Casino 3 for point spread function (PSF)

[zhlltu2026]

Hello everyone,

I am using CASINO 3.3.0.4 to simulate electron scattering in multilayer resist/substrate stacks in order to extract the PSF for an electron‑beam lithography application. I am using CASINO 3 specifically because it enables secondary‑electron (SE) generation.

However, I am finding that enabling or disabling SE generation produces no observable difference in the simulation results. The two PSF curves extracted from the CAS files look essentially identical. I've attached the PSF plot for reference.

My understanding is that when SE generation is enabled, both slow and fast SEs should contribute to the deposited energy distribution. In particular, fast SEs should travel tens of nanometers to perhaps a micron, which should increase the deposited energy density in that radial range compared to a simulation without SE generation. Because of this, I would expect the PSF with SEs enabled to show higher energy density in this region.

For these simulations, the energy‑density matrix was extracted from the CAS files using pycasiontools, specifically from the "_DEnergy_Density" block. I also noticed that when SE generation is enabled, the "_DDIffusionEnergy_Density" block remains empty, and I am unsure whether this block is supposed to contain SE‑related contributions.

The simulation parameters are:
– 100 kV beam energy
– 2 nm beam diameter
– Cylindrical energy distribution mesh (1 um in radius, and 2 um in Z)
– 500k electrons simulated

I've attached snapshots of the simulation setup as well.

I would greatly appreciate any insight into whether the two PSF curves should be identical in these two cases, or whether this indicates an issue of either my simulation setup or data extraction process.

Thank you in advance for your help.

You cannot view this attachment.You cannot view this attachment.You cannot view this attachment.You cannot view this attachment.You cannot view this attachment.