Structure and Electrochemical Performance of Glasses in the Li₂O-B₂O₃-V₂O₅-MoO₃ System

Abstract

Applying the melt quenching method (cooling rate 10¹–10² K/s), new multicomponent vanadate glasses were synthesized, containing different amounts of MoO₃ at the expense of B₂O₃ with the composition 20Li₂O:(30 − x)B₂O₃:50V₂O₅:xMoO₃, x = 10, 20 mol%. The obtained samples were characterized by X-ray diffraction, infrared spectroscopy, differential scanning calorimetry and impedance spectroscopy. The density of the glasses was measured by the Archimedes method, on the basis of which the physicochemical parameters molar volume, oxygen molar volume and oxygen packing density were calculated. It was found that the replacement of B₂O₃ with MoO₃ leads to changes in electrical conductivity, which are a consequence of the increase in non-bridging oxygen atoms in the amorphous structure. The electrochemical characterization of the 20Li₂O:(30 − x)B₂O₃:50V₂O₅:20MoO₃ glass obtained was performed by assembling an all-solid-state cell, employing 20Li₂O:(30 − x)B₂O₃:50V₂O₅:20MoO₃ glass as a cathode active material. The obtained results show that the studied glass compositions are interesting in view of their potential application as cathode materials in all-solid-state lithium-ion batteries.

1. Introduction

Currently, various types of batteries are widely used to power [1,2] consumer electronics, mobile devices, lighting components, medical equipment, etc. Batteries are used [1] in lifting and transport equipment (electric trucks, stackers, etc.), hybrid cars, electric cars, ships, and others. Stationary energy storage systems provide an alternative source of electricity for industrial and domestic users. At the same time, due to the high reliability [1,2] of lithium-ion batteries, their further development and the development of new generation samples are of great interest [3,4,5,6,7,8,9,10,11,12,13,14]. Many research groups [15,16,17,,18,19,20,21,22,23,24,25,26,27,28,29,30,,31,32] have studied the structure, physicochemical properties and electrochemical behavior of glass materials synthesized in oxide systems with the participation of various glass formers. In the Li₂O-P₂O₅-MoO₃ system, distinct regions (with different ratios and concentrations of oxide components) characterized by dominant ionic or electronic conductivity of the synthesized glasses were established [17,18,21]. The change in DC and AC conductivity [21] was studied by changing the composition of a series of phosphate glasses yLi₂O–(1 − y)[0.35(MoO₃)₂–0.65(P₂O₅)]. It was proven that modification of glassy materials from the Li₂O-P₂O₅-MoO₃ system [25], by introducing low concentrations of LiCl, causes a change in the conductivity mechanism from mixed (ionic–electronic) to predominantly ionic. In the case of glasses [19] with the composition xLi₂O–(100 − 2x)V₂O₅–xP₂O₅ (x = 15, 25, 35, 40 and 45 mol%), a transition from predominantly electronic to ionic conductivity was observed at a concentration of 40 mol% Li₂O. In glassy samples with the composition 50V₂O₅–(50 − x)P₂O₅–xA₂O (A = Li, Na, K), a decrease in DC conductivity was observed with an increase in the content of alkali oxides [24]. The effect of crystallization on the ionic and electronic conductivity of glasses prepared in the Li₂O–V₂O₅–P₂O₅ system was reported in [23]. A significant increase in electronic conductivity has been observed for a crystallized sample having composition 50Li₂O–25V₂O₅–25P₂O₅. The nanostructured features, as well as the ionic conductivity of Li_1.3Nb_0.3V_1.7(PO₄)₃ glass–ceramic materials, that may find application as cathode materials in battery production, were presented in [4]. At low V₂O₅ concentrations (0.05 ≤ x ≤ 0.5), it was discovered that the added divanadium pentoxide acts as a modifier that promotes the formation of boron-oxygen tetrahedra for glasses with compositions xV₂O₅–(1 − x)B₂O₃. When vanadium oxide content is higher (x > 0.5), a glass network composed of vanadium-oxygen tetrahedra is formed. Vanadium oxide acts as a glass former in the binary systems xMeO–(100 − x)V₂O₅ (where Me = Ba, Mg, and Sr) and creates an amorphous network of vanadium-oxygen polyhedra [20]. Vanadium oxide is mainly a modifier with the formation of VO₄ tetrahedra for (100 − x)[B₂O₃•Li₂O]–xV₂O₅ glasses_, at low x values [16]. With increasing content of V₂O₅, its structural role changes to a glass former and the formation of VO₅ trigonal bipyramids occurs [16]. A study on the local structure [28] of glasses from the Li₂O–B₂O₃–V₂O₅ system has been carried out, in which the co-ordination of boron (3 and 4) and vanadium (5 and 6) was identified, and a structural model of the amorphous network was proposed. An increase in the fraction of tetracoordinated boron was observed with increasing vanadium concentration. Glass samples [26] with compositions 0.3Li₂O–(0.7 − x)B₂O₃–xV₂O₅ and x = 0.3, 0.325, 0.35, 0.375, 0.4, 0.425, 0.45 and 0.475 were obtained, where the boron anomaly typical for borate glasses with a high Li₂O content was not detected. The dominant role of electronic conductivity, which increases exponentially (7.4 × 10⁻⁵ S cm⁻¹ at x = 0.475) with temperature and with the increase of vanadium content, has been confirmed [26]. The structure and electrochemical activity of glasses with compositions 20Li₂O-30V₂O₅-(50 − x)SiO₂-xB₂O₃ mol.% (x = 10, 20, 30, 40) were investigated [5]. For the composition of 20Li₂O-30V₂O₅-40SiO₂-10B₂O₃, the highest content of fraction V⁴⁺, increased electronic conductivity, and higher charge–discharge capacity were recorded compared to other samples. The reduction of the particle size by grinding (while preserving the amorphous structure of the material) stabilizes the charge–discharge cycle in the studied operating mode, lowers the impedance [5], facilitates the migration of Li⁺ ions and increases the conductivity. The possible states of vanadium from V⁵⁺ to V²⁺ allow [15] the realization of a complete charge–discharge cycle with adequate capacity under certain structural conditions. In most cases, vanadium ions exist in the form of V³⁺, V⁴⁺ and V⁵⁺ with different coordination environments [30]. In glassy materials, vanadium ions exhibit mainly +4 and +5 oxidation states. V⁴⁺ ions with four-coordination form tetrahedral structural units, while V⁵⁺ ions exist in the amorphous network with coordination 4, 5, and 6, allowing the formation of tetrahedra, bipyramids, and octahedra, respectively. DC conductivity in the interval 318 K–473 K was investigated for series of oxide glasses with compositions xMoO₃ + (30 − x)(V₂O₅) + 70(B₂O₃) [29]. The values were recorded to increase with the concentration of MoO₃ up to x = 0, 15 and then decrease at x > 0.15. With increasing MoO₃ concentration, a non-linear change in density and molar volume was found. Mott’s variable-range hopping model has been shown to allow adequate interpretation [29] of DC conductivity data obtained at T < θ_D/2.

