Resonant magnetic perturbations (RMPs)

Overview

Resonant magnetic perturbations (RMPs) are small, spatially varying magnetic fields intentionally applied to magnetically confined plasmas, primarily in tokamaks. These fields are non-axisymmetric, meaning they break the toroidal symmetry of the main confining field. Their primary purpose in modern fusion research is the control of edge-localized modes (ELMs), which are large, quasi-periodic bursts of energy and particles from the edge of high-confinement mode (H-mode) plasmas. While ELMs are a natural feature of H-mode operation, their transient heat loads pose a significant threat to the integrity of plasma-facing components (PFCs) in large-scale devices like ITER. The energy released by a single ELM in ITER is projected to be large enough to cause substantial erosion or melting of the tungsten divertor targets, severely limiting their operational lifetime [1].

RMPs offer a promising solution by altering the magnetic topology at the plasma edge. They are designed to be "resonant" with the helical structure of the magnetic field lines on specific rational magnetic flux surfaces. This interaction creates a network of small magnetic islands or a chaotic, stochastic layer in the pedestal region. This modified magnetic structure increases particle and heat transport across the plasma edge, effectively creating a continuous, gentle release of energy that prevents the edge pressure gradient from reaching the instability threshold that triggers a large ELM. This technique is a leading candidate for the primary ELM control strategy for ITER and future fusion power plants.

Physics / Mechanism

The fundamental principle of RMPs lies in the interaction between an externally applied magnetic perturbation and the plasma's intrinsic magnetic field structure. In a tokamak, magnetic field lines spiral around the torus on nested flux surfaces. The helicity of this spiral is characterized by the safety factor, q, which is the ratio of toroidal turns to poloidal turns a field line makes. A magnetic perturbation is considered resonant when its helical structure, defined by its poloidal (m) and toroidal (n) mode numbers, matches the helicity of the field lines on a flux surface where q = m/n.

External coils installed inside or outside the vacuum vessel generate these perturbations. By driving currents through these coils, a specific spectrum of (m, n) modes can be created. When this resonant field penetrates the plasma, it can cause magnetic reconnection at the corresponding rational q surface, forming a chain of magnetic islands. If the amplitude of the perturbation is sufficient, islands from adjacent rational surfaces can grow and overlap. According to the Chirikov criterion, this island overlap leads to the formation of a stochastic magnetic field layer, where field lines wander chaotically in the radial direction [2].

This stochastic layer at the plasma edge enhances radial transport. Instead of being well-confined until the pressure gradient collapses in an ELM, particles and heat can now leak out more readily along the chaotic field lines. This process, often called "density pump-out," reduces the pedestal pressure and density, keeping it below the critical value for the peeling-ballooning instability responsible for Type-I ELMs. The result is either complete ELM suppression or mitigation, where large ELMs are replaced by smaller, more frequent, and less harmful instabilities.

The plasma's response to the applied field is a critical aspect of the physics. Plasma rotation can induce screening currents that oppose the penetration of the RMP field, reducing its effectiveness. Successful ELM suppression is often achieved only within specific operational windows of plasma parameters like density, rotation, and collisionality, where the plasma screening is minimized, allowing the perturbation to penetrate to the top of the pedestal [3].

Historical development

The concept of using external magnetic perturbations to influence plasma stability has roots in early fusion research on tearing modes and error field correction. However, their specific application for ELM control is a more recent development. The pioneering experiments that demonstrated complete suppression of Type-I ELMs using RMPs were conducted on the DIII-D tokamak in 2003 [4]. Researchers at General Atomics used a set of internal coils (I-coils) to apply an n=3 perturbation to H-mode plasmas, successfully eliminating ELMs while maintaining good energy confinement.

This landmark result spurred a global research effort to understand the underlying physics and replicate the findings on other devices. Subsequent experiments on tokamaks such as JET (Joint European Torus) in the UK, ASDEX Upgrade in Germany, KSTAR in South Korea, and EAST in China have all explored RMP-based ELM control. These experiments have been crucial in building a database to characterize the operational window for ELM suppression and to test theoretical models of plasma response and transport. For instance, work on JET with its ITER-like wall confirmed that RMPs could mitigate ELMs in plasmas with metallic PFCs, a crucial finding for future reactors [5]. The development of sophisticated 3D magnetohydrodynamic (MHD) codes like JOREK and M3D-C1 has been essential for modeling the complex interaction between the RMP fields and the plasma response, helping to interpret experimental results and predict RMP performance in future devices like ITER.

Current status

As of 2026, RMPs are the baseline and most developed method for ELM control in ITER. The ITER design includes a set of 27 in-vessel coils specifically for this purpose. The international fusion community continues to refine the technique on existing tokamaks to prepare for ITER's operational phase. Research focuses on several key areas:

1. Optimizing the RMP Spectrum: Experiments systematically vary the toroidal mode number (n=1, 2, 3, 4) and the poloidal spectrum of the applied fields to find the most effective configurations for ELM suppression with minimal confinement degradation. For ITER, n=3 or n=4 perturbations are considered the most likely candidates [1].

2. Understanding Plasma Response: Advanced diagnostics and modeling are used to study how the plasma screens and amplifies the applied fields. This is crucial for predicting the required coil currents and optimal plasma conditions (e.g., rotation, density) for achieving suppression in ITER.

