About

Professor Andre Izidoro is a researcher affiliated with Rice University, focusing on planetary science and astrophysics. His research involves the study of planetary formation, dynamics, and evolution, contributing to the understanding of how planetary systems develop and change over time. He has mentored numerous students and postdoctoral researchers, including current group members and international students, and has a history of supervising undergraduate and graduate research projects. His academic background and professional activities are centered around advancing knowledge in planetary sciences, with a particular emphasis on simulation data and computational modeling.

Research topics

• Astronomy
• Physics
• Astrophysics

Selected publications

• The Role of Magnetospheric Rebound in Breaking Resonant Chains of Super-Earths and Mini-Neptunes

  The Astrophysical Journal · 2025-11-12 · 3 citations

  preprintOpen access

  Abstract Stellar magnetic fields are thought to truncate the inner regions of protoplanetary disks around T Tauri stars, creating a magnetospheric cavity near the star. As the disk evolves and disperses, the truncation radius is expected to move outward as the balance between magnetic and viscous forces shifts. Planets migrating inward can become trapped near the inner edge, but as the edge itself moves outward, the evolving disk torques can drive planets to migrate outward as well. We employ N -body simulations to assess the influence of magnetospheric cavity expansion on the dynamical evolution and orbital architectures of compact resonant chains of super-Earths and mini-Neptunes. Our results show that rebound-driven expansion of the disk’s inner edge plays a pivotal role in destabilizing resonant chains by spreading planetary systems outward, thereby triggering early dynamical instabilities and giant impacts. Despite this dynamical evolution, key observable properties of close-in planetary systems—such as the distribution of orbital period ratio, the intrasystem similarity in planet sizes (“radius uniformity”), and the bimodal distribution of planet radii known as the “radius valley”—remain largely consistent with those of systems formed without the rebound effect, in which the inner edge of the disk remains fixed. Thus, the primary consequence of the rebound appears to be the early disruption of resonant chains, rather than any significant alteration to the statistical properties of the resulting super-Earth and mini-Neptune populations.

  PublisherDOI

• Forming Uranus and Neptune through giant impacts: accretion scenarios with large and small impactor mass ratios in the early Solar System

