Unambiguous experimental determination of the ionic static structure factor from elastic x-ray Thomson scattering features in warm dense matter

In x-ray Thomson scattering (XRTS) on warm dense matter, the narrow quasi-elastic feature is the standard window onto ionic structure: in the cold, condensed-matter limit its weight factorizes as $W(\mathbf{q})=|N(\mathbf{q})|^{2}S_{ii}(\mathbf{q})$, with $N(\mathbf{q})$ the electronic form factor localized around each ion and $S_{ii}(\mathbf{q})$ the ionic static structure factor. Mo et al. (Phys. Rev. B 102, 195127 (2020)) showed from a first-principles Kramers-Heisenberg treatment that this factorization fails in the warm dense regime: partially occupied inner-shell states reduce and delocalize $N(\mathbf{q})$, and a new elastic contribution $S_{1}^{b}(\mathbf{q})$ from those states adds to the feature, so that $N(\mathbf{q})$ and $S_{ii}(\mathbf{q})$ become correlated unknowns that cannot be independently determined from the measured elastic scattering alone. The open problem is to achieve an unambiguous determination of ionic structure ($S_{ii}(\mathbf{q})$) from XRTS spectra in warm dense matter, via additional constraints or independent measurements. Later work has made the composite elastic weight (the Rayleigh weight $W_{R}(\mathbf{q})=S_{eI}^{2}(\mathbf{q})/S_{II}(\mathbf{q})$) itself a model-free experimental observable and improved forward models of $S_{ii}$, but no published method separates the ionic correlations from the electronic localization contribution using experimental data alone, and re-analyses of the same National Ignition Facility beryllium data with different model assumptions disagree on the inferred mass density by a factor of about 1.5 (about 34 versus about 22 g/cc).

Background

Warm dense matter (WDM) designates states at near-solid density and temperatures of roughly 1–100 eV, occurring in planetary interiors and on the compression path of inertial-confinement-fusion capsules; it is intermediate between a condensed solid and an ideal plasma, so neither limit's theory applies cleanly. X-ray Thomson scattering (XRTS) is a primary diagnostic for WDM: a keV x-ray probe scatters from the electrons, and the measured spectrum is (after convolution with the source and instrument function) proportional to the electronic dynamic structure factor $S(\mathbf{q},\omega)$, the Fourier transform of the electron density–density correlation function at momentum transfer $\mathbf{q}$ and energy transfer $\omega$. Spectra conventionally split into a narrow quasi-elastic peak centered on the probe energy (called the elastic, Rayleigh, or ion feature) and broad inelastic features (plasmons — collective oscillations of the electron density — and bound–free transitions).

The standard analysis framework is the Chihara decomposition (Chihara 1987; Gregori et al. 2003; Glenzer and Redmer 2009), which partitions electrons into bound and free parts. In this picture the elastic feature weight is written as $W(\mathbf{q})=|N(\mathbf{q})|^{2}S_{ii}(\mathbf{q})$, where $S_{ii}(\mathbf{q})$ is the ionic static structure factor — the Fourier transform of the ion–ion pair correlation function, the standard quantitative measure of ionic order — and $N(\mathbf{q})$ is an electronic form factor, the Fourier transform of the electron density assigned to each ion (bound electrons plus the screening cloud). Experiments are typically analyzed by fitting the elastic feature with $S_{ii}$ supplied by a model (hypernetted-chain theory, an integral-equation approximation for the ion pair-correlation function, or average-atom theory, a self-consistent mean-field model of a single ion embedded in its plasma environment) to infer temperature, density, and ionization; Starrett and Saumon (2015) documented substantial spread among such elastic-feature models for warm dense aluminum. Baczewski et al. (2016) showed that full XRTS spectra can instead be computed without the Chihara bound/free split, directly from time-dependent density functional theory — the extension of density functional theory (DFT, the standard framework in which the interacting electron system is replaced by effective single-particle Kohn-Sham orbitals) to time-dependent electron densities — evaluated on snapshots of finite-temperature DFT molecular dynamics, a forward-modeling route that still takes ionic configurations as simulation input rather than as an experimentally determined quantity.

