Computational insights into transition metal-based BaCoX3 (X = Cl, Br, I) halide perovskites for spintronics, photovoltaics, and renewable energy devices Authors: Arafat Rahman, Alamgir Kabir, Tareq Mahmud Published: 2024 ABSTRACT AbstractAb-initio simulations using density functional theory (DFT) were employed to investigate the structural, mechanical, electronic, magnetic, optical, and thermoelectric properties of halide perovskites (X = Cl, Br, I). Structural optimization and mechanical stability assessments confirm the reliability of these perovskites in a hexagonal Pmc symmetry. The stability of the ferromagnetic phase was validated through total crystal energy minimization via Murnaghan’s equation of state. Electronic band structures and density of states, derived from the generalized gradient approximation (GGA), reveal a semiconducting ferromagnetic nature in the spin up channel, spotlighting their potential in semiconductor spintronic applications. Phonon dispersion analysis of and revealed positive phonon modes throughout the entire Brillouin zone, confirming their dynamical stability. In contrast, demonstrated dynamical instability. The elastic constants confirm the mechanical stability and ductile nature of the perovskites. Optical and dielectric properties of these perovskites show significant UV absorption and photoconductivity, making them highly suitable for optoelectronic and solar cell applications. Finally, transport properties, such as the Seebeck coefficient, electrical conductivity, thermal conductivity, power factor, and figure of merit (ZT) unveil their exceptional thermoelectric performance. Combining half-metallic ferromagnetic traits with superior thermoelectric and optoelectronic performance positions compounds as exceptional candidates for applications in spintronics, optoelectronics, and thermoelectrics. This comprehensive investigation demonstrates their ability to excel across a diverse array of advanced technological applications.Keywords: Half-metallicity, Elastic properties, Optoelectronic, Thermoelectric, Phonon dispersion, DFTSubject terms: Electronic properties and materials, Magnetic properties and materials, Semiconductors, Spintronics, Materials for energy and catalysis, Materials for optics, Theory and computation The concept of half-metallic ferromagnets was first introduced in 1983 by R.A De Groot et al.1. Following this significant discovery, half-metallic ferromagnets rapidly became a focal point of theoretical and experimental research due to their potential applications in spintronics, magnetic sensors, magnetic random access memory (MRAM) and quantum computing.2–6 A half-metallic material displays metallic behavior in one spin channel and exhibits insulating or semiconducting characteristics in the other. This unique characteristic arises due to the presence of a band gap in proximity to the Fermi level, resulting in a complete spin polarization at Fermi level.7,8 The magnitude of spin polarization (SP) evaluated at the Fermi level is defined by, SP 9. Unlike conventional currents, which lose energy as heat, spin-polarized currents do not dissipate energy as heat, making them highly suitable for electronic devices that offer reduced power consumption, faster data processing speeds and cost-effectiveness10. Diamagnetic and paramagnetic materials possess symmetric densities of states (DOS) for spin-up and spin-down channels, hence they do not show spin polarization. In contrast, ferromagnetic materials exhibit asymmetric DOS; therefore, their spin polarization (SP) is always greater than zero but less than or equal to one. This property enables ferromagnetic materials to produce highly spin-polarized currents. As a result, the antiparallel magnetization alignment in half-metallic ferromagnetic layers results in extremely high resistance, while the parallel configuration yields zero resistance11. Therefore, the introduction of these materials into GMR (Giant magnetoresistance) devices would also lead to a significant increase in the efficiency of spintronic devices. Perovskite materials, primarily existing as oxides with a general formula , have been extensively studied due to their diverse chemical and physical properties. Here, A and B represent cations, while X denotes an anion, typically comprising oxides or halogens. The element A is commonly an alkali or alkaline metal, whereas B encompasses transition or inner transition elements. These materials have shown significant promise in various applications, including spintronics, optoelectronics, and thermoelectrics. In recent years, numerous half-metallic ferromagnets have been identified, including Heusler alloys (half, full, and quaternary12,13), 14, spinel 15,16, transition metal oxides such as 17, double perovskites like 18 and 19, and halide double perovskites (; M = Mn, Co, and Ni)20. Additionally, many metal halide perovskites have demonstrated half-metallic properties, with high spin-polarization at the Fermi level.21–24 Among these materials, those with transition metals at the B-site stand out for their remarkable half-metallic ferromagnetic behavior. For instance, compounds such as (M = Ir, Rh), (X = Cl, Br, I), (X = F , Cl, Br, I), and exhibit half-metallicity, with transition metals like Cr, Rh, and Ir playing a crucial role in their conducting properties25–28. Remarkably, the A-site atom does not affect the electronic structure of these materials, as the half-metallic character remains unchanged when varying the A-site from K to Rb to Cs. Furthermore, the discovery of ultralow thermal conductivity in 29 marked a significant breakthrough in the study of halide perovskites, driving extensive research into their transport properties. These materials are well-regarded for their exceptional optoelectronic and thermoelectric attributes, positioning them as highly promising candidates for next-generation energy applications30–34. Their room temperature Seebeck coefficient has been found to surpass that of other solution-processed materials, such as organics, making them even more attractive for thermoelectric devices35. Singh et al. recently demonstrated the high thermoelectric performance of (X = F and Cl) halide perovskites, reporting figure of merit (ZT) values of 2.4 for p-type and 2.1 for p-type at room temperature36. Further studies on other halide perovskites, such as (X = Cl, Br, I)37, (X = Cl, Br)38, and (X = Cl, Br, I)30, have highlighted their potential for thermoelectric efficiency. For instance, and were found to exhibit ZT values of 0.69 and 0.75 at room temperature, with improved performance at higher