Several previous studies have discussed the potential of glassy materials as solid-state battery electrolytes [33] or interfacial layers [34] due to their favorable stability against lithium metal, environmental durability, and moderate ionic conductivity. In particular, B₂O₃-based glasses, such as Li₂O–V₂O₅–SiO₂–B₂O₃, have been explored as promising cathode materials [5]. For example, composite electrodes based on V₂O₅–LiBO₂ glass and reduced graphene oxide (RGO) have demonstrated a high energy density of approximately 1000 Wh/kg, while maintaining a specific capacity of around 300 mAh/g over the first 100 cycles [3]. Building on this understanding, we designed an all-solid-state composite cathode by mixing the synthesized glass powder with a sulfide-based solid electrolyte and conductive carbon black [35,36,37]. The sulfide electrolyte, known for its high lithium-ion conductivity, mechanical softness, and excellent processability [38,39,40], plays a dual role. It minimizes interfacial resistance and mechanical mismatch between the electrode and electrolyte, and it provides efficient ionic pathways throughout the cathode. The composite can be fabricated without high-temperature sintering. This simplifies the manufacturing process and improves interfacial compatibility between the components.

The aim of the present work is to study the synthesis conditions, structure and impedance characteristics of new glassy materials from the Li₂O-B₂O₃-V₂O₅-MoO₃ system. The electrochemical performance of one selected glass composition was also investigated by assembling an all-solid-state cell, using the glass sample obtained as an active material.

2. Results and Discussion

2.1. XRD and DCS Studies

The phase formation of selected compositions from the Li₂O-B₂O₃-V₂O₅-MoO₃ system was studied by X-ray diffraction analysis (XRD). The first batch of compositions with reduced V₂O₅ content yielded opaque and visually non-uniform glasses (

Table 1. Nominal compositions, experimental conditions applied and visual observations of the quenched samples.

Sample	Li2O (mol%)	B2O3 (mol%)	V2O5 (mol%)	MoO3 (mol%)	Melting Temperature (°C)	Colling Rate K/s	Visual Assessment
1.	30	10	10	50	850	≤101–102	Hygroscopic/Non homogeneous glass
2.	30	30	10	30	850	≤101–102	Hygroscopic/Non homogeneous glass
3.	30	50	10	10	850	≤101–102	Hygroscopic/Non homogeneous glass

) (Figure 1a). According to the XRD data (Figure 1b), glass-crystalline material exhibiting an amorphous halo and diffraction peaks corresponding to the Li_0.66V₂O₅ (PDF # 01-074-0054) [41] and to the LiV₃O₈ (PDF # 01-072-1193) [41] was obtained from the ternary composition in the second series containing 50 mol% of V₂O₅ without MoO₃ (

Table 2. Nominal compositions, experimental conditions applied and visual observations of the quenched samples.

