In the various film deposition processes, atomic layer deposition (ALD) is an excellent process for thin film deposition because of its controllability of thickness, excellent step coverage and uniformity but it has some weak point as restriction for choice of precursor, low yield, poor films properties etc. Plasma-enhanced atomic layer deposition (PEALD) is an advanced ALD to solve the problems of conventional ALD process. $Ta_2O_5$ has been studied as new dielectric material to replace $SiO_2$ in DRAM capacitor and gate dielectrics but it shows poor leakage current characteristics. There have been several approach to reduce $Ta_2O_5$ leakage current level as annealing treatment or $SiO_2/Ta_2O_5$ stacked structure but these methods make global dielectric constant decrease. Cava reported that a specific composition of $Ta_2O_5$ and $SiO_2$ bulk ceramic mixture had higher dielectric constant than pure $Ta_2O_5$ ceramic so it proposed the possibility of Ta-Si-O films with high dielectric constant and low leakage current level.
We suggested Ta-Si-O thin films by PEALD method employed a ‘super-cycle’ concept and analyzed the growth behavior of ternary material in PEALD with super-cycle constitution. We defined a ‘super-cycle’ that consists of each PEALD sub-cycle of binary oxide $Ta_2O_5$ and $SiO_2$. The characteristics of super-cycle depend on the sub-cycle number ‘a’ for $Ta_2O_5$ sub-cycle and ‘b’ for $SiO_2$ sub-cycle. And we also showed that Ta-Si-O films by PEALD has good properties so it would be appropriate to alternative dielectric material for $SiO_2$ in DRAM capacitor and gate dielectric. Before Ta-Si-O PEALD, we researched $Ta_2O_5$ PEALD because the $Ta_2O_5$ PEALD is a part of super-cycle for Ta-Si-O PEALD.
We tested 2 plasma source gases, $O_2$ and $H_2$. $Ta_2O_5$ were grown at $260^\circ C$ by $O_2$ PEALD using $Ta(OC_2H_5)_5$ and $O_2$ plasma and $H_2$ PEALD using $Ta(OC_2H_5)_5$ and $H_2$ plasma. The properties of $Ta_2O_5$ films by $O_2$ P...