Band Structure And Dos

band structure is the relationship between the energy of electrons and their crystal momentum within a periodic solid. It is represented as a set of curves E(k) plotted along high‑symmetry directions of the reciprocal lattice. In both Quantum Espresso and VASP, the band structure is obtained after a self‑consistent field (SCF) calculation that determines the ground‑state charge density, followed by a non‑self‑consistent calculation (NSCF) that evaluates the eigenvalues at a chosen set of k‑points. The resulting eigenvalues are then post‑processed to generate the familiar “energy‑versus‑k” plot.

The density of states (DOS) quantifies how many electronic states are available at each energy level. Mathematically, it is defined as D(E)=∑_n∫_BZ δ(E−E_n(k)) dk, where the sum runs over all bands n and the integral extends over the entire Brillouin zone. The DOS provides a complementary view to the band structure: While the band diagram shows the dispersion of individual states, the DOS compresses that information into a one‑dimensional histogram that is directly comparable to experimental techniques such as photo‑electron spectroscopy.

Both the band structure and the DOS rely on the definition of the Brillouin zone, the primitive cell of the reciprocal lattice. High‑symmetry points such as Γ, X, L, K, and M are identified by the space‑group symmetry of the crystal. In practice, users generate a list of k‑points that sample the Brillouin zone using either uniform Monkhorst‑Pack grids or specialized paths that trace along the symmetry lines. Quantum Espresso provides the kpoints.X utility to create these grids, while VASP reads a KPOINTS file where the user explicitly lists the desired points and their weights.

The reciprocal lattice vectors are defined as a* = 2π(b×c)/V, b* = 2π(c×a)/V, and c* = 2π(a×b)/V, where a, b, c are the real‑space lattice vectors and V is the unit‑cell volume. These vectors form the basis for constructing the Brillouin zone and for expressing the crystal momentum k. Because the electronic wavefunctions are Bloch functions ψ_{n,k}(r)=e^{i k·r} u_{n,k}(r), the periodic part u_{n,k}(r) inherits the periodicity of the reciprocal lattice, allowing the problem to be solved in a finite domain.

A central concept in band‑structure analysis is the band gap. It is the energy difference between the top of the valence band and the bottom of the conduction band. If the minimum of the conduction band occurs at the same k‑point as the valence‑band maximum, the material exhibits a direct gap; otherwise it has an indirect gap. Silicon, for example, possesses an indirect gap of about 1.1 EV, whereas gallium arsenide shows a direct gap of roughly 1.4 EV. The nature of the gap strongly influences optical absorption, carrier recombination, and device performance.

The effective mass of carriers is derived from the curvature of the E(k) relationship near the band edges. In a parabolic approximation, m* = ℏ²/(∂²E/∂k²). Light effective masses lead to high carrier mobilities, which is why materials like graphene, with a linear dispersion near the Dirac point, are often described as having a “massless” charge carrier. Quantum Espresso can compute effective masses directly from the band curvature using the effective_mass.X post‑processing tool, while VASP users typically export the eigenvalues and fit the curvature manually or with external scripts.

The Fermi level (E_F) marks the chemical potential at zero temperature. In metals, E_F cuts through partially filled bands, leading to a non‑zero DOS at the Fermi energy. In insulators and semiconductors, E_F lies within the band gap, and the DOS at E_F is essentially zero. In practical calculations, the Fermi level is determined from the occupancy of the electronic states according to the chosen smearing method. Common smearing schemes include Methfessel‑Paxton, Gaussian, and the Fermi‑Dirac distribution. The smearing width (σ) must be selected carefully: Too large a σ artificially broadens the DOS and can obscure fine features, while too small a σ may cause convergence difficulties in metallic systems.