Sample	Li2O (mol%)	B2O3 (mol%)	V2O5 (mol%)	MoO3 (mol%)	Melting Temperature (°C)	Colling Rate K/s	Visual Assessment
1.	20	20	50	-	950	101–102	Glass + crystals
2.	20	20	50	10	950	101–102	Non hygroscopic/Homogeneous glass
3.	20	10	50	20	950	101–102	Non hygroscopic/Homogeneous glass
4.	20	10	40	30	950	101–102	Glass + crystals
5.	20	10	30	40	950	101–102	Glass + crystalls

). The addition of 10 and 20 mol% MoO₃ results in the formation of visually uniform X-ray amorphous samples (Figure 1b), while from the quaternary composition possessing higher MoO₃ content of 30 and 40 mol%, glass-crystalline specimens containing LiVMoO₆ (PDF # 01-082-8687) [41] as a crystalline phase were quenched (Figure 1c).

Thermal parameters of the obtained two homogeneous glass samples 20Li₂O:50V₂O₅:(30 − x) B₂O₃:xMoO₃, x = 10, 20 mol% were established by differential scanning calorimetry (DSC). Figure 2 compares the DSC curves of the two homogeneous glass samples x = 10 and x = 20 obtained.

The humps correspond to the glass transition temperature T_g, while the exothermic peaks are due to the glass crystallization temperature T_c. Their values and the calculated glass thermal stability ΔT = T_c − T_g are listed in

Table 3. Values of glass transition temperature T_g, crystallization temperatures (T_c), thermal stability ΔT and average single-bond enthalpy EB of investigated glasses.

Sample ID	Tg/°C	Tc/°C	ΔT/°C	EB/KJ mol−1
x = 10	219	245	24	612
x = 20	210	235	25	592

. As observed, T_g and T_c slightly decrease with the increasing MoO₃ content. Bearing in mind that the glass transition temperature is very sensitive to any alteration in the coordination number of network-forming atoms (i.e., short-range order) and to the formation of non-bridging oxygens [42,43], we can explain a slight reduction of T_g with the change in the short-range order and with slight depolymerization of the amorphous network (increasing number of non-bridging oxygens) with compositional changes. In addition, the values of T_g were correlated with average single bond enthalpy EB of glasses using the following relationship proposed in [14,15]:

$$EB = \frac{20E_{Li\text{-}O}+50E_{V\text{-}O}+x{E}_{Mo\text{-}O}+(30-x)E_{B\text{-}O}}{100}$$

(1)

where E_Li-O, E_V-O, E_Mo-O and E_B-O are the bond dissociation energies for the single bond Li-O, V-O, Mo-O and B-O, respectively, and the values of EB calculated are also included in Table 3. The higher EB value of the glass x = 10 can be connected with the difference in bond dissociation energies of metal oxide. Since the B-O bond enthalpy is 806 kJ mol⁻¹, exceeding Mo-O bond enthalpy (607 kJ mol⁻¹), the average single bond energy decreases with the addition of MoO₃ at the expense of B₂O₃, because of the formation of weaker Mo-O at the expense of stronger B-O bonds. Both glasses are characterized with a low thermal stability criterion ΔT of 25 °C that remains constant with the composition, indicating low thermal stability and poor glass-forming ability of the compositions [44].

2.2. IR Spectral Analysis

The structure of the two glasses obtained (x = 10 and x = 20) was investigated by IR spectroscopy (Figure 3a). The IR spectra of both glasses are extremely similar, as can be seen in the figure, and exhibit IR activity in three spectral regions: 1600–1150 cm⁻¹, 1150–850 cm⁻¹, and 850–400 cm⁻¹. To obtain more precise structural information, the spectra were deconvoluted and are displayed in Figure 3b,c.

