Beamline Simulations of a Dielectric Wall Accelerator for Low-Cost Proton Therapy
Abstract
Purpose
The accessibility of proton therapy is limited by the size and cost associated with traditional acceleration techniques. The dielectric wall accelerator (DWA) was proposed as a low-cost, compact proton accelerator capable of generating suitable beams for proton therapy. We have developed a simulated DWA beamline that accelerates protons from a standard ion source to clinical energies (20 - 226 MeV).
Methods
Two numerical models of the DWA were developed and validated: one in TRANSOPTR, a linear optics beam envelope code, and one in Warp, a particle-in-cell toolkit. TRANSOPTR was then used to design a source-to-patient DWA beamline for proton therapy applications.
Results
In our design, a conventional DC (i.e., continuous) proton source injects directly into a short DWA segment (the pre-accelerator) with a reduced field gradient of 10 MV/m. The accelerated bunches are filtered out of the DC current using a collimated bend section. Following the bend, the main DWA is directed towards the patient isocentre, avoiding the use of large dipoles necessary at high energies, allowing for a more compact design. The 2-RMS transverse beam size was optimized to prevent beam collisions with the wall, remaining below 1 cm through the main DWA by tuning the pre-accelerator field gradient and magnetic quadrupole strengths. Random timing errors of 20 ps from the ideal timing of each DWA module had minimal effects on the final energy (RMS deviation of 0.01% from the correctly timed simulation), with RMS deviations in the final bunch size and length of 3% and 15%, respectively.
Conclusion
The beamline design of the DWA addresses several challenges, including low-energy injection into high field gradients (risk of rapid defocusing) and timing errors, while maintaining a compact format capable of achieving clinical energies. Work is ongoing to characterize beam loss and bunch filtration using our Warp model.