The pseudopotential approximation replaces the all‑electron problem with an effective potential that reproduces the scattering properties of the valence electrons while eliminating the need to treat core electrons explicitly. Quantum Espresso typically uses norm‑conserving or ultrasoft pseudopotentials, whereas VASP employs the projector‑augmented‑wave (PAW) method. Both approaches dramatically reduce the plane‑wave basis size required for convergence. The choice of pseudopotential influences the accuracy of the band energies, particularly for elements with semi‑core states or strong relativistic effects. Users must verify that the pseudopotential includes the appropriate valence configuration (e.G., 3Dⁱ⁰4s² for copper) and that it is compatible with the exchange‑correlation functional employed.

The exchange‑correlation functional approximates the many‑body effects of electron‑electron interaction. The most common choices are the local density approximation (LDA) and the generalized gradient approximation (GGA) such as PBE. While LDA often underestimates lattice constants and band gaps, GGA improves structural properties but still suffers from the well‑known “band‑gap problem.” Hybrid functionals (e.G., HSE06) mix a fraction of exact Hartree–Fock exchange with GGA exchange, yielding more accurate band gaps at a higher computational cost. In VASP, the functional is selected via the INCAR tag (e.G., LHFCALC = .TRUE. for hybrid calculations). In Quantum Espresso, the input_dft variable controls the functional.

The plane‑wave basis set expands the periodic part of the Bloch wavefunction in a series of plane waves, ψ_{n,k}(r)=∑_G c_{n,k}(G) e^{i G·r}. The completeness of the basis is governed by the kinetic‑energy cutoff (E_cut). A higher E_cut includes more plane waves, improving accuracy but increasing computational expense. Convergence tests involve increasing E_cut until the total energy, forces, and band eigenvalues change by less than a chosen tolerance (e.G., 1 MeV/atom). Typical cutoffs range from 30 Ry for soft pseudopotentials to over 100 Ry for hard ones. In VASP, the analogous parameters are ENCUT and ENCUTGW for GW calculations.

The k‑point sampling determines how finely the Brillouin zone is discretized. For metallic systems, dense Monkhorst‑Pack grids (e.G., 16×16×16) Are required to capture the rapid variation of the DOS near the Fermi level. For insulators, coarser grids (e.G., 8×8×8) May suffice. The convergence of the total energy with respect to k‑point density is a standard check. In Quantum Espresso, the K_POINTS card defines the grid, while VASP reads the KPOINTS file where the user can specify automatic generation (e.G., “Automatic”, 8 8 8, 0 0 0) or explicit listings of points.

The self‑consistent field (SCF) cycle iteratively solves the Kohn‑Sham equations until the change in total energy and charge density between successive iterations falls below predefined thresholds. Convergence criteria are set by conv_thr in Quantum Espresso and by EDIFF in VASP. In metallic calculations, the smearing parameter must be compatible with the SCF convergence scheme; otherwise the charge density may oscillate and the calculation can stall. Mixing schemes (e.G., Broyden, Pulay) are controlled by mixing_beta (Quantum Espresso) or AMIX (VASP). Proper mixing improves stability, especially for systems with shallow energy minima or complex magnetic ordering.

The non‑self‑consistent field (NSCF) run uses the converged charge density from the SCF step but evaluates the eigenvalues on a different, often denser, k‑point mesh. This step is essential for accurate DOS calculations because a finer k‑mesh reduces artificial discretization of the energy levels. In Quantum Espresso, the bands.X utility can extract the band eigenvalues, while in VASP the NWRITE tag set to 11 triggers a detailed eigenvalue output. The resulting eigenvalues are then processed by the dos.X (Quantum Espresso) or the DOSCAR file (VASP) to produce the DOS histogram.

The projected density of states (PDOS) resolves the total DOS onto atomic orbitals, spin channels, or angular momentum components. PDOS is invaluable for interpreting the chemical bonding and for identifying the contributions of specific atoms or orbitals to the electronic structure. In Quantum Espresso, the projwfc.X program projects the wavefunctions onto spherical harmonics centered on each atom, generating files such as atom1.Pdos. In VASP, the LORBIT tag (set to 11 or 12) enables the generation of the PROCAR file, from which PDOS can be extracted. For example, the PDOS of a transition‑metal oxide may reveal that the valence band is dominated by O‑2p states, while the conduction band has significant transition‑metal d‑character.

