🔭 astrophysics

Are most detected tidal disruption events partial?

Using hydrodynamic simulations of partial tidal disruption events, this study demonstrates that such events can produce observable optical and X-ray signatures similar to full disruptions, suggesting that many detected TDEs may actually be partial and explaining the properties of repeating partial TDEs.

Original authors: Megha Sharma (Monash University), Daniel J. Price (Monash University, Univ. Grenoble Alpes), Alexander Heger (Monash University), Katie Auchettle (University of Melbourne, University of California, Sa

Published 2026-06-23

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Megha Sharma (Monash University), Daniel J. Price (Monash University, Univ. Grenoble Alpes), Alexander Heger (Monash University), Katie Auchettle (University of Melbourne, University of California, Santa Cruz)

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 dance floor where a massive, invisible giant (a supermassive black hole) sits in the center, and a lone dancer (a star) tries to waltz past. Usually, if the dancer gets too close, the giant's gravitational pull is so strong it rips the dancer apart completely. This is what astronomers call a "Tidal Disruption Event" (TDE).

But this paper asks a simple, yet profound question: What if the dancer isn't ripped to shreds? What if they survive the encounter, just losing a few layers of their outfit?

The authors, a team of astrophysicists, used powerful computer simulations to test this scenario. They asked: Are most of the TDEs we see in the sky actually "partial" events where the star survives, rather than total destruction?

Here is the breakdown of their findings using everyday analogies:

1. The "Peeling the Orange" Analogy

Think of a star like an orange.

Full TDE: If the orange gets too close to the giant, it gets crushed into a complete pulp. Nothing is left but juice and seeds flying everywhere.
Partial TDE: If the orange gets close but not too close, the giant's gravity just peels off the top layer of the rind. The core of the orange (the star) survives and keeps dancing, but it leaves a trail of peels (debris) behind.

The paper simulates a star (the orange) passing a black hole (the giant) at different distances. They found that even when the star loses up to 50% of its mass (a lot of peeling), the core often survives.

2. The "Foggy Cloud" Effect

When the star gets peeled, the debris doesn't just vanish. It swirls around the black hole.

The Old Idea: Scientists used to think this debris would fall straight into the black hole, creating a bright, hot flash of X-rays (like a spotlight).
The New Finding: The simulations show that for many of these partial events, the debris moves so fast and piles up so thickly that it creates a giant, opaque "fog" or "reprocessing layer" around the black hole.
The Result: This fog blocks the bright X-rays from escaping. Instead, the energy gets trapped, heats up the fog, and glows brightly in visible light and ultraviolet (like a glowing lantern covered by a thick blanket). This explains why we see so many TDEs in optical light rather than X-rays.

3. The "Speed of the Fall"

When a star is fully destroyed, the debris falls back to the black hole at a predictable speed, like water draining from a bathtub (following a specific mathematical rule called $t^{-5/3}$ ).

The Paper's Discovery: In these partial events, the "drainage" is different. The debris falls back more slowly and steadily, following a different rule ( $t^{-9/4}$ ). It's like the water is trickling out through a clogged pipe rather than a wide-open drain.

4. The "Survivor" and the "Repeat Offenders"

One of the most exciting parts of the paper is what happens to the star that survives.

The Remnant: The core of the star doesn't just fly away; it stays trapped in a very stretched, oval-shaped orbit around the black hole.
The Loop: Because it's still trapped, it might swing back around and get "peeled" again in the future.
The Connection: The authors found that their simulations of these "peeling" events look very similar to a specific group of observed TDEs that repeat themselves (like ASASSN-14ko or AT2020vdq). These are the "repeat offenders" that flash, fade, and flash again.

5. The Big Conclusion: "Are We Misidentifying the Crime?"

The paper suggests that we might be misclassifying many cosmic events.

The Claim: Many events that astronomers currently think are "Total Destruction" (Full TDEs) might actually be "Partial Peeling" (Partial TDEs).
Why it matters: If we think a star was totally destroyed when it was only partially peeled, our calculations for how heavy the black hole is might be wrong. It's like trying to guess the weight of a person by looking at their shadow; if you don't know if they are wearing a heavy coat (the debris layer), you might guess their weight incorrectly.

Summary

The paper uses computer models to show that partial tidal disruption events are likely very common. Even when a star loses half its mass, it can survive, create a thick cloud of debris that glows in visible light, and potentially return to get "peeled" again. This helps explain why we see so many optical TDEs and why some of them seem to repeat. The authors conclude that the universe might be full of "surviving stars" rather than just "dead stars," and we need to rethink how we identify these cosmic collisions.

Technical Summary: "Are most detected tidal disruption events partial?"

Problem Statement
Tidal disruption events (TDEs) occur when a star passes close enough to a supermassive black hole (SMBH) to be ripped apart by tidal forces. While theoretical models often assume "full" disruptions where the star is completely destroyed, recent observations suggest that "partial" TDEs—where a stellar remnant survives—may be more common due to their larger cross-section. However, distinguishing between full and partial TDEs observationally is difficult, as current lightcurve classifications often rely on decay rates that can be ambiguous. Furthermore, existing numerical simulations of TDEs have predominantly focused on full disruptions or have not generated synthetic lightcurves to directly compare with observational data. This study addresses the gap in understanding the observational signatures of partial TDEs, specifically testing whether they can reproduce the optical/UV and X-ray properties of detected events, including the emerging class of repeating partial TDEs.

