3.1. Structural properties, mechanical and thermodynamic properties
Initially we optimized the volume of Rb2LiGa(Br/I)6. Figure 1 (a, b) shows the volume optimization curve of Rb2LiGa(Br/I)6. The lattice parameters obtained are 11.13 Å for Rb2LiGaBr6 and 11.17 Å for Rb2LiGaI6. Additionally, the relaxed Rb2LiGa(Br/I)6 systems exhibit a cubic crystal phase with an Fm3m (No. 225) space group. The optimized crystal structure of Rb2LiGa(Br/I)6 is depicted in Fig. 2 (a, b), showing that each Rb atom is positioned at the centre of the octahedral cavity, sharing twelve Br atoms. Conversely, each Li and Ga atom is positioned at the centre of their respective octahedra, coordinated with six (Br/I) atoms [15]. Detailed structural analysis reveals that the Li(Br/I)6 and Ga(Br/I)6 octahedra are alternately arranged along the [001], [100], and [010] directions. The volume optimization curve of Rb2LiGa(Br/I)6, shown in Fig. 2 (a, b), is used to calculate the lattice parameters of these systems. We have calculated the tolerance factor (τ) as follows:
The τ values for Rb2LiGa(Br/I)6 are listed in Table 1, demonstrating their stability in the cubic phase [16]. Additionally, the charge density of Rb2LiGa(Br/I)6 along the (001) plane is illustrated in Fig. 2 (c, d).
Table 1
The estimated values of the tolerance factor (τ), lattice parameter (a), bond lengths (r), and bond angles of Rb2LiGa(Br/I)6.
Systems | a (Ȧ) | τ | r (Rb -Br/I) (Ȧ) | r (Li-Br/I) (Ȧ) | r (Ga-Br/I) (Ȧ) | Bond angles (0) |
---|
Rb2LiGaBr6 | 11.13 | 0.95 | 3.63 | 2.57 | 2.87 | α = β = γ = 900 |
Rb2LiGaI6 | 11.17 | 0.94 | 3.67 | 2.59 | 2.91 | α = β = γ = 900 |
To examine how the materials behave under external forces, we evaluated the mechanical properties of the Rb2LiGa(Br/I)6 double perovskites (DP). Initially, we determined the three main elastic constants: C44, C12, and C11 as detailed in Table 2. We found that the elastic constants for Rb2LiGaBr6 are greater than those for Rb2LiGaI6. Moreover, the elastic constants for both Rb2LiGa(Br/I)6 met the essential criteria: (a) C11 > B > C12, (b) C11 + 2C12 > 0, (c) C11 and C44 > 0, and (d) C11 – C12 > 0 [32]. Using these elastic constants in equations (9) through (14), we were unable to calculate the other mechanical parameters for Rb2LiGa(Br/I)6. The final numerical values are presented in Table 2.
$$\:B=\frac{{C}_{11}+2{C}_{12}}{3}$$
9
$$\:E=\frac{9BG}{3B+G}$$
10
$$\:{G}_{V}=\frac{{C}_{11}-{C}_{12}+3{C}_{44}}{5}$$
11
$$\:{G}_{R}=\frac{5{C}_{44}({C}_{11}-{C}_{12})}{4{C}_{44}+3({C}_{11}-{C}_{12})}$$
12
$$\:G=\frac{{G}_{V}+{G}_{R}}{2}$$
13
$$\:A=\frac{2{C}_{44}}{{C}_{11}-{C}_{12}}$$
14
Table 2 shows that Rb2LiGaI6 exhibits lower values for the bulk modulus (B), shear modulus (G), and Young's modulus (E) compared to Rb2LiGaBr6, indicating that Rb2LiGaI6 is more compressible under applied stress. It is noted that ductile materials typically have Pugh's ratio (B/G) and Poisson's ratio (ν) greater than 1.71 and 0.23, respectively [33]. The numerical values for both the B/G and Poisson's ratio (ν) for Rb2LiGa(Br/I)6 are greater than the cutoff which suggest that these materials display ductile characteristics. Moreover, the anisotropic factor (A) for both materials is not precisely 1, signifying their anisotropic nature [34].
Table 2
Mechanical parameter values obtained for Rb2LiGa(Br/I)6.
