Planetesimal Formation Research Articles

The Effect of Dust Evolution and Traps on Inner Disk Water Enrichment

Abstract Substructures in protoplanetary disks can act as dust traps that shape the radial distribution of pebbles. By blocking the passage of pebbles, the presence of gaps in disks may have a profound effect on pebble delivery into the inner disk, crucial for the formation of inner planets via pebble accretion. This process can also affect the delivery of volatiles (such as H2O) and their abundance within the water snow line region (within a few au). In this study, we aim to understand what effect the presence of gaps in the outer gas disk may have on water vapor enrichment in the inner disk. Building on previous work, we employ a volatile-inclusive disk evolution model that considers an evolving ice-bearing drifting dust population, sensitive to dust traps, which loses its icy content to sublimation upon reaching the snow line. We find that the vapor abundance in the inner disk is strongly affected by the fragmentation velocity (v f) and turbulence, which control how intense vapor enrichment from pebble delivery is, if present, and how long it may last. Generally, for disks with low to moderate turbulence (α ≤ 1 × 10−3) and a range of v f, radial locations and gap depths (especially those of the innermost gaps) can significantly alter enrichment. Shallow inner gaps may continuously leak material from beyond it, despite the presence of additional deep outer gaps. We finally find that for realistic v f (≤10 m s−1), the presence of gaps is more important than planetesimal formation beyond the snow line in regulating pebble and volatile delivery into the inner disk.

On Secular Gravitational Instability in Vertically Stratified Disks

Secular gravitational instability (GI) is one promising mechanism for explaining planetesimal formation. Previous studies of secular GI utilized a razor-thin disk model and derived the growth condition in terms of vertically integrated physical values such as dust-to-gas surface density ratio. However, in weakly turbulent disks where secular GI can operate, a dust disk can be orders of magnitude thinner than a gas disk, and analyses treating the vertical structures are necessary to clarify the interplay of the midplane dust motion and the upper gas motion. In this work, we perform vertically global linear analyses of secular GI with a vertical domain size of a few gas scale heights. We find that dust grains accumulate radially around the midplane while gas circulates over the whole vertical region. We obtain well-converged growth rates when the outer gas boundary is above two gas scale heights. The growth rates are underestimated if we assume the upper gas to be steady and regard it just as the source of external pressure to the dusty lower layer. Therefore, treating the upper gas motion is important even when the dust disk is much thinner than the gas disk. Conducting a parameter survey, we represent the growth condition in terms of the Toomre Q value for dust and dust-to-gas surface density ratio. The critical dust disk mass for secular GI is ∼10−4 M * for a dust-to-gas surface density ratio of 0.01, a Stokes number of 0.1, and a radial dust diffusivity of 10−4 c s H, where c s is the gas sound speed, and H is the gas scale height.

Evidence for Very Early Planetesimal Formation and 26Al/27Al Heterogeneity in the Protoplanetary Disk

We present a U-corrected Pb–Pb age of 4566.19 ± 0.20 Ma (1.11 ± 0.26 Myr after t 0) for the moderately volatile element rich, andesitic meteorite Erg Chech 002 (EC002). Our Al–Mg isochron defines a 26Al/27Al initial ratio of (8.65 ± 0.09) × 10−6 that corresponds to a 26Al/27Al ratio of × 10−5 for the parent body precursor at the time of solar system formation. Whereas the published bulk chemistry and our high-precision Ca isotope measurement correspond to those for inner solar system materials, the 26Al/27Al ratio overlaps that for outer solar system CI chondrites. This indicates that the carriers and/or processes responsible for the nucleosynthetic isotope compositions for inner and outer disk materials are different than those controlling the heterogeneous distribution of 26Al. A low μ 26Mg* initial value of −6.1 ± 1.7 ppm infers a source region with a subchondritic Al/Mg ratio until 1.1 Myr after t 0 such that melt generation must have immediately preceded its crystallization. With 26Al as the main heating source, a modeled temperature–time path for a 100 km radius parent body with our inferred 26Al abundance suggests that accretion must have occurred before 0.5 Myr after t 0 to reach melting temperatures at appropriate depths within 1.1 Myr. This requires that the parent body formed very early within the protoplanetary disk, consistent with predictions of rapid formation of planetesimals by streaming instabilities within high-density dust filaments during the earliest phase of the protoplanetary disk. Finally, an absence of initial Pb in this otherwise moderately volatile-rich achondrite implies Pb was effectively sequestered to the Fe–Ni core.

