Dyson (1971) showed that a classical binary ionic mixture in a rigid uniform electron background forms chemically bound ordered compounds — NaCl-type, CsCl-type, and ZY3-type lattices of point charges — in definite windows of the charge ratio x=Y/Zx=Y/Z, with binding energies of order 10−310^{-3} of the Coulomb energy, and proposed that an FeHe compound in the NaCl structure occurs in neutron-star crusts, with the ordered iron lattice shielding the helium nuclei from fusion. His determination was incomplete in four respects: (i) zero-point (phonon) energies, comparable to or larger than the binding itself at relevant densities, were treated only by a dimensional estimate; (ii) electron screening was included only to first order in the Thomas–Fermi expansion, while higher-order terms are of the same size as the binding; (iii) only three cubic structures were examined, and their dynamical (phonon) stability was not checked; and (iv) the suppression of pycnonuclear (density-driven, zero-temperature tunnelling) fusion of the light nuclei was estimated only through a dimensional Gamow-type factor. This problem record asks for the completion of that determination: a stability map of ordered binary Coulomb compounds over the candidate lattice structures and charge-ratio windows in which the sign and magnitude of the net binding are fixed with a controlled error budget after inclusion of zero-point phonon energies, verification of phonon stability, and electron-response corrections beyond first-order screening; and a computation of pycnonuclear reaction rates of light nuclei in ordered compound lattices accurate enough to decide whether such lattices preserve light nuclei for astrophysically long times. Witten (1974) settled the prototype compounds at the level of zero-point plus first-order screening; the general quantum-corrected determination and the compound-lattice pycnonuclear rates remain open.

Background

In the crust of a neutron star and the core of a white dwarf, matter is pressure-ionized: fully stripped nuclei (ions) of charge ZeZe move in a degenerate electron gas that is nearly uniform, because the electron Fermi energy greatly exceeds the Coulomb energy per particle. Below the melting temperature the ions freeze into a Coulomb lattice (a Wigner crystal); the electrostatic ground state of a single ionic species (the one-component plasma, OCP) is the body-centered cubic (bcc) lattice. The lattice energy is the Madelung energy — the electrostatic self-energy of point charges in a rigid uniform neutralizing background — of order Z2e2/aZ^2e^2/a per ion, where aa is the ion-sphere radius (3/4πn)1/3(3/4\pi n)^{1/3} and nn the ion number density.

A binary ionic mixture (BIM) contains two species with charges ZeZe and YeYe, Y<ZY<Z; the charge ratio x=Y/Zx=Y/Z (or its inverse α\alpha) is the main structure-determining parameter. The default comparison state is the phase-separated mixture of two pure OCP crystals, whose electrostatic energy is given to high accuracy by the linear mixing rule, the charge-weighted sum of the two OCP energies. The question of chemical binding is whether some ordered compound lattice — an alternating arrangement of the two species, such as the NaCl, CsCl, or ZY3ZY_3 structures borrowed from ionic crystals (ZY3ZY_3 denoting a cubic structure with one ZZ ion and three YY ions per formula unit) — has a lower energy than that mixture. Because the background is uniform, the classical energy ratio R(x)=Ecompound/EmixtureR(x)=E_{\rm compound}/E_{\rm mixture} is a function of xx alone, independent of density.

Dyson (1971) computed the classical Madelung energies of three cubic binary lattices and found stability windows with R(x)>1R(x)>1: the NaCl type for 4.46×10−4<x<0.19244.46\times10^{-4}<x<0.1924, with maximum binding R−1≃3.04×10−3R-1\simeq3.04\times10^{-3} near x=0.072x=0.072; the CsCl type for 0.413<x<10.413<x<1; and a ZY3ZY_3-type lattice for x<0.0182x<0.0182. Since helium and iron have x=2/26≃0.077x=2/26\simeq0.077, close to the NaCl optimum, Dyson proposed that a FeHe compound in the NaCl structure is the ground state of helium-rich matter at neutron-star-crust densities, bound by roughly 0.3%0.3\% of the Coulomb energy per ion (of order keV per helium nucleus at ρ∼108 g cm−3\rho\sim10^8\ \mathrm{g\,cm^{-3}} and ∼100\sim100 keV at 1012 g cm−310^{12}\ \mathrm{g\,cm^{-3}}), and that the ordered lattice, by keeping helium nuclei farther apart than in a pure helium crystal, might suppress their fusion.

Four effects, each potentially comparable to the binding, were beyond that determination.