The spin‑polarized calculation treats up‑ and down‑spin electrons independently, allowing the description of magnetic materials. In such calculations, the DOS is split into spin‑up and spin‑down components, often visualized as two separate curves. The net magnetic moment per unit cell is obtained by integrating the difference between the spin‑up and spin‑down DOS up to the Fermi level. Quantum Espresso activates spin polarization by setting nspin = 2, while VASP uses ISPIN = 2. For antiferromagnetic ordering, a larger magnetic unit cell must be constructed, and the initial magnetic moments are specified via the MAGMOM tag.

The band‑unfolding technique is used when supercells are employed to model defects, alloys, or surface reconstructions. The enlarged Brillouin zone folds the original band structure, making interpretation difficult. Unfolding maps the supercell bands back onto the primitive‑cell Brillouin zone, yielding an effective band structure that can be compared directly with experiment. In Quantum Espresso, the unfold.X utility (available in the EPW package) performs this operation, while VASP users often rely on external tools such as the BandUP code, which reads the EIGENVAL file and executes the unfolding algorithm.

The optical properties can be derived from the band structure and DOS by evaluating interband transition matrix elements. The imaginary part of the dielectric function ε₂(ω) is obtained from the joint density of states weighted by the transition dipole matrix elements, while the real part ε₁(ω) follows from a Kramers‑Kronig transformation. Both Quantum Espresso and VASP provide modules for such calculations: The epsilon.X utility in Quantum Espresso and the LOPTICS = .TRUE. flag in VASP. These calculations enable the prediction of absorption spectra, reflectivity, and refractive indices, which are crucial for photovoltaic and optoelectronic device design.

The total energy of a system, obtained from the SCF calculation, serves as a benchmark for structural optimization, phase stability, and reaction energetics. By comparing total energies of different crystal structures, one can predict the most stable polymorph. In practice, one performs a series of volume‑conserving relaxations (e.G., Using the vc-relax mode in Quantum Espresso or the ISIF = 3 setting in VASP) to locate the equilibrium lattice parameters. The equilibrium volume minimizes the total energy, and the curvature of the energy‑versus‑volume curve yields the bulk modulus.

The phonon dispersion is closely linked to the electronic band structure because electron‑phonon coupling can renormalize the electronic energies. Phonon calculations are typically performed using density‑functional perturbation theory (DFPT) in Quantum Espresso (via the ph.X program) or the finite‑difference approach in VASP (using the IBRION = 7 setting). The resulting phonon frequencies ω(q) can be combined with the electronic structure to assess superconducting transition temperatures via the Eliashberg function α²F(ω).

The GW approximation provides a many‑body correction to the Kohn‑Sham eigenvalues, improving the prediction of quasiparticle band gaps. In the GW method, the self‑energy Σ = iGW replaces the exchange‑correlation potential, and the Green’s function G and screened Coulomb interaction W are computed iteratively. Both Quantum Espresso (through the Yambo or BerkeleyGW interfaces) and VASP (with the ALGO = GW0 or GW settings) support GW calculations. The computational cost is high, requiring a large number of empty bands and a fine k‑mesh, but the resulting band gaps for semiconductors like Si and GaAs match experimental values within a few percent.

The Bethe‑Salpeter equation (BSE) extends GW by explicitly treating electron–hole interactions, enabling accurate optical spectra that include excitonic effects. In practice, one first performs a GW calculation to obtain the quasiparticle energies, then solves the BSE on top of those energies. VASP implements BSE via the LOPTICS = .TRUE. and NBANDSO/NBANDSV tags, while Quantum Espresso users can employ the Yambo code. The resulting absorption spectra show bound exciton peaks that are absent in independent‑particle calculations.