temperatures up to 1200 K. Investigations into the transport properties of and has demonstrated substantial thermoelectric potential, with these materials achieving a ZT value close to 0.9939. Moreover, halide perovskites featuring Ba at the A-site, such as (X = K, Rb), (X = Cl, Br, I), and (M = Ag, Cu), shows excellent optical properties40–42, while is characterized by its superior optical and thermoelectric properties43. These observations suggest that the inclusion of Ba significantly enhances the multifunctional performance of these perovskites. Despite the numerous applications of halide perovskites, there is a noticeable gap in the literature on perovskites featuring Ba at the A-site, combined with Co at the B-site, and a halogen anion (X = Cl, Br, I) at the X-site. This gap presents a significant opportunity to explore these materials, as the unique compositional structure of these (X = Cl, Br, I) perovskites could lead to exceptional electronic properties and diverse potential applications. Therefore, this study aims to explore the versatility of in advanced technologies such as spintronics, optoelectronics, and thermoelectrics using Density Functional Theory (DFT). Computational method The structural parameters, alongside the electronic, magnetic, elastic, and optical properties of (X= Br, Cl,I) perovskites, are determined by using spin-polarized density functional theory calculations with the Vienna Ab-initio Simulation Package (VASP)44–47. Projector Augmented Wave (PAW) potentials are utilized to model interactions between electrons and atomic cores which enhances the accuracy of DFT calculations by effectively representing the electron density near the atomic cores through augmented wavefunctions48. PAW potentials enable a more accurate description of electronic properties while also accounting for core electron interactions. The geometrical optimizations are carried out using the Generalized Gradient Approximation (GGA) for exchange-correlation interactions into the Perdew-Burke-Ernzerhof (PBE) functional49,50. The rationale for choosing the GGA-PBE functional is its computational efficiency and its much better accuracy than LDA functional over a wide range of systems. Conjugate-gradient algorithm is employed for the relaxation of ions into their lowest energy state configurations. The energies and forces on each ion are converged to within eV/atom and 0.02 eV/Å. The cutoff energy for the plane-wave basis set is extended up to 350 eV which is 1.3 times the maximum energy cutoff specified in the pseudopotentials. VASPKIT’s Monkhorst-Pack51 scheme is used to generate a 9 9 6 grid, considering the symmetry of the hexagonal structure.The lattice parameters were obtained by geometrical optimization using VASP, with the ISIF tag set to 3. This specific setting permits the relaxation of atomic positions, cell shape, and cell volume, thereby allowing all these parameters to vary during the optimization process. These criteria are standard in similar DFT studies and are sufficient to achieve accurate results52–55. Our calculations treat the following electrons as valence electrons: Ba-5 5 6, Co-3 4, Br-4 4, Cl-3 3 and I-5 5. To determine the thermodynamically stable configuration of hexagonal perovskites , structural optimization is conducted using Murnaghan’s equation of state56: where , , and represent the energy, bulk modulus, volume and pressure derivative of , respectively, in the equilibrium state. The orientational dependence of various elastic parameters was analyzed using ELATE, an elastic tensor analysis tool.57 Key mechanical parameters, such as the bulk modulus (B), Young’s modulus (E), shear modulus (G), and Poisson’s ratio (), were obtained through VASPKIT51. The transport coefficients were determined using the semiclassical Boltzmann theory as implemented in the BoltzTraP2 code58. Finally, the dynamic stability was confirmed through phonon dispersion curve using phonopy package59,60. Results and discussion Structural properties The (X = Cl, Br, I) perovskites examined in this study features a hexagonal crystal structure under the space group Pmc (No. 186). The optimized unit cell structures for each perovskite are illustrated in Fig. 1(a–c), showing the arrangement of Ba, Co, and X atoms at specific Wyckoff positions. Ba atoms are located at the 2b-Wyckoff coordinates (0.333, 0.666, 0.248), Co atoms at the 2a-Wyckoff coordinates (0.000, 0.000, 0.989), and X atoms at the 6c-Wyckoff coordinates (0.182, 0.344, 0.739). Energy-Volume optimization curves along with the unit cell structure of (a) (b) and (c) perovskite determined via spin-polarized calculations. Both the magnetic and non-magnetic phases for all the three perovskites were examined and the magnetic phase is found to be energetically more favorable than the non-magnetic phase. Figure 1(a–c) illustrates the optimization curves for both non-magnetic and magnetic phases of hexagonal perovskites. The parabolic shape observed in the optimization curves in Fig. 1(a–c) indicates the stability of these perovskites in the Pmc crystal structure. Since the magnetic phase consistently exhibits lower energy than the non-magnetic phase and is therefore thermodynamically more stable, we have used the magnetic phase for the evaluation of the remaining properties of the perovskites. The optimized structural parameters for the magnetic phase are summarized in Table 1. Table 1Optimized structural parameters viz. lattice constants (a, b, c) in Å, volume of unit cell in , ground state energy in eV, bulk modulus in GPa, derivative of bulk modulus and bond lengths (Ba-Co, Ba-X) of (X = Cl, Br, I) perovskites in ferromagnetic phase.Perovskitea (Å)b (Å)c (Å)Volume () (eV)/ atom (GPa)Bond lengthBa-Co (Å)Ba-X (Å)7.0587.0585.369231.61−4.30142.053.754.2863.3137.4817.4815.593271.13−3.90336.674.414.5353.4707.9937.9936.015332.82−3.47529.613.674.8473.734 Optimized structural parameters viz. lattice constants (a, b, c) in Å, volume of unit cell in , ground state energy in eV, bulk modulus in GPa, derivative of bulk modulus and bond lengths (Ba-Co, Ba-X) of (X = Cl, Br, I) perovskites in ferromagnetic phase. To be more specific, Fig. 1(a) shows that for , the magnetic phase is 0.1774 eV lower in energy than the non-magnetic phase. Similarly, for and , the energy differences between the magnetic and non-magnetic phases are 0.1627 eV and 0.1260 eV, respectively, as shown in Fig. 1(b,c). Additionally, the equilibrium cell volume for in the non-magnetic phase is 304.79 , whereas in the magnetic phase, it increases to 332.82 . A similar trend is observed for , with the equilibrium cell