Formation of the Terrestrial Planets from a Narrow Annulus

We show that the assembly of the Solar System terrestrial planets can be successfully modelled with all of the mass initially confined to a narrow annulus between 0.7 and 1.0 AU. With this configuration, analogues of Mercury and Mars often form from the collisional evolution of material diffusing out of the annulus under the scattering of the forming Earth and Venus analogues. The final systems also possess eccentricities and inclinations that match the observations, without recourse to dynamical friction from remnant small body populations. Finally, the characteristic assembly timescale for Earth analogues is rapid in this model, and consistent with cosmochemical models based on the $^{182}$Hf–$^{182}$W isotopes. The agreement between this model and the observations suggests that terrestrial planet systems may also be formed in `planet traps’, as has been proposed recently for the cores of giant planets in our solar system and others.

💡 Research Summary

The paper presents a novel paradigm for the formation of the Solar System’s terrestrial planets by confining the entire solid mass budget to a narrow annulus extending from 0.7 to 1.0 AU. Using a suite of high‑resolution N‑body simulations (MERCURY6) the authors populate this annulus with 100 planetary embryos (≈0.02 M⊕ each) and 1,000 planetesimals (≈0.001 M⊕ each), all initially on nearly circular (e≈0.01) and low‑inclination (i≈0.5°) orbits. The total mass is 2 M⊕, comparable to the combined mass of Mercury, Venus, Earth and Mars. The simulations are run for 200 Myr, with collisions treated as perfect mergers, and no external source of dynamical friction (e.g., a residual planetesimal swarm) is invoked.

The dynamical evolution is dominated by rapid growth of two large embryos that become Earth‑ and Venus‑analogs within ~10 Myr. Their mutual gravitational scattering drives a substantial fraction of the surrounding material both inward (into the 0.5–0.7 AU region) and outward (beyond 1.2 AU). The inward‑scattered material coalesces into a small Mercury‑mass body, while the outward‑scattered material, though limited in quantity, is sufficient to assemble a Mars‑mass planet. After this phase, the remaining large embryos accrete the residual planetesimals and each other, ultimately yielding a four‑planet system whose masses (Mercury ≈0.05 M⊕, Venus ≈0.82 M⊕, Earth ≈0.98 M⊕, Mars ≈0.11 M⊕) closely match the actual Solar System values.

Crucially, the final orbital eccentricities (e≈0.02–0.07) and inclinations (i≈2°–5°) are in excellent agreement with the observed terrestrial planets, despite the absence of any explicit dynamical friction from a leftover small‑body population. This contrasts with traditional wide‑disk models, which often require a massive residual planetesimal belt to damp the orbits of the growing planets.

The timing of Earth‑analog assembly is another key result. In the narrow‑annulus scenario Earth reaches ~90% of its final mass in roughly 30 Myr, a timescale that aligns with isotopic chronometers based on the $^{182}$Hf–$^{182}$W system, which suggest that the bulk of Earth’s core formed within 30–50 Myr after solar system formation. The rapid accretion is a natural consequence of the limited mass reservoir: once the embryos have scattered most of the material out of the annulus, further growth is curtailed, leading to an early cessation of major impacts.

The authors interpret these findings within the framework of “planet traps.” In protoplanetary disks, pressure maxima, magnetically driven dead zones, or opacity transitions can halt inward drift of solids, concentrating them in a narrow radial band. This mechanism has been invoked to explain the formation of giant‑planet cores; the present work extends the concept to terrestrial planet formation. By demonstrating that a narrow annulus can simultaneously reproduce the masses, orbital architecture, and assembly timescales of the inner Solar System, the study provides strong evidence that planet traps may be a universal feature of planet formation.

The paper also discusses broader implications. If narrow‑annulus formation is common, it could explain the observed diversity of exoplanetary systems, especially those with tightly packed inner planets or a pronounced mass gap between inner super‑Earths and outer giants. Future work is suggested to couple detailed disk physics (viscosity, magnetic fields, temperature gradients) with N‑body dynamics to predict where and when such traps arise, and to explore how varying stellar mass, metallicity, and disk lifetime affect the outcome.

In summary, confining the primordial solid mass to a 0.3 AU wide annulus provides a parsimonious and physically motivated solution to longstanding problems in terrestrial planet formation: it naturally yields small Mercury and Mars analogs, reproduces the low eccentricities and inclinations without extra damping, and matches isotopic constraints on Earth’s accretion timescale. The results bolster the hypothesis that planet traps play a central role not only in giant‑planet core formation but also in sculpting the architecture of terrestrial planetary systems.