Flat and magnetized beams have become increasingly important in particle accelerators. For instance, flat beams can be used to increase luminosity for future colliders. On the other hand, magnetized electron beams can improve electron cooling performance in hadron beams. Characterizing these kinds of beams requires the full 4D transverse phase space distribution (TPS), which is usually time consuming and/or requires specialized diagnostics. In this work, our generative phase space reconstruction method (GPSR) was implemented to efficiently reconstruct the 4D TPS of flat and magnetized beams at the Argonne Wakefield Accelerator with a simple quadrupole scan and a YAG screen. 

The figure shows the reconstructed 4D TPS for both the magnetized and flat beams. In both cases, the experimental data needed for the GPSR method consists of a few shots per step from an 11-step quadrupole scan. Magnetization values calculated from these reconstructed distributions were compared with traditional slit-mask measurements, showing good agreement within 10% error. Furthermore, x and y emittances were computed for the flat beam distribution and are in excellent agreement with conventional quadrupole scan measurements. These results show that GPSR can be used to reconstruct the 4D TPS distribution of flat and magnetized beams at the Argonne Wakefield Accelerator in an accurate and data-efficient way, leading to a more complete characterization of such beams. 

Reference:

S. Kim, J. P. Gonzalez-Aguilera, P. Piot, G. Chen, S. Doran, Y.-K. Kim, W. Liu, C. Whiteford, E. Wisniewski, A. Edelen, R. Roussel, and J. Power, “Four-dimensional phase-space reconstruction of flat and magnetized beams using neural networks and differentiable simulations,” Phys. Rev. Accel. Beams, vol. 27, no. 7, p. 074601, Jul. 2024, doi: 10.1103/PhysRevAccelBeams.27.074601. Available: https://link.aps.org/doi/10.1103/PhysRevAccelBeams.27.074601