Identification of a 4.3 billion year old asteroid family and planetesimal population in the Inner Main Belt

Context. Understanding the conditions that lead to the formation of planetesimals – the building blocks of planets - and their initial size distribution is a central problem of modern planetology. While most of these original planetesimals were accreted onto the terrestrial planets and the cores of the giant planets, some were also stranded in the main belt, where 4.5 Gyr of collisional evolution broke most of them into families of collisional asteroid fragments. However, some planetesimals survived, and are still hidden amongst asteroid fragments in the main belt. Aims. We make use of astronomical data to identify these leftover planetesimals amongst all other asteroids. Our search is based on separating planetesimal survivors from families of asteroids generated by collisions. Namely, we aim to identify and “clean” the main belt of collisional family members: by doing so, we would be left with the surviving members of the original planetesimals. Methods. We focus here on the inner portion of the main belt for asteroids with intermediate to high albedo. It is known that current asteroid family catalogs are not suitable for the aforementioned cleaning; they are conservative and only one-quarter of the known asteroids are associated with the approximately 120 distinct asteroid families. We therefore developed methods to inclusively link asteroids to known collisional families in order to better capture their extent. Namely, we apply a hierarchical clustering method (HCM) on asteroids filtered according to the V-shape of the Yarkovsky drift of each family in order to reassess family membership (V-shape-constrained HCM). The identified families were removed and the remaining background population was searched for previously undetected collisional families. Results. We succeed in using our V-shape-constrained HCM to link family “halos” to their cores. After removing these reassessed families from the asteroid population, our V-shape search reveals a previously unknown collisional family of S-type asteroids in the inner main belt with an age of 4.3 ± 1.7 Gyr and a significance level of 3.4σ. When this ancient collisional family is removed, 34 planetesimals are identified and their size-frequency distribution is presented. Conclusions. The asteroid belt has two components: planetesimals and collisional fragments. The cumulative size-frequency distribution of planetesimals has a steep power-law index for bodies larger than 100 km in diameter and a much smaller power-law index for planetesimals smaller than 100 km.

Ryugu’s Anhydrous Ingredients and Their Spectral Link to Primitive Dust from the Outer Solar System

Ryugu is a second-generation C-type asteroid formed by the reassembly of fragments of a previous larger body in the main asteroid belt. While the majority of Ryugu samples returned by Hayabusa2 are composed of a lithology dominated by aqueously altered minerals, clasts of pristine olivine and pyroxene remain in the least-altered lithologies. These clasts are objects of great interest for revealing the composition of the dust from which the original building blocks of Ryugu's parent asteroid formed. Here we show that some grains rich in olivine, pyroxene, and amorphous silicates discovered in one millimeter-sized stone of Ryugu have infrared spectra similar to the D-type asteroid Hektor (a Jupiter Trojan), to comet Hale–Bopp, and to some anhydrous interplanetary dust particles of probable cometary origin. This result indicates that Ryugu's primary parent body incorporated anhydrous ingredients similar to the building blocks of asteroids (and possibly some comets) formed in the outer solar system, and that Ryugu retained valuable information on the formation and evolution of planetesimals at different epochs of our solar system's history.

Planetesimal Initial Mass Functions Following Diffusion-regulated Gravitational Collapse

The initial mass function (IMF) of planetesimals is of key importance for understanding the initial stages of planet formation, yet theoretical predictions so far have been insufficient in explaining the variety of IMFs found in simulations. Here, we connect diffusion-tidal-shear limited planetesimal formation within the framework of a Toomre-like instability in the particle midplane of a protoplanetary disk to an analytic prediction for the planetesimal IMF. The shape of the IMF is set by the stability parameter Q p, which in turn depends on the particle Stokes number, the Toomre Q value of the gas, the local dust concentration, and the local diffusivity. We compare our prediction to high-resolution numerical simulations of the streaming instability and planetesimal formation via gravitational collapse. We find that our IMF prediction agrees with numerical results and is consistent with both the paradigm that planetesimals are born big and the power-law description commonly found in simulations.

