diffraction data were performed by using the FULLPROF program (version June 2020) [22]. The single crystals data were collected using an automatic four-circle Xcalibur Oxford Diffraction diffractometer (Agilent Technologies, Inc., Santa Clara, CA, USA) with CCD detector (graphite monochromatized Mo-radiation, ω-scan mode). The analytical absorption corrections were made by CrysalisRed [23]. The crystal structure was solved by direct methods and refined using the SHELX-97 program package (Georg August University of Göttingen, Göttingen, Germany) [24,25].
Crystallographic data (CIF and structure factors) for the structures reported in this paper has been deposited with the Cambridge Crystallographic Data Centre, no. CCDC-2062904 (for Y 5−x Pr x Sb 3−y M y Li z ) and CCDC-2062905 (for Y 5−x Pr x Sb 3−y M y Na z ). These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via http://www.ccdc.cam.ac.uk/data_request/cif (accessed on 13 February 2021) [26].

Electrochemical Investigations
Two series, each consisting of five electrodes of the corresponding compositions Y 5 Sb 3 , Y 5−x Pr x Sb 3−y Sn y , Y 5−x Pr x Sb 3−y Pb y , Y 5−x Pr x Sb 3−y Sn y -CNT, and Y 5−x Pr x Sb 3−y Pb y -CNT were prepared for electrochemical studies. Electrodes from Y 5 Sb 3 , Y 5−x Pr x Sb 3−y Sn y , and Y 5−x Pr x Sb 3−y Pb y were cut from alloys or prepared by grinding of bulk alloys to powders and the following mixed in an agate mortar of 90 wt.% alloys (acting as active material) with 10 wt.% polyvinylidene fluoride (PVDF binder), and finally, these mixtures were pressed onto a stainless-steel grid (about 5 mg, coin cell type electrode 11 mm in diameter and 0.8 mm thick). The electrode with CNT catalyst (Y 5−x Pr x Sb 3−y Sn y -CNT and Y 5−x Pr x Sb 3−y Pb y -CNT) were prepared by mixing powdered alloys with carbon nanotubes in ratio 95:5, 90:10, and 85:15, respectively, and subsequently, sintering at 670 K for 48 h. The carbon nanotubes NC7000 were manufactured by Nanocyl SA (Sambreville, Belgium), which were produced through catalytic chemical vapor deposition (CCVD). Their average diameter and length were 9.5 nm and 1.5 µm, respectively.
Subsequently, the sintering product was ground into a powder to a grain size of almost 1 µm and mixed with 10 wt.% PVDF binder and pressed onto a stainless-steel grid. This grain size of the powder of the active phase of the anode material was due to proportionality with the grain sizes of the powder of the cathode materials, which were commercial.
The lithium-ion cells contain prepared alloy as a working electrode with Li reference and LiCoO 2 counter electrodes. The sodium-ion cells contain Na as a reference and Na x CoO 2 as counter electrodes, respectively. The ethylencarbonate/dimethylcarbonate nonaqueous electrolyte, containing Li + or Na + ions were used. The anode and cathode were separated by Celgard 2320 separator (Celgrad, Charlotte, CA, USA) impregnated with an electrolyte.
Multi-cycle chronopotentiometry (CP), voltammetry (CVA) and electrochemical impedance spectroscopy (EIS) were performed in three-electrode Swagelok-type cells. All electro-chemical measurements were carried out by means of the electrochemical station from CH Instruments (Austin, TX, USA) and 8 Channel Battery Analyzer (BST8-MA, MTI Corporation, Richmond, CA, USA). Galvanostatic charge/discharge cycles were received at i ch = 2 mA cm −2 , i disch = 1 mA cm −2 . The cyclic voltammograms were registered at scan rate = 0.001 V/s. The impedance spectra were taken on electrodes cycled between 1 and 4.5 V using a 0.001 Hz-100 kHz frequency range. To determine the values of the standard deviation for discharge capacity and parameters of EIS measurements, the experiments were repeated three times. The relative standard deviation of discharge capacities does not exceed 8%. The relative standard deviation for all EIS parameters does not exceed 12%.