Mo et al. (2020) derived the entire XRTS spectrum, elastic and inelastic parts together, from first principles via the Kramers-Heisenberg formula (the exact second-order quantum-mechanical expression for the photon-scattering cross section), evaluated with Kohn-Sham orbitals carrying Fermi-Dirac occupations on ionic configurations from finite-temperature DFT molecular dynamics, plus linear-response TDDFT for the inelastic part, avoiding any empirical bound/free division. Their derivation recovers $W(\mathbf{q})=|N(\mathbf{q})|^{2}S_{ii}(\mathbf{q})$ only in the limit where the electron distribution is localized around individual ions and an additional elastic term is negligible. In the warm dense regime that limit fails in two coupled ways. First, inner-shell states become partially occupied and excited states are spatially extended, so $N(\mathbf{q})$ itself becomes temperature-dependent and delocalized; for warm dense beryllium, Mo et al. found $N(q)$ strongly reduced at electron temperatures above roughly 25 eV. Second, partially occupied core states generate an extra elastic contribution $S_{1}^{b}(\mathbf{q})\,\delta(\omega)$ to the scattering spectrum,

$$S_{1}^{b}(\mathbf{q})=\Big\langle\sum_{n}f_{n}(1-f_{n})\,\big|\langle n|e^{-i\mathbf{q}\cdot\mathbf{r}}|n\rangle\big|^{2}\Big\rangle\,,$$

where $f_{n}$ is the Fermi-Dirac occupation of single-particle state $n$ and the brackets denote an average over ionic configurations; the $\delta(\omega)$ factor marks the contribution as concentrated at zero energy transfer, so that $S_{1}^{b}(\mathbf{q})$, like $W(\mathbf{q})$, is an $\omega$-integrated elastic weight. This term vanishes at low temperature but grows with $T$ and can be comparable to $W(\mathbf{q})$ at elevated temperatures. Consequently the measured elastic feature determines, at each $\mathbf{q}$, essentially one composite number, while $N(\mathbf{q})$, $S_{ii}(\mathbf{q})$, and $S_{1}^{b}(\mathbf{q})$ remain correlated unknown functions: the inverse problem of reading ionic structure off the elastic feature is underdetermined. Modern treatments make this non-separability explicit through the exact identity $W_{R}(\mathbf{q})=S_{eI}^{2}(\mathbf{q})/S_{II}(\mathbf{q})$ for the Rayleigh weight, involving the electron–ion static structure factor $S_{eI}$ (a measure of electronic localization around ions) and the ion–ion structure factor $S_{II}$ (Dornheim et al. 2025); multi-component effects add further entanglement, since charge-state mixtures require partial structure factors and species-dependent form factors (Jin et al. 2021).