In this way, in the first spectral region between 1600–1150 cm⁻¹ several peaks were observed due to the stretching vibration of BO₃ groups in meta-, pyro-, orthoborate structures (bands in the range 1196–1461 cm⁻¹) and in superstructural units (BO₃+BO₄) (band at 1266–1270 cm⁻¹) [45,46]. In the second spectral range 1150–850 cm⁻¹, bands related to the vibrations of BO₄ tetrahedral units (bands at 1090 and at 1072 cm⁻¹) [47], VO₄ (band at 649–651 cm⁻¹), VO₅ (bands at 1000–993 cm⁻¹) and VO₆ (band at 932–916 cm⁻¹), as well as MoO₄ (bands at 910 cm⁻¹) and MoO₆ (bands at 923–916 cm⁻¹ and at 980–971 cm⁻¹) structural groups are situated [48,49,50]. Me-O-Me, Me = V, Mo bonding is also identified, manifested by the presence of the band at 885 cm⁻¹ [51]. The third spectral region from 850–400 cm⁻¹ contains bands connected with the bending vibrations of B-O-B bonds in meta-, pyro- and superstructural borate units (band at 697 cm⁻¹), V-O-V stretching vibrations (band at 767–764 cm⁻¹), bending, asymmetric and symmetric stretching mode of V(Mo)₂O₂ entities present in V(Mo)₂O₈ units (bands at 468–472 cm⁻¹, at 573 cm⁻¹ and at 728 cm⁻¹, respectively), and also vibrations of VO₄ tetrahedral groups (band at 649–651 cm⁻¹) [45,46,47,48,49,50]. The detailed assignments of the bands observed in the deconvoluted IR spectra of the present glasses are summarized in

Table 4. Infrared bands (in cm⁻¹) and their assignments for glasses 20Li₂O:50V₂O₅:(30 − x) B₂O₃:xMoO₃, x = 10, 20 mol%.

Infrared Bands Position (cm−1)	Assignment	Ref.
468–472	δ V2O2 (VO5)	[49]
573	νs (Me2O8), Me = V/Mo	[51]
649–651	νas (VO4)	[48]
697	ν3 (MoO4)	[52]
728	νs (Me2O8), Me = V/Mo	[51]
767–764	ν (V-O-V)	[49]
885	ν (Me-O-Me), (Mo/VO6) Me = V/Mo	[51]
910	ν1 (MoO4)	[52]
923–916	ν (Me = O), (Mo/VO6) Me = V/Mo	[48,52]
980–971	ν (Mo = O), MoO6	[50]
1000–993	ν (V = O), (VO5)	[49]
1090–1072	ν BØ4−	[47]
1196; 1242–1253	ν3 BO33−	[45]
1221; 1237–1330	B-O-B stretch in pyroborate units, BØO22−;B-O− stretch in in pyroborate units, BØO22−	[45]
1266–1270; 1352, 1392–1418; 1452–1464	BØ stretch in metaborate unuts, BØ2O + νas(B–O–B); B–O–B bridges connect BO3 units ν(B–O-) stretch in BØ2O− units	[45,46]

. With increasing MoO₃ content, MoO₄→MoO₆ transformation (disappearance of the band at 910 cm⁻¹ and appearance of the band at 448 cm⁻¹), and Me-O-Me, Me = V and Mo bonding (increased intensity of the band at 885 cm⁻¹) take place [51,52]. On the other hand, some depolymerization of the borate oxygen network and formation of orthoborate BO₃³⁻ groups also occurs, evidenced by the appearance of the band at 1196 cm⁻¹ and increased intensity of the band at 1253 cm⁻¹, both characteristic for the BO₃³⁻ units [45] in the spectrum of glass x = 20.

2.3. Density, Molar Volume, Oxygen Packing Density and Oxygen Molar Volume

Structural information about the glasses obtained was also acquired by measuring the glass density and calculating the molar volume (V_m), oxygen molar volume (V_o) and oxygen packing density (OPD) of the glass system, employing the following relations respectively [53].

$${\mathrm{V}}_{\mathrm{m}} = {\displaystyle \frac{\sum {\mathrm{x}}_{\mathrm{i}}{\mathrm{M}}_{\mathrm{i}}}{{\mathrm{\rho }}_{\mathrm{g}}}}$$

(2)

$${\mathrm{V}}_{\mathrm{o}} = {\mathrm{V}}_{\mathrm{m}}\times \left({\displaystyle \frac{1}{\sum {\mathrm{x}}_{\mathrm{i}}{\mathrm{n}}_{\mathrm{i}}}}\right)$$

(3)

$$\mathrm{O}\mathrm{P}\mathrm{D} = 1000\times \mathrm{C}\times \left({\displaystyle \frac{{\mathrm{\rho }}_{\mathrm{g}}}{\mathrm{M}}}\right)$$

(4)

where x_i is the molar fraction of each oxide component i, M_i the corresponding molecular weight, ρ_g the glass density, n_i is the number of oxygen atoms in each oxide, C is the number of oxygen atoms per formula units, and M is the total molecular weight of the glass compositions. The values obtained are listed in

Table 5. Values of physical parameters of glasses 20Li₂O:50V₂O₅:(30 − x) B₂O₃:xMoO₃, x = 10, 20 mol%: density (ρ_g), molar volume (V_m), oxygen molar volume (V_o), oxygen packing density (OPD).

Sample ID	ρ (g/cm3)	Vm (cm3/mol)	Vo (cm3/mol)	OPD (g atom/L)
x = 10	2.964 ± 0.006	39.90	11.08	90.22
x = 20	3.051 ± 0.002	43.48	12.08	82.80

.

