Omnigenous umbilic stellarators

This paper introduces and optimizes a new class of "umbilic stellarators" featuring a unique high-curvature edge to achieve omnigenous confinement, demonstrates the feasibility of their coil designs, and proposes a method to convert existing tokamaks into finite-beta diverted stellarators using single-stage optimization.

The Gist

Imagine trying to hold a swirling, super-hot ball of gas (plasma) in place using only invisible magnetic ropes. This is the goal of fusion energy. For decades, scientists have used two main shapes for these magnetic cages: the Tokamak (a perfect, symmetrical donut) and the Stellarator (a twisted, knotted donut).

While the twisted Stellarator is great because it doesn't need a massive electric current running through the plasma to stay stable, it has a tricky problem: the magnetic ropes can get tangled in a way that lets particles leak out.

This paper introduces a new, clever way to design these twisted cages, called "Umbilic Stellarators." Here is the breakdown using simple analogies:

1. The "Umbilic" Shape: A Twisted Bracelet

The authors took inspiration from a specific type of bracelet or torus (a ring shape) that looks like a three-pointed star when you slice it.

• The Analogy: Imagine a smooth, round donut. Now, imagine pinching the edge of that donut to create a sharp, jagged ridge that spirals around it. If you follow that sharp ridge, you have to go around the donut three times before the ridge connects back to itself.
• The Paper's Claim: They call this an "umbilic" shape. By creating this sharp, high-curvature ridge on the edge of the plasma, they can control how the magnetic field lines behave right at the boundary.

2. The "Omnigenity" Problem: Keeping the Ball Inside

In a perfect magnetic cage, particles should bounce around forever without hitting the walls. In Stellarators, they often drift outward.

• The Analogy: Think of the particles as marbles rolling inside a bowl. In a standard Stellarator, the bowl might have a slight tilt, causing the marbles to roll out the side. "Omnigenity" is a fancy word for designing the bowl so perfectly that no matter which way the marbles roll, they stay trapped inside.
• The Paper's Claim: The researchers used a super-computer (called DESC) to design these "umbilic" shapes. They found that by forcing the edge of the plasma to have this sharp, spiraling ridge, they could create a magnetic field that keeps particles trapped very well, even without the magnetic field looking perfectly symmetrical. They call this "piecewise omnigenity"—meaning the trapping works in sections, like a puzzle, rather than being perfect everywhere at once.

3. The "Divertor": The Trash Can for the Plasma

Fusion creates waste (like helium ash) that needs to be removed from the center of the plasma without ruining the reaction.

• The Analogy: In a standard donut-shaped reactor, you need a special "trash can" (a divertor) to catch the waste. In Tokamaks, this is easy because the magnetic field naturally creates a "hole" (an X-point) where the waste flows out. In Stellarators, creating this hole is hard.
• The Paper's Claim: The sharp, high-curvature ridge of the umbilic stellarator acts like a natural guide for the waste. Even though it doesn't create a perfect "hole" like a Tokamak, the magnetic field lines naturally flare out near this sharp ridge. This makes it a great place to put a "trash can" to catch the waste. The paper shows that even if the plasma inside gets a little wobbly (fluctuates), this edge structure stays stable and keeps the waste flowing in the right direction.

4. The "Umbilic Coil": A Magic Ribbon

Building a machine with these complex, sharp-edged shapes usually requires incredibly complicated, curved metal coils that are hard to build.

• The Analogy: Instead of building a whole new, complex cage, the authors propose adding just one special "ribbon" (a coil) wrapped around an existing machine.
• The Paper's Claim: They tested this idea by taking the shape of an existing machine (the HBT-EP tokamak) and adding a single, helical coil around it.
  • Case A (Opposite Current): When the coil pushes against the plasma, it creates a ridge but no "trash can" hole.
  • Case B (Same Current): When the coil pulls with the plasma, it creates a ridge that could form a "trash can" hole (an X-point).
  • The Result: This simple addition turned a standard, round plasma shape into the complex, sharp-edged "umbilic" shape, effectively turning a simple machine into a more advanced one without rebuilding the whole thing.

Summary

The paper proposes a new way to build fusion reactors by twisting the edge of the plasma into a sharp, spiraling ridge (like a specialized bracelet). This shape naturally traps particles well and guides waste out of the system. Most importantly, they show you can achieve this by simply wrapping a single, special coil around an existing machine, potentially making advanced fusion reactors easier to build and modify.

Technical Summary

Technical Summary: Omnigenous Umbilic Stellarators

Problem Statement
The development of a Stellarator Fusion Power Plant (FPP) requires not only excellent core confinement but also a predictable and optimized magnetic field structure outside the Last Closed Flux Surface (LCFS). While axisymmetric tokamaks utilize divertors to separate high-temperature plasma from Plasma Facing Components (PFCs) and exhaust impurities, stellarators have historically struggled with edge topology. Existing stellarator divertor concepts (helical, resonant, and non-resonant) face challenges regarding design complexity, sensitivity to magnetic shear, and the difficulty of predicting stochastic field regions. Furthermore, most stellarator optimizations focus solely on the LCFS, neglecting the edge field line structure necessary for effective particle and heat exhaust. The authors identify a need to understand how boundary shaping influences edge fieldline topology to facilitate X-point and island divertor designs in stellarators.