Results and Discussion
Disordered and high-entropy alloys are suitable for the development of advanced materials with unique properties that cannot be achieved by conventional materials based only on ordered phases. High-entropy alloys can be used to create materials with high heat resistance and oxidation resistance, with improved mechanical, magnetic, and electrochemical properties. The MD and MD-HEIP were received from Y 5 Sb 3 by concept based on two mechanisms: substitution and insertion. For the partial substitution of the Y atoms in crystallographic positions 4d and 6g, the Pr atoms were used, and Sn or Pb atoms were used for the partial substitution of Sn. As a result of this substitution, a MD phase of Y 5−x Pr x Sb 3−y M y is formed. The structure of the Y 5 Sb 3 binary phase contains typical empty octahedral voids in which lithium or sodium can be inserted. Moreover, atoms can be inserted in an octahedral void either thermally or electrochemically.
The results of X-ray powder diffraction and microprobe analysis show that the prepared alloys contain the hexagonal phase with Mn 5 Si 3 structure type [27]. In this structure type, the binary phase Y 5 Sb 3 crystallizes [28]. The X-ray powder pattern after Rietveld refinements and SEM micrographs of Y 5 Sb 3 , Y 5−x Pr x Sb 3−y Sn y , and Y 5−x Pr x Sb 3−y Pb y are presented in Figures 1 and 2a-c, respectively. The Rietveld refinements were carried out in P6 3 /mcm space group with the structure model corresponding to the Mn 5 Si 3 structure type. The structural data for the ordered Y 5 Sb 3 and disordered Y 5−x Pr x Sb 3−y M y (M = Sn, Pb) and experimental details of the structure determination are presented in Table 1. The refined unit cell dimensions of disordered Y 5−x Pr x Sb 3−y M y phases are increased in comparison to the binary Y 5 Sb 3 ordered phase; it correlates well with atomic radii of substituting atoms.
For detailed structural analysis and to confirm that the obtained phases were MD-HEIP, a thorough high-precision structural study was performed using single-crystal X-ray methods. Two single crystals were selected from thermally synthesized and homogenized alloys Y 4.5 Pr 0.5 Sb 2.5 Sn 0.5 Li and Y 4.5 Pr 0.5 Sb 2.5 Sn 0.5 Na and used to completely determine the crystal structure. Initial X-ray studies showed that all single crystals had the same hexagonal structure as Hf 5 CuSn 3 -type, which is a filled variant of the Mn 5 Si 3 -type, in which the binary phase of Y 5 Sb 3 crystallizes. Since these phases are multicomponent, it was important to more accurately determine the distribution of atoms in the crystallographic positions, which allows us to make an X-ray single crystal method. General information on the obtained structural results is summarized in Table 2. Fractional atomic coordinates and displacement parameters for Y 5−x Pr x Sb 3−y Sn y Li and Y 5−x Pr x Sb 3−y Sn y Na are presented in Table 3. Detailed structural refinements show that lithium and sodium can be inserted into octahedral voids.    For detailed structural analysis and to confirm that the obtained phases were MD-HEIP, a thorough high-precision structural study was performed using single-crystal X-ray methods. Two single crystals were selected from thermally synthesized and homogenized alloys Y4.5Pr0.5Sb2.5Sn0.5Li and Y4.5Pr0.5Sb2.5Sn0.5Na and used to completely determine the crystal structure. Initial X-ray studies showed that all single crystals had the same hexagonal structure as Hf5CuSn3-type, which is a filled variant of the Mn5Si3-type, in which the binary phase of Y5Sb3 crystallizes. Since these phases are multicomponent, it was important to more accurately determine the distribution of atoms in the crystallographic positions, which allows us to make an X-ray single crystal method. General information on the obtained structural results is summarized in Table 2. Fractional atomic coordinates and displacement parameters for Y5−xPrxSb3−ySnyLi and Y5−xPrxSb3−ySnyNa are presented in Table 3. Detailed structural refinements show that lithium and sodium can be inserted into octahedral voids.     