volume increasing from 252.59 in the non-magnetic phase to 271.13 in the magnetic phase. These significant increases in cell volumes indicate a strong magnetoelastic effect, where the magnetization induces lattice deformation. In contrast, exhibits less pronounced changes in equilibrium cell volume, with values of 222.19 in the non-magnetic phase and 231.61 in the magnetic phase. These variations suggest that the degree of magnetoelastic coupling differs among these materials, with and showing more pronounced structural responses to magnetization. This behavior highlights how these perovskite structures are influenced by changes in their magnetic states, which in turn affect their structural properties. Electronic and magnetic properties The spin-polarized electronic and magnetic properties of (X = Cl, Br, I) perovskites were investigated using the optimized structural parameters and employing the PBE-GGA functional approximation. Figure 2 illustrates the spin-polarized electronic band structures of the compunds, alongwith the corresponding DOS plots for its constituent atoms. The band structures of each perovskite in the spin-down channel clearly shows metallic behavior with bands crossing at the Fermi level. In contrast, the spin-up channel displays semiconductor characteristics, where the valence band maximum (VBM) and conduction band minimum (CBM) are seperated by a bang gap As depicted in Fig. 2(a), has a valence band maximum (VBM) located at and a conduction band minimum (CBM) at , resulting in a band gap of . Figure 2(b) shows that for , the VBM is at and the CBM is at , leading to a band gap of . In Fig. 2(c), displays a VBM at and a CBM at , corresponding to a band gap of . These results highlight the semiconducting nature of these materials in the spin-up channel. Therefore, these perovskites exhibit 100% spin polarization at the Fermi level, as evidenced by the presence of bands at the Fermi level in one spin channel and absence in the other spin channel. Spin-polarized band structure and density of states determined via GGA calculations: (a) BaCoCl3, (b) BaCoCl3, and (c) BaCoCl3, with the Fermi level set to zero. The distribution of energy bands reflects the semi-conducting character of the spin-up channel and metallic character of the spin-down channel. An important parameter in the context of spintronics is the half-metallic (HM) gap, which is defined as the energy difference between the bottom of the conduction band (since it is closer to the Fermi level as compared to the valence band for the semi-conducting state) and the Fermi level61. It represents the minimum energy required to induce spin excitation and generate holes in the minority spin channel. Usually, a small HM gap implies that half-metallicity is unstable and may be weakened under slight lattice strain. Hence, half-metallic materials with large HM gaps are preffered for practical spintronic applications62. According to the data presented in Table 2, exhibits the largest half-metallic (HM) gap of 0.630 eV, followed by with an HM gap of 0.418 eV, and with the smallest HM gap of 0.404 eV. also demonstrates a significantly larger HM gap compared to other half-metals, including half-Heusler compounds63, (M = Mg and Ca)10, and CoCrP and CoCrAs64. This enhanced HM gap suggests its superior suitability for spintronic applications, where a larger HM gap contributes to increased stability and efficiency in spin-dependent electronic processes62. Moreover, the large HM gap in indicates that this compound could outperform other materials in practical spintronic applications, such as magnetoresistive-based biomedical devices, nanomagnetic logic, spin-torque oscillators, quantum computing, spin field-effect transistors (spin FETs), and racetrack memory10,65–67. Therefore, the substantial HM gap observed in makes it the most suitable candidate for spintronic applications. Table 2Electronic and magnetic properties of the optimized (X = Cl, Br, I) perovskite structures viz. band gap (eV), half-metallic gap (eV), and total magnetic moments in units of .Materials (eV)2.01701.58161.1033HM gap (eV)0.6300.4040.4180.0040.0040.0041.8621.8641.8110.1050.1050.1200.0290.0270.0652.0002.0002.000 Electronic and magnetic properties of the optimized (X = Cl, Br, I) perovskite structures viz. band gap (eV), half-metallic gap (eV), and total magnetic moments in units of . To clearly understand the contributions of individual orbitals in the electronic band structure and the hybridization of the orbitals, the total DOS along with the partial DOS projected on to various atomic orbitals are examined. As illustrated in Fig. 3(a), the valence band of ranges from −8.5 to −4.5 eV and is mainly populated by Cl-3p states, with a slight presence of Co-3d states. The conduction band of extends from 0.6 to 7.5 eV, with the lower half (0.6 to 4 eV) being highly occupied by Ba-5d states and the upper half (4 to 7.5 eV) being occupied by both Cl-3p states and Ba-5d states. In Fig. 3(b), the valence band of extends from −8.0 to −3.5 eV, primarily occupied by Br-4p states with a minor contribution from Co-3d states. The conduction band of extends from 0.65 to 4.4 eV, predominantly occupied by Ba-5d states at the lower half of the conduction band, with an admixture of Br-4p states and Ba-5d states at the upper half. Contrarily, the valence band of the spin-up component of extends from −7.0 to −0.2 eV, while the spin-down component ranges from −7.0 to −2.0 eV as shown in Fig. 3(c). Both bands being significantly populated by a mixture of I-5p states and Co-3d states. The conduction band extending from to 4.1 eV is highly occupied by Ba-5d states, with a minor contribution from I-5p states. Figure 2(c) also illustrates that the narrower band gap of compared to and is primarily due to the Co-3d states in extending to higher energies and coming closer to the Ba-5d states. This proximity increases orbital overlap and hybridization, resulting in a smaller band gap for relative to and . Projected density of states for: (a) , (b) , and (c) It is evident that the Co-d states are responsible for the ferromagnetic half-metallic character of these perovskites. Positive values of the PDOS correspond to the spin-up channel, while negative values indicate the spin-down channel. It is noteworthy that the Co-3d states in each of the three (X = Br, Cl, I) perovskites cross the Fermi level for the spin-down channel, indicating partial occupancy and metallic character. In contrast, for the spin-up channel, the Co-3d states are located below the Fermi level. Therefore, the half-metallic ferromagnetic properties observed