The magnetic ordering can be probed by comparing the total energies of different spin configurations. For a ferromagnetic (FM) alignment, all magnetic moments point in the same direction, while antiferromagnetic (AFM) arrangements have alternating spin orientations. By constructing supercells that accommodate the AFM pattern and performing SCF calculations for each magnetic state, the energy difference ΔE = E_AFM − E_FM yields the exchange coupling constant J through a Heisenberg model. This approach is widely used to study transition‑metal oxides, rare‑earth compounds, and low‑dimensional magnetic materials.

The charge density obtained from the SCF run can be visualized to assess bonding characteristics. Isosurfaces of the charge density highlight regions of high electron localization, while difference density plots (Δρ = ρ_total − Σ ρ_atomic) reveal charge transfer between atoms. Both Quantum Espresso (via the pp.X utility) and VASP (through the CHGCAR file) generate charge‑density grids that can be imported into visualization tools such as VESTA or Xcrysden. Analyzing the charge density helps identify covalent, ionic, or metallic bonding regimes.

The partial charge analysis, often performed with Bader or Mulliken schemes, assigns electron populations to individual atoms. Bader analysis partitions space according to zero‑flux surfaces in the charge density gradient, providing a robust measure of atomic charges. In VASP, the CHGCAR file is processed by the external Bader code, while Quantum Espresso users can employ the bader.X utility. These atomic charges are useful for interpreting polarization, surface reactivity, and defect formation energies.

The defect formation energy is calculated by comparing the total energy of a supercell containing the defect with that of the pristine supercell, adding or removing atoms and electrons as required. The formula E_f = E_defect − E_bulk + Σ n_i μ_i + q(E_F + E_VBM) + ΔV accounts for chemical potentials μ_i, charge state q, and potential alignment ΔV. Accurate defect calculations demand large supercells (e.G., 3×3×3 Repetitions of the primitive cell) to minimize spurious interactions between periodic images. Convergence tests with respect to supercell size, k‑point density, and plane‑wave cutoff are essential.

The surface slab model is a common approach to study surfaces, adsorption, and catalysis. A slab consists of a finite number of atomic layers separated by a vacuum region large enough (typically >15 Å) to prevent interaction between periodic images. The slab is relaxed with the bottom layers constrained to mimic bulk behavior, while the top layers and adsorbates are allowed to move. In Quantum Espresso, the relax or vc-relax calculations can be employed, and the occupations tag is set to smearing for metallic surfaces. In VASP, the ISIF = 2 flag keeps the cell dimensions fixed while relaxing atomic positions.

The adsorption energy quantifies the strength of binding between an adsorbate and a surface. It is defined as E_ads = E_slab+adsorbate − E_slab − E_adsorbate(gas). A negative E_ads indicates exothermic adsorption. Accurate adsorption energies require careful convergence with respect to slab thickness, vacuum size, k‑point sampling, and the choice of exchange‑correlation functional (often hybrid or dispersion‑corrected functionals are needed to capture van der Waals interactions). Zero‑point energy corrections and entropy contributions may also be added to compare with experimental desorption temperatures.

The van der Waals corrections (e.G., DFT‑D2, DFT‑D3, Tkatchenko‑Scheffler) augment conventional functionals with an empirical dispersion term to account for long‑range correlation effects. In VASP, the IVDW tag selects the specific correction scheme, while Quantum Espresso users enable dispersion via the vdw_corr variable (e.G., vdw_corr = 'grimme-d2'). These corrections are essential for layered materials like MoS₂, molecular adsorption on metal surfaces, and organic crystals, where standard GGA functionals underestimate binding.

The k‑point path generator assists in constructing the high‑symmetry line set for band‑structure plots. Tools such as seekpath, the bandx.X script, or the VASP KPOINTS “Line‑mode” automatically identify the symmetry points based on the space group and output the coordinates and distances required for a smooth plot. The path must be continuous, and the distance between successive points should be fine enough (e.G., 0.01 Å⁻¹) to resolve subtle features like band crossings or avoided degeneracies.