Materials | C44 (GPa) | C12 (GPa) | C11 (GPa) | E (GPa) | B (GPa) | G (GPa) | A | B/G | V |
---|
Rb2LiGaBr6 | 24.33 | 7.050 | 67.14 | 64.82 | 29.15 | 27.27 | 0.76 | 1.87 | 0.27 |
Rb2LiGaI6 | 17.13 | 6.41 | 55.76 | 49.98 | 22.26 | 23.31 | 0.68 | 1.74 | 0.26 |
To evaluate thermal stability, we computed the root mean square deviation (RMSD) over time (in femtoseconds) using AIMD simulations. Ideally, the RMSD should remain at zero, but due to statistical uncertainties, some nonzero RMSD values are observed. Figure 3(a) and 3(b) illustrate these fluctuations in RMSD for both Rb2LiGaBr6 and Rb2LiGaI6. Despite these fluctuations, the minimal RMSD variations over time indicate that both systems are thermodynamically stable.
3.2 Electronic properties
Investigating electronic energy band dispersion is essential for assessing a material's suitability for photovoltaic and thermoelectric technologies. Figure 4(a) and 4(b) display the energy band dispersion spectra for Rb2LiGa(Br/I)6. Both systems exhibit similar band structures, likely due to the comparable electronic configurations of the elements in Rb2LiGa(Br/I)6. The valence band maxima and conduction band minima occur at the same symmetry point (┌). The fundamental energy band gap in these materials is direct, with estimated values of 1.19 eV for Rb2LiGaBr6 and 1.13 eV for Rb2LiGaI6. These suitable direct band gap values suggest that Rb2LiGa(Br/I)6 could be promising candidates for solar cell applications.
To gain a detailed understanding of the electronic properties of Rb2LiGa(Br/I)6, we have characterized their total and partial density of states, as shown in Fig. 5 (a-d). The valence state is primarily composed of Br-p and Ga-s orbitals in both systems. However, the conduction band minimum of Rb2LiGaBr6 is shaped by Br-p and Ga-s orbitals, while the conduction band of Rb2LiGaI6 is mainly influenced by I-p orbitals.
3.3 Optical properties
Studying optical properties is crucial for assessing a material's effectiveness in solar cells. Therefore, the optical characteristics of Rb2LiGa(Br/I)6 are analyzed across the 0–8 eV energy range. The optical absorption, conductivity, and reflectivity of a material can be determined from its frequency-dependent complex dielectric constant ε, which describes how the material responds to external electromagnetic radiation. The ɛ can be written as follows
ε(ω) = ε1 (ω) + iε2(ω) (9)
where ε1 (ω) and ε2(ω) are the real and imaginary parts of the dielectric constant, respectively. The imaginary part of the dielectric constant represents the electronic transition from the valence band to the conduction band, which can be expressed as [19]:
$$\:{\epsilon\:}_{2}\left(\omega\:\right)\:=\:\frac{8}{2\pi\:{\omega\:}^{2}\:}*{\sum\:P}_{nn{\prime\:}}^{2}\:\frac{{ds}_{k}}{{\nabla\:\omega\:}_{nn{\prime\:}}}$$
10
Here, Pnn′(k) represents the dipole matrix corresponding to the transition from the valence state n to the conduction state n′. On the other hand, Sk is the energy surface with a constant value. The frequency ωnn′ arises from the energy difference between En′(k) and En(k).
The real part of the dielectric constant, ε1(ω), can be derived using the Kramers-Kronig relation [20]
$$\:\:\:\:\:\:\:\:{\epsilon\:}_{1}\left(\omega\:\right)\:=\:\frac{2}{\pi\:\:}*\underset{0}{\overset{\alpha\:}{\int\:}}\omega\:{\prime\:}\:{\epsilon\:}_{2\:\:}\left(\omega\:\right)\:\frac{d\omega\:{\prime\:}}{{(\omega\:}^{2}\:-{\omega\:{\prime\:}}^{2})}$$
11
The real dielectric constant is key to comprehending a system's defect tolerance. Figure 6(a) illustrates the energy dependence of ϵ1(ω). Notably, the static dielectric constants ϵ1(0) for Rb2LiGaBr6 and Rb2LiGaI6 are 4.73 and 5.91, respectively. These values are consistent with Penn’s model, validating the accuracy of our calculations [21]. The relatively high \(\:{\epsilon\:}_{1}\left(0\right)\:\)values suggest that these materials exhibit good defect tolerance.