Methodology
The authors utilized the General Relativistic Smoothed Particle Hydrodynamics (GRSPH) code, Phantom, to simulate the disruption of a $1 M_\odot$ middle-aged main-sequence star by a $10^6 M_\odot$ Schwarzschild black hole.

Orbital Parameters: Simulations were conducted on zero-energy (parabolic) orbits with penetration factors ( $\beta \equiv r_t/r_p$ ) ranging from 0.6 to 1.6, corresponding to mass losses of $\le 50\%$ .
Stellar Model: A $1 M_\odot$ star was generated using the Kepler stellar evolution code and relaxed in a Minkowski metric before being placed on a geodesic orbit around the black hole.
Resolution and Techniques: The star was modeled with $10^6$ SPH particles. To maintain computational efficiency while resolving the debris stream, the authors employed particle splitting (using the splitpart module) to increase resolution in the debris stream post-disruption. The surviving stellar core was replaced by a softened point mass after 4 days to accelerate the simulation.
Thermodynamics: Two thermodynamic approaches were tested:
1. Adiabatic: Assuming all shock heating is trapped (appropriate for super-Eddington phases).
2. Isentropic: Assuming efficient radiative cooling (appropriate for sub-Eddington phases).
Synthetic Observables: Post-processing involved solving the radiative transfer equation to generate synthetic lightcurves and spectral energy distributions (SEDs). The authors assumed local thermal equilibrium, electron scattering opacity ( $\kappa \approx 0.34 \text{ cm}^2 \text{ g}^{-1}$ ), and applied a color correction factor ( $f_{col} = 1.7$ ) to account for deviations from a pure blackbody.

Key Results

Mass Fallback Rates: The simulations demonstrate that for $\beta \ge 0.8$ , the mass fallback rate can exceed the Eddington limit, leading to the formation of optically thick outflows and a reprocessing layer that obscures the accretion disc. The fallback rate decay is shallower than the canonical $t^{-5/3}$ law for full disruptions, tracking closer to the $t^{-9/4}$ power law predicted for partial TDEs.
Morphology and Outflows: Even in partial disruptions, stream-stream collisions at the apocentre drive kinetic outflows. In adiabatic simulations, these form a quasi-spherical envelope with temperatures of $\sim 10^4$ K, consistent with optical/UV emission. The size of the debris cloud and the reprocessing layer scales with $\beta$ .
Thermal Properties:
- Adiabatic Models: Yield optical bolometric luminosities of $\sim 10^{42}–10^{44} \text{ erg s}^{-1}$ , temperatures of $\sim 10^4$ K, and blackbody radii ranging from 10 to 100 au. These values align well with observed optical/UV TDEs.
- Isentropic Models: For $\beta = 0.8$ , the X-ray luminosity is $\sim 10^{44}–10^{45} \text{ erg s}^{-1}$ with radii smaller than the innermost stable circular orbit (ISCO), offering a potential explanation for X-ray TDEs with inferred radii below the Schwarzschild radius.
Comparison with Observations: The synthetic lightcurves and radii for the simulated partial TDEs overlap significantly with observed full TDEs and repeating partial TDEs (e.g., ASASSN-14ko, ASASSN-19dj, AT2020vdq). Specifically, lower $\beta$ values ( $\sim 0.6–0.8$ ) match the properties of repeating events, while higher $\beta$ values ( $\ge 1.0$ ) resemble full disruptions.

Significance and Claims
The paper argues that the observational properties of partial TDEs are sufficiently similar to full TDEs that many events currently classified as "full" disruptions may actually be partial.

Reclassification of Events: The results support the hypothesis that a significant fraction of detected TDEs are partial, consistent with theoretical cross-section arguments. The authors suggest that events currently identified as full TDEs could be re-evaluated as partial, particularly those exhibiting shallower decay rates or specific lightcurve morphologies.
Repeating TDEs: The study provides a physical basis for repeating partial TDEs, noting that the stellar remnants in their simulations ( $\beta \le 1.2$ ) remain bound on highly eccentric orbits, capable of undergoing recurrent disruptions.
Caution in Classification: The authors caution that misclassifying partial events as full TDEs in upcoming surveys (e.g., LSST) could lead to systematic biases in inferred black hole masses.
Limitations: The authors acknowledge that their simulations are limited to zero-energy orbits (whereas repeating TDEs likely involve bound, eccentric orbits), simplified equations of state, and a lack of detailed radiative transfer (recombination, detailed opacities). They note that their simulations converge only up to $\sim 100$ days post-disruption.

In conclusion, the study demonstrates that partial TDEs can produce the same optical/UV signatures and reprocessing layers as full TDEs, suggesting that the "partial" nature of a disruption is not always distinguishable by standard lightcurve analysis alone.