Validity domain of the VAMPIRES variational average-atom model in the strongly coupled, strong-overlap regime of dense plasmas

The VAMPIRES model (Blenski and Piron, 2023) describes a dense ion-electron plasma by minimizing an approximate free-energy functional under two structural hypotheses: the electron density is a superposition of a uniform background and identical spherical clouds centered on each nucleus, and a cluster expansion is truncated at one-center (non-overlapping-ion) terms. This couples the electronic structure of ions self-consistently to the ion-fluid pair correlations and yields pressure ionization without any ad hoc confinement cavity. The source paper leaves the model's strongly coupled, strong-overlap regime unassessed: the two hypotheses are expected to fail there; the coupled self-consistent-field and ion-fluid integral equations failed to converge for lithium below 6 eV at densities above 5 g/cm³ and below 7 eV above 8 g/cm³; and no quantitative comparison with canonical-ensemble DFT molecular dynamics has been published. The problem is to determine the model's quantitative validity domain in that regime: obtain convergent, thermodynamically consistent solutions of the model equations where they currently fail, characterize the liquid-like structural feedback that arises at strong coupling, and delimit — against explicit DFT molecular-dynamics benchmarks — where the superposition ansatz and one-center truncation remain quantitatively accurate and where they break down.

Background

Dense plasmas and warm dense matter — matter at near-solid or higher densities and temperatures from a few to a few hundred electronvolts, as encountered in inertial-confinement fusion and in the interiors of stars and giant planets — are awkward for both atomic physics and classical plasma physics: ions are strongly coupled (their mutual Coulomb energy exceeds their thermal energy, so they form a correlated liquid rather than an ideal gas), while electrons are neither fully bound in atoms nor fully free. Equation-of-state and opacity predictions for this regime nevertheless feed major simulations, which motivates average-atom models: rather than tracking every particle, one computes a single representative ion — a nucleus surrounded by a partially ionized electron cloud — embedded in a model of its plasma environment, and derives thermodynamic quantities (pressure, internal energy, mean ionization $Z^{*}$) from it.

The classical implementations of this idea, such as the INFERNO and VAAQP models, confine the ion's electronic structure to a Wigner-Seitz cavity — a sphere holding one ion's average volume, outside which the electrons form a uniform neutralizing background. Pressure ionization, the reduction and eventual suppression of bound states by the plasma environment (often discussed under the name of continuum lowering), is then enforced by the cavity geometry itself. The cavity is a useful heuristic but is poorly motivated physically, especially at weak and strong coupling, and defining the validity domain of atom-in-plasma models is described in the field's recent assessments as among the most challenging open issues. A related line, the quantum hyper-netted-chain (QHNC) approach, couples electronic structure to ion-ion correlations but requires an extra assumption to fix the free-electron density because it lacks a free-energy formulation.

VAMPIRES (variational atomic model of plasma accounting for ion radial correlations and electronic structure of ions) was constructed to remove these limitations. Its hypotheses are: nuclei are classical indistinguishable particles; the electron density takes the superposition form

$n(\mathbf{r})=n_{0}+\sum_{i}q(|\mathbf{r}-\mathbf{R}_{i}|),$

where $n_{0}$ is the uniform free-electron background, $q(r)$ the spherical electron cloud around a nucleus at position $\mathbf{R}_{i}$; the plasma free energy is approximated by a cluster expansion truncated at one-center terms, which presumes non-overlapping ions; and the resulting ion fluid is described by an approximate integral equation of classical liquid theory — the hypernetted-chain (HNC) equation or, at weaker coupling, its Debye-Hückel limit (the classical theory of linearized electrostatic screening). Minimizing the resulting approximate free-energy functional — with the electronic structure treated either by Kohn-Sham density-functional theory (a mean-field quantum method using an approximate exchange-correlation functional) or by its semiclassical Thomas-Fermi simplification — yields a closed, self-consistent set of equations: the cloud $q(r)$ determines an effective one-center electronic potential through a convolution with the ion pair correlation function $h(r)$ (the probability, relative to an ideal gas, of finding another ion at distance $r$), and $h(r)$ in turn responds to the effective ion-ion interaction built from $q(r)$. The construction satisfies the virial theorem (the exact pressure-energy consistency relation) by design and produces pressure ionization as an emergent consequence of the ion-fluid structure, with no cavity.

