Presentation Information
[PPS04-P31]Radial Drift of Sintered Dust Aggregates toward the Central Star
*YOSUKE SEGAWA1, Sin-iti Sirono1 (1.Nagoya University)
Keywords:
planetesimal,sintering,pressure bump
Protoplanetary disks consist of gas and solid dust with a mass ratio of approximately 100:1. The initial dust grain size is about 0.1 um, from which kilometer-sized planetesimals and eventually planets are formed. Dust particles grow through mutual collisions and sticking, forming irregular aggregates. Numerical simulations have shown that aggregates initially grow via hit-and-stick collisions, resulting in increasingly porous and low-density structures (Kataoka et al. 2013).
In this study, we focus on the process of sintering and its effects. Heating by the central star promotes sintering, leading to neck growth and grain growth, while the aggregate as a whole shrinks.
If sintering does not occur, aggregates can grow through collisional sticking. However, during the early stages of sintering, even partial neck growth increases the mechanical strength of aggregates. As a result, collisions between aggregates may lead to bouncing rather than sticking. Therefore, when sintering is considered, collisional growth can be inhibited due to bouncing.
We then consider pressure bumps, regions in the disk where dust accumulates. Dust particles drift toward pressure bumps, where their relative velocities are reduced, enabling collisional growth. Subsequently, streaming instability may be triggered, leading to planetesimal formation. Isotopic analyses of differentiated and undifferentiated meteorites suggest that planetesimals formed within 10^5~several 10^6 years after disk formation. However, highly porous aggregates experience slow radial drift, making it difficult for sufficient dust to accumulate at pressure bumps within this timescale.
If sintering proceeds further, dust particles grow and aggregates shrink, reducing gas drag and increasing their radial drift velocity. The objective of this study is to determine how long it takes for a sufficient amount of sintered aggregates to drift into a pressure bump.
In our model, we assume that streaming instability is triggered when the dust surface density within a pressure bump of width 0.1 au increases to ten times its initial value. We perform numerical calculations using three parameters: the initial aggregate size at the bouncing threshold aagg0, ranging from 0.1 um to 1 m; the fractal dimension x, ranging from 2.0 to 3.0; and the radial location of the pressure bump r_b, ranging from 2 to 50 au. Dust is assumed to consist of H_2O ice, and the fractal dimension is assumed to remain constant in time. For comparison, we also compute a case no sintering, in which the monomer size remains fixed at 0.1 um.
Our results show that when sintering is included, dust accumulates at pressure bumps more rapidly than in the no-sintering case. For pressure bumps located closer to the central star, the surrounding gas density is higher, which increases gas drag and slows radial drift. In the no-sintering case, this results in longer accumulation timescales. In contrast, when sintering is included, the effect of increased gas density in slowing radial drift is nearly balanced by the higher temperatures that accelerate sintering and thereby enhance drift velocity. As a result, the accumulation timescale becomes relatively insensitive to the location of the pressure bump. Beyond a certain radial distance, where temperatures are too low for sintering to proceed efficiently, the difference between the sintering and no-sintering cases disappears. Overall, including sintering can produce differences of up to two orders of magnitude in the accumulation timescale.
These results suggest that sintering is effective primarily within approximately 20 au. Previous numerical studies of direct collisional growth without sintering have suggested that planetesimals can form within about 10^5 years inside this region (Okuzumi et al. 2012). However, such models imply that small dust aggregates are rapidly depleted, making it difficult to reconcile with meteoritic evidence indicating planetesimal formation over several 10^6 years. By incorporating both sintering and pressure bumps, our model can reproduce longer formation timescales depending on parameter choices, suggesting that planetesimal formation consistent with meteoritic constraints may be achieved under these conditions.
In this study, we focus on the process of sintering and its effects. Heating by the central star promotes sintering, leading to neck growth and grain growth, while the aggregate as a whole shrinks.
If sintering does not occur, aggregates can grow through collisional sticking. However, during the early stages of sintering, even partial neck growth increases the mechanical strength of aggregates. As a result, collisions between aggregates may lead to bouncing rather than sticking. Therefore, when sintering is considered, collisional growth can be inhibited due to bouncing.
We then consider pressure bumps, regions in the disk where dust accumulates. Dust particles drift toward pressure bumps, where their relative velocities are reduced, enabling collisional growth. Subsequently, streaming instability may be triggered, leading to planetesimal formation. Isotopic analyses of differentiated and undifferentiated meteorites suggest that planetesimals formed within 10^5~several 10^6 years after disk formation. However, highly porous aggregates experience slow radial drift, making it difficult for sufficient dust to accumulate at pressure bumps within this timescale.
If sintering proceeds further, dust particles grow and aggregates shrink, reducing gas drag and increasing their radial drift velocity. The objective of this study is to determine how long it takes for a sufficient amount of sintered aggregates to drift into a pressure bump.
In our model, we assume that streaming instability is triggered when the dust surface density within a pressure bump of width 0.1 au increases to ten times its initial value. We perform numerical calculations using three parameters: the initial aggregate size at the bouncing threshold aagg0, ranging from 0.1 um to 1 m; the fractal dimension x, ranging from 2.0 to 3.0; and the radial location of the pressure bump r_b, ranging from 2 to 50 au. Dust is assumed to consist of H_2O ice, and the fractal dimension is assumed to remain constant in time. For comparison, we also compute a case no sintering, in which the monomer size remains fixed at 0.1 um.
Our results show that when sintering is included, dust accumulates at pressure bumps more rapidly than in the no-sintering case. For pressure bumps located closer to the central star, the surrounding gas density is higher, which increases gas drag and slows radial drift. In the no-sintering case, this results in longer accumulation timescales. In contrast, when sintering is included, the effect of increased gas density in slowing radial drift is nearly balanced by the higher temperatures that accelerate sintering and thereby enhance drift velocity. As a result, the accumulation timescale becomes relatively insensitive to the location of the pressure bump. Beyond a certain radial distance, where temperatures are too low for sintering to proceed efficiently, the difference between the sintering and no-sintering cases disappears. Overall, including sintering can produce differences of up to two orders of magnitude in the accumulation timescale.
These results suggest that sintering is effective primarily within approximately 20 au. Previous numerical studies of direct collisional growth without sintering have suggested that planetesimals can form within about 10^5 years inside this region (Okuzumi et al. 2012). However, such models imply that small dust aggregates are rapidly depleted, making it difficult to reconcile with meteoritic evidence indicating planetesimal formation over several 10^6 years. By incorporating both sintering and pressure bumps, our model can reproduce longer formation timescales depending on parameter choices, suggesting that planetesimal formation consistent with meteoritic constraints may be achieved under these conditions.