🔭 astrophysics

Dust Formation in Common Envelope Binary Interactions -- III. Lightcurves

This paper presents post-processing light-curve calculations for 3D hydrodynamic simulations of common envelope interactions between asymptotic giant branch stars and compact companions, revealing how dust formation drives luminous red nova-like transients characterized by an initial hot peak, a sharp decline, and a long-lived cool plateau that closely matches observed events such as OGLE-2002-BLG-360 and AT 2025abao.

Original authors: Chunliang Mu, Orsola De Marco, Luis C. Bermúdez-Bustamante, Ryosuke Hirai, Daniel J. Price, Lionel Siess, Miguel González-Bolívar, Mike Y. M. Lau, Nadejda Blagorodnova

Published 2026-06-26

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Chunliang Mu, Orsola De Marco, Luis C. Bermúdez-Bustamante, Ryosuke Hirai, Daniel J. Price, Lionel Siess, Miguel González-Bolívar, Mike Y. M. Lau, Nadejda Blagorodnova

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

The Big Picture: A Stellar Dance Gone Wrong

Imagine two stars orbiting each other like a pair of ice skaters holding hands. One is a giant, bloated star (like an elderly red giant), and the other is a smaller, dense companion. In this paper, the authors simulate what happens when the giant star gets so big that the smaller star crashes into it, getting swallowed up inside the giant's outer atmosphere. This event is called a Common Envelope (CE) interaction.

Think of it like a small boat crashing into a massive, fluffy cloud of cotton candy. The boat (the companion) gets stuck inside the cloud, spiraling inward, while the cloud gets tossed around violently.

The goal of this paper is to figure out what this cosmic crash would look like to an observer on Earth. Specifically, they want to predict the lightcurve: a graph showing how bright the object gets, how its color changes, and how long the whole show lasts.

The Simulation: A Cosmic Movie

The authors used powerful computers to run a 3D movie of this crash. They simulated two different scenarios:

A smaller giant (1.7 times the mass of our Sun) crashing into a companion.
A larger giant (3.7 times the mass of our Sun) crashing into a companion.

They watched these simulations for 44 years (in computer time) to see how the light changed.

The Three Acts of the Show

The paper describes the light from this event in three distinct "acts":

Act 1: The Bright Flash (The Outburst)

What happens: As the companion star spirals in, it stirs up the giant star's outer layers. The whole thing expands rapidly, like a balloon being blown up suddenly.
The Light: This causes a massive, bright flash. The object becomes incredibly hot and bright (millions of times brighter than our Sun) for a few years.
The Catch: The authors admit this early flash might be a bit "fake" or exaggerated in their computer model because the computer didn't have enough "pixels" (resolution) to see the tiny details of the star's surface perfectly. It's like trying to take a photo of a fluffy cat with a low-resolution camera; the edges look blurry and brighter than they really are.

Act 2: The Dust Curtain (The Decline)

What happens: About 1 to 3 years after the crash starts, the gas cools down enough for dust to form. Imagine the steam from a hot shower suddenly turning into a thick, gray fog.
The Light: This dust acts like a heavy curtain. It blocks the hot, bright light from the center. The object suddenly gets much dimmer and turns very red (cool).
The Twist: Because the dust doesn't form evenly (it's thicker in some directions than others), the brightness changes depending on where you are looking. If you look from the "side," the dust might block the view sooner than if you look from the "top."

Act 3: The Cool Plateau (The Aftermath)

What happens: After the initial drop, the object settles into a long, steady glow. It's no longer a hot, blue star; it's a cool, red, dusty ball.
The Light: The object stays at a steady, moderate brightness for decades.
The Problem: The authors note that their computer model is "adiabatic," meaning it doesn't let heat escape easily. In reality, the object would cool down faster and get dimmer. In the simulation, it stays too bright for too long, like a heater that won't turn off. They estimate the real object would be dimmer than what the computer shows.

The Long-Term Future: When the Fog Clears

The paper predicts what happens hundreds of years later.

The dust cloud keeps expanding, getting thinner and thinner, like smoke dispersing in a large room.
100 to 200 years after the crash: The dust becomes thin enough to see through again.
The Reveal: Once the dust clears, we would see the hot, inner core of the merged stars shining through again, but this time it would be a different, warmer object than the original giant.

Comparing to Real Life

The authors compared their computer movies to real astronomical events they have actually seen, specifically:

OGLE-2002-BLG-360: A fainter event that took a long time to rise and fall.
AT 2025abao: A very bright, recent event that flared up quickly.

They found that the AT 2025abao event looks very similar to their simulation of the larger, 3.7-solar-mass star. Both had a quick, bright flash followed by a cooling phase. However, the OGLE-2002 event was much dimmer than their simulations predicted, suggesting it might have started from a smaller, less massive star than the ones they modeled.

The Bottom Line

This paper is a "first draft" of understanding what these stellar crashes look like.

Success: They successfully showed how dust forms quickly, hides the star, and creates a long, cool afterglow.
Limitations: The computer models have two main flaws:
1. Blurry Start: The beginning of the event is hard to calculate accurately because the computer needs more "pixels" to see the star's surface clearly.
2. Overheated End: The models don't let heat escape fast enough, so the object stays too bright in the later years compared to reality.

Despite these flaws, the study provides a crucial roadmap for astronomers to identify these events in the sky and understand the violent birth of new binary stars or planetary nebulae.

