Fast simulation of stochastic exposure distribution in electron-beam lithography

The relative critical dimension variation of nanoscale features has become large enough to significantly affect the minimum feature size and maximum circuit density realizable in most lithographic processes. One source of such variation is the line edge roughness (LER). In the electron-beam lithographic process, the fluctuation of exposure (energy deposited) in the resist is one of the main factors contributing to the LER. It is essential to accurately estimate the exposure fluctuation for developing an effective method to reduce the LER. A possible method is to rely on the Monte Carlo simulation in computing the exposure distribution in a circuit pattern, i.e., generating a point spread function (PSF) for each point to be exposed, where the PSF is stochastic. While this approach can lead to a more realistic estimation, it is not practical due to its tremendous amount of computation required. In this paper, a new method to greatly reduce the number of PSF's to be generated without sacrificing the accuracy...