🔬 physics

Spacecraft heat shield study in the DIII-D tokamak

This paper reports a new experimental platform at the DIII-D tokamak that utilizes high-heat-flux carbon ablation tests to validate aerospace heat shield models and study material response under conditions simulating both fusion divertor environments and high-enthalpy atmospheric entry.

Original authors: Dmitri M. Orlov, Evdokiya G. Kostadinova, Igor Bykov, Dmitri L. Rudakov, Roman Smirnov, Jayson Barr, Gabrielle Bladon, Alessandro Bortolon, Justin Burzachiello, Lane Carlsson, Colin Chrystal, Jason Es

Published 2026-07-28

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Dmitri M. Orlov, Evdokiya G. Kostadinova, Igor Bykov, Dmitri L. Rudakov, Roman Smirnov, Jayson Barr, Gabrielle Bladon, Alessandro Bortolon, Justin Burzachiello, Lane Carlsson, Colin Chrystal, Jason Escalera, Jessica Eskew, Graeson Griffin, Michael O. Hanson, Georg Herdrich, Jeffrey Herfindal, Al Hyatt, Truell Hyde, Charles Lasnier, Claudio Marini, Lorin Matthews, Adam McLean, Christopher A. Mehta, Renato Perillo, Jens Schmidt, Filipo Scotti, Zola Spence, Hadith Taheri, Michael van Zeeland, Caitlyn Villareal, Huiqian Wang, Robert Wilcox, Theresa Wilks, Nandini Yadav, Daniel Zubovic

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 you are trying to build a shield for a spaceship that needs to dive into the atmosphere of a giant planet like Jupiter. The air there isn't just hot; it's a super-charged, super-fast soup of particles that can melt almost anything instantly. To survive, the spaceship needs a "heat shield" made of special materials that don't just melt, but slowly burn away in a controlled way, carrying the heat with them. This process is called "ablation." It's like a snowball rolling down a hot hill: the snow melts and flies off, but the core stays cool enough to keep rolling. Scientists have been trying to predict exactly how fast these shields melt and if they might suddenly shatter into pieces (a dangerous event called "spallation"). The problem is, it's incredibly hard to recreate the exact, violent conditions of a space dive in a normal lab. You can't easily make a wind tunnel that is hot enough, fast enough, and filled with the right kind of "plasma" (super-hot gas) to test these shields properly.

This is where a giant donut-shaped machine called a tokamak comes in. Think of a tokamak as a super-powerful oven that traps super-hot gas using giant magnets. Usually, scientists use these machines to try to create clean energy by fusing atoms together. But in this story, a team of researchers decided to use the tokamak's extreme heat to test heat shield materials instead. They wanted to see if they could recreate the "shock layer"—that super-hot, high-pressure bubble of air that forms in front of a speeding spacecraft—and use it to see how carbon-based heat shields really behave. If they can get it right, they can build better shields for future missions to the outer planets, where the heat is so intense that even our best guesses might be wrong.

The Experiment: A Donut Oven and a Space Probe

In this study, researchers at the DIII-D National Fusion Facility in San Diego turned their tokamak into a cosmic testing ground. They wanted to see how carbon samples would hold up against heat fluxes (the amount of heat hitting a surface) as extreme as 30 to 40 million watts per square meter. To put that in perspective, that's enough heat to melt through a thick steel plate in a split second. They focused on two main ways to test their materials, using two different "characters" in their experiment: stationary carbon rods and flying carbon pellets.

First, they stuck some carbon rods into the edge of the plasma, right near where the heat is most intense. They tried different shapes: some were round like pencils, some were wedge-shaped like the nose of the famous Galileo space probe, and some were concave (curved inward). They left these rods in the plasma for about 4.5 seconds. After the experiment, they took the rods out and used a super-precise 3D scanner to see how much they had melted away. They found that the rods lost mass at a rate of about (1 to 3) × 10⁻² grams per square centimeter per second. Interestingly, the wedge-shaped rods melted away much faster than the round ones. The scientists compared their results to two famous computer models used by aerospace engineers (the Park model and the Matsuyama model). They found that the Park model was a pretty good guess for the round rods, but both models struggled to predict just how fast the sharp-edged wedge rods would melt. It seems the sharp edges concentrated the heat in a way the old models didn't fully account for.

