🔭 astrophysics

Efficient Interstellar Grain Growth from High Sticking Coefficients on Amorphous Carbon Dust

This paper demonstrates through molecular dynamics simulations and experimental validation that amorphous carbon dust grains exhibit high sticking coefficients for gas-phase atoms, confirming that accretion in the interstellar medium is an efficient mechanism for rapid dust growth on astrophysical timescales.

Original authors: Clarke J. Esmerian, Duncan Bossion, Francois Dulieu, Saoud Baouche, Alexey Potapov, W. M. C. Sameera, Tom J. L. C. Bakx, Susanne Aalto, Kirsten K. Knudsen, Gunnar Nyman, Wouter Vlemmings

Published 2026-07-17

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Clarke J. Esmerian, Duncan Bossion, Francois Dulieu, Saoud Baouche, Alexey Potapov, W. M. C. Sameera, Tom J. L. C. Bakx, Susanne Aalto, Kirsten K. Knudsen, Gunnar Nyman, Wouter Vlemmings

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 the universe not as a vast, empty void, but as a bustling cosmic city filled with a thick, invisible fog. This fog is the Interstellar Medium (ISM), a soup of gas and tiny solid particles called "dust." While this dust makes up only a tiny fraction of the universe's mass—about 1%—it is the stage upon which the cosmic drama plays out. Without these specks of dust, stars couldn't form, planets couldn't coalesce, and life as we know it would never get a foothold. For decades, scientists have been trying to solve a mystery: how do these tiny dust grains grow so big? We know they are born in the fiery outbursts of dying stars, but once they drift into the cold, quiet spaces between stars, they need to get bigger to survive. The big question has been: do they just sit there, or do they actively "eat" gas atoms floating by to grow?

The key to this growth is something called a "sticking coefficient." Think of it like a cosmic game of catch. If a gas atom flies toward a dust grain, does it bounce off like a rubber ball hitting a wall, or does it stick like Velcro? If it sticks, the grain gets a little heavier. If it bounces, the grain stays the same size. For a long time, scientists weren't sure how often this "Velcro" effect actually happened. They worried that maybe the atoms were too energetic or the dust too cold to ever hold on. This uncertainty made it hard to explain how galaxies could have so much dust so quickly after the Big Bang.

Now, a team of researchers has taken a fresh look at this problem using powerful supercomputers and real-world lab experiments. They focused on a specific type of dust grain made of "amorphous carbon"—essentially a chaotic, jumbled pile of carbon atoms, like charcoal rather than a perfect diamond. They wanted to see how well this carbon dust grabs onto the most common elements in the universe: hydrogen, carbon, oxygen, silicon, iron, and a few others.

What they found is a game-changer for our understanding of the cosmos. Using molecular dynamics simulations—which are like ultra-fast-motion movies of atoms colliding—they discovered that the "Velcro" is incredibly sticky. For almost every element they tested, the chance of an atom sticking to the dust grain was surprisingly high, often greater than 20% (a sticking coefficient of ≳0.2), and sometimes much higher. This holds true even when the gas is hot or the dust is cold. In fact, for many elements, the stickiness actually increases when the gas atoms are moving faster, because the extra energy helps them break through surface barriers and form new chemical bonds.

The team didn't just rely on computer models; they also went into the lab to test their ideas. They created a porous carbon dust analog and shot hydrogen atoms at it. The results matched their simulations perfectly, confirming that the dust really does grab onto atoms efficiently.

When they plugged these high sticking numbers into equations to calculate how fast dust grains grow, the results were startling. In the dense, cold clouds where stars are born, a tiny dust grain could double its mass in as little as 10,000 to 1,000 years. Even in the more diffuse, warmer parts of space, the growth time is still short on a cosmic scale—usually less than 100 million years. This is a blink of an eye compared to the billions of years the universe has existed.

This suggests that gas-phase accretion—dust grains eating gas atoms—is likely the main way dust grows in the universe, perhaps even more important than the dust being born in dying stars. It also implies that interstellar dust isn't just pure carbon or pure rock; it's likely a "mixed salad" of elements. As grains drift through space, they probably pick up oxygen, iron, silicon, and sulfur, becoming complex, multi-element particles. This new picture means that when astronomers look at the light from distant galaxies, they might need to rethink what those dust grains are actually made of. The universe isn't just building dust; it's actively recycling and remixing it, turning the cold gas of space into the solid seeds of future worlds.

Technical Summary: Efficient Interstellar Grain Growth from High Sticking Coefficients on Amorphous Carbon Dust

Problem Statement
Cosmic dust, comprising carbonaceous and silicate grains, is a critical component of the interstellar medium (ISM), influencing galactic observables despite constituting only ~1% of ISM mass. A major uncertainty in astrophysical models concerns the efficiency of dust growth via gas-phase accretion versus direct production from evolved stars. While indirect evidence (e.g., interstellar depletions) suggests gas-phase accretion is significant, quantitative models have been hindered by large uncertainties in "sticking coefficients" ( $S_i$ )—the probability that an impacting gas-phase atom adheres to a dust grain. Previous theoretical estimates were limited by idealized assumptions (e.g., uniform graphite lattices) and lacked predictive power for chemisorption of carbon or elements beyond Hydrogen and Carbon. Consequently, the relative contribution of gas-phase accretion to the total dust mass of the universe remains poorly constrained, particularly regarding whether grains can grow on timescales comparable to the age of the universe or the lifetimes of giant molecular clouds.

