🔭 astrophysics

ALMA 873 $μ$ m Polarization Observations of the PDS~70 Disk

This study presents the first deep, full polarization observations of the PDS~70 protoplanetary disk at 873 $\mu$ m, revealing that dust self-scattering from grains up to $\sim$ 87–100 $\mu$ m in size within a marginally optically thick ring explains the observed polarization, suggesting that ice-coated dust coagulation is limited by fragmentation or bouncing.

Original authors: Hauyu Baobab Liu, Kiyoaki Doi, Simon Casassus, Akimasa Kataoka, Ruobing Dong, Jun Hashimoto, Philipp Weber

Published 2026-04-01

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hauyu Baobab Liu, Kiyoaki Doi, Simon Casassus, Akimasa Kataoka, Ruobing Dong, Jun Hashimoto, Philipp Weber

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a cosmic construction site where planets are being built. For a long time, astronomers have been trying to figure out exactly how the tiny dust grains in these construction sites stick together to eventually become rocks, then boulders, and finally, planets.

The star system PDS 70 is the most famous construction site we know of. It's special because we can actually see two giant planets currently being built there, like two toddlers playing in a sandbox. But to understand how they got there, we need to look at the "sand" (dust) they are made of.

This paper is like a high-tech, super-magnifying glass report on that sand. Here is the story of what the astronomers found, explained simply:

1. The New "Super-Snapshot"

Previously, we had pictures of PDS 70, but they were a bit blurry or only showed the heat (brightness) of the dust. In this study, the team used the ALMA telescope (a giant array of radio dishes in the Chilean desert) to take the deepest, sharpest picture ever taken of this system at a specific wavelength (873 micrometers).

Think of this as switching from a standard camera to a polarized camera. Just like how polarized sunglasses can cut through glare to see the road clearly, this telescope can see the direction the light waves are vibrating. This direction tells us about the shape and size of the dust grains, much like how the way light reflects off a smooth pebble is different from how it reflects off a jagged rock.

2. The Big Discovery: The Dust is "Stuck"

The astronomers found something surprising about the dust in the ring where the planets are forming.

The Expectation: You might think that in a place where planets are forming, the dust grains would have grown huge—like beach balls or even basketballs (millimeters to centimeters in size).
The Reality: The polarization data showed that the dust grains are actually quite small, roughly the size of sand grains or fine flour (about 100 micrometers, or 0.1 millimeters).

The Analogy: Imagine you are trying to build a sandcastle. You expect the sand to clump together into big, solid blocks. But instead, you find that the sand grains are coated in a layer of ice that makes them incredibly slippery. When they bump into each other, instead of sticking and growing bigger, they just bounce off or shatter.

3. The "Ice Coating" Problem

Why are the grains so small? The paper suggests it's because of the water snowline.

Inside a certain distance from the star, it's too hot for ice, so dust is dry and sticky (like wet sand).
Outside that line (where PDS 70's ring is), it's cold enough for water to freeze. The dust grains get coated in water ice.

In a laboratory, scientists have found that ice-coated dust isn't very sticky. It's like trying to build a tower out of frozen, slippery marbles. They bounce off each other rather than sticking. This paper provides the first real-world evidence that this "slippery ice" is happening in a real planet-forming disk.

4. The "Traffic Jam" of Light

The team also noticed that the ring of dust is so dense that it's almost "optically thick."

The Analogy: Imagine a crowded concert hall. If you stand in the back, you can't see the stage because there are too many people in front of you. The light from the back is blocked.
In PDS 70, the dust ring is so packed that light from the back of the ring can't get through to us easily. This "traffic jam" of dust helps explain why the grains haven't grown bigger; they are constantly colliding in a crowded space, but because they are slippery (ice-coated), they break apart instead of merging.

5. The "Bouncing Barrier"

The paper concludes that the growth of dust in this system is likely limited by bouncing.

The Metaphor: Imagine a game of dodgeball. If the balls are soft and sticky, they stick together when they hit. If the balls are hard and icy, they bounce off.
In PDS 70, the "dodgeballs" (dust grains) are hitting each other too hard and are too slippery to stick. They bounce around, breaking into smaller pieces or staying small, rather than growing into the massive building blocks needed to form the core of a planet.

Why Does This Matter?

This is a big deal for understanding how our own solar system formed.

If dust can't grow past the "bouncing barrier," how do planets ever form?
This study suggests that in cold, icy regions of space, planet formation might be much harder and slower than we thought. It implies that the planets we see in PDS 70 might have formed through a very difficult, "stop-and-start" process, or perhaps they formed in a different way than our current theories predict.

In a nutshell: The astronomers took a super-clear, polarized photo of a planet-building nursery and discovered that the building blocks (dust) are covered in slippery ice. Because of this, the blocks keep bouncing off each other instead of sticking together, making it a very tough job to build giant planets.

1. Problem Statement

The formation of rocky planets requires dust grains to grow from sub-micron sizes to at least $\gtrsim 10^3$ µm (millimeter to centimeter sizes) via coagulation. However, the efficiency of this growth, particularly for water-ice-coated grains found outside the water snowline, remains a subject of debate. Laboratory experiments suggest ice-coated grains may have low "stickiness," leading to fragmentation or bouncing barriers that prevent them from growing large.

