The composition of 75V2O5–(25-x) MgO–xLi2O (x = 0, 1.5, 3.0, 4.5, and 6.0) is synthesized by the melt quenching method. The effects of Li2O on devitrification, physical, thermal, structural, and conducting properties of the as-quenched samples are analyzed utilizing various experimental techniques. X-ray diffraction and differential thermal analysis confirmed the formation of phase-separated glasses up to 3.0 mol. % of Li2O. Above this concentration of Li2O, the samples are glass ceramic. With an increase in the concentration of Li2O, the density increases in all the samples. The Raman spectra demonstrate that as the concentration of Li2O increases, there is a transition of VO5 units into different structural units of vanadium oxide. The highest conductivity is observed for the composition with x = 4.5, i.e., 10−4 S/cm at 250 °C. The activation energy indicated that the present samples could be mixed conductors in nature. These samples can be used as cathode materials in energy storage devices due to their mixed conduction with an appropriate conductivity at 250 °C.
I. INTRODUCTION
V2O5 is a good electrical conductor that exhibits conductivity in the range of 10−3–10−5 S/cm at room temperature due to the variable transition states of vanadium, such as V5+/V4+/V3+, with its layered type crystal structure.1,2 This structural characteristic could be modified and applied in various alkali ion batteries and rechargeable vanadium redox flow batteries (VRFBs).3,4 A battery must have good conductivity and excellent storage capacity. The crystalline V2O5 is a predominant electronic conductor.5 It can be converted into mixed (electric–ionic) conductors to create disordering such as oxygen vacancies and other structural disorders such as glassy states. Hägg et al. made the first unsuccessful attempt to create glass from V2O5 in their research.6 V2O5–P2O5 glass with a high percentage of vanadium oxide is also reported.7 During the addition of alkali earth metal oxide to V2O5, VO5 crystalline polyhedral structural units are transformed into VO4 tetrahedral structural units.8 In a glass matrix, vanadyl ions often acquire threefold and fourfold symmetry.9,10 The twin-roller approach was used to create V2O5–MO (M = Mg, Ca, and Ba) glass, and a link between structural alterations and electrical characteristics was established.11 The thermal origin in the switching and formation of conducting channels in vanadium phosphate glasses is also reported.12 The addition of alkaline earth metal oxides in the vanadate network has a significant impact on electrical characteristics.13
These unpaired electrons induced polarization around the vanadium ion, which led to the formation of polarons.22 Basically, polaron hopping between different vanadium ions is responsible for conduction. It could be assumed that the presence of other components in glasses, such as alkali metal cation Li+, also contributes significantly to the overall electrical conductivity of glasses due to their lower charges and smaller sizes. Assuming that the motion of alkali ions and polarons hopping is independent of each other, one may expect electrical conduction to increase with the content of alkali ions in the system. Strong anomalies in conductivity at a certain amount of alkali are also reported.23 Alkali ion doped materials usually show a mixed conductivity, such as polaronic, electronic, and ionic conductivity, with respect to temperature.24,25 The ionic conductivity in alkali and alkaline earth containing borosilicate glasses is driven by the alkali content. It is expected that single-charged alkali ions will have higher diffusivities and simpler ion intercalation in comparison with alkaline earth ions. Different ionic radii and charges cause the site to rearrange and produce inherent flaws, which could be caused by the mixed cation effect.
In the present study, an attempt has been made to synthesize magnesium containing V2O5 glasses to promote the conduction of cathode materials in VRFBs. Moreover, these glasses can be used as a cathode material in lithium-ion batteries and rechargeable magnesium ion batteries.26,27 In this study, the addition of Li2O in place of MgO and its effects on physical, thermal, structural, and electrical properties of V2O5–MgO glasses and glass ceramics are reported.
