🔬 physics

Polaris: a flexible stellarator demonstration experiment with simple modular coils

This paper presents the design, construction, and initial plasma results of Polaris, a novel, flexible small-scale stellarator at the Swiss Plasma Center featuring a glass vacuum vessel and interchangeable modular coils to serve as a unique testbed for investigating fundamental stellarator edge physics.

Original authors: Simon P. H. Vincent, Joaquim Loizu, Matthieu Toussaint, Rémy Jacquier, Jérémy Salm, Philippe Guittienne, Matias Habib, Idil Sonmez, Christopher B. Smiet, Erol Balkovic, Rogerio Jorge, Alan G. Goodman

Published 2026-07-29

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Simon P. H. Vincent, Joaquim Loizu, Matthieu Toussaint, Rémy Jacquier, Jérémy Salm, Philippe Guittienne, Matias Habib, Idil Sonmez, Christopher B. Smiet, Erol Balkovic, Rogerio Jorge, Alan G. Goodman, Ivo Furno, Christian Moura, William Matthey-Dorey, Steve Couturier, Frédéric Dolizy

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 Great Magnetic Bubble Hunt

Imagine trying to hold a drop of water in your hand without it slipping through your fingers. Now, imagine that drop of water is actually super-hot gas, hotter than the center of the sun, and you are trying to keep it from touching anything solid, because if it does, it will melt your container instantly. This is the ultimate challenge of fusion energy: creating a star on Earth to power our future. To do this, scientists use invisible magnetic fields to create a "bottle" that traps the gas.

There are two main ways to build this magnetic bottle. One way is like a donut with a current running through the gas itself, twisting the magnetic field into a spiral. The other way, which is the focus of this story, is called a stellarator. Think of a stellarator as a complex, twisted pretzel made entirely of magnetic fields. Instead of relying on the gas to help twist the field, the machine uses a set of specially shaped magnets to do all the heavy lifting. This makes the stellarator incredibly stable and safe, but it also makes it very hard to design. The magnets have to be shaped with such perfect precision that even a tiny mistake can let the gas escape. For a long time, building these twisted pretzels was like trying to sculpt a masterpiece out of wet clay while wearing oven mitts—expensive, difficult, and not very flexible.

The Polaris Experiment: A "Fish Tank" for Stars

Enter Polaris, a new, small-scale experiment built at the Swiss Plasma Center. The team behind Polaris decided to try something radically different: instead of building a machine where the magnets are hidden inside a solid metal box, they built a giant, transparent glass fish tank and put the magnets inside it.

The goal was to create a "stellarator fish-tank"—a device so flexible that scientists could swap out the magnets like LEGO bricks to test different magnetic shapes without having to rebuild the whole machine. The paper describes the design, construction, and first successful plasma experiments of this unique device.

The Glass Tank and the Swappable Magnets
The vacuum vessel (the container holding the empty space) is made mostly of thick glass windows. This is a big deal because it lets scientists see the entire plasma from every angle, something that is usually impossible in metal-walled machines. Inside this glass tank, they installed six identical, circular copper coils. These coils are water-cooled and arranged in a specific pattern to twist the magnetic field into a stable, donut-like shape.

The researchers designed these coils using a computer method called "guided coil optimization." They wanted to see if they could get a good magnetic bottle using just six simple, round coils, rather than the dozens of weirdly shaped magnets usually required. The result was a magnetic field that successfully trapped plasma, creating a volume of magnetic surfaces about the size of a large bathtub (0.05 m³).

First Plasma: Lighting Up the Tank
Once the machine was built, the team turned on the power. They used a radio-frequency (RF) antenna, essentially a giant underwater heater, to zap the gas inside the tank and turn it into plasma. They successfully ignited plasma using argon, neon, and helium gases.

The results were exciting:

The Shape: The glowing plasma took on the exact twisted shape predicted by their computer models. You could see "lobes" of plasma forming between the coils, just like the theory said.
The Stability: Even though the magnetic field was created by simple, round coils, it held the plasma together quite well. The plasma density reached about $8 \times 10^{16}$ particles per cubic meter, with temperatures between 2 and 6 electron-volts (eV). While this is "cold" compared to a fusion reactor (which needs millions of degrees), it is hot enough to study the physics of how plasma behaves at the edges of larger machines.
The "Fish Tank" Advantage: Because the tank is glass, they could take photos and videos of the plasma from the top and sides, watching it glow and move in 3D.

Testing the Limits: What Happens When Things Go Wrong?
One of the most interesting parts of the paper is how the team tested the machine's flexibility and robustness.

The Wiggle Room: They simulated moving the coils slightly (by up to 1 cm) to see if the magnetic bottle would break. The computer models showed that the magnetic field is surprisingly tough; it can handle these small mistakes without losing the plasma. This suggests that building these machines doesn't require the impossible precision of a watchmaker, but rather the "good enough" precision of a carpenter.
The Big Mistake: To really test the system, they physically rotated one of the six coils by 180 degrees. This was a huge change, effectively breaking the perfect symmetry of the magnetic field. The result? The plasma didn't disappear, but it did get worse. The confinement time (how long the plasma stays trapped) dropped by about 30%. This proved that while the magnetic optimization helps a lot, the plasma can still survive in a "messy" field, though not as efficiently.

What They Found and What's Next
The paper concludes that Polaris is a success. It proved that you can build a flexible stellarator with simple, interchangeable coils inside a glass vessel. The plasma behaves as expected, filling the magnetic "cage" and showing signs of turbulence (wobbles in the plasma) that are similar to what happens in the edges of massive fusion reactors.

The authors suggest that this machine is a perfect "testbed" for studying the edge physics of stellarators—how the hot plasma interacts with the cooler gas and the walls of the machine. Because the plasma in Polaris is relatively cool and dense, it is dominated by collisions between particles, which is a regime that is hard to study in giant, super-hot machines.