The structural feature of these phases based on the possibility of insertion/deinsertion of lithium and sodium atoms into octahedral voids prompted the study of the possibility of carrying out these processes using electrochemical methods. The insertion of lithium or sodium atoms significantly increases the volume of the unit cell, more significantly along the lattice period c, since in this direction, the octahedron expands more during the insertion of atoms (Figure 3). At the insertion of Li into octahedra, the volume of the unit cell increases by 4.65 Å 3 , and there are even greater changes up to 11.36 Å 3 when sodium atoms are inserted. After several initial cycles of electrochemical lithiation/delithiation, as well as sodiation/desodiation of disk electrodes that were cut from the alloy, their amorphization was observed (Figure 2d-i), and their capacity decreased slightly. After 5-8 cycles of activation, the electrodes stabilized and could continue to operate for many cycles without significant loss of capacity. There are no significant differences when using electrodes that were made by pressing powders from crushed alloys. It should be noted that the electrodes from the ordered binary phaseY5Sb3 have a discharge capacity in the range of 200 mAh · g −1 (for LIBs) and 120 mAh · g −1 (for SIBs), respectively (Figure 4). The use of electrodes from the MD phases increases the discharge capacitance for both LIBs and SIBs by almost 10%. Moreover, slightly higher values of capacitance have electrodes in which the Sb atoms are replaced by Sn, in comparison with electrodes in which the Sb are replaced by Pb atoms. After several initial cycles of electrochemical lithiation/delithiation, as well as sodiation/desodiation of disk electrodes that were cut from the alloy, their amorphization was observed (Figure 2d-i), and their capacity decreased slightly. After 5-8 cycles of activation, the electrodes stabilized and could continue to operate for many cycles without significant loss of capacity. There are no significant differences when using electrodes that were made by pressing powders from crushed alloys. It should be noted that the electrodes from the ordered binary phaseY 5 Sb 3 have a discharge capacity in the range of 200 mAh · g −1 (for LIBs) and 120 mAh · g −1 (for SIBs), respectively (Figure 4). The use of electrodes from the MD phases increases the discharge capacitance for both LIBs and SIBs by almost 10%. Moreover, slightly higher values of capacitance have electrodes in which the Sb atoms are replaced by Sn, in comparison with electrodes in which the Sb are replaced by Pb atoms. The next important step was the modification of the electrodes by carbon nanotubes in order to increase their capacity and improve cyclic stability. This modification was performed for MD phases of Y5−xPrxSb3−ySny and Y5−xPrxSb3−yPby. Different amounts of CNT were added, namely 5, 10, and 15 wt.% to alloys with subsequent mixing, pressing, and sintering. For example, Figure 5 shows the SEM images of the initial alloy, nanotubes, and the product of their sintering. It is seen that the nanotubes wrap the grains of the alloy, filling the intergranular space. The next important step was the modification of the electrodes by carbon nanotubes in order to increase their capacity and improve cyclic stability. This modification was performed for MD phases of Y 5−x Pr x Sb 3−y Sn y and Y 5−x Pr x Sb 3−y Pb y . Different amounts of CNT were added, namely 5, 10, and 15 wt.% to alloys with subsequent mixing, pressing, and sintering. For example, Figure 5 shows the SEM images of the initial alloy, nanotubes, and the product of their sintering. It is seen that the nanotubes wrap the grains of the alloy, filling the intergranular space. Electrochemical studies have shown that the optimal amount of CNT is 10 wt.