Internal Transport Barriers in Fusion Plasmas

Updated 10 March 2026

Internal Transport Barriers (ITBs) are localized regions in magnetically confined plasmas where turbulent transport is sharply reduced, leading to steep temperature and density gradients.
They occur via mechanisms such as magnetic shear reversal, zonal flow shearing, and fast-ion effects, which collectively suppress turbulence and enhance energy confinement.
Researchers employ advanced gyrokinetic simulations and theoretical models to study ITB formation, optimize tokamak performance, and draw parallels with universal transport barrier phenomena.

An internal transport barrier (ITB) is a localized region in a magnetically confined plasma where turbulent transport of heat and particles is sharply reduced, enabling the development of steep temperature and density gradients, and thereby enhancing energy confinement. ITBs are central to the performance of advanced tokamak scenarios, as they enable elevated core pressure and may provide a route to steady-state, reactor-relevant operations by suppressing anomalous transport far below the “gyro-Bohm” scaling typical for fully turbulent plasmas. The following sections present a comprehensive technical account of the mechanisms, models, and experimental evidence relating to ITB formation, sustainability, and dynamics, grounded in the latest theoretical and computational literature.

1. Physical Mechanisms Leading to ITB Formation

The triggering and sustenance of ITBs are intimately linked to the interplay between magnetic shear, field-line topology, turbulence self-organization, and kinetic microphysics.

Magnetic Shear Reversal and Field-line Topology

ITBs preferentially occur in radial zones where the magnetic shear,

$\hat s(r) = \frac{r}{q(r)}\frac{dq}{dr},$

vanishes or becomes negative (reversed shear). When $\hat s \approx 0$ , the ballooning-localization of drift-wave turbulence is diminished, enabling turbulent structures (“eddies”) to extend tens to hundreds of poloidal turns along the field lines. This ultra-long parallel coherence allows eddies to “bite their own tail,” undergoing self-interaction that can lead to either constructive or destructive interference, profoundly altering the non-linear saturation and spectral transfer properties of the turbulence (Volčokas et al., 2022, Volčokas et al., 2024, Volčokas et al., 2024).

The causative processes are amplified at magnetic surfaces where the safety factor $q(r)$ assumes a low-order rational value $m/n$ . Near these surfaces, the closure condition for field-line wrapping optimally facilitates eddy self-encounter, provoking a strong resonant modification of the turbulent state. Precise alignment of $q_\text{min}$ (the minimum of the safety factor profile) with such rationals is often necessary for robust ITB onset (Giannatale et al., 2024).

Zonal Flows, Shearing, and Electromagnetic Effects

Large amplitude, axisymmetric $E \times B$ flows (“zonal flows”) are, in many cases, established or enhanced within the ITB region. Their shearing rates, $\gamma_\text{shear}$ , can readily exceed the linear growth rates of the dominant drift instabilities, cutting off eddy growth nonlinearly and strongly suppressing cross-field transport (Ma et al., 6 Nov 2025, Giannatale et al., 2024). Electromagnetic effects, i.e., finite plasma $\beta$ , further stabilize turbulence by introducing magnetic-field-line bending costs (Maxwell stress) and can drive the transition between different ITG branches, changing the dominant radial location of the most unstable modes. In fully formed ITBs, nonlinear interaction between global electromagnetic modes and zonal flows can even dominate over linear mechanisms in maintaining the barrier (Lin et al., 2023).

Fast Particle Effects and Wave-Particle Resonance

Energetic ions, produced via neutral beam injection (NBI) or ion cyclotron resonance heating (ICRH), can reduce turbulence by diluting the drive for thermal-ion ITG modes and via wave–particle resonance mechanisms—leading to anomalous barriers such as the fast-ion-induced transport barrier (F-ATB) (Siena et al., 2021). The critical threshold for barrier formation, width, and localization are sensitive to the amplitude and profile of the fast-ion population.

