Completing the 22Ne phase-separation problem for crystallizing carbon–oxygen white dwarfs: the ternary C/O/Ne phase diagram and the missing cooling delay behind the Gaia crystallization pile-up

White dwarfs are the most common stellar endpoints and standard cosmochronometers, yet their cooling ages depend on imperfectly known crystallization microphysics. Blouin et al. (2020) obtained the most accurate carbon–oxygen (C/O) phase diagram to date but found that the crystallization pile-up observed by Gaia in the luminosity function of 0.9–1.1 solar-mass hydrogen-atmosphere white dwarfs is still underestimated, concluding that phase separation of the trace isotope 22Ne during crystallization is likely the missing physics but could not be quantified because the ternary C/O/Ne phase diagram was poorly known. In the years since, the core microphysics has been largely settled: a high-precision Monte Carlo C/O/Ne phase diagram at small neon abundance was computed (Blouin et al. 2021), the resulting 22Ne 'distillation' mechanism was implemented in a white dwarf evolution code and shown to halt cooling for billions of years, resolving the ultramassive Q-branch cooling anomaly (Bédard et al. 2024), and precise diffusion coefficients were published (Caplan & Freeman 2021). The surviving open core is two-fold: (i) the ternary C/O/Ne phase diagram is still incomplete — its high-precision data are restricted to small 22Ne concentrations at a single representative pressure, and the composition trajectory followed during distillation is explicitly unknown; and (ii) the quantitative demonstration for standard-composition white dwarfs — detailed cooling sequences coupling 22Ne settling in the liquid, C/O phase separation, and 22Ne distillation, with population synthesis reproducing the amplitude and onset of the Gaia crystallization pile-up — has not been published, and the predicted additional standard-composition delay (1.6–2.0 Gyr from the binding-energy estimates of Blouin et al. 2021, characterized as ~1 Gyr by Bédard et al. 2024) awaits confirmation in such sequences.

Background

A white dwarf is the degenerate remnant left when a low- or intermediate-mass star exhausts its nuclear fuel; it cools and fades at a nearly predictable rate, which makes white dwarf cooling curves a standard chronometer for dating the Galactic disk, star clusters, and exoplanet host stars. The core of a typical white dwarf is a dense plasma of fully ionized carbon and oxygen (a C/O white dwarf) with trace amounts of neutron-rich nuclei, most importantly 22Ne, produced from nitrogen during helium burning in the progenitor star (about 1–2% by mass). Cooling is punctuated when the core becomes so strongly coupled that it crystallizes into a solid from the inside out. The strength of Coulomb interactions is measured by the coupling parameter $\Gamma$, the ratio of the electrostatic potential energy between neighboring ions to the thermal energy; crystallization sets in around $\Gamma\sim175$–$230$. The phase transition releases latent heat, and because oxygen preferentially enters the solid phase, the ensuing compositional readjustment ('O sedimentation') releases additional gravitational energy, temporarily slowing the cooling.

The quantitative input for this physics is the phase diagram of the dense multicomponent plasma: the liquidus (the temperature-composition boundary above which the mixture is fully liquid) and the solidus (below which it is fully solid). The separation between the two curves at fixed temperature determines how strongly the composition changes upon freezing, and hence how much gravitational energy is released. A point where liquid and solid compositions coincide, so that no fractionation occurs, is called an azeotrope. Modern phase diagrams for white dwarf interiors are computed with Monte Carlo (MC) simulations of the liquid and solid coupled by Gibbs–Duhem/Clapeyron integration: one integrates the Clapeyron differential equation along the liquid–solid coexistence curve, using MC simulations in the semi-grand canonical ensemble (particles may swap species at fixed fugacity ratio) to evaluate the enthalpies and compositions of each phase. This method, adapted to plasmas by Blouin and collaborators, superseded older density-functional and sparse two-phase molecular dynamics calculations.