Furthermore, we determined the imaginary dielectric constant spectra as a function of energy, as depicted in Fig. 6(b). The absorption edges for Rb2LiGaBr6 and Rb2LiGaI6 are 1.24 eV and 1.16 eV, respectively. These absorption edges result from electronic transitions from Br/I-p to Ga-s orbitals. Our detailed analysis shows a sharp rise in \(\:{\epsilon\:}_{2}\) beyond the optical band gap, reaching a peak at a specific photon energy. After this peak, \(\:{\epsilon\:}_{2}\) decreases, a typical behavior of semiconducting materials.
We also analyzed the absorption spectra as a function of photon energy. Both materials show significant absorption coefficients throughout the UV-Visible range, extending beyond the optical band gap [25]. The spectra display multiple absorption peaks associated with various electronic transitions. Photon absorption creates electron-hole pairs, thereby increasing optical conductivity. Figure 6(c) illustrates how optical conductivity changes with photon energy, showing a pattern similar to that of absorption behaviour [22]. This correlation arises because absorbed photons generate charge carriers, leading to increased conductivity.
Lastly, we assessed the reflectivity as a function of incoming energy, depicted in Fig. 6(d). Both materials exhibit low reflectivity throughout the entire energy range up to 5 eV. Overall, the high absorption coefficients, strong conductivity, and low reflectivity indicate that Rb2LiGa(Br/I)6 could be excellent candidates for photovoltaic applications.
3.4 Thermoelectric properties
Currently, most energy needs are met by non-renewable and environmentally harmful sources. As a result, there is an urgent need to explore renewable and eco-friendly energy alternatives. In this context, thermoelectric technology is promising as it converts waste heat into electricity. The effectiveness of a thermoelectric material is evaluated using the thermoelectric figure of merit ZT = S2σT / (ke + kl), where S represents the Seebeck coefficient, σ is the electrical conductivity, T is the temperature, ke is the electronic part of the thermal conductivity, and kl is the lattice part of the thermal conductivity. A high ZT value, approaching unity, is characteristic of a good thermoelectric material. This can be achieved when the material has a high Seebeck coefficient (S) and electrical conductivity (σ), while maintaining low thermal conductivity (k).
The relationship between S and T is shown in Fig. 7(a). It is evident from Fig. 7(a) that S decreases as the temperature increases. This decline is attributed to the increase in carrier density with rising temperature. Moreover, the positive Seebeck coefficient (S) confirms that both systems are p-type. Additionally, both compounds exhibit high values of S across the entire temperature range, which is advantageous for potential thermoelectric applications. Figure 7(b) illustrates how electrical conductivity changes with temperature. It shows a linear increase in conductivity with rising temperature, which is attributed to the increase in carrier concentration as the temperature rises. Significantly, the high carrier conductivity in both Rb2LiGa(Br/I)6 indicates their potential as effective thermoelectric materials. Figure 7(c) illustrates the variation in thermal conductivity for Rb2LiGaBr6. The thermal conductivity of these compounds increases linearly with temperature, in accordance with Wiedemann-Franz law. Notably, both systems exhibit exceptionally low thermal conductivity. Finally, Fig. 7(d) shows the variation of ZT with temperature. It is observed that ZT decreases gradually with rising temperature. Despite this, ZT remains high throughout the entire temperature range, indicating that Rb2LiGa(Br/I)6 could be promising materials for thermoelectric applications.
