Modeling Transverse Space Charge effects in IOTA with pyORBIT

Runze Li, Tanaji Sen, J.-f. Ostiguy

TL;DR

Researchers at Fermilab and the University of Wisconsin have successfully validated the pyORBIT simulation code for modeling space charge effects in the Integrable Optics Test Accelerator (IOTA). By benchmarking against established tools and theoretical predictions, they demonstrated that while particle-in-cell (PIC) codes face challenges with long-term stability, they are highly effective at predicting short-term beam growth and loss in complex, nonlinear accelerator lattices.

What Problem Does This Paper Address?

In high-intensity proton accelerators, "space charge"—the mutual electromagnetic repulsion between particles in a bunch—is a major headache. It causes the beam to spread out (emittance growth) and leads to particle loss, which can damage hardware and limit machine performance.

Fermilab’s IOTA ring is a unique laboratory designed to test "integrable optics," a method of using specific nonlinear magnets to make beams more robust against these effects. However, designing these systems requires simulation tools that can accurately handle both the complex nonlinear magnets and the collective repulsion of billions of protons. This paper addresses the need to validate pyORBIT, a specialized Particle-In-Cell (PIC) code, to ensure it can reliably predict the behavior of IOTA’s 2.5 MeV proton beams.

Methods

The authors employed a multi-stage validation strategy:

  • Benchmarking: They first compared pyORBIT’s single-particle tracking (without space charge) against MADX, the industry standard. They tested "symplecticity"—the code's ability to preserve phase-space volume—and "tune footprints" (how particle oscillation frequencies change with amplitude).
  • PIC Simulations: To model space charge, pyORBIT distributes "macro-particles" onto a grid, solves Poisson’s equation to find the electric field, and applies kicks to the particles.
  • Slow Initialization: To avoid the numerical "shock" of suddenly injecting a high-charge beam into a simulation, the authors used a technique where the charge is ramped up linearly over several turns. This allows the beam to reach a "steady state" more realistically.
  • Convergence Testing: They systematically varied the number of macro-particles, grid resolution, and the frequency of space charge "kicks" to find the optimal parameters for numerical accuracy.

Datasets

The study did not use external datasets but rather the specific lattice configuration of the IOTA proton ring.

  • Lattice: Included a "bare" linear lattice and a version with a string of 17 octupole magnets to create a partially integrable system.
  • Beam Parameters: Simulated a 2.5 MeV proton beam with intensities up to $9 \times 10^{10}$ protons per bunch.
  • Distributions: Used Gaussian transverse distributions and "Waterbag" longitudinal distributions for the macro-particles.

Key Insights and Ideas

  • The Symplecticity Barrier: The study confirms a known limitation of PIC codes: while single-particle tracking is highly symplectic (accurate over long durations), the introduction of a self-consistent space charge grid breaks this property. This makes these codes excellent for short-term studies ($10^3$ to $10^4$ turns) but less reliable for simulating millions of turns.
  • The "Slow Start" is Essential: Ramping up charge (slow initialization) over roughly 40 turns was found to be the "sweet spot" for IOTA. It drastically reduced artificial emittance growth and lowered particle losses from ~1% to nearly 0.01%.
  • Matching Matters: Much of the initial beam degradation in simulations is caused by a mismatch between the "bare" lattice functions and the actual equilibrium state of a beam under heavy space charge.
  • Code Refinement: The validation process uncovered and corrected several bugs in pyORBIT related to dipole edge focusing and solenoid handling, highlighting the importance of cross-code benchmarking.

Key Results

  • High Accuracy: In stable regions, pyORBIT and MADX matched particle tunes to within $10^{-3}$.
  • Numerical Convergence: The authors established that $5 \times 10^5$ macro-particles and a $128 \times 128 \times 5$ grid provide sufficient resolution for IOTA simulations.
  • Theoretical Alignment: The simulated space charge tune shifts (the shift in oscillation frequency due to repulsion) closely matched analytical predictions, confirming the code's physics model is sound.
  • Dynamic Aperture: The "stable region" for particles (dynamic aperture) was found to be remarkably similar between the two codes, even when including synchrotron motion.

Limitations and Future Work

  • Long-term Tracking: Due to the non-symplectic nature of the space charge model, the authors caution against using these results for long-term stability predictions beyond $10^4$ turns.
  • Dimensionality: This report focused primarily on transverse (2D) space charge effects. The authors note that longitudinal effects are a necessary next step for a complete 3D model.
  • Real-world Imperfections: The simulations assumed a "perfect" lattice. Future work will need to incorporate magnet alignment errors and more complex fringe field effects to fully mirror experimental conditions at IOTA.

Summary updated

Abstract

The role and mitigation of space charge effects are important aspects of the beam physics research to be performed at the Integrable Optics Test Accelerator (IOTA) at Fermilab. The impact of nonlinear integrability (partial and complete) on space charge driven incoherent and coherent resonances will be a new feature of this accelerator and has the potential to influence the design of future low energy proton synchrotrons. In this report we model transverse space charge effects using the PIC code pyORBIT. First we benchmark the single particle tracking results against MADX with checks on symplecticity, tune footprints, and dynamic aperture in a partially integrable lattice realized with a special octupole string insert. Our space charge calculations begin with an examination of the 4D symplecticity. Short term tracking is done first with the initial bare lattice and then with a matching of the rms values with space charge. Next, we explore slow initialization of charge as a technique to establish steady state and reduce emittance growth and beam loss following injection into a mismatched lattice. We establish values of space charge simulation parameters so as to ensure numerical convergence. Finally, we compare the simulated space charge induced tune shifts and footprints against theory.
BibTeX
@misc{li2021modeling,
  author = {Runze Li and Tanaji Sen and J.-F. Ostiguy},
  title = {{Modeling Transverse Space Charge effects in IOTA with pyORBIT}},
  year = {2021},
  eprint = {2106.03327},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2106.03327},
}
Figure