Interconnection of opposite conductivity type semiconductor regions

Abstract

A technique for making ohmic electrical interconnections between semiconductor regions of opposite conductivity type, without requiring metallic interconnection lines. This technique has applicability in any circuit using bipolar devices, and in particular is useful to provide a very dense static memory array of bipolar transistors. To join the opposite conductivity regions, intermediate layers are formed including a silicide of a refractory metal, such as W, Mo, Ta, etc. and at least one layer of doped polycrystalline semiconductor material. For a single crystal of silicon having N and P type regions, the refractory metal silicide forms an electrical connection to at least one doped polysilicon layer of a first conductivity type and to either a single crystal semiconductor region of the opposite conductivity type or to another polysilicon layer which also has the opposite conductivity type. As an example, an N type silicon region is interconnected to a P type silicon region by intermediate layers of N polysilicon--refractory metal silicide--P polysilicon.