in perovskites arise from the presence of Co-3d states near the Fermi level. The total magnetic moment ( ) and the magnetic moments of each atomic site of the perovskites were computed utilizing the GGA-PBE functional. The total magnetic moment of all (X = Cl, Br, I) perovskites is 2 as shown in Table 2. This integral magnetic moment value further affirms the half-metallic nature of these perovskites.68 The positive magnetic moment observed in different atoms indicates that their spins are aligned parallel to each other. The magnetic moment of the (X = Cl, Br, I) perovskites is primarily due to the unpaired electrons in the Co 3d-states. The Co atoms, having the highest number of unpaired electrons among the components, are the principal contributors to the total magnetic moment. In these materials, the Co atom’s contribution to the magnetic moment is approximately 1.86 for and , and slightly less for at 1.81 . Contributions from other atoms, specifically Ba and X (X= Cl, Br, I), are minimal, with Ba contributing around 0.004 and X contributing between 0.105 and 0.120 across the different compounds. Thus, the overall magnetic moment is predominantly influenced by the unpaired electrons in the Co 3d- states. Optical properties To explore the photonic and optoelectronic applications of (X = Cl, Br, I) perovskites, we have analyzed their optical properties over the energy range of eV. The dielectric properties can be expressed using the Kramers-Kronig relation for the complex dielectric function 69,70, where is the real part that helps to determine the polarization and dispersion of light, while is the imaginary part that describes the absorptive properties of the material. These components and form the basis for understanding the optical characteristics of these materials such as refractive index , extinction coefficient , optical conductivity , absorption coefficient , reflectivity as well as energy loss spectra The relation between the dielectric constants and other optical parameters are given by:71–74 The distinctive peaks that emerge in the optical spectra correspond to specific electronic transitions between the valence and conduction bands.75 As presented in Table 3, the greater static dielectric constant of implies that it possesses the highest polarizability compared to the other perovskites. Furthermore, the values of show an inverse relationship with the band gap energy of the compounds, consistent with the Penn model 76. An analogous trend was observed by Redi Kristian et al., where exhibited a larger static dielectric constant compared to and 77. The static dielectric constant significantly influences the charge carrier recombination rate, which in turn affects the performance of optoelectronic devices.78 Materials with a high dielectric constant exhibit a lower charge carrier recombination rate, thereby enhancing device performance. Therefore, , with its larger static dielectric constant, is anticipated to have superior optoelectronic properties compared to other perovskites. It can also be seen from Fig. 4(a) that the real part of the dielectric function, , starts to increase with the increase in energy and reaches the maximum value of 5.40 at 5.55 eV (), 5.45 at 2.39 eV () and 6.42 at 2.61 eV (). It then decreases, reaching a minimum value of -0.94 at 12.54 eV (), −1.68 at 9.06 eV () and −2.56 at 6.6 eV (). The negative values of the dielectric constant in the ultraviolet region of the electromagnetic spectrum indicate metallic nature, resulting in maximum reflectivity as electromagnetic waves are reflected back by the medium in this energy range73. The value of again starts to increase and approaches unity for higher energy values. Figure 4(b) displays the imaginary part of the dielectric function, , which characterizes the absorption properties of the perovskite materials and is intrinsically linked to their electronic band structure. The onset of occurs at photon energies around 1.72 eV for , 1.20 eV for , and 1.04 eV for , correlating well with the materials’ band gap energies. This indicates that the materials begin to absorb photons at energies corresponding to their band gaps, leading to electronic transitions from the valence band to the conduction band. Table 3Optical properties of (X=Br, Cl, I) perovskite structures at zero energy, including static dielectric constant , static refractive index n(0), and static reflectivity R(0).Perovskiten(0)R(0) %4.572.1413.154.792.1913.835.742.4016.91 Optical properties of (X=Br, Cl, I) perovskite structures at zero energy, including static dielectric constant , static refractive index n(0), and static reflectivity R(0). Variation in the optical properties of perovskites (X=Br, Cl, I) against photon energy: (a) Real part of Dielectric constant; (b) Imaginary part of Dielectric constant; (c) Refractive index; (d) Extinction coefficient; (e) Optical conductivity; (f) Absorption spectra; (g) Reflectivity; (h) Electron energy loss spectra . Prominent peaks in are observed around 7.00 eV for , 5.91 eV for , and 4.75 eV for . These peaks indicate significant interband transitions primarily involving Co-3d states, as the projected density of states (PDOS) reveals the presence of Co-3d states within these energy intervals. Additionally, the sharp decline in for suggests minimal interaction between the material’s surface and the incident electromagnetic radiation79. At energies above 20 eV, approaches zero, suggesting that these perovskites exhibit transparency with only minor absorption at high energy ranges. This transparency at high energies implies that the materials do not support electronic transitions at these energies, likely due to the absence of available electronic states for such transitions. As expected, the refractive index of as a function of energy closely mirrors the shape of real part of the dielectric function , as depicted in Fig. 4(c). The calculated static refractive index n(0) is presented in Table 3. This table clearly indicates that n(0) follows an inverse relationship with the band gap, which aligns with the findings as deported by N.M. Ravindra et al.80. The values of are notably high in the range 2.5−6 eV, due to the presence of halogen (X = Cl, Br, I) p-states at this energy range. Furthermore, the maximum values of refractive index are observed to be 2.45 near 6.09 eV for 2.43 near 5.02 eV for , and 2.60 near 2.56 eV for . It is significant to highlight that due to high refractive index of and in the ultraviolet (UV) region, they can serve effectively as inner layer coatings in UV-absorbing devices81. Subsequently, the refractive index gradually