(1) Zero-point energy. Ions are not fixed on lattice sites; in the harmonic approximation each normal mode ll of wave vector kk contributes 12ℏωl(k)\frac{1}{2}\hbar\omega_l(k), so the zero-point energy is an integral over the phonon spectrum obtained by diagonalizing the lattice dynamical matrix (the matrix of interionic force constants whose eigenvalues are the squared phonon frequencies). For the HeFe mixture it is Pm≃1.5×10−4[ρ/(1 g cm−3)]1/6CmP_m\simeq1.5\times10^{-4}[\rho/(1\ \mathrm{g\,cm^{-3}})]^{1/6}C_m per nucleus (Witten 1974), about 1.5%1.5\% of the Coulomb energy CmC_m at ρ=1012 g cm−3\rho=10^{12}\ \mathrm{g\,cm^{-3}} — several times the compound binding — so only the difference of zero-point energies between compound and mixture decides the sign, and phonon spectra must be integrated to ∼10−4\sim10^{-4} relative accuracy. The phonon calculation also tests dynamical stability: a lattice is dynamically stable only if all ωl(k)\omega_l(k) are real (no soft modes), and compound lattices are known to develop phonon instabilities in parts of parameter space (e.g., the CsCl-type lattice becomes phonon-unstable when the inverse charge ratio α=Z/Y\alpha=Z/Y exceeds 3.596, while the NaCl lattice is phonon-stable only for α>3.9\alpha>3.9).

(2) Electron screening. The electron background is compressible and polarizes around the ions. In the Thomas–Fermi (TF) approximation the response is exponential screening with wave number kTFk_{\rm TF}; expanding in the small parameter (kTFa)2(k_{\rm TF}a)^2 gives a first-order correction, which Dyson included, and second- and higher-order terms of relative size set by the ratio C/FC/F of electrostatic to electron Fermi energy, comparable to the binding at crust densities. The full response is the random-phase-approximation (RPA, or Lindhard) dielectric function of the degenerate electron gas.

(3) Structure completeness. Only three cubic binary structures of point ions were considered; real ions have finite charge radii at outer-crust densities, and later surveys have identified many more candidate ordered structures (Kozhberov 2020; Kozhberov 2024; Engstrom, Yoder, and Crespi 2016 for ternaries).

(4) Pycnonuclear reactions. At zero temperature, fusion proceeds by quantum tunnelling through the Coulomb barrier, driven by the zero-point motion of ions in the lattice ("pycnonuclear" = density-driven fusion, independent of temperature). The rate is exponentially sensitive to the inter-nuclear equilibrium distance and to the lattice vibrational frequencies, so the microstructure of the Coulomb solid controls the rate. Whether an ordered compound lattice in fact preserves its light nuclei for astrophysically long times (∼1010\sim10^{10} yr) is a quantitative question about a multidimensional tunnelling problem that Dyson estimated only dimensionally through a Gamow-type factor.

Whether such compounds occur is constrained by nuclear equilibrium. In a cold catalyzed (nonaccreting) crust — the composition minimizing the total energy including nuclear masses, reached by weak (electron-capture, beta-decay) and strong nuclear processes — the equilibrium nuclides at each pressure have Z∼30Z\sim30–4040, so stoichiometric mixtures like FeHe are not in nuclear equilibrium (Jog and Smith 1982), although CsCl-type compounds can be favored at interfaces between adjacent pure layers (Jog and Smith 1982; Chamel and Fantina 2016, who found such compounds in very narrow density ranges between neighboring nuclides with charge ratios q≈0.93q\approx0.93–1.51.5). In accreted crusts, where X-ray-burst ashes are buried faster than nuclear equilibrium can be re-established, multicomponent solids with impurities are the norm (Horowitz and Berry 2009), and the reaction kinetics of the light species remains a live question (Yakovlev et al. 2006; Jain et al. 2023).

