Static-Screened Hartree-Fock Approximation

• Static-screened Hartree-Fock approximation is a method that replaces the bare Coulomb exchange with a zero-frequency screened interaction, mitigating divergence issues in electron gas models.
• It employs polarization bubble resummation to generate a self-consistent one-particle spectrum that avoids anomalously high Fermi velocities.
• While it improves spectral properties by softening exchange interactions, its static nature may yield higher total energies compared to fully dynamic treatments.

Searching arXiv for the cited papers and closely related screened-HF work to ground the article. {"query":"Static-screened Hartree-Fock approximation homogeneous electron gas screened exchange Hartree-Fock arXiv (Oca, 2012, Sabet et al., 2021, Caldarelli et al., 2024, Chang et al., 17 Jul 2025)","max_results":10,"sort_by":"relevance"} I found several directly relevant arXiv papers, including "Modifying the HF procedure to include screening effects" (Oca, 2012), "Minimal length implications on the Hartree-Fock theory" (Sabet et al., 2021), "Variational formulation of dynamical electronic response functions in presence of nonlocal exchange interactions" (Caldarelli et al., 2024), and "Harmonic generation of graphene quantum dots in Hartree-Fock approximation" (Chang et al., 17 Jul 2025). Searching for additional adjacent arXiv work on static screening in Hartree, hybrid, and dynamically screened Fock frameworks. {"query":"static screening Hartree Fock reduced Hartree Fock hybrid functionals screening band gap bilayer graphene arXiv (Koller et al., 2013, Levitt, 2018, Engdahl et al., 7 May 2025)", "max_results": 10, "sort_by": "relevance"} I also found adjacent papers relevant for positioning the topic: "Hybrid functionals for solids with an optimized Hartree-Fock mixing parameter" (Koller et al., 2013), "Screening in the finite-temperature reduced Hartree-Fock model" (Levitt, 2018), and "Screening of the band gap in electrically biased bilayer graphene: From Hartree to Hartree-Fock" (Engdahl et al., 7 May 2025). The static-screened Hartree-Fock approximation is a screened extension of Hartree-Fock in which the bare Coulomb interaction is replaced, at least in the exchange sector, by a statically screened interaction, typically evaluated at zero frequency. In the formulation proposed for the homogeneous electron gas, the approximation preserves the self-consistent one-particle-orbital structure of Hartree-Fock while incorporating screening effects usually associated with many-body perturbation theory through polarization-bubble dressing of Coulomb lines (Oca, 2012). In this sense it sits between ordinary Hartree-Fock, which uses bare exchange, and fully dynamical screened approaches, which retain retardation and frequency dependence. The topic is also closely connected to screened exchange, static Bethe-Salpeter kernels, and dielectric-dependent hybrid approximations, although several of those frameworks are only conceptually adjacent and are not identical to a static-screened Hartree-Fock scheme (Caldarelli et al., 2024, Koller et al., 2013).

1. Definition and formal scope

In the screened generalization of Hartree-Fock proposed in (Oca, 2012), the starting point is not the Slater determinant expectation value alone, but an energy functional built from the adiabatic switching-on of the Coulomb and nuclear interactions starting from a Slater determinant. Because the exact adiabatically connected functional is too complicated, the approximation is imposed diagrammatically by retaining only Wick contractions with a “screening” structure, namely diagrams in which Coulomb lines are dressed by polarization loops. The restricted functional is written as

$$E= \frac{\langle \Phi_0|[U_\alpha^{(c)+}HU_\alpha^{(c)}]_{WS}|\Phi_0\rangle} {\langle \Phi_0|[U_\alpha^{(c)+}U_\alpha^{(c)}]_{WS}|\Phi_0\rangle}.$$

Within this construction the screened Coulomb interaction is generated as a bubble-resummed potential,

$$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$

with polarization loop

$P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$

The simplest implementation for jellium is the static-screened Hartree-Fock approximation, obtained by replacing the screening kernel by its zero-frequency value. In that limit,

$$v_s(\vec q)=\frac{v_c(\vec q)}{1+P(\vec q,0)v_c(\vec q)},$$

so the approximation is explicitly non-retarded and the one-particle spectrum must be found self-consistently together with the static polarization (Oca, 2012).

