Dynamic Model of Electron Beam Induced Deposition (EBID) of Residual Hydrocarbons in Electron Microscopy

Abstract

Adsorbed species surface diffusion Electron beam induced deposition (EBID) of residuals carbon can be either a contamination problem or can provide a basis for 3-D nanofabrication and nanoscale metrology. In this process a solid deposit is formed at the point of impact of the electron beam due to the decomposition of residual hydrocarbon species adsorbed on the solid substrate. The first observation of EBID can be traced to miscroscopists who noticed the growth of thin films of carbon while imaging using an electron microscope. The process was referred to as "contamination" because of its adverse effects on the microscope's imaging quality. Later, it has been demonstrated that with appropriate control of the electron beam this problematic contamination can be exploited to deposit three dimensional nanostructures with the spatial resolution down to 10nm. Numerous researchers have experimentally explored various factors influencing EBID growth rate and geometry of the deposit. To date, the most comprehensive theoretical model predicting the shape of the deposit in EBID is due to Silvis-Cividjian[1]. However, this model accounts for electron transport only. A few, fairly rudimentary models have also been developed for mass transport in EBID, but usually limited to rather simplistic treatment of electron transport. To this end, we have developed a comprehensive dynamic model of EBID coupling mass transport, electron transport and scattering, and species decomposition to predict deposition of carbon nano-dots. The simulations predict the local species and electron density distributions, as well as the 3-D profile and the growth rate of the deposit. Since the process occurs in a high vacuum environment surface diffusion is considered as the primary transport mode of surface-adsorbed hydrocarbon precursor. Transport, scattering, and absorption of primary electron as well as secondary electron generation are treated using the Monte Carlo methods. Low energy secondary electrons (SE) are the major contributors to hydrocarbon decomposition due to their energy range matching peak dissociation reaction cross section energies for precursor molecules. The local SE flux at the substrate and at the free surface of the growing deposit is computed using the Fast Secondary Electron (FSE) model. When combined with the total dissociation reaction corssection and the local hydrocarbon surface concentration, this allows us to compute the local deposition rate. The deposition rates are then used to predict the shape profile evolution of the deposit. Simulation results are compared with an AFM imaging of carbon EBID.