Locked mode

Overview

A locked mode is a type of large-scale magnetohydrodynamic (MHD) instability in magnetically confined plasmas, most notably in tokamaks. It occurs when a naturally rotating magnetic perturbation, usually an (m=2, n=1) tearing mode, ceases to rotate and becomes stationary, or "locked," with respect to the device's vacuum vessel. This locking is driven by the interaction between the plasma's helical magnetic island and small, unavoidable, non-axisymmetric magnetic fields known as error fields. These error fields arise from minute imperfections in the construction and alignment of the main magnetic field coils.

The onset of a locked mode is a critical concern in fusion research because it severely degrades plasma confinement by creating a magnetic short-circuit that allows heat and particles to escape rapidly from the core to the edge. The stationary nature of the mode leads to intense, localized plasma-wall interaction, which can damage vessel components. Most significantly, a locked mode is a common and reliable precursor to a major disruption, a catastrophic event that terminates the plasma discharge in milliseconds and can inflict severe structural damage on the fusion device. Understanding, predicting, and mitigating locked modes is therefore essential for the successful operation of current and future tokamaks, particularly for large-scale devices like ITER.

Physics / Mechanism

The formation of a locked mode involves the interplay of plasma rotation, tearing mode stability, and external magnetic fields. The process can be understood through a sequence of events:

1. Tearing Mode Formation: The process begins with the formation of a tearing mode. This is a resistive MHD instability that grows on a rational magnetic surface, where the safety factor q is a rational number m/n (e.g., q=2/1). The instability tears and reconnects magnetic field lines, forming a chain of rotating magnetic islands. In the absence of external influences, these islands rotate toroidally with the bulk plasma fluid, driven by E×B drifts and momentum from heating systems like neutral beam injection (NBI).

2. Interaction with Error Fields: All tokamaks have small, static, non-axisymmetric magnetic fields, or "error fields." These arise from sources such as slight misalignments of poloidal and toroidal field coils, coil winding imperfections, and current feeds. A rotating magnetic island passing through this static error field experiences an electromagnetic torque. The interaction is resonant; the (m,n) component of the error field exerts a torque on the (m,n) tearing mode.

3. Mode Slowdown and Locking: The torque exerted by the error field opposes the mode's rotation. Simultaneously, as the mode rotates, it induces eddy currents in the surrounding conductive structures, like the vacuum vessel wall. These eddy currents create a magnetic field that also produces a drag torque, further slowing the mode's rotation. As the mode slows, the shielding effect of the plasma is reduced, allowing the external error field to penetrate more deeply and exert an even stronger torque. This creates a positive feedback loop: the mode slows, the torque increases, causing it to slow further. When the error field torque and wall drag overcome the plasma's viscous torque that drives rotation, the mode's rotation frequency drops rapidly to zero. The mode becomes locked in phase with the external error field.

4. Mode Growth and Disruption: Once locked, the mode often grows rapidly in amplitude. The static error field provides a constant, resonant forcing that drives the instability. The large, stationary magnetic island flattens the temperature and density profiles across a significant portion of the plasma radius, crippling energy confinement. This degradation can cool the plasma edge, causing the current profile to contract, which further destabilizes the tearing mode. This cascade of events frequently culminates in a major disruption, where thermal and magnetic energy is lost to the walls on a timescale of milliseconds.

The critical error field amplitude required to cause locking depends on plasma parameters, particularly density and rotation speed. Higher plasma density and faster rotation provide greater inertia and viscous forces, increasing the plasma's resilience to locking. This relationship is a key factor in defining operational limits for tokamaks.

Historical development

The phenomenon of locked modes has been observed and studied since the early days of tokamak research. Initial observations in the 1970s on devices like the Princeton Large Torus (PLT) identified large, non-rotating magnetic perturbations preceding disruptions. However, the connection to external error fields was not fully established until the 1980s.

Pioneering experiments on the DIII-D tokamak in the late 1980s and early 1990s were instrumental in elucidating the physics. Researchers used an external coil set (the "n=1 coil") to deliberately apply a static, resonant magnetic perturbation. These experiments systematically demonstrated that an external field could slow and lock a naturally occurring tearing mode, and that a critical error field threshold existed for this to occur. The work at DIII-D, along with similar studies at COMPASS-C and JET, established the direct causal link between error fields, mode locking, and disruptions.

These findings spurred the development of detailed theoretical models, such as the one by Fitzpatrick, which described the torque balance between the plasma's natural rotation, wall drag, and the external resonant field. This theoretical framework provided a quantitative basis for predicting the error field thresholds observed in experiments. The scaling of this threshold with plasma parameters (e.g., density, magnetic field, and machine size) became a major area of research, as it was crucial for extrapolating the requirements for future, larger devices like ITER. This research led to the consensus that active error field correction would be an absolute necessity for ITER and future fusion power plants.

Current status

As of 2026, the understanding of locked mode physics is mature, and mitigation techniques are standard practice on all major tokamaks. The primary strategy is active error field correction (EFC). This involves using a set of non-axisymmetric saddle coils located outside the plasma to generate a small magnetic field that precisely cancels the intrinsic error field.