  2025-07-09

  preprintOpen access

  1São Paulo State University, FEG, Department of Mathematics, Guaratinguetá, São Paulo, Brazil (leandro.esteves@unesp.br) 2Rice University, Department of Earth, Environmental and Planetary Sciences, Houston, Texas, USA The formation of Uranus and Neptune remains one of the outstanding problems in planetary science. Unlike Jupiter and Saturn, the ice giants possess relatively low masses and significantly tilted spin axes, with obliquities of ~98° and ~30°, respectively. These characteristics suggest that they experienced one or more giant impacts during their formation. However, the specific nature of these collisions—the number, mass ratio, and dynamical conditions of the impactors—remains debated. Previous studies have explored the accretion of Uranus and Neptune through giant impacts between planetary embryos with comparable masses, typically around 5 M⊕. These scenarios can successfully reproduce the current masses and mass ratio between the two planets, as well as their large obliquities, assuming stochastic impacts (Izidoro et al. 2015). However, these impacts often lead to the formation of planets with excessively rapid rotation, due to the large angular momentum delivered in approximately equal-mass collisions. This inconsistency with the present-day rotation periods of Uranus (~17.2 hours) and Neptune (~16.1 hours) presents a significant challenge. An alternative hypothesis involves impacts between bodies with large mass ratios—for instance, a proto-Uranus (~13 M⊕) and a much smaller embryo (~1 M⊕). Smooth Particle Hydrodynamics simulations indicate that such large mass ratio collisions can dissipate more angular momentum and result in slower rotating planets, with spin periods more consistent with Uranus and Neptune (Reinhardt et al. 2020). In addition, depending on the impact geometry and location, these impacts can still generate large obliquities, particularly for Uranus, without significantly altering the mass ratio or total mass of the planets. In this work, we explore both scenarios using a large suite of N-body simulations that incorporate key processes relevant to planet formation in a gaseous protoplanetary disk. Our simulations start with a population of planetary embryos with masses ranging from ~1 to 13 M⊕ and include the effects of type-I migration, as well as eccentricity and inclination damping from the gas disk. We investigate how the mass distribution of impactors and the dynamical environment influence the frequency and outcomes of collisions that can reproduce the observed characteristics of Uranus and Neptune. The high mass ratio scenario (HMR) simulations start with two massive protoplanets (~13 M⊕) and several small embryos (0.5–3 M⊕). The I15 scenario, based on Izidoro et al. (2015), involves only similar-mass embryos (~5 M⊕). Figure 1 illustrates the initial conditions used in simulations. The top panel shows the gas surface density as a function of radial distance, with the blue curve denoting how the giant planets shape the protoplanetary disk, following Morbidelli & Crida 2007. The vertical lines mark the approximate orbits of Jupiter, Saturn, and the range of distribution for embryos. The middle panel displays the normalized resultant torque, where negative values indicate inward migration toward the Sun and positive values represent outward migration. Lines are color-coded to denote varying body masses. The bottom panel depicts the approximate initial positions of Jupiter, Saturn, and the embryos, distributed between approximately 10 and 35 AU. Our results show that scenarios with high mass ratio impactors are more likely to yield planets with slower spin rates, alleviating the angular momentum problem present in equal-mass collision (I15) scenario. However, these same simulations exhibit a significantly reduced probability of such collisions occurring. This is because gas damping is relatively inefficient for low-mass embryos (≲1 M⊕), which tend to be dynamically excited and scattered by more massive protoplanets instead of merging with them. Consequently, although the final spin states are more favorable, the rarity of such collisions limits the overall success rate of this formation path. Conversely, simulations involving similar-mass impactors result in a higher frequency of collisions and a greater number of systems that match the final masses of Uranus and Neptune. Nonetheless, most of these planets end up with excess angular momentum, highlighting the trade-off between collision frequency and rotational outcomes in these different formation scenarios. Figure 2 shows the distribution of rotation periods for Uranus/Neptune analogues from simulations. The light-blue and dark-blue vertical lines represent the actual rotation periods of Uranus and Neptune, respectively. The four upper panels display planets that collided with specific small embryos in the High Mass Ratio (HMR) scenario simulations. The bottom panels show results from the I15 scenario with embryo masses of 6 M⊕ and 4-8 M⊕. The percentage plotted in red indicates the fraction of simulations where at least two protoplanets collided with embryos, reached masses close to those of Uranus and Neptune, and preserved the early Solar System architecture. Despite these contrasting dynamics, our statistical analysis shows that the overall probability of simultaneously reproducing the observed masses, mass ratio, and spin periods of Uranus and Neptune is comparable between the two scenarios, differing by no more than a factor of ~2. In both cases, the likelihood of achieving such an outcome remains low, on the order of 0.1–1%. These findings suggest that both the large and small mass ratio impact scenarios remain viable from a planet formation perspective. The ultimate pathway may depend on additional factors such as the structure and evolution of the protoplanetary disk, the timing of giant planet migration, and stochastic dynamical interactions in the outer solar system. Future work incorporating improved models of gas disk evolution, pebble accretion, and spin-orbit coupling may further constrain the plausibility of these scenarios. References: This work: Esteves, L., Izidoro, A. & Winter, O.C., 2025. Accretion of Uranus and Neptune: Confronting different giant impact scenarios. Icarus, 429, p.116428. Izidoro, A. et al., 2015. Accretion of Uranus and Neptune from inward-migrating planetary embryos blocked by Jupiter and Saturn. A&A, 582, A99. Morbidelli, A. & Crida, A., 2007. The dynamics of Jupiter and Saturn in the gaseous protoplanetary disk. Icarus, 191(1), pp.158–171. Reinhardt, C. et al., 2020. Bifurcation in the history of Uranus and Neptune: the role of giant impacts. MNRAS, 492(4), pp.5336–5353.

  PublisherDOI

• A Tale of Dynamical Instabilities and Giant Impacts in the Exoplanet Radius Valley

  The Astrophysical Journal Letters · 2025-06-17 · 3 citations

  articleOpen accessSenior authorCorresponding

  Abstract The size distribution of planets with radii between 1 R ⊕ and 4 R ⊕ peaks near 1.4 R ⊕ and 2.2 R ⊕ , with a dip around 1.8 R ⊕ —the so-called “radius valley.” Recent statistical analyses suggest that planets within this valley (1.5 < R < 2 R ⊕ ) tend to have slightly higher orbital eccentricities than those outside it. The origin of this dynamical signature remains unclear. We revisit the “breaking the chains” formation model and propose that late dynamical instabilities—occurring after disk dispersal—may account for the elevated eccentricities observed in the radius valley. Our simulations show that subvalley planets ( R < 2 R ⊕ ) are generally rocky, while those beyond the valley ( R > 2 R ⊕ ) are typically water-rich. Rocky planets that undergo strong dynamical instabilities and numerous late giant impacts have their orbits excited and their radii increased, ultimately placing them into the radius valley. In contrast, the larger, water-rich planets just beyond the valley experience weaker instabilities and fewer impacts, resulting in lower eccentricities. This contrast leads to a peak in the eccentricity distribution within the valley. The extent to which planets in the radius valley are dynamically excited depends sensitively on the orbital architecture before the orbital instability. Elevated eccentricities among radius valley planets arise primarily in scenarios that form a sufficiently large number of rocky planets within 100 days (typically ≳5) prior to instability, and that also host external perturbers ( P > 100 days), which further amplify the strength of dynamical instabilities.

