Comments on "A numerical analysis of the storage times of dynamic random-access memory cells incorporating ultrathin dielectrics"

This correspondence discusses two practical aspects which have to be includ ed in the storage time analysis of the above referenced paper. First of all , the value of the diffusion current is overestimated by three orders of ma gnitude in calculating the log time dependence. Secondly, in the total junc tion leakage current, the generation current near the edges of the n(+)-Si/ SiO2 interface is a more important factor than the bulk generation current. Inclusion of those two practical aspects could lead to a different predict ion of storage times in future DRAM tells with thin dielectrics.