Instabilities in dusty non-isothermal protoplanetary discs

ABSTRACT Protoplanetary discs (PPDs) can host a number of instabilities that may partake directly or indirectly in the process of planetesimal formation. These include the vertical shear instability (VSI), convective overstability (COS), streaming instability (SI), and dust settling instability (DSI), to name a few. Notably, the VSI and COS have mostly been studied in purely gaseous discs, while the SI and DSI have only been analysed in isothermal discs. How these instabilities operate under more general conditions is therefore unclear. To this end, we devise a local model of a PPD describing a non-isothermal gas interacting with a single species of dust via drag forces. Using this, we find that dust drag sets minimum length-scales below which the VSI and COS are suppressed. Similarly, we find that the SI can be suppressed on sufficiently small scales by the gas’ radial buoyancy if it cools on roughly a dynamical time-scale. We show that the DSI can be effectively stabilized by vertical buoyancy, except at special radial and vertical length-scales. We also find novel instabilities unique to a dusty, non-isothermal gas. These result in a dusty analogue of the COS that operates in slowly cooled discs, and a dusty version of the VSI that is strongly enhanced by dust settling. We briefly discuss the possible implications of our results on planetesimal formation.

Formation of Terrestrial Planets from the Viewpoints of Astrophysics and Material Science – Formation of Planetesimals by Chemical Reactions at Contact Points between Solids

Formation of Terrestrial Planets from the Viewpoints of Astrophysics and Material Science – Formation of Planetesimals by Chemical Reactions at Contact Points between Solids

Massive Protostellar Disks as a Hot Laboratory of Silicate Grain Evolution

Typical accretion disks around massive protostars are hot enough for water ice to sublimate. We here propose to utilize the massive protostellar disks for investigating the collisional evolution of silicate grains with no ice mantle, which is an essential process for the formation of rocky planetesimals in protoplanetary disks around lower-mass stars. We, for the first time, develop a model of massive protostellar disks that includes the coagulation, fragmentation, and radial drift of dust. We show that the maximum grain size in the disks is limited by collisional fragmentation rather than by radial drift. We derive analytic formulae that produce the radial distribution of the maximum grain size and dust surface density in the steady state. Applying the analytic formulae to the massive protostellar disk of GGD27-MM1, where the grain size is constrained from a millimeter polarimetric observation, we infer that the silicate grains in this disk fragment at collision velocities above ≈10 m s−1. The inferred fragmentation threshold velocity is lower than the maximum grain collision velocity in typical protoplanetary disks around low-mass stars, implying that coagulation alone may not lead to the formation of rocky planetesimals in those disks. With future measurements of grain sizes in massive protostellar disks, our model will provide more robust constraints on the sticking property of silicate grains.

Siderophile element and Hf-W isotope characteristics of the metal-rich chondrite NWA 12273 – Implication for its origin and chondrite metal formation

Metal phases in primitive chondrites contain important information for understanding early Solar System processes. Northwest Africa (NWA) 12273 is an ungrouped chondrite and, due to its preternatural high modal abundance of metal grains (>60%), provides a unique perspective on the formation and evolution of primitive planetesimals. To investigate the formation of metal phases in early solar system materials we report detailed petrography, siderophile geochemistry and Hf-W isotope geochronology for NWA 12273. The primitive textures and heterogeneous mineral chemistry of chondrules, as well as siderophile element compositions of Ni-rich metal, all indicate an affinity of NWA 12273 to unequilibrated (type 3.8) ordinary chondrites that experienced limited thermal metamorphism. Modeling shows that crucial inter-element fractionations among refractory siderophile elements for kamacite in NWA 12273, were most likely caused by solid metal-liquid metal partitioning, for example during partial melting of Fe-Ni metal containing ∼3.9 wt.% of S. Analysis of the metal fractions within NWA 12273 gives a Hf-W model age of ∼2.4 Ma after CAI formation (ε182W6/4=−3.29±0.11), older than metal formation in most of H chondrites and IIE irons. In comparison with published data for chondrites and ‘non-magmatic’ iron meteorites, siderophile element data suggest that the metal phases in NWA 12273 may have originated from similar precursor materials and/or accreted in adjacent nebular locations to IIE irons and H chondrites. These primary planetesimals experienced impact-related evaporation and mixing followed by rebuilding of the secondary body. This scenario is consistent with that proposed for IIE and other silicate-bearing iron meteorites, indicating that the building blocks of some chondrite parent bodies were not entirely primitive (undifferentiated) materials.