II. EXPERIMENTAL PROCEDURE
The composition of 75V2O5–(25-x) MgO–xLi2O (x = 0, 1.5, 3.0, 4.5, 6.0) was synthesized by the melt quenching process. A stoichiometric amount of V2O5, MgO, and Li2O (mol. %) was taken and mixed using an agate mortar and pestle. The purity of all chemicals was 98%–99%. They were melted in re-crystallized alumina crucibles at 900 °C in an electric furnace. The melt was held at 900 °C for 0.5 h to increase the diffusivity among various ingredients present in the composition. Furthermore, the melts were quenched in air using two heavy copper plates. The as-quenched samples were labeled as MVL-0 (x = 0), MVL-1.5 (x = 1.5), MVL-3.0 (x = 3.0), MVL-4.5 (x = 4.5), and MVL-6.0 (x = 6.0). The standard Archimedes technique was used to determine the density (ρ) of the as-quenched sample at room temperature with xylene as the immersing liquid at 25 °C. The molar volume (Vm) was determined via Vm = M/ρ (where M is the molecular weight of the composition). The x-ray diffraction (XRD) of powdered samples was taken in the angular range of 10°–80° with a scanning speed of 10° min−1, and the step size is 0.01 on a SmartLab SE with Cu-Kα radiation (λ = 1.54 Å). The as-quenched powdered samples were subjected to differential thermal analysis (DTA) measurements using a Pyris Diamond system (PerkinElmer) in a N2 atmosphere at a heating rate of 10 °C min−1 from room temperature to 800 °C. The Raman spectra of the powder sample were measured using a LabRAM HR Evolution Raman spectrometer with an excitation wavelength of 532 nm. For conductivity measurements, the samples were dried at 150 °C for 1 h in an oven; after that, both sides of the as-quenched samples were coated with Au using a 3000FC auto-fine coater from JEOL. A Solartron impedance analyzer (Model TSVHI2/50/305) was used to measure the impedance of the quenched samples at RT–250 °C. The voltage was set to ±1 V, while the frequency was maintained between 102 and 106 Hz. The temperature was raised from the ambient temperature to 250 °C in steps of 10 °C. The temperature controller’s lowest count was 1 °C. Before performing any impedance measurements, the furnace was kept at a set temperature to provide adequate temperature stability.
III. RESULTS AND DISCUSSION
A. Density
Various physical parameters, i.e., experimental density (ρexp.), theoretical density (ρthr), molar volume (Vm), inter-ionic concentration (N), and interatomic radii (R) of the samples.
Sample id . | ρexp. (g/cm3) . | ρthr (g/cm3) . | Vm(103 × cm3) . | NLi(×10−4) Å−1 . | NMg(×104) Å−1 . | RLi (Å) . | RMg (Å) . |
---|---|---|---|---|---|---|---|
MVL-0 | 2.57 | 3.37 | 5.69 | ⋯ | 20.5 | ⋯ | 7.89 |
MVL-1.5 | 2.63 | 3.37 | 5.56 | 1.62 | 19.3 | 18.23 | 8.04 |
MVL-3.0 | 2.75 | 3.36 | 5.32 | 3.40 | 18.1 | 14.27 | 8.20 |
MVL-4.5 | 2.89 | 3.34 | 5.05 | 5.36 | 16.9 | 12.27 | 8.41 |
MVL-6.0 | 2.91 | 3.34 | 5.01 | 7.21 | 15.7 | 11.14 | 8.61 |
Sample id . | ρexp. (g/cm3) . | ρthr (g/cm3) . | Vm(103 × cm3) . | NLi(×10−4) Å−1 . | NMg(×104) Å−1 . | RLi (Å) . | RMg (Å) . |
---|---|---|---|---|---|---|---|
MVL-0 | 2.57 | 3.37 | 5.69 | ⋯ | 20.5 | ⋯ | 7.89 |
MVL-1.5 | 2.63 | 3.37 | 5.56 | 1.62 | 19.3 | 18.23 | 8.04 |
MVL-3.0 | 2.75 | 3.36 | 5.32 | 3.40 | 18.1 | 14.27 | 8.20 |
MVL-4.5 | 2.89 | 3.34 | 5.05 | 5.36 | 16.9 | 12.27 | 8.41 |
MVL-6.0 | 2.91 | 3.34 | 5.01 | 7.21 | 15.7 | 11.14 | 8.61 |
Variation in density and molar volume of MVL-0, MVL-1.5, MVL-3.0, MVL-4.5, and MVL-6.0.
Variation in density and molar volume of MVL-0, MVL-1.5, MVL-3.0, MVL-4.5, and MVL-6.0.
The ionic concentration is seen to increase with the Li2O content, while the interatomic distance between cation and anion decreases. The glass network is altered by Li2O, which also boosts iconicity. As the covalency leads to an increase in density and the structure becomes more compact due to the evolution of crystalline phases, the molar volume of the samples decreases.34 Table I shows the change in ionic concentration (Ni) and the inter-ionic distance (Ri) with Li2O concentration.