The spin‑orbit coupling (SOC) introduces relativistic interactions that split degenerate bands, especially in heavy elements. SOC is included by adding a non‑collinear term to the Hamiltonian, which mixes spin‑up and spin‑down components. In Quantum Espresso, SOC is activated by setting lspinorb = .True. and noncolin = .True., while VASP requires LSORBIT = .TRUE. and SAXIS to define the spin quantization axis. SOC can open gaps at Dirac points (e.G., In topological insulators) and alter the band topology, making it indispensable for studying materials like Bi₂Se₃, WTe₂, and perovskite oxides.

The topological invariants such as the Z₂ index or Chern number are derived from the band structure and its symmetry properties. Calculating these invariants often involves tracking the evolution of Wannier charge centers or evaluating the Berry curvature across the Brillouin zone. In practice, one constructs maximally localized Wannier functions using the wannier90 code interfaced with both Quantum Espresso and VASP, then computes the Berry phase with the postw90.X utility or the VASP WANNIER90 interface. The resulting topological classification guides the search for quantum spin Hall materials and Weyl semimetals.

The band alignment between two materials, such as a semiconductor and an insulator, is crucial for heterojunction design. The lineup is typically referenced to the vacuum level, requiring the calculation of the average electrostatic potential in the bulk region of each material and the surface dipole for a slab model. By aligning the potentials, one obtains the conduction‑band offset (CBO) and valence‑band offset (VBO). In both Quantum Espresso and VASP, the planar‑averaged electrostatic potential can be extracted (via pp.X or the LOCPOT file) and post‑processed to determine the offsets. Accurate band alignment informs the design of photovoltaic cells, tunnel diodes, and field‑effect transistors.

The magnetic anisotropy energy (MAE) quantifies the dependence of the total energy on the direction of magnetization. MAE is computed by performing two SCF calculations with the magnetization constrained along different crystallographic axes (e.G., [001] And [100]). The energy difference ΔE = E_[100] − E_[001] yields the anisotropy. Inclusion of SOC is mandatory, as MAE arises from relativistic effects. In VASP, the SAXIS tag defines the magnetization direction, while Quantum Espresso uses the starting_magnetization and noncolin settings together with SOC. MAE values are essential for spintronic applications and permanent magnet design.

The carrier mobility can be estimated from the effective mass and scattering rates using the Boltzmann transport equation within the relaxation‑time approximation. The BoltzTraP code, compatible with both Quantum Espresso and VASP outputs, takes the eigenvalues on a dense k‑mesh and computes transport coefficients such as electrical conductivity, Seebeck coefficient, and Hall coefficient as functions of temperature and chemical potential. These predictions guide the selection of thermoelectric materials and high‑mobility semiconductors.

The thermodynamic stability of a compound is assessed using the convex hull construction in composition space. By calculating the formation energies of all relevant phases (including competing binaries and ternaries) and plotting them against composition, one identifies whether a given structure lies on the convex hull (stable) or above it (metastable). This analysis often employs databases such as the Materials Project, which store pre‑computed total energies from VASP. Nevertheless, for novel compounds, the user must generate the necessary total‑energy calculations and feed them into a hull‑building script (e.G., pymatgen's PhaseDiagram module).

The phonon‑limited thermal conductivity is derived from the phonon dispersion and lifetimes obtained via anharmonic force constants. In the ShengBTE or Phono3py packages, the second‑order (harmonic) and third‑order (anharmonic) force constants are extracted from VASP or Quantum Espresso calculations (using finite‑displacement supercells). Solving the Boltzmann transport equation for phonons yields the lattice thermal conductivity κ_L, a key metric for thermoelectric materials and heat‑management applications.

