Interaction-suppressed equilibrium populations of atomic bound states in gases and plasmas of finite density: the occupation-probability problem originating in Fermi's 1924 excluded-volume theory

In 1924, Fermi showed that the finite volume of an atom suppresses the equilibrium population of its excited quantum states in a gas of finite density, giving the first physical truncation of the otherwise divergent atomic internal partition function. His derivation dropped the attractive part of the van der Waals interaction, assumed a small excluded volume ($b \ll V$), retained only the ground-state population in the suppression factor, modeled the state volume classically as a sphere of the Kepler-orbit radius, treated only hydrogen, and lacked experimental verification. A century of later work — quantum statistics and the Beth–Uhlenbeck virial expansion, the Planck–Larkin partition function, the Hummer–Mihalas occupation-probability formalism in stellar equation-of-state practice, and recent ionization-potential-depression measurements — has settled the weak-coupling (Debye–Hückel) limit and produced widely used heuristic frameworks, but the general core remains open: a thermodynamically consistent, quantum-mechanical determination of the occupation probabilities of atomic bound states in non-ideal (strongly coupled, pressure-ionized) matter for general atoms and ions, where any satisfactory resolution should be confrontable with, and ideally discriminated by, experimental data on the suppression of highly excited states across densities and temperatures. Current reviews still pose 'how to account for these states?' as an open question, existing models 'exhibit largely varying results', and first-principles approaches have been demonstrated only for hydrogen.

Background

Statistical mechanics assigns an atom in thermal equilibrium at temperature $T$ the probability $p_i \propto g_i e^{-E_i/kT}$ of occupying a bound state (a 'quantum state') of energy $E_i$ and degeneracy $g_i$. For hydrogen, the old quantum theory pictured the states as Kepler ellipses whose semi-major axis grows as the square of the principal quantum number $n$, with energies $E_n = -R/n^2$ ($R$ is the Rydberg energy) and weights $g_n \propto n^2$. Because $e^{-E_n/kT} \to 1$ as $n \to \infty$ while the number of states grows without bound, the internal partition function $Q_{\mathrm{int}} = \sum_i g_i e^{-E_i/kT}$ — the sum over an atom's bound levels that governs level populations and, through the Saha equation (the equilibrium relation between the densities of successive ionization stages of a gas), all ionization equilibria — diverges formally. Any theory of a real gas of atoms must therefore explain which excited states survive at a given density.

Fermi (1924) — elaborating a 1916 heuristic truncation argument of Herzfeld — gave the first treatment grounded in free-energy minimization. He computed the equilibrium among the states of the same atom taking the atom's finite volume into account: using the van der Waals equation of state without the attractive term $A/v^2$, i.e. $p(V-b) = NkT$ with $N$ the number of atoms and $b$ the total excluded volume, generalized so that each state $i$ contributes its own volume $v_i$, he minimized the Helmholtz free energy. In the approximations that $b \ll V$ and that only the ground state is significantly populated, and modeling the volume of the hydrogen state $n$ as a sphere of radius equal to the semi-major axis of the Kepler ellipse (so $v_n \propto n^6$), he found that each Boltzmann weight must be multiplied by an 'a priori probability' factor of the form $\exp(-N v_n/V)$, with $N/V$ the atom number density: large, weakly bound states are realized less often than ordinary statistics predicts, and the partition function converges. The paper's own limitations define the open problem: attractive interactions dropped; $b \ll V$ assumed; a single populated state; a classical rather than quantum-mechanical state volume; hydrogen only; no experimental verification; no link to the quantum statistics of identical particles (Fermi–Dirac statistics appeared two years later).