B. XRD analysis
The XRD patterns of the as-prepared samples are presented in Fig. 2. The MVL-0, MVL-1.5, and MVL-3.0 samples exhibit a completely amorphous nature. In the case of MVL-4.5, some broad and pronounced humps are observed. Meanwhile, in the case of MVL-6.0, some fully resolved XRD peaks embedded in the glassy matrix are observed. These XRD peaks were indexed with orthogonal Mg0.01V2O5 (ICCD card no. 01-089-0610) having Pmn21 space group and monoclinic Li1.2V3O8 (ICCD card no. 01-080-0071) having P21/m space group. Moreover, the stoichiometry of the formed crystalline phases in MVL-6.0 indicated that the vanadium is in mixed oxidation states. The stoichiometric chemical formula of these crystalline phases is satisfied only when the vanadium is in mixed states. According to the XRD analysis, replacing MgO with Li2O in the current system increases the devitrification in glasses. Similar outcomes were seen in our earlier research paper on samples that contained Li2O and V2O5.35
X-ray diffraction pattern of (a) MVL-0, (b) MVL-1.5, (c) MVL-3.0, (d) MVL-4.5, and (e) MVL-6.0.
X-ray diffraction pattern of (a) MVL-0, (b) MVL-1.5, (c) MVL-3.0, (d) MVL-4.5, and (e) MVL-6.0.
C. Thermal analysis
Li2O breaks down the bridging oxygen in the network former by a chemical reaction, creating NBOs. This chemical reaction shortens the average length of the macromolecular chains. When the amount of modifier is increased, the glass becomes less stable and the glass transition temperature decreases in silicate or borosilicate glasses.36
Sharp crystallization and melting occur in the case of the MVL-0 sample as shown in Fig. 3. A dip around 266 °C confirms the glass transition temperature of MVL-0 glass. In the case of MVL-1.5, MVL-3.0, and MVL-4.5, the two exothermic peaks (Tc1 and Tc2) confirm that Li2O addition may induce phase separation in the samples. This phase separation alters the properties and induces non-linearity in the glass.37 Among these two peaks, Tc1 is more intense in all the samples; this peak is used to deduce other thermal parameters. Furthermore, two or more melting peaks are also observed in these glasses, which correspond to the melting of different crystalline phases or different glass matrices at different temperatures. This tendency is more pronounced in the MVL-6.0 sample since it has two crystalline phases with an amorphous matrix. From the above-mentioned results, it can be concluded that the substitution of MgO by Li2O leads to devitrification in the glasses and forms crystalline phases along the glass matrix.
In addition, Tc − Tg, a measure of the capacity to create glass, decreases with the increase in Li2O, whereas Tm − Tc increases. The values of both the factors indicate a poor capacity to create glass with Li2O content, as shown in Table II. Furthermore, the glass forming efficiency parameter, i.e., Hruby’s parameter (Tc − Tg/Tm − Tc), was computed. MVL-0 has the highest value of Hruby’s parameter, which suggests the highest glass forming ability as compared to other samples in the present system. This observation is also confirmed in the XRD pattern of these samples.
Thermal parameters of MVL-samples obtained from DTA thermographs, where Tg represents the glass transition temperature, Tc represents the crystallization temperature, Tm represents the melting temperature, F represents the thermodynamic fragility index, and Hr represents the Hruby parameter.