Methodology
The paper introduces a new class of stellarator equilibria called "umbilic stellarators" (US), characterized by a single continuous high-curvature edge on the plasma boundary that wraps toroidally multiple times before closing on itself. The methodology involves:

1. Parametrization: The authors define an analytical parametrization for an umbilic-torus-like surface inspired by the "umbilic bracelet" (a deltoid curve cross-section). This shape features a sharp edge defined by a floor function in the poloidal angle, creating a ridge that closes after $n$ toroidal turns.
2. Approximation and Solving: Since fully spectral solvers like DESC (Dudt & Kolemen 2020) struggle with perfectly sharp edges, the authors approximate the umbilic boundary using a Fourier-Zernike spectral basis. They solve the ideal MHD force balance equation within this volume.
3. Optimization Strategy: A two-stage optimization process is employed using the DESC suite:
   • Stage One: Simultaneously optimizes the plasma boundary shape and a specific curve lying on the boundary (the umbilic edge). The objective function minimizes aspect ratio, enforces a target rotational transform ($\iota \approx n_{FP}m/n$), ensures omnigenity (via effective ripple minimization), and imposes a high negative principal curvature ($\kappa_{2,\rho}$) along the umbilic curve.
   • Stage Two: For vacuum cases, modular coils are designed using REGCOIL to minimize the normal magnetic field error ($B \cdot \hat{n}$) on the fixed boundary.
   • Single-Stage Optimization: For finite-$\beta$ cases involving coil modifications, the authors perform a single-stage optimization that simultaneously varies the plasma boundary and the shape/current of an added "umbilic coil" to minimize normal field errors and pressure discontinuities.
4. Sensitivity Analysis: To test robustness, a line current source is introduced at the magnetic axis to emulate plasma fluctuations, and field line tracing is performed to assess the stability of the edge topology.

Key Contributions and Results

• Definition of Umbilic Stellarators: The paper defines and generates the first set of omnigenous umbilic stellarators. These configurations naturally favor "piecewise omnigenity" (Velasco et al. 2024) rather than omnigenity with a specific helicity, achieving low neoclassical transport without enforcing quasisymmetry.
• Vacuum Equilibrium (US131): A single-field-period ($n_{FP}=1, n=3, m=1$) vacuum stellarator was optimized. The resulting boundary exhibits a high-curvature ridge where the magnetic field lines align closely. Despite the lack of specific helicity in the Boozer plots, the effective ripple ($\epsilon_{eff}^{3/2}$) remains below 2% for most of the volume. Modular coil designs were generated, showing that while the coils must assume high curvature to match the ridge, the resulting field line structure "flares" near the edge, suggesting a favorable location for divertors.
• Finite-$\beta$ Equilibrium (US252): A two-field-period ($n_{FP}=2, n=5, m=2$) finite-$\beta$ ($\beta=0.5\%$) stellarator was generated. This configuration achieved even lower ripple transport ($\epsilon_{eff}^{3/2} < 0.001$) in the core. The optimization demonstrated that high curvature can be maintained along the umbilic curve while the rotational transform is partially derived from plasma shaping, reducing reliance on external current drive.
• Edge Robustness: Sensitivity analysis on the US131 equilibrium revealed that while internal fluctuations (simulated by an on-axis current) cause the inner flux surfaces to become stochastic, the field line structure at the edge remains resilient. The "flaring" behavior near the high-curvature ridge persists, suggesting that divertor placement would not require significant modification under such perturbations.
• Umbilic Coil Application: The authors propose and demonstrate a method to convert an existing tokamak (specifically the HBT-EP experiment) into a diverted stellarator by adding a single helical "umbilic coil."
  • In the counter-I case (current opposite to plasma), the coil creates a high-curvature ridge and a stochastic layer but does not form an X-point.
  • In the co-I case (current in the same direction), a single-stage optimization simultaneously modified the plasma boundary and the coil shape. This resulted in a configuration where the umbilic coil contributes to the edge rotational transform and creates a ridge capable of potentially forming an X-point, effectively converting the limited tokamak into a diverted stellarator.

Significance and Claims
The paper claims that umbilic stellarators offer a novel approach to controlling edge magnetic topology. By imposing high curvature along a specific curve on the boundary, the authors demonstrate that it is possible to create a continuous divertor option with long connection lengths without relying on the complex stochastic regions of traditional helical divertors or the precise island structures of resonant divertors.

The authors modestly claim that while these configurations do not produce a perfectly sharp X-point or X-line like a tokamak, the high-curvature edges serve as favorable locations for divertors and limiters. The technique of simultaneously optimizing a curve and a surface to control edge sharpness is presented as a generalizable method that could be applied to design ridges for non-resonant divertors or to fine-tune rotational transform for island divertors. The proposed modification of the HBT-EP experiment serves as a proof-of-concept for converting limited tokamaks into diverted stellarators using a single helical coil, potentially offering a simpler pathway to stellarator-like exhaust solutions.