Method of fabricating field-effect transistor utilizing improved gate sidewall spacers
AssigneeNorth American Philips Corporation, Signetics Division
InventorTeh-Yi J. Chen
Life of this patent
Estimating the term…
Abstract
Gate sidewall spacers are created by a two-step procedure in fabricating a field-effect transistor using a protective material such as silicon nitride to prevent gate-electrode oxidation. In the first step, a layer (32) of insulating material is conformally deposited and then substantially removed except for small spacer portions (34) adjoining the sidewalls of a doped non-monocrystalline semiconductor layer (20A) destined to become the gate electrode (36). The second step consists of performing an oxidizing heat treatment to increase the thickness of the spacer portions. No significant gate dielectric encroachment occurs. Also, the spacers achieve a profile that substantially avoids electrical shorts.
Claims15
A method comprising the steps of:
- forming a structure in which a patterned electrically insulating cover layer overlies a similarly patterned conductively doped layer of non-monocrystalline semiconductor material lying on a dielectric layer that lies along the upper surface of a major doped region of a semiconductor body;
- depositing a further layer of electrically insulating material on the upper surface of the structure;
- removing largely all of the further layer except for small spacer portions adjoining the sidewalls of the doped layer;
- and
- performing an oxidizing heat treatment to oxidize parts of the doped layer and parts of the major region along its upper surface, and implanting a main semiconductor dopant into selected parts of the major region using the spacer portions and the material of the patterned layers to substantially block the main dopant from being implanted into the part of the major region below the spacer portions and the material of the patterned layers;
- characterized in that the cover layer comprises a protective material that substantially prevents the underlying non-monocrystalline material of the doped layer, except for the non-monocrystalline material along its sidewalls, from being oxidized during the heat treatment, portions of the doped layer along its sidewalls being oxidized during the heat treatment to increase the thickness of the spacer portions.
A method wherein;
- a structure is formed in which a patterned electrically insulating cover layer overlies a similarly patterned conductively doped non-monocrystalline silicon layer lying on an electrically insulating layer that lies along the upper surface of a major region of a first conductivity type of a monocrystalline silicon body;
- a further layer consisting principally of silicon dioxide is deposited on the upper surface of the structure;
- largely all of the further layer is removed except for small spacer portions that adjoin the sidewalls of the doped layer and extend fully across its thickness;
- and
- an oxidizing heat treatment is performed to oxidize parts of the doped layer and parts of the major region along its upper surface, and ions of a main dopant of a second conductivity type opposite to the first conductivity type are implanted into selected parts of the major region through its upper surface using the spacer portions and the material of the patterned layers as a shield that substantially blocks ions of the main dopant from being implanted into the part of the major region below the spacer portions and the material of the patterned layers;
- characterized in that the cover layer comprises a protective material that substantially prevents the underlying non-monocrystalline silicon of the doped layer, except for the non-monocrystalline silicon along its sidewalls, from being oxidized during the heat treatment, portions of the doped layer along its sidewalls being oxidized during the heat treatment to increase the thickness of the spacer portions.
Dependencies are inferred from the claim text.