The chemical composition of the Sun

The solar chemical composition is an important ingredient in our understanding of the formation, structure and evolution of both the Sun and our solar system. Furthermore, it is an essential reference standard against which the elemental contents of other astronomical objects are compared. In this review we evaluate the current understanding of the solar photospheric composition. In particular, we present a re-determination of the abundances of nearly all available elements, using a realistic new 3-dimensional (3D), time-dependent hydrodynamical model of the solar atmosphere. We have carefully considered the atomic input data and selection of spectral lines, and accounted for departures from LTE whenever possible. The end result is a comprehensive and homogeneous compilation of the solar elemental abundances. Particularly noteworthy findings are significantly lower abundances of carbon, nitrogen, oxygen and neon compared with the widely-used values of a decade ago. The new solar chemical composition is supported by a high degree of internal consistency between available abundance indicators, and by agreement with values obtained in the solar neighborhood and from the most pristine meteorites. There is, however, a stark conflict with standard models of the solar interior according to helioseismology, a discrepancy that has yet to find a satisfactory resolution.

💡 Research Summary

The paper presents a comprehensive reassessment of the solar photospheric chemical composition using state‑of‑the‑art three‑dimensional (3D), time‑dependent hydrodynamical simulations of the solar atmosphere. The authors begin by emphasizing that the Sun’s elemental abundances serve as a fundamental reference for a wide range of astrophysical topics, from models of solar structure and evolution to the chemical benchmarking of other stars, nebulae, and planetary systems. Traditional abundance determinations have relied on one‑dimensional (1D) static model atmospheres, which treat convection and radiative transfer through simplified, often empirical, parameters such as micro‑ and macroturbulence. While computationally convenient, these 1D models cannot capture the intrinsically three‑dimensional, highly dynamic nature of solar granulation, leading to systematic uncertainties in line formation, especially for lines formed in the upper photosphere or those sensitive to temperature inhomogeneities.

To overcome these limitations, the authors employ a realistic 3D radiative‑hydrodynamical model that solves the full set of Navier‑Stokes equations together with non‑grey radiative transfer on a fine Cartesian grid. The model reproduces observed granulation patterns, line asymmetries, and convective blueshifts without invoking ad‑hoc turbulence parameters. For each element, a carefully curated set of spectral lines is selected based on high‑resolution solar atlases, laboratory oscillator strengths, and the absence of significant blends. The line list spans neutral and ionized species across the periodic table, ensuring that nearly every element with observable photospheric lines is represented.

A crucial aspect of the analysis is the treatment of departures from local thermodynamic equilibrium (LTE). Whenever reliable atomic models are available, the authors perform full 3D non‑LTE (NLTE) radiative transfer calculations, incorporating detailed collisional cross‑sections and radiative rates. NLTE effects are found to be particularly important for light elements (C, N, O, Na, Mg) and iron‑peak species, where they can modify line strengths by up to 0.2 dex. By applying NLTE corrections on top of the 3D atmospheric structure, the derived abundances become largely independent of the specific line chosen, thereby enhancing internal consistency.

The resulting solar abundances differ markedly from the widely used values compiled a decade ago (e.g., Grevesse & Sauval 1998). Carbon, nitrogen, oxygen, and neon are reduced by 20–40 % relative to those older estimates, leading to a lower overall metallicity (Z) and a Z/X ratio of about 0.0185. This downward revision brings the solar composition into excellent agreement with the elemental ratios measured in CI chondritic meteorites and with the abundances of nearby F‑ and G‑type stars, which are thought to reflect the present‑day composition of the local interstellar medium. The authors demonstrate a high degree of internal consistency: different ionization stages, molecular lines, and forbidden transitions yield concordant abundances within the quoted uncertainties (typically 0.03–0.05 dex).

However, the new composition creates a pronounced tension with helioseismology. Standard solar interior models that previously matched the observed sound‑speed profile, convection‑zone depth, and surface helium abundance relied on the higher metallicity of the older abundance set. When the revised, lower Z is adopted, the computed sound‑speed deviations increase to ≈0.5 % and the convection‑zone base shifts outward, contradicting precise helioseismic inversions. This “solar abundance problem” suggests that either (i) additional physics—such as enhanced opacity, magnetic inhibition of convection, or rotational mixing—must be incorporated into interior models, or (ii) further refinements in the photospheric abundance analysis (e.g., even more sophisticated NLTE treatments or 3D magneto‑hydrodynamics) are required.

In the discussion, the authors outline several pathways to resolve the discrepancy. They propose targeted opacity experiments, refined atomic data for key opacity contributors (e.g., Fe, Ni), and the development of 3D magneto‑hydrodynamic solar models that could alter the temperature stratification in the line‑forming region. They also stress the importance of extending NLTE calculations to a broader set of elements and transitions, as well as performing independent abundance determinations using alternative diagnostics such as solar wind and solar energetic particle measurements.

In summary, this work delivers the most homogeneous and physically realistic solar abundance table to date, grounded in a 3D hydrodynamical atmosphere and extensive NLTE corrections. While it resolves many long‑standing inconsistencies with meteoritic and stellar data, it simultaneously sharpens the conflict with helioseismic constraints, thereby opening a new frontier for solar physics and stellar astrophysics. The paper serves as both a benchmark for future abundance studies and a catalyst for revisiting the microphysics of the solar interior.