Later work recast this into the internal-partition-function or occupation-probability problem of non-ideal gases and plasmas. In modern vocabulary, an occupation probability $w_i \le 1$ multiplies each level's Boltzmann weight, representing environmental destruction of the level; on the energy side, continuum lowering or ionization-potential depression (IPD) denotes the reduction of the ionization energy by the plasma environment. A rigorous low-density treatment exists: the Beth–Uhlenbeck quantum virial expansion expresses the second virial coefficient — the first density correction to the ideal-gas equation of state — through two-body bound-state and scattering contributions, and the associated Planck–Larkin partition function is convergent; however, this perturbative route is in practice restricted to the Debye–Hückel regime, the mean-field theory of weakly coupled plasmas in which charges are screened over the Debye length. At the higher densities where suppression and pressure ionization actually matter, standard practice relies on heuristic models: the Hummer–Mihalas occupation probabilities (Fermi's idea extended to all elements and ionization stages, with neutral and charged-particle interactions) inside the Mihalas–Hummer–Däppen (MHD) equation of state for stellar envelopes, or ion-sphere/continuum-lowering pictures in which each ion sits at the center of a Wigner–Seitz cavity — a sphere of the volume per ion, filled with a uniform free-electron background that neutralizes the central ion. These models disagree with one another, their statistical-mechanical foundations are contested, and the regimes they target — dense partially ionized plasmas, warm dense matter (near-solid density at electron-volt temperatures), stellar interiors, and inertial-confinement-fusion fuel — are precisely those of greatest current experimental and astrophysical interest.

Problem Statement

Determine, from quantum statistical mechanics, the equilibrium probabilities of the bound quantum states of atoms and ions in matter of finite density, in the source-faithful generality of Fermi (1924) and its modern descendants. Concretely, the problem asks for: (i) a derivation, from the interacting many-body system of nuclei and electrons, of the level-dependent occupation probabilities (equivalently, a convergent internal partition function) that replace the bare Boltzmann weights of isolated atoms — including both neutral (excluded-volume and attractive van der Waals) and charged-particle (screening) interactions — without the restrictions to negligible excluded volume ($b \ll V$), to a single populated state, or to the weak-coupling Debye–Hückel limit, and therefore valid also in strongly coupled and pressure-ionized regimes where non-ideality corrections are comparable to or exceed the binding energies of weakly bound states; (ii) demonstration that the result is thermodynamically consistent (extensive, satisfying the Maxwell and Gibbs–Duhem relations) rather than resting on an ad hoc factorization; and (iii) applicability to general atoms and ions with multiple simultaneously populated excited states and arbitrary ionization stages, not only hydrogen — understood as a formulation general in form for arbitrary atoms, ions, and ionization stages, with its extension procedure stated and demonstrated for at least one multi-electron species and one non-hydrogenic ionization stage (it does not require tabulated results for every element). These deliverables jointly support one overarching objective — the statistically consistent prediction of which atomic bound states exist and are populated in interacting matter — and all three are required together for a resolution, so a result that addresses only part of them does not resolve the task. On the experimental side, existing checks of the predicted suppression of highly excited states (level dissolution / ionization-potential depression) — including the direct ionization-potential-depression measurements of Ciricosta et al. (2012) and Hoarty et al. (2013) — are scarce, carry large uncertainties, and do not discriminate among models; a satisfactory resolution should therefore be confrontable with, and ideally discriminated by, experimental data on bound-state suppression across a range of densities and temperatures.