The PDS 70 system is a unique laboratory for studying this process because it hosts two confirmed accreting protoplanets (PDS 70b and 70c) within a protoplanetary disk. While previous studies have resolved the disk's substructures, the maximum dust grain size ( $a_{max}$ ) in the planet-forming region remains poorly constrained. Traditional methods using spectral indices ( $\alpha$ ) suffer from degeneracies and low precision. The authors aim to break this degeneracy by utilizing dust self-scattering polarization, a technique that can precisely constrain $a_{max}$ when grains are in the Mie scattering regime ( $a_{max} \sim \lambda/2\pi$ ).

2. Methodology

Observations

The study presents the first deep, full-polarization observations of the PDS 70 disk at 873 µm (ALMA Band 7) with an angular resolution of $\lesssim 0.1''$ .

New Data:
- Band 7 (873 µm): Full polarization (Stokes I, Q, U) observations using high-resolution and short-spacing configurations to ensure no missing flux.
- Band 3 & 4 (3075 µm & 2068 µm): Total intensity (Stokes I) observations to complement the spectral coverage.
Archival Data:
- Band 6 (1226 µm & 1287 µm): Retrieved from ALMA archives (Project 2019.1.01619.S) to complete the spectral energy distribution (SED) coverage.

Data Analysis & Modeling

Imaging: The authors performed multi-frequency synthesis imaging and primary beam correction. Polarization percentages were de-biased to account for thermal noise.
Spectral Indices: Spatially resolved spectral indices ( $\alpha_{b6-b7}$ ) were calculated between 873 µm and 1226 µm to trace optical depth and grain size variations.
Radiative Transfer Modeling:
- A Monte Carlo Radiative Transfer (MCRT) model was constructed using the RADMC-3D code.
- The model included a small-dust disk (sub-µm grains) to reproduce optical/IR SEDs, and distinct structures for the (sub)millimeter ring (narrow and wide Gaussian rings) and two dust crescents (Northwest and Southwest).
- MCMC Optimization: The Markov Chain Monte Carlo (MCMC) method was used to optimize free parameters (dust mass surface density, grain size $a_{max}$ , scale height, and truncation radii) to fit the observed Stokes I intensities and polarization fractions.
- The model assumed compact, water-ice-coated dust grains (DSHARP composition) outside the snowline.

3. Key Contributions

First Polarization Detection: This is the first detection of linear polarization in a disk hosting confirmed protoplanets via direct imaging.
Constraining Grain Size: By combining polarization position angles and fractions with spectral indices, the study provides a robust constraint on the maximum dust grain size ( $a_{max}$ ) in the PDS 70 ring, overcoming the degeneracies of spectral index-only analyses.
Physical Interpretation of Crescent Structures: The study distinguishes between the physical properties of the main ring and the asymmetric crescent features, linking them to dust growth barriers.

4. Key Results

Polarization Observations

Detection: Linear polarization of $\sim 1\%$ – $2.5\%$ was detected across the bulk of the $\sim 55$ –$100$ AU ring.
Orientation: The polarization position angles align preferentially with the projected minor axis of the disk. This alignment is characteristic of dust self-scattering in an inclined, marginally optically thick disk (rather than a geometrically thin ring where angles would be radial).
Azimuthal Variation: The polarization fraction is maximized in the west and minimized at the Northwest (NW) crescent.

Grain Size and Optical Depth

Maximum Grain Size ( $a_{max}$ ): The observed polarization and spectral indices are best reproduced by a model with $a_{max} \approx 87$ µm in the main ring and $a_{max} \approx 79$ µm in the NW crescent.
Optical Depth: The ring is marginally optically thick at 873 µm in the azimuthal direction. The NW crescent is highly optically thick, which suppresses the polarization fraction there not due to grain size differences, but due to high optical depth.
Spectral Indices: The spectral index $\alpha_{b6-b7}$ shows significant azimuthal variation, dropping to $\sim 1.86$ near the NW crescent, confirming the high optical depth in that region.

Fragmentation Velocity

Using the derived $a_{max}$ and assuming dust growth is limited by fragmentation, the authors calculated the fragmentation velocity ( $v_{frag}$ ).
The results suggest $v_{frag} \lesssim 1$ m s $^{-1}$ for the water-ice-coated dust in the ring.
This low velocity is consistent with laboratory experiments suggesting that water-ice-coated grains are not sufficiently "sticky" to overcome bouncing or fragmentation barriers to grow beyond $\sim 100$ µm.

5. Significance

Planet Formation Constraints: The finding that $a_{max} \lesssim 100$ µm in a disk hosting massive planets implies that dust coagulation in the PDS 70 system is likely halted by fragmentation or bouncing barriers rather than efficient growth to kilometer-sized planetesimals in the gas-rich mid-plane.
Implications for Planetary Composition: Since the disk is outside the water snowline, the inability of grains to grow large suggests that planets forming here may have difficulty accreting massive solid cores efficiently, or that the observed planets (PDS 70b/c) formed via different mechanisms (e.g., gravitational instability or rapid migration of larger bodies not currently resolved).
Methodological Advancement: The study validates the use of multi-wavelength polarization observations combined with MCRT modeling as a powerful tool to break degeneracies in dust grain size estimation, offering a new standard for characterizing protoplanetary disks.

Conclusion

The paper concludes that the PDS 70 disk, despite hosting two massive planets, is dominated by dust grains smaller than 100 µm. The observed polarization patterns and spectral properties indicate that the growth of ice-coated dust is limited by low fragmentation velocities ( $v_{frag} \lesssim 1$ m s $^{-1}$ ), supporting the hypothesis that bouncing and fragmentation act as significant barriers to planetesimal formation in the outer regions of protoplanetary disks.