Problem Statement

Give additional constraints or independent measurements under which the ionic static structure factor $S_{ii}(\mathbf{q})$ — and thereby the ionic structure — of warm dense matter is determined unambiguously from measured x-ray Thomson scattering spectra. The elastic XRTS feature is a composite quantity: in the localized-electron, low-temperature limit its weight factorizes as $W(\mathbf{q})=|N(\mathbf{q})|^{2}S_{ii}(\mathbf{q})$, but in the warm dense regime the electronic form factor $N(\mathbf{q})$ is itself delocalized and temperature-dependent (partially occupied core states, spatially extended excited states), an additional elastic contribution $S_{1}^{b}(\mathbf{q})$ from partially occupied inner-shell states enters and can be comparable to $W(\mathbf{q})$ at elevated temperatures, and $N(\mathbf{q})$ and $S_{ii}(\mathbf{q})$ become correlated quantities that cannot be independently determined from the measured elastic scattering alone. Resolving this requires information beyond the composite elastic-feature weight at each $\mathbf{q}$: additional constraints or independent measurements. The admissible experimental information is the measured XRTS spectrum at one or more momentum transfers — including both its elastic (Rayleigh/ion) feature and its inelastic features, which may be used, e.g., via sum rules or the elastic-to-inelastic ratio — optionally supplemented by independent measurements or auxiliary constraints on the same or a companion shot (e.g., a characterized source-and-instrument function, or independently known temperature or density). A complete answer must therefore supply, without assuming away the entanglement, additional constraints or independent measurements (combined probes, multi-angle or multi-q data, auxiliary diagnostics) under which $S_{ii}(\mathbf{q})$ is determined from experimental data with quantified uncertainty, including a controlled treatment of $N(\mathbf{q})$ and $S_{1}^{b}(\mathbf{q})$. Re-expressing the composite elastic weight alone (for example, extracting the Rayleigh weight $W_{R}(\mathbf{q})=S_{eI}^{2}(\mathbf{q})/S_{II}(\mathbf{q})$ as a model-free observable, or fitting it with simulation-supplied $S_{II}$) does not by itself constitute a solution, because the ionic correlations are then still fixed by a model rather than by the measurement.

