Dust retention in protoplanetary disks

Context: Protoplanetary disks are observed to remain dust-rich for up to several million years. Theoretical modeling, on the other hand, raises several questions. Firstly, dust coagulation occurs so rapidly, that if the small dust grains are not replenished by collisional fragmentation of dust aggregates, most disks should be observed to be dust poor, which is not the case. Secondly, if dust aggregates grow to sizes of the order of centimeters to meters, they drift so fast inwards, that they are quickly lost. Aims: We attempt to verify if collisional fragmentation of dust aggregates is effective enough to keep disks ‘dusty’ by replenishing the population of small grains and by preventing excessive radial drift. Methods: With a new and sophisticated implicitly integrated coagulation and fragmentation modeling code, we solve the combined problem of coagulation, fragmentation, turbulent mixing and radial drift and at the same time solve for the 1-D viscous gas disk evolution. Results: We find that for a critical collision velocity of 1 m/s, as suggested by laboratory experiments, the fragmentation is so effective, that at all times the dust is in the form of relatively small particles. This means that radial drift is small and that large amounts of small dust particles remain present for a few million years, as observed. For a critical velocity of 10 m/s, we find that particles grow about two orders of magnitude larger, which leads again to significant dust loss since larger particles are more strongly affected by radial drift.

💡 Research Summary

The paper tackles a long‑standing discrepancy between observations of protoplanetary disks, which remain dust‑rich for several million years, and theoretical models that predict rapid dust depletion. Two main theoretical problems are highlighted: (1) dust coagulation proceeds so quickly that, without a replenishment mechanism, the population of small grains should vanish, contradicting observations; (2) once aggregates grow to centimeter‑ to meter‑size, they experience strong aerodynamic drag and drift inward on timescales much shorter than the disk lifetime, leading to rapid loss of solid material. The authors set out to test whether collisional fragmentation can simultaneously (a) regenerate small grains and (b) limit the growth of large aggregates enough to suppress excessive radial drift.

To address this, they develop a new, implicitly integrated numerical code that couples dust coagulation, fragmentation, turbulent mixing, and radial drift with the one‑dimensional viscous evolution of the gas disk. The implicit scheme allows all processes to be solved together at each time step, ensuring numerical stability over the multi‑Myr integration required. The model adopts a standard α‑disk prescription for turbulence (α≈10⁻³–10⁻²) and tracks the dust size distribution across many mass bins.

A critical parameter in the model is the collision velocity threshold above which aggregates fragment rather than stick. Laboratory experiments on silicate aggregates suggest a low threshold of about 1 m s⁻¹, while some theoretical works have used higher values (∼10 m s⁻¹). The authors therefore run two sets of simulations: one with a critical velocity of 1 m s⁻¹ and another with 10 m s⁻¹, keeping all other disk parameters identical.

The results are strikingly different. In the 1 m s⁻¹ case, fragmentation is extremely efficient. Whenever relative velocities exceed the threshold, aggregates break apart, keeping the bulk of the dust mass in particles smaller than a few tens of microns. These small grains are tightly coupled to the gas (low Stokes numbers), so their radial drift velocity is negligible. Consequently, the disk retains a substantial reservoir of fine dust for the entire simulated period (several Myr), matching the observed spectral energy distributions of young disks. In contrast, with a 10 m s⁻¹ threshold, aggregates can grow to sizes two orders of magnitude larger (centimeter to decimeter scale) before fragmenting. Larger particles decouple from the gas, acquire higher Stokes numbers, and drift inward rapidly. The simulation shows a pronounced depletion of dust mass as these large bodies are lost to the star, reproducing the “drift barrier” problem that has plagued earlier models.

The authors also explore the sensitivity of the outcome to the turbulent α parameter. Even for relatively strong turbulence (α≈10⁻²), the low‑velocity fragmentation case still maintains a small‑grain dominated population, indicating that the fragmentation threshold, rather than turbulence strength, is the controlling factor for dust retention. Moreover, the study reveals a self‑regulating feedback loop: as particles grow, collision speeds increase, which in turn raises the probability of fragmentation, breaking large aggregates back into small fragments and preventing runaway growth.

In summary, the paper provides compelling evidence that a low critical fragmentation velocity (≈1 m s⁻¹), as measured in laboratory experiments, is sufficient to keep protoplanetary disks dusty over Myr timescales. This mechanism simultaneously curtails the formation of large, fast‑drifting bodies and replenishes the small‑grain population that dominates the observed infrared emission. Conversely, higher fragmentation thresholds lead to significant dust loss, highlighting the importance of accurate laboratory constraints on aggregate strength. The work bridges the gap between observations and theory, offering a robust framework for future studies of dust evolution, planetesimal formation, and the early stages of planet building.