Next, the team got more adventurous. Instead of sticking rods in place, they launched tiny carbon pellets (about the size of a grain of sand or a small pea) straight up into the heart of the plasma. These pellets traveled from the cooler edge all the way into the super-hot core, completely ablating as they went. This was like watching a snowball fly through a furnace and disappear. By using high-speed cameras and infrared sensors, they tracked the pellets' paths and how fast they vanished. They discovered that the pellets didn't just fly in a straight line; they were pushed and pulled by invisible magnetic forces (specifically the $j \times B$ force) that only become important for objects this size. If they ignored these forces in their computer simulations, the pellets would hit the wall too early. But when they added the magnetic push, the simulations matched the real-life flight paths perfectly.

The Big Surprise: When Shields Shatter

The most exciting and slightly scary discovery came from the pellets made of "glassy carbon." This material is hard and brittle, kind of like a piece of glass. As these pellets flew through the plasma, the researchers saw something unexpected: tiny fragments were flying off the surface, not just melting away. This is called "spallation." In the past, scientists thought spallation only happened if the heat was above a certain threshold (around 146 MW m⁻²). But here, the heat was lower than that, yet the glassy carbon still shattered. This suggests that the material didn't just melt; it cracked and broke apart due to the stress of heating up so quickly. This is a crucial finding because if a heat shield shatters, it loses its protection much faster than if it just melts.

The team also noticed that as the pellets melted, they created a cloud of gas around themselves. This cloud acted like a shield, protecting the pellet from some of the heat and slowing down the melting process. Their computer models, which included this "cloud shielding" effect, matched the real-world data much better than models that ignored it.

What This Means

This paper doesn't claim to have solved the mystery of space heat shields forever. Instead, it suggests that tokamaks are a fantastic new tool for testing them. The researchers showed that they can recreate the intense heating of a planetary entry and measure exactly how materials react. They proved that old models need to be updated to account for sharp shapes and that materials can shatter even when the heat isn't "supposed" to be high enough to do so. By combining real experiments with advanced computer simulations, they are building a better roadmap for designing heat shields that can keep future space probes safe as they dive into the fiery atmospheres of gas giants like Jupiter. It's a step forward in turning the extreme heat of a fusion reactor into a safety net for the next generation of space explorers.

Technical Summary: Spacecraft Heat Shield Study in the DIII-D Tokamak

Problem Statement
Atmospheric entry into planets with thick atmospheres, such as Jupiter, subjects spacecraft to extreme heating environments characterized by high heat fluxes, strong plasma flows, and high stagnation pressures. The Galileo probe's entry into Jupiter in 1995 serves as a critical case study, where peak heat fluxes reached 300 MW m⁻² and shock layer temperatures peaked near 16,000 K. Post-flight analysis of the Galileo heat shield revealed significant discrepancies between predicted and measured mass loss rates; models overestimated ablation at the stagnation point while underestimating it in downstream regions. Furthermore, approximately 2.9% of the probe's mass loss was attributed to mechanical spallation (ejection of material fragments), a failure mode difficult to predict and replicate in ground-based laboratory tests. Current thermal protection system (TPS) designs rely on large safety margins to account for these uncertainties, increasing launch mass and reducing scientific payload capacity. There is a need for experimental platforms capable of replicating the specific combination of high parallel heat flux and strong plasma flows found in planetary entry shock layers to validate and refine ablation models.

Methodology
The authors utilized the DIII-D National Fusion Facility to create a unique experimental platform for studying carbon ablation under conditions relevant to both fusion reactor divertors and high-enthalpy atmospheric entries. The study employed two complementary approaches using the Divertor Material Evaluation System (DiMES):

Stationary Carbon Rods: Carbon rods (ATJ graphite, some SiC-coated) with varied geometric profiles (cylindrical, wedge-shaped, and concave) were inserted into the scrape-off layer (SOL) near the divertor strike point. These rods were exposed to parallel heat fluxes of 30–40 MW m⁻² for approximately 4.5 seconds per discharge.
Slow-Launch Carbon Pellets: Spherical carbon pellets (porous "tidal" graphite and dense glassy carbon) were launched vertically from the DiMES port into the edge and core plasma. These pellets traversed the separatrix, experiencing heat fluxes up to ~100 MW m⁻² in the core.