Revisiting Collisional Dust Growth in Class 0/I Protostellar Disks: Sweep-up Can Convert a Few 10 M ⊕ of Dust into Kilogram Pebbles in 0.1 Myr

Recent observations suggest that the first stages of planet formation likely take place in the Class 0/I phase of young stellar object evolution, when the star and the disk are still embedded in an infalling envelope. In this study we perform grain coagulation calculations to investigate the very first stage of planet formation, the collisional growth of dust grains, in Class 0/I disks. We find that the slow increase in grain mass by high-velocity collision with much smaller grains (“sweep-up”) allows ∼50 M ⊕ of grains to grow well beyond the fragmentation barrier into ∼kilogram pebbles by the end of Class 0/I (0.1 Myr). We analyze the linear growth and saturation of sweep-up to understand our results quantitatively, and test whether the sweep-up outcome is sensitive to disk parameters and details of the grain coagulation model. The sweep-up pebble population could be important for planet formation, because they are less well-coupled to the gas (compared to the main population below the fragmentation barrier) and therefore more favorable to known mechanisms of dust clump formation (which initiate planetesimal formation). It also contains enough mass to form all planet cores, based on observational estimates of the planet mass budget. Our findings motivate future studies of grain growth and planetesimal formation in Class 0/I disks, including the subsequent evolution of this sweep-up population.

The Hot Main Kuiper Belt Size Distribution from OSSOS

Using the absolute detection calibration and abundant detections of the Outer Solar System Origins Survey project, we provide population measurements for the main Kuiper Belt. For absolute magnitude H r < 8.3, there are 30,000 nonresonant main-belt objects, with twice as many hot-component objects than cold, and with total mass of 0.014 M ⊕, only one-seventh of which is in the cold belt (assuming a cold-object albedo about half that of hot-component objects). We show that trans-Neptunian objects with 5.5 < H r < 8.3 (rough diameters 400–100 km) have indistinguishable absolute magnitude (size) distributions, regardless of being in the cold classical Kuiper Belt (thought to be primordial) or the “hot” population (believed to be implanted after having been formed elsewhere). We discuss how this result was not apparent in previous examinations of the size distribution due to the complications of fitting assumed power-law functional forms to the detections at differing depths. This shared size distribution is surprising in light of the common paradigm that the hot-population planetesimals formed in a higher density environment much closer to the Sun, in an environment that also (probably later) formed larger (dwarf planet and bigger) objects. If this paradigm is correct, our result implies that planetesimal formation was relatively insensitive to the local disk conditions and that the subsequent planet-building process in the hot population did not modify the shape of the planetesimal size distribution in this 50–300 km range.

The Evolution of Activity and Chemical Composition in Rosetta’s Comet Targets across Multiple Apparitions: Complications for CS2 as the CS Parent in Comet Nuclei

Jupiter-family comets are ephemeral small bodies injected into the inner solar system from the Kuiper Belt, doomed to either sublimate all their volatiles and become inert or violently shatter from the activity. We investigate two target candidates of the ESA Rosetta mission, comets 46P/Wirtanen and 67P/Churyumov-Gerasimenko, which had favorable apparitions for Earth-based observations in 2018–2019 and 2021, respectively. Using the Hubble Space Telescope STIS and COS instruments, we observed OH and CS emissions to characterize production rates of H2O and CS, established Af ρ values, and placed upper limits on the production rate of C2 and its parent. We find CS/H2O relative abundances that are significantly (5σ–7σ) larger than previous remote near-UV (NUV) measurements of 46P and 67P at similar heliocentric distances and CS/H2O values larger than those obtained via contemporaneous submillimeter observations for the same apparitions. We also find that for 67P the remote derivations of CS2/H2O ratios are substantially (∼50×) higher than the values measured by the ROSINA mass spectrometer on board the Rosetta spacecraft for all NUV-derived CS2 production rates. The discrepancy points toward an unidentified CS parent or parents with contributing factors from uncertainties with the fluorescence efficiencies of the CS (0,0) band of the A 1Π–X 1Σ+ system around 2580 Å. Given the significance of understanding the chemistry and dissociation physics of sulfur-bearing molecules in comets for tracing planetesimal formation environments, as well as the limited studies in this area, we propose several hypotheses to explain this discrepancy and outline future studies to address these issues.