2. Theoretical Frameworks and Modeling Approaches

Gyrokinetic Simulations and Critical-Gradient Theory

State-of-the-art nonlinear gyrokinetic codes (GENE, GTC, ORB5, GYRO) are the primary computational tools used to explore ITB dynamics. In direct numerical experiments, the critical gradient required to trigger turbulence ( $a/L_{T,\text{crit}}$ for ions or electrons) is significantly upshifted—by factors of 2–3 or more—within ITBs due to reversed shear or electromagnetic effects (Peterson et al., 2011, Giannatale et al., 2024). For electrostatic turbulence, nonlinearly driven off-midplane streamers and profile corrugations are observed, while electromagnetic runs reveal robust suppression of both linear and nonlinear transport at enhanced $\beta$ (Ma et al., 6 Nov 2025).

Hamiltonian and Non-Twist KAM Theory

Transport-barrier onset in reduced models is linked to the existence of shearless (non-twist) invariant tori in Hamiltonian maps of field-line or particle dynamics (Firpo et al., 2011, Jr et al., 2012, Osorio et al., 2021, Marcus et al., 2018). Zeros of the local twist provide robust structures against stochasticity for large ranges of perturbation amplitude, but are vulnerable to destruction when the rotational transform approaches low-order rationals—reflecting in experimental observations of ITB collapse at those points.

Turbulence-Generated Magnetic Corrugations

Recent electromagnetic gyrokinetic simulations demonstrate that ion-scale turbulence can drive zonal parallel currents, generating stationary corrugations in vector potential $A_\parallel$ and leading to a stepped $q(r)$ profile with extended $s \approx 0$ plateaus. This feedback increases eddy self-interaction and reduces turbulent heat flux by factors of 3–4 (Volčokas et al., 2024, Giannatale et al., 2024). The barrier formation is therefore an emergent property of the turbulence–current feedback loop, particularly amplified near rational $q$ surfaces in low-shear environments.

3. Experimental Observations and Diagnostic Signatures

ITBs have been robustly observed in most major tokamaks (JET, JT-60U, DIII-D, HL-2A, EAST, ASDEX Upgrade) and are linked experimentally to the following signatures:

Transport reduction: Ion and electron thermal diffusivities $\chi_{i,e}$ are suppressed by up to an order of magnitude within the barrier, confirmed through modulated heating, heat-pulse propagation, and profile analysis (Xie et al., 2022, Ma et al., 6 Nov 2025).
Profile corrugation: Steep (often off-axis) gradients in $T_i$ , $T_e$ , and density $n$ form within or adjacent to the low-shear, rational- $q$ region (Xie et al., 2022, Giannatale et al., 2024).
Barrier localization: Barrier positions correlate with $q \approx$ low-order rationals and with the zero or reversal of the local magnetic shear (Xie et al., 2022, Giannatale et al., 2024).
Threshold phenomena: Rapid, bursty changes in local transport are observed as equilibria (e.g., via current drive or q-profiling) are scanned across rational values; small adjustments in $q$ can abruptly trigger or extinguish ITBs (Volčokas et al., 2022).
Role of external control: Launching off-axis ECH at specific rational $q$ surfaces has been used to tailor barrier location and sustainment in DIII-D and similar devices (Xie et al., 2022).

4. Generalizations to Other Contexts

The concept of internal transport barriers transcends magnetically confined plasma systems:

Wall turbulence: Objective barriers to diffusive momentum transport in wall-bounded turbulence—the so-called “momentum transport barriers” (MTBs)—are mathematically analogous to ITBs, acting as invariant manifolds of the Laplacian field that block the viscous momentum flux and organizing the underlying coherent structures (Aksamit et al., 2021).
Stochastic models: In the sharp-interface limit, heat or particle transport in a turbulent system with an idealized barrier can be represented as Brownian motion with a “hard membrane”—obtained as the scaling limit of a diffusion equation with a vanishingly thin, minimally diffusive barrier. This framework captures the essential “snapping-out” dynamics at the barrier, providing a mathematical justification for effective ITB behavior as observed macroscopically (Aryasova et al., 18 Oct 2025).