%. has been experimentally established that if the content of carbon nanotubes in th electrodes exceeds 10 wt.%, for example, 15 wt.%, the capacity and cyclic characteristic of the electrodes begin to deteriorate due to the formation of grains of carbide phase during sintering. It is also important to ensure the mechanical stability of the electrode so that they do not crack during numerous charge-discharge cycles. This is realized b adding a binder material, the optimal amount of which was 10 wt.%. The addition o CNTs and PVDF binder to electrodes improved capacity and their cycles' stability. Th CNTs provided diffusion pathways for Li+ or Na+ ions to most of the grains inside th material, improving charge/discharge kinetics, and the PVDF binder provided goo adhesion between the electrode materials and current collectors, giving excellen electrochemical, mechanical, and thermal stability of electrodes. Modification of alloy with carbon nanotubes allowed us to significantly increase the discharge capacity of bot types of batteries as LIBs and SIBs (Figure 4). Modification of Y5−xPrxSb3−yMy (M = Sn, Pb alloys by carbon nanotubes allowed us to significantly increase the discharge capacity o both types of batteries, which reached 280 mAh · g −1 (for LIBs) and 160 mAh · g −1 (fo SIBs), respectively. These maximum capacities are for Y5-xPrxSb3-ySny alloy electrodes i which antimony is partially replaced by tin. For Y5-xPrxSb3-yPby alloys in which antimon is replaced by lead, these capacities are slightly smaller and are 270 mAh · g −1 (for LIBs and 155 mAh · g −1 (for SIBs), respectively. The study of cyclic stability shows that fo CNT-modified electrodes after 50 cycles of charge-discharge, capacity decreases by n more than 8% ( Figure 6).
The analysis potentiodynamic polarization curves (Figure 7) show that tested Y5Sb Y5−xPrxSb3−yMy (M = Sn, Pb), and Y5−xPrxSb3−yMy (M = Sn, Pb)-CNT electrodes passivate i Electrochemical studies have shown that the optimal amount of CNT is 10 wt.%. It has been experimentally established that if the content of carbon nanotubes in the electrodes exceeds 10 wt.%, for example, 15 wt.%, the capacity and cyclic characteristics of the electrodes begin to deteriorate due to the formation of grains of carbide phases during sintering. It is also important to ensure the mechanical stability of the electrodes so that they do not crack during numerous charge-discharge cycles. This is realized by adding a binder material, the optimal amount of which was 10 wt.%. The addition of CNTs and PVDF binder to electrodes improved capacity and their cycles' stability. The CNTs provided diffusion pathways for Li+ or Na+ ions to most of the grains inside the material, improving charge/discharge kinetics, and the PVDF binder provided good adhesion between the electrode materials and current collectors, giving excellent electrochemical, mechanical, and thermal stability of electrodes. Modification of alloys with carbon nanotubes allowed us to significantly increase the discharge capacity of both types of batteries as LIBs and SIBs (Figure 4). Modification of Y 5−x Pr x Sb 3−y M y (M = Sn, Pb) alloys by carbon nanotubes allowed us to significantly increase the discharge capacity of both types of batteries, which reached 280 mAh · g −1 (for LIBs) and 160 mAh · g −1 (for SIBs), respectively. These maximum capacities are for Y 5-x Pr x Sb 3-y Sn y alloy electrodes in which antimony is partially replaced by tin. For Y 5-x Pr x Sb 3-y Pb y alloys in which antimony is replaced by lead, these capacities are slightly smaller and are 270 mAh · g −1 (for LIBs) and 155 mAh · g −1 (for SIBs), respectively. The study of cyclic stability shows that for CNT-modified electrodes after 50 cycles of charge-discharge, capacity decreases by no more than 8% ( Figure 6).  