Because 22Ne has more neutrons per electron than C or O, it is heavier per unit charge and gravitationally settles in the liquid core, releasing energy; more importantly, if the solid formed during crystallization excludes 22Ne, the 22Ne-poor crystals are lighter than the surrounding liquid, float upward and melt, while 22Ne-rich liquid is displaced downward. This 'distillation' cycle can transport most of the star's 22Ne deep into the potential well and release a large amount of gravitational energy, potentially delaying cooling by gigayears. Whether, when, and how strongly this happens is dictated by the ternary C/O/Ne phase diagram at the small neon concentrations ($x_{\mathrm{Ne}}\sim0.01$, by number) relevant to real white dwarfs.

The Gaia astrometry mission transformed this field by delivering an unbiased color-absolute-magnitude diagram of hundreds of thousands of white dwarfs. Two discrepancies emerged. First, Tremblay et al. (2019) detected the crystallization pile-up — an excess of stars at the luminosities where the core freezes — in the luminosity function (the number of stars per luminosity interval) of 0.9–1.1 solar-mass hydrogen-atmosphere white dwarfs, but theoretical models underestimated the pile-up amplitude. Second, Cheng et al. (2019) showed that about 6% of ultramassive white dwarfs on the Q branch — an overdensity of high-mass white dwarfs in the Gaia color–magnitude diagram — require an extra cooling delay of roughly 8 Gyr not predicted by any standard model. Blouin, Daligault, Saumon, Bédard and Brassard (2020) computed their new high-accuracy C/O phase diagram, incorporated it into population simulations, and confirmed that latent heat plus O sedimentation still underpredict the crystallization pile-up; they concluded that 22Ne phase separation is probably the missing ingredient, but that the ternary C/O/Ne phase diagram was known only from the semi-analytic calculation of Segretain (1996) and a handful of two-phase molecular dynamics points (Hughto et al. 2012), far too coarsely to implement the effect in evolution models. Their Letter is the source of the present problem, which also inherits the long-running idea of Isern et al. (1991) that minor species separated during crystallization could matter for white dwarf cooling.

Problem Statement

The problem, in its surviving form after the 2021–2024 resolution of the ultramassive Q-branch strand, is to complete the quantitative account of 22Ne phase separation in crystallizing carbon–oxygen white dwarfs at the precision required for Gaia-era cosmochronology. Its overarching objective is to close, microphysically, the 22Ne contribution to the cooling of standard-composition C/O white dwarfs; two linked results serve that objective, and a resolution requires both. First, complete the ternary C/O/Ne phase diagram under white dwarf interior conditions beyond the domain of the existing high-precision data: the published Clapeyron-integration Monte Carlo results cover 11 temperatures at a single representative pressure $P=10^{24}\,\mathrm{dyn\,cm^{-2}}$, at high resolution only for small 22Ne number fractions $x_{\mathrm{Ne}}\le0.05$ (the analytic fit implemented in evolution codes is valid only for $x_{\mathrm{Ne}}<0.04$). What is still required are the liquidus/solidus surfaces, liquid–solid 22Ne partition coefficients, azeotropic and density-crossover (distillation) boundaries, and — most importantly — the composition trajectory followed by the crystal-forming region during distillation, from its initial state at the onset of distillation (the local liquid composition, at small neon fraction) to the liquidus–solidus meeting (azeotropic) composition of the C/Ne limit as determined by the completed diagram. Current estimates place that endpoint near $(x_{\mathrm{C}},x_{\mathrm{O}},x_{\mathrm{Ne}})\approx(0.8,0.0,0.2)$, by number, at $\Gamma_{\mathrm{C}}\approx208$: a value inherited from the earlier coarse C/Ne binary calculations (Segretain 1996 obtains $(0.78,0.0,0.22)$ at $T_a=0.85\,T_C$; Hughto et al. 2012) and adopted as a working assumption by Blouin et al. (2021) and Bédard et al. (2024), while the trajectory between onset and endpoint is unknown and current evolution codes must guess it (a linear prescription is assumed). The exact azeotropic endpoint and any density or pressure dependence are therefore part of what the completed diagram must determine, not a boundary condition imposed in advance. Second, produce detailed evolutionary cooling sequences for standard-composition C/O white dwarfs (canonical mass fraction $X(^{22}\mathrm{Ne})\approx0.014$; higher abundances and radial profiles arise from progenitor metallicity and merger histories) that self-consistently couple 22Ne gravitational settling in the liquid, C/O phase separation, and 22Ne distillation, and determine through population synthesis whether they quantitatively reproduce the amplitude and onset of the Gaia crystallization pile-up in the 0.9–1.1 solar-mass hydrogen-atmosphere luminosity function — the discrepancy that motivated the source Letter — together with the predicted additional standard-composition delay (1.6–2.0 Gyr from the binding-energy estimates of Blouin et al. 2021, characterized as ~1 Gyr by Bédard et al. 2024; the delay inferred to be missing from current C/O cooling tracks is ~0.6–1 Gyr, arising once roughly 60% of the core has crystallized). The same coupled modeling may instead show that 22Ne distillation cannot supply the missing delay. Observational differential-age measurements in enlarged samples of coeval wide binaries containing a crystallized member (e.g., Barrientos et al. 2024) would corroborate the second result, but the samples needed for a statistically significant detection are not currently available.