In the published numerical application — lithium at compressions from 1/10 to 20 of normal density (mass densities up to about 10 g/cm³) and temperatures from 5 to 300 eV, using the quantum version with HNC — the model yields mean ionizations and pressures broadly consistent with INFERNO and VAAQP at weak and moderate coupling, and displays an intriguing strong-coupling behavior: the pair correlation function develops liquid-like oscillations that carve repulsive regions (electron-cloud overlap) into the effective potential, and pressure ionization sets in together with the transition to liquid-like ion structure. However, the authors did not succeed in converging the coupled equations at the lowest temperatures and highest densities (below 6 eV above 5 g/cm³, and below 7 eV above 8 g/cm³), and no comparison with density-functional-theory molecular dynamics (DFT-MD) — in which the many-ion problem is simulated directly with thermostatted (Nosé-Hoover) canonical dynamics — has been published. Precisely in this strongly coupled, strong-overlap regime the two structural hypotheses, the superposition ansatz and the one-center cluster truncation, are expected to become inaccurate; the authors' own later assessments (a 2024 review and habilitation thesis) describe the strong-coupling results as puzzling and the validity assessment as unfinished work.

Problem Statement

Determine the quantitative validity domain of the VAMPIRES average-atom model in the strongly coupled, strong-overlap regime of dense plasmas — the regime in which its two structural hypotheses are expected to fail and in which the model has so far been neither solved nor validated. Concretely, for dense plasmas such as the lithium test case of the source paper:

1. Solvability and numerical robustness. Establish whether the coupled VAMPIRES equations — the self-consistent-field problem for the spherical electron cloud $q(r)$ and the liquid-integral equation (hypernetted-chain, or Debye-Hückel at weak coupling) for the ion pair correlation function $h(r)$, obtained by minimizing the approximate free-energy functional under the ansatz $n(\mathbf{r})=n_{0}+\sum_{i}q(|\mathbf{r}-\mathbf{R}_{i}|)$ — admit solutions in the regime where the published computations failed to converge (lithium: below 6 eV at matter densities above 5 g/cm³, and below 7 eV above 8 g/cm³), and, where such solutions exist, provide a numerical scheme that reaches them with controlled precision while preserving the model's exact consistency properties, in particular virial-theorem agreement between the pressure routes at the percent level achieved elsewhere.

2. Structural behavior. Characterize the strong-coupling solutions, in which liquid-like oscillations of $h(r)$ feed back into the electronic structure through the effective potential, and decide whether the static average-environment picture — each ion's electrons feeling the surrounding ions only through the mean pair correlation function — remains physically meaningful there or breaks down.

3. Quantitative validity. By systematic comparison with canonical-ensemble DFT molecular-dynamics reference calculations, quantify the error induced by the two core hypotheses (the electron-density superposition ansatz and the one-center cluster truncation) in equation-of-state quantities, mean ionization, and the ion pair correlation function, as a function of density and temperature, and delimit the region of the phase diagram within which VAMPIRES predictions are quantitatively reliable.

These three components jointly support one objective — a defensible validity domain for the model — and a complete answer must address all of them: a converged solver without benchmarking, or a benchmark restricted to the already-accessible moderate-coupling regime, does not settle where the model's approximations hold. The extensions announced by the authors (detailed, level-resolved accounting of ion excited states; plasma mixtures; two-temperature systems) are separate development tracks and are not part of this problem's completion boundary.

References

1. T. Blenski and R. Piron, 'A variational atomic model of plasma accounting for ion radial correlations and electronic structure of ions (VAMPIRES)', Physical Review E 107, 035209 (2023), DOI: 10.1103/PhysRevE.107.035209, arXiv:2211.06258, https://doi.org/10.1103/PhysRevE.107.035209
2. R. Piron, 'Atomic Models of Dense Plasmas, Applications, and Current Challenges', Atoms 12(4), 26 (2024), DOI: 10.3390/atoms12040026, arXiv:2311.15418, https://doi.org/10.3390/atoms12040026
3. R. Piron, 'Atoms in Dense Plasmas: Models, Applications, and Current Challenges', habilitation thesis, arXiv:2411.18357 (2024), DOI: 10.48550/arXiv.2411.18357, https://arxiv.org/abs/2411.18357
4. R. Piron and T. Blenski, 'Variational average-atom model of electron-ion plasma with correlations and quantum bound electrons', arXiv:1904.13289 (2019), DOI: 10.48550/arXiv.1904.13289, https://arxiv.org/abs/1904.13289