Density Functional Theory: Hohenberg-Kohn Theorems Explained

The Local Density Approximation (LDA) approximates the exchange-correlation energy as a local functional of the electron density at each point in space. For exchange, LDA uses the exact expression for an homogeneous electron gas. For correlation, numerical calculations are performed and fitted to analytic forms. LDA works remarkably well for many systems but has systematic errors, particularly for systems with strong non-local effects. The Generalized Gradient Approximation (GGA) improves upon LDA by including the gradient of the electron density in addition to the local density, accounting for non-local effects. Popular GGA functionals include PBE (Perdew-Burke-Ernzerhof). Hybrid functionals mix a fraction of exact Hartree-Fock exchange with GGA-type correlation. Meta-GGA functionals go further by including the kinetic energy density. Each functional level offers trade-offs between accuracy and computational cost.

In Density Functional Theory (DFT), the exchange-correlation energy functional is unknown and must be approximated. The Local Density Approximation (LDA) assumes exchange-correlation depends only on local electron density, derived from homogeneous electron gas models, making it computationally efficient but systematically underestimating band gaps. The General Gradient Approximation (GGA) improves upon LDA by incorporating the gradient of electron density, providing better accuracy for non-metallic systems and semiconductors. Popular LDA functionals include VWN, PZ, and PW92, while popular GGA functionals include B88, BLYP, and PBE. Both approaches have systematic errors: LDA overestimates correlation energy by 100-200% and underestimates exchange energy by ~20%, while GGA reduces these errors but still underestimates band gaps. The choice between LDA and GGA depends on the specific application, with GGA generally preferred for band structure calculations despite higher computational cost.

The exchange-correlation functional determines DFT accuracy and is classified by sophistication: Local Density Approximation (LDA) treats density at single points; Generalized Gradient Approximation (GGA) considers density gradients and inhomogeneity; Meta-GGA incorporates density, gradient, and second derivatives; Hybrid Meta-GGA adds fixed Hartree-Fock exchange fractions; Random Phase Approximation (RPA) utilizes all occupied and unoccupied orbitals for maximum accuracy. Hybrid functionals combine LDA, Hartree-Fock, B88, LYP, and VWN parameters. Higher sophistication correlates with greater computational expense but improved accuracy. The choice of functional is critical as it determines calculation accuracy for any system.

Exchange-correlation functionals determine DFT accuracy. Local Density Approximation (LDA) uses the uniform electron gas model, with Slater exchange E_x^LDA = C∫ρ^(4/3)dr and correlation from Monte Carlo simulations. LDA overestimates correlation by up to factor of 2, leading to too-large bond strengths. Generalized Gradient Approximation (GGA) incorporates density gradients (∇ρ) to account for non-uniform electron distributions in molecules. Meta-GGA extends further by including second derivatives or kinetic energy density τ(r). Hybrid functionals combine Hartree-Fock exchange with DFT correlation, introduced by Becke in 1993 to achieve chemical accuracy (±1 kcal/mol).

Density Functional Theory (DFT) solves the Schrödinger equation by focusing on electron density rather than wave functions, transforming the many-body problem into a computationally tractable single-variable optimization. Developed by Nobel laureates Hohenberg, Kohn, and coworkers, DFT treats electrons as a fluid-like distribution, enabling practical calculations for molecules and solids. The method has revolutionized material science, chemistry, and physics through applications in biochemistry (enzyme reactions), surface science (gas-surface interactions), catalysis (activation barriers), semiconductors (band gaps), superconductors, batteries (lithium diffusion), and water research. Despite its power, DFT has limitations: it is a zero-temperature theory unsuitable for temperature-dependent phenomena, and standard calculations underestimate semiconductor band gaps. Success stories include elucidating superconductivity mechanisms in MgB2 and revealing interlayer states in superconductors.

Density Functional Theory (DFT) revolutionized computational chemistry by replacing wave function-based methods with electron density as the fundamental variable. Developed from the Hohenberg-Kohn theorem (1964) and LDA approximation (1970s), DFT became practical in the 1980s and earned Dr. Walter Kohn the Nobel Prize. The self-consistent field algorithm iteratively solves the Kohn-Sham equations until electron density converges. DFT enables calculation of electronic structures, band gaps, absorption spectra, and catalytic properties through pseudopotentials that focus on valence electrons. This framework now routinely predicts material properties across chemistry, physics, and materials science.

Practical DFT implementation requires careful computational choices: basis set selection (plane waves, Gaussians, LAPW), self-consistent field iterations, and Brillouin zone sampling with symmetry-adapted k-point grids. Total energy calculations demand extreme numerical precision due to cancellation of large positive and negative terms. Advanced methods extend DFT capabilities: band structure calculations reveal electronic dispersion and band gaps; density functional perturbation theory efficiently computes elastic, piezoelectric, and phonon properties; and systematic screening enables materials discovery. Researchers have used DFT to predict novel ferroelectric oxy-nitrides with desirable nonlinear optical properties, demonstrating predictive power for materials design. These applications highlight DFT's role in guiding experimental synthesis and enabling rational materials development, though fundamental limitations persist requiring complementary approaches like Quantum Monte Carlo for higher accuracy.