Methodology
The authors employed a multi-pronged approach combining atomistic molecular dynamics (MD) simulations with laboratory experiments:

Molecular Dynamics Simulations:
- System: Simulations modeled gas-phase atoms of cosmologically abundant elements (H, C, O, Al, Si, S, Fe, Ni) impacting amorphous carbon (a-C) grain surfaces.
- Potential: The Reactive Force Field (ReaxFF) algorithm was used to capture bond formation and breaking, distinguishing between chemisorption (covalent bonding) and physisorption (van der Waals interactions).
- Conditions: Simulations covered a wide range of gas temperatures ( $T_g \in [3, 10^4]$ K) and dust temperatures ( $T_D \in [3, 300]$ K), constrained by the Cosmic Microwave Background and the hottest dense ISM phases.
- Setup: 50 collisions were simulated per temperature combination on five different cubic a-C structures ( $14 \times 14 \times 14$ Å), ensuring statistical averaging over surface sites and angles.
- Analysis: Sticking coefficients were calculated based on binding events. For H, C, and O, asymptotic total chemisorption values were estimated by accounting for the probability of physisorbed atoms eventually chemisorbing over longer timescales, based on previous work [9].
Laboratory Experiments:
- Experiments measured the recombination coefficient of Hydrogen on a porous, amorphous carbon dust analog (produced in the Jena Dust Machine) under ultra-high vacuum.
- Surface temperatures of 10 K, 100 K, and 300 K were tested with an atomic H beam ( $T_g \approx 250 \pm 100$ K).
- The recombination coefficient served as a lower bound for the sum of physisorption and chemisorption sticking probabilities, providing an empirical check on the MD results.
Growth Rate Modeling:
- The calculated sticking coefficients were integrated into a grain growth timescale equation ( $\tau$ ) to estimate growth rates across different thermodynamic phases of the ISM (e.g., WNM, CNM, Dense H $_2$ ).

Key Results

High Sticking Coefficients: The simulations reveal high sticking coefficients ( $\gtrsim 0.2$ ) for all non-inert elements across all relevant gas and dust temperatures. Specifically, total chemisorption sticking coefficients are predicted to be greater than 0.2 for all elements.
Temperature Dependence:
- Sticking coefficients are primarily determined by gas temperature ( $T_g$ ), with no clear trend observed regarding dust temperature ( $T_D$ ).
- H, C, and O exhibit a non-monotonic dependence on $T_g$ , decreasing from $\sim 10$ K to $\sim 100-1000$ K before increasing again at higher temperatures.
- Other elements (Al, Si, S, Fe, Ni) show approximately monotonically increasing or constant sticking coefficients with increasing $T_g$ , generally exceeding 0.5.
- The increase in sticking at high temperatures is attributed to the ability of high-kinetic-energy atoms to dislodge existing bonds, creating free valence shells for new covalent bond formation.
Comparison with Previous Work: The results represent a qualitative improvement over 40-year-old perturbative quantum mechanical calculations [8], which assumed uniform graphite lattices, failed to predict carbon chemisorption, and incorrectly modeled the temperature dependence of sticking. The new results align well with recent theoretical estimates [9] where $T_g = T_D$ .
Experimental Validation: Laboratory recombination measurements for Hydrogen yielded lower limits for the combined sticking probability ( $S_{H,chem} + S_{H,phys}$ ) of $\ge 0.25$ at $T_D = 100-300$ K and $\ge 0.56$ at $T_D = 10$ K. These values are consistent with the MD predictions, lending confidence to the numerical results.
Grain Growth Timescales: Using the derived sticking coefficients, the characteristic grain growth timescale ( $\tau$ ) for a 50 Å grain is calculated to be less than 100 Myr for all ISM phases, dropping to $10^4$ – $10^3$ years in the densest phases. Even for larger grains (0.1 µm), growth timescales remain below 1 Gyr for most phases, excluding the most diffuse Warm Neutral Medium.

Significance and Claims
The paper claims that gas-phase accretion onto dust grains is likely an efficient mechanism for interstellar dust mass growth on astrophysical timescales. The findings suggest that:

Efficiency: Grain growth via accretion can occur on timescales comparable to or shorter than the lifetimes of giant molecular clouds and the age of the universe, supporting the hypothesis that a significant fraction of cosmic dust mass is acquired through this process rather than solely from stellar ejecta.
Composition: The high sticking coefficients for elements such as O, Fe, Si, S, Al, and Ni imply that "mixed" composition carbon-carrying grains are likely to form readily in the ISM. This challenges the standard two-species paradigm (separate amorphous hydrocarbon and silicate populations) and suggests models must account for a wider range of constituent materials.
Caveats: The authors note three primary caveats:
- Ice Mantles: In molecular clouds, ice mantles may inhibit accretion onto the refractory core, though the extent of this coverage is uncertain.
- Coulomb Repulsion: In diffuse phases, positive grain charges may repel ions, potentially reducing accretion rates for larger grains, though this is less prohibitive for smaller grains.
- Hydrogenation: The simulations used pure a-C, which is maximally reactive. Real ISM grains (a-C:H) may have lower sticking coefficients due to hydrogen saturation, but radiative desorption and other processes likely de-hydrogenate grains, allowing for the accretion of heavier elements over cosmic time.

The study concludes that the process of gas-phase accretion is plausibly central to the evolutionary life-cycle of interstellar grains at all cosmic epochs, necessitating updates to models of dust composition and observable properties.

More like this