References

1. Blouin, S., Daligault, J., Saumon, D., Bédard, A., & Brassard, P. (2020). Toward Precision Cosmochronology: A New C/O Phase Diagram for White Dwarfs. Astronomy & Astrophysics 640, L11. DOI: 10.1051/0004-6361/202038879. https://doi.org/10.1051/0004-6361/202038879 (arXiv:2007.13669)
2. Blouin, S., Daligault, J., & Saumon, D. (2021). 22Ne Phase Separation as a Solution to the Ultramassive White Dwarf Cooling Anomaly. The Astrophysical Journal Letters 911, L5. DOI: 10.3847/2041-8213/abf14b. https://doi.org/10.3847/2041-8213/abf14b (arXiv:2103.12892)
3. Blouin, S., & Daligault, J. (2021). Direct evaluation of the phase diagrams of dense multicomponent plasmas by integration of the Clapeyron equations. Physical Review E 103, 043204. DOI: 10.1103/PhysRevE.103.043204. https://doi.org/10.1103/PhysRevE.103.043204
4. Bédard, A., Blouin, S., & Cheng, S. (2024). Buoyant crystals halt the cooling of white dwarf stars. Nature 627, 286–288. DOI: 10.1038/s41586-024-07102-y. https://doi.org/10.1038/s41586-024-07102-y (arXiv:2409.04419)
5. Shen, K. J., Blouin, S., & Breivik, K. (2023). The Q Branch Cooling Anomaly Can Be Explained by Mergers of White Dwarfs and Subgiant Stars. The Astrophysical Journal Letters 955, L33. DOI: 10.3847/2041-8213/acf57b. https://doi.org/10.3847/2041-8213/acf57b (arXiv:2308.04559)
6. Caplan, M. E., & Freeman, I. (2021). Precise diffusion coefficients for white dwarf astrophysics. Monthly Notices of the Royal Astronomical Society 505, 45–49. DOI: 10.1093/mnras/stab1259. https://doi.org/10.1093/mnras/stab1259 (arXiv:2105.05863)
7. Saumon, D., Blouin, S., & Tremblay, P.-E. (2022). Current challenges in the physics of white dwarf stars. Physics Reports 988, 1–63. DOI: 10.1016/j.physrep.2022.09.001. https://doi.org/10.1016/j.physrep.2022.09.001 (arXiv:2209.02846)
8. Tremblay, P.-E., Bédard, A., O'Brien, M. W., Munday, J., Elms, A. K., Gentillo Fusillo, N. P., & Sahu, S. (2024). The Gaia white dwarf revolution. New Astronomy Reviews 99, 101705. DOI: 10.1016/j.newar.2024.101705. https://doi.org/10.1016/j.newar.2024.101705 (arXiv:2402.14960)
9. Barrientos, M., Kilic, M., Bergeron, P., Blouin, S., Brown, W. R., & Andrews, J. J. (2024). Fundamental Tests of White Dwarf Cooling Physics with Wide Binaries. The Astrophysical Journal 973, 88. DOI: 10.3847/1538-4357/ad6303. https://doi.org/10.3847/1538-4357/ad6303
10. Castro-Tapia, M., & Cumming, A. (2025). Three-component Phase Separation for Ultramassive White Dwarf Models. The Astrophysical Journal 991, 64. DOI: 10.3847/1538-4357/adf745. https://doi.org/10.3847/1538-4357/adf745
11. Bauer, E. B. (2023). Carbon–Oxygen Phase Separation in Modules for Experiments in Stellar Astrophysics (MESA) White Dwarf Models. The Astrophysical Journal 950, 115. DOI: 10.3847/1538-4357/acd057. https://doi.org/10.3847/1538-4357/acd057