Sample id . | Tg . | Tc . | Tm . | ∆Tg . | F . | Tc − Tg . | Tm − Tc . | . |
---|---|---|---|---|---|---|---|---|
MVL-0 | 266.62 | 341.31 | 656.36 | 16.28 | 0.41 | 74.69 | 315.05 | 0.23 |
MVL-1.5 | 262.33 | 306.04 | 649.49 | 22.10 | 0.28 | 43.71 | 343.45 | 0.13 |
337.53 | ||||||||
MVL-3.0 | 251.48 | 295.19 | 582.48 | 23.14 | 0.24 | 43.70 | 350.31 | 0.12 |
324.98 | 645.50 | |||||||
MVL-4.5 | 247.62 | 280.18 | 586.52 | 23.20 | 0.25 | 32.56 | 294.62 | 0.11 |
314.11 | 613.88 | |||||||
MVL-6.0 | ⋯ | 266.62 | 588.85 | ⋯ | ⋯ | ⋯ | 319.03 | ⋯ |
306.66 | 640.63 |
Sample id . | Tg . | Tc . | Tm . | ∆Tg . | F . | Tc − Tg . | Tm − Tc . | . |
---|---|---|---|---|---|---|---|---|
MVL-0 | 266.62 | 341.31 | 656.36 | 16.28 | 0.41 | 74.69 | 315.05 | 0.23 |
MVL-1.5 | 262.33 | 306.04 | 649.49 | 22.10 | 0.28 | 43.71 | 343.45 | 0.13 |
337.53 | ||||||||
MVL-3.0 | 251.48 | 295.19 | 582.48 | 23.14 | 0.24 | 43.70 | 350.31 | 0.12 |
324.98 | 645.50 | |||||||
MVL-4.5 | 247.62 | 280.18 | 586.52 | 23.20 | 0.25 | 32.56 | 294.62 | 0.11 |
314.11 | 613.88 | |||||||
MVL-6.0 | ⋯ | 266.62 | 588.85 | ⋯ | ⋯ | ⋯ | 319.03 | ⋯ |
306.66 | 640.63 |
D. Microstructure analysis
Furthermore, the results of XRD and DTA are confirmed using a field emission scanning electron microscope (FESEM). Figure 4 shows some particles embedded in the glass matrix. It confirms the formation of crystalline phases along with the glass matrix in the MVL-6.0 sample.
E. Raman spectra analysis
Figure 6 exhibits some broad bands, and shoulder bands are observed in the case of MVL-0, MVL-1.5, MVL-3.0, and MVL-4.5, which could be attributed to the glassy structure. These shoulder bands were not observed in the case of the MVL-6 sample. Rather, some sharp peaks were observed in this particular sample, which correspond to the crystalline nature. The Raman spectra show bands around 144 (B1g), 283 (B2g), 412 (Ag), 492 (Ag), 529 (Ag), 639 (Ag), 699 (Ag), 839(A3g), 998(A3g), and 1016 cm−1(A3g). Among the Raman modes mentioned, 529 and 699 cm−1 belong to V–O3 and V–O2, respectively. Due to the difference in composition, some shifting was also observed in the position of the bands and their intensity. A broad diffused peak was observed around 999 and 1016 cm−1 in the case of MVL-1.5, MVL-3.0, and MVL-4.5 samples. In contrast, MVL-6.0 sample has a significant and sharp peak at 998 cm−1. The sharp Raman band was related to a stable V–O1 bond.38,39 The major Raman band was related to the terminal oxygen stretching mode.40,41 The V–O1, V–O2, and V–O3 Raman band intensities increase with the increase in the concentration of Li2O. At lower concentrations of Li2O, there is a doubled Raman band in this region, and with increasing Li2O concentrations, this band is converted into a single Raman band. These observations are found to be in agreement with the earlier results in terms of stoichiometry and structural quality.40,41 In the vanadate system, the V5+ = O stretching mode of terminal oxygen was often assigned to a band between 960 and 1000 cm−1.42 The intensity of this band increases with Li2O content because this leads to the multilayer formation of V2O5.43 In addition, an alternative proof is offered to support the formation of flawed crystal structures to transform into stable crystal structures. Lower valence state vanadium is developed in the unit cell i.e. V–O1. The Raman band positions with the increasing concentration of Li2O are presented in Fig. 5. The lower wavenumber shifts may associate with the oxidation of the valance states of cation (V4+ to V5+) and convert NBOs to BOs. In addition, it is the induction of the creation of various polyhedra rather than the actual crystal structure polyhedra.44 An increase in intensity with respect to increasing concentration of Li2O demonstrated the ordering in the samples.
