Abstract:
On a first semiconductor material substrate, an overlying sacrificial layer formed of a second semiconductor material is deposited. In a first region, a first semiconductor material region is formed over the sacrificial layer. In a second region, a second semiconductor material region is formed over the sacrificial layer. The first semiconductor material region is patterned to define a first FinFET fin. The second semiconductor material region is patterned to define a second FinFET fin. The fins are each covered with a cap and sidewall spacer. The sacrificial layer formed of the second semiconductor material is then selectively removed to form an opening below each of the first and second FinFET fins (with those fins being supported by the sidewall spacers). The openings below each of the fins are then filled with a dielectric material that serves to isolate the semiconductive materials of the fins from the substrate.
Abstract:
An analog integrated circuit is disclosed in which short channel transistors are stacked on top of long channel transistors, vertically separated by an insulating layer. With such a design, it is possible to produce a high density, high power, and high performance analog integrated circuit chip including both short and long channel devices that are spaced far enough apart from one another to avoid crosstalk. In one embodiment, the transistors are FinFETs and the long channel devices are multi-gate FinFETs. In one embodiment, single and dual damascene devices are combined in a multi-layer integrated circuit cell. The cell may contain various combinations and configurations of the short and long-channel devices. A high density cell can be made by simply shrinking the dimensions of the cells and replicating two or more cells in the same size footprint as the original cell.
Abstract:
The presence of a facet or a void in an epitaxially grown crystal indicates that crystal growth has been interrupted by defects or by certain material boundaries. Faceting can be suppressed during epitaxial growth of silicon compounds that form source and drain regions of strained silicon transistors. It has been observed that faceting can occur when epitaxial layers of certain silicon compounds are grown adjacent to an oxide boundary, but faceting does not occur when the epitaxial layer is grown adjacent to a silicon boundary or adjacent to a nitride boundary. Because epitaxial growth of silicon compounds is often necessary in the vicinity of isolation trenches that are filled with oxide, techniques for suppression of faceting in these areas are of particular interest. One such technique, presented herein, is to line the isolation trenches with SiN to provide a barrier between the oxide and the region in which epitaxial growth is intended.
Abstract:
A resistive random access memory (RRAM) structure is formed on a supporting substrate and includes a first electrode and a second electrode. The first electrode is made of a silicided fin on the supporting substrate and a first metal liner layer covering the silicided fin. A layer of dielectric material having a configurable resistive property covers at least a portion of the first metal liner. The second electrode is made of a second metal liner layer covering the layer of dielectric material and a metal fill in contact with the second metal liner layer. A non-volatile memory cell includes the RRAM structure electrically connected between an access transistor and a bit line.
Abstract:
A shallow trench is formed to extend into a handle substrate of a semiconductor-on-insulator (SOI) layer. A dielectric liner stack of a dielectric metal oxide layer and a silicon nitride layer is formed in the shallow trench, followed by deposition of a shallow trench isolation fill portion. The dielectric liner stack is removed from above a top surface of a top semiconductor portion, followed by removal of a silicon nitride pad layer and an upper vertical portion of the dielectric metal oxide layer. A divot laterally surrounding a stack of a top semiconductor portion and a buried insulator portion is filled with a silicon nitride portion. Gate structures and source/drain structures are subsequently formed. The silicon nitride portion or the dielectric metal oxide layer functions as a stopping layer during formation of source/drain contact via holes, thereby preventing electrical shorts between source/drain contact via structures and the handle substrate.
Abstract:
A method for forming a complementary metal oxide semiconductor (CMOS) semiconductor device includes forming laterally adjacent first and second active regions in a semiconductor layer of a silicon-on-insulator (SOI) wafer. A stress inducing layer is formed above the first active region to impart stress thereto. Trench isolation regions are formed bounding the first active region and adjacent portions of the stress inducing layer. The stress inducing layer is removed leaving the trench isolation regions to maintain stress imparted to the first active region.
Abstract:
Trenches are formed through a top semiconductor layer and a buried insulator layer of a semiconductor-on-insulator (SOI) substrate. A selective epitaxy is performed to form bulk semiconductor portions filling the trenches and in epitaxial alignment with the semiconductor material of a handle substrate. At least one dielectric layer is deposited over the top semiconductor layer and the bulk semiconductor portions, and is patterned to form openings over selected areas of the top semiconductor layer and the bulk semiconductor portions. A semiconductor alloy material is deposited within the openings directly on physically exposed surfaces of the top semiconductor layer and the bulk semiconductor portions. The semiconductor alloy material intermixes with the underlying semiconductor materials in a subsequent anneal. Within each of the SOI region and the bulk region, two types of semiconductor material portions are formed depending on whether a semiconductor material intermixes with the semiconductor alloy material.
Abstract:
Insulating layers can be formed over a semiconductor device region and etched in a manner that substantially reduces or prevents the amount of etching of the underlying channel region. A first insulating layer can be formed over a gate region and a semiconductor device region. A second insulating layer can be formed over the first insulating layer. A third insulating layer can be formed over the second insulating layer. A portion of the third insulating layer can be etched using a first etching process. A portion of the first and second insulating layers beneath the etched portion of the third insulating layer can be etched using at least a second etching process different from the first etching process.
Abstract:
On a substrate formed of a first semiconductor layer, an insulating layer and a second semiconductor layer, a silicon oxide pad layer and a silicon nitride pad layer are deposited and patterned to define a mask. The mask is used to open a trench through the first semiconductor layer and insulating layer and into the second semiconductor layer. A dual liner of silicon dioxide and silicon nitride is conformally deposited within the trench. The trench is filled with silicon dioxide. A hydrofluoric acid etch removes the silicon nitride pad layer along with a portion of the conformal silicon nitride liner. A hot phosphoric acid etch removes the silicon oxide pad layer, a portion of the silicon oxide filling the trench and a portion of the conformal silicon nitride liner. The dual liner protects against substrate etch through at an edge of the trench between the first and second semiconductor layers.
Abstract:
A transistor includes a body and a semiconductor region configured to stress a portion of the body. For example, stressing a channel of the transistor may increase the mobility of carriers in the channel, and thus may reduce the “on” resistance of the transistor. For example, the substrate, source/drain regions, or both the substrate and source/drain regions of a PFET may be doped to compressively stress the channel so as to increase the mobility of holes in the channel. Or, the substrate, source/drain regions, or both the substrate and source/drain regions of an NFET may be doped to tensile stress the channel so as to increase the mobility of electrons in the channel.