It can be observed that both density and molar volume of the glass possessing higher MoO₃ content at the expense of B₂O₃ content is greater as compared to the glass with lower MoO₃ concentration. In general, density and molar volume are inversely related. However, this anomalous behavior has been reported in other glass systems [54]. In the present system, the increase in glass density can be attributed to the increase in the average molecular mass of the glasses [55]. Since the molar volume (volume occupied by a mass of the glass equal to 1 mole) is significantly influenced by the ionic radii of the incorporated ionic species in the glass, the higher V_m for glass x = 20 as compared with V_m of glass x = 10 is due to the substitution of B³⁺ ions with smaller ionic radii (0.27 Å) by Mo⁶⁺ ions known to have a higher ionic radius (0.59 Å), resulting in the generation of excess free space, which increases the overall molar volume of this glass. Oxygen molar volume (V_o) and OPD are two parameters that provide information on the packing of the oxygen ions in the glass structure [56]. Low V_o and high OPD are indications of a high degree of network connectivity. It is observed that glass x = 20 is characterized with a greater V_o and reduced OPD values compared with the values of these parameters estimated for the glass x = 10, indicating lower degree of connectivity (increased number of non-bridging oxygens (NBOs)) of the network of the glass having higher MoO₃ content. This assumption is in line with the DSC and IR spectral data revealing the accumulation of BO₃ groups with NBOs (i.e., BO₃³⁻) in the glass structure of glass x = 20.

2.4. Diffuse Reflectance Spectra (DRS)

UV-vis diffuse spectroscopy is also used to obtained information about the local structure of glasses via ligand to metal charge transfer (LMCT) band position and the corresponding edge energy (E_g) values. Figure 4 represents the diffuse reflectance spectra (a) and the band gap energy E_g (b) and (c) of glasses under investigation.

The diffuse reflectance spectra of the studied glasses consist of bands in the range 200–600 nm, with maxima at 260, 350 nm, due to the ligand–metal charge transfer (LMCT) from oxygen ligands to Mo⁶⁺ in octahedral coordination and/or V⁵⁺ ions in tetrahedral coordination, and also bands at 380 nm due to the charge transfer band of the V = O double bond, and bands at 410 and 450 nm due to polymerized VO₆ with octahedral coordination [50,57].

The absence of any absorption in the visible range (600–800 nm) indicates that Mo⁵⁺ and V⁴⁺ species are not present in the investigated sample. Assuming the glass obtained to be indirect semiconductors, the plot of the transformed Kubelka–Munk function versus the energy of light (Tauc plot) provides band gap energies, E_g of 1.86 eV for x = 10 and 1.13 eV for x = 20 glasses, respectively. The decrease of E_g values with increase in MoO₃ content is related to the changes in bonding during vitrification, and increase in the number of non-bridging oxygens in the glass structure [58].

2.5. Electrical Properties

The samples with x = 10 and x = 20 which proved to be glasses were heated with a heating rate of 10 K/min to 180 °C, i.e., below the T_g values obtained by DSC, and then the impedance spectra were recorded at different temperatures and frequencies on cooling. After equilibrating the sample at each measurement temperature, the frequency dependence of the impedance modulus and the phase angle was measured. The two samples showed purely resistive behavior at all frequencies in the whole temperature range, i.e., the phase angle was always between 0 and −25°. Thus, the measured samples exhibited behavior which allowed assumption of the impedance modulus at 1 Hz equal to the sample resistance. From this resistance, taking into account the sample thickness and electrode surface, the dc-conductivity σ [S/m] is calculated from Equation (5):

$$\sigma = {\displaystyle \frac{1}{\rho }} = {\displaystyle \frac{1}{R}}{\displaystyle \frac{d}{S}} = {\displaystyle \frac{1}{\left|Z\right|}}{\displaystyle \frac{d}{S}}$$

(5)

where ρ [Ω.m]—sample resistivity, S [m²]– electrode surface, d [m]—sample thickness, |Z| [Ω]—impedance modulus. The impedance modulus, |Z| [Ω] and phase angle, φ [°] values at 1 Hz, as well as the calculated dc-conductivity σ and the respective activation energies, E_a, assuming, for the glasses x = 10 and x = 20 are given in

Table 6. Data from the electrical measurements: impedance modulus, |Z| and phase angle, φ at 1 Hz, calculated dc-conductivity, σ and the respective activation energies E_a for the samples x = 10 and 20.