3. Impact on Confinement and Impurities: A primary trade-off of RMPs is a potential reduction in energy confinement, typically around 10-20%, and a significant drop in pedestal density (density pump-out). Current research aims to minimize these effects. There is also active investigation into how RMPs affect the transport of impurities, particularly the tungsten eroded from the divertor, as enhanced impurity accumulation in the core could degrade plasma performance.

4. Integration with other Scenarios: RMPs are being tested in various operational scenarios, including those with high density and low torque injection, which are relevant for a reactor. The compatibility of RMPs with other control systems is also under investigation.

Notable implementations

• ITER: The international ITER project has fully integrated RMPs into its design as the primary ELM control system. A set of 27 coils (9 upper, 9 equatorial, 9 lower) is being installed inside the vacuum vessel to provide flexibility in applying n=1 to n=4 perturbations [1].

• DIII-D (USA): Operated by General Atomics, DIII-D has been a world leader in RMP research since the first demonstration of ELM suppression. Its flexible coil sets and advanced diagnostic capabilities continue to provide critical data for validating physics models.

• ASDEX Upgrade (Germany): The Max Planck Institute for Plasma Physics operates ASDEX Upgrade, which has a full tungsten wall. Its experiments are vital for understanding RMP physics in a metallic environment, providing insights into the challenges of impurity control and heat load management in future reactors [6].

• KSTAR (South Korea): The Korea Superconducting Tokamak Advanced Research (KSTAR) device, with its long-pulse superconducting magnets, is used to study RMP effects in steady-state-relevant scenarios. It has demonstrated ELM suppression for extended periods.

• JET (UK): Before its decommissioning, the Joint European Torus provided crucial data on RMP ELM mitigation with an ITER-like beryllium and tungsten wall, testing the technique at a scale and with parameters closer to those of ITER than any other machine [5].

Open challenges

Despite significant progress, several scientific and engineering challenges remain for the successful application of RMPs in a fusion power plant.

• Confinement Degradation: The reduction in energy and particle confinement associated with RMPs remains a primary concern. If the degradation is too severe, it could make achieving the required Lawson criterion for net energy gain more difficult. Mitigating this degradation while maintaining ELM control is a key research priority.

• Predictive Capability: While models have improved, a fully predictive, first-principles understanding of the plasma response to RMPs is still lacking. Accurately predicting the optimal RMP configuration and the required coil currents for a new device like ITER, without empirical tuning, is a major challenge [7].

• Access to ELM Suppression at High Density: In many experiments, ELM suppression becomes more difficult to achieve at higher plasma densities, which are required for reactor operation. The operational window for suppression tends to narrow, posing a challenge for robust control in a power plant scenario.

• Divertor Heat Flux Management: The 3D magnetic structures created by RMPs can lead to non-uniform, or "striated," heat and particle fluxes onto the divertor targets. These localized hot spots could cause damage and must be managed, potentially through dynamic rotation of the applied RMP field [8].

• Integration with Core Scenarios: The effect of edge RMPs on the core plasma, including toroidal rotation braking and potential triggering of other MHD instabilities like neoclassical tearing modes (NTMs), must be fully understood and managed.

Outlook

The 5-15 year trajectory for RMP research is tightly coupled to the timeline of ITER and the design of future demonstration power plants (DEMOs). In the near term (5 years), research on existing tokamaks will focus on resolving the open challenges, particularly on developing control strategies to minimize confinement degradation and demonstrating robust ELM control in reactor-relevant high-density, low-rotation regimes. This period will see continued validation and refinement of 3D MHD and transport models against experimental data.

In the medium term (5-10 years), the focus will shift to the first plasma operations at ITER. The initial commissioning and use of ITER's RMP coils will be a critical test of our current understanding. These experiments will provide the first data on ELM control in a burning plasma environment, validating the design choices and informing operational strategies. The results from ITER will be definitive in assessing the viability of RMPs as a reactor-scale ELM control solution.

Looking further ahead (10-15 years), the experience from ITER will directly inform the design of DEMO-class reactors. If successful, RMPs will be a standard feature. Research may evolve towards more advanced applications, such as using RMPs for impurity control or tailoring the plasma edge profile for optimal performance. The development of real-time feedback systems that dynamically adjust the RMP spectrum based on plasma conditions could lead to more robust and efficient control, solidifying the role of resonant magnetic perturbations as an indispensable tool for steady-state tokamak operation.

References

1. ITER Physics Basis — Nuclear Fusion (2007)
2. Stochasticity in Hamiltonian Systems — Physics Reports (1979)
3. Physics of RMP ELM control in ITER — Plasma Physics and Controlled Fusion (2015)
4. Suppression of Large Edge-Localized Modes in High-Confinement DIII-D Plasmas with a Stochastic Magnetic Boundary — Physical Review Letters (2003)
5. ELM control with RMPs in JET with the ITER-like wall — Nuclear Fusion (2015)
6. Understanding the operational space of ELM suppression by resonant magnetic perturbations in ASDEX Upgrade — Nuclear Fusion (2017)
7. Key role of the plasma response to resonant magnetic perturbations for edge-localized mode control — Nuclear Fusion (2012)
8. Three-dimensional heat deposition patterns on the DIII-D divertor tiles during ELM suppression by resonant magnetic perturbations — Journal of Nuclear Materials (2011)