References

1. Chongjie Mo, Zhen-Guo Fu, Ping Zhang, Wei Kang, Weiyan Zhang, and X. T. He, 'First-principles method for x-ray Thomson scattering including both elastic and inelastic features in warm dense matter', Physical Review B 102, 195127 (2020), DOI: 10.1103/PhysRevB.102.195127, https://doi.org/10.1103/PhysRevB.102.195127
2. J. Chihara, 'Difference in X-ray scattering between metallic and non-metallic liquids due to conduction electrons', Journal of Physics F: Metal Physics 17, 295–304 (1987), DOI: 10.1088/0305-4608/17/2/002, https://doi.org/10.1088/0305-4608/17/2/002
3. G. Gregori, S. H. Glenzer, W. Rozmus, R. W. Lee, and O. L. Landen, 'Theoretical model of x-ray scattering as a dense matter probe', Physical Review E 67, 026412 (2003), DOI: 10.1103/PhysRevE.67.026412, https://doi.org/10.1103/PhysRevE.67.026412
4. Siegfried H. Glenzer and Ronald Redmer, 'X-ray Thomson scattering in high energy density plasmas', Reviews of Modern Physics 81, 1625–1663 (2009), DOI: 10.1103/RevModPhys.81.1625, https://doi.org/10.1103/RevModPhys.81.1625
5. C. E. Starrett and D. Saumon, 'Models of the elastic x-ray scattering feature for warm dense aluminum', Physical Review E 92, 033101 (2015), DOI: 10.1103/PhysRevE.92.033101, https://doi.org/10.1103/PhysRevE.92.033101
6. A. D. Baczewski, L. Shulenburger, M. P. Desjarlais, S. B. Hansen, and R. J. Magyar, 'X-ray Thomson Scattering in Warm Dense Matter without the Chihara Decomposition', Physical Review Letters 116, 115004 (2016), DOI: 10.1103/PhysRevLett.116.115004, https://doi.org/10.1103/PhysRevLett.116.115004
7. Yang Jin, Ping Zhang, Yong-Jun Li, Yong Hou, Jiao-Long Zeng, and Jian-Min Yuan, 'Influence of different charge-state ion distribution on elastic X-ray scattering in warm dense matter', Acta Physica Sinica 70, 073102 (2021), DOI: 10.7498/aps.70.20201483, https://doi.org/10.7498/aps.70.20201483
8. Maximilian Schörner, Hannes R. Rüter, Martin French, and Ronald Redmer, 'Extending ab initio simulations for the ion-ion structure factor of warm dense aluminum to the hydrodynamic limit using neural network potentials', Physical Review B 105, 174310 (2022), DOI: 10.1103/PhysRevB.105.174310, https://doi.org/10.1103/PhysRevB.105.174310
9. Maximilian Schörner, Mandy Bethkenhagen, Tilo Döppner, Dominik Kraus, Luke B. Fletcher, Siegfried H. Glenzer, and Ronald Redmer, 'X-ray Thomson scattering spectra from density functional theory molecular dynamics simulations based on a modified Chihara formula', Physical Review E 107, 065207 (2023), DOI: 10.1103/PhysRevE.107.065207, https://doi.org/10.1103/PhysRevE.107.065207
10. T. Döppner, M. Bethkenhagen, D. Kraus, P. Neumayer, D. A. Chapman, B. Bachmann, R. A. Baggott, M. P. Böhme, L. Divol, R. W. Falcone, L. B. Fletcher, O. L. Landen, M. J. MacDonald, A. M. Saunders, M. Schörner, P. A. Sterne, J. Vorberger, B. B. L. Witte, A. Yi, R. Redmer, S. H. Glenzer, and D. O. Gericke, 'Observing the onset of pressure-driven K-shell delocalization', Nature 618, 270–275 (2023), DOI: 10.1038/s41586-023-05996-8, https://doi.org/10.1038/s41586-023-05996-8
11. Thomas Gawne, Zhandos A. Moldabekov, Oliver S. Humphries, Karen Appel, Carsten Bähtz, Victorien Bouffetier, Erik Brambrink, Attila Cangi, Sebastian Göde, Zuzana Konôpková, Mikako Makita, Mikhail Mishchenko, Motoaki Nakatsutsumi, Kushal Ramakrishna, Lisa Randolph, Sebastian Schwalbe, Jan Vorberger, Lennart Wollenweber, Ulf Zastrau, Tobias Dornheim, and Thomas R. Preston, 'Ultrahigh resolution x-ray Thomson scattering measurements at the European X-ray Free Electron Laser', Physical Review B 109, L241112 (2024), DOI: 10.1103/PhysRevB.109.L241112, https://doi.org/10.1103/PhysRevB.109.L241112
12. T. Dornheim, H. M. Bellenbaum, M. Bethkenhagen, S. B. Hansen, M. P. Böhme, T. Döppner, L. B. Fletcher, Th. Gawne, D. O. Gericke, S. Hamel, D. Kraus, M. J. MacDonald, Zh. A. Moldabekov, T. R. Preston, R. Redmer, M. Schörner, S. Schwalbe, P. Tolias, and J. Vorberger, 'Model-free Rayleigh weight from x-ray Thomson scattering measurements', Physics of Plasmas 32, 052712 (2025), DOI: 10.1063/5.0238630, https://doi.org/10.1063/5.0238630 (preprint arXiv:2409.08591, https://arxiv.org/abs/2409.08591)
13. T. Dornheim, 'Towards the Model-Free Interpretation of X-Ray Thomson Scattering Measurements of Warm Dense Matter', 2025 IEEE Pulsed Power & Plasma Science (PPPS), pp. 1–2 (2025), DOI: 10.1109/PPPS56198.2025.11248477, https://doi.org/10.1109/PPPS56198.2025.11248477
14. Sebastian Schwalbe, Hannah M. Bellenbaum, Tilo Döppner, Maximilian P. Böhme, Thomas Gawne, Dominik Kraus, Michael J. MacDonald, Zhandos A. Moldabekov, Panagiotis Tolias, Jan Vorberger, and Tobias Dornheim, 'Static linear density response from X-ray Thomson scattering measurements: A case study of warm dense beryllium', High Energy Density Physics 60, 101302 (2026), DOI: 10.1016/j.hedp.2026.101302, https://doi.org/10.1016/j.hedp.2026.101302 (preprint arXiv:2504.13611, https://arxiv.org/abs/2504.13611)