x (mol%)	t (°C)	|Z| (Ω)	φ (°)	σ (S/m)	Ea (eV)
20	RT	1.0699 × 103	−0.03	0.0121	0.257 ± 0.001
	40	7.8543 × 102	−0.03	0.0165
	50	6.4880 × 102	−0.03	0.02
	60	4.7676 × 102	−0.04	0.0273
	70	3.5601 × 102	−0.05	0.0365
	80	2.8176 × 102	−0.05	0.0461
	90	2.1960 × 102	−0.06	0.0592
	100	1.7293 × 102	−0.05	0.0751
	110	1.3925 × 102	−0.05	0.0933
	120	1.1285 × 102	−0.06	0.115
	130	9.2586 × 101	−0.06	0.14
	140	7.7259 × 101	−0.07	0.168
	150	6.4038 × 101	−0.07	0.203
	160	5.3605 × 101	−0.08	0.242
	170	4.5850 × 101	−0.10	0.283
	180	4.3496 × 101	−0.11	0.299
10	RT	3.0617 × 104	−0.05	4.16 × 10−4	0.307 ± 0.002
	40	2.1214 × 104	−0.03	6 × 10−4
	50	1.7194 × 104	−0.03	7.41 × 10−4
	60	1.1905 × 104	−0.03	0.00107
	70	8.5299 × 103	−0.03	0.00149
	80	6.2686 × 103	−0.03	0.00203
	90	4.5696 × 103	−0/03	0.00279
	100	3.4185 × 103	−0.04	0.00373
	110	2.6089 × 103	−0.04	0.00488
	120	2.0284 × 103	−0.04	0.00628
	130	1.5909 × 103	−0.05	0.00801
	140	1.2780 × 103	−0.05	0.00997
	150	1.0293 × 103	−0.06	0.0124
	160	8.4183 × 102	−0.06	0.0151
	170	7.1487 × 102	−0.08	0.0178
	180	7.3508 × 102	−0.11	0.0173

for the different temperatures. In Figure 5, the Arrhenius plots are shown for the glasses x = 10 and x = 20. From the slopes of the Arrhenius plots, the dc-conductivity activation energy E_a is determined and also given in Table 6. The results from the estimation of the activation energies show that the sample with lower MoO₃ concentration has higher activation energy, which could be attributed to the structural changes occurring in the glass network with increasing molybdenum concentration, i.e., the occurrence of more NBOs as also observed in the IR spectra of the same glass compositions.

Furthermore, the obtained activation energy is the energy of ionic conductivity established by the Li⁺ ions under the influence of the applied external electrical field, as the results from the diffuse reflectance spectroscopy reveal that no V⁴⁺ and Mo⁶⁺ ions are present in the two studied glasses, which excludes the possibility for the existence of an electronic component of conduction. V₂O₅-containing glasses are known for their electronic conduction established by the phonon-assisted hopping of small polarons [59]; however, the hopping conduction mechanism is only possible to observe if both V⁴⁺ and V⁵⁺ ions are present in glass, as for example reported in [22,,60]. The same suggestion can be made concerning the hopping of polarons between polyvalent molybdenum ions—as only Mo⁶⁺ and no Mo⁵⁺ ions are present in the two investigated glasses, the electronic component of conduction can be excluded.

2.6. All-Solid-State Battery

The electrochemical characterization of the x = 20 glass (glass sample with lower activation energy) obtained was performed by assembling an all-solid-state cell, employing 20Li₂O–50V₂O₅–20MoO₃–10B₂O₃ glass as a cathode active material. Figure 6a illustrates the impedance characteristics of cells with two different active material contents (C35 and C60) prior to electrochemical testing. The interfacial resistance increases from 223 Ω for the C35 electrode to 308 Ω for the C60 electrode. The C35 electrode contains a higher proportion of the 75Li₂S-25P₂O₅ matrix, which provides better encapsulation and densification of the glass particles, thereby reducing the interfacial resistance between the cathode materials. Figure 6b presents the galvanostatic charge–discharge curves of the C35-based cell at two different current levels. The first cycle, conducted at 0.007 mA (corresponding to 2 mA g⁻¹) within a voltage range of 1.6–4.2 V, delivers a charge capacity of 53.42 mA h g⁻¹ and a discharge capacity of 43.31 mA h g⁻¹, with a Coulombic efficiency of 81.07%. In the second cycle, performed at 0.035 mA (10 mA g⁻¹) within a voltage range of 0.9–4.2 V, the cell exhibits a charge capacity of 24.74 mA h g⁻¹ and a discharge capacity of 24.87 mA h g⁻¹, achieving a nearly 100% Coulombic efficiency. However, a noticeable increase in the charge voltage and a decrease in the discharge voltage are observed compared to the first cycle. This behavior suggests that under higher current conditions, Li₂S in the cell becomes electrochemically active, leading to the degradation of the active material. Figure 6c displays the galvanostatic charge–discharge profiles of the C60-based cell at various voltages under a constant current of 0.012 mA (2 mA g⁻¹). In the first cycle (2.0–4.2 V), the cell achieves a charge capacity of 8.86 mA h g⁻¹, a discharge capacity of 7.53 mA h g⁻¹, and a Coulombic efficiency of 84.99%. In the second cycle (2.0–4.6 V), the charge and discharge capacities increase to 10.53 and 10.01 mA h g⁻¹, respectively, with a Coulombic efficiency of 95.06%. In the third cycle (2.0–5.0 V), further improvements are observed, with a charge capacity of 15.06 mA h g⁻¹, a discharge capacity of 14.80 mA h g⁻¹, and a Coulombic efficiency of 98.27%. The C60 cell demonstrates stable cycling performance at low current densities, maintaining effective charge–discharge behavior even under high voltage conditions up to 5 V. While the increase in voltage enhances the charge capacity, it also raises the risk of side reactions due to material limitations. To further evaluate the cycling stability, the C60 cell was reassembled and subjected to 50 charge–discharge cycles at 60 °C, within a voltage range of 2.0–4.2 V and at a current of 0.05 mA (8.33 mA g⁻¹). As shown in Figure 6e, the cell exhibits a gradual improvement in capacity over the cycling process. The charge capacity increases from 7.68 to 9.47 mA h g⁻¹, and the discharge capacity rises from 6.88 to 9.09 mA h g⁻¹, while the Coulombic efficiency improves from 89.58% to 95.99%. The higher content of 75Li₂S-25P₂O₅ in the cathode facilitates the effective encapsulation of 20Li₂O–50V₂O₅–20MoO₃–10B₂O₃ glass particles, which enhances the overall ionic transport capability and reduces the interfacial resistance of the battery. As a result, the electrode exhibits improved ionic conductivity. Therefore, under the same current density, the C35 cell demonstrates superior charge–discharge capacity compared to the C60 cell at equivalent charging voltages. However, as the current increases, the C35 cell rapidly fails. Adverse internal reactions deactivate the active material, leading to a sharp drop in discharge voltage below 2 V.