Diagnostics and Analysis
The experiments were supported by a comprehensive suite of diagnostics:

Imaging: Visible fast-framing cameras and infrared (IR) thermography provided real-time data on surface temperature evolution and ablation dynamics.
Spectroscopy: The Multichordal Divertor Spectrometer (MDS) monitored carbon emission (specifically C2 Swan bands) to determine ablation onset and duration.
Plasma Characterization: Divertor Thomson Scattering (DTS) and Langmuir probes measured electron temperature, density, and parallel heat flux to reconstruct the plasma environment.
Interferometry: CO₂ laser interferometry measured line-integrated electron density to infer mass loss rates for moving pellets.
Post-Exposure Analysis: Optical 3D surface profilometry and precision mass measurements quantified surface recession and total mass loss.
Modeling: The UEDGE fluid code was coupled with the DUSTT 3D Monte-Carlo dust transport code to simulate pellet trajectories, accounting for gravity, ion drag, and electromagnetic ( $j \times B$ ) forces. Ablation rates were compared against semi-empirical models by Park and Matsuyama.

Key Results

Stationary Rod Ablation:
- Measured mass loss rates were in the range of $(1–3) \times 10^{-2}$ g cm⁻² s⁻¹.
- The wedge-shaped rod exhibited the highest ablation rate, significantly exceeding predictions, likely due to geometry-dependent heat flux concentration at sharp leading edges.
- Cylindrical and concave rods showed lower, more uniform ablation rates.
- Comparison with semi-empirical models indicated that the Park model provided a closer baseline approximation for cylindrical and concave geometries (within 20–40% of measurements), whereas the Matsuyama model systematically underpredicted ablation rates by 20–50%.
Pellet Dynamics and Modeling:
- Stereoscopic imaging allowed for the reconstruction of 3D pellet trajectories.
- Simulations demonstrated that the inclusion of the $j \times B$ force was critical for millimeter-scale conducting pellets; omitting this force resulted in premature wall loss, whereas including it allowed pellets to penetrate deeper into the core, matching experimental observations.
- Ablation cloud shielding was found to be a dominant factor. Simulations including radiation-cloud shielding predicted significantly reduced peak mass-loss rates and extended pellet lifetimes compared to unshielded models, consistent with interferometric data.
Spallation Observation:
- Spallation (ejection of micron-scale fragments) was observed in glassy carbon pellets, despite local parallel heat fluxes (~100 MW m⁻²) remaining below the nominal steady-state spallation threshold of 146 MW m⁻² derived from Galileo data.
- This suggests that transient thermal stress gradients and the brittle, cross-linked microstructure of glassy carbon can lower the threshold for mechanical fracturing compared to steady-state predictions.

Significance and Claims
The paper establishes the DIII-D tokamak edge plasma as a unique high-enthalpy test environment capable of validating ablation models for carbon-based protective materials. The study claims to provide the first systematic investigation of carbon ablation in a fusion environment specifically aimed at informing TPS modeling for giant-planet entries.

Key contributions include:

Validation of Models: Providing experimental data to refine semi-empirical ablation models (specifically highlighting the utility of the Park model for blunt geometries and the need to account for geometric effects for wedges).
Physics of Pellet Transport: Demonstrating the necessity of including electromagnetic ( $j \times B$ ) forces and ablation-cloud shielding in transport models for millimeter-scale particles in magnetized plasmas.
Spallation Mechanisms: Identifying that spallation can occur under heat flux conditions below traditional steady-state thresholds, driven by transient thermal stresses and material microstructure.
Platform Utility: Establishing a pathway to integrate controlled plasma experiments, real-time diagnostics, and validated simulations to improve predictive capabilities for future planetary entry missions, potentially reducing the safety margins and mass of thermal protection systems.

The authors note that while the tokamak environment replicates the plasma temperature, density, and heat flux of a gas giant entry shock layer, it does not simulate the high stagnation pressures (up to 7 atm) encountered during flight. Consequently, the platform focuses on validating thermochemical ablation, ionization effects, and plasma-surface interactions rather than mechanical pressure loading.

More like this