The analysis potentiodynamic polarization curves (Figure 7) show that tested Y 5 Sb 3 , Y 5−x Pr x Sb 3−y M y (M = Sn, Pb), and Y 5−x Pr x Sb 3−y M y (M = Sn, Pb)-CNT electrodes passivate in LiPF 6 based ethylencarbonate/dimethylcarbonate electrolyte, electrolyte solution, and their passivation range extends from +1.07 V to +0.70 V. The electrode corrosion potential (E corr ) is equal to +1.07 V for Y 5 Sb 3 and shifts to +0.97 V for Y 5−x Pr x Sb 3−y Pb y and +0.84 V for Y 5−x Pr x Sb 3−y Sn y . For the catalyst activated by CNT, Y 5−x Pr x Sb 3−y Pb y -CNT and Y 5−x Pr x Sb 3−y Sn y -CNT electrodes, the E corr shifts to +0.77 V and +0.70 V, respectively, and these shifts indicate an increase of corrosion resistance in the electrolyte solution, which causes a higher specific capacity and better cyclic stability. Figure 8 shows the Nyquist plot from electrochemical impedance spectroscopy (EIS) data of Y 5 Sb 3 , Y 5-x Pr x Sb 3-y M y (M = Sn, Pb), and Y 5−x Pr x Sb 3−y M y -CNT (M = Sn, Pb) electrodes for both LIBs and SIBs. The semi-circle areas correspond to the charge transfer limitation. Generally, the kinetics of Li+ ions intercalation into electrodes is faster than Na+ because of the smaller radius and lighter mass of Li+ compared to Na+, resulting in smaller charge transfer resistances in the LIBs (Figure 8a) than in the SIBs (Figure 8b). As shown in Figure 8 for LIBs, the semi-circle areas of Y 5−x Pr x Sb 3−y M y -CNT (M = Sn, Pb) electrodes are smaller than those of the Y 5 Sb 3 and Y 5−x Pr x Sb 3−y M y , which demonstrates that the charge-transfer reaction is facilitated for the modified electrode. Similar behavior of these electrodes is observed for SIBs, but the higher values of the resistances indicate that the processes of charge transfer in this system are more difficult.
The electrochemical lithiation and sodiation of Y 5−x Pr x Sb 3−y M y phase can be written, respectively, as:    . EIS data analyzed by fitting to an equivalent electrical circuit scheme (inset) in which R1 is the electrolyte solution resistance, R2 is the charge transfer resistance, CPE1 is the double-layer capacitance, and Wo1 is related to the diffusion-controlled Warburg impedance.
CNT additives fill the intergranular space during sintering and improve the electrical connectivity within the electrode, thus, improving capacity by connecting and maintaining more electron conduction pathways between the active material and the current collector and then, finally, increasing the specific capacities and cycle stability of electrodes.

Conclusions
The maximally disordered Y 5−x Pr x Sb 3−y M y (M = Sn, Pb) phases with stable hexagonal Mn 5 Si 3 structure type were prepared and investigated as the anode materials for LIBs and SIBs. During the electrochemical lithiation and sodiation, the formation of Y 5−x Pr x Sb 3−y M y Li z and Y 5−x Pr x Sb 3−y M y Na z maximally disordered-high entropy intermetallic phases, as the result of the insertion of Li/Na into octahedral voids, were observed. Carbon nanotubes are an effective additive to improve the cycle stability of the Y 5−x Pr x Sb 3−y M y (M= Sn, Pb) anodes for lithium-ion and sodium-ion batteries. Modification of Y 5−x Pr x Sb 3−y M y (M = Sn, Pb) alloys by structure disordering and carbon nanotube additives allowed us to significantly increase the discharge capacity of both types of batteries, which reached 280 mAh · g −1 (for LIBs) and 160 mAh · g −1 (for SIBs), respectively. A CNT-modified electrode also shows higher corrosion resistance in electrolyte solutions and, as a result, demonstrates an increased battery life. The effect of the addition of CNTs and PVDF binder on the improvement of capacity and cycle stability of electrodes is due to the following factors: CNTs provide diffusion pathways for Li+ or Na+ ions to most of the grains inside the material, improving charge/discharge kinetics; PVDF provides good adhesion between the electrode materials and current collectors, giving excellent electrochemical, mechanical, and thermal stability of electrodes.