  PublisherDOI

• A tale of dynamical instabilities and giant impacts in the radius valley

  ArXiv.org · 2025-05-29

  preprintOpen accessSenior author

  The size distribution of planets with radii between 1 and $4 R_\oplus$ peaks near 1.4 and $2.2R_\oplus$, with a dip around $1.8 R_\oplus$ -- the so-called "radius valley." Recent statistical analyses suggest that planets within this valley ($1.5 < R < 2R_\oplus$) tend to have slightly higher orbital eccentricities than those outside it. The origin of this dynamical signature remains unclear. We revisit the "breaking the chains" formation model and propose that late dynamical instabilities -- occurring after disk dispersal -- may account for the elevated eccentricities observed in the radius valley. Our simulations show that sub-valley planets ($R < 2 R_\oplus$) are generally rocky, while those beyond the valley ($R > 2 R_\oplus$) are typically water-rich. Rocky planets that undergo strong dynamical instabilities and numerous late giant impacts have their orbits excited and their radii increased, ultimately placing them into the radius valley. In contrast, the larger, water-rich planets just beyond the valley experience weaker instabilities and fewer impacts, resulting in lower eccentricities. This contrast leads to a peak in the eccentricity distribution within the valley. The extent to which planets in the radius valley are dynamically excited depends sensitively on the orbital architecture before the orbital instability. Elevated eccentricities among radius valley planets arise primarily in scenarios that form a sufficiently large number of rocky planets within 100 days (typically $\gtrsim 5$) prior to instability, and that also host external perturbers ($P > 100$ days), which further amplify the strength of dynamical instabilities.

  PublisherOA PDFDOI

• The Link between Athor and EL Meteorites Does Not Constrain the Timing of the Giant Planet Instability

  The Astrophysical Journal Letters · 2025-09-08 · 2 citations

  articleOpen access1st authorCorresponding

  Abstract The asteroid Athor, residing today in the inner main asteroid belt, has been recently identified as the source of EL enstatite meteorites to Earth. It has been argued that Athor formed in the terrestrial region—as indicated by a similarity in isotopic compositions between Earth and EL meteorites—and was implanted in the belt ≳60 Myr after the formation of the solar system. A recently published study modeling Athor’s implantation in the belt further concluded, using an idealized set of numerical simulations, that Athor could not have been scattered from the terrestrial region and implanted at its current location unless the giant planet dynamical instability occurred after Athor’s implantation (≳60 Myr). In this work, we revisit this problem with a comprehensive suite of dynamical simulations of the implantation of asteroids into the belt during the terrestrial planet accretion. We find that Athor-like objects can in fact be implanted into the belt long after the giant planets’ dynamical instability. The probability of implanting Athor analogs when the instability occurs at ≲15 Myr is at most a factor of ∼2 lower than that of an instability occurring ∼100 Myr after the solar system formation. Moreover, Athor’s implantation can occur up to ≳100 Myr after the giant planet instability. We conclude that Athor’s link to EL meteorites does not constrain the timing of the solar system’s dynamical instability.

  PublisherOA PDFDOI

• Reassessing the origin and evolution of Ecliptic Comets in the Planet-9 Scenario

  ArXiv.org · 2025-01-28

  preprintOpen access

  A group of newly observed extreme trans-Neptunian objects exhibit unexpected orbital confinement, characterized by the alignment of orbital angular momentum vectors and apsidal lines. It is proposed that an undiscovered giant planet, named Planet-9, exists in the solar system's outer regions and causes this clustering. Initial studies suggested Planet-9 could have a mass of 15 Earth masses. However, such a massive planet strongly interacts with scattered disk objects (SDOs; 50 < a < 1000 au) and influences the orbits of short-period comets, resulting in orbital inclinations inconsistent with observations. This study models the formation and long-term evolution of trans-Neptunian object populations and the Oort cloud during the solar system's dynamical instability, using revised parameters for Planet-9. Simulations assume Planet-9 has a mass of 7.5 Earth masses, an inclination of ~20 degrees, a semi-major axis of ~600 au, and an eccentricity of ~0.3. Results suggest a less massive Planet-9 aligns with observed trans-Neptunian object inclinations and the number of ecliptic comets (D > 10 km). Distant Kuiper belt objects with 40 < q < 100 au and 200 < a < 500 au, particularly with significant inclinations, are more likely to align apsidally with Planet-9, with an anti-aligned-to-aligned ratio of 0.5-0.7. Lower inclination objects (<20 degrees) exhibit significant apsidal anti-alignment, with an anti-aligned-to-aligned ratio of 2-4. These findings offer a new observational direction to refine the search for Planet-9.