Spins and shapes of basaltic asteroids and the missing mantle problem

Basaltic V-type asteroids are common in the inner part of the Main Asteroid Belt and much less abundant in the mid and outer parts. They are of scientific interest because they sample crusts and mantles of theoretically plentiful differentiated planetesimals that existed in the Solar System four billion years ago. Some Solar System theories suggest that those objects formed in the terrestrial planet region and were then implanted in the main asteroid belt. In consequence, we should observe a large number of fragments of multiple differentiated planetesimals in the inner Main Belt. That region of the Asteroid Belt is filled with V-type fragments; however, they are difficult to tell apart from typical Vestoids and Vesta fugitives. In this work, we focus on physical and dynamical characterization of V-types in the inner Main-Belt and aim to reconcile those properties with the planetesimal formation and evolution theories.We conducted an observing campaign over the years 2013–2022 and obtained photometric observations of V-type asteroids located mostly outside the Vesta family at specific locations of the inner Main Belt (the so-called Cells I and II). The total number of partial dense photometric lightcurves obtained in this study was ∼ 2910. We were able to model ∼100 V-types. We further supplement those data with 133 spins of V-types from the DAMIT database and 237 objects derived from Gaia DR3 (Ďurech & Hanuš 2023). We found 78%±11% and 38%±13% retrograde rotators in Cell I and II, respectively. This statistic is remarkably consistent with the numerical simulations of the escape paths of Vesta fugitives that predict 81% retrograde rotators in Cell I and 40% in Cell II after the dynamical integration of 2 Gys. Based on our statistics we conclude that if there are non-Vestoids in the inner main belt, they are likely to be very few. This is consistent with the small fraction of anomalous HED meteorites in meteorite collections, small number of non-Vestoids in the middle and outer Main Belt and points to planetesimal formation location close to the Sun.

Deuterium-enriched water ties planet-forming disks to comets and protostars.

Water is a fundamental molecule in the star and planet formation process, essential for catalysing the growth of solid material and the formation of planetesimals within disks1,2. However, the water snowline and the HDO:H2O ratio within proto-planetary disks have not been wellcharacterized because water only sublimates at roughly 160 K (ref. 3), meaning that most water is frozen out onto dust grains andthat the water snowline radii are less than 10 AU (astronomical units)4,5. The sun-like protostar V883 Ori (M* = 1.3 M⊙)6 is undergoing an accretion burst7, increasing its luminosity to roughly 200 L⊙ (ref. 8), and previous observations suggested that its water snowline is 40-120 AU in radius6,9,10. Here we report the direct detection of gas phase water (HDO and [Formula: see text]) from the disk of V883 Ori. We measure a midplane water snowline radius of approximately 80 AU, comparable to the scale of the Kuiper Belt, and detectwater out to a radius of roughly 160 AU. We then measure the HDO:H2O ratio of the disk to be (2.26 ± 0.63) × 10-3. This ratio is comparable to thoseofprotostellar envelopes and comets, and exceeds that of Earth's oceans by 3.1σ. We conclude that disks directly inherit water from the star-forming cloud and this water becomes incorporated into large icy bodies, such as comets, without substantial chemical alteration.

Planetesimal formation during protoplanetary disk buildup (Corrigendum)

Models of dust coagulation and subsequent planetesimal formation are usually computed on the backdrop of an already fully formed protoplanetary disk model. At the same time, observational studies suggest that planetesimal formation should start early, possibly even before the protoplanetary disk is fully formed. In this paper, we investigate under which conditions planetesimals already form during the disk buildup stage, in which gas and dust fall onto the disk from its parent molecular cloud. We couple our earlier planetesimal formation model at the water snow line to a simple model of disk formation and evolution. We find that under most conditions planetesimals only form after the buildup stage when the disk becomes less massive and less hot. However, there are parameters for which planetesimals already form during the disk buildup. This occurs when the viscosity driving the disk evolution is intermediate ($\alpha_v \sim 10^{-3}-10^{-2}$) while the turbulent mixing of the dust is reduced compared to that ($\alpha_t \le 0.03 \cdot \alpha_v$), and with the assumption that water vapor is vertically well-mixed with the gas. Such $\alpha_t \ll \alpha_v$ scenario could be expected for layered accretion, where the gas flow is mostly driven by the active surface layers, while the midplane layers, where most of the dust resides, are quiescent.

Formation of the First Planetesimals via the Streaming Instability in Globally Turbulent Protoplanetary Disks?

Using self-consistent models of turbulent particle growth in an evolving protoplanetary nebula of solar composition, we find that recently proposed local metallicity and Stokes number criteria necessary for the streaming instability to generate gravitationally bound particle overdensities are generally not approached anywhere in the disk during the first million years, an epoch in which meteoritic and observational evidence strongly suggests that the formation of the first planetesimals and perhaps giant planet core accretion are already occurring.