2. Motivation from the failures of unscreened Hartree-Fock

The standard motivation is the pathology of bare-exchange Hartree-Fock for the homogeneous electron gas. For plane waves, the single-particle energy can be written as

$$\varepsilon(\mathbf{k})=\frac{\hbar^2 k^2}{2m} -\frac{2e^2}{\pi}k_{\rm F}F_0\!\left(\frac{k}{k_{\rm F}}\right),$$

with

$$F_0(x)=\frac12+\frac{1-x^2}{4x}\ln\left|\frac{1+x}{1-x}\right|, \qquad x=\frac{k}{k_{\rm F}}.$$

Two defects are emphasized repeatedly in later work: the divergence of the Fermi velocity at the Fermi surface and the prediction of an anomalous bandwidth not confirmed experimentally (Sabet et al., 2021). The divergence originates in the logarithmic singularity

$\ln\left|\frac{1+x}{1-x}\right|,$

which becomes singular as $x\to 1$. Since

$$\mathbf v_{\rm F}=\frac{1}{\hbar}\nabla_{\mathbf k}\varepsilon(\mathbf k)\Big|_{k=k_{\rm F}},$$

the derivative of the exchange-induced logarithm produces the divergent slope at the Fermi wavevector (Sabet et al., 2021).

For the same unscreened model, the Hartree-Fock ground-state energy per particle is

$$\frac{E}{N} = \frac{e^2}{2a_0} \left[ \frac35 (k_{\rm F}a_0)^2-\frac{3}{2\pi}(k_{\rm F}a_0) \right] = \left[ \frac{2.21}{(r_s/a_0)^2}-\frac{0.916}{(r_s/a_0)} \right]\mathrm{Ry}.$$

Static screening is introduced precisely to soften the long-range exchange kernel responsible for these spectral pathologies.

3. Self-consistent screened equations for jellium

In the homogeneous electron gas, translational invariance collapses the generalized formalism to a screened-exchange problem. The direct term cancels against the positive jellium background, so only kinetic and exchange contributions remain. The momentum-space screened interaction is

$$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$0

The polarization entering this interaction is of Lindhard type but is evaluated with the self-consistent single-particle energies: $$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$1

After imposing the static approximation,

$$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$2

The self-consistent cycle is then explicit. One starts with

$$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$3

so the first iteration reproduces ordinary Hartree-Fock: $$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$4 Then one computes $$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$5, updates $$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$6, recomputes the dispersion, and iterates to self-consistency (Oca, 2012). A plausible implication is that the method is best viewed as a static screened-exchange fixed-point problem rather than as ordinary Hartree-Fock with a perturbative correction.

4. Static approximation, spectral consequences, and energetic limitations

The principal success of the static-screened approximation in jellium is spectral rather than energetic. In ordinary Hartree-Fock, the derivative singularity at $$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$7 implies a vanishing density of states at the Fermi level. In the self-consistent static-screened scheme, the dispersion becomes much closer to free-electron-like and the derivative at $$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$8 is finite, so the density of states at the Fermi level no longer vanishes (Oca, 2012).

The same work emphasizes that the static approximation is extreme and non-retarded. In this limit both direct and exchange interactions are strongly screened, and the total energy is higher than the one given by the usual Hartree-Fock scheme (Oca, 2012). The stated interpretation is that the static approximation may over-screen exchange, whereas inclusion of retardation and dielectric-response sum rules in a more exact treatment can lead to energy lowering. Accordingly, static-screened Hartree-Fock is presented as a first approximation and a stepping stone toward a dynamically screened self-consistent scheme rather than as a final quasiparticle theory (Oca, 2012).

Later analysis of Hartree-Fock pathologies reinforces this interpretation. A review of minimal-length corrections to Hartree-Fock explicitly states that “another method to eliminate the divergence of Fermi velocity is the screening theory,” referring to Thomas-Fermi and Lindhard screening, while also stressing that merely presenting the static dielectric function is not equivalent to deriving a screened-exchange quasiparticle dispersion (Sabet et al., 2021). This distinction is central: a dielectric function by itself is an ingredient of static-screened Hartree-Fock, not the completed approximation.