5. Energetic Particle Barriers and Effective Safety Factor

For energetic particles (EPs), the conventional $q$ -profile does not dictate the location or strength of barriers: instead, the effective safety factor $q_\text{eff}(r)$ , incorporating gradients and curvature drifts, determines EP orbit topology. Transport barriers for EPs can emerge even in regions where field-line chaos persists, through two mechanisms (Ogawa et al., 2016):

Resonance displacement: Grad- $B$ drift shifts the resonant locations away from chaotic zones, reconstructing invariant tori (barriers).
Resonance elimination: Curvature drift can annihilate resonances via saddle-node bifurcation, removing local stochasticity and embedding barriers amid field-line chaos.

This highlights that ITBs for bulk species and EPs may be spatially and dynamically distinct, complicating the barrier robustness landscape in burning plasmas.

6. Neoclassical Effects and Transport Equilibria

In established ITBs, suppressed turbulence brings neoclassical transport to the foreground. Under strong-gradient conditions (with $L_{n,T}\sim\rho_{\theta i}$ ), poloidal variations of density and potential become significant, introducing nonlinearities and “multiple root” bifurcations in the transport equations (Trinczek et al., 22 Feb 2026). The resulting system may exhibit multiple steady-state solutions (low- and high-transport branches), with abrupt H–L back-transitions corresponding to root-jumps—a mechanism for intrinsic barrier formation, sustainment, and collapse even in the absence of significant turbulence.

7. Parametric Dependencies, Thresholds, and Control

The formation, robustness, and collapse of ITBs are sensitive to the following parameters and physical controls:

Parameter/Control	Barrier Onset/Collapse	Notes
$q_\text{min}$ at low-order rational	Onset at $\Delta q \lesssim \rho^* q_0/n$	Empirically, ITBs typically form for $q_\text{min}=2,3$
Magnetic shear $\hat{s}$	Onset for $\|\hat s\|\lesssim 0.1$	$L_\parallel\sim 1/\|\hat s\|$ scaling
Fast-ion log-gradient amplitude $A$	Threshold $A_\text{thres}\sim 42$ (GENE units)	Controls F-ATB width and position (Siena et al., 2021)
Plasma $\beta$	$>0.5\%$ on-axis suppresses high-frequency ITG	Essential for EM barrier sustainment (Ma et al., 6 Nov 2025)
ECH deposition radius	Alignment with rational $q$ surface required	ITB location controlled in DIII-D (Xie et al., 2022)
Zonal current strength	Sets width of $q$ -flattened region	$\Delta r_\text{fl} \sim$ 10–14 $\rho_i$ (Giannatale et al., 2024)

The sensitivity to these controls offers flexible, real-time strategies for ITB optimization, including current-profile control, rational $q$ targeting, tailored heating profiles, and exploitation of fast-particle populations.

8. Outlook: Unified Paradigm and Open Problems

Recent work establishes a unified paradigm: ITBs form through multi-scale, dynamic interactions between magnetic topology (shear and rational surfaces), turbulence nonlinearity (eddy self-interaction, squeezing), zonal field amplification (flows and currents), and kinetic/electromagnetic stabilization. Precise simulation benchmarks (Volčokas et al., 2022, Giannatale et al., 2024, Ma et al., 6 Nov 2025), reduced-Hamiltonian theory (Firpo et al., 2011, Osorio et al., 2021, Jr et al., 2012), and cross-disciplinary analogs in fluid turbulence (Aksamit et al., 2021, Aryasova et al., 18 Oct 2025) collectively illuminate the universality of barrier formation in complex turbulent systems.

Outstanding challenges include the prediction and control of ITB collapses (e.g., via resonant field perturbations), the integration of multi-species and impurity transport physics, the effects of energetic particles in reactor-relevant regimes, and the exploitation of feedback loops (e.g., zonal current—magnetic shear—eddy self-interaction) for robust, reproducible barrier sustainment.

Key References: