An Investigation of the Deposition and Characterization of Materials Formed by Electron Beam Induced Deposition

This dissertation deals with EBID – Electron Beam Induced Deposition – a novel bottom up nanofabrication technique. Since EBID was first employed for nano-patterning, a number of empirical factors were investigated to control deposition process. Meanwhile a few theoretical models were proposed based on some fundamental assumptions. So far little work has been done to verify the validity of these assumptions. The main objective of my PhD study, therefore, was to answer whether these assumptions are valid so that unifying the empirical factors and theoretical models would be possible. Electrical resistivity of deposited materials was another interesting topic included in this work.

Verification of the assumptions was the core of this work. It started with the first assumption that the reaction happens on the surface. The substrate temperature was changed under controlled beam conditions to study how the growth rate was affected and the result provided evidence for the first assumption. The same data confirmed the second assumption that the mean lifetime of adsorbed molecules follows an Arrhenius relation. To verify the third assumption that secondary electrons drive the dissociation reaction, a variety of experiments were designed. It was attempted to suppress secondary emission by biasing the substrate or change secondary emission yield by controlling the experimental conditions. No correlation with deposition rate was found. But the experiments led to a better understanding of the role of interaction volume in deposition. Time-resolved sample current monitoring was then adopted to study electron scattering during deposition and a characteristic trend was found in the measurements. NISTMonte, was employed to simulate electron scattering for comparison. The estimation from the simulation agreed with the experiment. The role of backscattered electrons was also investigated by depositing lines across the boundary between a bulk material and thin film. The substrate thickness was found no influence on the linewidth of the deposits, which was explained by the transition of interaction volume. At last, a theoretical estimation suggested that backscattered electrons predominate the deposition for keV range primary electrons.

The second part of this work is to measure resistivity of the deposited materials. Four-point measurement circuits were fabricated on a wafer through thin film process. A probe station was used to measure the resistance and SEM and AFM were adopted to provide dimensional information. The resistivity of the deposited materials showed that the deposited materials have resistivity in semiconducting material range and the deposited material formed at slow scan speed had a lower resistivity than that formed at fast scan speed.