decreases and falls below unity at high energies, indicating weakened interaction between incident electromagnetic radiation and the material. The variation of the extinction coefficient as a function of photon energy is shown in Fig. 4(d). The materials show sharp peaks at 7.09 eV for , 6.12 eV for , and 6.46 eV for , indicating their significant capability for UV light absorption. This UV absorption range enables deep penetration of incident photons into the compound, promoting maximum diffusion of photons within this spectral region. Therefore, the material exhibits considerable potential for photocatalysis applications, attributed to its strong optical absorption in the UV spectrum of electromagnetic radiation82. Optical conductivity reflects the generation of free charge carriers resulting from bond breaking due to electron-photon interactions, such as the photoelectric effect79. Greater light absorption correlates with higher optical conductivity, as photons supply the necessary energy for bond dissociation. The photoconductivity spectra of is illustrated in Fig. 4(e), showing distinct peaks in the UV region for all the materials. It is noteworthy that exhibits the maximum optical conductivity at 7.02 eV with a magnitude of 3.76 , highlighting its strong response to UV electromagnetic radiation. This high optoconductive response of indicates that it is a promising candidate for ultraviolet (UV) sterilization, a disinfection method that uses UV light to kill or damage microorganisms, such as bacteria, viruses, and fungi. The absorption spectra reflects the attenuation of light intensity as it traverses a unit distance in an absorbing medium. Figure 4(f) shows the absorption spectrum of (X = Cl, Br, I), with primary peaks observed at 12.41 eV for , 9.05 eV for , and 6.55 eV for . Hence, all the compounds have high absorbance and minimal electron losses in the UV region. As a result, these perovskites serve as efficient UV absorbers, counteracting the harmful effects of UV radiation and offering potential applications in UV phototherapy and photovoltaic cells. Furthermore, the superior absorption coefficient along with their greater optical conductivity compared to other halide perovskites such as inrognaic Ge-Based Halide perovskites83, ( X = Cl, Br)84, ( X = F,Cl)85, (B=Ge,Sn; X=Cl, Br, I)86, and 87, highlight their potential for optoelectronic applications requiring efficient UV absorption. As shown in Fig. 4(f), the absorption coefficient remains negligible across the visible spectrum and increases progressively in the UV region. This optical transparency in the visible region suggests a significant potential for using these materials as a hole transport layer (HTM) in solar cells88,89. Reflectivity is an optical property that provides insights into how the surface of a material interacts with electromagnetic radiation. It is defined as the amount of light energy reflected from the surface of a material with respect to the amount of light energy incident on the surface of the material. The reflectivity spectra of (X = Cl, Br, I) is illustrated in Fig. 4(g). In can be noted that, exhibits the highest reflectivity within the 0–25 eV range, followed by and . This trend aligns with findings reported in similar studies38,77,83,90. The reflectivity at zero photon energy for the studied perovskites is presented in Table 3. It is observed that in the visible region, reflectivity of all the compounds is below 20%. This low reflectivity, coupled with minimal absorption in the visible spectrum further supports their application as hole transport layer (HTM) in solar cell technologies as mentioned earlier. Notably, the larger values of reflectivity are achieved where attains negative values indicating metallic behavoiur in the compounds. Furthermore, the reflectivity of the studied halide perovskites is greater compared to other halide perovskites, such as CsPbX (X = Cl, Br, I)91, double perovskite CsYInX (X = Cl, Br, I)90, and CsPdX (X = Cl, Br)84. The electron energy loss spectra (EELS) describe the energy loss of a fast-moving electron within a medium. When electrons in the material are excited by absorbed photon energy, a high-speed electron loses part of its energy, inducing collective electron oscillations identifiable in the spectra70. Fig. 4(h) illustrates multiple peaks in the UV region of the spectra, indicating significant electron losses in this region compared to the visible region. When the incident frequency of the electromagnetic wave is lower than the plasma frequency, electrons in the material can move rapidly enough to generate plasma oscillations that effectively screen or cancel out the external electric field92. This results in an increased reflectivity, characterized by peaks in the UV region of spectra that correspond to minima in the reflectivity spectrum (Fig. 4(g). Elastic properties To manufacture materials, it is crucial to understand their elastic and mechanical properties since these materials are subjected to various forces during fabrication and processing. Knowledge of these properties aids in optimizing material selection, improving fabrication techniques, and predicting material behavior under operational conditions. Therefore, in our comprehensive analysis of the elastic and mechanical properties of (X = Cl, Br, I) hexagonal perovskites, various parameters were explored to understand their structural stability, bonding nature, and potential applications. To date, no experimental or theoretical data for the elastic constants of the (X = Cl, Br, I) compound are available to be compared with our theoretical results. Thus, our results can provide reference data for future investigations. As shown in Table 4, the bulk modulus values calculated using the relation B = are 42.040 GPa for , 35.130 GPa for , and 32.250 GPa for . These values are in good agreement with those obtained from the Murnaghan equation of state, which are 42.05 GPa for , 36.67 GPa for , and 29.61 GPa for , as presented in Table 1. This consistency suggests the reliability and accuracy of the predicted elastic constants for the (X = Cl, Br, I) compounds. To verify the mechanical stability of all three perovskites, we employed the Born stability criteria.93 The elastic constants must satisfy the following criteria for mechanical stability: , , , , , and . The calculated elastic constants for each perovskite, as presented in Table 4, meet these conditions, ensuring their mechanical stability. To determine whether the material is ductile or brittle and to understand the angular nature of atomic bonding, we have calculated the