A broad shoulder band was observed at around 839 cm−1, which was relevant to isolated VO4 tetrahedral structural units on the surface of the sample, which was related to the symmetric stretching mode (VO2).45 Bands in this area were used to analyze the amorphous V2O5. However, no such band is observed in the case of MVL-6.0 as crystallization occurs in this case, which may be associated with different oxidation states of vanadium. So, instead of vanadium in V5+, it exists in a number of different oxidation states in MVL-6.0. A small shoulder band around 639 cm−1 was also observed in MVL-0, MVL-1.5, MVL-3.0, and MVL-4.5. However, it is absent in the case of MVL-6.0. This band was mostly caused by oxygen ions stretching and vibrating while bridging between the three vandal centers.38,39
The stretching of Li–O may possibly be the cause of the band at 412 cm−1. The surface vanadyl group’s deformation mode is represented by a band at 283 cm−1. Around 492 cm−1, there is a shoulder band that is connected to the stretching vibration of the V–O–V band in V2O5.38,39 With the increase in the concentration of Li2O, this band split up into two peaks: one at 492 cm−1 and another at 529 cm−1, which is explained by the presence of the crystalline V2O5 distinctive layer structure as shown in Fig. 6 for the MVL-6.0 sample. The band at 144 cm−1 was associated with the [VO5] vibration and strongly accompanied by a layered structure, which was observed at a lower wavenumber due to the heavy [VO5] unit.46 So, the presence of different bands corresponding to different structural units in the MVL-6.0 sample as compared to other samples confirms that with the increase in the concentration of Li2O, crystalline nature increases and different crystalline phases are formed. These results are also confirmed by XRD analysis and DTA of these samples.
Raman spectra of (a) MVL-0, (b) MVL-1.5, (c) MVL-3.0, (d) MVL-4.5, and (e) MVL-6.0.
Raman spectra of (a) MVL-0, (b) MVL-1.5, (c) MVL-3.0, (d) MVL-4.5, and (e) MVL-6.0.
Based on the XRD, DTA, and Raman spectra, the mechanisms shown in Fig. 7 are proposed for the present samples. Different polyhedra are responsible for nucleating the different crystalline phases with increasing concentrations of Li2O.
Change in network structure of V2O5 with the addition of alkaline earth metal oxide and alkali metal oxide.
Change in network structure of V2O5 with the addition of alkaline earth metal oxide and alkali metal oxide.
F. Conductivity analysis

In the case of MVL-0, an extra-depressed semicircle was also observed, which may correspond to the electrode polarization effect, but with the increase in the concentration of Li2O, this semicircle decreases and becomes more depressed. This is associated with the interfacial impedance that may be caused by the accumulation of Li+ at the electrodes. The equivalent impedance circuit corresponds to the semicircular portion of the plots containing a parallel combination of resistor and capacitor as given in the inset of Fig. 8. The impedance behavior is comparable to a circuit with a parallel-connected resistor and capacitor for polarization.48 The depressed semicircles are toward the Z′ axis. Consequently, a constant phase element (CPE) may be used to replace the capacitor in the equivalent circuit. The resistance is represented as Re in the equivalent circuits as shown in the inset of Fig. 8.
With an increase in temperature, the region beneath the semicircle part of the figure shrinks, indicating that the overall impedance of the samples has decreased. Furthermore, the Nyquist plot suggests that the point of intersection on the Z′ axis shifts toward the origin with the increase in temperature. Moreover, the diameter of the semicircle decreases, resulting in a decrease in resistance and an increase in conductivity. The diameter of the semicircle decreases as temperature increases. It is attributed to an increase in Li+ concentration and mobility of the Li+ ions.
This may be due to an increase in Warburg resistance (Wd), which is related to the diffusion of charge carriers at the interference of the electrode. However, this behavior is not observed in the case of MVL-1.5 glass. The sudden decrease in conductivity observed in this particular glass may be due to the strong interaction between ions and polarons. The generation of polarons and mobile electrons occur when V4+ grab moving electrons are attracted to the Li+ ions with the opposite charge. This newly generated cation–polaron pair moves as a neutral particle. Therefore, the movement of these pairs has no impact on electrical conductivity as a whole. A similar behavior is also observed for sodium molybdenum phosphate glasses.49
From the Nyquist plots, it is observed that for the MVL-6.0 sample, the area under semicircular plot increases, which suggests that the impedance increases. It could be related to inhomogeneity in this sample created by the power of two crystalline phases along with an amorphous matrix in Fig. 2.
Semicircle depicts interfacial effects at lower frequencies, while at higher frequencies, it reflects the involvement of bulk conductivity.50 The radius of semicircles shrinks as temperature increases, indicating that the current samples’ ac conductivity was a thermally stimulated process. Higher temperatures cause the thermal migration of ions to speed up, which lowers conductivity. The mobility of modifying ions (especially monovalent alkali ions such as Li+) was the main factor for determining the electrical conductivity of the glasses and glass ceramics.51 The conductivity of glasses and glass ceramics is additionally influenced by network stiffness, size, charge, mobility, cation concentration, porosity, and other parameters.52
Furthermore, the Arrhenius plots were used to determine the activation energies as well as the types of conductors in the present samples. The temperature dependent conductivity follows the Arrhenius equation as given in Refs. 52 and 53.