Planetesimal Formation by the Gravitational Instability of Dust Ring Structures

We investigate the gravitational instability (GI) of dust ring structures and the formation of planetesimals by their gravitational collapse. The normalized dispersion relation of a self-gravitating ring structure includes two parameters that are related to its width and line mass (the mass per unit length). We survey these parameters and calculate the growth rate and wavenumber. Additionally, we investigate the formation of planetesimals by growth of the GI of the ring that is formed by the growth of the secular GI of the protoplanetary disk. We adopt a massive, dust-rich disk as a disk model. We find the range of radii for fragmentation by the ring GI as a function of the width of the ring. The innermost radius for the ring GI is smaller for a smaller ring width. We also determine the range of the initial planetesimal mass resulting from the fragmentation of the ring GI. Our results indicate that the planetesimal mass can be as large as 1028 g at its birth after the fragmentation. It can be as low as about 1025 g if the ring width is 0.1% of the ring radius, and the lower limit increases with the ring width. Furthermore, we obtain approximate formulae for the upper and lower limits of the planetesimal mass. We predict that the planetesimals formed by the ring GI have prograde rotations because of the Coriolis force acting on the contracting dust. This is consistent with the fact that many trans-Neptunian binaries exhibit prograde rotation.

Constraining the turbulence and the dust disk in IM Lup: Onset of planetesimal formation

Context. Observations of protoplanetary disks provide information about the processes of planetary formation and reasons behind the diversity of planetary systems. The key to understanding planetary formation is the study of dust evolution from small grains to pebbles. Smaller grains (~1 µm) are well coupled to the gas dynamics and their distribution is significantly extended above the disk midplane. Larger grains settle much faster and are efficiently formed only in the midplane. By combining near-infrared polarized light and millimeter observations, it is possible to constrain the spatial distribution of both the small and large grains. Aims. Our aim is to construct detailed models of the size distribution and the vertical and radial structure of the dust particles in protoplanetary disks based on observational data. In particular, we are interested in recovering the dust distribution in the IM Lup protoplanetary disk. Methods. We created a physical model for the dust distribution of protoplanetary disks. We then simulated the radiative transfer of the millimeter continuum and the near-infrared polarized radiation. Using a Markov chain Monte Carlo method, we compared the derived images to the observations available for the IM Lup disk to constrain the best physical model for IM Lup and to recover the vertical grain size distribution. Results. The millimeter and near-infrared emission tightly constrain the dust mass and grain size distribution of our model. We find size segregation in the dust distribution, with millimeter-sized grains in the disk midplane. These grains are efficiently formed in the disk, possibly by sedimentation-driven coagulation, in accordance with the short settling timescales predicted by our model. This also suggests a high dust-to-gas ratio at smaller radii in the midplane, possibly triggering streaming instabilities and planetesimal formation in the inner disk. We obtained a turbulent α parameter of 3 × 10−3.

JWST/MIRI Spectroscopy of the Disk of the Young Eruptive Star EX Lup in Quiescence

EX Lup is a low-mass pre-main-sequence star that occasionally shows accretion-related outbursts. Here, we present JWST/MIRI medium-resolution spectroscopy obtained for EX Lup 14 yr after its powerful outburst. EX Lup is now in quiescence and displays a Class II spectrum. We detect a forest of emission lines from molecules previously identified in infrared spectra of classical T Tauri disks: H2O, OH, H2, HCN, C2H2, and CO2. The detection of organic molecules demonstrates that they are back after disappearing during the large outburst. Spectral lines from water and OH are for the first time deblended and will provide a much-improved characterization of their distribution and density in the inner disk. The spectrum also shows broad emission bands from warm, submicron-size amorphous silicate grains at 10 and 18 μm. During the outburst, in 2008, crystalline forsterite grains were annealed in the inner disk within 1 au, but their spectral signatures in the 10 μm silicate band later disappeared. With JWST we rediscovered these crystals via their 19.0, 20.0, and 23.5 μm emission, the strength of which implies that the particles are at ∼3 au from the star. This suggests that crystalline grains formed in 2008 were transported outwards and now approach the water snowline, where they may be incorporated into planetesimals. Containing several key tracers of planetesimal and planet formation, EX Lup is an ideal laboratory to study the effects of variable luminosity on the planet-forming material and may provide an explanation for the observed high crystalline fraction in solar system comets.