Cauchy pressure ()94. As shown in Table 5 calculated values of the Cauchy pressure indicate that the materials are ductile and exhibit metallic bonding95,96. Table 4Elastic constants obtained from DFT calculations and the average elastic parameters of perovskites, evaluated using the Voigt-Reuss-Hill method.PerovskiteElastic Constant (GPa)Bulk Modulus B (GPa)Young’s Modulus E (GPa)Shear Modulus G (GPa)Poisson Ratio 58.12931.19434.83660.3807.08242.04041.93241.98729.28026.99328.13910.5809.69110.1340.3800.3930.38846.57224.15229.17558.04210.23635.13034.27034.70129.67029.40229.53810.92010.83310.8740.3600.3570.35851.58122.59724.89742.3292.41432.25032.07132.16224.04013.38918.8188.7404.6806.7090.3800.4300.402 Elastic constants obtained from DFT calculations and the average elastic parameters of perovskites, evaluated using the Voigt-Reuss-Hill method. Table 5Calculated Pugh’s ratio (B/G), Cauchy pressure (GPa), Kleinman’s parameter , and average wave velocity, v ().PerovskitePugh’s ratio (B/G)Cauchy pressure (GPa)Kleinman’s parameterAverage wave velocity (m/s)Debye Temperature (K)4.1424.100.891717.45180.13.1913.900.861606.15159.24.7920.200.741194.80112.3 Calculated Pugh’s ratio (B/G), Cauchy pressure (GPa), Kleinman’s parameter , and average wave velocity, v (). The bulk modulus (B), which quantifies a material’s resistance to volumetric deformation under pressure, varies among the compounds. exhibits the lowest bulk modulus (32.071 GPa), indicating its relative flexibility and ability to undergo significant volume changes under pressure. This property makes a promising candidate for applications like thin-film fabrication in solar cells, where flexibility and resistance to pressure are essential86,97. It is noteworthy to mention that there exists a clear proportionality between the band gap of the materials and their bulk modulus, a finding consistent with the results reported by Reddy et al.98. Young’s modulus (E) characterizes the stiffness and resistance to deformation under tensile or compressive forces. Among the studied compounds, has the highest Young’s modulus (29.538 GPa), making it the most robust against deformation. This high stiffness is advantageous for applications requiring structural integrity, where materials must resist breaking under stress. The shear modulus (G), which measures a material’s resistance to shear deformation, is highest for , with a value of 10.874 GPa. This suggests that is relatively more resistant to shear forces compared to the other compounds, making it suitable for applications involving shaping or molding, such as in flexible electronics. Poisson’s ratio (), which provides insights into the bonding nature and plasticity of a material, varies slightly among the compounds, with values ranging from 0.358 to 0.402. These values indicate the presence of central forces within the crystal lattice99. The ductility and brittleness of the perovskites can be further assessed through their Pugh’s ratio. As shown in Table 5, the Pugh’s ratio for all the materials are above the critical value (1.75), confirming the ductile nature of these perovskites which is also confirmed by Cauchy pressure.This ductility is crucial for their application in devices that require materials to sustain mechanical deformation without fracturing, such as flexible electronics and photovoltaic cells97,100,101. We have also investigated the anisotropic nature of the (X = Cl, Br, I) perovskites to evaluate the directional dependence of various elastic parameters. This analysis is presented in Table 6 and further illustrated in Fig. 5 using the ‘ELATE: elastic tensor analysis’ tool. Table 6The maximum and minimum values, as well as anisotropy ratio (), of various elastic moduli. Perovskite Bulk Modulus B (GPa)Young’s Modulus E (GPa)Shear Modulus G (GPa)Poisson Ratio Linear compressibilityMinMax MinMax MinMax MinMax MinMax 38.71950.0871.29422.75834.7721.5287.12613.4631.8890.2250.5972.6546.6558.6091.294 27.75364.7262.33228.62233.9711.18710.23611.3491.1090.2970.4301.4465.15012.0112.332 25.97636.3251.3988.77634.9543.9832.41414.4445.9830.0810.82210.1059.17712.8321.398 The maximum and minimum values, as well as anisotropy ratio (), of various elastic moduli. Directional dependence of various elastic parameters such as Young’s modulus, linear compressiblity, Poisson’s ratio and shear modulus for: (a)–(d) , (e)–(h) , (i)-(l) . For , the Young’s modulus reveals a lower anisotropy with an of 1.187, suggesting a more isotropic behavior compared to the other perovskites. This is shown in Fig. 5(a). The linear compressibility of has a relatively high anisotropy ratio of 2.332, indicating considerable directional dependence, as depicted in Fig. 5(b). The Poisson’s ratio for also shows anisotropy with an of 1.446, as shown in Fig. 5(c), and the shear modulus exhibits an of 1.109, reflecting strong directional dependence, illustrated in Fig. 5(d). In the case of , the Young’s modulus (E) demonstrates significant anisotropy with an anisotropy ratio () of 1.528, indicating a pronounced directional dependence along different crystallographic axes. This behavior is visually represented in Fig. 5(e). The linear compressibility also shows notable anisotropy with an of 1.294, as depicted in Fig. 5(f). Additionally, the Poisson’s ratio () for exhibits a high anisotropy ratio of 2.654, reflecting substantial variation in compressibility across directions, as shown in Fig. 5(g). The shear modulus (G) of is characterized by significant anisotropy, with an of 1.889, indicating directional dependence in its shear response, as illustrated in Fig. 5(h). For , the Young’s modulus demonstrates the highest anisotropy among the studied perovskites, with an of 3.983, indicating substantial variation in mechanical stiffness along different directions, as shown in Fig. 5(i). The linear compressibility has an anisotropy ratio of 1.398, suggesting moderate directional dependence, as depicted in Fig. 5(j). The Poisson’s ratio for is notably anisotropic with an of 10.105, implying significant differences in the material’s response to uniaxial stress, as illustrated in Fig. 5(k). Additionally, the shear modulus shows an of 5.983, indicating pronounced directional dependence, as demonstrated in Fig. 5(l). The Kleinman parameter () is a crucial indicator of the internal strain in materials, with values ranging between 0 and 1 for solids. A value of corresponds to minimizing bond bending, whereas is associated with minimizing bond stretching. The calculated Kleinman parameters for , , and are 0.89, 0.86, and 0.74, respectively, as shown in Table 5. These values suggest that