12. Camisassa, M. E., Baiko, D. A., Torres, S., & Rebassa-Mansergas, A. (2024). Exsolution process in white dwarf stars. Astronomy & Astrophysics 683, A101. DOI: 10.1051/0004-6361/202348344. https://doi.org/10.1051/0004-6361/202348344
13. Cheng, S., Cummings, J. D., & Ménard, B. (2019). A Cooling Anomaly of High-mass White Dwarfs. The Astrophysical Journal 886, 100. DOI: 10.3847/1538-4357/ab4989. https://doi.org/10.3847/1538-4357/ab4989
14. Tremblay, P.-E., Fontaine, G., Gentile Fusillo, N. P. G., Dunlap, B. H., Gänsicke, B. T., Hollands, M. A., Hermes, J. J., Marsh, T. R., Cukanovaite, E., & Cunningham, T. (2019). Core crystallization and pile-up in the cooling sequence of evolving white dwarfs. Nature 565, 202–205. DOI: 10.1038/s41586-018-0791-x. https://doi.org/10.1038/s41586-018-0791-x
15. Segretain, L. (1996). Three-body crystallization diagrams and the cooling of white dwarfs. Astronomy & Astrophysics 310, 485. arXiv:astro-ph/9510118. https://arxiv.org/abs/astro-ph/9510118
16. Hughto, J., Horowitz, C. J., Schneider, A. S., Medin, Z., Cumming, A., & Berry, D. K. (2012). Direct molecular dynamics simulation of liquid-solid phase equilibria for a three-component plasma. Physical Review E 86, 066413. DOI: 10.1103/PhysRevE.86.066413. https://doi.org/10.1103/PhysRevE.86.066413
17. Isern, J., Hernanz, M., Mochkovitch, R., & García-Berro, E. (1991). The role of the minor chemical species in the cooling of white dwarfs. Astronomy & Astrophysics 241, L29–L32. ADS bibcode 1991A&A...241L..29I. https://ui.adsabs.harvard.edu/abs/1991A&A...241L..29I
18. Caplan, M. E., Horowitz, C. J., & Cumming, A. (2020). Neon Cluster Formation and Phase Separation during White Dwarf Cooling. The Astrophysical Journal Letters 902, L44. DOI: 10.3847/2041-8213/abbda0. https://doi.org/10.3847/2041-8213/abbda0
19. Bauer, E. B., Schwab, J., Bildsten, L., & Cheng, S. (2020). Multi-gigayear White Dwarf Cooling Delays from Clustering-enhanced Gravitational Sedimentation. The Astrophysical Journal 902, 93. DOI: 10.3847/1538-4357/abb5a5. https://doi.org/10.3847/1538-4357/abb5a5
20. Rui, N. Z., & Fuller, J. (2026). Forging Neon-distilling White Dwarfs in the Stellar Engulfments of Helium White Dwarfs. The Astrophysical Journal 998, 64. DOI: 10.3847/1538-4357/ae31e8. https://doi.org/10.3847/1538-4357/ae31e8
21. Pathak, P., Blouin, S., & Herwig, F. (2025). Quantifying Systematic Uncertainties in White Dwarf Cooling Age Determinations. The Astrophysical Journal 983, 158. DOI: 10.3847/1538-4357/adc10f. https://doi.org/10.3847/1538-4357/adc10f
22. Xu, M., Smith, A. A., Prokhorov, L., Guarrera, V., & Barontini, G. (2026). Measuring impurity-induced shifts in Coulomb crystallization. arXiv:2603.14980. https://arxiv.org/abs/2603.14980