The slope of ln σT vs 1000/T of the Arrhenius plot is attributed to the change in the charge carriers with an increase in the temperature. The Arrhenius plot is used to calculate the activation energy (Fig. 9) by linear fitting. The graph indicated that activation energies varied between high- and low-temperature zones, as shown in Table III.
Conductivity and activation energy of all the prepared samples at low and high temperatures, respectively.
Sample id . | Conductivity (S cm−1) RT . | Ea2 (eV) (50–150 °C) . | Conductivity (S cm−1) 250 °C . | Ea1 (eV) (150–250 °C) . |
---|---|---|---|---|
MVL-0 | 2.48 × 10−6 | 0.23 | 1.23 × 10−4 | 0.38 |
MVL-1.5 | 5.68 × 10−7 | 0.28 | 5.75 × 10−5 | 0.40 |
MVL-3.0 | 1.64 × 10−6 | 0.30 | 2.53 × 10−4 | 0.52 |
MVL-4.5 | 3.26 × 10−6 | 0.31 | 4.32 × 10−4 | 0.64 |
MVL-6.0 | 3.72 × 10−8 | 0.31 | 7.95 × 10−6 | 0.65 |
Sample id . | Conductivity (S cm−1) RT . | Ea2 (eV) (50–150 °C) . | Conductivity (S cm−1) 250 °C . | Ea1 (eV) (150–250 °C) . |
---|---|---|---|---|
MVL-0 | 2.48 × 10−6 | 0.23 | 1.23 × 10−4 | 0.38 |
MVL-1.5 | 5.68 × 10−7 | 0.28 | 5.75 × 10−5 | 0.40 |
MVL-3.0 | 1.64 × 10−6 | 0.30 | 2.53 × 10−4 | 0.52 |
MVL-4.5 | 3.26 × 10−6 | 0.31 | 4.32 × 10−4 | 0.64 |
MVL-6.0 | 3.72 × 10−8 | 0.31 | 7.95 × 10−6 | 0.65 |
For ionic conductors, the activation energy typically ranges from 0.9 to 1.5 eV.54,55 Below this range, glasses and glass ceramics may have mixed conduction (electronic + ionic) followed by predominantly electronic conductivity at lower temperature regions. The calculated activation by the Arrhenius plot lies in the range of 0.38–0.60 eV for the present samples. These activation energy values suggest that the present samples exhibit mixed conductivity. Vanadium ions exist in different oxidation states and act as localized hoping centers in the present samples. Electrical conduction changes at higher temperatures as a result of structural modifications, such as crystalline phase transitions or the reduction/oxidation of the transition ions. As a result, non-linearity was observed in the Arrhenius plots of the present samples.
IV. CONCLUSION
The replacement of MgO by Li2O increases the tendency of devitrification and phase separation in the present glasses. The evolution of crystalline phases in glass network with Li2O content increases the density of samples with decreasing molar volume. The chemical formula of crystalline phases indicated that vanadium is in a mixed oxidation state in the present samples. The Raman spectra confirmed the presence of different structural units of V2O5 in the samples. The current samples exhibit two types of conduction mechanisms at high temperatures (150–250 °C) where ionic conduction predominates; however, at lower temperatures, electronic conduction dominates. The highest conductivity (4.32 × 10−4 S cm−1) is observed for MVL-4.5 at 250 °C. All samples exhibit mixed conduction with good conductivity in the required range of cathode materials for possible battery application and act as electrolytes in rechargeable vanadium redox flow batteries.
ACKNOWLEDGMENTS
This work received financial support from Thapar Institute of Engineering and Technology (TIET), Patiala. The authors acknowledge the DST-TIET for providing mutual characterization facilities under the FIST-II scheme. The authors thank Mr. Santosh Kumar for his valuable suggestions and discussion. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
AUTHOR DECLARATIONS
Conflict of Interest
The authors have no conflicts to disclose.
Author Contributions
Vimi Dua: Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). K. Singh: Conceptualization (equal); Investigation (equal); Supervision (equal).
DATA AVAILABILITY
The data that support the findings of this study are available from the corresponding author upon reasonable request.