Modern EFC systems rely on detailed magnetic measurements from the device to characterize the intrinsic error field before plasma operations. During a plasma discharge, feedback control systems can dynamically adjust the EFC coil currents to respond to changes in the plasma's response or to suppress specific MHD modes. For example, JET and DIII-D have demonstrated that optimized EFC can significantly expand the operational space to lower densities and lower q95 values, regimes that were previously inaccessible due to locked mode disruptions.

The error field tolerance of a plasma is now a well-established figure of merit. Multi-machine scaling studies have been conducted to project the requirements for ITER. These studies indicate that ITER's intrinsic error field must be corrected to a level of B_error / B_T < 10^-4 to avoid locked modes in its high-performance scenarios. This has driven stringent engineering tolerances on the manufacturing and alignment of ITER's massive superconducting magnet system.

Research continues to refine locked mode avoidance strategies. This includes using non-resonant magnetic fields for disruption avoidance and exploring the use of electron cyclotron current drive (ECCD) to shrink or eliminate the tearing mode island before it can lock.

Notable implementations

Virtually every major tokamak in operation employs systems and strategies to manage locked modes. Key examples include:

• DIII-D (General Atomics, USA): A world leader in locked mode and error field research. Its flexible set of internal (I-coils) and external (C-coils) non-axisymmetric coils has been used for foundational EFC experiments and for developing advanced techniques using resonant magnetic perturbations (RMPs) for both error field correction and edge localized mode (ELM) suppression.

• JET (UKAEA, UK): As the largest operating tokamak for many years, JET's experience with locked modes at large scale has been invaluable for ITER. It is equipped with an Error Field Correction Coil (EFCC) system that has been critical for achieving high-performance discharges, particularly in low-density operational regimes.

• KSTAR (NFRI, South Korea): The Korea Superconducting Tokamak Advanced Research device features a sophisticated set of in-vessel control coils (IVCCs). These coils have a fast time response, enabling advanced feedback control schemes to dynamically suppress MHD instabilities, including the precursors to locked modes, in real-time.

• ITER (International): The design of ITER has been profoundly influenced by locked mode physics. The device will be equipped with a powerful set of 27 correction coils (6 upper, 9 mid-plane, 12 lower) specifically for EFC and other MHD control functions. The successful commissioning and operation of this system are considered a critical prerequisite for achieving ITER's mission goals.

Open challenges

Despite significant progress, several challenges remain in the study and control of locked modes, particularly in the context of a burning plasma environment like that expected in a fusion power plant.

• Error Field Amplification: The plasma itself can amplify the externally applied magnetic field, a phenomenon known as plasma response. Accurately predicting this response is complex and is an active area of research. An imperfect understanding can lead to sub-optimal EFC, where the correction field itself becomes a source of instability.

• Dynamic Error Fields: While EFC can correct static error fields, transient magnetic perturbations from other sources (e.g., ELMs, sawteeth) can also trigger mode locking. Developing control systems that can react to these dynamic events on the relevant timescale is a continuing challenge.

• Burning Plasma Effects: In a reactor-grade plasma, a significant fraction of the heating will come from energetic alpha particles produced by fusion reactions. The interaction of these alpha particles with tearing modes is not fully understood. It is possible they could have a stabilizing or destabilizing effect, altering the conditions under which locked modes form.

• Integration with Other Control Systems: EFC must be integrated with a suite of other control systems, including those for plasma position and shape, density, and impurity control. The actuators for one system (e.g., magnetic coils) can affect the others, requiring sophisticated, multi-variable control algorithms to avoid conflicts and ensure stable operation.

Outlook

The 5-15 year trajectory for locked mode research and control is focused on preparing for and operating next-generation devices like ITER and the first fusion power plants. The primary goal is to transition from disruption mitigation to disruption avoidance.

In the near term (5 years), research will focus on validating predictive models for error field amplification against experimental data from multiple devices. This will be crucial for developing robust, feed-forward EFC scenarios for ITER's first plasma. The use of machine learning and AI-based controllers, trained on vast databases of past discharges, is expected to become more prevalent for real-time instability prediction and avoidance.

Looking further ahead (10-15 years), the focus will shift to operating within the ITER environment. The initial ITER campaigns will be a critical testbed for EFC at reactor scale. The data gathered will be used to refine control strategies for high-fusion-power deuterium-tritium (D-T) operation. A key objective will be to develop integrated control schemes where actuators like ECCD are used preemptively to stabilize tearing modes long before they grow large enough to be at risk of locking. Success in these areas will be a fundamental enabling step for the design and operation of Demonstration (DEMO) fusion power plants, where reliable, disruption-free operation is not just a research goal but an economic and safety necessity.

References

1. Nonlinear magnetohydrodynamics of tearing modes in the presence of sheared flows — Physics of Fluids B: Plasma Physics (1992)
2. Error field thresholds, mode penetration and seed island formation in various tokamaks — Nuclear Fusion (2007)
3. Locked modes in DIII-D and a method for prevention of the low density locked mode — Nuclear Fusion (1991)
4. ITER Physics Basis Editors, Chapter 3: MHD stability, operational limits and disruptions — Nuclear Fusion (1999)
5. Error field physics in tokamaks — Plasma Physics and Controlled Fusion (2009)
6. Feedback control of resistive wall modes and error fields in the DIII-D tokamak — Nuclear Fusion (2008)
7. MHD stability and disruption physics in JET — Plasma Physics and Controlled Fusion (2009)
8. Review of error field control in KSTAR — Fusion Engineering and Design (2017)