  PublisherOA PDFDOI

• Reassessing the origin and evolution of Ecliptic Comets in the Planet-9 Scenario

  Icarus · 2025-02-19 · 1 citations

  article
  PublisherDOI

• Size-Frequency Distribution of Terrestrial Leftover Planetesimals and S-complex Implanted Asteroids

  ArXiv.org · 2025-04-15

  preprintOpen access

  The isotopic composition of meteorites linked to S-complex asteroids has been used to suggest that these asteroids originated in the terrestrial planet's region, i.e., within 1.5 au, and later got implanted into the main asteroid belt (MAB). Dynamical models of planet formation support this view. Yet, it remains to be demonstrated whether the currently observed size-frequency distribution (SFD) of S-complex bodies in the MAB can be reproduced via this implantation process. Here we studied the evolution of the SFD of planetesimals during the accretion of terrestrial planets with the code LIPAD self-consistently accounting for growth and fragmentation of planetesimals. In our simulations we vary the initial surface density of planetesimals, the gaseous disk lifetime, and the power slope of the initial planetesimals' SFD. We compared the final SFDs of leftover planetesimals in the terrestrial planet region with the SFD of observed S-complex MAB objects (D $>$ 100km). We found that the SFDs of our planetesimal populations and that of S-complex MAB objects show very similar cumulative power index (i.e., q $\approx$ 3.15 in N($>$D)$~\propto$ D$^{-q}$) for slopes in the diameter range 100 km $<$ D $<$ 400 km by the end of our simulations. Our results support the hypothesis of S-complex MAB implantation from the terrestrial planet forming region, assuming implantation is size-independent, and implies that implantation efficiency is smaller than $\mathcal{O}$(10$^{\rm -2}$--10$^{\rm -4}$) to avoid over-implantation of (4) Vesta-sized objects or larger.

  PublisherOA PDFDOI

• Size–Frequency Distribution of Terrestrial Leftover Planetesimals and S-complex Implanted Asteroids

  The Astrophysical Journal · 2025-06-12 · 4 citations

  articleOpen accessCorresponding

  Abstract The isotopic composition of meteorites linked to S-complex asteroids has been used to suggest that these asteroids originated in the terrestrial planet’s region, i.e., within 1.5 au, and later got implanted into the main asteroid belt (MAB). Dynamical models of planet formation support this view. Yet it remains to be demonstrated whether the currently observed size–frequency distribution (SFD) of S-complex bodies in the MAB can be reproduced via this implantation process. Here we studied the evolution of the SFD of planetesimals during the accretion of terrestrial planets with the code LIPAD self-consistently accounting for growth and fragmentation of planetesimals. In our simulations we vary the initial surface density of planetesimals, the gaseous disk lifetime, and the power slope of the initial planetesimals’ SFD. We compared the final SFDs of leftover planetesimals in the terrestrial planet region with the SFD of observed S-complex MAB objects ( D &gt; 100 km). We found that the SFDs of our planetesimal populations and that of S-complex MAB objects show very similar cumulative power index (i.e., q ≈ 3.15 in N (&gt; D ) ∝ D − q ) for slopes in the diameter range 100 km &lt; D &lt; 400 km by the end of our simulations. Our results support the hypothesis of S-complex MAB implantation from the terrestrial planet forming region, assuming implantation is size independent, and imply that implantation efficiency is smaller than <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mi class="MJX-tex-calligraphic" mathvariant="script">O</mml:mi> </mml:math> (10 –2 –10 –4 ) to avoid overimplantation of (4) Vesta-sized objects or larger.

  PublisherDOI

• Very-wide-orbit planets from dynamical instabilities during the stellar birth cluster phase

  Nature Astronomy · 2025-05-27 · 4 citations

  articleOpen access1st authorCorresponding
  PublisherOA PDFDOI

Frequent coauthors

• Sean N. Raymond

  Laboratoire d'Astrophysique de Bordeaux

  244 shared

• Alessandro Morbidelli

  Centre National de la Recherche Scientifique

  96 shared

• M. Ollivier

  68 shared

• A. Pierens

  55 shared

• Daniel Rouan

  Université de Versailles Saint-Quentin-en-Yvelines

  48 shared

• F. Hersant

  Université de Bordeaux

  42 shared

• Bertram Bitsch

  University College Cork

  41 shared

• O. C. Winter

  Universidade Estadual Paulista (Unesp)

  41 shared

Labs

• Andre Izidoro LabPI

Education

• PhD

  Sao Paulo State University