5. Static screening kernels and neighboring screened-exchange formalisms

Several later arXiv works clarify the broader formal landscape into which static-screened Hartree-Fock fits. In a generalized linear-response setting with nonlocal exchange interactions, the nonlocal exchange operator is written as

$$v^s(x,x')= \left(v_c\frac{1}{1+Pv_c}\right)(x,x') = \left(\frac{1}{1+v_cP}v_c\right)(x,x'),$$9

with ordinary Hartree-Fock recovered for $P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$0 and screened exchange recovered when $P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$1 is taken as static (Caldarelli et al., 2024). In that formulation, the associated interaction kernel has the direct-minus-exchange structure

$P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$2

This is structurally identical to a static screened Fock operator in the particle-hole channel.

A finite-system realization appears in graphene quantum dots, where semiconductor Bloch equations are derived under a static-screened Hartree-Fock approximation. There the effective Hamiltonian is

$P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$3

with local Hartree potential and nonlocal Fock potential, and with the static screened interaction defined from

$P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$4

In that paper, the Hartree channel is associated with local-field or plasmonic effects, while the screened Fock term is identified with excitonic effects (Chang et al., 17 Jul 2025).

At the same time, several neighboring methods must be distinguished from static-screened Hartree-Fock. In biased bilayer graphene, the Fock self-energy is evaluated with a frequency-dependent RPA interaction

$P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$5

so the method is a self-consistent Hartree plus dynamically screened Fock scheme, not ordinary static-screened exchange (Engdahl et al., 7 May 2025). Dielectric-dependent screened hybrids are also only conceptually adjacent: they scale exact exchange with the static dielectric constant through

$P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$6

but do not construct a microscopic screened Fock operator $P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$7 (Koller et al., 2013).

6. Conceptual boundaries, common misconceptions, and rigorous background

A recurring misconception is to equate any use of static dielectric screening with a full static-screened Hartree-Fock approximation. The distinction is explicit in (Sabet et al., 2021). That work reviews the static Lindhard susceptibility,

$P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$8

and the corresponding static dielectric function,

$P(x,x')=\frac{-i}{\hbar}G(x,x')G(x',x).$9

but it does not actually insert $$v_s(\vec q)=\frac{v_c(\vec q)}{1+P(\vec q,0)v_c(\vec q)},$$0 into the Hartree-Fock exchange self-energy and derive the resulting screened dispersion (Sabet et al., 2021). The historical static-screened Hartree-Fock idea is therefore more specific than a review of Lindhard screening alone.

A second misconception is to treat rigorous results for screened Hartree models as if they already established screened exchange. In the finite-temperature reduced Hartree-Fock model for a periodic crystal with a small defect, the dielectric operator is defined as

$$v_s(\vec q)=\frac{v_c(\vec q)}{1+P(\vec q,0)v_c(\vec q)},$$1

and the linear screened response is

$$v_s(\vec q)=\frac{v_c(\vec q)}{1+P(\vec q,0)v_c(\vec q)},$$2

That paper proves total screening of small defect perturbations and shows that the corresponding linear screened interaction $$v_s(\vec q)=\frac{v_c(\vec q)}{1+P(\vec q,0)v_c(\vec q)},$$3 has an exponentially decaying kernel, but the model is reduced Hartree-Fock or Hartree, without exchange (Levitt, 2018). It therefore provides rigorous background for the static screened Coulomb sector, not a full static-screened Hartree-Fock theory.

Taken together, these results delimit the subject precisely. Static-screened Hartree-Fock is a self-consistent screened-exchange approximation in which the bare Coulomb interaction entering Hartree-Fock is replaced by a statically screened interaction, usually generated by polarization-bubble resummation and then frozen at $$v_s(\vec q)=\frac{v_c(\vec q)}{1+P(\vec q,0)v_c(\vec q)},$$4. Its canonical role is to regularize the exchange singularities of bare Hartree-Fock in the electron gas, especially the divergent Fermi velocity and vanishing density of states at the Fermi level. Its principal limitation is equally clear: in the extreme non-retarded approximation it improves the spectrum but may worsen the energy, which is why later developments repeatedly move toward dynamical screening, screened exchange in response theories, or hybridized dielectric-dependent approximations rather than stopping at the static limit (Oca, 2012, Engdahl et al., 7 May 2025).