This section covers the practical implementation of Density Functional Theory. The total energy functional decomposes into E_known (analytical terms: kinetic energy, electron-nucleus, electron-electron, and nucleus-nucleus interactions) and E_XC (exchange-correlation functional containing difficult quantum effects). The Kohn-Sham equations describe single-particle orbitals, including kinetic energy, electron-nucleus attraction, and Hartree potential. The Hartree potential contains self-interaction errors requiring correction via the exchange-correlation potential V_XC. Solving these equations requires iteration: initial trial density → solve Kohn-Sham equations → calculate new density → repeat until convergence. The Local Density Approximation (LDA) uses the exact solution for uniform electron gas, setting exchange-correlation potential based on local density. The Generalized Gradient Approximation (GGA) improves upon LDA by incorporating local density gradients, with PW91 and PBE being widely used functionals. Different functionals yield different results for the same atomic configuration, making functional specification essential in DFT calculations.

Computational Materials Science applies quantum mechanics to predict material properties without empirical parameters. Density Functional Theory (DFT) calculations are first-principles methods that solve quantum mechanical equations to obtain structural, vibrational, and electronic properties. Three foundational concepts underpin DFT: wave functions describe quantum states, operators represent measurable quantities, and the ground state represents the lowest energy configuration. The ultimate goal is solving the many-body Schrödinger equation to determine material properties from fundamental principles.

Time-Dependent Density Functional Theory (TDDFT) extends standard DFT by allowing the external potential to become time-dependent, enabling study of excited states and dynamic electronic processes. The Gross theorem establishes a one-to-one correspondence between time-dependent external potential and electronic density, similar to static DFT but extended to time evolution. Unlike static DFT, observables in TDDFT depend on both density and initial state because time evolution requires boundary conditions. The variational principle applies: stationary points of the action functional provide exact solutions. The Kohn-Sham approach reduces the many-body problem to single-particle equations with an effective potential. Linear response theory treats small perturbations on top of static DFT solutions. Polarizability connects induced density to applied potential through Δρ = χ(ΔV_ext + ΔV_Hxc). The fundamental equation χ_real = χ_0 / (1 - v_Hxc χ_0) relates independent particle polarizability to real polarizability via the exchange-correlation kernel. The microscopic dielectric function differs from classical experiments because classical test charges don't obey Pauli exclusion. Non-diagonal polarizability matrix elements are crucial for inhomogeneous systems.

Time Dependent Density Functional Theory (TDDFT) extends density functional theory to time-dependent systems by establishing a one-to-one correspondence between time-dependent external potentials and time-dependent electron densities (Runge-Gross theorem), enabling practical calculations through the Kohn-Sham formalism where the exchange-correlation potential is approximated, typically using the adiabatic approximation; TDDFT can be applied via two main approaches: direct time propagation of the Kohn-Sham equations for real-time dynamics, or linear response theory for excitation energies by solving the Casida equation derived from the Dyson equation for the polarizability.

Time-dependent density functional theory (TDDFT) extends static DFT to describe electronic dynamics under time-varying external fields. Unlike static DFT limited to ground states, TDDFT enables simulation of excited states and electron motion during light-matter interactions. The mathematical framework uses time-dependent Schrödinger equations with time-dependent kinetic, Hartree, and exchange-correlation potentials. The Runge-Gross theorem establishes the theoretical foundation, analogous to Hohenberg-Kohn for static DFT. Most implementations use the adiabatic approximation, employing ground-state functionals evaluated at each instant. TDDFT originated in nuclear physics for atomic nucleus studies and has become essential for optical science, enabling calculations of electronic excitations, optical absorption, and non-linear phenomena across chemistry, physics, and materials science.

Time-Dependent Density Functional Theory (TDDFT) is a computational quantum mechanical framework that extends Density Functional Theory to describe electronic excited states and time-dependent phenomena in condensed matter systems, enabling predictions of optical properties, energy loss spectra, and other spectroscopic observables for neutral excitations through the linear response formalism.

TDDFT extends ground-state DFT by establishing one-to-one correspondence between time-dependent electron density and external potential for fixed initial states. This enables mapping interacting systems onto non-interacting ones reproducing density and its first time derivative. The Kohn-Sham potential comprises external, Hartree, and exchange-correlation (xc) potentials. The xc potential depends on initial states and density history, leading to non-local spatial dependence. The adiabatic approximation ignores memory dependence by using instantaneous density in ground-state functionals, but fails for non-adiabatic processes. To study whether non-adiabatic features persist in real 3D systems, researchers examine 3D helium atoms using close-coupling methods with partial wave expansions in spherical harmonics, discretizing radial degrees of freedom with finite element discrete variable representation.