The high‑throughput workflow automates the generation, execution, and analysis of many DFT calculations. Frameworks such as AiiDA, FireWorks, and Atomate orchestrate the sequence of SCF, NSCF, DOS, and band‑structure steps, storing inputs, outputs, and provenance in databases. In a typical high‑throughput study, one defines a set of crystal structures, assigns appropriate pseudopotentials, sets convergence criteria, and launches parallel jobs on a computing cluster. The workflow then parses the resulting EIGENVAL, DOSCAR, and OUTCAR files, extracts band gaps, effective masses, and DOS features, and writes them to a CSV file for downstream data mining.

The machine‑learning models trained on DFT‑derived descriptors (e.G., Band gap, DOS at the Fermi level, orbital character) can predict properties for unexplored compounds. Features such as the average electronegativity, bond length distribution, and partial DOS moments are fed into regression algorithms (e.G., Random forests, kernel ridge regression) to accelerate materials discovery. The quality of the training data depends on the consistency of the DFT calculations; therefore, standardized settings for cutoffs, k‑points, and functionals are essential to avoid systematic bias.

The convergence testing protocol is a systematic approach to ensure that calculated quantities (total energy, forces, band gap, DOS) are insensitive to computational parameters. A typical protocol involves: (1) Raising the plane‑wave cutoff in steps (e.G., 30 Ry, 40 Ry, 50 Ry) while keeping the k‑mesh fixed; (2) increasing the k‑mesh density (e.G., 4×4×4, 6×6×6, 8×8×8) At the highest cutoff; (3) checking the effect of smearing width on metallic systems; and (4) verifying that the total energy changes by less than a chosen tolerance (e.G., 1 MeV/atom). For DOS calculations, the smearing parameter used to broaden the delta functions should be reduced until the DOS features converge, typically to within 0.01 States/eV.

The vacuum level alignment is crucial when evaluating work functions or electron affinities. By constructing a slab with a sufficiently thick vacuum region, one extracts the planar‑averaged electrostatic potential and identifies the flat region far from the surface. The difference between this vacuum potential and the Fermi level yields the work function Φ = V_vacuum − E_F. In VASP, the LOCPOT file contains the necessary potential, and post‑processing scripts compute the average in the vacuum region. Accurate work functions enable the design of electron emitters, photocathodes, and contact materials.

The charge‑density difference for an adsorbate–surface system highlights regions of electron accumulation and depletion upon adsorption. It is obtained by subtracting the sum of the isolated substrate and adsorbate charge densities from the combined system: Δρ(r) = ρ_total(r) − ρ_substrate(r) − ρ_adsorbate(r). Visualizing Δρ with isosurfaces reveals the formation of chemical bonds, charge transfer direction, and dipole formation. Both Quantum Espresso and VASP provide the charge density files (charge-density.Dat or CHGCAR) needed for this analysis.

The partial wave analysis in the context of PAW potentials decomposes the wavefunction into angular‑momentum channels (s, p, d, f). This decomposition is used to generate the projected DOS and to calculate the magnetic moments associated with each orbital. In VASP, the LORBIT flag controls the level of detail; setting LORBIT = 11 yields orbital‑resolved PDOS, while LORBIT = 10 provides only atomic‑resolved PDOS. Quantum Espresso’s projwfc.X offers analogous functionality, allowing users to select specific angular‑momentum channels for projection.

The band‑gap underestimation is a well‑known limitation of standard DFT. LDA and GGA functionals typically predict gaps that are 30–50 % smaller than experimental values. To mitigate this, one may employ hybrid functionals, GW quasiparticle corrections, or DFT+U approaches for correlated d‑ or f‑electron systems. The DFT+U method adds an on‑site Coulomb term U to the localized orbitals, correcting the self‑interaction error. In VASP, the LDAU = .TRUE. flag activates the method, and the LDAUU and LDAUL tags specify the U values and orbital types. In Quantum Espresso, the Hubbard_U variable serves the same purpose.

The supercell approach for point defects, surfaces, and interfaces introduces artificial periodicity that can affect the calculated properties. Finite‑size corrections, such as the Makov‑Payne correction for charged defects, compensate for the spurious electrostatic interactions. The correction term scales as q²α/(2εL), where α is the Madelung constant, ε the dielectric constant, and L the supercell dimension. Implementations exist in post‑processing tools (e.G., sxdefectalign) that read the VASP or Quantum Espresso output and apply the correction automatically.