Problem Statement

Determine, with a controlled error budget at the level of 10−310^{-3} of the mixture Coulomb energy CmC_m (the size of the classical compound binding), the ground-state stability of ordered binary Coulomb compounds in dense degenerate matter — ions of charges ZeZe and YeYe in a degenerate electron gas, as in neutron-star crusts and white-dwarf cores — and the survival of their light nuclei against pycnonuclear fusion. Concretely:

(A) Quantum-corrected stability map. For the candidate ordered binary lattice structures (the classical set — NaCl-type, CsCl-type, and ZY3ZY_3-type — and the other ordered binary lattices tabulated in the structural surveys of Kozhberov (2020, 2024) and Engstrom, Yoder, and Crespi (2016); any further lattice the determination finds to be competitive may be reported in addition, but the required structure coverage for part (A) is that tabulated set), compute the sign and magnitude of the net binding of the compound relative to the phase-separated mixture,

B=(Cc−Cm)+(Sc−Sm)+(Pc−Pm)+⋯ ,B = (C_c - C_m) + (S_c - S_m) + (P_c - P_m) + \cdots,

where CC is the classical Madelung energy, SS the electron-screening correction, and P=12∑k,lℏωl(k)P=\frac{1}{2}\sum_{k,l}\hbar\omega_l(k) the phonon zero-point energy, the ellipsis denoting any further terms required for the stated accuracy (e.g., finite ion charge radius, exchange). The determination must (i) verify the dynamical (phonon) stability of every lattice claimed bound, excluding regions with soft modes; (ii) include the electron response beyond first-order Thomas-Fermi screening — the second-order term or the full RPA/Lindhard response — with an explicit bound on the neglected remainder; and (iii) delimit, in the plane of charge ratio x=Y/Zx=Y/Z and mass density ρ\rho, the regions in which ordered binary compounds are the ground state (the zero-point corrections introduce the density dependence through the ρ1/6\rho^{1/6} scaling).

(B) Pycnonuclear kinetics in the compound lattice. Compute the pycnonuclear (zero-temperature, tunnelling-driven) fusion rate of the light species in an ordered binary Coulomb compound lattice — treating the multidimensional tunnelling of a like pair in the actual lattice environment (equilibrium separations, phonon frequencies, screened potential, and the lattice and electron response to the approaching pair) — with accuracy sufficient to decide, for the compounds found stable in (A) at neutron-star-crust and white-dwarf-core conditions, whether the light nuclei are preserved for astrophysically long times (e.g., ∼1010\sim10^{10} yr) or burned on observable timescales.

The determination comprises (A) and (B): for each candidate structure and composition, whether the compound is bound and dynamically stable once quantum corrections are included, and whether it preserves its light nuclei. The two parts are linked: the same lattice dynamics that fixes the zero-point energy fixes the tunnelling attempt frequencies, and the astrophysical relevance of any stable compound depends on its reaction kinetics.

