Spatially Nonlocal Dielectric Functions
- Spatially nonlocal dielectric functions describe material responses where the electric displacement depends on both local and surrounding electric fields, capturing non-pointwise screening effects.
- They modify wave propagation and plasmon resonance behaviors in systems ranging from biophysical solvation to quantum electronic materials.
- Advanced computational methods, including Fourier transforms and surface-integral techniques, enable practical modeling of nonlocal effects in complex geometries.
A spatially nonlocal dielectric function describes electromagnetic or electrostatic material response in which the electric displacement field D at a point depends not merely on the local electric field E, but also on field values in a finite surrounding region. This nonlocality, or spatial dispersion, is essential for accurate theoretical modeling of media where the screening, polarization, or collective electronic phenomena cannot be captured by pointwise constitutive relations. It arises across a spectrum of physical systems, from biophysical solvation to quantum plasmonics, and modifies the fundamental behavior of wave propagation, energy storage, and fluctuation-induced forces.
1. Constitutive Law and Mathematical Framework
The general nonlocal dielectric constitutive law is expressed in real space as
where the kernel encodes the spatial correlation of polarization response. For isotropic and homogeneous media,
Transformation to Fourier space renders the convolution into simple multiplication: The choice of kernel, and the analytic properties of (or for susceptibility), are dictated by physical and causality requirements, as well as the microscopic characteristics of the medium (Bardhan et al., 2012, Llosa et al., 2019).
2. Prototypical Nonlocal Models
2.1 Lorentz (Dogonadze–Kornyshev) Model in Electrostatics
In molecular and continuum electrostatics, the Lorentz model introduces a finite correlation length for dipolar solvent polarization: The spectral (Fourier-space) permittivity
interpolates smoothly from at long wavelengths () to at large , reflecting reduced screening for fields varying on atomic scales. This framework enables analytical or semi-analytical solutions in separable geometries (e.g., spheres), via expansion in surface spherical harmonics and the diagonalization of boundary integral operators (Bardhan et al., 2012).
2.2 Hydrodynamic Drude Model in Plasmonics
In metallic systems, especially at the nanoscale, the hydrodynamic Drude model captures nonlocality by including a pressure term for the conduction electron gas: where is the bulk plasma frequency, the damping rate, and the effective speed. This modifies classical plasmon resonance conditions, regularizes divergences in the local field at nanometric separations, and introduces new longitudinal (volume) plasmon modes inaccessible in local models (0912.4746, Yan et al., 2013).
2.3 Hopfield–Thomas and Halevi–Fuchs Oscillator Models
For dielectrics with spatial dispersion, the susceptibility tensor admits
with model functions
allowing independent control over the transverse and longitudinal response branches. This structure is vital at boundaries, interfaces, and in energy-density calculations (Churchill et al., 2016, Singer et al., 2015).
2.4 Nonlocal Dielectric Response in Quantum Materials
First-principles calculations using quantum field theory (e.g., via polarization tensors in graphene) yield dielectric functions with fully nonlocal, frequency and wavenumber dependence. Notably, in graphene the transverse dielectric function exhibits a double pole in at low frequency and finite , driving physically significant corrections in fluctuation-induced forces such as the Casimir effect (Mostepanenko et al., 15 Jan 2026).
3. Boundary-Value Formulations and Interface Phenomena
Spatial nonlocality necessitates advanced methodologies at interfaces, as the standard set of Maxwell boundary conditions becomes insufficient. Additional boundary conditions (ABC) are required, often cast as constraints on normal polarization or its derivatives (Churchill et al., 2016, Singer et al., 2015): with parameters reflecting microscopic surface interaction models (e.g., Pekar, Fuchs–Kliewer). In integral-equation approaches for spheres or arbitrary bodies, the use of eigenfunction expansions (spherical harmonics, for instance) allows reduction to modewise algebraic systems (Bardhan et al., 2012). Boundary-integral and surface Green-function methods have been generalized to handle the extra equations and singular behavior of nonlocal kernels (Yan et al., 2013).
Metasurface/boundary implementations (e.g., for wide-angle antireflection) now employ spatially dispersive sheet impedances , equivalently representing nonlocal permittivity at the interface and enabling functionalities unattainable with local models (Zhuravlev et al., 16 Jul 2025).
4. Analytic Properties, Causality, and Physical Constraints
The extension of Kramers–Kronig relations to nonlocal media introduces deep constraints on permissible forms of (Llosa et al., 2019). Finite-propagation speed implies analyticity in two complex variables, enforced by
thus excluding noncausal forms. The Hilbert-transform (Kramers–Kronig) relation for spatially dispersive response becomes
with dependence on both frequency and wavenumber. Physically, these requirements guarantee correct signal propagation and the preservation of sum rules (e.g., –sum rule) adapted to nonlocality.
