Hamiltonian Particle Dynamics in Fusion Plasmas: Orbital Tomography and Spectrum Analysis for Energy and Momentum Transport under Resonant Non-Axisymmetric Perturbations

Scientific Impact Assessment

Paper type: PhD thesis (theoretical/computational plasma physics) synthesizing four peer-reviewed journal publications (J. Plasma Phys. 2021, 2024; Phys. Plasmas 2024 ×2).

1. Core Contribution

The thesis develops a low-computational-cost framework, grounded in Hamiltonian Action-Angle (AA) formalism, for predicting resonant mode-particle interactions in toroidal fusion plasmas. Its central novelty is the Drift Center (DC) approximation, which yields closed-form analytical expressions (in Jacobi elliptic functions and complete elliptic integrals) for guiding-center orbital frequencies and a kinetic q-factor as a function of the three constants of motion (E, µ, P_ζ). This enables *a priori* identification of resonance island-chain locations, the number of islands per chain, island widths, and transport-barrier locations (via local extrema of q_kin) — all without expensive orbit-following of perturbed trajectories. The work extends this from analytical Large-Aspect-Ratio (LAR) equilibria to realistic reconstructed equilibria (ASDEX Upgrade, DTT) via a semi-analytical contouring (Shoelace-formula) method, and finally constructs the Arnold web analytically to demonstrate Arnold diffusion under time-dependent perturbations. The problem it solves is practically important: existing tools (ORBIT, SPIRAL, HAGIS) require costly Poincaré-map construction and blind sampling of initial conditions, which is prohibitive for the parameter scans needed in scenario modeling.

2. Methodological Rigor

The methodology is sound and rests on well-established canonical Hamiltonian theory (Kaufman, Littlejohn, White-Boozer-Chance). Analytical predictions are systematically validated against direct numerical integration (Poincaré surfaces of section), showing excellent agreement in resonance locations and island counts. The domain of validity of the DC approximation is honestly investigated and quantified in terms of drift-orbit width (Eq. 5.22), with the authors clearly noting larger discrepancies for trapped and high-energy particles. The first application of the Smaller Alignment Index (SALI) to plasma chaos quantification is a rigorous choice with justified advantages over Lyapunov exponents. The peer-reviewed provenance of the constituent papers reinforces confidence. A limitation: validation is largely internal (analytical vs. numerical within the same code framework) rather than against experimental data or independent codes.

3. Potential Impact

The framework is directly relevant to energetic-particle confinement — a load-bearing concern for ITER, DEMO, DTT, and JT-60SA, since alpha-particle and NBI-ion losses threaten reactor performance. The compact COM-space resonance diagrams offer an intuitive, aggregated overview enabling "mode-engineering" strategies to mitigate fast-ion losses. The computational efficiency is the key selling point: it could plausibly be integrated into transport simulation workflows and, as the author outlines, into ML/PINN surrogate models for real-time resonance mapping. Impact is likely to accrue within the energetic-particle/MHD subfield rather than broadly across physics.

4. Timeliness & Relevance

The work addresses a genuine current bottleneck — efficient prediction of RMP/ELM/AE-induced fast-ion transport as devices move toward burning-plasma regimes. RMP-driven ELM control and AE-driven fast-ion losses are active experimental frontiers, so the tool is well-timed. The specific application to DTT (under construction) adds forward-looking relevance.

5. Strengths & Limitations

Strengths: Analytical tractability where prior work relied on numerics; unified treatment of passing and trapped particles (a gap left open in Shinohara 2020); careful prediction of the non-trivial island-number/mode-number correspondence; the first analytical Arnold-web construction for GC motion; thorough, pedagogically complete exposition. Limitations: The DC approximation degrades for large-drift energetic particles — precisely the population of greatest interest — though the semi-analytical method for realistic equilibria is exact and mitigates this. The approach treats quasi-linear (test-particle) interactions only; self-consistent mode evolution is deferred to future work. No code release limits immediate reproducibility. The physics phenomena confirmed (transport barriers, kinetic vs. magnetic chaos) are consistent with, rather than overturning, established understanding.

Additional Observations

The thesis is a consolidation of already-published, peer-reviewed results rather than presenting wholly new claims, which raises confidence but caps surprisingness. The contribution is best characterized as a *methodological building block* — a valuable, reusable computational primitive for a specialist community — rather than a paradigm shift. Resource intensity is low (analytical/semi-analytical, laptop-to-small-cluster scale), lowering the barrier for others to adopt it. Interdisciplinary reach is modest, spanning plasma physics and nonlinear/Hamiltonian dynamics (the SALI import).