Heavy-element abundances in planetary nebulae (PNe) derived from faint optical recombination lines (ORLs) systematically exceed those derived from bright collisionally excited lines (CELs) by factors of about 2–3 on average — quantified by the abundance discrepancy factor (ADF) — and reach more than an order of magnitude in extreme objects, accompanied by electron temperatures from recombination diagnostics far below the CEL values of about 10,000 K. The standard interpretation posits cold, dense, metal-rich, hydrogen-deficient inclusions embedded in hot, chemically normal gas, but Wang & Liu (2007), in a deep spectroscopic survey of Galactic bulge PNe, stated that the nature and origin of this high-metallicity gas remain elusive. The subsequent literature establishes the link between extreme ADFs and close-binary central stars with an orbital-period threshold near 1.15 days (Corradi et al. 2015; Wesson et al. 2018), reports spatially resolved and kinematic evidence for a distinct centrally concentrated ORL-emitting phase (Gómez-Llanos et al. 2024; Huang et al. 2025; Henney et al. 2025), and offers alternative frameworks based on temperature inhomogeneities or departures from Case-B recombination (Méndez-Delgado & García-Rojas 2023; Méndez-Delgado et al. 2024). The physical mechanism that produces the hydrogen-deficient plasma, and hence which abundance scale is actually correct, remains unidentified — a 2026 review still lists the elemental abundance discrepancy among the unsolved problems of PN research — so the problem is open.

Background

A planetary nebula (PN) is the ionized envelope expelled by a low- or intermediate-mass star at the end of its life, shining because ultraviolet radiation from the exposed hot stellar core ionizes the expanding gas. Elemental abundances measured in PNe are a primary tracer of the chemical evolution of the Galaxy, including its bulge. Two families of emission lines give abundance results that disagree. Collisionally excited lines (CELs) are forbidden transitions, such as the [O III] lines at 4959 and 5007 Å, excited by electron impacts; they are bright but strongly temperature-sensitive, so the abundance they yield depends on an electron temperature estimated from other line ratios, classically [O III] 4363/5007. Optical recombination lines (ORLs), such as the O II permitted lines near 4650 Å, arise when free electrons recombine with ions; they are typically hundreds of times fainter but nearly temperature-independent. Since the 1940s it has been known that heavy-element abundances from ORLs exceed those from CELs by factors of about 2–3 on average; the ratio of the two determinations is called the abundance discrepancy factor (ADF), reaching values above 10 (called extreme) and, in the most extreme known cases, approaching three orders of magnitude. In parallel, temperatures from recombination diagnostics — O II ORL ratios, the He I recombination spectrum, and the Balmer jump (the discontinuity at the edge of the hydrogen recombination continuum near 3646 Å) — come out systematically below CEL temperatures, in extreme objects dropping to only a few hundred kelvin.

Liu et al. (2000), from a deep spectroscopic study of NGC 6153, proposed the interpretation that now frames the field: the nebula contains two plasma phases — the hot (about 10,000 K), chemically normal, diffuse gas traced by CELs, and cold, dense, metal-rich, hydrogen-deficient inclusions that emit almost all the heavy-element ORLs while contributing essentially no CELs, because at their low temperature collisional excitation is suppressed. Wang & Liu (2007) extended the same ORL/CEL analysis to 25 Galactic bulge PNe and found the same phenomena, adding key population-level constraints: the ADF is comparable for different elements within one nebula and yields CEL- and ORL-compatible C/O ratios, while magnesium — the only third-row element measured via an ORL (Mg II λ4481) — shows no enhancement regardless of ADF, and objects with large ADFs show large temperature discrepancies, suggesting one underlying process. They concluded that all evidence points to a cold, metal-rich, hydrogen-deficient component embedded in the diffuse nebula, while stating that the nature and origin of the high-metallicity gas remain elusive.

Competing or complementary explanations exist. Temperature fluctuations (parameterized by the t2t^2 mean-square temperature fluctuation) attribute the discrepancy to the CEL analysis being biased by a hot minority of the gas, and implicitly assume the higher ORL abundances are the correct ones — a position Wang & Liu (2007) argue against, since CEL abundances cluster near slightly sub-solar values while ORL abundances reach more than ten times solar. Other proposed contributors include electron energy distributions deviating from Maxwellian, and departures from Case-B recombination (the standard assumption that recombinations directly to the ground state are locally reabsorbed, so they can be excluded from the emitted spectrum). The literature also ties the extreme end of the ADF distribution to binary evolution: Wesson et al. (2018) proposed that a nova-like, common-envelope-triggered enriched eruption — with fallback of material after common-envelope ejection, a binary-evolution phase in which one star engulfs its companion and the shared envelope is expelled at the cost of orbital energy — expels hydrogen-poor, metal-enriched material into the inner nebula, and argued on statistical grounds against a very late thermal pulse (a final helium-shell flash after the star has begun descending the white-dwarf cooling track) as the origin of the extreme-ADF objects, while noting that the eruptive scenario itself remains ill-defined. Which line family, if either, yields the true bulk heavy-element abundances of PNe remains unsettled, with direct consequences for studies of Galactic chemical evolution.

Problem Statement

Wang & Liu (2007) concluded their survey of Galactic bulge planetary nebulae by stating that, although the diagnostics all point to a cold, metal-rich, hydrogen-deficient plasma component embedded in the hot, normal-composition gas — on the two-phase interpretation, the component that emits the heavy-element optical recombination lines and drives the correlated ORL–CEL abundance and temperature discrepancies — the nature and origin of this high-metallicity gas remain elusive. The problem is to determine, physically, what this plasma is and where it comes from, or whether the two-phase interpretation fails altogether: identifying the specific stellar-evolutionary or nebular process responsible for the correlated ORL–CEL abundance and temperature discrepancies, where a resolution may either explain how the cold, hydrogen-deficient, metal-rich material is produced or show that no distinct hydrogen-deficient component is required and the discrepancies arise entirely within a single-phase plasma; predicting the composition, mass, and physical conditions of the emitting material — the distinct cold component, if one exists, or the single-phase gas otherwise — well enough to reproduce the observed diagnostics, including the ADF distribution of the PN population and its extreme tail, the systematically low recombination-line and Balmer-jump temperatures, the central concentration of ORL emission relative to CEL emission, the unenhanced magnesium abundance and CEL-compatible C/O ratios, and the statistical association between extreme ADFs and short-period binary central stars; and thereby settling which line family, if either, yields the true bulk heavy-element abundances of planetary nebulae. The question is posed for planetary nebulae generally, not only the bulge population, and is unresolved as of 2026.

References

  1. Wang, W., & Liu, X.-W. (2007). Elemental abundances of Galactic bulge planetary nebulae from optical recombination lines. Monthly Notices of the Royal Astronomical Society, 381, 669–701. DOI: 10.1111/j.1365-2966.2007.12198.x. https://doi.org/10.1111/j.1365-2966.2007.12198.x
  2. Liu, X.-W., Storey, P. J., Barlow, M. J., Danziger, I. J., Cohen, M., & Bryce, M. (2000). NGC 6153: a super-metal-rich planetary nebula? Monthly Notices of the Royal Astronomical Society, 312, 585–628. DOI: 10.1046/j.1365-8711.2000.03167.x. https://doi.org/10.1046/j.1365-8711.2000.03167.x
  3. Corradi, R. L. M., García-Rojas, J., Jones, D., & Rodríguez-Gil, P. (2015). Binarity and the abundance discrepancy problem in planetary nebulae. The Astrophysical Journal, 803(2), 99. DOI: 10.1088/0004-637X/803/2/99. https://doi.org/10.1088/0004-637X/803/2/99
  4. Wesson, R., Jones, D., García-Rojas, J., Boffin, H. M. J., & Corradi, R. L. M. (2018). Confirmation of the link between central star binarity and extreme abundance discrepancy factors in planetary nebulae. Monthly Notices of the Royal Astronomical Society, 480, 4589–4613. DOI: 10.1093/mnras/sty1871. https://doi.org/10.1093/mnras/sty1871
  5. Wesson, R., Jones, D., García-Rojas, J., Boffin, H. M. J., & Corradi, R. L. M. (2018). Close binaries and the abundance discrepancy problem in planetary nebulae. Galaxies, 6(4), 110. DOI: 10.3390/galaxies6040110. https://doi.org/10.3390/galaxies6040110
  6. Gómez-Llanos, V., & Morisset, C. (2020). Bi-abundance photoionization models of planetary nebulae: determining the amount of oxygen in the metal-rich component. Monthly Notices of the Royal Astronomical Society, 497, 3363–3380. DOI: 10.1093/mnras/staa2157. https://doi.org/10.1093/mnras/staa2157
  7. Espíritu, J. N., & Peimbert, A. (2021). Physical conditions and chemical abundances in PN M 2-36. Results from deep echelle observations. Monthly Notices of the Royal Astronomical Society, 508, 2668–2687. DOI: 10.1093/mnras/stab2746. https://doi.org/10.1093/mnras/stab2746
  8. Gómez-Llanos, V., García-Rojas, J., Morisset, C., Monteiro, H., Jones, D., Wesson, R., Boffin, H. M. J., & Corradi, R. L. M. (2024). MUSE spectroscopy of the high abundance discrepancy planetary nebula NGC 6153. Astronomy & Astrophysics, 689, A228. DOI: 10.1051/0004-6361/202450822. https://doi.org/10.1051/0004-6361/202450822
  9. Méndez-Delgado, J. E., & García-Rojas, J. (2023). The abundance discrepancy in ionized nebulae: which are the correct abundances? arXiv:2311.10280. https://arxiv.org/abs/2311.10280
  10. Méndez-Delgado, J. E., Skillman, E. D., Aver, E., Morisset, C., Esteban, C., García-Rojas, J., Kreckel, K., Rogers, N. S. J., Rosales-Ortega, F. F., Arellano-Córdova, K. Z., Flury, S. R., Reyes-Rodríguez, E., Orte-García, M., & Tan, S. (2024). Generalized TeT_e([OIII])-TeT_e(HeI) discrepancies in ionized nebulae: possible evidence of Case B deviations and temperature inhomogeneities. arXiv:2410.17381. https://arxiv.org/abs/2410.17381
  11. Henney, W. J., López, J. A., García-Díaz, M. T., Richer, M. G., González-Buitrago, D. H., Zharikov, S. V., Noroozi, S., & Pozo Nuñez, F. (2025). Kinematics and physical structure of Ou 5: A bipolar planetary nebula with extreme abundance discrepancy and an eclipsing binary core. arXiv:2512.09012. https://arxiv.org/abs/2512.09012
  12. Huang, H., Fang, X., García-Rojas, J., Tu, Z., Liu, J., & Liu, X. (2025). A morpho-kinematic study of Galactic high-ADF PNe based on the VLT/UVES deep spectroscopy. arXiv:2512.04533. https://arxiv.org/abs/2512.04533
  13. Kwok, S., Balick, B., Chu, Y.-H., Hrivnak, B. J., López, A., Parker, Q., Sahai, R., & Zijlstra, A. (2026). Current unsolved problems in planetary nebulae research. Galaxies, 14(2), 30. DOI: 10.3390/galaxies14020030. https://doi.org/10.3390/galaxies14020030