References

1. E. Fermi, 'Über die Wahrscheinlichkeit der Quantenzustände', Zeitschrift für Physik 26, 54–56 (1924). DOI: 10.1007/BF01327311. https://doi.org/10.1007/BF01327311 (primary source)
2. K. F. Herzfeld, 'Zur Statistik des Bohrschen Wasserstoffatommodells', Annalen der Physik 356, 261–284 (1916). DOI: 10.1002/andp.19163561903. https://doi.org/10.1002/andp.19163561903 (1916 precursor to Fermi's argument)
3. H. C. Urey, 'The Distribution of Electrons in the Various Orbits of the Hydrogen Atom', The Astrophysical Journal 59, 1 (1924). DOI: 10.1086/142787. https://doi.org/10.1086/142787
4. Max Planck, 'Zur Quantenstatistik des Bohrschen Atommodells', Annalen der Physik 380, 673–684 (1924). DOI: 10.1002/andp.19243802302. https://doi.org/10.1002/andp.19243802302
5. E. Fermi, 'Zur Quantelung des idealen einatomigen Gases', Zeitschrift für Physik 36, 902–912 (1926). DOI: 10.1007/BF01400221. https://doi.org/10.1007/BF01400221
6. G. E. Uhlenbeck and E. Beth, 'The quantum theory of the non-ideal gas I. Deviations from the classical theory', Physica 3, 729–745 (1936). DOI: 10.1016/S0031-8914(36)80346-2. https://doi.org/10.1016/S0031-8914(36)80346-2
7. E. Beth and G. E. Uhlenbeck, 'The quantum theory of the non-ideal gas. II. Behaviour at low temperatures', Physica 4, 915–924 (1937). DOI: 10.1016/S0031-8914(37)80189-5. https://doi.org/10.1016/S0031-8914(37)80189-5
8. F. J. Rogers, 'Occupation numbers for reacting plasmas — The role of the Planck–Larkin partition function', The Astrophysical Journal 310, 723 (1986). DOI: 10.1086/164725. https://doi.org/10.1086/164725
9. D. G. Hummer and D. Mihalas, 'The equation of state for stellar envelopes. I — An occupation probability formalism for the truncation of internal partition functions', The Astrophysical Journal 331, 794–814 (1988). DOI: 10.1086/166600. https://doi.org/10.1086/166600
10. D. Mihalas, W. Däppen and D. G. Hummer, 'The equation of state for stellar envelopes. II — Algorithm and selected results', The Astrophysical Journal 331, 815–825 (1988). DOI: 10.1086/166601. https://doi.org/10.1086/166601
11. A. Nayfonov, W. Däppen, D. G. Hummer and D. Mihalas, 'The MHD Equation of State with Post-Holtsmark Microfield Distributions', The Astrophysical Journal 526, 451–464 (1999). DOI: 10.1086/307972. https://doi.org/10.1086/307972
12. M. R. Zaghloul, 'Inconsistency in Fermi's probability of the quantum states', The European Physical Journal H 36, 401–406 (2011). DOI: 10.1140/epjh/e2011-20018-8. https://doi.org/10.1140/epjh/e2011-20018-8 (preprint: arXiv:1109.2479, https://arxiv.org/abs/1109.2479)
13. O. Ciricosta, S. M. Vinko, H.-K. Chung, B.-I. Cho, C. R. D. Brown, T. Burian, J. Chalupský, K. Engelhorn, R. W. Falcone, C. Graves, V. Hájková, A. Higginbotham, L. Juha, J. Krzywinski, H. J. Lee, M. Messerschmidt, C. D. Murphy, Y. Ping, D. S. Rackstraw, A. Scherz, W. Schlotter, S. Toleikis, J. J. Turner, L. Vysín, T. Wang, B. Wu, U. Zastrau, D. Zhu, R. W. Lee, P. Heimann, B. Nagler and J. S. Wark, 'Direct Measurements of the Ionization Potential Depression in a Dense Plasma', Physical Review Letters 109, 065002 (2012). DOI: 10.1103/PhysRevLett.109.065002. https://doi.org/10.1103/PhysRevLett.109.065002
14. D. J. Hoarty, P. Allan, S. F. James, C. R. D. Brown, L. M. R. Hobbs, M. P. Hill, J. W. O. Harris, J. Morton, M. G. Brookes, R. Shepherd, J. Dunn, H. Chen, E. Von Marley, P. Beiersdorfer, H. K. Chung, R. W. Lee, G. Brown and J. Emig, 'Observations of the Effect of Ionization-Potential Depression in Hot Dense Plasma', Physical Review Letters 110, 265003 (2013). DOI: 10.1103/PhysRevLett.110.265003. https://doi.org/10.1103/PhysRevLett.110.265003
15. R. Piron, 'Atomic Models of Dense Plasmas, Applications, and Current Challenges', Atoms 12(4), 26 (2024). DOI: 10.3390/atoms12040026. https://doi.org/10.3390/atoms12040026 (preprint: arXiv:2311.15418, https://arxiv.org/abs/2311.15418)
16. M. Bonitz and L. Kordts, 'Ionization Potential Depression and Fermi Barrier in Warm Dense Matter: A First-Principles Approach', Contributions to Plasma Physics 65, e70001 (2025). DOI: 10.1002/ctpp.70001. https://doi.org/10.1002/ctpp.70001