3.5 Modelling of heterojunction solar cell
Motivated by higher absorption coefficient and optical conductivity of Rb2LiGa(BrI)6 , we have Modelled Au/Cu2O/Rb2LiGa(Br/I)6/TiO2/FTO solar cell. Figure 8 exhibited the p-i-n structure Rb2LiGa(Br/I)6 based solar cell. In this structure, Au works as a back contact, Cu2O as a hole transparent layer, Rb2LiGa(BrI)6 is the absorber layer, TiO2 is the hole blocking layer and FTO is the front contact. The input parameters used for simulation is summarized in Table 3. Here, \(\:\chi\:\) represents the electron affinity, Eg is the band gap, \(\:{\epsilon\:}_{\text{r}}\) represents the dielectric permittivity, \(\:{\mu\:}_{\text{e}}\) is the electron mobility, \(\:{\mu\:}_{\text{p}}\) is the hole mobility, \(\:{\text{N}}_{\text{C}}\:\:\text{r}\text{e}\text{p}\text{r}\text{e}\text{s}\text{n}\text{e}\text{t}\text{s}\:\text{t}\text{h}\text{e}\:\)conduction band density of states, \(\:{\text{N}}_{\text{V}}\)represents the valence band density of states, \(\:{\text{N}}_{\text{D}}\) is the donor concentration, and \(\:{\text{N}}_{\text{T}}\) is the defect concentration. All simulation were performed at 300 K with illumination \(\:\:of\:1000{\text{W}\text{m}}^{-2}\) and at AM 1.5G light spectrum. It is worthy to be mentioned here that Rb2LiGaI6/TiO2 and Rb2LiGaI6 /Cu2O interface layers are used to perfect a more accurate perovskite solar cell. We have considered that both interfaces have neutral defect densities of \(\:{10}^{14}{\text{\:}\text{c}\text{m}}^{-2}\).The back contact Au has a work function value of \(\:5.1\text{e}\text{V}\:\text{i}\text{s}\:\text{c}\text{o}\text{n}\text{s}\text{i}\text{d}\text{e}\text{r}\text{e}\text{d}.\:\)
Figure 9 presented the J-V spectrum of Au/Cu2O/Rb2LiGa(BrI)6/TiO2/FTO solar cell. The results are summarised Table 4. We can see that Au/Cu2O/Rb2LiGaI6/TiO2/FTO is having higher efficiency than Au/Cu2O/Rb2LiGaBr6/TiO2/FTO. Also, the obtained efficiency of Au/Cu2O/Rb2LiGaI6/TiO2/FTO is higher than the previous report [39]. Thus, this study will provide the guidelines for construction of double perovskites based efficient solar cell.
Table 3
Used parameters on SCAPS simulation
Properties | TiO2 [26 − 3] | \(\:{\mathbf{C}\mathbf{u}}_{2}\mathbf{O}\) [26, 30,31 ] | Rb2LiGaBr6 | Rb2LiGaI6 |
---|
thickness (nm) | 30 nm | 30 nm- | 600nm | 700nm |
\(\:{E}_{\text{g}}\left(\text{e}\text{V}\right)\) | 3.2 | \(\:2.17\) | 1.19 | 1.13 |
\(\:\chi\:\left(\text{e}\text{V}\right)\) | 4 | 3.0 | 3.87 | 3.94 |
\(\:{\epsilon\:}_{\text{r}}\left(\text{e}\text{V}\right)\) | 100 | 7.5 | 4.1 | 4.13 |
\(\:{N}_{\text{v}}\left({\text{c}\text{m}}^{-3}\right)\) | 1x1021 | \(\:1.1\times\:{10}^{19}\) | \(\:2.21\times\:{10}^{18}\) | \(\:2.23\times\:{10}^{18}\) |
\(\:{N}_{\text{c}}\left({\text{c}\text{m}}^{-3}\right)\) | 2x1021 | \(\:2\times\:{10}^{18}\) | \(\:5.82\times\:{10}^{18}\) | \(\:5.81\times\:{10}^{18}\) |
\(\:{\mu\:}_{e}\left({\text{\:}\text{c}\text{m}}^{2}/\left(\text{V}\text{s}\right)\right)\) | 0.006 | 200 | 100 | 100 |
\(\:{\mu\:}_{\text{p}}\left({\text{c}\text{m}}^{2}/\left(\text{V}\text{s}\right)\right)\) | 0.006 | 80 | 100 | 100 |
\(\:{N}_{\text{D}}\left({\text{c}\text{m}}^{-3}\right)\) | 0 | o | 0 | 0 |
\(\:{N}_{\text{A}}\left({\text{c}\text{m}}^{-3}\right)\) | 5.06\(\:\times\:{10}^{19}\) | \(\:1\times\:{10}^{15}\) | 10X1010 | 10X1010 |
\(\:{N}_{\text{t}}\left({\text{c}\text{m}}^{-3}\right)\) | \(\:10\times\:{10}^{15}\) | \(\:1\times\:{10}^{14}\) | 10X1010 | 10X1010 |
Table 4
The obtained solar cell parameters.
Solar cell | Jsc (mA/cm2) | Voc (v) | FF | PCE (%) |
---|
Au/Cu2O/Rb2LiGaBr6/TiO2/FTO | 32.1 | 1.06 | 78 | 23.1 |
Au/Cu2O/Rb2LiGaI6/TiO2/FTO | 32.2 | 1.19 | 79 | 25.6 |
Previour report |
TCO/IDL1/Cs2AuBiCl6/IDL2/Cu2O | 31.69 | 0.86 | 81.17 | 22.18 |