Technical Summary: Dust Formation in Common Envelope Binary Interactions — III. Lightcurves

Problem Statement
Luminous Red Novae (LRNe) are transient events resulting from common envelope (CE) binary interactions, yet predicting their observable lightcurves from hydrodynamic simulations remains challenging. Previous attempts to model these events often treated dust formation as an instantaneous process or relied on opacity tables that did not explicitly resolve the nucleation timescale. Furthermore, accurately bridging the gap between 3D hydrodynamic simulations and observational data requires robust post-processing radiative transfer methods that account for the complex, asymmetric, and evolving nature of the CE ejecta, particularly the formation of an optically thick dust shell.

Methodology
The authors perform post-processing light-curve calculations on two 3D smoothed particle hydrodynamics (SPH) simulations of CE interactions, originally described in Bermúdez-Bustamante et al. (2024). The simulations model the interaction between a compact companion ( $0.6 M_\odot$ ) and asymptotic giant branch (AGB) stars of two distinct masses: $1.7 M_\odot$ and $3.7 M_\odot$ . Crucially, these simulations include explicit carbon-based dust nucleation.

The methodology for deriving the lightcurves involves:

Radiative Transfer: Applying the radiative transfer equation along rays projected from the simulation domain to an observer. The specific intensity is calculated using a grey opacity approximation assuming local thermodynamic equilibrium (LTE).
Opacity Treatment: The total opacity is the sum of gas opacity (derived from MESA tables for $T > 1450$ K) and dust opacity (calculated via Mie theory for carbon dust). For temperatures below 1450 K, where dust forms, the gas opacity is set to a low constant value, allowing dust opacity to dominate.
Luminosity Calculation: The bolometric luminosity is derived by integrating the specific intensity over a $256 \times 256$ pixel grid. The source function is treated as a grey blackbody.
Resolution and Uncertainty Analysis: The authors explicitly quantify uncertainties arising from low surface resolution (where few SPH particles define the photosphere) and the adiabatic assumption (lack of radiative cooling). They introduce a resolution metric ( $N_{res}$ ) based on the number of particles contributing to the luminosity integration per ray to define error bars.

Key Results
The simulations, run for 44 years under adiabatic conditions, yield the following lightcurve characteristics:

Early Phase (Rise to Peak): Both models exhibit a bright, hot peak ( $L \sim 2-4 \times 10^6 L_\odot$ , $T_{eff} \sim 6000-7000$ K) occurring 1–5 years after the onset of Roche lobe overflow. This rise is driven primarily by the rapid expansion of the photosphere due to the inspiral and surface instabilities, rather than significant temperature changes. The timing and shape of the peak vary with viewing angle and progenitor mass; the more massive ( $3.7 M_\odot$ ) model shows a broader, more irregular peak due to a lower mass ratio ( $q=0.16$ ) and delayed dust formation.
Dust Formation and Decline: Dust nucleation begins approximately 1–3 years after the interaction starts. Shortly thereafter, the bolometric luminosity undergoes a sharp decline as the dust shell becomes optically thick, obscuring the hot gas photosphere. This is followed by a partial recovery and a plateau phase with an effective temperature of $\sim 400$ K.
Photospheric Evolution: The "geometric" photosphere (defined by $\tau = 2/3$ ) expands linearly, reaching $\sim 250$ au by the end of the simulation. However, the "observer" photosphere (derived from SED fitting) reveals a complex evolution where a cooler, dust-dominated component eventually outshines the hotter gas component.
Long-term Dispersion: Extrapolating the free expansion of the dust shell suggests that the dust will become optically thin at visible wavelengths between 100 and 200 years post-interaction, revealing the inner, warmer photosphere again.
Comparison with Observations: The authors compare their results to OGLE-2002-BLG-360 and AT 2025abao. The $3.7 M_\odot$ model provides a closer match to AT 2025abao in terms of progenitor mass and peak luminosity, though the observed rise time for AT 2025abao is significantly faster than the simulation predicts. The $1.7 M_\odot$ model shares some characteristics with OGLE-2002, particularly regarding dust mass and photospheric radius at late times.

Limitations and Uncertainties
The paper identifies two primary sources of uncertainty that limit the quantitative accuracy of the lightcurves:

Surface Resolution: The first 1–2 years of the lightcurves are heavily affected by insufficient SPH resolution at the stellar surface. The current resolution (1.3 million particles) results in only 1–2 particles defining the photosphere at $t=0$ , leading to large uncertainties in the initial luminosity and rise time. The authors note that achieving converged resolution would require hundreds of millions to billions of particles.
Adiabatic Assumption: The simulations are adiabatic, lacking radiative cooling. Consequently, the late-time plateau luminosity ( $\sim 10^5 L_\odot$ ) is likely overestimated by an order of magnitude compared to physical expectations, as the system cannot radiate away the excess internal energy.

Significance
This work represents a first step in bridging the gap between 3D CE hydrodynamic simulations with explicit dust nucleation and observable lightcurves. While the authors acknowledge that their models are not yet fully converged or physically complete (due to the adiabatic assumption), they successfully demonstrate that:

Dust formation in CE interactions is a non-instantaneous process that fundamentally alters the lightcurve morphology, creating a distinct decline and plateau phase.
The asymmetry of the interaction leads to significant viewing-angle dependencies in the observed lightcurve.
The timing of dust formation and the subsequent transition from a gas-dominated to a dust-dominated photosphere can be tracked, providing a framework for interpreting future LRN observations.

The paper concludes that while current limitations prevent precise quantitative matching to specific transients, the qualitative features of the lightcurves (peak, decline, plateau, and dust dispersion timescales) offer a valuable template for understanding the evolution of CE events and the formation of planetary nebulae. Future improvements will require adaptive particle refinement to resolve the surface and the implementation of radiative cooling to correct the late-time energetics.