The analytic structure further ensures proper limiting behavior: as the nonlocal kernel's spatial scale (or ), models recover the standard local dielectric relations, while at large spatial dispersion leads to cutoff or saturation effects (e.g., attenuation suppression in near-field heat transfer) (Singer et al., 2015).
5. Physical Consequences and Experimental Implications
Spatially nonlocal dielectric functions yield several key physical consequences:
- Mode-Dependent Screening: In electrolyte and soft-matter models (e.g., Lorentz model), different angular harmonic orders experience effective dielectric constants ranging between the static and high-frequency limits, with high- (short-wavelength) modes weakly screened (Bardhan et al., 2012).
- Plasmonic Volume Modes and Blueshifts: Nonlocality is necessary to describe the existence and quantitative features of longitudinal plasmon modes in metallic nanostructures, leading to blueshifts of resonance frequencies and regularization of near-contact divergences (0912.4746, Yan et al., 2013).
- Suppression of Divergences: In near-field thermal and EM energy density calculations, nonlocality regularizes the divergence in the local energy density as from a boundary, cutting off unphysically large densities and ensuring thermodynamic consistency (Churchill et al., 2016).
- Fluctuation-Induced Forces: The behavior of the transverse dielectric function at low and finite is pivotal in resolving the long-standing Casimir "Drude vs plasma" puzzle; nonlocal models (both phenomenological and field-theoretical) predict correct temperature and material dependences in both metals and graphene (Mostepanenko et al., 15 Jan 2026, Klimchitskaya et al., 2021, Klimchitskaya et al., 2021).
- Wave Manipulation and Metasurfaces: Practical designs of metasurfaces for angularly robust antireflection rely critically on synthesizing nonlocal, spatially dispersive impedance profiles, matched to the phase accumulation in thick slabs (Zhuravlev et al., 16 Jul 2025).
6. Computational and Experimental Methods
Numerically, handling spatially nonlocal dielectric functions is challenging due to dense convolution integrals. Efficient strategies include:
- Trefftz Basis and Pseudo-Trefftz Methods: Using basis functions that satisfy the nonlocal governing equations exactly on small patches, enabling sparse local assembly and rapid convergence (Tsukerman, 2020).
- Plane-Wave and FFT Convolutions: Fourier-based approaches facilitate efficient treatment of convolutional dielectric kernels in high-dimensional problems, as seen in nonlocal DFT-embedded solvation (Islam et al., 2023).
- Surface-Integral (BEM/GSIM) and Mode Decomposition: Inclusion of nonlocal longitudinal modes in combination with regularization of singularities allows for robust computations in arbitrary geometries (Yan et al., 2013).
- FDTD Implementations: Nonlocal response enters as Laplacian terms in auxiliary current equations, yielding explicit and stable time-domain updates for Maxwell equations in complex structures (0912.4746).
For experimentally relevant system design, the ability to map microstructure to effective enables engineering of spectral windows for transparency, index control, and loss minimization in composites (Torquato et al., 2020).
7. Broader Applications and Theoretical Implications
Spatially nonlocal dielectric functions extend beyond traditional optics, with demonstrated or potential impact in:
- Biomolecular Electrostatics: Quantitative modeling of solvation energetics, reaction potentials, and interfacial capacitance in proteins, electrolytes, and electrochemical interfaces (Bardhan et al., 2012, Tsukerman, 2020, Islam et al., 2023).
- Nano-Optics and Metamaterials: Control of scattering, mode dispersion, and field confinement in metasurfaces, nanoantennas, and all-dielectric photonic platforms; improved discrete-dipole models for Mie-resonant systems incorporating nonlocal corrections (Bobylev et al., 2020).
- Fundamental Fluctuation Phenomena: Accurate predictions of Casimir and van der Waals interactions, near-field radiative heat transfer, energy-density distributions, and spontaneous emission rates, unified under the formalism of nonlocal response (Mostepanenko et al., 15 Jan 2026, Klimchitskaya et al., 2021, Klimchitskaya et al., 2021, Singer et al., 2015, Churchill et al., 2016).
- Quantum Electronic Materials: Ab initio QFT analysis of 2D materials, revealing features (e.g., double pole in ) not captured in semi-phenomenological treatments (Mostepanenko et al., 15 Jan 2026).
These advances have unified previously divergent experimental and theoretical results (e.g., Casimir force discrepancies), and opened routes for predictive, microstructure-driven design in both classical and quantum electromagnetic systems.
References:
(Bardhan et al., 2012, Tsukerman, 2020, Torquato et al., 2020, 0912.4746, Yan et al., 2013, Churchill et al., 2016, Mostepanenko et al., 15 Jan 2026, Islam et al., 2023, Zhuravlev et al., 16 Jul 2025, Klimchitskaya et al., 2021, Llosa et al., 2019, Singer et al., 2015, Bobylev et al., 2020, Klimchitskaya et al., 2021, Grigoryan, 2017).