The electron‑phonon coupling constant λ is obtained by integrating the Eliashberg spectral function α²F(ω) over phonon frequencies. In the EPW code interfaced with Quantum Espresso, one computes the electron‑phonon matrix elements on coarse k‑ and q‑grids, interpolates them onto dense meshes using Wannier functions, and evaluates λ. VASP users can extract the electron‑phonon coupling via the IBRION = 8 setting combined with the LEPSILON = .TRUE. flag, though the workflow is less automated. Λ influences superconducting transition temperatures through the Allen‑Dynes formula.

The surface energy γ is defined as γ = (E_slab − N E_bulk)/2A, where E_slab is the total energy of the slab, N the number of bulk units, E_bulk the bulk energy per unit, and A the surface area. The factor of two accounts for the two surfaces of the slab. Accurate surface energies require convergence with respect to slab thickness, vacuum size, and k‑point sampling. Surface energies guide the prediction of crystal habit, faceting, and equilibrium shapes via the Wulff construction.

The band‑structure visualization typically involves plotting the eigenvalues along the high‑symmetry path with energy referenced to the Fermi level. Tools such as gnuplot, matplotlib, or dedicated software like VasP and XCrySDen read the EIGENVAL or band.Dat files and generate smooth curves. For spin‑polarized calculations, separate lines for spin‑up and spin‑down are plotted, often using different colors or line styles. Adding the DOS as a subplot beneath the band diagram provides a comprehensive picture of the electronic structure.

The partial DOS integration yields the number of electrons contributed by a specific orbital within a chosen energy window. By integrating the PDOS from the bottom of the valence band up to the Fermi level, one obtains the orbital occupation, which can be compared with nominal oxidation states. This analysis is particularly useful for mixed‑valence compounds, where the charge distribution deviates from simple integer values. In practice, scripts parse the PDOS files, perform numerical integration (e.G., Trapezoidal rule), and output the electron counts.

The magnetic moment extraction from VASP is performed by reading the OUTCAR file, where the total and atom‑resolved magnetic moments are listed after each SCF step. The final converged values are reported under the “magnetization (x)” section. Quantum Espresso records the magnetization in the output.Xml file, which can be parsed with the pwtools library. These magnetic moments are essential for characterizing ferromagnetic, antiferromagnetic, and ferrimagnetic orderings.

The charge‑state correction for defects uses the alignment of the electrostatic potential far from the defect. By comparing the planar‑averaged potential of the defective supercell with that of the pristine cell, one extracts the shift ΔV and applies it to the formation energy expression. This procedure removes the artificial band‑bending caused by periodic image interactions. In VASP, the VASP output includes the planar‑averaged potential (LOCPOT), which can be processed by the align.Py script from the PyCDT package.

The band‑structure interpolation using Wannier functions enables the generation of ultra‑dense k‑point meshes without performing explicit DFT calculations at each point. After constructing maximally localized Wannier functions with wannier90, one can interpolate the Hamiltonian to arbitrary k‑points, yielding smooth band curves and high‑resolution DOS. This technique is valuable for studying fine features such as Rashba splitting, Weyl points, and Dirac cones. Both Quantum Espresso and VASP provide interfaces to wannier90, facilitating seamless workflow integration.

The spin‑texture analysis maps the spin orientation of Bloch states across the Brillouin zone, revealing phenomena such as spin‑momentum locking in topological insulators. By computing the expectation values of the Pauli matrices for each eigenstate, one obtains the spin components S_x(k), S_y(k), and S_z(k). In VASP, the LSORBIT and ISPIN = 2 settings allow extraction of spinor wavefunctions, while Quantum Espresso’s projwfc.X can output spin‑projected PDOS. Visualizing spin textures helps design spintronic devices that exploit spin‑polarized currents.