References

  1. F. J. Dyson, 'Chemical binding in classical Coulomb lattices', Annals of Physics 63, 1-11 (1971). DOI: 10.1016/0003-4916(71)90294-6, https://doi.org/10.1016/0003-4916(71)90294-6
  2. T. A. Witten, Jr., 'Compounds in Neutron-Star Crusts', The Astrophysical Journal 188, 615-625 (1974). DOI: 10.1086/152755, https://doi.org/10.1086/152755
  3. C. J. Jog and R. A. Smith, 'Mixed lattice phases in cold dense matter', The Astrophysical Journal 253, 839 (1982). DOI: 10.1086/159685, https://doi.org/10.1086/159685
  4. T. Igarashi, N. Nakao, and H. Iyetomi, 'Chemical Stability of Binary Coulomb Compounds', Contributions to Plasma Physics 41, 319-322 (2001). DOI: 10.1002/1521-3986(200107)41:4<319::AID-CTPP319>3.0.CO;2-J, https://doi.org/10.1002/1521-3986(200107)41:4<319::AID-CTPP319>3.0.CO;2-J
  5. Takahiro Igarashi and Hiroshi Iyetomi, 'Phase characteristics and elastic properties of binary Coulomb compounds', Journal of Physics A: Mathematical and General 36, 6197-6206 (2003). DOI: 10.1088/0305-4470/36/22/348, https://doi.org/10.1088/0305-4470/36/22/348
  6. A. A. Kozhberov and D. A. Baiko, 'Physical Features of Binary Coulomb Crystals: Madelung Energy, Collective Modes and Phonon Heat Capacity', Contributions to Plasma Physics 52, 153-156 (2012). DOI: 10.1002/ctpp.201100091, https://doi.org/10.1002/ctpp.201100091
  7. A. A. Kozhberov and D. A. Baiko, 'Coulomb crystal mixtures in white dwarf cores and neutron star crusts', Physics of Plasmas 22, 092903 (2015). DOI: 10.1063/1.4930215, https://doi.org/10.1063/1.4930215
  8. D. G. Yakovlev, L. R. Gasques, A. V. Afanasjev, M. Beard, and M. Wiescher, 'Fusion reactions in multicomponent dense matter', Physical Review C 74, 035803 (2006). DOI: 10.1103/PhysRevC.74.035803, arXiv:astro-ph/0608488, https://doi.org/10.1103/PhysRevC.74.035803
  9. C. J. Horowitz and D. K. Berry, 'Structure of accreted neutron star crust', Physical Review C 79, 065803 (2009). DOI: 10.1103/PhysRevC.79.065803, arXiv:0904.4076, https://doi.org/10.1103/PhysRevC.79.065803
  10. A. I. Chugunov and H. E. DeWitt, 'Nuclear fusion reaction rates for strongly coupled ionic mixtures', Physical Review C 80, 014611 (2009). DOI: 10.1103/PhysRevC.80.014611, arXiv:0905.3844, https://doi.org/10.1103/PhysRevC.80.014611
  11. T. A. Engstrom, N. C. Yoder, and V. H. Crespi, 'Crystal chemistry of three-component white dwarfs and neutron star crusts: phase stability, phase stratification, and physical properties', The Astrophysical Journal 818, 183 (2016). DOI: 10.3847/0004-637X/818/2/183, arXiv:1508.05088, https://doi.org/10.3847/0004-637X/818/2/183
  12. N. Chamel and A. F. Fantina, 'Binary and ternary ionic compounds in the outer crust of a cold nonaccreting neutron star', Physical Review C 94, 065802 (2016). DOI: 10.1103/PhysRevC.94.065802, arXiv:2103.06073, https://doi.org/10.1103/PhysRevC.94.065802.
  13. A. A. Kozhberov, 'Electrostatic properties and stability of Coulomb crystals', Contributions to Plasma Physics 60, e202000021 (2020). DOI: 10.1002/ctpp.202000021, https://doi.org/10.1002/ctpp.202000021
  14. N. Chamel, 'Binary and ternary ionic compounds in the outer crust of accreted neutron stars', Journal of Physics: Conference Series 932, 012039 (2017). DOI: 10.1088/1742-6596/932/1/012039, https://doi.org/10.1088/1742-6596/932/1/012039
  15. A. A. Kozhberov and A. Y. Potekhin, 'Electrostatic energy of Coulomb crystals with polarized electron background', Physical Review E 103, 043205 (2021). DOI: 10.1103/PhysRevE.103.043205, arXiv:2104.09964, https://doi.org/10.1103/PhysRevE.103.043205
  16. D. A. Baiko, 'Phase diagrams of binary ionic mixtures and white dwarf cooling', Monthly Notices of the Royal Astronomical Society 517, 3962-3973 (2022). DOI: 10.1093/mnras/stac2693, arXiv:2209.11563, https://doi.org/10.1093/mnras/stac2693
  17. D. Barba-González, C. Albertus, and M. A. Pérez-García, 'Crystallization in single- and multicomponent neutron star crusts', Physical Review C 106, 065806 (2022). DOI: 10.1103/PhysRevC.106.065806, arXiv:2207.14323, https://doi.org/10.1103/PhysRevC.106.065806
  18. N. N. Shchechilin, M. E. Gusakov, and A. I. Chugunov, 'Accreting neutron stars: composition of the upper layers of the inner crust', Monthly Notices of the Royal Astronomical Society (2023), article stad1731. DOI: 10.1093/mnras/stad1731, arXiv:2303.10003, https://doi.org/10.1093/mnras/stad1731
  19. R. Jain, E. F. Brown, H. Schatz, A. V. Afanasjev, M. Beard, L. R. Gasques, S. S. Gupta, G. W. Hitt, W. R. Hix, R. Lau, P. Möller, W. J. Ong, M. Wiescher, and Y. Xu, 'Impact of Pycnonuclear Fusion Uncertainties on the Cooling of Accreting Neutron Star Crusts', The Astrophysical Journal 955, 51 (2023). DOI: 10.3847/1538-4357/acebc4, arXiv:2306.12451, https://doi.org/10.3847/1538-4357/acebc4
  20. A. A. Kozhberov, 'Electrostatic energy of solid binary ionic mixtures', Physical Review E 110, 045206 (2024). DOI: 10.1103/PhysRevE.110.045206, arXiv:2411.03863, https://doi.org/10.1103/PhysRevE.110.045206
  21. Mingyao Xu, Aaron A. Smith, Leonid Prokhorov, Vera Guarrera, and Giovanni Barontini, 'Measuring impurity-induced shifts in Coulomb crystallization' (2026). arXiv:2603.14980, https://arxiv.org/abs/2603.14980