Vienna, Austria

ESTRO 2023

Session Item

New technologies
Poster (Digital)
Physics
Longitudinal stability in the dielectric wall accelerator through time-varying accelerating fields
Jan Seuntjens, Canada
PO-1847

Abstract

Longitudinal stability in the dielectric wall accelerator through time-varying accelerating fields
Authors:

Christopher Lund1, Paul M Jung2, Morgan J Maher1, Julien Bancheri1, Thomas Planche2, Rick Baartman2, Jan Seuntjens3

1McGill University, Medical Physics Unit, McGill University Health Centre, Montreal, Canada; 2TRIUMF, Beam Physics, Vancouver, Canada; 3University of Toronto, Princess Margaret Cancer Centre, Radiation Medicine Program, University Health Network, Toronto, Canada

Show Affiliations
Purpose or Objective

As part of the development of a dielectric wall accelerator (DWA)-based compact proton therapy (PT) device, this work aims to demonstrate longitudinal particle stability in a DWA for the first time. This is important for Bragg peak sharpness, maximum dose rate, and other PT dose characteristics. In particular, we test whether linear time variations in the accelerating fields can produce a corrective force gradient about the reference trajectory during acceleration.

Material and Methods

The particle-in-cell code Warp was used to track a 1 µA stream of 25 keV protons through a novel numerical DWA model. This model consists of: (i) a multipole expansion describing the spatial distribution of each field, (ii) a time profile (1 ns FWHM) describing the ramp-up, linear variation Γ', and ramp-down of each field, and (iii) a set of differential equations used to describe the reference trajectory, i.e. the dynamics of an idealized fictitious particle, and to coordinate the relative timing of each field. See Figure 1. The phase space distributions and their second moments were tracked through a 15 cm (150 modules), 10 MeV/m DWA segment (representative of a pre-accelerator) and compared for Γ [-2,2] MeV/m/ns.


Results

For all Γ', acceleration up to the final reference energy of the DWA segment (1.5 MeV) was demonstrated. However, significant differences were observed in the phase space distributions, especially in the vicinity of the reference trajectory. We define the outer boundary of this region as a maximum allowed energy deviation from the reference energy, i.e. an energy cutoff. As this boundary is not clearly defined, energy cutoffs between 0.5% and 10% were considered. The following trends were visible for all applied cutoffs, with the effects larger for stricter cutoffs.

Compared to the baseline (Γ'=0), a Γ'>0 led to higher bunch charge, reduced temporal width, and a weaker (less negative) correlation between deviations in particle arrival time and energy. For Γ'<0, the trends were inverted. In both cases, the effects increased with the magnitude of Γ'. See Table 1.


Conceptually, for Γ'>0, particles that arrive later than the reference particle at a given point experience a larger field, while those that arrive earlier experience a weaker field. Due to the linear variation, the larger the temporal deviation, the stronger this corrective force. As a result, a set of stable orbits is established about the reference trajectory in time-energy (i.e. longitudinal) phase space. For Γ'<0, existing deviations are exacerbated rather than diminished. The resulting divergence in time-energy space leads to fewer accelerated particles, increased temporal width, and a stronger time-energy correlation.

Conclusion

These data demonstrate that while time-varying fields are not necessary for acceleration over short distances, a positive linear time variation leads to longitudinal stability in the vicinity of the reference trajectory. This is important for the design of a DWA-based PT device.