Looking ahead, the team plans to use the glass windows to take even faster pictures of the plasma to study waves and instabilities. They also mention a future dream: upgrading the copper coils to superconducting ones to create even stronger magnetic fields. For now, however, Polaris stands as a shining example of how a simple, flexible, and transparent design can open up new windows into the complex world of fusion energy.

Technical Summary of "Polaris: a flexible stellarator demonstration experiment with simple modular coils"

Problem and Motivation
Stellarators are re-emerging as viable concepts for future fusion power plants, driven by advanced optimization techniques that yield high-performance magnetic configurations. However, there is a recognized need for university-scale stellarator experiments to serve as testbeds for basic plasma physics, model validation, and education. Existing small-scale stellarators often suffer from limited flexibility in their magnetic configurations, utilizing complex helical coils, interlinked coils, or large numbers of permanent magnets. Furthermore, the experimental space for configurations that achieve good neoclassical confinement with low coil complexity remains largely unexplored. The authors aim to address this by constructing a highly flexible, small-scale stellarator capable of testing simple coil configurations and providing unprecedented optical access for diagnostics and outreach.

Methodology and Experimental Design
The authors designed and constructed Polaris, a small-scale stellarator (major radius $R \approx 0.4$ m) located at the Swiss Plasma Center. The device is defined by three primary design choices:

Optimized Magnetic Configuration: Using the "guided coil optimization" method (SIMSOPT), the authors designed a configuration using only six identical, circular, water-cooled copper coils. The optimization targeted a specific volume of vacuum magnetic surfaces and a rotational transform ( $\iota \approx 0.3$ ) with a field periodicity of $N_{fp}=3$ . The design allows for a total current of $\sim 5$ kA per coil (currently operated at $\sim 150$ A continuous, up to 510 A pulsed), generating an on-axis magnetic field of $\sim 0.03$ T.
Glass Vacuum Vessel: To maximize flexibility and optical access, the vacuum vessel is predominantly constructed of glass windows (hexagonal base, six side windows, and a top window). This design allows the entire upper structure to be lifted, providing full access to the internal components. The vessel accommodates interchangeable sets of coils installed on replaceable aluminum base plates.
Plasma Generation: Plasma is produced via an inductively coupled plasma (ICP) source using a water-cooled RF antenna operating at 13.56 MHz with up to 2.5 kW of power. The antenna is housed inside the vacuum chamber and protected by a MACOR ceramic casing and a Faraday shield to prevent arcing and capacitive coupling dominance.

The experimental setup includes double Langmuir probes for measuring density ( $n_e$ ), electron temperature ( $T_e$ ), and floating potential ( $V_f$ ), as well as fast imaging capabilities enabled by the glass vessel.

Key Results
The paper presents the first plasma experiments and characterization of Polaris:

Plasma Formation and Shape: Argon, neon, and helium plasmas were successfully ignited. Visual observations confirm that the plasma shape closely follows the theoretical last closed magnetic surface, including the formation of lobes between coils. The plasma fills the volume defined by the magnetic surfaces despite the large magnetic mirror ratio ( $B_{max}/B_{min} \approx 10$ ).
Plasma Parameters: At a neutral pressure of $5 \times 10^{-3}$ mbar and coil current of 150 A, the plasma reaches densities of $\sim 8 \times 10^{16}$ m $^{-3}$ (in argon) with electron temperatures between 2 and 3 eV. The density peaks near the magnetic axis.
Fluctuations and Instabilities: Measurements of floating potential fluctuations reveal dominant low-frequency modes around 109 kHz and a broad peak near 50 kHz. These frequencies are consistent with drift or interchange instabilities, which are expected to dominate cross-field transport in this low-temperature, high-collisionality regime.
Confinement Time: The global energy confinement time is estimated to be on the order of milliseconds ( $\tau_2 \approx 1$ ms). A faster decay component ( $\tau_1 \approx 30-60$ $\mu$ s) is attributed to radiative losses from electron-neutral collisions.
Robustness and Perturbation Studies: The magnetic configuration was tested for robustness against coil displacement (up to 1 cm) and a deliberate major perturbation (rotating one coil by 180°). While the optimized configuration is robust to small errors, the 180° rotation significantly degraded confinement, reducing the confinement time by approximately 30% and altering density profiles. This demonstrates that while stellarator optimization improves confinement, it is not the sole factor in this collisional regime.

Significance and Claims
The authors claim that Polaris serves as a "stellarator fish-tank," offering a unique platform for:

Flexibility: The ability to easily interchange coil sets allows for the experimental exploration of various magnetic configurations, including those with simple coil geometries that are difficult to test in larger, fixed devices.
Optical Access: The all-glass vacuum vessel provides 360-degree visual access, enabling fast imaging diagnostics and serving as a powerful tool for scientific outreach by visualizing toroidal confinement without plasma current.
Edge Physics Testbed: The low-temperature, low-density, high-neutral-pressure plasmas generated in Polaris are relevant to the edge physics of larger stellarators. The device is positioned as a first-of-its-kind testbed for investigating fundamental stellarator edge phenomena, such as turbulent transport, radiation, and neutral-plasma interactions, in a regime dominated by collisions.

The paper concludes that while magnetic optimization plays an important role in confinement, the specific low-temperature regime of Polaris is heavily influenced by collisional transport and radiation, motivating further investigation into the interplay between magnetic configuration and turbulence in stellarator edge physics. Future work includes upgrading to superconducting coils and utilizing fast imaging to study toroidal modes.

The Great Magnetic Bubble Hunt

The Polaris Experiment: A "Fish Tank" for Stars

More like this