Under such low discharge voltage conditions, the Li₂S participates in redox reactions, which further compromises the battery’s stability and performance. In contrast, the charge–discharge capacity of the C60 cell is constrained by the limited voltage window, making it difficult to achieve higher capacity. Furthermore, increasing the charging voltage poses a greater risk of triggering side reactions. Nevertheless, within a safe voltage range, the C60 cell exhibits excellent cycling stability and maintains reliable electrochemical performance. These findings highlight that optimizing interfacial resistance to achieve higher charge–discharge capacities at lower charging voltages is a crucial direction for the further development of this class of cathode materials. Enhancing ionic conductivity and mitigating parasitic reactions through interfacial engineering will be essential to unlocking the full potential of glass-based cathodes for advanced solid-state batteries.

3. Materials and Methods

3.1. Sample Preparation

The glass formation ability of two series of compositions with lower and higher V₂O₅ content from the Li₂O-B₂O₃-V₂O₅-MoO₃ system is investigated by applying the melt quenching method. Reagent grade Li₂CO₃, H₃BO₃, V₂O₅ and MoO₃ are used as raw materials. The homogenized batches (5 g) from the first series of compositions with reduced V₂O₅ content (10 mol%) were melted at 850 °C for 15 min in a platinum crucible in air. The melts were cast into graphite molds (cooling rates ≤ 10¹–10² K/s). The homogenized batches (3 g) from the second series of compositions with a higher V₂O₅ concentration (30–50 mol%) were melted at 950 °C for 20 min in a platinum crucible in air. The melts were quenched by pouring and pressing between two copper plates (cooling rates 10¹–10² K/s). Figure 7 shows a schematic representation of the preparation steps adopted to obtain the glass and glass–crystalline samples under investigation. The nominal compositions investigated, the experimental conditions applied and the visually observed results are listed in Table 1 and Table 2.