bond bending and bond stretching are nearly balanced for and , while bond bending is slightly more dominant than bond stretching in . Additionally, Table 5 highlights that has the highest average wave velocity of 1717.45 m/s, followed by with 1606.15 m/s, and with 1194.80 m/s. Thermoelectric properties Understanding the thermoelectric properties of materials is essential for converting waste heat into useful energy. About 20–50% of industrial energy input is dissipated as waste heat every year102. Thermoelectric materials can directly convert this heat energy into electrical power, thereby enhancing overall energy efficiency. Therefore, in this section we have studied the thermoelectric properties of (X = Cl, Br, I) in the range of temperature from 300 to 1450 K. The constant relaxation time approximation was used to calculate thermoelectric properties such as Seebeck coefficient (S), electronic thermal conductivity per relaxation time (), electrical conductivity per relaxation time () and magnetic susceptibility () utilizing BoltzTrap2 code58.. Due to the half-metallic nature of perovskites, these parameters give different results in different spin channels, so the net effect of these parameters can be calculated as,103 where and represent the corresponding thermoelectric parameters in the spin-up and spin-down channels, respectively. The Seebeck coefficient determines the magnitude of thermoelectric voltage generated when a temperature gradient is applied across a material. Figure 6(a) shows the variation of total Seebeck coefficient (S) with temperature for all the three studied compounds, it is evident that S increases with an increase in temperature. Furthermore, as shown in Fig. 7(a–c), the Seebeck coefficient for all the compounds gradually decreases for the spin-up channel and increases for the spin-down channel, demonstrating semiconducting nature in the spin-up channel and metallic nature in the spin-down channel68. Additionally, the positive value of throughout the temperature range indicates p-type behavior of the half-metals in both channels103. Notably, exhibits the highest Seebeck coefficient of 51.9 V/K among the compounds at room temperature, followed by with 37.5 V/K and with 24.9 V/K, as outlined in Table 7. Variation in the thermoelectric properties of p-type (X = Cl, Br, I) with temperature. (a) Seebeck coefficient (S). (b) Thermoelectric conductivity per relaxation time ()7. (c) Electrical conductivity() per relaxation time. (d) Magnetic susceptibility (). (e) Power Factor (PF). (f) Figure of Merit (ZT). Variation of Seebeck coefficient with temperature (for both the spin channels and their combined contribution) for (a) (b) and (c) . Table 7Thermoelectric properties of halide perovskites (X = Cl, Br, and I) at room temperature.PerovskiteS (V/K) (xW/mKs) (x)51.91.893.250.3524.90.941.730.8237.50.610.991.61 Thermoelectric properties of halide perovskites (X = Cl, Br, and I) at room temperature. The electronic thermal conductivity () per relaxation time () is another essential parameter for assessing a material’s thermoelectric efficiency, as it measures the material’s capability to conduct heat through electron transport. This conductivity arises from lattice vibrations occuring within the compounds during the movement of thermally excited electrons, including both electronic () and phononic () components. In the present study, thermal conductivity is calculated while neglecting the phononic component (), despite its contribution to the total thermal conductivity of the material. This is because BoltzTraP code does not incorporate this contribution. This exclusion has also been acknowledged in several related studies due to the same computational constraint104,105. The electronic thermal conductivity per relaxation time as a function of temperature is shown in Fig. 6(b). From the figure, we can infer that the electronic thermal conductivity follows an increasing trend because with increase in temperature the molecular vibrations increase. Furthermore, as shown in Table 5, possesses the highest Debye temperature among the titled compounds. According to the Bloch-Grüneisen formula, this implies that should exhibit the highest thermal conductivity106. The Bloch-Grüneisen formula is expressed as follows: This equation indicates that materials with higher Debye temperatures will have lower resistivity at a given temperature, resulting in higher thermal conductivity. This theoretical prediction aligns well with our findings, as depicted in Fig. 6(b). According to Fig. 8(a–c), the electrical conductivity of all compounds increases with temperature for the spin-up channel. This is because, as temperature rises, the charge carriers gain sufficient energy to jump from the valence band to the conduction band, increasing the number of charge carriers and resulting in higher conductivity. This behavior confirms the semiconducting nature of the materials in this channel. In contrast, the electrical conductivity decreases for the spin-down channel as temperature increases. This is due to the increased vibration of metal ions at higher temperatures, which increases resistance and, therefore, decreases conductivity, thereby confirming the metallic nature of the materials in this channel. The electronic conductivity as a function of temperature for all the three compounds is also shown in Fig. 6(c). As shown in Table 7, exhibits significantly greater conductivity (3.25 than both (1.73 and (0.99 at room temperatures. This is due to the fact that the larger atomic sizes of I and Br relative to Cl lead to increased Coulomb repulsion, which hinders the movement of charge carriers and thereby reduces conductivity. The ratio is found to be on the order of , indicating that these compounds are promising thermoelectric materials according to the Wiedemann-Franz law. Variation of electrical conductivity with temperature (for both the spin channels and their combined contribution) for for (a) (b) and (c) . Remarkably, , which possesses the highest Seebeck coefficient (51.9 V/K) and electrical conductivity (3.25 ), exhibits excellent potential for thermoelectric applications107. Additionally, its relatively low electronic thermal conductivity per relaxation time () compared to the other compounds enhances the figure of merit (ZT), which is crucial for thermoelectric device efficiency. The higher ZT values of and at elevated temperatures further indicate that these materials are well-suited for high-temperature thermoelectric devices, making them viable candidates for waste heat recovery systems, medical devices, microelectronics, automotive systems, remote power generation, solar-thermal systems, and temperature or gas sensing108,109. The variation of magnetic susceptibility () with temperature is depicted in Fig. 6(d). Magnetic susceptibility is a measure of how much a material will become magnetized when exposed to an external magnetic field. As shown in Fig. 6(d), decreases with increasing temperature, which is a characteristic of ferromagnetic materials. This phenomenon can be explained by the thermal agitation of magnetic moments within the material. At lower temperatures, the magnetic domains are more aligned, resulting in higher magnetization and thus a higher magnetic susceptibility. However, as temperature increases, thermal energy disrupts this alignment by increasing the random motion of magnetic moments. This leads to a reduction in the net magnetization of the material and, consequently, a decrease in magnetic susceptibility. This behavior aligns with the Curie-Weiss law, which describes the temperature dependence of magnetic susceptibility in ferromagnetic materials. We have also calculated the power factor (PF) which quantifies a material’s capability to convert temperature differentials into electrical power, mathematically expressed as PF. As shown in Fig. 6(e), the power factor shows nonlinear variation, increasing initially with temperature up to 1100 K, and then decreasing as a result of reduced electrical conductivity at higher temperatures. At room temperature, the power factor is approximately W/ms for , W/ms for and W/ms for indicating good thermoelectric efficiency. For a more accurate evaluation of the thermal to electrical energy conversion, the figure of merit (ZT) was also determined and plotted in Fig. 6(f). It quantifies the maximum conversion efficiency that can be achieved by a thermoelectric device. The maximum values of ZT are shown in Table 7 which occur at around 1150 K for (0.35), 850 K for (0.82) and 1050 K for (1.61). This observation aligns with the results reported by Yadav et al., which show that possesses the highest ZT value, followed by and 110. Additionally, Fig. 9 presents a bar graph comparing ZT values at 300 K across the perovskite compounds, alongside their respective maximum ZT values. At 300 K, exhibits the highest magnitude of 0.125, followed by (0.062) and (0.031). However, it is important to note that the achieved ZT values can be further improved through the application of strain effects or optimization of hole doping strategies111. A bar graph comparing the values at 300 K across three different compounds: , , and , along with the maximum value observed in the temperature range from 300 K to 1450 K. In comparison to other halide perovskites, the investigated compounds stand out with a significantly higher figure of merit (ZT). For instance, (X = Cl, Br, I) and (A = Li, Rb, Cs) achieve maximum ZT values of 0.79 and 0.99, respectively110,112, while reaches a maximum ZT of 1.61. Additionally, other halide perovskites, such as (X = F, Cl, Br)113, 39, and (M = Pb, Sn)33, exhibit even lower ZT values than and . These findings suggest that the half-metallic (X = Cl, Br, I) perovskites are highly promising thermoelectric materials, making them suitable for various applications in spintronics and energy conversion technologies. Dynamic stability Figure 10 presents the phonon band spectra for the studied perovskites, computed using density functional perturbation theory (DFPT) as implemented in VASP. In a crystal lattice with n atoms per unit cell, there are theoretically 3n phonon branches, with 3 being acoustic and the remaining being optical. For the perovskites under investigation, which contain five atoms per unit cell, this results in a total of 15 phonon modes, distributed as three acoustic and 12 optical phonons. The phonon dispersion curves illustrated in Figure 10(a) for and (b) for exhibit positive frequency values, confirming their dynamical stability. In contrast, the dispersion curve for , presented in Fig. 10(c), reveals a slight negative phonon mode, indicating the dynamical instability of this material. This instability is also observed in other iodine-based perovskites, such as and 68,114. Phonon dispersion curves for hexagonal perovskites: (a) , (b) , and (c) . Conclusion In this study, we conducted a thorough investigation of the electronic, magnetic, elastic, optical, and thermoelectric properties of halide perovskites (X = Cl, Br, I), employing density functional theory (DFT). The analysis of spin-polarized band structures and density of states (DOS) reveal that these compounds exhibit half-metallicity, characterized by a significant band gap in the spin-up channels and 100% spin polarization at the Fermi level. The spin magnetic moment value for (X = Cl, Br, I) was found to be 2 B for all compounds, primarily due to the hybridization of Co 3d orbitals and p orbitals of the Cl, Br, and I halogens. Among all the perovskites, , in particular, demonstrates outstanding performance across several properties, making it the prime candidate for practical applications. Its large half-metallic gap is essential for the effective performance of spintronic devices, as it facilitates efficient spin polarization and transport. The material also exhibits excellent elastic and mechanical properties, showing ductility, anisotropy, and compliance with Born stability criteria, which ensures its mechanical integrity in various applications. Moreover, the high optical conductivity of , particularly in the UV region, positions it as a strong candidate for optoelectronic devices. Additionally, its notable thermoelectric parameters, including a high Seebeck coefficient, electrical and thermal conductivity, and power factor, highlight its potential in energy conversion and thermoelectric applications. Author contributions A.R: Conceptualization, methodology, formal analysis, investigation, data curation, software, writing the main manuscript, A.K: Conceptualization, methodology, formal analysis, software, writing the main manuscript, supervision, project administration, T.M: Formal analysis, supervision, project administration. All authors reviewed the manuscript. Data availability The datasets generated and analyzed during the current study are available in the Google Drive repository: https://drive.google.com/drive/folders/1PDVcTKyJgvaJRYJ0aU2gAnZhytScV1EA?usp=drive_link. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets generated and analyzed during the current study are available in the Google Drive repository: https://drive.google.com/drive/folders/1PDVcTKyJgvaJRYJ0aU2gAnZhytScV1EA?usp=drive_link.