3.2. Sample Characterization

The phase formation of the samples was established by X-ray phase analysis using a Bruker D8 advance diffractometer with Cu Kα radiation in the 10 < 2θ < 60° range. The glass transition (T_g) and crystallization peak maximum (T_c) temperatures of the glasses were determined by means of Differential Scanning Calorimetry (DSC) using a Netzsch 404 F3 Pegasus instrument, Germany in the temperature range 25–750 °C at a heating rate of 10 K/min in argon atmosphere. The density of the obtained glasses at room temperature was measured by the Archimedes principle using toluene (ρ = 0.867 g/cm³) as an immersion liquid on a Mettler Toledo electronic balance of sensitivity 10⁻⁴ g. The IR spectra of the glasses were measured using the KBr pellet technique on a Nicolet-320 FTIR spectrometer with a resolution of ±4 cm⁻¹, by collecting 64 scans in the range 1600–400 cm⁻¹. A random error in the center of IR bands was found as ±3 cm⁻¹. The optical spectra of powder samples at room temperature were recorded by a spectrometer (Evolution 300 UV-vis Spectrophotometer) employing the integration sphere diffuse reflectance attachment. The spectra of the samples were recorded in the wavelength (λ) range of 190–800 nm with a magnesium oxide reflectance standard used as the baseline. The uncertainty in the observed wavelength is about ±1 nm. The Kubelka–Munk function (F(R_∞)) was calculated from the UV-vis diffuse reflectance spectra. The band gap energy (E_g) for the direct allowed transition (n = 2) was determined by preparing a Tauc plot where the function (F(R_∞)hν)ⁿ was plotted versus hν (incident photon energy). The tangent to the inflection point was determined, and the band gap energy was found as the intersection of the tangent with the horizontal axis, i.e., at [hνF(R_∞)]² = 0. The electrical properties of selected samples were studied by utilizing measurement of the impedance modulus and the phase angle as a function of frequency and temperature by an impedance analyzer (Zahner IM6, Zahner Elektrik, Kronach, Germany) and two-contacting points measurement. After gold electrodes were thermally evaporated onto the sample base surface, the 1–3 mm thick samples were mounted in the holder, introduced into a resistive furnace with a DC power supply and then connected to the impedance analyzer. The temperature was measured by an Al-Ni-Cr (K-type) thermocouple with an accuracy of 1 °C. The impedance modulus and the phase angle were measured as a function of frequency in the range from 1 Hz to 100 kHz and as a function of temperature from room temperature to 180 °C, i.e., below the T_g. An AC voltage with amplitude 500 mV was applied during the impedance measurements. The accuracy of the impedance modulus measurement is ≤±5% and that of the phase angle ±0.2°. Taking into account the sample geometry, the conductivity as well as several dielectric characteristics of the measured samples were estimated.

3.3. Preparation of Cathode Materials

For the preparation of cathode materials, all sample preparation procedures were carried out in an argon-filled glovebox to prevent moisture and oxygen contamination. Two types of cathode composites were prepared: one consisting of 60 mg of 20Li₂O-50V₂O₅-20MoO₃-10B₂O₃ glass, 35 mg of 75Li₂S-25P₂O₅ solid electrolyte, and 5 mg of acetylene black (denoted as C60), and the other consisting of 35 mg of 20Li₂O-50V₂O₅-20MoO₃-10B₂O₃ glass, 60 mg of 75Li₂S-25P₂O₅, and 5 mg of acetylene black (denoted as C35). Each mixture was thoroughly homogenized by dry grinding in a small agate mortar for 10 min to ensure uniform dispersion of the active components.

3.4. Battery Cell Assembly

For Battery Cell Assembly, 100 mg of 75Li₂S-25P₂O₅ solid electrolyte was placed into a PEEK die and pressed under 80 MPa for 2 min to form a dense electrolyte pellet. After removing the upper punch, 10 mg of the pre-mixed cathode composite was evenly spread onto the electrolyte surface and pressed again under 80 MPa for 2 min. The entire assembly was then compressed under 360 MPa for 3 min to enhance interfacial contact and mechanical integrity. For the anode, indium foil with a thickness of 0.1 mm was punched into 8 mm diameter discs and placed on the counter electrode side. The full cell was further pressed under 80 MPa for 2 min to ensure good contact. The assembled cells were then fixed in a customized clamp in preparation for electrochemical testing. All electrochemical measurements were conducted at room temperature using a Princeton Applied Research PMC200 electrochemical workstation. Galvanostatic charge–discharge tests were performed within a voltage window of 0.5–6 V at various current densities ranging from 2 to 10 mA g⁻¹. The voltage limits were adjusted according to the cathode composition and the applied current density in order to evaluate the maximum charge–discharge capacities under different conditions.

4. Conclusions

Novel multicomponent vanadate glasses with the composition 20Li₂O:(30 − x)B₂O₃:50V₂O₅:xMoO₃, x = 10, 20 mol% were obtained by using the conventional melt quenching method. Several structural units building up the amorphous network, specifically VO₄, VO₅, VO₆, MoO₆, MoO₄, V(Mo)₂O₈, BO₃³⁻, BØO₂²⁻ BØ₂O⁻ and BØ₄⁻, were determined by combining IR and DRS spectroscopies. The values of the optical band gap (E_g) of the present glass samples fall within the semiconductor range (1.86 eV and 1.13 eV), which render these glass compositions intriguing for exploration as cathode materials in Li-ion batteries. The prepared glassy material exhibits excellent charge–discharge performance, delivering a reversible capacity of 43.31 mAh g⁻¹ at room temperature. Moreover, the material demonstrates remarkable cycling stability, maintaining its capacity without noticeable degradation over 50 cycles at 60 °C, highlighting its potential as an active cathode material for solid-state batteries. However, optimizing interfacial impedance and enhancing the interfacial compatibility between the cathode and solid electrolyte remain significant challenges in all-solid-state systems. This study offers new insights into the application of glassy materials as active cathode components and contributes to the development of all-solid-state lithium metal batteries.